Cryo-EM Structure of the Human FLCN-FNIP2-Rag-Ragulator Complex

mTORC1 controls anabolic and catabolic processes in response to nutrients through the Rag GTPase heterodimer, which is regulated by multiple upstream protein complexes. One such regulator, FLCN-FNIP2, is a GTPase activating protein (GAP) for RagC/D, but despite its important role, how it activates the Rag GTPase heterodimer remains unknown. We used cryo-EM to determine the structure of FLCN-FNIP2 in a complex with the Rag GTPases and Ragulator. FLCN-FNIP2 adopts an extended conformation with two pairs of heterodimerized domains. The Longin domains heterodimerize and contact both nucleotide binding domains of the Rag heterodimer, while the DENN domains interact at the distal end of the structure. Biochemical analyses reveal a conserved arginine on FLCN as the catalytic arginine finger and lead us to interpret our structure as an on-pathway intermediate. These data reveal features of a GAP-GTPase interaction and the structure of a critical component of the nutrient-sensing mTORC1 pathway.

Conformational Landscape of the p28-Bound Human Proteasome Regulatory Particle

The proteasome holoenzyme is activated by its regulatory particle (RP) consisting of two subcomplexes, the lid and the base. A key event in base assembly is the formation of a heterohexameric ring of AAA-ATPases, which is guided by at least four RP assembly chaperones in mammals: PAAF1, p28/gankyrin, p27/PSMD9, and S5b. Using cryogenic electron microscopy, we analyzed the non-AAA structure of the p28-bound human RP at 4.5 Å resolution and determined seven distinct conformations of the Rpn1-p28-AAA subcomplex within the p28-bound RP at subnanometer resolutions. Remarkably, the p28-bound AAA ring does not form a channel in the free RP and spontaneously samples multiple "open" and "closed" topologies at the Rpt2-Rpt6 and Rpt3-Rpt4 interfaces. Our analysis suggests that p28 assists the proteolytic core particle to select a specific conformation of the ATPase ring for RP engagement and is released in a shoehorn-like fashion in the last step of the chaperone-mediated proteasome assembly.

Atomistic view of the conformational activation of Src kinase using the string method with swarms-of-trajectories

The inactive-to-active conformational transition of the catalytic domain of human c-Src tyrosine kinase is characterized using the string method with swarms-of-trajectories with all-atom explicit solvent molecular dynamics simulations. The activation process occurs in two main steps in which the activation loop (A-loop) opens first, followed by the rotation of the alphaC helix. The computed potential of mean force energy along the activation pathway displays a local minimum, which allows the identification of an intermediate state. These results show that the string method with swarms-of-trajectories is an effective technique to characterize complex and slow conformational transitions in large biomolecular systems.

Ultrastructural nuclear import assay

Electron microscopy (EM) has been used for several decades to study the mechanisms of nuclear transport. In early studies of nuclear import, gold-conjugated nuclear proteins were microinjected into cells and followed by EM. As the components of the nuclear pore complex (NPC) and soluble mediators of nuclear import were cloned and characterized, gold-conjugated antibodies were utilized to sublocalize the components of the nuclear transport machinery by immuno-EM. Further, gold-conjugated recombinant proteins were used to probe permeabilized cells or isolated nuclear envelopes and characterize binding sites for these proteins at the NPC. More recently, recombinant gold-conjugated nuclear proteins were used in in vitro nuclear import assays to help dissect the mechanisms of nuclear import. We have used this ultrastructural nuclear import assay to study the nuclear import of the transcription factor PU.1. The results showed that this import requires energy but is carrier-independent. In the presence of energy, gold-conjugated PU.1 shifted to the nuclear side of the NPC and the inside of the nucleus. In conjunction with biochemical assays, these results indicated that this shift involved Ran-dependent binding of PU.1 to NUP153, a nucleoporin situated at the nuclear side of the NPC. Here we describe in detail the methods used in the ultrastructural nuclear import assay including preparation of recombinant protein, gold conjugation, in vitro nuclear import assay, electron microscopy, and data analysis.

