NFAT
NFAT sites are read by five proteins with a Rel-homology DNA-binding domain: the calcineurin-regulated NFATC1, NFATC2, NFATC3 and NFATC4, plus the structurally distinct, calcium-independent NFAT5 (TonEBP). Unlike NF-kB, the calcineurin-dependent NFATs bind as monomers to a short GGAAA core, usually written TGGAAA and only about six informative bases long. Because that is too little information to specify a site on its own, NFATs typically work in composite elements, most famously the NFAT-AP1 element of the interleukin-2 promoter, where NFAT and a Fos-Jun dimer touch each other on adjacent DNA. NFAT activity is controlled by dephosphorylation and nuclear import, which is why cyclosporin A and tacrolimus, which inhibit calcineurin, are immunosuppressive.
The factors
Five human proteins carry an NFAT DNA-binding domain. NFATC1 (NFAT2), NFATC2 (NFAT1), NFATC3 (NFAT4) and NFATC4 (NFAT3) share a Rel-homology region related to that of NF-kB, preceded by a long regulatory domain packed with serine-rich motifs and SP boxes. In a resting cell that domain is heavily phosphorylated and the protein is cytoplasmic; a sustained rise in cytosolic calcium activates the phosphatase calcineurin, which dephosphorylates it and exposes a nuclear localisation signal 8397339,12975316. Nuclear export kinases such as GSK3, CK1 and DYRK reverse this, so NFAT shuttles continuously and its nuclear concentration tracks calcium in near-real time 12975316. NFATC1 and NFATC2 dominate in lymphocytes, NFATC3 in thymocytes, NFATC4 in heart, skeletal muscle and brain 9143705. NFAT5 (TonEBP) is the outlier: it keeps the Rel-homology fold but lacks the calcineurin-docking regulatory domain, is activated by hypertonicity rather than calcium 10051678, and binds DNA as an obligate homodimer that completely encircles the double helix 11780147. That structural work also made the family's relationship to NF-kB explicit, since DNA-bound NFAT5 adopts an NF-kB-like arrangement 11780147.
The motif
The core is GGAAA, normally written with its 5' T as TGGAAA, and the databases converge on a very short model: the representative in vitro-supported HOCOMOCO entry is ATGGAAAAT, of whose nine positions about six carry real information. The Rel-homology domain contacts the GGAA bases through basic residues in the equivalent of NF-kB's L1 loop 9510247, but a monomer covers only one half-site's worth of DNA, so total specificity is low - TGGAAA occurs roughly every 2 kb in random sequence. Two features add length in practice. First, many NFAT models in these databases are two half-sites in opposed orientation (for example TTTCCATGGAAA, or TTTCCNNTAACGGAAAA), reflecting genuine NFAT dimers on pseudo-palindromic, kB-like sites; the HIV-1 LTR tandem kB element is the structural example 12949493,18462673. Second, the classic arrangement is the NFAT-AP1 composite element of the interleukin-2 distal antigen-receptor response element, where NFAT and Fos-Jun bind adjacent sites and form a continuous protein-DNA interface across 15 bp 9510247. That partnership leaves a visible mark on the motif collections: the ChIP-derived NFATC1 and NFATC2 models here read TGGAAAAANTGAGTCA - the NFAT core spliced to a canonical AP-1 site, which is a property of the element, not of the protein. NFAT5 differs slightly, preferring a TonE-like TGGAAANNYNY, but not enough to separate it from NFATC1-4 by score. Within NFATC1-4 there is no usable discrimination at all: family cohesion is 0.87 and several members' models are effectively identical.
