ligD (PA2138, Q9I1X7) notes

Identity

Swiss-Prot entry, 840 aa, "Multifunctional non-homologous end joining protein
LigD". Three autonomous domains: N-terminal 3'-phosphoesterase (PE, 7-162),
central ATP-dependent ligase (219-521, active-site Lys238 N6-AMP), C-terminal
polymerase/primase (Pol, 549-793). NCBIfam TIGR02776 NHEJ_ligase_prk, PANTHER
PTHR42705:SF2. [file:PSEAE/ligD/ligD-uniprot.txt "N6-AMP-lysine intermediate"]

Literature (Shuman laboratory; all abstracts only except where noted)

Zhu and Shuman 2005, PMID:15520014. Purified LigD; autonomous polymerase and
ligase domains; templated primer extension and non-templated addition.
PMID:15520014
PMID:15520014
Zhu and Shuman 2005, PMID:15897197. PE domain has 3'-ribonuclease and
3'-phosphatase activities, Mn2+-dependent.
PMID:15897197
Zhu and Shuman 2010, PMID:20018881. Gap filling; Ku stimulates LigD and
promotes plasmid end joining by full-length LigD.
PMID:20018881
PMID:20018881

The ligase activity itself is not described in the 15520014 abstract beyond
"autonomous DNA ligase domain"; UniProt (curated, ECO:0000269 PubMed:15520014)
records that the Lig domain "functions as an independent domain", and the EC
6.5.1.1 nick-sealing activity of PaeLigD is well established. The IDA is
accepted (defer to curator; abstract-only).

PseudoCAP IBA-seed assessment

Seeds from .cache/paint/IBD.gaf; Mtb LigD (P9WNV3) experimental rows from
QuickGO 2026-11-05.

PTN node	node term	n_seeds	leaf IBAs	seed annotation (term, evidence, PMID)	verdict	rationale	other seeds	IBD stands without this seed?
PTN001627042	GO:0003910 DNA ligase (ATP) activity	2	15 (10 taxa)	GO:0003910, IDA, PMID:15520014 (PseudoCAP; same row from UniProt)	SOUND	Purified PaeLigD with an autonomous ATP-dependent ligase domain; Ku-stimulated end joining (PMID:20018881). Exact activity, suits a LigD clade.	M. tuberculosis LigD P9WNV3 (GO:0003910 IDA PMID:12215643, 16023671, 16476729)	Yes
PTN001627042	GO:0003887 DNA-directed DNA polymerase activity	2	15 (10 taxa)	GO:0003887, IDA, PMID:15520014 (PseudoCAP; same row from UniProt)	SOUND	Templated extension of a primer-template by the purified Pol domain, abolished by D669A/D671A.	P9WNV3 (GO:0003887 IDA PMID:15499016)	Yes

Other IBDs at this node that ligD receives but does not seed:
- GO:0006303 NHEJ (seed P9WNV3 only): sound, and true of PaeLigD (PMID:20018881). ACCEPT.
- GO:0006269 DNA replication, synthesis of primer (seed P9WNV3 only, MTBBASE
IDA PMID:16023671): LigD Pol has in vitro primase-like activity, but LigD acts
in NHEJ DSB repair, not chromosomal replication. The node term does not suit
the clade. MARK_AS_OVER_ANNOTATED with a GO_CENTRAL CHECK (now PROBABLE
after the tree check below; the Mtb paper was still not read).

No PSEUDOCAP defects. One GO_CENTRAL defect (GO:0006269, PROBABLE).

PANTHER tree check (2026-11-05)

Topology of PTN001627042 in PTHR42705 fetched with
projects/PSEUDOCAP/panther_tree.py node PTN001627042 --depth 5. The node is a
SPECIATION node (LUCA), 15 leaves, PTHR42705:SF2 12 + SF3 3.

Child node	Event	Leaves	Subfamilies	Contents	Verdict
PTN005127089	SPECIATION (Eukaryota)	2	SF2	Monosiga MONBRDRAFT_36321 "DNA ligase D", Gossypium LOC107941224 "DNA ligase (ATP)"	LigD-type ligase
PTN000048875	SPECIATION (Eubacteria)	13	SF2 10, SF3 3	the two clades below	LigD-type ligases
PTN008085115	SPECIATION (Terrabacteria)	7	SF2 4, SF3 3	PTN000048869 (DUPLICATION, 6: Streptomyces SCO5308/SCO6709/SCO6498 + Mtb LigD P9WNV3), B. subtilis LigD O34398	NHEJ LigD + Streptomyces ligase paralogs
PTN005127091	SPECIATION (Proteobacteria)	6	SF2	Bradyrhizobium bll6773; PTN005127092 (Gammaproteobacteria, 5: Coxiella CBU_1934, two Xanthomonas ligases, and PTN005127093 HORIZ_TRANSFER with Ricinus RCOM_0053280 and PAO1 LigD Q9I1X7)	NHEJ LigD

Verdict: the clade is homogeneous - every leaf is a LigD-type or
ATP-dependent DNA ligase, the two fully named ones (Mtb, B. subtilis) are
bifunctional NHEJ proteins, and there is no replicative primase or DnaG-type
enzyme anywhere below the node. So the concern is not a node-placement problem
that a narrower clade would solve: GO:0006269 is the wrong kind of term for the
whole family, and the fix is to remove the IBD rather than move it. The two
activity IBDs at the same node (GO:0003910, GO:0003887) and the GO:0006303 NHEJ
IBD do suit the clade and already cover what the primase assay shows.

Confidence raised SUSPECTED -> PROBABLE: the tree is primary data and settles
that relocation is not the remedy, but whether the Mtb seed paper
(PMID:16023671) describes something genuinely replicative was not checked, and
only reading it would settle that.

Review decisions

ACCEPT: DNA ligase (ATP) activity (IBA, IDA, IEA); DNA-directed DNA
polymerase activity (IBA, IDA); NHEJ (IBA).
MODIFY: RNA exonuclease activity -> 3'-5'-RNA exonuclease activity
(GO:0000175; removes the 3'-terminal ribonucleoside).
NEW: polynucleotide 3'-phosphatase activity (GO:0046403), IDA PMID:15897197.
TOO_GENERAL: DNA repair (NHEJ accepted); nucleotidyltransferase activity
and DNA biosynthetic process (entailed by DNA-directed DNA polymerase activity).
KEEP_AS_NON_CORE: ATP binding.
MARK_AS_OVER_ANNOTATED: DNA recombination (IEA), DNA replication synthesis
of primer (IBA).