KDM7A

KDM7A is a dual-specificity JmjC-domain histone demethylase that removes mono- and di-methyl marks from H3K9 and H3K27, thereby relieving transcriptional repression at target gene promoters to regulate neural differentiation, mesenchymal lineage commitment, bone homeostasis, hepatic lipid metabolism, and neuronal immediate-early gene expression (PMID:20084082, PMID:30614617, PMID:38346941, PMID:37565374). KDM7A directly binds promoters of context-specific target genes—including FGF4 in neural induction, C/EBPα and Sfrp1 in adipogenesis, DGAT2 in hepatocytes, FAP and RANKL in osteoprogenitor cells, and Fscn1 in reward-circuit neurons—and its catalytic activity is required for transcriptional activation, as demonstrated by catalytic-dead mutant controls (PMID:20084082, PMID:30614617, PMID:34681759, PMID:38346941, PMID:39836528). Beyond canonical epigenetic functions, KDM7A stabilizes ICAM1 protein by suppressing TFEB-mediated lysosomal degradation independently of its demethylase activity on chromatin (PMID:27565733). KDM7A is transcriptionally regulated by TGF-β signaling through SMAD2/SMAD4 binding at its own promoter and by nuclear miR-451a acting at an enhancer in an AGO2-dependent manner (PMID:34249916, PMID:38943031).

1. 2010 High

   Establishing that KDM7A possesses intrinsic dual-specificity demethylase activity toward H3K9me2 and H3K27me2 resolved the enzymatic identity of this JmjC-domain protein and linked it to neural differentiation via FGF4 transcriptional activation.

   Evidence In vitro demethylase assay with catalytic-dead mutant control, mouse ESC knockdown/overexpression, and chick embryo electroporation with FGF4 epistasis rescue

   PMID:20084082 PMID:20981832

   Open questions at the time

   • Structural basis for dual H3K9/H3K27 substrate recognition not determined
   • Whether H3K9me1 and H3K27me1 are also substrates was not systematically tested in vitro at this stage
   • Upstream signals controlling KDM7A during neural induction were unknown
2. 2011 Medium

   Demonstrating that KDM7A suppresses tumor angiogenesis under nutrient stress expanded its biology beyond development, though the direct chromatin targets mediating angiogenic factor downregulation remained undefined.

   Evidence Stable overexpression/siRNA in cancer cells, xenograft assays with immunohistochemistry for CD31 and CD11b

   PMID:22143793

   Open questions at the time

   • No ChIP evidence identifying direct promoter targets of KDM7A at angiogenic factor loci
   • Catalytic-dead mutant not tested, so demethylase-dependence not confirmed
   • Single lab without independent replication
3. 2016 Medium

   Discovery that KDM7A regulates ICAM1 protein stability through lysosomal degradation (TFEB pathway) without affecting ICAM1 mRNA revealed a non-canonical, non-epigenetic function for this demethylase.

   Evidence siRNA knockdown in brain microvascular endothelial cells, cycloheximide chase, lysosome/proteasome inhibitors

   PMID:27565733

   Open questions at the time

   • Whether this non-epigenetic function requires demethylase catalytic activity is unknown
   • Direct physical interaction between KDM7A and TFEB or lysosomal machinery not shown
   • Not independently replicated
4. 2018 Medium

   Showing that KDM7A is recruited to androgen receptor target-gene promoters upon androgen stimulation to remove H3K27me2 established KDM7A as a hormone-responsive transcriptional coactivator in prostate cancer.

   Evidence shRNA knockdown in LNCaP cells, ChIP for KDM7A and H3K27me2 at AR target promoters, xenograft growth

   PMID:30183076

   Open questions at the time

   • Direct physical interaction between KDM7A and AR not demonstrated
   • Whether H3K9me2 demethylation also occurs at AR target loci was not addressed
   • Single lab
5. 2019 High

   ChIP-based identification of C/EBPα and Sfrp1 promoters as direct KDM7A targets—together with catalytic-dead mutant failure to drive adipogenesis—established that enzymatic H3K9me2/H3K27me2 demethylation controls mesenchymal lineage commitment between adipogenic and osteogenic fates.

   Evidence ChIP, overexpression/knockdown with catalytic-dead mutant in ST2 and primary marrow stromal cells, differentiation assays

   PMID:30614617

   Open questions at the time

   • In vivo bone/fat phenotype from conditional knockout not yet reported at this time
   • Whether KDM7A acts downstream of specific signaling cues in marrow was not identified
6. 2019 Medium

   KDM7A knockdown in porcine embryos increased global H3K9me1/2 and H3K27me1/2 and impaired blastocyst formation, establishing that KDM7A-dependent demethylation is necessary for early embryonic development and first lineage specification.