Nucleoporins as components of the nuclear pore complex core structure and Tpr as the architectural element of the nuclear basket

The vertebrate nuclear pore complex (NPC) is a macromolecular assembly of protein subcomplexes forming a structure of eightfold radial symmetry. The NPC core consists of globular subunits sandwiched between two coaxial ring-like structures of which the ring facing the nuclear interior is capped by a fibrous structure called the nuclear basket. By postembedding immunoelectron microscopy, we have mapped the positions of several human NPC proteins relative to the NPC core and its associated basket, including Nup93, Nup96, Nup98, Nup107, Nup153, Nup205, and the coiled coil-dominated 267-kDa protein Tpr. To further assess their contributions to NPC and basket architecture, the genes encoding Nup93, Nup96, Nup107, and Nup205 were posttranscriptionally silenced by RNA interference (RNAi) in HeLa cells, complementing recent RNAi experiments on Nup153 and Tpr. We show that Nup96 and Nup107 are core elements of the NPC proper that are essential for NPC assembly and docking of Nup153 and Tpr to the NPC. Nup93 and Nup205 are other NPC core elements that are important for long-term maintenance of NPCs but initially dispensable for the anchoring of Nup153 and Tpr. Immunogold-labeling for Nup98 also results in preferential labeling of NPC core regions, whereas Nup153 is shown to bind via its amino-terminal domain to the nuclear coaxial ring linking the NPC core structures and Tpr. The position of Tpr in turn is shown to coincide with that of the nuclear basket, with different Tpr protein domains corresponding to distinct basket segments. We propose a model in which Tpr constitutes the central architectural element that forms the scaffold of the nuclear basket.

Direct evidence for membrane pore formation by the apoptotic protein Bax

Direct imaging of the interaction of the apoptotic protein, Bax, with membrane bilayers shows the presence of toroidal-shaped pores using atomic force microscopy. These pores are sufficiently large to allow passage of proteins from the intermitochondrial space. Both the perturbation of the membrane and the amount of protein bound to the bilayer are increased in the presence of calcium. The results from the imaging are consistent with leakage studies from liposomes of the same composition. The work shows that Bax by itself can form pores in membrane bilayers.

c-Cbl-dependent EphA2 protein degradation is induced by ligand binding

The EphA2 receptor protein tyrosine kinase is overexpressed and functionally altered in a large number of human carcinomas. Despite its elevated levels in cancer, the EphA2 on the surface of malignant cells demonstrates lower levels of ligand binding and tyrosine phosphorylation than the EphA2 on non-transformed epithelial cells. In our present study, we demonstrate that ligand-mediated stimulation causes EphA2 to be internalized and degraded. The mechanism of this response involves ligand-mediated autophosphorylation of EphA2, which promotes an association between EphA2 and the c-Cbl adaptor protein. We also show that c-Cbl promotes stimulation-dependent EphA2 degradation. These findings are important for understanding the causes of EphA2 overexpression in malignant cells and provide a foundation for investigating EphA2 as a potential target for therapeutic intervention.

Structures and dynamics of Drosophila Tpr inconsistent with a static, filamentous structure

Here we report immunofluorescence localizations of the Drosophila Tpr protein which are inconsistent with a filament-forming protein statically associated with nuclear pore complex-associated intranuclear filaments. Using tissues from throughout the Drosophila life cycle, we observe that Tpr is often localized to discontinuous, likely granular or particulate structures in the deep nuclear interior. These apparent granules have no obvious connectivity to pore complexes in the nuclear periphery, and are often localized on the surfaces of chromosomes and to the perinucleolar region. Most strikingly, after 1 h of heat shock, the great majority of the Tpr in the deep nuclear interior accumulates at a single heat shock puff, while Tpr in the nuclear periphery appears unchanged. This heat shock puff, 93D, is a known repository for many components of pre-mRNA metabolism during heat shock. Although we do not observe Tpr at sites of transcription under normal conditions, the 93D heat shock result leads us to favor a role for Tpr in mRNA metabolism, such as the transport of mRNA through the nuclear interior to nuclear pore complexes. Consistent with this, we observe networks of Tpr containing granules spanning between the nucleolus and the nuclear periphery which are also decorated by an anti-SR protein antibody. Since we also observe Drosophila Tpr in reticular or fibrous structures in other nuclei, such as salivary gland polytene nuclei, these results indicate that Tpr can exist in at least two structural forms, and suggest that Tpr may relocalize or even change structural forms in response to cellular needs.

