MNN14 (YJR061W, P40355) — curation notes

Journal for the AI GO-annotation review of Saccharomyces cerevisiae MNN14.
Understudied ("dark") gene. Primary deliverable is a rigorous knowledge_gaps section.
Every assertion below carries inline provenance.

Identity (verified)

• UniProt: P40355 (MNN14_YEAST), 935 aa, MW 108,427.
• SGD: S000003822; systematic/ordered-locus name YJR061W; ORF name J1736.
• RefSeq NP_012595.1.
• Protein name in UniProt: "Mannosyltransferase regulator 14"
  [ECO:0000303|PubMed:28101612]. The name was coined in the 2017 glyco-engineering
  paper; it does NOT itself establish a mannosyltransferase catalytic activity.

Family / domains — IMPORTANT attribution point

The task brief anticipated a GT15/MNN1 alpha-1,3-mannosyltransferase. This is not what
the record shows. MNN14 is NOT in the MNN1/GT15 family.

• UniProt SIMILARITY: "Belongs to the MNN4 family." [UniProt P40355, "SIMILARITY: Belongs to the MNN4 family."]
• Pfam: PF04991 (LicD) — a nucleotidyl-/phosphotransferase-type domain
  [UniProt P40355, "Pfam; PF04991; LicD; 1."].
• InterPro: IPR007074 (LicD/FKTN/FKRP nucleotidyltransferase) and
  IPR009644 (FKTN/MNN4/W02B3.4-1)
  [UniProt P40355, "InterPro; IPR007074; LicD/FKTN/FKRP_NTP_transf." and
  "InterPro; IPR009644; FKTN/MNN4/W02B3.4-1."].
• PANTHER: PTHR15407 (FUKUTIN-RELATED)
  [UniProt P40355, "PANTHER; PTHR15407; FUKUTIN-RELATED; 1."].

So the fold is the LicD/fukutin-related nucleotidyltransferase superfamily, shared by MNN4
and the metazoan fukutin (FKTN)/FKRP ribitol-phosphate transferases — a phosphotransferase-type
architecture, distinct from the KRE2/MNT1/GT15 mannosyltransferases.

Topology (type II Golgi membrane protein)

• TOPO_DOM 1..21 Cytoplasmic; TRANSMEM 22..42 (signal-anchor, type II); TOPO_DOM 43..935 Lumenal.
  [UniProt P40355 FT lines]. Consistent with a Golgi-lumen-acting glycan-modifying protein.
• KW: Golgi apparatus; Membrane; Signal-anchor; Transmembrane.

DXD motif (catalytic-motif reasoning)

• MOTIF 498..500 "DXD" [ECO:0000250|UniProtKB:P36044] — inferred by similarity to MNN4 (P36044).
• UniProt DOMAIN comment: "The conserved DXD motif is essential for the function, which could
  be an indication that MNN14 has transferase activity." [UniProt P40355,
  "The conserved DXD motif is essential for the function, which could be an indication that MNN14 has transferase activity."]
• DXD motifs coordinate a divalent cation and the nucleotide-sugar in many GT-A / phosphotransferase
  enzymes. Its presence is consistent with, but does NOT prove, a catalytic transferase role for
  MNN14 — see knowledge gap. The same wording is used for MNN4, whose curated MF is regulator,
  not transferase (below).

Function — KNOWN vs NOT known

KNOWN (experimentally, PMID:28101612 — abstract-only in cache; IGI in GOA)

• MNN14 is an MNN4 paralog required for full N-glycan mannosylphosphorylation in
  S. cerevisiae. PMID:28101612
• Partial redundancy with MNN4: single deletions leave residual mannosylphosphate; the
  MNN4+MNN14 double deletion abolishes N-glycan mannosylphosphorylation.
  PMID:28101612
• Biotechnology relevance: eliminating mannosylphosphate (an och1Δmnn1Δmnn4Δmnn14Δ strain) is a
  step toward human-compatible glycoproteins in yeast. [PMID:28101612 abstract]
• INDUCTION: expression is repressed by RIM101 [UniProt P40355, "Expression is repressed by RIM101."],
  from PMID:12509465 (Lamb & Mitchell 2003; Rim101 represses NRG1/SMP1). This is a regulatory-network
  observation, not a molecular function.
• Localization: Golgi apparatus membrane (ECO:0000305; by SubCell + topology; IEA in GOA).
  [UniProt P40355, "SUBCELLULAR LOCATION: Golgi apparatus membrane"]

NOT known (the real gaps)

