MYB / MYB-like
MYB factors bind the Myb binding site (MBS), a six-base element written YAACKG, through two or three imperfect ~52-residue repeats that each fold into a helix-turn-helix-like unit. Humans have three true MYB proteins: MYB (c-Myb), MYBL1 (A-Myb) and MYBL2 (B-Myb). Only repeats R2 and R3 are required for sequence-specific binding, and their recognition helices sit side by side in one major groove. MYB proteins control hematopoietic progenitor expansion, spermatogenesis and, for MYBL2, mitotic gene expression through the MuvB complex. MYB is duplicated in T-cell acute lymphoblastic leukemia and fused to NFIB in adenoid cystic carcinoma. This family key also collects Myb/SANT-fold archetypes whose proteins are not MYB factors and do not bind the MBS, which is the main source of confusion here.
The factors
Three human genes encode true MYB transcription factors: MYB (c-Myb), MYBL1 (A-Myb) and MYBL2 (B-Myb); HOCOMOCO labels them MYB, MYBA and MYBB, and the Vierstra archetypes use the group name MYBL. Their DNA-binding domain is built from three imperfect repeats of about 52 residues, R1, R2 and R3, each containing three regularly spaced tryptophans and folding into a helix-turn-helix-related unit 8415700,7954830. R1 is dispensable: R2 and R3 alone give specific binding 8415700,7954830, which is why plant and fungal homologues are usually R2R3 proteins. MYB is expressed in immature hematopoietic cells, colonic crypt epithelium and some neural progenitors; MYBL1 is highest in testis and germinal-center B cells; MYBL2 is a cell-cycle gene, rising in S phase and peaking at G2/M. This atlas key is deliberately broader than the three MYB proteins, because the Myb/SANT fold is reused by proteins that are not MYB factors. In the current data those extras are DMTF1 (DMP1, a cyclin D-binding Myb-like protein) 8887674, MYPOP (a Myb-related protein that partners profilin) and TTF1 (the RNA polymerase I transcription termination factor, not the thyroid factor NKX2-1, which is unfortunately also called TTF-1) 7597036. TERF1 and TERF2, whose telobox is a Myb-like repeat, are held in a separate TELOBOX family here; SNAPC4, which uses Myb repeats to bind the snRNA proximal sequence element, has no motif in these databases. The SANT domain itself, found in chromatin remodelers and corepressors, has the Myb fold but is thought to read histone tails rather than DNA sequence 15040448.
The motif
The literature consensus for the MBS is YAACKG, six bases, usually quoted as PyAAC(G/T)G 3185713. The invariant part is AAC followed by G or T then G; the identity comes from the recognition helix of R3 inserting into the major groove with R2 packed alongside, so the two repeats read one continuous site 7954830,8415700 and the cavity in R2 rearranges on binding. Database renderings differ mainly by strand and by doubling. JASPAR MYB (MA0100.4) and the corresponding CIS-BP model give CAACTG, which is YAACKG with Y=C and K=T; the HOCOMOCO ChIP model GGCAGTTG is the same site read on the other strand. The data consensus for the family as a whole, CCGTTA, is the reverse complement of TAACGG, again the same core. In vitro selection on MYBL1 and MYBL2 returns longer, near-palindromic models such as AACCGTTAACGG and CCGTTAAACSG 23332764: these are two MBS half-sites in inverted orientation and reflect how the isolated domains bind selected DNA, not a different specificity. Within the three MYB proteins nothing is separable by sequence. The non-MYB members are separable, and they are not MBS models: DMTF1's grade-A model, backed by ChIP-seq and four in vitro assays, is CCCGGATG 8887674, MYPOP's is GCGCAA, and TTF1's is the rDNA Sal box terminator 7597036, recognizable by its internal GACCAG. Their nearest neighbours across the four databases fall outside this family, and the family's cohesion (0.72 over 22 motifs) is lower than for a single-specificity family for exactly that reason.
Where it acts
MYB sites are short and occur in both promoters and enhancers. In hematopoietic cells c-Myb occupies thousands of sites and cooperates with CEBP factors, RUNX1, GATA2 and the coactivators CBP/p300; classic targets include MYC, KIT, CCNB1 and, in T cells, TCR genes. MYBL2 behaves differently: it is recruited to mitotic promoters mainly as a subunit of the MYB-MuvB (MMB) complex, which hands over from the repressive DREAM complex at G1/S 22391450, and MuvB's own sequence anchor is the CHR element read by LIN54, not the MBS 22064854. Consequently much of the MYBL2 signal in chromatin immunoprecipitation sits on CHR-containing promoters where the underlying sequence is not a Myb site at all 22064854,22391450. TTF1 sites are confined to the rDNA repeat, where the Sal box terminates Pol I transcription 7597036, so its motif is genomically almost absent outside the rDNA and its nucleolar-organizer copies.
