ZDHHC23 is a multipass membrane protein S-acyltransferase of the DHHC family (PANTHER subfamily PTHR22883:SF475). It transfers palmitate from palmitoyl-CoA onto cysteine residues of substrate proteins through an autoacylated cysteine in its DHHC cysteine-rich domain (C283 in this horse product). In mammalian cells ZDHHC23 resides mainly in Golgi and trans-Golgi network membranes. With ZDHHC22, it palmitoylates the intracellular S0-S1 loop of the BK potassium channel alpha subunit KCNMA1. This palmitoylation is needed for efficient exit of the channel from the trans-Golgi network and for its delivery to the plasma membrane. Rodent ZDHHC23 (NIDD) also binds the PDZ domain of neuronal nitric-oxide synthase through a C-terminal -EDIV motif. The 448-residue horse product described here is the Ensembl canonical gene model. Its C-terminal tail comes from an alternative last exon and lacks the -EDIV motif.
Summary: Golgi membrane is the best-supported location for ZDHHC23 and the compartment where it acts on its known substrate.
Reason: In HEK293 cells, ZDHHC23 co-localizes mainly with the Golgi marker GM130 and, more weakly, with the TGN marker TGN38. The KCNMA1 palmitoylation it controls gates channel exit from the TGN. The horse model keeps all of the multipass transmembrane architecture of the subfamily (RESULTS-rereview.md). The horse UniProt SubCell line behind this IEA is an ARBA prediction and is not used as support. This location rests on ortholog data within PTHR22883:SF475.
zDHHC23 being predominantly localized at the Golgi apparatus and TGN, in agreement with previous studies (3, 4, 21), whereas the thioesterases were predominantly cytosolic (12, 20), supports a role for dynamic regulation of BK channel palmitoylation at the level of the TGN.
Summary: ER residence is inherited from the deep DHHC-family node (PTN000525256). The direct ZDHHC23 data place the enzyme mainly in Golgi and TGN.
Reason: The PAINT node is the root of the DHHC family. Its donors are paralogs outside the ZDHHC23 subfamily, such as yeast Erf2 (SGD:S000004236), and the ZDHHC23 orthologs are not among them. Where ZDHHC23 has been examined it sits mainly at the Golgi and TGN, with no reported ER pool. That is not evidence of loss, because a multipass enzyme passes through the ER and DHHC enzymes often distribute across ER and Golgi. ER activity is therefore kept as a family-level, non-core location rather than removed.
Summary: Golgi activity is correct for ZDHHC23. The integral-membrane enzyme is better described by Golgi membrane, which this product already carries.
Reason: The family-level IBA agrees with the direct ZDHHC23 co-localization data (GM130 and TGN38). ZDHHC23 is a multipass integral membrane enzyme that modifies cytoplasmic cysteines of membrane cargo, so the specific compartment is the Golgi membrane (GO:0000139), which is part_of the Golgi apparatus and is already annotated to this product. The term is refined rather than kept, as the repo guidance asks for correct but general terms.
Propagation Review
Root cause:TERM SCOPING PROBLEM
Failure modes:GRANULARITY MISMATCH
Sources checked:
PANTHER:PTN000525256SUPPORTS TRANSFER
DHHC-family root node. Golgi residence is compatible with the direct ZDHHC23 co-localization data, so the propagation is sound. Only the granularity is refined, to the membrane of the compartment.
Summary: S-palmitoylation-dependent membrane targeting is inherited from the DHHC-family root node. For ZDHHC23 the characterized effect is plasma-membrane delivery of KCNMA1, captured by GO:0072659.
Reason: The donors are yeast Erf2 and rice DHHC9, whose substrates gain membrane association through palmitoylation. The DHHC enzyme does part of the work of that process because it installs the lipid anchor. For ZDHHC23, the substrate KCNMA1 is already an integral membrane protein, and palmitoylation of its S0-S1 loop governs exit from the TGN and surface delivery, not membrane insertion. The family-level term is plausibly true of ZDHHC23 activity on other substrates but has no ZDHHC23-specific support. The ZDHHC23-specific process is the separately annotated GO:0072659, which is not a descendant of this term, so the term is kept as non-core rather than refined.
Furthermore, we reveal reversible protein palmitoylation is a critical step in controlling exit of BK channels from the trans-Golgi network (TGN) and thus a major checkpoint for regulating cell surface expression of the channel.
Summary: Correct but general. The DHHC domain specifies protein-cysteine S-palmitoylation.
Reason: The InterPro DHHC signature (IPR001594) and Pfam PF01529 cover residues 251-380. The horse product keeps the DHHC motif with the active-site cysteine at C283, which corresponds to human C289. The ortholog activity is S-palmitoylation of protein cysteines. The specific term GO:0019706, which is already annotated to this product, should replace the generic parent.
Summary: Correct but very general. The acyl transfer is S-palmitoylation of protein cysteines.
Reason: The UniRule (DHHC palmitoyltransferase family) supports DHHC acyltransferase membership. The ZDHHC23 subfamily activity is protein-cysteine S-palmitoylation, so the term should be refined to GO:0019706, which this product already carries.
Summary: Core activity. It is inherited from the DHHC-family root node, whose donors include human ZDHHC23 itself, and the horse product keeps the catalytic DHHC cysteine.