Where it acts
NFAT sites are enhancer elements far more often than promoter elements, and they are not CpG-island features. Occupancy is strongly cell-type restricted because it depends both on calcium signalling and on chromatin a lineage factor has already opened; in T cells the relevant regions are co-occupied by AP-1, and frequently by RUNX1, ETS1, GATA3 or TBX21 depending on the subset. Composite elements with recognised names include the interleukin-2 ARRE-2 (NFAT-AP1) 7565783,9510247 and the tandem kB/NFAT element of the HIV-1 long terminal repeat, which NFAT and NF-kB can each occupy 12949493. Cooperation with AP-1 is not optional for much of the programme: NFAT mutants that cannot recruit Fos and Jun activate a different, largely non-overlapping set of genes 10970869.
Biology and disease
The calcineurin-NFAT axis is the main transcriptional output of antigen-receptor signalling, driving interleukin-2, interleukin-4, interferon-gamma, CD40LG and FASLG in T cells 9143705,15928679, and contributing to the anergy programme when NFAT acts without AP-1 15928679. Beyond immunity, NFATC1 is required for osteoclast differentiation downstream of RANK and for heart valve development, and NFATC3 and NFATC4 contribute to cardiac hypertrophy 12975316. The pathway is a first-line drug target: cyclosporin A and tacrolimus block calcineurin and are standard immunosuppressants in transplantation, and their renal and neurological toxicity reflects how broadly the pathway is used 12975316. NFAT5-null mice show that the lymphoid microenvironment is osmotically stressful, with impaired lymphocyte function and renal medullary defects 15247420, and NFAT5 drives osmolyte-transporter genes under hypertonic conditions 10051678.
Reading a hit
A bare TGGAAA match is weak evidence. With six informative positions the site occurs hundreds of thousands of times in the human genome, so one match carries essentially no information unless supported by clustering, conservation, or an adjacent partner site. Look for composites: a TGGAAA within about 15 bp of a TGASTCA is far more convincing than either alone, and tandem or inverted GGAAA repeats indicate the dimeric mode. A hit cannot distinguish NFATC1 from NFATC2, NFATC3 or NFATC4, and only marginally separates NFAT5, so report it as NFAT-family. Check the evidence code before trusting a model: HOCOMOCO entries with S or M in the fourth field come from HT-SELEX or methyl-HT-SELEX and are statements about the protein, whereas the ChIP-only NFATC1 and NFATC2 models here are demonstrably contaminated by the composite element's AP-1 site - scan with those and you are scoring AP-1 as much as NFAT. Note also that the Vierstra archetype AC0682 is labelled NFATC/ZNF because unrelated zinc-finger motifs clustered with it, not because those factors share a specificity.
How this was worked out
NFAT-1 is identified as a complex, formed only with nuclear extracts of activated T cells, that binds a functional sequence of the interleukin-2 enhancer and appears minutes before IL-2 transcription; a similar activity binds the HIV-1 LTR.
Still open then: The proteins in the complex, including whether AP-1 was part of it, were unknown.
NFATp is shown to be a substrate of the calcium-activated phosphatase calcineurin and to bind DNA together with Fos and Jun, linking calcium signalling to the NFAT-AP-1 composite site.
Still open then: The DNA-binding domain and its relationship to other families had not been defined.
The structure of the NFAT, Fos and Jun DNA-binding domains on the IL-2 distal ARRE shows an extended NFAT-AP-1 interface and a continuous groove recognizing 15 bp, explaining why the composite element binds cooperatively.
Still open then: How NFAT binds sites that have no adjacent AP-1 site, and whether it can dimerize, were not addressed.
TonEBP (NFAT5) is cloned as a Rel-like protein that binds the tonicity-responsive enhancer and activates transcription in response to hypertonicity, extending the family beyond calcium signalling.
NFAT1 mutants that bind DNA normally but cannot recruit Fos-Jun still induce TNF and IL-13 but not IL-2, IL-3, IL-4, GM-CSF, MIP1a or Fas ligand, separating AP-1-dependent and AP-1-independent NFAT programmes in one cell type.
DNA-bound TonEBP/NFAT5 is an obligate dimer that adopts an NF-kB-like structure and completely encircles its DNA target, giving a direct structural link between the NFAT and NF-kB families.