• Molecular function is undetermined. MNN14 has NO curated MF term other than the ND root.
  The two live hypotheses, neither established for MNN14:
• Enzyme regulator/activator of the mannosylphosphate transferase — the role its paralog
  MNN4 is assigned. In yeast the actual catalytic mannosylphosphate transferase is
  Mnn6/Ktr6 (KRE2/MNT1 family), and MNN4 is its positive regulator whose amount is
  rate-limiting [Odani et al. 1997, PMID:9459307 "Mannosylphosphate transfer to cell wall
  mannan is regulated by the transcriptional level of the MNN4 gene"; Wang et al., MNN6=KTR6
  is the mannosylphosphate transferase]. MNN4's curated MF is enzyme activator activity
  (GO:0008047, IMP:SGD); UniProt (P36044) states "While MNN4 seems to have a regulatory role
  in N-glycan mannosylphosphorylation, a transferase activity of MNN4 cannot be ruled out."
• Transferase activity of its own — suggested only by the conserved DXD motif (by similarity).
• Direct acceptor substrate / exact reaction of MNN14 is unknown (which mannan/glycan position;
  whether it transfers mannose-1-phosphate at all vs. regulates the enzyme that does).
• Basis of the MNN4/MNN14 redundancy is unknown (paralog sub-/neo-functionalization;
  condition-, substrate-, or acceptor-position specificity).
• Standalone loss-of-function phenotype beyond the glyco-profile is uncharacterized; MNN14 is
  non-essential and there is no described growth/stress phenotype for mnn14Δ alone.

GOA annotations to review (5)

1. GO:0009101 glycoprotein biosynthetic process — IBA (GO_REF:0000033); IBA panel includes
   SGD:S000001684 (MNN4). BP is correct (mannosylphosphorylation is glycoprotein biosynthesis);
   generic but defensible. Not the most specific but IBA-appropriate. -> KEEP_AS_NON_CORE / ACCEPT.
2. GO:0000139 Golgi membrane — IEA (SubCell). Supported by topology + SubCell. -> ACCEPT.
3. GO:0006491 N-glycan processing — IGI (PMID:28101612), with SGD:S000001684 (MNN4). This is the
   experimental genetic-interaction annotation matching the double-deletion result. -> ACCEPT (core BP).
4. GO:0003674 molecular_function — ND (root). MF genuinely unknown. -> KEEP_AS_NON_CORE.
5. GO:0005575 cellular_component — ND (root). Superseded by Golgi membrane; but ND placeholder. -> KEEP_AS_NON_CORE.

Note: GO:0006491 "N-glycan processing" is defined as conversion of N-linked glycan to mature form by
glycosidases/glycosyltransferases [OLS GO:0006491]. Mannosylphosphorylation is an N-glycan
outer-chain maturation/modification, so this is an appropriate (if slightly generic) BP.

Candidate MF terms considered (NOT asserted as MNN14 core)

• GO:0000031 mannosylphosphate transferase activity — this is the catalytic activity of Mnn6/Ktr6;
  assigning it to MNN14 would be over-annotation because (a) MNN14's own catalytic activity is unproven
  and (b) its paralog MNN4 is curated as a regulator, not a transferase. Listed only in knowledge_gaps.
• GO:0008047 enzyme activator activity — the MF of the paralog MNN4 (IMP:SGD); a plausible-by-orthology
  hypothesis for MNN14 but not experimentally shown for MNN14. Listed only in knowledge_gaps.

References gathered

• PMID:28101612 — Kim et al. 2017, Appl Microbiol Biotechnol. Primary experimental (abstract-only cache).
  The single functional paper directly on MNN14; source of IGI GO:0006491.
• PMID:12509465 — Lamb & Mitchell 2003 (RIM101 represses NRG1/SMP1). Source of the INDUCTION note;
  MNN14 mentioned as a Rim101-repressed target. Secondary/regulatory context.
• PMID:9459307 — Odani et al. 1997, FEBS Lett. MNN4 = positive regulator of mannosylphosphorylation;
  establishes the MNN4-family regulator paradigm (background for the MF gap).
• UniProt:P40355 — domain/family/topology/DXD evidence.
• UniProt:P36044 — MNN4 paralog record (regulator MF; DXD ambiguity) for attribution.

Web verification log

• rest.uniprot.org P40355 (full record downloaded to MNN14-uniprot.txt): family=MNN4, Pfam LicD,
  DXD 498-500, type II Golgi, FUNCTION = "role in N-glycan mannosylphosphorylation... partially
  redundant with MNN4."
• rest.uniprot.org P36044 (MNN4): MF enzyme activator activity (IMP:SGD); "seems to have a
  regulatory role... transferase activity cannot be ruled out."
• WebSearch (Odani 1997 PMID:9459307; Wang MNN6=KTR6): MNN6/Ktr6 = the mannosylphosphate transferase;
  MNN4 = its positive regulator, Mnn4p amount rate-limiting.
• OLS: GO:0006491, GO:0009101, GO:0000031, GO:0008047 definitions confirmed.