Gfat1 (zeppelin, CG12449) encodes glutamine:fructose-6-phosphate amidotransferase 1, the enzyme that catalyses the first committed step of hexosamine biosynthesis (L-glutamine + D-fructose 6-phosphate to L-glutamate + D-glucosamine 6-phosphate), supplying UDP-N-acetylglucosamine for chitin synthesis and protein glycosylation. Full-length Gfat1 products (e.g. the 694-residue Gfat1-PA) combine an N-terminal glutaminase (class II glutamine amidotransferase, Ntn-hydrolase) domain with a C-terminal pair of SIS sugar-isomerase domains; fly Gfat1 activity is feedback inhibited by UDP-GlcNAc and stimulated by PKA, it is expressed in chitin-synthesising embryonic tissues and larval salivary glands, and null mutants die by the first larval instar. This accession is the Gfat1-PF isoform (434 residues), an N-terminally truncated product that starts at the equivalent of Met261 of Gfat1-PA. It is identical to the C-terminal 434 residues of the full-length protein, keeps both SIS sugar-phosphate-binding domains, and lacks the catalytic Cys2 nucleophile together with all but the last 40 residues of the glutaminase domain. It therefore cannot by itself carry out the glutamine-dependent amidotransferase reaction. Whether it binds sugar phosphates, pairs with full-length subunits, or has any physiological role has not been tested.
Summary: The full-length Gfat1 products have this activity, but the Gfat1-PF isoform has lost the glutaminase domain and its catalytic Cys2, so it cannot perform the glutamine-dependent reaction.
Reason: The activity is well supported for the gene: fly GFAT1 expressed in yeast had activity regulated by UDP-GlcNAc and PKA (PMID:11716769), and full-length Gfat1 (Q7PLC7, PANTHER PTHR10937:SF0) falls under PAINT node PTN000101302, where GO:0004360 was placed with fly Gfat1 itself among the seeds. Those experiments and that node describe a protein with both domains. The reaction needs the N-terminal glutaminase to hydrolyse glutamine and pass the ammonia through a channel to the isomerase site (PMID:16339762). A8Y5A1 (Gfat1-PF) is exactly the C-terminal 434 residues of the 694-residue Gfat1-PA. It keeps only 40 of the 299 residues of the glutamine-amidotransferase domain and lacks the Cys2 nucleophile. This automated mapping, made from the family signature (IPR005855) and the EC number, therefore assigns this gene product an activity its sequence cannot carry out. The removal applies only to this accession and not to the gene; the activity should stay on the full-length products. Whether PF could form a mixed dimer with full-length subunits and contribute to the shared sugar site is untested and is raised as a question, not annotated.
Supporting Evidence:
file:DROME/Gfat1/Gfat1-bioinformatics/RESULTS.md
Only 40 of 299 residues of the annotated glutamine-amidotransferase domain remain, and its N-terminal catalytic Cys2 is absent.
Glucosamine-6P synthase catalyzes the synthesis of glucosamine-6P from fructose-6P and glutamine and uses a channel to transfer ammonia from its glutaminase to its synthase active site.
Expressing Dmel/GFAT1 in yeast we showed that Dmel/GFAT1 activity is controlled by UDP-N-acetylglucosamine and PKA in the yeast total protein extract system.
Summary: The gene takes part in hexosamine (amino-sugar) metabolism through its full-length enzyme. No step of that process has been shown to be carried out by the truncated PF isoform.
Reason: Gfat1's role in amino-sugar metabolism comes from the catalytic full-length products (PMID:11716769; PMID:32615907). PF lacks the glutaminase module, so it cannot catalyse the committed step. A non-catalytic contribution, such as sugar-phosphate binding or regulating or partnering with full-length subunits, is possible but has never been tested for this isoform. Losing catalysis does not prove that the isoform takes no part in the process, so the row is left undecided rather than removed.