Biology and disease
Mice lacking c-Myb die around embryonic day 15 with a failure of fetal hepatic hematopoiesis 1709592, the sharpest demonstration that MYB is needed for progenitor expansion rather than for lineage identity. Mybl1-null male mice are sterile with a meiotic arrest, and Mybl2 loss is lethal at the blastocyst stage. In cancer, MYB is more often amplified, translocated or driven by a new enhancer than mutated: tandem duplication of MYB occurs in T-cell acute lymphoblastic leukemia 17435759, and a recurrent MYB-NFIB fusion defines most adenoid cystic carcinomas of salivary gland and breast 19841262. MYBL2 overexpression is a component of proliferation signatures in many tumors and is largely a readout of cell-cycle activity.
Reading a hit
A YAACKG match is six informative bases and will occur by chance roughly every few hundred base pairs on either strand, so a single hit is close to meaningless without other evidence. The three MYB proteins cannot be separated from sequence, so a hit means "a MYB-family site" at best, and expression context is what narrows it: MYB in immature hematopoietic cells, MYBL1 in testis and germinal centers, MYBL2 in cycling cells. The bigger hazard in this key is the label. Several members are Myb/SANT-fold proteins that do not bind the MBS, and a match scored against the DMTF1, MYPOP or TTF1 models is a match to an unrelated sequence that happens to share the family key. For MYBL2, treat ChIP-derived enrichment with care: recruitment through MuvB to CHR elements means a B-Myb peak often marks a LIN54 site rather than a Myb site 22064854.
How this was worked out
The v-Myb oncoprotein is a sequence-specific DNA-binding protein that recognizes PyAACG/TG, the Myb-binding site (MBS).
Still open then: Which of the three repeats contact the site, and how.
c-myb knockout mice become severely anemic and die around embryonic day 15: yolk-sac erythropoiesis is normal but fetal-liver hematopoiesis fails across lineages.
NMR structure of the R2R3 domain on DNA: the recognition helices of R2 and R3 pack together in the major groove and read AACNG cooperatively, while R1 makes no specific contacts, consistent with earlier mutational mapping that put R3 on the core AAC.
DMP1 (DMTF1), a cyclin D-binding Myb-like protein, binds a different nonamer, CCCG(G/T)ATGT, showing that a Myb-like fold does not imply the MBS.
The t(6;9) translocation in adenoid cystic carcinomas of breast and head and neck fuses MYB to NFIB, removing MYB 3' sequences that carry microRNA target sites.
B-Myb reaches late cell-cycle promoters as part of the MuvB complex, and the B-Myb-MuvB complex binds the CHR element independently of any Myb site, so B-Myb promoter occupancy is largely recruitment rather than MBS binding.
Still open then: Whether B-Myb's own repeats also contact DNA at these promoters.
HT-SELEX on MYBL1 and MYBL2 returns longer near-palindromic models built from two inverted MBS half-sites.
Still open then: Whether the inverted double site is used at genomic sites.
Still unresolved
- Whether the K position separates paralogues or assays is unresolved: the c-Myb models here favor CAACTG (K = T), the HT-SELEX models for MYBL1 and MYBL2 favor TAACGG (K = G), and the HOCOMOCO MYBL1 ChIP model accepts either, so with no in vitro c-Myb model in these data, paralogue and method are confounded.
- Whether MYB proteins use paired, inverted MBS half-sites in cells is unknown: HT-SELEX on MYBL1 and MYBL2 recovers near-palindromic double sites, whereas the literature MBS and the ChIP-derived models are single six-base elements.
- How much B-Myb ChIP signal reflects direct DNA contact is unresolved: B-Myb reaches late cell-cycle promoters through MuvB bound to the CHR, and any contribution from its own repeats binding an MBS at those promoters has not been separated from that recruitment.
Two problems compound here. First, the MBS is six bases, so chance matches are abundant and the three MYB paralogues are indistinguishable by sequence. Second, this key is a fold class, not a single specificity: DMTF1 (CCCGGATG), MYPOP (GCGCAA) and TTF1 (the rDNA Sal box) are Myb/SANT-fold proteins whose motifs are unrelated to the MBS, and the CIS-BP MYPOP model is transferred from mouse. The family cohesion of 0.72 reflects this mixture. For MYBL2, ChIP-derived signal is frequently recruitment through the MuvB complex onto LIN54's CHR element rather than direct MBS binding, so a B-Myb peak is not evidence of a Myb site.
