Motif Tools
Family atlas / CCHC zinc finger

MYT1 / ST18 (CCHC zinc finger)

CCHC zinc fingerAAAGTTT3 motifsaka MYT/NZF family site, AAAGTTT, neural zinc finger factor site
TCGAGATAGTCAGT
representative: MYT1L.H14CORE.0.P.C · data consensus aAAGTT
members tight · 0.89stated AAAGTTT · data aAAGTT1 in vitro1 ChIP-only1 cross-species15 references, all PMIDs verified

MYT1, MYT1L and ST18 are the three human members of the MYT/NZF family, which uses an unusual non-classical zinc finger (described as C2HC or CCHHC, not the C2H2 fold of most zinc-finger transcription factors) arranged in two widely separated clusters of fingers. Each cluster binds DNA independently and recognizes the same short AT-rich element, reported as AAAGTTT, and affinity rises sharply when two copies are present, so the functional element is effectively bipartite. The family matters most in neurons: MYT1L is one of the three factors that convert fibroblasts to induced neurons, where it works by repressing non-neuronal programs. MYT1L haploinsufficiency causes a human neurodevelopmental syndrome with obesity. The available motif models for this family are weak.

The factors

MYT1 was cloned in 1992 by its binding to cis-regulatory elements of the myelin proteolipid protein gene and named myelin transcription factor I 1280325. Its zinc fingers do not belong to the classical C2H2 class: the original description assigned them a Cys-X5-Cys-X12-His-X4-Cys arrangement (C2HC) 1280325, while later structural work on the same domain in this family describes it as a non-classical CCHHC-type finger 24820620, and both names appear in the literature. MYT1 has six such fingers arranged in two widely separated clusters. Critically, each cluster binds DNA independently and recognizes the same sequence, which led to the proposal that one protein can bridge a proximal and a distal site in the same promoter 1280325.

Humans have three members: MYT1 (also MyT1, NZF-2, PLPB1), MYT1L (MYT2) and ST18 (MYT3). ST18 was cloned as the third member of the neural zinc finger factor family, NZF-3; unlike MyT1 and NZF-1, which activate transcription in reporter assays, NZF-3 represses 9478997. Expression across mouse nervous system development is broadly overlapping but not identical among the three 24214099. MYT1 and MYT1L are the neural members in practice 8530187,24214099; MYT1 and ST18 are additionally expressed in pancreatic islet beta cells 32359405. Only MYT1 and MYT1L have motif models in the databases behind this family; ST18 is assigned here on domain grounds.

The motif

The consensus reported for these proteins is AAAGTTT, and the key measurement is that a single copy binds, but a bipartite element containing two copies of the consensus is bound with markedly higher relative affinity 9478997. That is the expected behavior for a protein with two independent finger clusters of identical specificity, and it means the real recognition unit is a pair of short AT-rich cores rather than one longer site. Recognition chemistry has been examined for the NZF-1 finger domain, where hydrogen bonding contributes to base discrimination 24820620.

The motifs actually available are thin. HOCOMOCO's MYT1 model, the only one in this family with an A grade, has the consensus AAACTTTTBNAAA and rests on two in vitro assays, HT-SELEX and GHT-SELEX; its MYT1L model is ChIP-only at grade C with the consensus AAAGTT; and the CIS-BP human MYT1L motif is not measured on a human protein at all but inherited from a Caenorhabditis elegans relative at 79% DNA-binding-domain identity. The data consensus for the family, AAAGTT, therefore reflects six weakly informative positions, one of them from a nematode. A six- or seven-base AT-rich core carries roughly the same information as a TATA box, which sets a hard ceiling on what any MYT PWM can resolve.

Where it acts

MYT1L occupancy, mapped during fibroblast-to-neuron reprogramming, is unusual in two ways: the binding sites are largely the same in neurons and in fibroblasts, and they are preferentially in already-open chromatin 28379941. In other words this factor is not selecting sites by cell-type-specific accessibility; its targets are distributed across the somatic lineage programs it represses. Sites are found at promoters and in accessible regulatory regions of genes belonging to non-neuronal lineages, including several components of Notch signaling such as Hes1 28379941. Because the core is short and AT-rich, matches are dense in AT-rich intergenic and intronic DNA and overlap in scoring space with other AT-preferring models, including homeodomain and forkhead sites and poly-A tracts.

Biology and disease

MYT1L is one of the three transcription factors, with ASCL1 and POU3F2/BRN2, that convert mouse fibroblasts into functional induced neurons 20107439, and it is dispensable for the initial steps but important for the quality and stability of the resulting neurons 24243019. Its mechanism is repressive: MYT1L directly represses many different somatic lineage programs while sparing the neuronal program, and it does so by recruiting a SIN3B-containing complex through an N-terminal domain 28379941. That makes it a counterexample to the assumption that developmental repressors are lineage-specific.

In humans, heterozygous MYT1L deletions and truncating variants (2p25.3) cause MYT1L syndrome: intellectual disability, autistic features and obesity 25232846,32065501. Mouse models reproduce the behavioral and metabolic phenotype, including obesity 35538503,34614421, and human neuron models show maturation defects, some of which have been reversed pharmacologically in these systems 36782060. On the metabolic side, the Myt factors act together in beta cells, where their combined loss allows over-activation of stress-response genes and impairs postnatal beta-cell proliferation and function 32359405. ST18 has been studied mainly outside the nervous system, as a candidate tumor suppressor and in beta-cell function 15489893.

