Jun-B is a negative regulator of c -jun (Chiu et al. 1989). Since there are three 
mammalian Jun proteins (c-Jun, Jun-B, Jun-D) and at least four Fos family 
members (c-Fos, Fos-B, Fra-1 , Fra-2) to which the Jun proteins can bind, 
there exist many possible combinations of Fos-Jun proteins. In addition, 
there is also the possibility of Jun-Jun and Fos-Fos homodimers with 
unknown potentials. Other members of the two families also probably 
have unrealized potentials. For example, it has recently been shown that 
a truncated form of Fos-B inhibits the transcriptional activation of Fos/Jun 
heterodimers (Nakabeppu and Nathans 1991). Similarly, Fra-1 can combine 
with Jun and bind to the AP-1 recognition element, raising the possibility that 
this Fos family member might play a role in regulation at the AP-1 site (Cohen 
et al. 1989). Finally, it is now becoming apparent that Fos/Jun can regulate 
other receptors; for example, Fos and Fos/Jun can inhibit the glucocorticoid 
receptor (Touray et al. 1991), and in turn, other factors, such as retinoic acid 
(Schule et al. 1991), can regulate AP-1 -responsive genes. 
The issue is further confused because in many studies, especially for neural 
tissue, stimulation leads to activation of a variety of Fos and Jun family 
members (in addition to other lEGs). For example, stimulation with D1 
dopamine agonists can cause rapid expression of c -fos, c -jun, and jun - B in 
striatum. Are these lEGs all in the same cell, or more attractively, are c -fos 
and c -jun expressed in one type of cell leading to transcriptional activation 
while in another cell c -fos and jun - B expression is producing transcriptional 
inhibition? It is clear that the answers to such questions are of great 
importance. However, in some tissues, such as the dentate granule cells after 
seizure activity, it seems clear that c -fos, c -jun, and jun - B activation all occur in 
the same cells (B.J. Chiasson and H.A. Robertson, unpublished observations). 
In addition to this complexity, it is now suspected that the so-called flanking 
sequences to the AP-1 site play an important role in the binding of Jun proteins 
to the AP-1 site (Ryseck and Bravo 1991). It is important to note that AP-1 -like 
sites are sometimes common, and it is not immediately obvious which sites 
might be active (e.g., Hengerer et al. 1990). It is also clear that differential 
regulation of lEGs does occur, and there are several instances in which c -fos 
is expressed in the absence of c -jun or together with jun - B rather than c -jun 
(Bartel et al. 1989; Naranjo et al. 1991). There are instances in which the c -jun 
message is expressed following stimulation but no c-Jun protein appears to 
be expressed, suggesting, among other possibilities, that IEG protein levels 
might be regulated at the transcription level (Rusak et al., in press). 
In summary, it is now clear that the picture of activation of lEGs that encode 
nuclear proteins that regulate gene expression is misleading. The more 
accurate picture has been described as regulation by committee, a series of 
58 