Molecular complexes that contain both c-Cbl and c-Src associate with Golgi membranes

Cbl is an adaptor protein that is phosphorylated and recruited to several receptor and non-receptor tyrosine kinases upon their activation. After binding to the activated receptor, Cbl plays a key role as a kinase inhibitor and as an E3 ubiquitin ligase, thereby contributing to receptor down-regulation and internalization. In addition, Cbl translocates to intracellular vesicular compartments following receptor activation. We report here that Cbl also associates with Golgi membranes. Confocal immunofluorescence staining of Cbl in a variety of unstimulated cells, including CHO cells, revealed a prominent perinuclear colocalization of Cbl and a Golgi marker. Both the prominent Cbl staining and the Golgi marker were dispersed by brefeldin A. Subcellular fractionation of CHO cells demonstrated that about 10% of Cbl is stably associated with membranes, and that Golgi-enriched membrane fractions produced by isopycnic density centrifugation and free-flow electrophoresis are also enriched in Cbl, relative to other membrane fractions. The membrane-bound Cbl was hyperphosphorylated and it co-immunoprecipitated with endogenous Src. By immunofluorescence, some Src colocalized with Cbl and Golgi markers, and Src, like Cbl, was present in the Golgi-enriched fraction prepared by sequential density centrifugation and free-flow electrophoresis. Transfection of an activated form of Src, but not wild-type Src, increased the amount of Src that co-immunoprecipitated with Cbl, and increased the intensity of Cbl staining on the Golgi. This result, together with the increased tyrosine phosphorylation of the membrane-associated Cbl, suggests that Golgi-associated Cbl could be part of a molecular complex that contains activated Src. The localization and interaction of Src and Cbl at the Golgi and the regulation of the interaction of Cbl with Golgi membrane suggest that this complex may contribute to the regulation of Golgi function.

Highly efficient cellular uptake of c-myb antisense oligonucleotides through specifically designed polymeric nanospheres

c-myb antisense oligonucleotides (AS ODNs) were reversibly immobilized to a novel polymeric core shell nanosphere and their cellular uptake and inhibitory effect on HL60 leukemia cell proliferation studied. The nanosphere surface was so designed as to directly bind ODNs via ionic interactions and reversibly release them inside the cells. Compared with the cellular uptake of free oligonucleotide, the use of AS ODN (immobilized to the nanospheres) produced a 50-fold increase in the intracellular concentration. Specifically, a single dose of 320 nM of AS ODN immobilized to the nanospheres was capable of inhibiting HL60 cell proliferation with the same degree of efficiency obtained using a 50-fold higher dose of free AS ODN. Flow cytometric experiments with fluoresceinated ODNs showed a temperature-dependent uptake, which was detectable as early as 2 h after the beginning of treatment. The inhibitory effect on cell proliferation was maintained for up to 8 days of culture. Moreover, the level of c-Myb protein decreased by 24% after 2 days and by 60% after 4 days of treatment, thus indicating a continuous and sustained release of non-degraded AS ODN from the nanospheres inside the cells.

The neurotrophin receptor, gp75, forms a complex with the receptor tyrosine kinase TrkA

The high-affinity NGF receptor is thought to be a complex of two receptors , gp75 and the tyrosine kinase TrkA, but direct biochemical evidence for such an association had been lacking. In this report, we demonstrate the existence of such a gp75-TrkA complex by a copatching technique. Gp75 on the surface of intact cells is patched with an anti-gp75 antibody and fluorescent secondary antibody, the cells are then fixed to prevent further antibody-induced redistributions, and the distribution of TrkA is probed with and anti-TrkA antibody and fluorescent secondary antibody. We utilize a baculovirus-insect cell expression of wild-type and mutated NGF receptors. TrkA and gp75 copatch in both the absence and presence of NGF. The association is specific, since gp75 does not copatch with other tyrosine kinase receptors, including TrkB, platelet-derived growth factor receptor-beta, and Torso (Tor). To determine which domains of TrkA are required for copatching, we used a series of TrkA-Tor chimeric receptors and show that the extracellular domain of TrkA is sufficient for copatching with gp75. A chimeric receptor with TrkA transmembrane and intracellular domains show partial copatching with gp75. Deletion of the intracellular domain of gp75 decreases but does not eliminate copatching. A point mutation which inactivates the TrkA kinase has no effect on copatching, indicating that this enzymatic activity is not required for association with gp75. Hence, although interactions between the gp75 and TrkA extracellular domains are sufficient for complex formation, interactions involving other receptor domains also play a role.