Supporting Evidence:
file:DROME/Gfat1/Gfat1-bioinformatics/RESULTS.md
Free-ammonia use, trans-complementation and a regulatory role are not demonstrated and are not assumed.
first committed step of the hexosamine biosynthesis pathway (HBP) using
GO:0006048 UDP-N-acetylglucosamine biosynthetic process
IEA GO_REF:0000041
UNDECIDED
Summary: This UniPathway mapping is correct for the catalytic Gfat1 enzyme, but the glutaminase-deficient PF isoform has not been shown to do any part of UDP-GlcNAc synthesis.
Reason: The full-length GFAT reaction is the committed first step of hexosamine biosynthesis (PMID:32615907). Fly Gfat1 activity is feedback inhibited by UDP-GlcNAc (PMID:11716769), and gfat1 null mutants die (PMID:32615907); that lethality shows the gene is required, not what PF does. PF lacks the catalytic glutaminase domain (Cys2 and most of the domain), so the IEA's assumption of a complete enzyme does not hold for this accession. No evidence shows any other contribution by PF, and none rules one out.
Supporting Evidence:
file:DROME/Gfat1/Gfat1-bioinformatics/RESULTS.md
Only 40 of 299 residues of the annotated glutamine-amidotransferase domain remain, and its N-terminal catalytic Cys2 is absent.
Summary: PF keeps both SIS sugar-isomerase domains, whose conserved ligands are sugar phosphates (carbohydrate derivatives), so carbohydrate derivative binding is the more accurate term.
Reason: Both SIS domains are intact in PF (reference residues 372-512 and 543-684). In the GFAT/GlmS family, the isomerase domain binds hexose 6-phosphates. The isolated human GFAT isomerase domain was crystallised with glucose 6-phosphate and glucosamine 6-phosphate (PMID:19059404). The relationship to the human enzyme rests on membership of family PTHR10937 and on descent from PAINT node PTN000101302, where the GFAT activity, and so this isomerase-site binding, was placed above fly Gfat1, E. coli GlmS (both in SF0) and human GFPT1/GFPT2 (SF12/SF10). The human protein sits in a different subfamily, so only this family-wide sugar-phosphate binding is transferred. The ligands are sugar phosphates, which are carbohydrate derivatives rather than free carbohydrates, so the claim is better expressed by GO:0097367 carbohydrate derivative binding, a term this product already carries. Binding by PF itself has not been measured.
We now report the first structures of the isomerase domain of the human GFAT in the presence of cyclic glucose-6-phosphate and linear glucosamine-6-phosphate.
Summary: Both C-terminal SIS sugar-phosphate-binding domains are intact in PF, which supports this broad binding term.
Reason: PF keeps the whole C-terminal isomerase region. Ligand-bound structures of the homologous human isomerase domain (same PANTHER family PTHR10937, below the GFAT PAINT node PTN000101302) show hexose-phosphate binding at this site (PMID:19059404). The broad term matches what the retained domains imply. A more specific ligand term, such as fructose-6-phosphate binding, is not used, because the site binds substrate, product and a glucose 6-phosphate analogue, and nothing has been measured for PF. The term does not claim the complete glutamine-dependent reaction or any pathway flux.
Supporting Evidence:
file:DROME/Gfat1/Gfat1-bioinformatics/RESULTS.md
Both SIS domains are fully retained (reference 372–512 and 543–684).
We now report the first structures of the isomerase domain of the human GFAT in the presence of cyclic glucose-6-phosphate and linear glucosamine-6-phosphate.
GO:1901135 carbohydrate derivative metabolic process
IEA GO_REF:0000002
UNDECIDED
Summary: This is inferred from the SIS domains alone. It is true of the gene, but no step of the process has been shown to be carried out by the truncated PF isoform.
Reason: Having SIS domains does not show that this product takes part in sugar-phosphate metabolism. The catalytic Gfat1 enzyme does (PMID:11716769), but PF lacks the glutaminase module needed for the reaction. A non-catalytic contribution by PF is possible but untested, so the row is left undecided rather than removed.
