Post-translational C/EBP modifications and epigenetic functions
Several years ago, we found that C/EBPα,β may instruct even non-hematopoietic cells, such as skin fibroblasts, to express myeloid genes. Others have shown that C/EBPs may reprogram lymphocytes into myeloid cells. Accordingly, C/EBPs entail epigenetic competence and gene regulatory functions to determine cell fate. We had also shown that cellular signaling cascades regulate the activity of C/EBPβ to convert extracellular information into gene regulation. C/EBPβ is a ras/ MAP kinase signal sensitive transcription factor that regulates genes involved in metabolism, proliferation, differentiation,immunity, senes cence, and tumorigenesis. The functional capacity of C/EBPβ is governed by proteininteractions that depend on post-translational C/EBPβ modifications. In a proteome-wide interaction screen, the histone-lysine N-methyl trans ferase, H3 lysine-9 specific 3 (G9a) was found to directly inter act with the C/EBPβ transactivation domain (TAD). G9a, but not a defective G9a mutant abrogated the transactivation potential of wild type C/EBPβ. Metabolic labeling showed that C/EBPβ is post-translationally modified by methylation of its TAD. A conserved lysine residue (K39) in the C/EBPβ-TAD served as a substrate for G9a mediated methylation. A C/EBPβ K39 alanine exchange mutant was resistant to G9a mediated inhibition and conferred super-activation of myeloid genes. These data identified C/EBPβ as a direct substrate of G9a that alters the functional properties of C/EBPβ by posttrans lational lysine methylation.Mass spectrometry of cell derived C/EBPβ (in collaboration with Gunnar Dittmar, MDC) revealed extensive methylation of N-terminal arginine residues in C/EBPβ. The protein arginine methyl-transferase 4 (PRMT4) was found to interact with C/EBPβ and to di-methylate the conserved arginine residue (R3) in the C/EBPβ TAD. Phosphorylation of the regulatory domain of C/EBPβ by ras/ MAP kinase signaling abrogated the inter action between C/EBPβ and PRMT4. Differential proteomic screening with R3-methylated and un-methylated C/EBPβ peptides, protein interaction studies, and mutationalanalysis revealed that methylation of R3 constrained the interaction between C/EBPβ with SWI/SNFand Mediator complexes. Both complexes play essential roles in chromatin remodeling and transcription initiation by polymerase II and were previously implicated in C/EBPβ functions. Mutation of the R3-C/EBPβ methylation site alters the ability of C/EBPβ to induce myeloid and adipogenic differentiation. Thus, phospho ry lation of the transcription factor C/EBPβ couples ras signaling to arginine methylation and regulates the interaction of C/EBPβ with epigenetic gene regulatory protein complexes during cell differentiation.
A number of implications and conceptual advances are contained in these results. An “indexing code” of post-translational transcription factor modifications has recently been suggested (Sims & Reinberg, 2008), although experimental evidence is currently still scarce. Covalent modifications by “writers” (here, R-methylation by PRMT4, and K-methylation by G9a) determine modification dependent docking of “readers” (SWI/SNF; Mediator; G9a). Our results imply that writing/reading such an indexing code is downstream of receptor tyrosine kinase signaling, relaying signals to epigenetic events that finally determine cell fate. This concept is an important extension of the Histone Code hypothesis to non-histone transcription factor proteins, which come first in gene regulation and before chromatin modifications occur. Many mechanistic (gene regulation in development and disease) and medical implications (pharmacology) are implied.