Generated by rgs_gap_residues.py (run python rgs_gap_residues.py to regenerate;
a fresh run must reproduce this file and results.json exactly).
UniProt annotates two RGS domains in AKAP10 (FT DOMAIN 125..369 "RGS 1",
379..505 "RGS 2", both ECO:0000255). An RGS domain name implies
GTPase-activator activity on heterotrimeric Galpha subunits, which is a catalytic
claim. This tests whether the residues that make an RGS box a Galpha GAP are
present in AKAP10.
Each domain is aligned (BLOSUM62, global, gap -11/-1) to the RGS box of human
RGS4 (P49798, residues 68-184). Five RGS4 positions that the
RGS4:Gi-alpha-1 structure and mutagenesis identify as Galpha transition-state
contacts are scored: S85, E87, N88, N128, R167.
RGS4 Asn128 is decisive — N128A abolishes GAP activity without unfolding the
domain. A site counts as retained only when the alignment column is a true
aligned pair (no gap on either side) and the residue is identical.
Controls run in the same pass: RGS16 and RGS9 (established Galpha GAPs, must
score 5/5 or the script aborts) and AXIN1 (an RGS fold that is not a Galpha
GAP).
| domain | role | identity vs RGS4 box | contacts retained | residue aligned to RGS4 N128 | N128 retained |
|---|---|---|---|---|---|
| AKAP10 RGS-A | subject | 30.3% | 0/5 | P315 | no |
| AKAP10 RGS-B | subject | 24.3% | 0/5 | G452 | no |
| RGS16 | positive control (Galpha GAP) | 56.6% | 5/5 | N131 | yes |
| RGS9 | positive control (Galpha GAP) | 38.0% | 5/5 | N367 | yes |
| AXIN1 | negative control (RGS fold, no Galpha GAP) | 31.5% | 0/5 | gap | no |
Both AKAP10 RGS domains score 0/5, exactly like the AXIN1 negative control
and unlike both positive controls, and the decisive asparagine is replaced by
proline in RGS-A and glycine in RGS-B. The fold is present; the Galpha catalytic
machinery is not.
Caveat stated rather than hidden: RGS-B sits at 24.3% identity, below the range
where a residue-level call is safe on alignment alone. RGS-A at 30.3% is directly
comparable to the AXIN1 control at 31.5%, and the RGS-A N128->Pro substitution is
independently disruptive. The computational result therefore corroborates, and
does not replace, the two published statements: that not all of the charged
residues conserved in RGS proteins are conserved in D-AKAP2 (PMID:9326583), and
that "D-AKAP2 has so far demonstrated no specific interactions with heterotrimeric
G proteins, and no ability to activate them" (PMID:19797056).
No GTPase-activator (GO:0005096) or Galpha-binding annotation is warranted for
AKAP10 from its RGS domains. GOA carries none, so this is recorded as a
hypothesis that did not confirm rather than as a defect found.