Endogenous mouse Adissp is present in brown-adipocyte conditioned medium, and a Flag knock-in at the native locus confirms the endogenous protein is secreted.

Secretion is non-classical. Brefeldin A plus monensin, which blocks the ER-to-Golgi route, increased Adissp secretion while inhibiting adiponectin in the same experiment.

The human protein was itself shown to be secreted: adenoviral expression of human ADISSP in mouse liver put it into the circulation at about 0.45 ug/mL.

The receptor is inferred from binding but never identified; the authors describe it as putative throughout.

The link to a heterotrimeric G protein rests on a single melittin experiment; melittin is a membrane-active bee-venom peptide rather than a selective Gs inhibitor.

Adipose-specific knockout impairs white-fat browning and produces diet-induced obesity and hyperglycaemia.

The thermogenic phenotype persists at thermoneutrality, so the process regulated is not specifically cold-induced thermogenesis.

To purify secreted protein the authors had to graft a heterologous signal peptide onto the construct, which is independent evidence that the native protein has none.

States that ADISSP is secreted by human as well as mouse adipocytes and is present in human circulation, which is human-specific support for the extracellular region rows.

Endogenous Adissp is required for systemic glucose homeostasis.

Recombinant full-length protein activates Akt and AS160 dose-dependently from 25 nM, identifying a second, insulin-independent signalling output.

The activity is a property of the protein rather than of the tag or the preparation: untagged recombinant protein reproduces it.

Human SGBS adipocytes respond to the protein, so the responding system exists in human cells.

C20orf27/ADISSP co-immunoprecipitates with the PP1 catalytic subunit, in an assay using two transfected tagged plasmids rather than endogenous protein.

The proposed mechanism is competition: binding PP1c reduces its association with the regulatory subunit GADD34.

The NF-kappaB readouts are canonical markers only - phospho-IKK, phospho-IkappaB and phospho-p65/RelA - measured in both the overexpression and the silencing direction. NIK, p100/p52 processing and RelB are not assayed anywhere in the paper.

Silencing in two further lines reduces the same canonical markers, giving the reciprocal direction.

Human and mouse ADISSP are the same length and 90.2% identical with no indels, computed at run time, so the ISS and Ensembl Compara transfers are well licensed. The comparator carries a self-identity positive control and refuses unequal-length input rather than truncating.

Neither orthologue has a SIGNAL peptide feature, so the leaderless topology is a shared property of the pair rather than an assumption carried across.

Four of ADISSP's thirteen distinct IntAct protein partners are PP1-module proteins, and all three catalytic subunits are recovered, which is why the GO term is correctly left isoform-agnostic. The denominator counts protein partners only: of 27 IntAct records, 26 are interactions of the ADISSP protein and one pairs the ADISSP mRNA with a miRNA, which is excluded rather than counted as a partner.

The null that makes the seven-publication count interpretable: PP1 catalytic subunits are affinity-proteomics hubs, so recurrence in tag pulldowns is weak on its own and only the subject-centric count is informative.