René St-Arnaud and Isabelle Quélo

Genetics Unit, Shriners Hospital, and Departments of Surgery and Human Genetics, McGill University, Montréal (Québec) Canada H3G 1A6

Received 3/27/98 Accepted 4/22/98

3. AP-1

The acronym AP-1 stands for ‘activating protein-1’ and was first coined by the group of Robert Tjian to identify proteins regulating the expression of the human metallothionein promoter (33). Purification of these proteins using DNA affinity chromatography revealed that several proteins co-purified following the sequence-specific chromatographic step (34), suggesting that more than one protein could recognize the regulatory DNA binding site. A series of elegant biochemical studies by several laboratories subsequently identified these proteins as members of the fos and jun proto-oncogene families.

3.1. the fos family

The c-fos proto-oncogene is the cellular homolog of the transforming oncogene of the FBJ and FBR murine sarcoma viruses (35). These viruses induce osteosarcomas in vivo, hence the name of the gene, fos, for FBJ/FBR osteosarcoma. The Fos protein is a nuclear phosphoprotein of around 55000 Da molecular weight (36). Immunoprecipitation using Fos-specific antibodies demonstrated that a second nuclear protein of Mr 39 000 Da (p39) co-precipitate with the Fos protein (37). The p39 protein was subsequently identified as the c-Jun protein (38).

The antisera raised against Fos-derived peptides allowed to identify two additional members of the family, fos-related antigen-1 (fra-1) and -2 (fra-2) (39). Yet another member of the family, fosB, was cloned by screening a library of serum-inducible genes with a probe corresponding to the DNA-binding domain of c-Fos (40). All fos family members are expressed in osteoblasts, although the composition of osteoblastic AP-1 complexes appears to vary as a function of bone cell proliferation and differentiation (32).

There is extensive sequence similarity between the different family members within the basic DNA binding domain and the dimerization interface, a protein motif that has been dubbed the leucine zipper (41). This motif consists of heptad repeats of leucine residues which align along one face of an alpha helix. When aligned in parallel, the hydrophobic faces of two complementary helices form a coiled coil (42). Leucine zipper dimerization serves to juxtapose adjacent regions of each of the dimer ’s partners that are rich in basic amino acid residues and that serve as the DNA binding domain of the dimer (43, 44). Thus Fos is recognized as the paradigm of the bZIP (basic region — leucine ZIPper) class of transcription factors, which include many different subfamilies with distinct dimerization and DNA binding properties.

A number of dimerizing partners for c-Fos have been identified. These include the jun family members (45, 46), ATF (activating transcription factors) proteins (47, 48), and the recently identified Maf (the cellular homolog of the transforming oncogene from the AS42 avian transforming retrovirus), and Nrl (retina specific gene) gene families (49). All of these fos heterodimers can bind the canonical AP-1 site (see below), although with different affinities (46-49). Moreover, the c-fos/ATF heterodimers can bind to the CRE site (cAMP response element; ref. 47).

3.2. the jun family

The v-jun oncogene was identified as the transforming oncogene of avian sarcoma virus-17 (jun is the Japanese word for 17) (50). The observed sequence similarity between v-jun (and its cellular counterpart c-jun) and the DNA-binding domain of the yeast transcription factor GCN4 immediately suggested that the c-jun protein could function as a DNA-binding transcriptional regulator (51). Several laboratories were quick to recognize that the GCN4 DNA binding site was very similar to the DNA sequence utilized to purify AP-1 proteins by DNA-affinity chromatography. This observation launched a series of studies which culminated in the identification of the Fos and Jun proteins as components of the AP-1 complexes (38; 52).

Differential screening of cDNA libraries from serum-stimulated cells allowed the identification of two additional family members which were named junB and junD (53, 54). Jun family members are also bZIP transcription factors and heterodimerize with fos family members through their leucine zippers. The diversity of active AP-1 complexes is further increased by the fact that jun family members can homodimerize and heterodimerize amongst themselves (46; 55). Additional dimerization partners for jun family members include the CREB/ATF (cAMP response element binding protein/activating transcription factor), Maf, and Nrl families of bZIP proteins (49; 56).

The heterodimers formed between the different Jun and Fos proteins have an increased binding activity compared to the Jun/Jun homodimers (55). The binding affinities of the different Jun dimers vary as follows: c-Jun/c-Jun > JunD/JunD > JunB/JunB (55). It is interesting to note that in certain assays, JunB has been shown to act as a negative modulator of c-Jun function (57, 58). Thus several parameters can affect the expression of an AP-1 responsive gene: the response element itself (see below), the relative binding affinities of the various AP-1 complexes, as well as their specific composition at any given time.

3.3 the AP-1 binding site

The consensus AP-1 binding site is the palindrome TGA(C/G)TCA (33, 34). The AP-1 sites are responsible for mediating induction by tumor promoters such as 12-O-tetradecanoylphorbol-b-acetate (TPA) and are therefore sometimes confusingly labeled TRE, for TPA-response element. Many genes contain AP-1 sites within their promoter region. The AP-1-responsive genes that are of particular relevance to bone cell biology include collagenase (59) and osteocalcin (60).

Biochemical analysis using in vitro-translated AP-1 proteins has revealed that the sequences adjacent to the core binding site can influence the stability of the Jun/DNA complexes (55). The AP-1 site is also very similar to the cAMP response element (CRE): TGACGTCA. Indeed, Jun family members have been shown to bind CREs alone (55) or when heterodimerized with ATF/CREB factors (49; 56).

3.4 post-translational modulation of AP-1 function

AP-1 activation is caused by: (i) increased expression of Fos and Jun proteins; and (ii) post-translational modification of Fos and Jun by phosphorylation (61, 62). The complex subject of the regulation of c-fos expression is adequately summarized in the review by Hipskind and Bilbe in this issue of Frontiers in Bioscience (63). Jun family members are similarly induced by a variety of signals including serum, cytokines, tumor promoters, and genotoxic stresses such as treatment with alkylating agents, ultraviolet (UV) light and ionizing radiation. It is the analysis of the Jun response to UV irradiation that led to the cloning of a member of the stress-activated protein kinases (SAPKs), the c-Jun amino-terminal kinase (JNK) (64, 65).

The JNK protein kinases phosphorylate the NH₂-terminal activation domain of c-Jun on residues Ser63 and Ser73 (65). The transcriptional activity of c-Jun is increased following this post-translational modification (61). Site-directed mutagenesis confirmed that phosphorylation of these serine residues is essential for stimulation of c-Jun-dependent transcription (61; 66). Thus any transcriptional co-factor implicated in mediating the c-Jun transcriptional activation function must be able to interact with the phosphorylated form of the protein.