C/EBPα:C/EBPα and C/EBPα:Jun Leucine Zipper Complexes Are Detectable in Myeloid Cells Via Binding to Distinct DNA Elements.

Abstract 1458

Poster Board I-481

Mice lacking C/EBPα have reduced granulocyte-monocyte progenitors, and C/EBPα plays a role in the maturation of GMP along the granulocyte or monocyte lineages. C/EBPα is a member of the bZIP family of transcription factors, dimerizing with itself or other C/EBP proteins via its leucine zipper (LZ) domain to bind palindromic DNA elements (e.g. ATTGCGCAAT, designated αα sites) via the adjacent basic region (BR). AP-1 proteins such as c-Fos and c-Jun represent a distinct bZIP subfamily that heterodimerize amongst themselves to bind AP-1 sites (e.g. TGACTCA, designated JJ sites) to also mediate monocytic differentiation in hematopoietic cell lines or transduced marrow cells. In addition, we previously demonstrated that C/EBPα or C/EBPβ can zipper with c-Jun, JunB, or c-Fos (Cai et al 2008). To accomplish this we employed hybrid proteins in which, for example, the c-Jun LZ was swapped in place of the C/EBPα LZ and binding to a C/EBP site was then assessed after expression in 293T cells. We also provided data suggesting that C/EBPα:c-Jun heterodimers bind preferentially to hybrid αJ sites (e.g. TGACGCAAT). To accomplish this we forced heterodimerization by replacing the LZ in C/EBPα with an acidic zipper containing multiple glutamic acid residues (LZE) and replaced the LZ in c-Jun with a basic zipper containing multiple lysines (LZK). We have now extended these initial findings in two ways. First, we demonstrate that the intact proteins, not only those containing artificial acid or basic leucine zippers, maintain binding site specificity. In particular, when expressed in 293T cells or when generated by coupled in vitro transcription/translation, C/EBPα alone preferentially binds an αα site whereas C/EBPα co-expressed with c-Jun prefers the αJ site. Neither C/EBPα alone nor C/EBPα:c-Jun bound a JJ site. Consistent with the weak affinity of C/EBPα for c-Jun, formation of sufficient C/EBPα:c-Jun heterodimers to allow detection of preferential binding to the αJ site required a 4:1 c-Jun:C/EBPα ratio, with input proteins normalized via a shared N-terminal myc tag. Super-shift assays confirmed that a single gel shift band contains both proteins under these conditions. Co-expression of c-Jun with c-Fos yielded complexes that bound JJ sites, but interestingly c-Jun:c-Fos bound αJ sites with even greater affinity. Second, we have used biotinylated αα and αJ oligonucleotides to demonstrate the presence of both endogenous C/EBPα:C/EBPα and C/EBPα:c-Jun complexes in myeloid cell lines. When 32Dcl3 cells are placed in G-CSF for 24 hrs to stimulate granulopoiesis, Western blotting indicates that C/EBPα levels are induced while c-Jun levels remain constant, and binding of C/EBPα to the αα biotin-oligo increases, likely reflecting the presence of endogenous C:EBPα homodimers, as no interaction of c-Jun with the αα probe was detected. Both endogenous C/EBPα and c-Jun interact the αJ biotin-oligo, suggesting binding by an endogenous C/EBPα:c-Jun complex, and binding of each of these proteins to this probe also increases upon G-CSF induction of C/EBPα. Further support for the presence of an endogenous C/EBPα:c-Jun complex comes from the observation that interaction of C/EBPα with the αJ oligo is several-fold stronger than with the αα oligo, before or after G-CSF addition, whereas gel shift assay results described above indicate that C/EBPα homodimers prefer the αα probe. We previously found that C/EBPαLZK:C/EBPαLZE homodimers favor granulopoiesis whereas C/EBPαLZK:c-JunLZE heterodimers favor monopoiesis of marrow progenitors (Cai et al 2008). In summary, new data indicate that when c-Jun is present in excess, co-expressed C/EBPα and c-Jun preferentially binds hybrid αJ site, and C/EBPα:c-Jun complexes are detected endogenously in myeloid cells. These new findings provide biochemical support for the idea that C/EBPα:AP-1 heterodimers bind novel DNA elements to help mediate monopoiesis, with C/EBPα:C/EBPα homodimers potentially favoring granulopoiesis. Our future focus will be to further define the cellular levels of endogenous C/EBP, AP-1, and C/EBP:AP-1 complexes in immature and mature monocytic and granulocytic cells and to elucidate the spectrum of genes these complexes bind and regulate during myelopoiesis.