DPF2

DPF2 (BAF45d) is a histone-reader subunit of the BAF (SWI/SNF) chromatin-remodeling complex that couples recognition of modified histone tails to the regulation of cell-type-specific transcriptional programs (PMID:28533407, PMID:30609396). Its tandem PHD finger module directly binds acetylated histone H3 and H4 tails through bipartite binding pockets, and structure-guided mutation of these pockets abolishes DPF2 chromatin recruitment and its ability to block myeloid differentiation (PMID:28533407). This reader activity extends beyond acetylation to other acyl marks: DPF2 engages crotonylated H3K14 through its PHD1/PHD2 pockets while embedded in the intact BAF complex (PMID:39184054), and binds H3K14 lactylation to drive transcription of oncogenic genes, where disrupting the DPF2–H3K14la interaction blunts cancer-gene expression and cell survival (PMID:39636855). In embryonic stem cells DPF2 co-occupies enhancers with OCT4, SOX2, p300, and BRG1, and is required for Tbx3 expression via H3K27ac at its distal enhancer, antagonizing the PRC2 subunit EED to enable mesendodermal differentiation (PMID:30609396); an analogous enhancer-licensing role operates in macrophages, where DPF2 sustains H3K27ac/H3K4me1 at the Cacna1d enhancer to promote calcium-dependent MAPK signaling and inflammatory polarization (PMID:41223220). DPF2 also acts as a transcriptional co-repressor for ERRα in association with HDAC1 (PMID:20400511), promotes K48-linked polyubiquitination and proteasomal degradation of OCT4 in a manner requiring its PHD2 domain (PMID:26417682), and functions in the non-canonical NF-κB pathway to suppress type I interferon induction during influenza infection (PMID:28404846). Heterozygous missense variants in the PHD1/PHD2 domains cause Coffin-Siris syndrome, impairing histone-tail binding and acting through a dominant-negative mechanism that sequesters wild-type DPF2 and BRG1 (PMID:29429572).

1. 2010 Medium

   Established that DPF2 is not merely a structural subunit but a transcriptional regulator, acting as a co-repressor that links a nuclear receptor to histone deacetylase activity.

   Evidence Biochemical purification, Co-IP/GST pulldown, ChIP, and siRNA knockdown in myoblasts

   PMID:20400511

   Open questions at the time

   • Did not resolve whether repression depends on PHD-domain histone reading versus HDAC1 recruitment
   • Generality beyond ERRα targets untested
2. 2011 Medium

   Defined the structure of the DPF2 C2H2 zinc finger, raising the possibility of a DNA-contacting surface in addition to histone reading.

   Evidence X-ray crystal structure of the C2H2 domain

   PMID:21888896

   Open questions at the time

   • DNA-binding capacity inferred from surface charge only, not experimentally validated
   • Functional relevance of the proposed DNA contacts unknown
3. 2015 Medium

   Revealed a non-chromatin enzymatic role: DPF2 directs K48-linked ubiquitination and proteasomal degradation of the pluripotency factor OCT4, connecting it to control of the pluripotency network.

   Evidence Co-IP, GST pulldown, in vitro ubiquitination, knockdown/overexpression in 293 cells

   PMID:26417682

   Open questions at the time

   • Whether DPF2 is the direct E3 ligase or a recruiting adaptor not resolved
   • PHD2 requirement for ubiquitination mechanistically unexplained
4. 2015 Medium

   Placed DPF2 within a functionally required BAF subunit set for MLL-rearranged leukaemia, framing it as a dependency in oncogenic transcriptional maintenance.

   Evidence shRNA knockdown with gene expression profiling and in vitro/in vivo leukaemia models

   PMID:26571505

   Open questions at the time

   • DPF2-specific contribution partially conflated with SMARCA4 and SMARCD2
   • Direct target genes of DPF2 in this context not isolated
5. 2017 High

   Provided the structural basis for DPF2 as a dual-acetyl reader, showing the tandem PHD module binds acetylated H3 and H4 tails and that this reading is functionally required to block myeloid differentiation.

   Evidence 1.6-Å crystal structure, histone peptide binding, targeted mutagenesis, in vivo myeloid differentiation assay

   PMID:28533407

   Open questions at the time

   • Did not address acyl marks beyond acetylation
   • Did not map genome-wide targets controlled by this reading activity
6. 2017 Medium

   Identified a signaling role beyond chromatin, showing DPF2 in the non-canonical NF-κB pathway restrains type I interferon antiviral responses.