   Evidence mRNA knockdown in IVF/PA/SCNT porcine embryos, immunofluorescence, RT-PCR for pluripotency genes

   PMID:31216927

   Open questions at the time

   • Specific promoter targets in early embryos not identified by ChIP
   • Rescue with wild-type or catalytic-dead KDM7A not performed
   • Porcine system; relevance to human embryogenesis not directly tested
7. 2021 High

   Identification of MKL1 as a physical partner that recruits KDM7A to the RHOJ promoter, and of SMAD2/SMAD4 as direct transcriptional activators of KDM7A itself, placed KDM7A within TGF-β signaling and explained how its expression and chromatin targeting are regulated during cancer cell invasion.

   Evidence Co-immunoprecipitation, ChIP at RHOJ and KDM7A promoters, knockdown, xenograft assays

   PMID:34249916

   Open questions at the time

   • Whether MKL1-KDM7A interaction is direct or bridged was not resolved
   • Genome-wide scope of MKL1-KDM7A co-occupancy not mapped
8. 2021 Medium

   ChIP evidence that KDM7A removes H3K9me2/H3K27me2 at the DGAT2 promoter to drive triglyceride accumulation linked the demethylase to hepatic lipid metabolism and steatosis in vivo.

   Evidence Overexpression/knockdown in AML12 cells, ChIP at DGAT2 promoter, adenoviral overexpression in mouse liver

   PMID:34681759

   Open questions at the time

   • Catalytic-dead mutant not tested in the hepatocyte system
   • Conditional hepatocyte knockout not generated
   • Upstream signals inducing KDM7A in steatotic conditions not identified
9. 2023 High

   Genome-wide CUT&Tag and RNA-seq in neurons combined with in vivo AAV knockdown showed that KDM7A maintains H3K9me1/2 and H3K27me1/2 at TSS regions and is essential for immediate-early gene expression, neuronal activity, and memory, establishing a CNS-specific functional role.

   Evidence CUT&Tag-seq, RNA-seq in N2A cells, AAV-mediated hippocampal knockdown in mice, behavioral assays

   PMID:37565374

   Open questions at the time

   • Direct promoter targets among IEGs not validated by individual ChIP
   • Cell-type specificity within hippocampal subpopulations not resolved
   • Whether catalytic activity is required for IEG regulation was not tested with mutant
10. 2024 High

    Conditional deletion of Kdm7a in osterix+ osteoprogenitors increased bone mass and protected against ovariectomy-induced bone loss by reducing FAP and RANKL at the promoter level, providing definitive in vivo genetic evidence for KDM7A in bone homeostasis.

    Evidence Osterix-Cre conditional knockout, ChIP at Fap/Rankl promoters, bone histomorphometry, recombinant FAP rescue, co-culture assays

    PMID:38346941

    Open questions at the time

    • Therapeutic targeting of KDM7A in osteoporosis not tested with specific inhibitors
    • Whether KDM7A also regulates osteoclast-intrinsic programs is unknown
11. 2024 High

    Identification of Fscn1 as a direct KDM7A target in mPFC neurons linked the demethylase to dendritic spine remodeling and reward memory consolidation, extending its neuronal role beyond hippocampal IEGs.

    Evidence AAV knockdown in mouse mPFC, Nanopore RNA-seq, ChIP at Fscn1 promoter, dendritic spine imaging, conditioned place preference

    PMID:39836528

    Open questions at the time

    • Whether KDM7A-Fscn1 axis generalizes to other forms of memory not tested
    • Catalytic-dead rescue not performed in this brain region
12. 2024 Medium

    Nuclear miR-451a was found to activate KDM7A transcription via enhancer interaction in an AGO2-dependent manner, revealing a non-canonical RNA-mediated mechanism of KDM7A gene regulation beyond the previously known SMAD-dependent pathway.

    Evidence ChIRP, small RNA sequencing, FISH, siRNA knockdown in HNSCC cells

    PMID:38943031

    Open questions at the time

    • Whether miR-451a/AGO2 mechanism operates in non-cancer contexts is unknown
    • Direct causal link between KDM7A upregulation and cetuximab resistance not fully dissected at the epigenetic level

• Key unresolved questions include the structural basis for KDM7A's dual H3K9/H3K27 substrate specificity, the full repertoire of its non-epigenetic functions, and whether selective KDM7A inhibitors can be developed for therapeutic applications in bone loss, neurological disorders, or cancer.
• No crystal structure of KDM7A in complex with nucleosomal substrates
• Mechanism of the non-canonical lysosomal/TFEB-related function remains unresolved
• Genome-wide mapping of KDM7A chromatin occupancy across tissues is lacking
• No selective KDM7A inhibitor with validated in vivo pharmacokinetics