鈥淚F YOU can describe what is happening in my brain as I picture the Mona
Lisa, I鈥檒l quit,鈥� says Eric Kandel, a neurobiologist at Columbia University in
New York City. 鈥淚鈥檒l consider the problem of memory nearly solved.鈥�
As the leader in his field, Kandel knows his job is safe for a long while
yet. Indeed, much of what we know about the molecular basis of memory comes from
his decades of pioneering work with an animal that has a simple nervous system,
the giant marine snail Aplysia.
Give Aplysia a single electric shock to its tail and it becomes
hypersensitive for several minutes, retracting its breathing siphon and gill at
the slightest touch. If the shock is repeated several times, the snail stays
hypersensitive for days.
Advertisement
In the mid-1970s, Kandel鈥檚 team discovered that the snail鈥檚 neurons need a
neurotransmitter called serotonin to recall anything at all. Serotonin triggers
a chain reaction by activating the enzyme cAMP-dependent protein kinase A. PKA
modifies other proteins, which strengthen the electrical connections between
neurons for several minutes.
This is analogous to short-term memory. For long-term memory, a factor that
switches on genes is needed, and in 1990 Kandel鈥檚 team found it. They discovered
that PKA also turns on the CREB1 protein, which activates genes that manufacture
proteins. These, in turn, help build more synapses. When Kandel鈥檚 team blocked
CREB1, the neurons were only capable of short-term memory. Conversely, if they
injected activated CREB1 into neurons, the cells grew new synapses with far less
stimulation. Indeed, boosting a CREB1-like molecule in flies gives the insects
the equivalent of a photographic memory鈥攖hey learn in one training session
what normal flies need ten sessions to learn.
The potential for a drug that could enhance human memory was not lost upon
Kandel and his team. In December 1998, they injected mice with Rolipram, a drug
that increases the activity of PKA and therefore indirectly boosts CREB1. The
treated animals remembered associations better (Proceedings of the National
Academy of Sciences, vol 95, p 15 020; vol 96, p 5280). Rolipram isn鈥檛
suitable as a memory drug for humans, but Kandel imagines that other drugs might
be available in 10 to 15 years.
Knowledge of neuron-shaping molecules won鈥檛 completely explain memory.
Consider Kandel鈥檚 visit to the Louvre. How does his brain capture that
particular experience? The brain might be thought of as a collection of
different systems, one memorising spatial information鈥攚here the painting
was hanging鈥攚hile others record the sounds and smells of the crowd. But
how do these systems interact?
An answer might come through brain imaging. If a defined region in the human
brain lights up when someone admires a piece of art, for example, it should be
possible to study genes that are active in the same region in mice. Kandel鈥檚 lab
is already busy with this approach (Cell, vol 83, p 1211). But even if
we could systematically map the brain, it would not fully explain what happens
when Kandel remembers the Mona Lisa. For that we would need to understand
consciousness. Will science ever solve that problem? 鈥淟et鈥檚 talk again in a few
decades,鈥� says the 70-year-old neurobiologist.
