MECOM / EVI1
MECOM, better known as EVI1, is a 145-kDa C2H2 zinc-finger protein with two separate finger clusters that bind two different sequences. The N-terminal cluster of seven fingers reads a 15-bp GATA-like site, GA(C/T)AAGA(T/C)AAGATAA, whose core is GACAAGATA; the C-terminal cluster of three fingers reads the unrelated GAAGATGAG. EVI1 is normally expressed in hematopoietic stem cells, kidney and lung, and is silenced in most differentiated blood cells. It is one of the strongest adverse prognostic markers in acute myeloid leukemia, where 3q26 rearrangements such as inv(3)(q21q26) do not create a fusion protein but move a GATA2 enhancer next to MECOM, switching the gene on. Germline MECOM mutations cause radioulnar synostosis with amegakaryocytic thrombocytopenia.
The factors
MECOM (MDS1 and EVI1 complex locus, 3q26.2) encodes EVI1, first found as the site of retroviral integration in murine myeloid leukemia cell lines 2842066. The protein carries two physically separate arrays of Cys2-His2 zinc fingers, seven at the N terminus and three near the C terminus 2842066,2017172, plus a CtBP-interaction motif and an acidic domain; it can act as a repressor or an activator depending on partners. A longer read-through transcript, MDS1-EVI1, adds a PR/SET domain and is the reason EVI1 was once numbered PRDM3. Unlike the PRDM family proper, the two finger clusters of EVI1 are used as independent DNA-binding modules, which is why this atlas treats MECOM separately. EVI1 binds as a monomer, not an obligate dimer, and one protein can engage both site types. Expression is restricted: hematopoietic stem and progenitor cells, kidney, lung and developing heart and limb, with silencing on myeloid and lymphoid differentiation. There is only one human gene, so there is no paralogue ambiguity within the family, but there is considerable ambiguity with GATA factors.
The motif
The proximal (N-terminal, fingers 1-7) cluster recognizes a 15-nucleotide consensus, GA(C/T)AAGA(T/C)AAGATAA, which contains a tandem pair of GATA-like half-sites 8321231. Fingers 4-7 make the sequence-specific contacts; fingers 1-3 do not detectably bind DNA on their own but bias the first position toward GACAA over GATAA and are needed for optimal occupancy of the full 15-mer 8321231. The functional minimal core is GACAAGATA 8557637, and a single amino-acid change (R205N) abolishes high-affinity binding to it 16006653. Because GATAA is embedded in the site, EVI1 and GATA factors have overlapping but not identical preferences: EVI1 binds erythroid GATA-1 target sites only weakly, while GATA-1 can bind and activate through a GACAAGATAA element 8557637. The distal (C-terminal, three-finger) cluster binds a completely different and shorter consensus, GAAGATGAG 8183551.
The database consensus in this atlas, TTATCTTATCT (and the 14-bp TATCTTATCWTATC archetype), is the reverse complement of the proximal site written as tandem AGATAAGATAA half-sites; JASPAR renders the same thing as AGATAAGATAA. So the literature consensus and the data agree once strand and finger cluster are accounted for, and the longer Vierstra archetype simply captures more half-site repeats. No widely used matrix in these databases represents the distal GAAGATGAG site, which is worth remembering: scanning with a MECOM matrix tests only one of EVI1's two specificities.
Where it acts
EVI1 occupancy in fibroblasts and myeloid cells maps to conserved GACAAGATA-like motifs within about 10 kb of transcription start sites of a defined target set, including GATA2, ZFPM2/FOG2, GADD45G, SKIL, KLF5, DCN and MAP3K14 16006653. Sites are found in both promoters and distal elements. Because the proximal site is a doubled GATA half-site, EVI1 peaks overlap GATA1/GATA2 peaks in hematopoietic cells, and in erythroid or megakaryocytic chromatin a WGATAA-rich region is ambiguous between the two families. EVI1 also functions through protein interactions with a large partner network, including CtBP and polycomb-associated proteins, so a fraction of its ChIP signal is likely tethered rather than sequence-driven.
