JMJ22 (JUMONJI 22, At5g06550) is a nuclear Fe(II)/2-oxoglutarate-dependent JmjC-domain protein with an N-terminal F-box, belonging to the JMJD6/JMJD8-related PANTHER family PTHR12480 (subfamily SF21). It acts as a histone arginine demethylase that removes the repressive H4R3me2 mark. Its expression is repressed in darkness or far-red light by the zinc-finger protein SOMNUS and released when phytochrome B is activated by red light. Together with its paralog JMJ20, with which it is genetically redundant, JMJ22 then demethylates histone arginines at the gibberellin biosynthesis genes GA3ox1 and GA3ox2. This raises bioactive gibberellin levels and promotes light-induced seed germination. JMJ22 also has a role that JMJ20 does not share: loss-of-function alleles partially suppress the dwarf phenotype of the thermospermine-deficient acl5 mutant and stabilise mRNAs of the SAC51 family. The mechanism of that role is unknown. No DNA-binding domain is present, and the F-box has not been shown to work in an SCF ubiquitin ligase.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000976 transcription cis-regulatory region binding | IPI PMID:25533953 An Arabidopsis gene regulatory network for secondary cell wa... | MARK AS OVER ANNOTATED | Summary: This comes from a high-throughput yeast one-hybrid (Y1H) screen of secondary cell wall gene promoters against a library of transcription factor clones. JMJ22 was recovered as binding the KOR1 (AT5G49720) promoter. JMJ22 is not discussed in the main text, and the interaction was not validated. Reason: JMJ22 has only an F-box and a JmjC/cupin demethylase domain, with no recognisable DNA-binding domain. Its characterised function is enzymatic, at chromatin (H4R3me2 removal at GA3ox1/GA3ox2). A single unvalidated Y1H hit from a screen that is prone to false positives is weak support for cis-regulatory DNA binding as an enabled molecular function. I do not REMOVE it, because the Y1H readout is real data, but it should not be read as evidence that JMJ22 is a transcription factor. Supporting Evidence: PMID:25533953 against a library of 1,664 full-length Arabidopsis transcription factors file:ARATH/JMJ22/JMJ22-uniprot.txt Histone demethylase that demethylates 'Arg-3' (H4R3me) of histone H4 with a specific activity for H4R3me2 |
| GO:0000987 cis-regulatory region sequence-specific DNA binding | IBA GO_REF:0000033 | MARK AS OVER ANNOTATED | Summary: PAINT IBD at PTN001753861 (PTHR12480). The only seed is JMJ22's own IDA from Cho et al. 2012. Reason: JMJ22 appearing as the donor of its own IBA is expected, not circular. The problem is the underlying claim. JMJ22 lacks a DNA-binding domain, and the demethylase action at GA3ox1/GA3ox2 described by Cho et al. ("at GA3ox1/GA3ox2") is chromatin occupancy, not sequence-specific DNA recognition. This row matches the IDA row it was propagated from (MARK_AS_OVER_ANNOTATED). Because the term is the only MF that PAINT placed at this node, transfers to SF21 orthologs (for example wheat) should not be read as evidence of TF-like DNA binding. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: ROLE CONFLATION Sources checked: PANTHER:PTN001753861 · PTN001753861 SOURCE WEAK OR INFERRED The IBD for GO:0000987 is seeded only by the JMJ22 (AT5G06550) IDA from Cho et al. 2012. The underlying observation is chromatin-level action at GA3ox1/GA3ox2 by a JmjC demethylase that has no DNA-binding domain, and the sequence-specific DNA-binding MF over-states it. The histone demethylase activity from the same donor was not placed at this node. Supporting Evidence: PMID:22483719 removal of repressive histone arginine methylations at GA3ox1/GA3ox2 file:ARATH/JMJ22/JMJ22-uniprot.txt Histone demethylase that demethylates 'Arg-3' (H4R3me) of histone H4 with a specific activity for H4R3me2 |
| GO:0000987 cis-regulatory region sequence-specific DNA binding | IDA PMID:22483719 Control of seed germination by light-induced histone arginin... | MARK AS OVER ANNOTATED | Summary: TAIR IDA from Cho et al. 2012. The cached record is abstract-only, and the abstract places JMJ22's action at the GA3ox1/GA3ox2 loci; the paper's MeSH terms list chromatin immunoprecipitation and EMSA. The assay behind this annotation could not be inspected. Reason: Occupancy of the GA3ox1/GA3ox2 regulatory regions by JMJ22 is consistent with its role as a histone arginine demethylase acting there. However, the protein (F-box plus JmjC) has no sequence-specific DNA-binding domain, and the abstract says JMJ22 acts by removing histone methylation, not by recognising a DNA sequence. Most likely JMJ22 is recruited to these loci, so "sequence-specific DNA binding" over-states the molecular function. The full text was not readable, so I flag the row as an over-annotation rather than removing it. Supporting Evidence: PMID:22483719 removal of repressive histone arginine methylations at GA3ox1/GA3ox2 |
