MLLT3 (AF9) and its homolog MLLT1 (ENL) were initially identified as chromosome translocation partners of the MLL (KMT2A) gene observed in Mixed Lineage Leukemia (MLL). The amino termini of MLLT3, and MLLT1 proteins contain a nearly identical chromatin-binding YEATS domain which preferentially binds crotonylated histone sites (Kcr). This distinguishes YEATS domains as crotonylation reader modules in contrast to other acetylation reader modules, such as bromodomains. The MLLT3 YEATS domain directly links histone Kcr readout to active gene transcription, but mechanisms underlying specific recruitment to direct target genes are not understood. Work from different laboratories, including ours, has revealed roles of MLLT3 and MLLT1 in at least four different complexes with critical gene regulatory functions based on direct binding to the C-terminal ANC1 homology domain (AHD). The canonical functions of two of these complexes (AF4-containing Super Elongation Complex; DOT1L) are to activate gene transcription whereas the other two (CBX8, BCOR) most often function in gene repression. The factors that decide which of these four different complexes are recruited, and whether recruitment of one complex facilitates or inhibits recruitment of another are not understood.
Aim 1 : Functional effects of CBX8 and BCOR recruitment on MLL-MLLT3/1 (MLL-AF9/-ENL) function. We have determined 3D structures of MLLT3 AHD-CBX8 and AHD-BCOR complexes and used the structural information to develop point mutations to selectively disrupt recruitment of CBX8 and BCOR. These will be used to specifically delineate the role of direct recruitment of CBX8 and BCOR to MLL-MLLT3 and MLL-MLLT1 in altering gene expression and driving leukemia, as we have done previously for the AF4 and DOT1L interactions.
Aim 2 : MLLT3 (AF9) YEATS domain is a dual reader of H3K9 (and K18, K27) crotonylation and RNA. We have used a biochemical approach to show that the MLLT3 YEATS domain also binds to RNA, in addition to specific binding to crotonylated H3, indicating this domain is a dual reader of both epigenetic marks and RNA. We are proposing to fully characterize the role of the RNA binding of this domain in MLLT3 function. This includes delineation of the RNA binding specificity, structural studies of a YEATS domain-H3K9cro-RNA ternary complex, and development of point mutations which can selectively disrupt RNA binding and H3 peptide binding to probe the functional role of these interactions. Similar studies will be carried out with MLLT1.
Aim 3 : MLLT3 (AF9) and MLLT1 (ENL) have non-redundant roles in hematooietic stem and progenitor cell (HSPC) gene regulation which require their YEATS domain and C-terminal AHD functions. Using wildtype and point mutant forms of MLLT3 and MLLT1 which can selectively disrupt either histone or RNA binding, we will probe the functional role of the H3Kcr and RNA interactions via ChIP-seq, RNA-seq, and effects on in vitro and in vivo HSPC functions. Wildtype and mutant MLLT3 and MLLT1 that specifically disrupt binding to AF4, DOT1L, BCOR, and CBX8 will probe the roles of these interactions on gene expression and HSPC functions.
AF9 is an important protein for the development of blood cells. In addition, a portion of AF9 is fused to the protein MLL in a type of leukemia that has a very poor prognosis. We will study the roles of functional portions of AF9 in normal blood development and of the AF9 portion of the MLL-AF9 fusion protein in causing leukemia. Our efforts will increase understanding of the disease and may identify novel avenues for drug development.
