Cell Biological Studies of Memory 
Eric R. Kandel, M.D. — Senior Investigator 
Dr. Kandel is also University Professor of Physiology and Psychiatry at the Columbia University College of 
Physicians and Surgeons. He was born in Vienna, Austria; graduated from Harvard College, having ma- 
jored in history and literature; and received his M.D. degree from New York University School of Medicine. 
He took postdoctoral training with Wade Marshall at the Laboratory of Neurophysiology at the NIH and 
with Ladislav Tauc at the Institut Morey in Paris. He was founding director of the Center for Neurobiology 
and Behavior at Columbia. Dr. Kandel is a member of the National Academy of Sciences and counts 
among his honors the Lasker Award, the Gairdner Award, and the National Medal of Science. 
LEARNING is commonly divided into two ma- 
jor types, declarative and reflexive. Declara- 
tive learning refers to the acquisition of informa- 
tion about persons, places, or things. Reflexive 
learning refers to the acquisition of procedures 
and motor skills. 
In the past our laboratory has focused primarily 
on elementary forms of reflexive learning as man- 
ifest in the gill-withdrawal reflex of the marine 
snail Aplysia. We showed that this simple reflex 
can be modified by both nonassociative and asso- 
ciative learning, giving rise to short- and long- 
term memory, whose duration is a function of the 
number of training trials. 
To analyze the relationship between the short- 
and long-term processes for nonassociative learn- 
ing (sensitization), we focused on one compo- 
nent of the neural circuit of this reflex — the 
connections between the siphon sensory neuron 
and the gill motor neurons. Here we found that 
both the short- and long-term processes involve 
an increase in transmitter release. Whereas the 
short-term process reflects enhanced transmitter 
release from preexisting synaptic connections 
due to covalent modification of preexisting pro- 
teins, the long-term process results from alter- 
ations in gene expression and the growth of syn- 
aptic connections. 
What molecular mechanisms contribute to de- 
clarative forms of learning? In an attempt to com- 
pare the biochemical mechanisms underlying re- 
flexive forms of learning with those underlying 
declarative learning, we have turned to the hip- 
pocampus in the mammalian brain. 
Since Milner's pioneering work in the late 
1950s, the hippocampus has been known to be 
important for aspects of long-term declarative 
memory storage in humans and other mammals. 
The hippocampus may be essential for initially 
storing long-term memory for days or weeks be- 
fore the memory trace is consolidated elsewhere, 
perhaps in the cerebral cortex. In 1973 Bliss and 
L0mo first demonstrated that a brief, high- 
frequency train of action potentials in the perfor- 
ant path increases the excitatory synaptic poten- 
tial in the granule cells. The increase can last for 
hours and, under some circumstances, even for 
weeks. They called this facilitation long-term po- 
tentiation, or LTP. Later studies showed that LTP 
occurs at each of the three major synaptic path- 
ways in the hippocampus. 
Recent studies in the CAl region, by Tim Bliss, 
Charles Stevens (HHMI, the Salk Institute), 
Richard Tsien, and their colleagues, have pro- 
vided insights into the cellular mechanisms in- 
volved in the acquisition and maintenance of 
LTP. Despite important differences, this form of 
synaptic plasticity in the mammalian brain may 
bear certain similarities on the cellular level to 
presynaptic facilitation that accompanies sensiti- 
zation in Aplysia. In both cases synaptic connec- 
tions are strengthened through an enhancement 
of transmitter release, and both the short- and the 
long-term changes occur at the same synaptic 
locus. 
Therefore, Tom O'Dell, Seth Grant, and I have 
begun to address the question. What are the mo- 
lecular steps involved in LTP? It has been known 
for several years that calcium influx through a 
glutamate receptor of the TV-methyl-o-aspartate 
type is critical for the induction of LTP. Less is 
known, however, about the subsequent biochem- 
ical steps responsible for LTP expression and 
maintenance. Several studies using various kinase 
inhibitors have implicated as important events 
following calcium influx the activation of two 
serine-threonine protein kinases: kinase C (PKC) 
and calcium/calmodulin-sensitive kinase II 
(CaMK-II). Consistent with the involvement of 
PKC and CaMK-II in LTP, high levels of these ki- 
nases are found in hippocampal pyramidal cells. 
In addition to these serine-threonine kinases, 
hippocampal neurons also express high levels of 
protein tyrosine kinase activity. Indeed, the brain 
is a richer source of tyrosine kinases than any 
other organ, and these kinases are particularly 
enriched in the hippocampus. Subcellular frac- 
tionation studies showing highest levels of pro- 
tein tyrosine kinase activity in crude synaptic vesi- 
cle fractions have suggested a role for tyrosine 
233 