| GO:0005634 nucleus | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT IBD at PTN001753861, seeded by JMJ22 itself and two other family members. Reason: This agrees with the experimental nuclear localisation of JMJ22 and with its chromatin function as a histone demethylase. Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt SUBCELLULAR LOCATION: Nucleus {ECO:0000269|PubMed:22483719}. |
| GO:0005634 nucleus | IDA PMID:22483719 Control of seed germination by light-induced histone arginin... | ACCEPT | Summary: Experimental nuclear localisation of JMJ22 (Cho et al. 2012, curated by UniProt and TAIR). Reason: Nuclear localisation is required for its role in demethylating histone H4R3me2 at target loci. Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt SUBCELLULAR LOCATION: Nucleus {ECO:0000269|PubMed:22483719}. |
| GO:0005634 nucleus | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular location mapping, which rests on the experimental IDA. Reason: This is consistent with the experimental nuclear localisation. Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt SUBCELLULAR LOCATION: Nucleus {ECO:0000269|PubMed:22483719}. |
| GO:0005634 nucleus | ISM GO_REF:0000122 | ACCEPT | Summary: AtSubP sequence-based localisation prediction. Reason: The prediction agrees with the experimental nuclear localisation. Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt SUBCELLULAR LOCATION: Nucleus {ECO:0000269|PubMed:22483719}. |
| GO:0010030 positive regulation of seed germination | IGI PMID:22483719 Control of seed germination by light-induced histone arginin... | ACCEPT | Summary: This genetic interaction with JMJ20 (AT5G63080) rests on the jmj20 jmj22 double mutant, which has reduced germination after phyB activation by a red-light pulse. Reason: This is the best-established biological role of JMJ22. JMJ20 and JMJ22 act redundantly, so the phenotype is seen in the double mutant. JMJ22 contributes directly by demethylating H4R3me2 at GA3ox1/GA3ox2, which raises GA levels that promote germination. Supporting Evidence: PMID:22483719 JMJ22, act redundantly as positive regulators of seed germination file:ARATH/JMJ22/JMJ22-uniprot.txt Plants missing both JMJ20 and JMJ22 exhibit reduced seed germination efficiency during PHYB activation after red light (R)-pulse treatment |
| GO:0010030 positive regulation of seed germination | IMP PMID:22483719 Control of seed germination by light-induced histone arginin... | ACCEPT | Summary: Mutant analysis showing that JMJ22 promotes light-induced seed germination (Cho et al. 2012). Reason: This is consistent with the IGI row. JMJ22 promotes germination downstream of phyB by derepressing the GA biosynthesis genes GA3ox1/GA3ox2. Supporting Evidence: PMID:22483719 JMJ22, act redundantly as positive regulators of seed germination |
| GO:0010099 regulation of photomorphogenesis | IMP PMID:22483719 Control of seed germination by light-induced histone arginin... | KEEP AS NON CORE | Summary: TAIR IMP from Cho et al. 2012. The abstract describes only phyB-dependent seed germination. Any seedling photomorphogenesis assay would be in the full text, which was not readable. Reason: Light-induced germination downstream of phyB is a light-controlled developmental response, so the curator's annotation is plausible. I defer to the curator, who read the full text. The term is peripheral to the core chromatin and germination function and is better captured by positive regulation of seed germination. Supporting Evidence: PMID:22483719 upon PHYB activation, JMJ20/JMJ22 are derepressed |
| GO:0010114 response to red light | IEP PMID:22483719 Control of seed germination by light-induced histone arginin... | KEEP AS NON CORE | Summary: JMJ22 expression is induced by red light: SOMNUS represses it when phyB is inactive, and the repression is lifted when phyB is activated. Reason: The red-light induction is well documented, but it describes how JMJ22 is regulated rather than what JMJ22 does. Non-core. Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt Repressed by the zinc-finger protein SOMNUS when PHYB is inactive in far-red (FR) conditions, but derepressed upon PHYB activation by red light (R). |
| GO:0010114 response to red light | IMP PMID:22483719 Control of seed germination by light-induced histone arginin... | KEEP AS NON CORE | Summary: The jmj20 jmj22 double mutant has reduced germination after a red-light pulse. Reason: JMJ22 links red-light/phyB perception to GA biosynthesis in seeds, so the term is true. The specific output is captured by positive regulation of seed germination. Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt Plants missing both JMJ20 and JMJ22 exhibit reduced seed germination efficiency during PHYB activation after red light (R)-pulse treatment |
