ABL1 notes

No notes file existed for this COMPLETE review; this file was created for the
section below. Automated deep research was not run for this section; it is based
on papers retrieved from PubMed, cached with just fetch-pmid, and read.

Premetazoan origin (ORIGINS_OF_MULTICELLULARITY, 2026-10-01)

Sources read

PMID:26090675 (Aleem, Craddock and Miller 2015, PLoS One), full text cached.
The only biochemical characterisation of a premetazoan Abl (MbAbl2, from the
choanoflagellate Monosiga brevicollis). This is the key paper.
PMID:18621719 (Manning et al. 2008, PNAS), full text cached. The M. brevicollis
tyrosine kinome.
PMID:22550341 (Suga et al. 2012, Sci Signal), abstract only. The tyrosine kinomes of
the filastereans Capsaspora owczarzaki and Ministeria vibrans.
PMID:18273011 (King et al. 2008, Nature), full text cached. The M. brevicollis
genome paper.
PMID:27746046 (Sebé-Pedrós et al. 2016, Dev Cell), abstract only. Capsaspora
phosphoproteomics.
PubMed searches for Abl in choanoflagellates, Capsaspora, ichthyosporeans,
sponges or cnidarians found no other primary study of a non-bilaterian Abl
("(Monosiga OR choanoflagellate OR Capsaspora) AND Abl" returned only PMID:26090675).

Abl family distribution

Abl is one of four cytoplasmic tyrosine kinase families with clear
choanoflagellate orthologs
PMID:18621719.
The cytoplasmic tyrosine kinase repertoire was already in place before
filastereans split from choanoflagellates and animals
PMID:22550341.
Abl itself is found in choanoflagellates, filastereans and ichthyosporeans, but
these unicellular Abls lack the long C-terminal region
PMID:26090675.
M. brevicollis has two Abl paralogs, MbAbl1 and MbAbl2
PMID:26090675.
Background: the choanoflagellate genome has the full phosphotyrosine signalling
toolkit (kinase, PTP and SH2 domains)
PMID:18273011.

What MbAbl2 does (in vitro and in mammalian cells)

Active tyrosine kinase with the Abl substrate preference
PMID:26090675.
Activation-loop autophosphorylation is conserved
PMID:26090675.
No N-terminal myristoyl/cap autoinhibition, and constitutively active
PMID:26090675.
In mouse NIH3T3 cells it phosphorylates many proteins and transforms cells more
potently than c-Abl; it is poorly inhibited by imatinib, which binds the
inactive conformation
PMID:26090675.
No F-actin binding domain
PMID:26090675.
The authors conclude that the F-actin- and DNA-binding C-terminus is an animal
addition
PMID:26090675.
Caveats stated by the authors:
The second paralog, MbAbl1, has a possible N-terminal myristoylation glycine and
has not been tested
PMID:26090675.
The cellular role of Abl in unicellular holozoans is unknown
PMID:26090675.
Substrates of MbAbl2 in M. brevicollis were not identified. The transformation
and phospho-blot data are from mouse cells, so they say nothing about native
substrates.
Capsaspora phosphoproteomics shows that tyrosine kinases are among the proteins
whose phosphorylation changes across life stages
PMID:27746046.
Only the abstract was read, and it does not name Abl.

Core functions of the existing review: ancestral or animal-specific

Core function (existing review)	Classification	Evidence
GO:0004715 non-membrane spanning protein tyrosine kinase activity	Ancestral (choanoflagellate, filasterean and ichthyosporean Abl present; MbAbl2 shown active with Abl-type substrate preference and autophosphorylation)	PMID:26090675, PMID:18621719, PMID:22550341
Tight autoinhibition (myristoyl/cap clamp; not a GO core function but stated in the description)	Animal-specific, at least for MbAbl2; MbAbl1 untested	PMID:26090675
...directly_involved_in GO:0032956 regulation of actin cytoskeleton organization	Unresolved. No unicellular Abl has been tested for an actin role. The direct actin-binding route through the C-terminal domain is absent in all premetazoan Abls, so any ancestral actin role would have to go through substrates	PMID:26090675
...directly_involved_in GO:0006281 DNA repair	Unresolved / probably animal-specific. No data from unicellular holozoans. The DNA-binding C-terminus is absent	PMID:26090675
GO:0051015 actin filament binding	Animal-specific. No F-actin binding domain in any premetazoan Abl examined	PMID:26090675
GO:0003677 DNA binding, with DNA damage response and apoptotic process	Animal-specific for DNA binding (C-terminal DNA-binding region absent). DNA damage response unresolved. The apoptotic machinery ABL1 feeds into (TP73 and others) is animal; its age was not tested here	PMID:26090675
Locations nucleus, cytosol, cytoskeleton, plasma membrane	Unresolved. No localisation data for any unicellular Abl. The ABL1 NLS/NES and myristoyl membrane anchor sit in regions that MbAbl2 lacks; this inference is mine and is not stated in the paper	none