NFAT1 binds the HIV-1 LTR kB site cooperatively as a dimer whose asymmetric interface differs from the symmetric Rel dimer, showing that the calcium-regulated NFATs, previously viewed as monomeric binders, can also occupy kB-like sites as dimers.
Still open then: How often NFAT dimers form on genomic sites, as opposed to viral and model elements.
Still unresolved
- Whether NFAT prefers a methylated CpG immediately 5' of its core is unresolved: the literature and ChIP-derived models read TGGAAA, but several HOCOMOCO models built from methyl-HT-SELEX for NFATC1-4 and NFAT5 read ACGGAAA, consistent with a 5-methylcytosine standing in for the thymine, and whether methylated CGGAAA sites are bound in cells has not been shown in the work cited here.
- Which binding mode NFAT uses at a given genomic site - a monomer on a lone GGAAA, a cooperative complex with Fos-Jun, or a dimer on a kB-like site - cannot be predicted from sequence: all three are structurally established, but their genome-wide shares are unknown.
This is a short, low-information motif and that dominates everything else. Six informative bases make chance matches overwhelmingly more common than real sites, so an isolated TGGAAA should not be reported as an NFAT site; require a composite (usually AP-1 within about 15 bp), a tandem or inverted pair, or independent occupancy data. No model separates NFATC1-4, and NFAT5 only weakly, so a hit is family-level. Several ChIP-derived NFATC1 and NFATC2 models in these databases include the AP-1 TGAGTCA alongside the NFAT core: that is the partner's motif riding along, and it inflates apparent specificity. Finally, occupancy requires calcium signalling, so matches in an unstimulated cell are mostly inactive.
References
- Jain J, McCaffrey PG, Miner Z, et al. The T-cell transcription factor NFATp is a substrate for calcineurin and interacts with Fos and Jun. Nature 1993;365:352-5. PMID 8397339
- Rooney JW, Sun YL, Glimcher LH, et al. Novel NFAT sites that mediate activation of the interleukin-2 promoter in response to T-cell receptor stimulation. Mol Cell Biol 1995;15:6299-310. PMID 7565783
- Rao A, Luo C, Hogan PG. Transcription factors of the NFAT family: regulation and function. Annu Rev Immunol 1997;15:707-47. PMID 9143705
- Chen L, Glover JN, Hogan PG, et al. Structure of the DNA-binding domains from NFAT, Fos and Jun bound specifically to DNA. Nature 1998;392:42-8. PMID 9510247
- Miyakawa H, Woo SK, Dahl SC, et al. Tonicity-responsive enhancer binding protein, a rel-like protein that stimulates transcription in response to hypertonicity. Proc Natl Acad Sci U S A 1999;96:2538-42. PMID 10051678
- Macián F, García-Rodríguez C, Rao A. Gene expression elicited by NFAT in the presence or absence of cooperative recruitment of Fos and Jun. EMBO J 2000;19:4783-95. PMID 10970869
