granule cells (Dragunow and Robertson 1987a; White and Gall 1987; Dragunow 
and Robertson 1987b). There is a continuing discussion over the possible 
role of c -fos and other lEGs in the phenomenon of long-term potentiation 
(LTP), which is the persistent (up to several weeks) increase in the excitatory 
postsynaptic potential elicited at synapses following a high-frequency 
stimulation of afferent neurons. Early studies suggested that LTP could 
develop in the absence of c -fos induction (Douglas et al. 1988), and this 
was confirmed by Cole and colleagues (1989) and Wisden and coworkers 
(1990) who did, however, demonstrate a correlation between LTP and 
induction of NGFI-A. Surprisingly, LTP induced using “burst” stimulation 
involves c -fos induction, but LTP induced by the same number of stimulations 
evenly spaced over the same train duration (i.e., LTP induced by the same 
number of stimulations but with different spacing) is not associated with 
increases in Fos protein (Dragunow et al. 1989). However, all earlier studies 
were done on anesthetized animals, and more recent studies on the 
development of LTP in unanesthetized animals revealed that, in this situation, 
c-fos induction accompanies LTP induction (Jeffery et al. 1990). From all this 
work, several points are clear. First, it is already known that LTP develops in 
milliseconds, whereas c -fos mRNA only appears between 10 and 15 minutes 
later, with c-Fos protein appearing shortly after. The same is true for other 
lEGs. Thus, although it remains possible that lEGs play a role in maintenance 
of LTP (but see Jeffery et al. 1 990), it seems unlikely that they play any role in 
induction of LTP. Second, some stimulation parameters produce LTP with c -fos 
activation, but others produce LTP without c -fos activation, which suggests (as 
was already suspected) that there are different types of LTP, differing perhaps 
in persistence or other properties. 
The kindling model is interesting in that, unlike LTP, induction of kindling 
appears to be a permanent change in the brain (for a review of the differences 
between kindling and LTP, see Cain 1989). A kindling stimulus will activate 
c -fos in the dentate granule cells and in the hippocampal pyramidal neurons 
(Dragunow and Robertson 1987a; White and Gall 1987). Other lEGs 
(c-jun t jun- B, NGFI-A) are also activated by the kindling stimulus (H.A. 
Robertson, unpublished data). It is known that kindling stimuli lead to 
synaptic reorganization in the molecular layer of the dentate gyrus of the 
hippocampus (Sutula et al. 1988). Kindling has also been shown to produce 
increased synthesis of neurotrophic factors in the dentate gyrus (Ernfors 
et al. 1991; B.J. Chiasson and H.A. Robertson, unpublished observations), 
and Funabashi and colleagues (1988) demonstrated that injections of 
antibody against NGF into the cerebral ventricles during stimulation would 
prevent kindling. In at least one other system, lEGs appear to play a role 
in the regulation of NGF synthesis (Hengerer et al. 1990, see below). Thus, 
it is reasonable to suggest that I EG induction may be the first step in a 
60 
