proteins are exclusively located in the cell nucleus. A further question is 
whether the induced proteins lead to an increase in the amount of AP-1 
complex formation in the cell nucleus. Fos is a member of the leucine 
zipper family of transcription factors and is known to bind as a heterodimer 
with members of the Jun transcription factor family to the AP-1 enhancer 
element (but also with other proteins, see Hai and Curran 1991). The AP-1 
consensus sequence is TGAGTCA. We used a small segment of the GALV 
enhancer that contains the AP-1 consensus in a gel shift reaction with extracts 
of nuclei isolated from normal and cocaine-induced striatum to determine 
whether the elevation in Fos protein content led to an increase in AP-1 
complex formation. The gel mobility shift assay yielded an AP-1 shift in 
control striatum and an elevated amount of AP-1 complex within 1 hour 
following cocaine injection (figure 13). In both cases, the primary AP-1 
shift contained Fos-immunoreactive proteins since the entire complex could 
be shifted further still by addition of the Fos antibody. The entire amount of 
complex was further retarded, which suggests that a Fos-immunoreactive 
protein was an element in all the complexes formed in both the control and 
cocaine-induced conditions. Thus, the elevated capacity for striatal extracts 
to form AP-1 complexes in the cocaine-treated condition is consistent with the 
Northern blot, Western blot, and immunocytochemical data demonstrating an 
increase in c -fos mRNA and Fos/Fra protein(s). 
The most interesting effect of cocaine on IEG expression occurred when the 
effects of a multiple dosing regimen were examined (figure 14). With several 
doses given over the course of a single day (4 injections of 30 mg/kg IP, at 
2-hour intervals), we observed a biphasic effect: The initial injection caused 
the familiar marked (fivefold to sevenfold) increase in c -fos mRNA, but the 
response to subsequent injections was nearly completely attenuated. Figure 
15 shows that the c -fos induction was completely suppressed in response to 
the fourth injection of a series of 30 mg/kg injections given 2 hours apart over 
the course of a single day. We have mapped the time course of suppression 
following a single acute cocaine injection and have found that a short-lived 
period (approximately 1 to 2 hours) of suppression (or refractoriness to 
induction) occurs following a single injection of cocaine (not shown). However, 
the duration of suppression becomes much more prolonged with multiple 
injections. In the case illustrated in figure 15, where four successive injections 
were given, the duration of suppression lasted for 3 to 4 days. The long 
duration required for recovery of full sensitivity to cocaine was mapped out by 
giving challenge doses of cocaine at progressively greater intervals following 
the subchronic series of injections (figures 16 and 17). Our data indicate that 
the duration of suppression cumulates with multiple injections. Thus, multiple 
injections of cocaine produce a long-term uncoupling of dopaminergic 
neurotransmission from dopamine-regulated gene expression. 
199 