Antibody-induced growth inhibition is mediated through immunochemically and functionally distinct epitopes on the extracellular domain of the c-erbB-2 (HER-2/neu) gene product p185

Over-expression of the c-erbB-2 (HER-2/neu) gene product p185 occurs in 30% of breast and ovarian cancers. The p185 protein might serve as a target for serotherapy in that antibodies against different epitopes on the extracellular domain of p185 can inhibit growth of tumor cells in the absence of cellular or humoral effector mechanisms. To define epitopes of functional relevance, 11 monoclonal antibodies (MAbs) were evaluated for their ability to bind to the extracellular domain of p185. Results of competition studies with 125I-labeled and non-labeled antibodies indicated that 10 of 11 epitopes were grouped in a linear array. Antibodies against 7 epitopes inhibited anchorage-independent growth and antibodies against 2 of these epitopes also inhibited anchorage-dependent growth of SKBr3 breast-cancer cells that over-expressed p185. Treatment with antibodies exerted cytotoxic rather than cytostatic effects. When antibodies were used in combination, additive or supra-additive inhibition of anchorage-independent and anchorage-dependent growth was observed between pairs of antibodies. Growth inhibition did not relate to the affinity of the antibody or its isotype. Two antibodies that inhibited both anchorage-dependent and anchorage-independent growth also blocked binding of the HER-2/neu ligand, whereas 5 antibodies that inhibited only anchorage-independent growth had no effect on ligand binding. Inhibition of cell growth did not correlate with internalization of p185 or down-regulation of p185 on the cell surface. Fab fragments of active antibodies could also inhibit anchorage-independent growth of SKBr3. Thus, murine MAbs and their fragments recognized both immunochemically distinct and functionally distinct epitopes on the p185 molecule. Whereas inhibition of anchorage-dependent growth correlated with the ability of antibodies to block ligand binding, inhibition of anchorage-independent growth did not correlate with effects on ligand binding, internalization, cell-surface expression or cross-linking of p185.

Solution structure of a DNA-binding unit of Myb: a helix-turn-helix-related motif with conserved tryptophans forming a hydrophobic core

The DNA-binding domain of the c-myb protooncogene product consists of three imperfect tandem repeats of 51 or 52 amino acids, each of which contains three conserved tryptophans, spaced 18 or 19 amino acids apart. The structure of the third repeat, which is essential for sequence-specific DNA binding, has been determined by NMR with distance geometry calculation. It includes three well-defined helices (residues 149-162, 166-172, and 178-187) maintained by a hydrophobic core that includes the three conserved tryptophans, together with two histidines. Helices 2 and 3 form a structure related to but distinct from a canonical helix-turn-helix motif. In particular, the turn between these helices is one amino acid longer than the corresponding turn in bacterial repressors and homeodomains and contains a proline residue. In addition, the architecture of the three helices is different from those of homeodomains and DNA-binding domains of bacterial repressors. Based on the present structure, the binding mode of Myb repeat 3 with a specific DNA is also discussed.

Integrity of FOS B leucine zipper is essential for its interaction with JUN proteins

fos B encodes a nuclear protein with 70% homology to c-fos, whose expression is transiently induced during the G0/G1 transition. Immunoprecipitation studies demonstrated that FOS B protein forms a complex in vitro with c-JUN, JUN B, and JUN D. We have mutated some of the leucines of the 'leucine zipper' present in the FOS B protein and determined their effect in the interaction with JUN proteins and their binding to an AP-1 containing sequence. The exchange of either leucine 1, 3, or 5 of the leucine repeat of FOS B to a proline dramatically inhibits its association with JUN proteins. However, a more conserved substitution to isoleucine has only a 50% inhibition. These results demonstrate that any major alteration in the alpha-helical structure of the 'leucine zipper' completely inhibits the interaction of FOS B with any of the three JUN proteins.

The src protein contains multiple domains for specific attachment to membranes

The proteins encoded by the oncogene v-src and its cellular counterpart c-src (designated generically here as pp60src) are tightly associated with both plasma membranes and intracellular membranes. This association is due in part to the amino-terminal myristylation of pp60src, but several lines of evidence suggest that amino-terminal portions of the protein itself are also involved. We now report that pp60src contains at least three domains which, in conjunction with myristylation, are capable of mediating attachment to membranes and determining subcellular localization. We identified these domains by fusing various portions of pp60src to pyruvate kinase, which is normally a cytoplasmic protein. Amino acids 1 to 14 of pp60src are sufficient to mediate both myristylation and the attachment of pyruvate kinase to cytoplasmic granules. In contrast, amino acids 38 to 111 mediate association with the plasma membrane and perinuclear membranes, whereas amino acids 204 to 259 mediate association primarily with perinuclear membranes. We conclude that pp60src contains independent domains that target the protein to distinctive subcellular locations and thus may facilitate diverse biological functions of the protein.