- Stroud JC, Lopez-Rodriguez C, Rao A, et al. Structure of a TonEBP-DNA complex reveals DNA encircled by a transcription factor. Nat Struct Biol 2002;9:90-4. PMID 11780147
- Giffin MJ, Stroud JC, Bates DL, et al. Structure of NFAT1 bound as a dimer to the HIV-1 LTR kappa B element. Nat Struct Biol 2003;10:800-6. PMID 12949493
- Hogan PG, Chen L, Nardone J, et al. Transcriptional regulation by calcium, calcineurin, and NFAT. Genes Dev 2003;17:2205-32. PMID 12975316
- Go WY, Liu X, Roti MA, et al. NFAT5/TonEBP mutant mice define osmotic stress as a critical feature of the lymphoid microenvironment. Proc Natl Acad Sci U S A 2004;101:10673-8. PMID 15247420
- Macian F. NFAT proteins: key regulators of T-cell development and function. Nat Rev Immunol 2005;5:472-84. PMID 15928679
- Bates DL, Barthel KK, Wu Y, et al. Crystal structure of NFAT bound to the HIV-1 LTR tandem kappaB enhancer element. Structure 2008;16:684-94. PMID 18462673
- Shaw JP, Utz PJ, Durand DB, et al. Identification of a putative regulator of early T cell activation genes. Science 1988;241:202-5. PMID 3260404
29 motifs in this family — JASPAR 5 · HOCOMOCO 15 · CIS-BP 5 · Vierstra 4
| motif | database | consensus | bits |
|---|---|---|---|
| M03449_2.00 | CIS-BP | gGgAAAatwatTTTCCAT | 20.9 |
| AC0680:NFATC:Rel | Vierstra | TTTCCattaayGGAA | 17.9 |
| NFAT5.H14CORE.1.PS.APSA | HOCOMOCO | gTGGAAAATTaC | 17.2 |
| AC0683:NFATC:Rel | Vierstra | TTTCCATGGAAA | 16.9 |
| NFAC2.H14CORE.4.M.BMB | HOCOMOCO | TTtCcntaACGGAAAat | 16.5 |
| NFAC1.H14CORE.2.SM.BSMB | HOCOMOCO | TTtCnnywACGGAAAAt | 16.4 |
| NFAC2.H14CORE.2.S.CSC | HOCOMOCO | tTtCcnyaATGGAAAaT | 15.7 |
| NFAC1.H14CORE.0.P.BPB | HOCOMOCO | TGGAAAaanTGAgTCA | 13.7 |
| NFAC1.H14CORE.3.SM.BSMB | HOCOMOCO | aCGGAAAdTtC | 13.1 |
| NFAC3.H14CORE.0.SM.BSMB | HOCOMOCO | AcGGAAAat | 13 |
| AC0681:NFAT:Rel | Vierstra | GTAmtTTTCCA | 13 |
| NFAC2.H14CORE.0.P.BPB | HOCOMOCO | TGGAAAaanTgAgTca | 12.8 |
| NFAT5.H14CORE.0.SM.BSMB | HOCOMOCO | AcGGAAAattaC | 12.6 |
| NFAC4.H14CORE.1.SM.BSMB | HOCOMOCO | ACGGAAA | 12.4 |
| NFAC1.H14CORE.1.PS.APSA | HOCOMOCO | ATGGAAAat | 12.2 |
| M05709_2.00 | CIS-BP | GTAcntTTCCac | 11.9 |
| NFATC4 (MA1525.3) | JASPAR | aTGGAAAat | 11.8 |
| NFAC2.H14CORE.3.M.CMC | HOCOMOCO | ACGGAAA | 11.7 |
| Nfat5 (MA0606.3) | JASPAR | ATGGAAAA | 11.2 |
| Nfatc2 (MA0152.3) | JASPAR | AATGGAAA | 11.1 |
| NFAC4.H14CORE.0.P.CPC | HOCOMOCO | GGAAAAttt | 11.1 |
| Nfatc1 (MA0624.3) | JASPAR | TGGAAA | 10.7 |
| NFAC2.H14CORE.1.PS.APSA | HOCOMOCO | aaTGGAAAaa | 10.4 |
| M01511_2.00 | CIS-BP | TTTCCG | 9.4 |
| NFAC3.H14CORE.1.P.CPC | HOCOMOCO | aGGAAAa | 9.1 |
| AC0682:NFATC/ZNF:Rel | Vierstra | TTTCCa | 8.9 |
| M02440_2.00 | CIS-BP | TTTCCa | 8.8 |
| NFATC3 (MA0625.3) | JASPAR | tGGAAA | 8.1 |
| M01306_2.00 | CIS-BP | tttcca | 2.7 |
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.