A0A8B8L1Z3 (Abrus precatorius LOC113860185) notes

2026-10-20 — Full re-review, independent of the ProtNLM2 predictions

This re-review did not open or use any ProtNLM2 prediction file.

UniProt A0A8B8L1Z3 is an unreviewed TrEMBL entry (PE 4) of 474 aa, from RefSeq
XP_027348664.1 (gene LOC113860185).
It is not a fragment or an isoform. Its length matches the full-length Arabidopsis
SF19 relative EIP9/SDJ1 (At5g64360, 464 aa), and the J domain (70-135) sits at the
same N-terminal position.
GOA has no annotations for this accession, so existing_annotations holds only the
proposed NEW rows.

PANTHER 19.0 places it in PTHR45496:SF19.
The only Arabidopsis member of SF19 is At5g64360 (Q9FMF3). The same gene is called
EIP9 in PMID:21700722, DNAJ1 in PMID:30523112 and SDJ1 in PMID:30589221.
Paralog subfamilies: SDJ2 (At2g01710) is in SF1, SDJ3 (At1g62970) in SF10, and
ADMETOS (At4g11940) in SF32.
PAINT node PTN002896004 (taxon:3398, flowering plants) is seeded from all four of
these subfamilies, so the Abrus protein descends from it. The node carries:
nucleus (GO:0005634);
GO:0001228 DNA-binding transcription activator activity, RNA polymerase II-specific.
Global alignment identity of the query to its relatives:
SF19 members: 0.46 to EIP9/SDJ1 and 0.53 to the Populus proteins;
other subfamilies: 0.17-0.36.
See A0A8B8L1Z3-bioinformatics/RESULTS-rereview.md.
The J domain has no HPD motif (...LALLLEPHRNPF...). EIP9/SDJ1, SDJ3 and ADMETOS also
lack HPD; SDJ2 is the only one that has it.

Harris et al. 2018 PMID:30523112: SUVH1 and SUVH3 bind methylated DNA and
recruit DNAJ1/DNAJ2 ["DNAJ1 and DNAJ2 lack any discernible methyl-binding domain, but they
are robustly associated with SUVH1 and SUVH3"].
Zhao et al. 2019 PMID:30589221.
Park et al. 2011 [PMID:21700722 "EMF1-EIP9 heterodimers were localized in the nucleus"
(abstract wording: "EMF1-EIP1, EMF1-EIP6 and EMF1-EIP9 heterodimers were localized in the
nucleus")]: eip9 mutants flower early. This is an organismal phenotype and is not
transferred.
Miao et al. 2021 PMID:34185870: sdj1/2/3 triple mutants are hypomethylated at a
subset of RdDM sites. This shows the SDJs are needed for methylation there, not that
they take part in it, so no RdDM process term is proposed.

Description rewritten. The old text claimed a "conserved HPD motif" (it is absent) and
a cytoplasmic or ER-excluded Hsp70 cochaperone role. It also leaned on the IPR053052
"Imprinting Balance Regulator" name, which is an unchecked LLM-generated InterPro
description derived from ADMETOS.
Core function changed:
before: protein folding chaperone / protein folding / cytoplasm;
after: transcription coactivator activity (GO:0003713) / positive regulation of
DNA-templated transcription / nucleus.
Added NEW rows (ISS, relying on the PTN002896004 placement) for nucleus and for
transcription coactivator activity.
The coactivator term deliberately replaces PAINT's GO:0001228 (DNA-binding
transcription activator activity). SDJ proteins have no DNA-binding domain and reach DNA
only when SUVH1/SUVH3 recruit them. SDJ2 already has an IDA for GO:0003713. This is
raised as a suggested question for the PAINT node.
Not transferred:
flowering time (an organismal phenotype);
gene silencing by RdDM (necessity-only evidence);
SUVH1/SUVH3 binding (bare protein binding).
The deep-research (falcon) report is generic J-domain material and missed the SDJ
literature. It is kept in the references but no longer used as support.

interpro/panther/PTHR45496/PTHR45496-review.yaml lists the target as the SF19
representative and leaves functional coherence UNKNOWN.
No disagreement about the subfamily placement.
The family file's preferred name "ADMETOS-related" puts ADMETOS (SF32) first. The
characterized function relevant to SF19 is the SUVH-SDJ transcriptional activation of
SDJ1, so that summary could be improved. The family file was not edited.

No module in modules/ covers the SUVH-SDJ complex or RdDM-linked anti-silencing.