Please note: this site relies heavily on the use of javascript. Without a javascript-enabled browser, this site will not function correctly. Please enable javascript and reload the page, or switch to a different browser.
209  structures 800  species 5  interactions 18302  sequences 911  architectures

Family: PHD (PF00628)

Summary: PHD-finger

Pfam includes annotations and additional family information from a range of different sources. These sources can be accessed via the tabs below.

This is the Wikipedia entry entitled "PHD finger". More...

PHD finger Edit Wikipedia article

PDB 1f62 EBI.jpg
PHD zinc finger. Zinc atoms shown in grey
Symbol PHD
Pfam PF00628
Pfam clan CL0390
InterPro IPR019787
SCOP 1f62
OPM superfamily 62
OPM protein 1vfy

The PHD finger was discovered in 1993 as a Cys4-His-Cys3 motif in the plant homeodomain (hence PHD) proteins HAT3.1 in Arabidopsis thaliana and maize ZmHox1a. The PHD finger motif resembles the metal binding RING domain (Cys3-His-Cys4) and FYVE domain. It occurs as a single finger, but often in clusters of two or three, and it also occurs together with other domains, such as the chromodomain and the bromodomain.

Role in epigenetics

The PHD finger, approximately 50-80 amino acids in length, is found in more than 100 human proteins. Several of the proteins it occurs in are found in the nucleus, and are involved in chromatin-mediated gene regulation. The PHD finger occurs in proteins such as the transcriptional co-activators p300 and CBP, Polycomb-like protein (Pcl), Trithorax-group proteins like ASH1L, ASH2L and MLL, the autoimmune regulator (AIRE), Mi-2 complex (part of histone deacetylase complex), the co-repressor TIF1, the JARID1-family of demethylases and many more.


The NMR structure of the PHD finger from human WSTF (Williams Syndrome Transcription Factor) shows that the conserved cysteines and histidine coordinate two Zn2+ ions. In general, the PHD finger adopts a globular fold, consisting of a two-stranded beta-sheet and an alpha-helix. The region consisting of these secondary structures and the residues involved in coordinating the zinc-ions are very conserved among species. The loop regions I and II are variable and could contribute functional specificity to the different PHD fingers.


Recently the PHD fingers of some proteins, including ING2, YNG1 and NURF, have been reported to bind to histone H3 tri-methylated on lysine 4 (H3K4me3), while other PHD fingers have tested negative in such assays. Interestingly, a protein called SMCX (or JARID1C) has a PHD finger, which has been reported to bind histone H3 tri-methylated lysine 9 (H3K9me3). Based on these recent publications, binding to tri-methylated lysines on histones may therefore be a property widespread among PHD fingers. Domains that bind to modified histones, are called epigenetic readers as they specifically recognize the modified version of the residue and binds to it. The modification H3K4me3 is associated with the transcription start site of active genes, while H3K9me3 is associated with inactive genes. The modifications of the histone lysines are dynamic, as there are methylases that add methyl groups to the lysines, and there are demethylases that remove methyl groups. The SMCX protein is actually a histone H3 lysine 4 demethylase, which means it is an enzyme that can remove the methyl groups of lysine 4 on histone 3 (making it H3K4me2 or H3K4me1). One can only speculate if the H3K9me3-binding of SMCX PHD domain provides a crosstalk between trimethylation of H3K9 and the demethylation of H3K4me3. Such crosstalks have been suggested earlier with other domains involved in chromatin regulation, and may provide a strictly coordinated regulation.

Another example is the PHD finger of the BHC80/PHF21A protein, which is a component of the LSD1 complex. In this complex, LSD1 specifically demethylates H3K4me2 to H3K4me0, and BHC80 binds H3K4me0 through its PHD finger to stabilize the complex at its target promoters, presumably to prevent further re-methylation. This is the first example of a PHD finger recognizing lysine methyl-zero status.


  • Schindler U, Beckmann H et al. (1993) "HAT3-1, a novel Arabidopsis homeodomain protein containing a conserved cystein-rich region", Plant J, 4, 137-150
  • Aasland R, Gibson T, Stewart AF (1995) "The PHD finger: implications for chromatin-mediated transcriptional regulation", TIBS, 20, 56-59
  • Pascual J, Martinez-Yamout M et al. (2000)," Structure of the PHD Zinc Finger from Human Williams-Beuren Syndrome Transcription Factor", Journal of Molecular Biology, 304, 723-729
  • Peña PV, Davrazou1 F, Shi X et al. (2006),"Molecular mechanism of histone H3K4me3 recognition by plant homeodomain of ING2", Nature, 442, 100 - 103
  • Li H, Ilin S, Wang W et al. (2006),"Molecular basis for site-specific read-out of histone H3K4me3 by the BPTF PHD finger of NURF", Nature, 442, 91-95
  • Iwase S, Lan F (2007),"The X-Linked Mental Retardation Gene SMCX/JARID1C Defines a Family of Histone H3 Lysine 4 Demethylases", Cell, 128, 1-12
  • Lan F, Collins RE (2007), "Recognition of unmethylated histone H3 lysine 4 links BHC80 to LSD1-mediated gene repression." Nature, 448, 718-22