Supporting Evidence:
file:DROME/Gfat1/Gfat1-bioinformatics/RESULTS.md
Free-ammonia use, trans-complementation and a regulatory role are not demonstrated and are not assumed.
GO:1901137 carbohydrate derivative biosynthetic process
IEA GO_REF:0000002
UNDECIDED
Summary: The GFAT family signature implies glucosamine-6-phosphate synthesis, which PF cannot carry out without the glutaminase domain. Any other contribution by PF is untested.
Reason: IPR005855 marks the complete GFAT enzyme, whose biosynthetic role is the committed hexosamine step (PMID:32615907). PF matches the signature through its C-terminal region but lacks the catalytic Cys2 and most of the glutaminase domain. As with the other process rows, catalytic loss is established, but whether PF contributes in a non-catalytic way is not known.
Supporting Evidence:
file:DROME/Gfat1/Gfat1-bioinformatics/RESULTS.md
Only 40 of 299 residues of the annotated glutamine-amidotransferase domain remain, and its N-terminal catalytic Cys2 is absent.
Core Functions
Gfat1-PF keeps the C-terminal SIS sugar-isomerase region that binds hexose phosphates in GFAT-family enzymes. Without the glutaminase domain it cannot carry out the glutamine-dependent amidotransferase reaction of full-length Gfat1, and its native role is unresolved.
We now report the first structures of the isomerase domain of the human GFAT in the presence of cyclic glucose-6-phosphate and linear glucosamine-6-phosphate.
Q: Is the Gfat1-PF (434 aa, N-terminally truncated) protein made in vivo, and if so in which tissues and in what amount relative to the full-length Gfat1 isoforms?
Q: Can Gfat1-PF form mixed dimers with full-length Gfat1 or Gfat2 subunits, and would that add to or reduce GFAT activity at the shared sugar site?
Q: Since glucosamine-6-phosphate rescues gfat2 but not gfat1 null flies, does any Gfat1 product, including truncated isoforms such as PF, have a role beyond producing GlcN6P?
Suggested Experiments
Experiment: Express recombinant Gfat1-PF and full-length Gfat1-PA. Measure fructose-6-phosphate and glucosamine-6-phosphate binding, glutamine-dependent and ammonia-dependent GlcN6P synthesis, and co-purification with full-length subunits.
Experiment: Use isoform-specific tagging or targeted proteomics of the unique Met261 start to test whether the PF polypeptide exists in vivo, and test whether a PF-only transgene rescues any gfat1 null phenotype.
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.
The ProtNLM2 API snapshot retrieved 2026-09-08 contains no GO-term predictions for A8Y5A1, the native 434-residue Gfat1-PF product. Omission of the complete glutamine-dependent amidotransferase activity (GO:0004360) is not a supported deficiency: this exact product lacks the catalytic cysteine and most of the glutaminase module needed for that reaction. The retained SIS domains and ligand-bound structures of homologous human isomerase domains support carbohydrate derivative binding (GO:0097367) as a broad, qualified omission candidate, not proof of target-specific binding or catalysis; GOA already carries this term for A8Y5A1 and the gene review accepts it, so its omission costs no novel annotation. Published work on fly Gfat1 (PMID:11716769, PMID:32615907, PMID:35159258) characterises the full-length enzyme and does not test this short product. Participation in UDP-N-acetylglucosamine biosynthesis remains unresolved for this isoform. The separate FUNCTION claim of the complete glutamine-dependent reaction is NPI because of the missing catalytic architecture. This completed record assesses the absence of GO output; predictions is empty because no GO or EC prediction was emitted. The narrative judgment remains in the linked function review and is not a score assigned to the omission.
Glutamine:fructose-6-phosphate aminotransferase 1 (Gfat1) encodes an enzyme that catalyzes the transamination of fructose 6-phosphate to D-glucosamine 6-phosphate. It is involved in the biosynthesis of UDP-N-acetylglucosamine, a precursor for chitin biosynthesis and protein glycosylation.