Summary: the ancestral Abl was an SH3-SH2-kinase module with Abl-type substrate
specificity and activation-loop autophosphorylation. Animals added two things to
it: tight N-terminal autoinhibition, and the long C-terminus with F-actin and DNA
binding. Both of ABL1's best-known cellular roles, direct actin cytoskeleton
remodelling and the nuclear DNA damage response, depend on the animal additions.
No paper tests what Abl does in a unicellular holozoan. The repository's SRC and
CSK reviews reach the same conclusion for Src: the catalytic module is ancestral,
and stringent negative regulation is a later addition (disputed for Csk-Src).
This fits the general model the Miller lab proposes for Src, Abl and Pak.

Action changes

None. No row's rationale depends on premetazoan evidence. The IBA
GO:0005886 plasma membrane row cites myristoylation of isoform IB. MbAbl2 has
no myristoylation site, but that does not bear on whether human ABL1 is at the
plasma membrane.

Track C: IBA node reach into unicellular holozoans (QuickGO, 2026-10-01)

Query: annotation/search?withFrom=PANTHER:<PTN>&taxonId=<T>&taxonUsage=descendants,
T = 28009 Choanoflagellata, 2687318 Filasterea, 127916 Ichthyosporea.

PTN node (ABL1 IBA terms)	Choanoflagellata	Filasterea	Ichthyosporea
PTN002521457 (GO:0004713 protein tyrosine kinase activity; GO:0005886 plasma membrane)	105 rows on 59 M. brevicollis (taxon 81824) proteins: GO:0004713 IBA x46, GO:0005886 IBA x59	0	0
PTN002815481 (GO:0007229 integrin-mediated signaling pathway; GO:0010595 positive regulation of endothelial cell migration)	0	0	0
PTN008405112 (GO:0007173 epidermal growth factor receptor signaling pathway)	0	0	0

No animal tissue, organ or developmental process term from an ABL1 IBA node
reaches a unicellular holozoan. The endothelial-migration and integrin terms sit on
PTN002815481, which reaches no choanoflagellate, filasterean or ichthyosporean
protein.
PTN002521457 is a broad tyrosine kinase node, not an Abl node. Its MF transfer to
M. brevicollis is consistent with the biochemistry above. Its CC transfer
(plasma membrane) reaches all 59 proteins. That includes cytoplasmic kinases, and
for unicellular Abl there is no localisation evidence at all. This is a possible
localisation over-reach worth flagging in the propagation audit; it is not a
multicellular process term.
No Salpingoeca rosetta, Capsaspora or ichthyosporean protein receives any of
these IBAs, even though UniProt places S. rosetta and Capsaspora tyrosine
kinases in PTHR24418. This probably reflects which proteomes are in the PANTHER
tree; I did not check that.

2026-10-24 ORGANELLES batch 2 (mitochondrion rows)

Rows on GO:0005739: IEA GO_REF:0000044 (UniProt keyword SL-0173), previously
KEEP_AS_NON_CORE; NAS PMID:24522549 (ParkinsonsUK-UCL), previously ACCEPT. The two
actions were inconsistent. No sub-compartment rows.

Checked:
- UniProt location is By similarity [file:human/ABL1/ABL1-uniprot.txt "Localizes to
mitochondria in response to oxidative stress (By similarity)."]; the FUNCTION text
reads "ABL1 targets to mitochondria in response to oxidative stress and thereby
mediates mitochondrial dysfunction and cell death."
- The primary papers (fetched, abstract-only): PMID:11350980 and PMID:11509666.
- PMID:24522549 is an abstract-only review that lists c-Abl among PINK1-associated
proteins in mitochondrial networks; the old supporting_text was the title only and
was replaced with abstract text.
- ABL1 has no presequence and no TM segment; it is a cytoplasmic/nuclear kinase
recruited under stress. No cached paper resolves a sub-compartment; an outer-surface
location is plausible but not shown, and the ER-to-mitochondria translocation could
not be separated from MAM from the abstract alone.

Outcome: root. Both rows KEEP_AS_NON_CORE (NAS changed from ACCEPT for consistency);
reasons record the check; PMID:11350980 and PMID:11509666 added to references.