Structure of the guanine-nucleotide-binding domain of the Ha-ras oncogene product p21 in the triphosphate conformation

The crystal structure of the guanine-nucleotide-binding domain of p21 (amino acids 1-166) complexed to the guanosine triphosphate analogue guanosine-5'-(beta, gamma-imido)triphosphate (GppNp) has been determined at a resolution of 2.6 A. The topological order of secondary structure elements is the same as that of the guanine-nucleotide-binding domain of bacterial elongation factor EF-Tu. Many interactions between nucleotide and protein have been identified. The effects of point mutations and the conservation of amino-acid sequence in the guanine-nucleotide-binding proteins are discussed.

Changing fos oncoprotein to a jun-independent DNA binding protein with GCN4 dimerization specificity by swapping "leucine zippers"

A structural motif for DNA-binding proteins, the 'leucine zipper', has been proposed for the jun, fos and myc gene products, the yeast transcriptional activator GCN4, and the C/EBP enhancer-binding protein. These proteins all contain a region with four or five leucine residues spaced exactly seven amino acid residues apart whose sequence is consistent with the formation of an amphipathic alpha-helix. It has been proposed that the leucine zipper consists of two interdigitated alpha-helices, one from each monomer, that constitute the dimerization function necessary for high-affinity binding to DNA; an adjacent region of basic residues is thought to be responsible for specific protein-DNA contacts. In support of this model, substitution of the leucine residues within the motif can abolish dimerization and DNA-binding, and a synthetic peptide corresponding to the GCN4 leucine zipper forms alpha-helical dimers. Despite the conserved leucine residues, however, each protein has a distinct dimerization specificity. Specifically, GCN4 homodimer, Jun homodimer and Fos-Jun heterodimer proteins bind to the same DNA site, whereas Fos is unable to form homodimers, bind DNA, or interact with GCN4 (refs 8-14). Here, we alter the dimerization specificity of Fos by precisely replacing its leucine zipper with that from GCN4. This Fos-GCN4 chimaeric protein is able to bind to the target site in the absence of Jun, and can form DNA-binding heterodimers with GCN4 but not with Jun. These results indicate that the leucine zipper is sufficient to confer dimerization specificity and strongly suggest that Fos contacts DNA directly.

The mouse type IV c-abl gene product is a nuclear protein, and activation of transforming ability is associated with cytoplasmic localization

The subcellular localization of the mouse type IV c-abl protein was determined by indirect immunofluorescence of nontransformed NIH 3T3 fibroblasts that overexpress the protein. Unlike the viral transforming protein p160gag/v-abl, which has cytoplasmic and plasma membrane localization, a large fraction of the c-abl (IV) protein is nuclear, with the remainder in the cytoplasm and plasma membrane. Deletion of a small N-terminal regulatory region of the c-abl (IV) protein, sufficient to activate its transforming potential fully, changes the distribution of the protein from the nucleus to the cytoplasm. Mapping of an amino acid sequence responsible for the nuclear localization of the c-abl (IV) protein reveals a nuclear localization signal similar to that of SV40 large T antigen.

Conformation change of effector-region residues in antiparallel beta-sheet of human c-Ha-ras protein on GDP----GTP gamma S exchange: a two-dimensional NMR study

The conformations of a truncated human c-Ha-ras gene product [ras(1-171) protein] in the GDP-bound form and in the GTP gamma S-bound form were compared by two-dimensional nuclear Overhauser effect spectroscopy (NOESY). As for the GDP-bound ras(1-171) protein, three NOESY cross peaks were observed in the region of 4.5-6.0 ppm, indicating a regular antiparallel beta-sheet structure. On the ligand exchange from GDP to GTP gamma S, one of the three NOESY cross peaks disappeared and the other two cross peaks were appreciably shifted. By analysis of the effects of specific deuteration of leucine residues and the homonuclear Hartmann-Hahn spectroscopy, the antiparallel beta-sheet was found to consist of residues 38-44 and residues 51-57. The conformations around Ser-39 and Leu-56 are of the regular antiparallel beta-strand type in the GDP-bound state, and largely distorted in the GTP gamma S-bound state, which is probably related to the conformational activation of the effector region of ras proteins by ligand exchange from GDP to GTP gamma S.

