Acetylation of a Conserved Lysine Residue within Activation Domain One of E2A Transcription Factors Plays a Role in Transcriptional Regulation.

The E2A locus encodes transcription factors, called E12 and E47, involved in lineage-specific cellular differentiation. The locus is also involved in chromosomal translocations associated with acute lymphoblastic leukemia, the most common of which results in expression of the oncoprotein E2A-PBX1. We showed recently that direct interaction between E2A and the histone acetyl-transferase (AT) and transcriptional co-activator proteins CBP/p300 is required in leukemogenesis by E2A-PBX1. E2A proteins have also been shown to interact with another AT/co-activator, p/CAF. Interaction with these AT proteins results in acetylation of E2A proteins themselves. Here we map the acetylated lysine residues within the oncogenic portion of E2A proteins and begin to elucidate some functional correlates of E2A acetylation. Our results indicate that the isolated AT domain of p/CAF as well as full-length p/CAF were capable of acetylating E2A. Interestingly, full-length p300 was capable of acetylating E2A while an isolated AT domain was unable to acetylate E2A, suggesting that additional domains of CBP/p300 are required to mediate E2A acetylation. We demonstrate that both p300 and p/CAF can interact directly with E2A, independent of the known interaction between p300 and p/CAF. These co-activators do, however, appear to co-operate to achieve maximal E2A acetylation. Mutagenesis-based mapping studies indicate that several lysines are substrates for acetylation. Of particular interest, a conserved lysine residue (K34) located within the N-terminal transcriptional activation domain (AD1) is acetylated in vitro by p/CAF. K34 is located within a GKXXP consensus sequence, suggested to be a recognition motif for acetylation by several AT enzymes including CBP/p300 and p/CAF. Conservative replacement of K34 with arginine (i.e., K43R) substantially impairs transcriptional activation of a luciferase reporter gene by E2A, suggesting that post-translational modification of this residue may play an important functional role. Consistent with a role for acetylation, relative to some other lysine-dependent modification, we were unable to demonstrate sumoylation or ubiquitination of the N-terminus of E2A. Therefore, we have found that acetylation by AT/co-activator proteins contributes to transcriptional regulation by the functionally critical N-terminal activation domain (AD1) of E2A proteins. The mechanisms by which this acetylation event is regulated and how it contributes to target gene induction by E2A are not clear. It seems plausible that the acetylation status of AD1 could be determined by upstream signaling events and acetylation of E2A could modulate interactions with transcriptional co-regulators, DNA or chromatin. Further studies to investigate these possibilities are underway. In particular, results using an amino acid substitution that mimics acetylation of AD1 (K34Q) will be presented.