Reason: PAINT placed protein-cysteine S-palmitoyltransferase activity at the DHHC-family root (PTN000525256). Human ZDHHC23 (Q8IYP9) is among the experimentally annotated descendants, and its appearance as a donor reflects its own experimental grounding. A0A9L0T4E4 sits in the same PANTHER subfamily as human ZDHHC23 (SF475) and keeps the DHHC motif with C283 and the full cysteine-rich domain, so there is no sign of catalytic loss.
Summary: The same core activity, also assigned by combined automated methods (EC 2.3.1.225 via UniRule).
Reason: This agrees with the IBA row and with the conserved DHHC catalytic domain. The UniRule family assignment is valid membership-level support, and the activity is experimentally established for the human subfamily member.
We identify zDHHC22 and zDHHC23 as the palmitoyltransferases and LYPLA1 and LYPLAL1 as acyl thioesterases controlling BK channel palmitoylation.
GO:0072659 protein localization to plasma membrane
IBA GO_REF:0000033
ACCEPT
Summary: This is the ZDHHC23-specific process. It is inherited from the vertebrate ZDHHC23 node PTN008583751, whose donors are human and mouse ZDHHC23.
Reason: The PAINT node is narrow. Its donors are human ZDHHC23 (Q8IYP9, IMP) and mouse Zdhhc23 (MGI:2685625), so this is ortholog-clade inheritance within SF475, not a paralog transfer. ZDHHC23 does part of the work of the process because it palmitoylates the S0-S1 loop of KCNMA1, the step that licenses TGN exit and surface delivery. siRNA depletion reduced, and overexpression increased, surface expression of the palmitoylation reporter. The horse product keeps the catalytic and transmembrane core that this activity needs.
First, we demonstrate that zDHHC22 and zDHHC23 specifically control palmitoylation of the S0–S1 loop of BK channels to control surface trafficking.
file:human/ZDHHC23/ZDHHC23-uniprot.txt
Palmitoyltransferase that mediates
Core Functions
Protein-cysteine S-palmitoyltransferase acting at Golgi and trans-Golgi network membranes. It palmitoylates the intracellular S0-S1 loop of the BK channel subunit KCNMA1 and so promotes exit of the channel from the TGN and its localization to the plasma membrane.
Q: Is the A0A9L0T4E4 alternative last exon, which encodes a 49-residue tail without the -EDIV PDZ-binding motif, supported by horse transcript data? Do horse tissues instead mainly express the -EDIV isoform found in the other Ensembl models (A0A3Q2KPV4, A0A9L0R7E2, A0A9L0SCZ1)?
Q: Does the C-terminal tail (-EDIV, GLLG or the alternative horse tail) change ZDHHC23 substrate choice, for example KCNMA1 compared with nNOS-associated targets?
Suggested Experiments
Experiment: Express horse ZDHHC23 isoforms carrying the A0A9L0T4E4 tail or the -EDIV tail in HEK293 cells with the KCNMA1 S0-S1-YFP reporter. Measure acyl-biotin exchange palmitoylation and surface expression, and use a DHHS (C283S) mutant as the catalytic-dead control.
External Prediction Reviews
These computational predictions are reviewed separately from the GOA annotation set used for this review. The assessments below are from this project and do not constitute official GO annotations or endorsement by GO/UniProt. They are not included in the existing annotation review above.
Assessment of 1 ProtNLM GO predictions for the selected horse ZDHHC23 protein, using mammalian experimental findings and an explicit comparison of the horse sequence.
Prediction method: ProtNLM2 · Version: UniProt API snapshot 2026-09-08
Review rationale: Human ZDHHC23-dependent KCNMA1 palmitoylation is experimentally supported (PMID:22399288). The horse sequence retains the DHHC catalytic cysteine (human289/horse283), with 91.07% identity over 98.53% of the human sequence. Transfer of protein-cysteine S-palmitoyltransferase activity is justified and agrees with the exact horse IBA/IEA term. This judgment rests on substrate experiments and sequence conservation, not ARBA agreement. Training-set membership is unknown.
Supporting Evidence:
file:human/ZDHHC23/ZDHHC23-uniprot.txt: "CC -!- FUNCTION: Palmitoyltransferase that could catalyze the addition of CC palmitate onto various protein substrates and be involved in a variety CC of cellular processes (Probable). Palmitoyltransferase that mediates CC palmitoylation of KCNMA1, regulating localization of KCNMA1 to the CC plasma membrane. May be involved in NOS1 regulation and targeting to CC the synaptic membrane. {ECO:0000269|PubMed:22399288, CC ECO:0000305|PubMed:22399288}."
file:HORSE/ZDHHC23/ZDHHC23-bioinformatics/RESULTS.md: "The horse sequence A0A9L0T4E4 (448 residues) aligns to human Q8IYP9 (409 residues) with 91.07% identity across 403 paired residues. Paired coverage is 98.53% of the human sequence and 89.96% of the horse sequence."
PMID:22399288: "Here, we demonstrate that palmitoylation of the intracellular S0-S1 loop of BK channels is controlled by two of the 23 mammalian palmitoyl-transferases, zDHHC22 and zDHHC23. Palmitoylation by these acyl transferases is essential for efficient cell surface expression of BK channels."