This page is based on a Wikipedia article. The text is available under the Creative Commons Attribution/Share-Alike License.

This tab holds the annotation information that is stored in the Pfam database. As we move to using Wikipedia as our main source of annotation, the contents of this tab will be gradually replaced by the Wikipedia tab.

PHD-finger Provide feedback

PHD folds into an interleaved type of Zn-finger chelating 2 Zn ions in a similar manner to that of the RING and FYVE domains [2]. Several PHD fingers have been identified as binding modules of methylated histone H3 [3].

Literature references

  1. Aasland R, Gibson TJ, Stewart AF; , Trends Biochem Sci 1995;20:56-59.: The PHD finger: implications for chromatin-mediated transcriptional regulation. PUBMED:7701562 EPMC:7701562

  2. Pascual J, Martinez-Yamout M, Dyson HJ, Wright PE; , J Mol Biol 2000;304:723-729.: Structure of the PHD zinc finger from human williams-beuren syndrome transcription factor PUBMED:11124022 EPMC:11124022

  3. Shi X, Kachirskaia I, Walter KL, Kuo JH, Lake A, Davrazou F, Chan SM, Martin DG, Fingerman IM, Briggs SD, Howe L, Utz PJ, Kutateladze TG, Lugovskoy AA, Bedford MT, Gozani O; , J Biol Chem. 2006; [Epub ahead of print]: Proteome-wide analysis in S. cerevisiae identifies several PHD fingers as novel direct and selective binding modules of histone H3 methylated at either lysine 4 or lysine 36. PUBMED:17142463 EPMC:17142463

Internal database links

External database links

This tab holds annotation information from the InterPro database.

InterPro entry IPR019787

Zinc finger (Znf) domains are relatively small protein motifs which contain multiple finger-like protrusions that make tandem contacts with their target molecule. Some of these domains bind zinc, but many do not; instead binding other metals such as iron, or no metal at all. For example, some family members form salt bridges to stabilise the finger-like folds. They were first identified as a DNA-binding motif in transcription factor TFIIIA from Xenopus laevis (African clawed frog), however they are now recognised to bind DNA, RNA, protein and/or lipid substrates [PUBMED:10529348, PUBMED:15963892, PUBMED:15718139, PUBMED:17210253, PUBMED:12665246]. Their binding properties depend on the amino acid sequence of the finger domains and of the linker between fingers, as well as on the higher-order structures and the number of fingers. Znf domains are often found in clusters, where fingers can have different binding specificities. There are many superfamilies of Znf motifs, varying in both sequence and structure. They display considerable versatility in binding modes, even between members of the same class (e.g. some bind DNA, others protein), suggesting that Znf motifs are stable scaffolds that have evolved specialised functions. For example, Znf-containing proteins function in gene transcription, translation, mRNA trafficking, cytoskeleton organisation, epithelial development, cell adhesion, protein folding, chromatin remodelling and zinc sensing, to name but a few [PUBMED:11179890]. Zinc-binding motifs are stable structures, and they rarely undergo conformational changes upon binding their target.

This entry represents the PHD (homeodomain) zinc finger domain [PUBMED:7701562], which is a C4HC3 zinc-finger-like motif found in nuclear proteins thought to be involved in chromatin-mediated transcriptional regulation. The PHD finger motif is reminiscent of, but distinct from the C3HC4 type RING finger.

The function of this domain is not yet known but in analogy with the LIM domain it could be involved in protein-protein interaction and be important for the assembly or activity of multicomponent complexes involved in transcriptional activation or repression. Alternatively, the interactions could be intra-molecular and be important in maintaining the structural integrity of the protein. In similarity to the RING finger and the LIM domain, the PHD finger is thought to bind two zinc ions.

Gene Ontology

The mapping between Pfam and Gene Ontology is provided by InterPro. If you use this data please cite InterPro.

Domain organisation

Below is a listing of the unique domain organisations or architectures in which this domain is found. More...

Loading domain graphics...

