Smart drugs

Memory is one of the greatest feats of the human brain, yet it has proven a peculiarly stubborn phenomenon to understand – until now. The ability to memorise and recall what has happened in the past is a key facet of who we are, but how the brain does this is largely a mystery. However, in recent years a set of experiments on the humble fruit fly has at last given scientists an important insight into the memory machinery of the mind. If all goes to plan, the research could soon lead to a class of new memory-enhancing drugs that have already been dubbed "Viagra for the brain".

The modern era of research into memory began nearly 40 years ago, but it is only in the past decade or so that the most startling advances have been made. One of the scientists at the forefront is Tim Tully, professor of genetics at Cold Spring Harbor Laboratory on Long Island, New York, who is beginning to build a new understanding of memory based on studies of the fruit fly. He has discovered a number of strains of mutant fruit flies with unusual memories, and has started to name them after the dogs used in some of the first memory experiments carried out more than 60 years ago by the great Russian behavioural psychologist Ivan Pavlov. "I applied the Pavlovian notion to this fruit-fly task, and essentially developed Pavlov's flies," Tully says.

What Pavlov showed when he rang a bell before feeding his experimental dogs was something called associative memory. The dogs learnt to associate the sound of the bell with food. The result was that their bodies eagerly anticipated being fed as soon as they heard the bell. It manifested itself in both their general excitement and their unconscious physiology, namely, salivating. It is the sort of associative memory displayed by every pet cat and dog that has ever rushed to the kitchen on hearing the sound of a tin being opened.

Tully has taken research into associated memory to a new level in his studies of fruit flies. He uses a device called the "teaching machine" to separate the flies that have a good memory from those that don't. The machine exposes a group of flies to an odour followed by a mild but unpleasant electric shock transmitted through the floor of the machine. Another odour is then wafted over the same batch of flies, but this time there is no electric shock. Later on, each fly in the group can choose which odour they prefer by deciding on one of two compartments they can move into. Those with good memories are more likely to choose the compartment with the second odour, whereas those without such good memories are unlikely to show a preference. Tully found that most flies do in fact learn to move away from the smell associated with the shock, but a few don't, and it is these that he studied more closely.

One strain of "dumb" fly turned out to carry a defect in a gene for a protein known to be active in the fly's nervous system, called Creb (cyclic adenosine monophosphate response element binding protein). The protein was known, from earlier experiments on sea slugs by Professor Eric Kandel of Columbia University, to be essential for converting short-term memory into long-term memory. Tully went further than Kandel, who won a Nobel prize in 2000 for his work, by showing that boosting levels of Creb in fruit flies gave them the equivalent of a photographic memory.

Tully genetically engineered fruit flies with the Creb gene permanently switched into the "on" position – normally, the gene can be switched on and off. Normal flies took about 10 times in the teaching machine to learn to associate a particular odour with an electric shock. Tully's "superflies" learned after just one "lesson". Creb was evidently helping the conversion of short-term into long-term memory. Similar experiments conducted on mice showed much the same thing, indicating that the Creb gene plays much the same role in lower and higher organisms, almost certainly including humans.

It is, of course, the human aspect of this research that is proving most interesting to biotechnology companies in search of "smart drugs" to improve memory. Tully says that the tiny fruit fly can lead the way. "We can do gene discovery in simple systems like the fly with every expectation that it is a fast track, an economy of scale, to come up with candidate genes involved in the process in humans," says Tully.

Although the memory of a fly and a person are in quite different leagues, they both rely on similar neural circuitry involving essentially the same neurochemicals, such as Creb. As Tully explains: "Flies are Philco radios, and we are Macintosh computers, but both run on transistors. The brain of the fly works very much like the brain of other animals, including humans."

Long-term memory is aptly named. If you can remember anything from last week, it is stored in long-term memory, whereas anything that happened a few minutes ago is still in short-term memory, which could easily be lost with time. Long-term memory is a particular problem for people with some forms of dementia, such as Alzheimer's disease, which is why scientists are so interested in understanding how it works in order that it may be corrected or enhanced with drugs. Evidence from a number of studies suggests that long-term memory requires genes to be turned on to make proteins such as Creb, whereas short-term memory does not. Long-term memory is also known to involve the strengthening of connections (the synapses) between nerves in the brain, almost certainly involving proteins such as Creb.

In another set of experiments, Tully showed that long-term memories are essentially stored as part of a chemical process (rather than electrical activity) within the nervous system. Yet the retrieval of these memories is an electrical process involving the firing of nerve cells across nerve connections.

People with exceptional long-term memories are renowned. Such "memory men" include the famous Solomon Veniaminovich Shereshevsky, a Russian mnemonist in the 1930s who carried out astounding feats of recall. In one test, he could remember an extremely complex set of mathematical formulae perfectly (down to a deliberately introduced error) 15 years after seeing them for the first and only time.

In spite of his exceptional long-term memory, however, Shereshevsky could not remember two separate conversations with the same person, making him socially inept and incapable of holding down a job, Tully says. "These observations suggest that having too much long-term memory may not be a good thing," he says.

It is certainly the case that the brain cannot have an infinite memory. Some experiences in life have to be forgotten, otherwise the brain's "hard drive" – composed of some 70 trillion nerve connections – would soon be clogged up. Tully suggests that the Creb switch might have evolved to function as an "information filter" for most circuits in the brain. "Exposure to one experience on a single day may not be sufficient to activate the switch and induce long-term memory formation, thereby preserving some synaptic [nerve] connections for more important, recurrent events," he says.

Tully's most recent scientific paper, published in Current Biology, explains how he used gene chips – small plates to identify those active genes switched into the "on" position – to count the number of fruit-fly genes involved in long-term memory. He now has a list of nearly 100 genes that play some sort of role in learning and memory. Many, if not all of these will have analogues in humans, making them invaluable in finding the fundamental mechanism responsible for memory formation and retrieval in people – and finding chemicals for potential smart drugs.

"Humans always thought that learning and memory were something special, almost spiritual, but people are now realising that it is just as biological as kidney function. In the near future, I anticipate the development of traditional pharmaceutical treatments for long-term memory loss," Tully says. "The first of these may be ready for use sometime in the next 10 years."

So, if Tully's work continues to advance as it has over the past decade, memory loss could soon be as treatable as kidney damage. Finding Viagra for the mind ultimately rests on knowing that memories are made of... molecules.