MBD1 - one protein with a methyl-CpG reader and three CXXC fingers, only the third of which binds unmethylated CpG, and whose CXXC fails the MLL-AF9 swap

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

• MBD1 is the one family member a CXXC-directed molecule might genuinely spare, since its isolated CXXC did not bind DNA in the one assay that put it head to head with MLL's. That is a single assay and should be re-measured before it is used as a selectivity argument.
• Blocking unmethylated-CpG reading genome-wide would break the CXXC3-dependent arm of MBD1 and leave the MBD arm intact, which the optochemical study says would suppress TET1 oxidation kinetics - i.e. push methylation up. That is the same direction that silenced Hoxa9 in the CXXC-swap experiment, so an effect intended as anti-leukaemic and an effect known to be anti-leukaemic would be confounded.

1. The MBD1-TET1 cycle - CXXC3 recruits TET1, TET1 changes the DNA, the change evicts MBD1 - is the only reciprocal occupancy mechanism documented inside this family. Does the KMT2A fusion participate in any equivalent cycle, or is it inert to the chemistry it reads?
2. Which MBD1 isoform is expressed in KMT2A-rearranged blasts? The CXXC3-containing and CXXC3-lacking isoforms have opposite effects on TET1, and twelve isoforms are annotated.
3. MBD1-L arises from reduced WTAP rather than a splicing-factor mutation. Is WTAP expression altered in KMT2A-rearranged leukaemia, and does the same mis-spliced MBD1 appear there?

1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   MBD1 (Q9UIS9, 605 aa) carries an MBD at 1-69 and three CXXC-type zinc fingers at 169-216, 217-263 and 330-378; its curated DOMAIN comment states that the THIRD mediates binding to DNA containing unmethylated CpG; its principal annotated function is transcriptional repression at methylated CpG islands; and it carries a CAUTION recording that its SETDB1 replication-coupling papers have been retracted.

   UniProtKB Q9UIS9 (MBD1_HUMAN), Methyl-CpG-binding domain protein 1 / CXXC-type zinc finger protein 3 · UniProt Consortium · UniProtKB 2026
   ▸What this evidence cannot show

   Fetched 2026-09-21; accession resolved by gene_exact:MBD1 AND organism_id:9606 AND reviewed:true, one row, primary gene name read back as MBD1 and length 605 from the same record. Twelve isoforms are annotated and the feature coordinates are canonical-sequence coordinates; several isoforms differ precisely in the CXXC region, which is the biologically interesting part.

   ▸Evidence that argues against this (1)

   • The four subcellular-location terms (nucleus, nucleus matrix, nucleus speckle, chromosome) are all ECO:0000269 and all UNSCOPED, while the functional literature shows the CXXC3-containing and CXXC3-lacking isoforms behave oppositely - so the unscoped annotation flattens the distinction the biology turns on. Zhang P et al. 2017
2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   MBD1's third CXXC domain was measured alongside every other human CXXC domain in a systematic comparison of DNA-binding selectivity supported by co-crystal structures.

   DNA Sequence Recognition of Human CXXC Domains and Their Structural Determinants. · Xu C, Liu K, et al. · Structure 2018
   PMID 29276034DOI
   ▸What this evidence cannot show

   Isolated domains on short duplexes in vitro; the tandem CXXC-PHD modules in the family are handled separately from single CXXC domains, and MBD1's three-finger arrangement is neither.

   ▸Evidence that argues against this (1)

   • In the MLL-AF9 swap study, the isolated MBD1 CXXC domain could not bind the test DNA under the conditions used, which is hard to reconcile with a measured selectivity for unmethylated CpG and suggests the result depends strongly on substrate and assay. Risner LE et al. 2013
3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Substituting the MBD1 CXXC domain, or the MBD1 MBD domain, for the MLL CXXC in MLL-AF9 abrogated colony-forming ability and leukaemogenicity, and the isolated MBD1 CXXC and MBD domains were unable to bind the same DNA that MLL, DNMT1 and CGBP CXXC domains bound.

   Functional specificity of CpG DNA-binding CXXC domains in mixed lineage leukemia. · Risner LE, Kuntimaddi A, et al. · J Biol Chem 2013
   PMID 23990460DOI
   ▸What this evidence cannot show

   Murine MLL-AF9 myeloid system, retroviral transduction of primary bone-marrow progenitors and transplantation. Because the substituted domain does not bind DNA, this arm cannot distinguish 'a foreign CXXC cannot substitute' from 'a non-binding module cannot substitute'. Nothing human, B-lineage or KMT2A::AFF1.

   ▸Evidence that argues against this (1)

   • The CGBP/CXXC1 arm of the same experiment DID bind unmethylated DNA and still abrogated leukaemogenesis, so the informative negative in this paper is CGBP's, not MBD1's. Risner LE et al. 2013
4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Mbd1 enhances Tet1-mediated 5mC oxidation by enhancing Tet1 - not Tet2 or Tet3 - localisation at heterochromatic DNA; the effect depends on the Mbd1 CXXC3 domain's ability to bind unmethylated CpG; Tet1 recruitment and activity then displace Mbd1 from methylated DNA; and an Mbd1 isoform lacking CXXC3 blocks the conversion, giving opposite biological effects for the two isoforms.