Conformational changes occurring in N-ras p21 in response to binding of guanine nucleotide and metal ions probed by proteolysis performed under controlled conditions

Variations in susceptibility to proteolysis by trypsin and chymotrypsin have been used as indicators of conformational changes taking place in N-ras p21 in response to ligand binding. It has been observed that changes occur in undenatured protein, rendering it more resistant to degradation, in the presence of divalent cations such as Mg2+ and Ca2+ (suggesting direct binding of metals to the polypeptide) and even more markedly in the presence of GDP and/or Mg2+ GDP. Monovalent cations (Na+ or K+) cannot substitute for Mg2+ or Ca2+. Some capacity to bind guanine nucleotide is also retained by p21 treated with 7 M urea, as evidenced by increased resistance to proteolytic degradation, but the ability to bind divalent cations is irreversibly lost following denaturation. Protein prepared under denaturing conditions from a eukaryotic source, however, never regains the resistance to proteolysis shown by the bacterial p21 indicating irreversible changes in secondary and tertiary structure produced under these conditions.

Inactivation of the N-myc gene product by single amino acid substitution of leucine residues located in the leucine-zipper region

To verify the importance of the hypothetical leucine-zipper structure in the N-myc protein, a series of mutants of the mouse N-myc gene were constructed, in which codons for the first and second leucine residues within this structure were systematically replaced by other amino acids. The expression plasmids which contained the mutated and wild type N-myc genes were cotransfected into rat embryo cells with activated c-Ha-ras gene and their transforming abilities were compared. It was shown that single amino acid substitutions in the leucine-zipper region inactivate the transforming ability of the N-myc gene product. In particular, proline, which is known to disrupt an alpha-helical structure, completely inactivated the transforming activity even when it was substituted for another amino acid located between these two leucine residues. Among several amino acid species used for substitution of the leucine residues, only methionine was able to retain the transforming activities in both the first and second leucine positions, although the activity was reduced as compared with wild-type N-myc gene product. It also appeared that the integrity of the first leucine is more important than the second leucine. Our results provide experimental evidence for the physiological importance of the hypothetical leucine-zipper structure.

Localization of H-ras mRNA in oral squamous cell carcinomas

Expression of the H-ras oncogene was examined in 5 SCC of the oral cavity. The presence of oncogene mRNA was detected and localized in the tumor tissue by in situ hybridization. Tumor tissues were also examined for the presence of the oncogene product, p21, by immunohistochemical techniques. H-ras mRNA and p21 were detected in all 5 tumors in cells that were also positive for both keratin mRNA and keratin protein. The distribution of the mRNA was not uniform throughout the tumor tissue. Distinct spatial localizations of the H-ras mRNA were present in regions of the tumors with both higher proliferative and invasive potential. The results suggest that the distribution of the oncogene mRNA may be related to the pattern of development and progression of oral SCC.

The structure of the amino terminal transforming segment of the p21 protein, Tyr4-Thr20 (with Asp12), by two-dimensional NMR

The structure of a peptide from the transforming region (residues 4-20) of the p21 protein has been determined using two-dimensional NMR. In the normal protein, this segment contains a Gly residue at the critical 12 position; any substitution, other than Pro, at this position results in a transforming protein. Previously performed energy calculations indicated that this peptide segment is a structured one. In this study we find that the Asp12 containing peptide has a surprisingly well-defined structure in solution which has more similarity to the GDP-binding loop region in EF-tu than to that in p21.

c-Jun dimerizes with itself and with c-Fos, forming complexes of different DNA binding affinities

The c-Jun and c-fos proto-oncogenes encode proteins that form a complex which regulates transcription from promoters containing AP-1 activation elements. c-Jun has specific DNA binding activity, while c-Fos has homology to the putative DNA binding domain of c-Jun. Following in vitro translation, c-Jun binds as a homodimer to the AP-1 DNA site, while c-Fos fails to dimerize and displays no apparent affinity for the AP-1 element. Cotranslated c-Jun and c-Fos proteins bind 25 times more efficiently to the AP-1 DNA site as a heterodimer than does the c-Jun homodimer. These experiments suggest that in growth factor-stimulated cells c-Jun binds DNA as a dimer with c-Fos as its natural partner. However, overexpression of c-Jun protein in the absence of c-Fos may result in formation of aberrant homodimeric transcription complexes, which could abrogate the normal mechanisms controlling gene expression.