   Evidence RNAi screen, siRNA knockdown, viral growth kinetics, IFN-β/STAT1 readouts in influenza-infected cells

   PMID:28404846

   Open questions at the time

   • Molecular link between DPF2 and NF-κB components undefined
   • Unclear whether this requires BAF complex or histone-reading activity
7. 2018 Medium

   Established DPF2 as a Coffin-Siris syndrome gene and connected disease variants mechanistically to loss of histone-tail binding and dominant-negative sequestration of BAF.

   Evidence Recombinant pull-down with histone peptides, overexpression and immunofluorescence in HEK293/COS7

   PMID:29429572

   Open questions at the time

   • Dominant-negative aggregate behavior shown by overexpression, not endogenous variants
   • Tissue-specific developmental consequences not modeled
8. 2018 Medium

   Linked DPF2 to an RNA-splicing feedback loop with PTBP1, expanding its regulatory reach from transcription into splicing in glioblastoma.

   Evidence Splicing array and functional siRNA/overexpression studies in glioblastoma samples and cell lines

   PMID:29373718

   Open questions at the time

   • Biochemical basis of the DPF2–PTBP1 interaction not detailed
   • Direct versus indirect reciprocal regulation unresolved
9. 2019 High

   Positioned DPF2 in the ESC enhancer network, showing it co-occupies pluripotency enhancers and licenses Tbx3 expression by antagonizing PRC2 to enable mesendodermal differentiation.

   Evidence ChIP-seq, ATAC-seq, Dpf2 KO ESCs with Tbx3 rescue and Dpf2/Eed epistasis

   PMID:30609396

   Open questions at the time

   • How DPF2 reading translates into H3K27ac deposition at the enhancer not fully resolved
   • Mechanism of antagonism with EED at the intragenic enhancer not detailed
10. 2024 Medium

    Extended DPF2 reading specificity to crotonylation, demonstrating PHD-mediated engagement of H3K14Cr is preserved within the assembled BAF complex.

    Evidence Hydroxyl radical protein footprinting MS with binding assays on isolated DPF domain and full BAF complex

    PMID:39184054

    Open questions at the time

    • Functional/transcriptional consequence of H3K14Cr reading not tested
    • Single-lab footprinting method not independently replicated
11. 2024 High

    Defined DPF2 as a histone lactylation reader and tied this directly to oncogenic transcription, showing disruption of the H3K14la interaction impairs cancer-gene expression and cell survival.

    Evidence Photoaffinity proteomics, biochemical binding, CUT&Tag, structure-guided mutagenesis

    PMID:39636855

    Open questions at the time

    • Identity of the full oncogenic gene set governed by H3K14la reading incomplete
    • In vivo tumor dependence not established
12. 2024 Low

    Mapped a developmental-stage shift in DPF2 chromatin targeting, showing preferential binding of anterior/trunk HOX loci, NES, and TBX6 in spinal cord neural stem cells.

    Evidence Chromatin binding profiling (CUT&RUN-type) comparing ESCs and spinal cord NSCs

    PMID:38168763

    Open questions at the time

    • Descriptive binding data without functional perturbation establishing causal role
    • Single lab, no rescue or knockout validation
13. 2025 High

    Demonstrated an in vivo enhancer-licensing role in immune cells, where DPF2 sustains Cacna1d enhancer marks to drive calcium-dependent MAPK signaling and inflammatory macrophage polarization.

    Evidence Mouse KO, scRNA-seq, spatial transcriptomics, ChIP for histone marks, MAPK assays, patient organoids

    PMID:41223220

    Open questions at the time

    • Direct DPF2 reading of a specific acyl mark at the Cacna1d enhancer not pinpointed
    • Whether MAPK attenuation is solely Cacna1d-dependent untested

• How DPF2's distinct activities — multivalent histone acyl reading, BAF recruitment, ERRα co-repression, OCT4 ubiquitination, and NF-κB signaling — are integrated and switched between contexts remains unresolved.
• No unifying model linking chromatin-reading to the ubiquitin-ligase activity
• Determinants of cell-type-specific target selection (HOX vs. enhancers vs. oncogenes) unknown
• Whether NF-κB and splicing roles depend on BAF complex membership untested