   Methyl-CpG binding domain protein 1 regulates localization and activity of Tet1 in a CXXC3 domain-dependent manner. · Zhang P, Rausch C, et al. · Nucleic Acids Res 2017
   PMID 28449087DOI
   ▸What this evidence cannot show

   Mammalian cell lines with fluorescent-protein fusions, live imaging and overexpression; the DNA compartment analysed is pericentromeric heterochromatin, not CpG islands. Displacement is inferred from imaging dynamics rather than from ChIP. No leukaemia and no KMT2A fusion anywhere in it.

   ▸Evidence that argues against this (1)

   • Arand et al. report that the full-length Tet1 zinc finger tethers Tet1 AWAY from heterochromatin, which is the opposite direction from MBD1-enhanced heterochromatic Tet1 localisation; the two can be reconciled only if the MBD1 interaction overrides the intrinsic exclusion, which neither paper tests. Arroyo M et al. 2022
5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   An MDS-associated mis-spliced MBD1 isoform, MBD1-L, produced by reduced WTAP expression rather than by a splicing-factor mutation, recapitulates archetypal MDS defects in healthy human HSPCs through an isoform-specific switch in MBD1 binding behaviour.

   Aberrant splicing of MBD1 reshapes the epigenome to drive convergent myeloerythroid defects in MDS. · Chen HTT, Joshi P, et al. · Blood 2026
   PMID 42322117DOI
   ▸What this evidence cannot show

   Overexpression in healthy human HSPCs plus xenotransplantation; MDS, not KMT2A-rearranged leukaemia. Abstract read this session; full text not read. Recorded because it is the first myeloid-malignancy role for MBD1 and because it is a third independent instance of the isoform-switch theme in this family.

   ▸Evidence that argues against this (1)

   • Overexpression of an isoform in healthy HSPCs demonstrates sufficiency for a phenotype, not that the isoform drives disease at endogenous levels in patients. Chen HTT et al. 2026
6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

   MBD1 shows no dependency in KMT2A-rearranged cell lines: difference -0.0193, rank 6,616 of 17,728, pan-essentiality +0.0753 across 1,208 lines.

   Genome-wide KMT2A-rearranged selectivity scan, 17,728 genes (DepMap CRISPR Chronos via the Open Targets Platform GraphQL API) · Oncofusion knowledge-graph project · analyses/depmap-kmt2a-selective-scan-full/results.json, generated 2026-09-19; Open Targets dataVersion year 26 month 06, re-read from the API on 2026-09-21 2026
   ▸What this evidence cannot show

   The scan compares 11 Cellosaurus-annotated KMT2A-rearranged lines against 43 other leukaemia lines; none of the 11 is infant-derived and the arm is myeloid-enriched, so a difference mixes fusion status with lineage. Both the difference and the pan-essentiality across all 1,208 lines are quoted, because a gene required everywhere is not a selective dependency. Open Targets dataVersion 26.06; DepMap's own release identifier is not returned by that endpoint.

   ▸Evidence that argues against this (1)

   • A positive pan-essentiality means MBD1 knockout is very slightly advantageous on average, which is what a repressor with tumour-suppressor character would give and says nothing about a gain-of-function isoform. Chen HTT et al. 2026
7. E7moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   Using a light-activatable human TET1, the MBD1 MBD domain was shown to negatively regulate TET1-catalysed 5mC oxidation kinetics while the third CXXC domain promotes them, dependent on its ability to bind non-methylated CpG.

   Optochemical Control of TET Dioxygenases Enables Kinetic Insights into the Domain-Dependent Interplay of TET1 and MBD1 while Oxidizing and Reading 5-Methylcytosine. · Lin TC, Palei S, et al. · ACS Chem Biol 2022
   PMID 35709470DOI
   ▸What this evidence cannot show

   A chemical-biology system with a genetically encoded photocaged serine in overexpressed TET1; kinetics are measured in cells but under non-physiological expression. No leukaemia.

   ▸Evidence that argues against this (1)

   • Both modules are in the same protein, so a net effect at any real locus depends on the local 5mC/unmethylated-CpG ratio, which neither this study nor the imaging study manipulates. Zhang P et al. 2017

• related to →The CXXC domain (aa 1147-1195, PDB 2KKF) is a chromatin address, not a compartment address — CpG mutants lose transformation but keep their subnuclear position

  Reads both states of the address - methylated CpG through its MBD, unmethylated CpG through its third CXXC finger.

• related to →CXXC to unmethylated CpG DNA: nine mapped contact residues, and an affinity that spans 44 uM to 33 nM depending on the construct, the DNA and the salt

  Its CXXC domain was swapped into MLL-AF9 in place of this interface's domain and abrogated leukaemogenesis - but did not bind DNA, so the arm is a control rather than a test.

• regulates →TET1 (LCX / CXXC6) - a CXXC reader that is also a KMT2A fusion partner, and whose leukaemic fusion throws its own CXXC domain away

  MBD1's CXXC3 enhances TET1 heterochromatin localisation and 5mC oxidation; TET1 activity then displaces MBD1 - the family's one documented reciprocal occupancy cycle.

• part of →The human CXXC/CpG-reading family is exactly twelve proteins, and TET2 is not one of them

  One of the twelve, and the only one carrying three CXXC fingers.

• related to →CXXC1 (CFP1) - a CpG-reading COMPASS subunit that binds MLL1 and recruits it to HoxA7, and whose CXXC domain cannot substitute for KMT2A's in a fusion

  The two heterologous CXXC domains that failed the MLL-AF9 swap; only CGBP's failure is informative, because MBD1's did not bind DNA.