| GO:0010476 gibberellin mediated signaling pathway | IMP PMID:22483719 Control of seed germination by light-induced histone arginin... | MODIFY | Summary: Cho et al. show that JMJ20/JMJ22 raise gibberellin levels by derepressing the GA biosynthetic genes GA3ox1 and GA3ox2. Reason: The evidence places JMJ22 upstream of gibberellin biosynthesis (GA3ox1/GA3ox2 expression and endogenous GA levels), not within GA signal transduction (GA perception, DELLA degradation). The precise process is positive regulation of gibberellin biosynthetic process. Any effect on GA signalling is indirect, through hormone levels. Proposed replacements: positive regulation of gibberellin biosynthetic process Supporting Evidence: PMID:22483719 increased gibberellic acid levels through the removal of repressive histone arginine methylations at GA3ox1/GA3ox2 file:ARATH/JMJ22/JMJ22-uniprot.txt impaired H4R3me2 removal-dependent derepression of GA3ox1 and GA3ox2 causing lower endogenous gibberellic acid (GA) biosynthesis |
| GO:0032452 histone demethylase activity | IGI PMID:22483719 Control of seed germination by light-induced histone arginin... | MODIFY | Summary: Cho et al. 2012 identify JMJ20/JMJ22 as histone arginine demethylases. UniProt curates H4R3me2-specific demethylation and the RHEA:58352 reaction (dimethyl-Arg to monomethyl-Arg) as experimental (ECO:0000269) from this paper. Reason: The activity is right, but the experimentally supported specificity is for histone H4 Arg3 (H4R3me2), so the more specific child term GO:0033749 histone H4R3 demethylase activity is warranted. The JmjC domain keeps the Fe(II)-binding triad (H324, D326, H407), consistent with catalytic competence. PAINT did not propagate this term from JMJ22 at PTN001753861, although it propagated nucleus and DNA binding from the same donor. I take that as a deliberate decision not to treat histone demethylase activity as an inherited SF21-wide function. It is established for JMJ22 itself. Proposed replacements: histone H4R3 demethylase activity Supporting Evidence: file:ARATH/JMJ22/JMJ22-uniprot.txt Histone demethylase that demethylates 'Arg-3' (H4R3me) of histone H4 with a specific activity for H4R3me2 PMID:22483719 the histone arginine demethylases, JMJ20 and JMJ22 |
| GO:0040029 epigenetic regulation of gene expression | IDA PMID:22483719 Control of seed germination by light-induced histone arginin... | ACCEPT | Summary: JMJ22 removes repressive histone arginine methylation (H4R3me2) at GA3ox1/GA3ox2 chromatin, which derepresses these genes. Reason: Removing a repressive histone mark to switch on target genes is a direct chromatin mechanism of gene regulation and is the core process in which JMJ22 acts. Supporting Evidence: PMID:22483719 removal of repressive histone arginine methylations at GA3ox1/GA3ox2 |
Loading supporting content…
Download this section (compressed HTML)Q: How is JMJ22 recruited to GA3ox1/GA3ox2 chromatin, given that it lacks a DNA-binding domain? Is the TAIR "sequence-specific DNA binding" IDA based on ChIP occupancy rather than direct, sequence-specific binding?
Suggested experts: Plant chromatin biologists, TAIR curators
Q: Does the N-terminal F-box of JMJ22 assemble into an SCF E3 ligase (binding ASK/SKP1)? If so, does that function bear on its histone demethylase or thermospermine-related roles?
Suggested experts: Plant ubiquitin-proteasome researchers
Q: What is the molecular basis of the JMJ22-specific (not JMJ20/JMJ21) role in thermospermine signalling and SAC51-family mRNA stability? Is it histone demethylation, or a different JmjC hydroxylase/demethylase activity on RNA-associated proteins?
Suggested experts: Polyamine and vascular development researchers
Q: Is H4R3me2 demethylase activity conserved across the F-box-containing SF21 orthologs (for example in cereals), given that PAINT did not propagate the activity from JMJ22?
Suggested experts: PAINT/PANTHER curators
Experiment: Assay recombinant wild-type JMJ22 and an H324A/D326A mutant on H4R3me2 peptides and nucleosomes by mass spectrometry, and test whether the mutant fails to complement the jmj20 jmj22 germination defect.
Hypothesis: JMJ22 demethylates H4R3me2 directly and requires its Fe(II)-binding triad.
Type: in vitro enzyme assay plus genetic complementation
Experiment: Compare EMSA/DAP-seq of purified JMJ22 with ChIP-seq of tagged JMJ22, and identify chromatin partners by IP-MS.
Hypothesis: JMJ22 is recruited to GA3ox promoters through a partner protein rather than by binding DNA directly.
Type: biochemistry and ChIP-seq
Experiment: Test whether catalytically dead JMJ22 complements sac59/jmj22-2 suppression of acl5, and measure SAC51 mRNA stability and translation.
Hypothesis: The thermospermine-related phenotype of jmj22 reflects a catalytic activity of JMJ22.
Type: genetic complementation
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)