Pfam Clan

This family is a member of clan zf-FYVE-PHD (CL0390), which has the following description:

Superfamily contains a number of zinc-fingers, of the FYVE/PHD type, which are found in several groups of proteins including myelin-associated oligodendrocytic basic proteins (MOBP) Rabphilins, melanophilins, exophilins and myosin-VIIA and Rab-interacting protein families.

The clan contains the following 10 members:

FYVE FYVE_2 PHD PHD_2 PHD_Oberon RAG2_PHD zf-HC5HC2H zf-HC5HC2H_2 zf-PHD-like zf-piccolo


We store a range of different sequence alignments for families. As well as the seed alignment from which the family is built, we provide the full alignment, generated by searching the sequence database (reference proteomes) using the family HMM. We also generate alignments using four representative proteomes (RP) sets, the UniProtKB sequence database, the NCBI sequence database, and our metagenomics sequence database. More...

View options

We make a range of alignments for each Pfam-A family. You can see a description of each above. You can view these alignments in various ways but please note that some types of alignment are never generated while others may not be available for all families, most commonly because the alignments are too large to handle.

Representative proteomes UniProt
Jalview View  View  View  View  View  View  View  View  View 
HTML View                 
PP/heatmap 1                

1Cannot generate PP/Heatmap alignments for seeds; no PP data available

Key: ✓ available, x not generated, not available.

Format an alignment

Representative proteomes UniProt

Download options

We make all of our alignments available in Stockholm format. You can download them here as raw, plain text files or as gzip-compressed files.

Representative proteomes UniProt
Raw Stockholm Download   Download   Download   Download   Download   Download   Download   Download   Download  
Gzipped Download   Download   Download   Download   Download   Download   Download   Download   Download  

You can also download a FASTA format file containing the full-length sequences for all sequences in the full alignment.

HMM logo

HMM logos is one way of visualising profile HMMs. Logos provide a quick overview of the properties of an HMM in a graphical form. You can see a more detailed description of HMM logos and find out how you can interpret them here. More...


This page displays the phylogenetic tree for this family's seed alignment. We use FastTree to calculate neighbour join trees with a local bootstrap based on 100 resamples (shown next to the tree nodes). FastTree calculates approximately-maximum-likelihood phylogenetic trees from our seed alignment.

Note: You can also download the data file for the tree.

Curation and family details

This section shows the detailed information about the Pfam family. You can see the definitions of many of the terms in this section in the glossary and a fuller explanation of the scoring system that we use in the scores section of the help pages.

Curation View help on the curation process

Seed source: Prosite
Previous IDs: none
Type: Domain
Author: Pascual J, Bateman A
Number in seed: 74
Number in full: 18302
Average length of the domain: 50.00 aa
Average identity of full alignment: 29 %
Average coverage of the sequence by the domain: 5.85 %

HMM information View help on HMM parameters

HMM build commands:
build method: hmmbuild -o /dev/null HMM SEED
search method: hmmsearch -Z 26740544 -E 1000 --cpu 4 HMM pfamseq
Model details:
Parameter Sequence Domain
Gathering cut-off 28.8 28.8
Trusted cut-off 28.8 28.8
Noise cut-off 28.7 28.7
Model length: 52
Family (HMM) version: 28
Download: download the raw HMM for this family

Species distribution

Sunburst controls


Weight segments by...

Change the size of the sunburst


Colour assignments

Archea Archea Eukaryota Eukaryota
Bacteria Bacteria Other sequences Other sequences
Viruses Viruses Unclassified Unclassified
Viroids Viroids Unclassified sequence Unclassified sequence


Align selected sequences to HMM

Generate a FASTA-format file

Clear selection

This visualisation provides a simple graphical representation of the distribution of this family across species. You can find the original interactive tree in the adjacent tab. More...

Loading sunburst data...

Tree controls


The tree shows the occurrence of this domain across different species. More...


Please note: for large trees this can take some time. While the tree is loading, you can safely switch away from this tab but if you browse away from the family page entirely, the tree will not be loaded.


There are 5 interactions for this family. More...

JmjC PHD Bromodomain Bromodomain RRM_1


For those sequences which have a structure in the Protein DataBank, we use the mapping between UniProt, PDB and Pfam coordinate systems from the PDBe group, to allow us to map Pfam domains onto UniProt sequences and three-dimensional protein structures. The table below shows the structures on which the PHD domain has been found. There are 209 instances of this domain found in the PDB. Note that there may be multiple copies of the domain in a single PDB structure, since many structures contain multiple copies of the same protein sequence.

Loading structure mapping...