Biology and disease
Conditional deletion of Evi1 in mice depletes hematopoietic stem cells and impairs repopulation, and also blocks transformation by leukemic fusion oncogenes, making EVI1 required for both normal and malignant stem-cell self-renewal 18682242. High EVI1 expression marks roughly 6-11% of adult acute myeloid leukemia and is associated with poor outcome 26729571. The clearest mechanism is enhancer hijacking: in inv(3)(q21q26) and t(3;3)(q21;q26), the breakpoint does not create a fusion protein but relocates a distal GATA2 hematopoietic enhancer to the MECOM locus, driving EVI1 expression while leaving the other GATA2 allele haploinsufficient 24703711. Two independent studies established this, one by mapping the rearranged enhancer in patient genomes 24703711 and one by deleting the orthologous mouse enhancer 24703906. Germline missense mutations in MECOM, one of them de novo and all three clustered in the eighth zinc finger of the C-terminal domain, cause radioulnar synostosis with amegakaryocytic thrombocytopenia 26581901, a bone-marrow-failure syndrome with a skeletal phenotype, confirming a developmental role beyond blood.
Reading a hit
A MECOM match is a doubled GATA-like element, so the first alternative hypothesis is a GATA factor, not EVI1 - especially in erythroid, megakaryocytic or endothelial cells where GATA1/GATA2 are abundant and EVI1 is usually off. The 15-bp full site is long enough to be reasonably specific, but the shorter matrices in these databases (10-11 bp, essentially two WGATAA half-sites) will match tandem GATA sites indiscriminately. A hit on the distal GAAGATGAG specificity will not be found at all with the available matrices. There is only one human MECOM gene, so a true EVI1 hit has no paralogue problem, but the CIS-BP model is inferred from mouse Evi1 at 98% DNA-binding-domain identity rather than measured on the human protein 25215497, and the Vierstra archetypes are named for MECOM without claiming separability from GATA. Treat a hit as evidence for a GATA-type element and use expression data to decide whether EVI1 is a plausible occupant.
How this was worked out
Evi-1 is identified as a common retroviral integration site activated in IL-3-dependent mouse myeloid leukemia lines, encoding a protein with two zinc-finger domains of seven and three fingers.
Still open then: Whether the protein bound DNA, and what sequence, was unknown.
Purified Evi-1 is shown to be a sequence-specific DNA-binding protein with a high-affinity site, TGACAAGATAA, protected in DNase I footprints.
Still open then: Which of the ten fingers made the contacts was not resolved.
The N-terminal seven-finger domain recognizes a 15-nucleotide consensus, GA(C/T)AAGA(T/C)AAGATAA; four of the seven fingers are required, and fingers 1-3 bias the first position toward GACAA.
The C-terminal three-finger domain is shown to bind a different consensus, GAAGATGAG, which, unlike the N-terminal site, follows the standard C2H2 recognition rules.
Still open then: The site did not support reporter activation, leaving its function in cells unclear.
- 1996highMethylation interference; quantitative gel shift; genomic fragment selection; cotransfection8557637
Fingers 1-7 require the GACAAGATA core and bind erythroid GATA-1 target sites only weakly, arguing against simple occupation of GATA sites by EVI1.
A single-residue mutant (R205N) abolishes high-affinity binding, and native EVI1 occupies GACAAGATA-like motifs near target genes including Gata2, Zfpm2 and Gadd45g in myeloid cells.
- 2014highFunctional genomics and enhancer excision in AML cells; BAC transgenic mice with and without the enhancer24703711 24703906
Two groups independently show that inv(3)/t(3;3) leukemias activate EVI1 by repositioning a distal GATA2 hematopoietic enhancer to the MECOM locus rather than by creating a fusion protein.
A de novo MECOM missense mutation and two further missense mutations, all clustered in the eighth zinc finger (C-terminal domain), are found in radioulnar synostosis with amegakaryocytic thrombocytopenia.
Still open then: The mutants were tested at an ETS-like EVI1 site rather than at GAAGATGAG, so which C-terminal specificity matters in vivo was left open.
Still unresolved
- What the C-terminal finger cluster reads in cells is unsettled: in vitro selection gave GAAGATGAG, whereas the disease-causing mutations in finger 8 were assayed at an ETS-like EVI1 site, and no matrix in these databases represents either specificity.
