Other C2H2 zinc fingers

The remainder of the C2H2 zinc-finger class: factors that are neither KRAB zinc-finger proteins nor members of a family with its own entry. Ninety-six named genes appear here, from well-studied cases such as ZNF335, ZNF341, ZFAT and ZFTA to dozens of ZNF numbers with almost no literature. What they share is the fold, and little else: the 168 matrices in this entry are mutually unrelated, with a cohesion of 0.38 and a median length of 12 bp. The evidence behind them is nonetheless much better than for the KRAB group, with 39 of 93 HOCOMOCO models at the top grade and many supported in vitro. A hit here identifies one named protein at best, never a family motif.

How this was worked out

1. 2009mediumcuration of sequence and expression data19274049

   A manually curated census of 1,391 human transcription factors established C2H2 zinc fingers as the largest class and documented how few of them had been studied individually, which is the condition this entry still records.

   Still open then: Binding specificities were unknown for the great majority of the class.

2. 2012highhuman genetics, null and conditional mouse knockouts, RNA interference, postmortem human tissue23178126

   ZNF335 was identified as a cause of severe microcephaly with small somatic size and neonatal death, shown to be required for neural progenitor self-renewal, and placed in a vertebrate-specific H3K4-methylation complex that directly regulates REST/NRSF, the first clear disease mechanism for a member of this leftover group.

3. 2014highprotein-binding microarrays, inference by DNA-binding-domain identity25215497

   Measuring preferences for more than 1,000 factors across 131 eukaryotes showed that closely related DNA-binding domains almost always share motifs, which is the basis for the cross-species inferred models that fill out CIS-BP, including the mouse-derived entries in this family.

   Still open then: How far identity can be pushed before a transferred motif becomes wrong was set by convention, not by measurement.

4. 2015highbacterial one-hybrid, protein-binding microarray, ChIP-seq25690854

   A recognition code derived from thousands of natural C2H2 domains, with protein-binding microarray and ChIP data, supplied motifs for a large share of the proteins in this entry and showed that the class recognizes more distinct motifs than all other human transcription factors combined.

5. 2016mediumChIP-seq, affinity purification-mass spectrometry, knockdown expression profiling27852650

   Direct functional analysis of 131 C2H2 proteins produced motif models for 78 that had none and found that their co-activator and co-repressor partners are nearly as diverse as their motifs, so members of this group share neither a site nor an effector mechanism.

6. 2018highhuman genetics, patient cells, reporter and binding assays29907691

   Loss-of-function mutations in the previously uncharacterized ZNF341 were shown to cause a recessive hyper-IgE syndrome, with ZNF341 binding the STAT3 promoter and its loss abolishing STAT3 autoinduction, demonstrating that an unstudied zinc finger can carry a specific, non-redundant regulatory role.

7. 2026highmultiple in vitro and in vivo binding assays; ChIP-seq in living cells42557334 42557260

   Systematic characterization gave motifs for 177 of 332 poorly characterized human factors, while in-cell profiling of 166 of them found about half at promoters and enhancers, frequently co-localizing with each other at the same conserved motif matches, and the rest at unique repeat-rich sites in closed chromatin.

   Still open then: Co-localization at shared motif matches leaves open which factor makes the DNA contact at any given occupied site.

Still unresolved

• Whether the short GC-rich models in this entry, such as those for ZNF740, ZNF395, SLC2A4RG, ZXDA and ZXDB, represent genuinely distinct specificities or one degenerate GC-box read by many proteins is unresolved: each has its own in vitro model, yet they cannot be separated from one another or from the SP/KLF and MAZ motifs by sequence.
• The simple-repeat models for ZNF646 and ZNF865, both (TG)n, carry grade-A evidence including in vitro assays, so it is unclear whether these proteins truly recognize dinucleotide microsatellites or whether repeat-rich bound regions dominate both the in vitro selection and the ChIP peaks.
• The JASPAR BNC2 matrix is the canonical AP-1 heptamer at 0.99 similarity, and nothing published settles whether basonuclin 2 occupies AP-1 sites through its own fingers, through co-binding with a bZIP partner, or whether the model is simply mislabelled.
• Because about half of the poorly characterized factors surveyed in cells converge on the same conserved motif matches, it is not known for most members here whether an occupied site reflects direct recognition by the named protein or co-binding with another factor at the same element.

References

• Vaquerizas JM, Kummerfeld SK, Teichmann SA, et al. A census of human transcription factors: function, expression and evolution. Nat Rev Genet 2009;10:252-63. PMID 19274049
• Lambert SA, Jolma A, Campitelli LF, et al. The Human Transcription Factors. Cell 2018;172:650-665. PMID 29425488
• Schmitges FW, Radovani E, Najafabadi HS, et al. Multiparameter functional diversity of human C2H2 zinc finger proteins. Genome Res 2016;26:1742-1752. PMID 27852650
• Najafabadi HS, Mnaimneh S, Schmitges FW, et al. C2H2 zinc finger proteins greatly expand the human regulatory lexicon. Nat Biotechnol 2015;33:555-62. PMID 25690854
• Weirauch MT, Yang A, Albu M, et al. Determination and inference of eukaryotic transcription factor sequence specificity. Cell 2014;158:1431-1443. PMID 25215497
• Yang YJ, Baltus AE, Mathew RS, et al. Microcephaly gene links trithorax and REST/NRSF to control neural stem cell proliferation and differentiation. Cell 2012;151:1097-112. PMID 23178126
• Béziat V, Li J, Lin JX, et al. A recessive form of hyper-IgE syndrome by disruption of ZNF341-dependent STAT3 transcription and activity. Sci Immunol 2018;3. PMID 29907691
• Parker M, Mohankumar KM, Punchihewa C, et al. C11orf95-RELA fusions drive oncogenic NF-κB signalling in ependymoma. Nature 2014;506:451-5. PMID 24553141
• Jolma A, Laverty KU, Fathi A, et al. An expanded codebook of human transcription factor DNA-binding specificity. Nature 2026;657:275-283. PMID 42557334
• Razavi R, Fathi A, Yellan I, et al. Extensive binding of poorly characterized human transcription factors to genomic dark matter. Nat Commun 2026;17. PMID 42557260
• Vorontsov IE, Eliseeva IA, Zinkevich A, et al. HOCOMOCO in 2024: a rebuild of the curated collection of binding models for human and mouse transcription factors. Nucleic Acids Res 2024;52:D154-D163. PMID 37971293

The factors, elsewhere

Other C2H2 zinc finger families