References
- Biedenkapp H, Borgmeyer U, Sippel AE, et al. Viral myb oncogene encodes a sequence-specific DNA-binding activity. Nature 1988;335:835-7. PMID 3185713
- Tanikawa J, Yasukawa T, Enari M, et al. Recognition of specific DNA sequences by the c-myb protooncogene product: role of three repeat units in the DNA-binding domain. Proc Natl Acad Sci U S A 1993;90:9320-4. PMID 8415700
- Ogata K, Morikawa S, Nakamura H, et al. Solution structure of a specific DNA complex of the Myb DNA-binding domain with cooperative recognition helices. Cell 1994;79:639-48. PMID 7954830
- Mucenski ML, McLain K, Kier AB, et al. A functional c-myb gene is required for normal murine fetal hepatic hematopoiesis. Cell 1991;65:677-89. PMID 1709592
- Lahortiga I, De Keersmaecker K, Van Vlierberghe P, et al. Duplication of the MYB oncogene in T cell acute lymphoblastic leukemia. Nat Genet 2007;39:593-5. PMID 17435759
- Persson M, Andrén Y, Mark J, et al. Recurrent fusion of MYB and NFIB transcription factor genes in carcinomas of the breast and head and neck. Proc Natl Acad Sci U S A 2009;106:18740-4. PMID 19841262
- Sadasivam S, Duan S, DeCaprio JA. The MuvB complex sequentially recruits B-Myb and FoxM1 to promote mitotic gene expression. Genes Dev 2012;26:474-89. PMID 22391450
- Boyer LA, Latek RR, Peterson CL. The SANT domain: a unique histone-tail-binding module?. Nat Rev Mol Cell Biol 2004;5:158-63. PMID 15040448
- Hirai H, Sherr CJ. Interaction of D-type cyclins with a novel myb-like transcription factor, DMP1. Mol Cell Biol 1996;16:6457-67. PMID 8887674
- Evers R, Grummt I. Molecular coevolution of mammalian ribosomal gene terminator sequences and the transcription termination factor TTF-I. Proc Natl Acad Sci U S A 1995;92:5827-31. PMID 7597036
- Jolma A, Yan J, Whitington T, et al. DNA-binding specificities of human transcription factors. Cell 2013;152:327-39. PMID 23332764
- Weirauch MT, Yang A, Albu M, et al. Determination and inference of eukaryotic transcription factor sequence specificity. Cell 2014;158:1431-1443. PMID 25215497
- Müller GA, Quaas M, Schümann M, et al. The CHR promoter element controls cell cycle-dependent gene transcription and binds the DREAM and MMB complexes. Nucleic Acids Res 2012;40:1561-78. PMID 22064854
22 motifs in this family — JASPAR 3 · HOCOMOCO 9 · CIS-BP 4 · Vierstra 6
| motif | database | consensus | bits |
|---|---|---|---|
| MYBL2 (MA0777.1) | JASPAR | AaCCGTTAAACsGtc | 16.8 |
| DMTF1.H14CORE.0.PSGIB.APSGIBA | HOCOMOCO | aggaCCCGGATG | 16.3 |
| MYBB.H14CORE.0.S.BSB | HOCOMOCO | AACCGTTAACgg | 16 |
| M03345_2.00 | CIS-BP | aaCcGTTAACcGTT | 15.4 |
| MYBL1 (MA0776.1) | JASPAR | ACCGTTAACggt | 15.3 |
| MYBA.H14CORE.2.S.BSB | HOCOMOCO | ACCGTTAAC | 15.3 |
| AC0279:MYBL:Myb/SANT | Vierstra | CcGTTAdAACCGTT | 15.1 |
| TTF1.H14CORE.0.PS.APSA | HOCOMOCO | GGthgACCAGmaGcc | 14.4 |
| AC0282:MYBL:Myb/SANT | Vierstra | ccGTTAACGGT | 13.8 |
| AC0281:MYBL:Myb/SANT | Vierstra | CCGTTAAACsG | 13 |
| M00808_2.00Mus musculus 0.96 | CIS-BP | TTGCGCc | 12 |
| M03349_2.00 | CIS-BP | cvGTTTAACgGt | 11.4 |
| M08150_2.00 | CIS-BP | CAACTG | 11.3 |
| MYB (MA0100.4) | JASPAR | CAACTG | 11.1 |
| MYBA.H14CORE.0.P.BPB | HOCOMOCO | GGCvGTTg | 10.4 |
| MYPOP.H14CORE.0.PSGB.APSGBA | HOCOMOCO | GCGCaA | 9.7 |
| MYBA.H14CORE.1.PS.APSA | HOCOMOCO | gGCmGTTg | 9.6 |
| MYB.H14CORE.0.P.BPB | HOCOMOCO | GgCaGTTg | 9.4 |
| MYBB.H14CORE.1.S.BSB | HOCOMOCO | CCGTTA | 9.3 |
| AC0280:MYBL:Myb/SANT | Vierstra | AACCGTTA | 9.3 |
| AC0275:MYB/MYBL:Myb/SANT | Vierstra | cAACtGcc | 8.5 |
| AC0276:MYB:Myb/SANT | Vierstra | CAACTG | 8 |
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
These entries are written from the primary literature and every citation is checked against PubMed, but this one can still be out of date, too confident, or simply wrong — especially where a motif’s published site and its database matrices disagree. Corrections and additions are welcome, and they are the fastest way to make this page better for everyone.