- Whether EVI1 antagonizes GATA factors by competing for shared sites is unresolved: GATA-1 can bind and activate through the GACAAGATAA element, but EVI1 binds erythroid GATA-1 sites only weakly, so direct site competition and indirect interference remain alternative models.
The proximal EVI1 site is a tandem GATA-like element, so matches are shared with GATA1-GATA6 and TRPS1; in blood and endothelium a GATA factor is usually the more likely occupant, since EVI1 is off in most differentiated cells. The shorter matrices here are effectively two WGATAA half-sites and match tandem GATA sites indiscriminately. EVI1's second specificity, GAAGATGAG, is not represented by any matrix in these databases, so a MECOM scan tests only one of two binding modes. The CIS-BP model is inferred from mouse Evi1 (98% DBD identity), not measured on human EVI1. Part of EVI1's ChIP signal is likely tethered through its large protein-interaction network rather than direct.
References
- Morishita K, Parker DS, Mucenski ML, et al. Retroviral activation of a novel gene encoding a zinc finger protein in IL-3-dependent myeloid leukemia cell lines. Cell 1988;54:831-40. PMID 2842066
- Perkins AS, Fishel R, Jenkins NA, et al. Evi-1, a murine zinc finger proto-oncogene, encodes a sequence-specific DNA-binding protein. Mol Cell Biol 1991;11:2665-74. PMID 2017172
- Delwel R, Funabiki T, Kreider BL, et al. Four of the seven zinc fingers of the Evi-1 myeloid-transforming gene are required for sequence-specific binding to GA(C/T)AAGA(T/C)AAGATAA. Mol Cell Biol 1993;13:4291-300. PMID 8321231
- Funabiki T, Kreider BL, Ihle JN. The carboxyl domain of zinc fingers of the Evi-1 myeloid transforming gene binds a consensus sequence of GAAGATGAG. Oncogene 1994;9:1575-81. PMID 8183551
- Perkins AS, Kim JH. Zinc fingers 1-7 of EVI1 fail to bind to the GATA motif by itself but require the core site GACAAGATA for binding. J Biol Chem 1996;271:1104-10. PMID 8557637
- Yatsula B, Lin S, Read AJ, et al. Identification of binding sites of EVI1 in mammalian cells. J Biol Chem 2005;280:30712-22. PMID 16006653
- Goyama S, Yamamoto G, Shimabe M, et al. Evi-1 is a critical regulator for hematopoietic stem cells and transformed leukemic cells. Cell Stem Cell 2008;3:207-20. PMID 18682242
- Gröschel S, Sanders MA, Hoogenboezem R, et al. A single oncogenic enhancer rearrangement causes concomitant EVI1 and GATA2 deregulation in leukemia. Cell 2014;157:369-381. PMID 24703711
- Yamazaki H, Suzuki M, Otsuki A, et al. A remote GATA2 hematopoietic enhancer drives leukemogenesis in inv(3)(q21;q26) by activating EVI1 expression. Cancer Cell 2014;25:415-27. PMID 24703906
- Niihori T, Ouchi-Uchiyama M, Sasahara Y, et al. Mutations in MECOM, Encoding Oncoprotein EVI1, Cause Radioulnar Synostosis with Amegakaryocytic Thrombocytopenia. Am J Hum Genet 2015;97:848-54. PMID 26581901
- Hinai AA, Valk PJ. Review: Aberrant EVI1 expression in acute myeloid leukaemia. Br J Haematol 2016;172:870-8. PMID 26729571
- Weirauch MT, Yang A, Albu M, et al. Determination and inference of eukaryotic transcription factor sequence specificity. Cell 2014;158:1431-1443. PMID 25215497
4 motifs in this family — JASPAR 1 · CIS-BP 1 · Vierstra 2
| motif | database | consensus | bits |
|---|---|---|---|
| M02667_2.00Mus musculus 0.98 | CIS-BP | TTATCTTaTCT | 17.9 |
| Mecom (MA0029.2) | JASPAR | AGAtAAGATAA | 17.3 |
| AC0242:MECOM:C2H2_ZF | Vierstra | TTATCTTaTC | 14.9 |
| AC0244:MECOM:C2H2_ZF | Vierstra | TATCTTaTCwTATC | 14.1 |
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.