Reading a hit

Be conservative. The motif is a short AT-rich core, so matches are common by chance and cannot be distinguished from other AT-preferring factors on sequence alone; a single AAAGTTT is weak evidence even in an otherwise plausible context. A tandem pair of cores in one element is a better signal, because affinity for the bipartite arrangement is substantially higher. A hit does not identify the paralogue: MYT1, MYT1L and ST18 share one specificity and the databases cannot separate them, and ST18 has no measured human matrix here at all. Two provenance problems are worth stating plainly: the CIS-BP human MYT1L motif is a cross-species transfer from a C. elegans protein, and the HOCOMOCO MYT1L model is ChIP-only at the lowest quality grade. Given that MYT1L's binding sites are mostly in pre-existing open chromatin and that its repressive function works through a recruited SIN3B complex, a MYT1L ChIP peak without a motif is a normal observation rather than an anomaly, and a motif hit without accessibility is not worth much.

How this was worked out

  1. 1992mediumExpression screening with DNA probes, EMSA with zinc chelation, sequence and secondary-structure analysis1280325

    MyTI is cloned by binding to the myelin proteolipid protein gene; its six fingers use a Cys-X5-Cys-X12-His-X4-Cys arrangement unlike C2H2 or C2C2 fingers, and the two separated finger clusters bind DNA independently and read the same sequence.

    Still open then: The recognized sequence was defined only within the PLP regulatory elements, and no structure of the finger existed.

  2. 1998mediumcDNA cloning, EMSA on single and bipartite elements, cotransfection reporter assays9478997

    The third family member, NZF-3/ST18, is cloned and shown to bind a single AAAGTTT copy but with markedly higher relative affinity to a bipartite element with two copies, and to repress rather than activate reporters, so the family contains both signs of output on one consensus.

    Still open then: Whether the bipartite preference reflects one protein using both clusters or two proteins binding adjacent sites.

  3. 2010mediumCombinatorial factor screening with electrophysiology on induced neurons20107439

    MYT1L is identified as one of three factors (with ASCL1 and POU3F2/BRN2) that convert mouse fibroblasts directly into functional neurons, which moved the family from myelin-gene regulation to cell-fate control.

    Still open then: What MYT1L contributes mechanistically to reprogramming was not known.

  4. 2013mediumChIP-seq of reprogramming factors, chromatin accessibility, staged reprogramming experiments24243019

    In the reprogramming hierarchy ASCL1 acts first as the on-target pioneer factor, with MYT1L dispensable for initiating conversion but required for the quality and maturation of the induced neurons.

  5. 2014mediumPeptide binding studies with mutational and spectroscopic analysis24820620

    Biophysical analysis of the NZF-1 finger domain characterizes the non-classical CCHHC finger and the hydrogen bonds that contribute to base discrimination, giving the family's recognition chemistry its first direct treatment.

    Still open then: No structure of a MYT finger cluster bound to DNA has been reported.

  6. 2015mediumArray-based deletion mapping and clinical phenotyping of a patient series25232846

    Mapping of 2p25.3 deletions in patients narrows the critical interval to MYT1L and links haploinsufficiency to intellectual disability with obesity, establishing the human phenotype.

  7. 2017mediumChIP-seq and ATAC-seq in fibroblasts and neurons, transcriptomics, domain mapping and complex purification28379941

    MYT1L is shown to work as a repressor of many different non-neuronal lineage programs at once, acting through a recruited SIN3B complex, and to bind largely the same, already-accessible sites in fibroblasts and neurons, which overturns the expectation that a cell-fate factor selects cell-type-specific sites.

    Still open then: Why occupancy is so similar between cell types while the transcriptional effect is not.

  8. 2022mediumMouse behavioral and metabolic phenotyping, human iPSC-derived neuron models with drug treatment35538503 34614421 36782060

    Mouse Myt1l haploinsufficiency reproduces the human phenotype, including obesity and behavioral alterations, and human neuron models show maturation defects that can be reversed genetically and pharmacologically in those systems.

Still unresolved

How much to trust a hit

The evidence base is weak for a whole family. The recognition unit is a six- to seven-base AT-rich core, which occurs constantly by chance and overlaps homeodomain, forkhead and poly-A signals; only a tandem pair of cores is a meaningfully specific element. MYT1, MYT1L and ST18 cannot be separated by sequence, and ST18 has no human matrix here. The CIS-BP human MYT1L motif is transferred from a Caenorhabditis elegans protein at 79% DNA-binding-domain identity, and the HOCOMOCO MYT1L model is ChIP-only at grade C; only the MYT1 model has in vitro support. Since MYT1L binds mostly in already-open chromatin and represses via a recruited SIN3B complex, peaks without motifs are expected.

References

3 motifs in this family — HOCOMOCO 2 · CIS-BP 1
motifdatabaseconsensusbits
MYT1.H14CORE.0.SG.ASGAHOCOMOCOaAACTTTtbnaaA14.1
M00650_2.00Caenorhabditis elegans 0.79CIS-BPAAGTTT9.8
MYT1L.H14CORE.0.P.CPCHOCOMOCOaAAGTT9.3

HOCOMOCO tags read as assay + quality grade: P ChIP-seq, S HT-SELEX, M methyl-HT-SELEX, G GHT-SELEX, I SMiLE-seq, B PBM; grade A means two or more assay types agreed. A P-only matrix is a record of where the protein was crosslinked, which need not be a sequence it binds itself.

The factors, elsewhere

MYT1MYT1L
Find these in JASPAR →in HOCOMOCO →this entry as JSON →
how this page was made Drafted by a language model reading the primary literature, then checked: every cited PMID is resolved against PubMed and its citation text regenerated from PubMed's own record, and every claim about the motifs themselves is checked against the matrices in the four source databases. Methods and limits →
See this family on the motif map →Search your motif against the databases →