Seroatlas · Human Serome Atlas

GAD2

Glutamate decarboxylase 2

Also known as: DCE2_HUMAN, GAD65

Cross-references: UniProt · Ensembl · Human Protein Atlas · GeneCards · NCBI Gene

Protein identityUniProt · HPA

UniProt accession
Q05329
Gene
GAD2
Ensembl
ENSG00000136750
Chromosome
10
Canonical length
585 aa
Protein class
Enzymes, Metabolic proteins, Predicted intracellular proteins
Quaternary structure
Homodimer

OverviewNCBI Gene

This gene encodes one of several forms of glutamic acid decarboxylase, identified as a major autoantigen in insulin-dependent diabetes. The enzyme encoded is responsible for catalyzing the production of gamma-aminobutyric acid from L-glutamic acid. A pathogenic role for this enzyme has been identified in the human pancreas since it has been identified as an autoantibody and an autoreactive T cell target in insulin-dependent diabetes. This gene may also play a role in the stiff man syndrome. Alternative splicing results in multiple transcript variants that encode the same protein. [provided by RefSeq, Oct 2008]

Canonical amino-acid sequenceUniProt

585 residues, UniProt reviewed canonical sequence.

>Q05329|GAD2
     1  MASPGSGFWS FGSEDGSGDS ENPGTARAWC QVAQKFTGGI GNKLCALLYG DAEKPAESGG
    61  SQPPRAAARK AACACDQKPC SCSKVDVNYA FLHATDLLPA CDGERPTLAF LQDVMNILLQ
   121  YVVKSFDRST KVIDFHYPNE LLQEYNWELA DQPQNLEEIL MHCQTTLKYA IKTGHPRYFN
   181  QLSTGLDMVG LAADWLTSTA NTNMFTYEIA PVFVLLEYVT LKKMREIIGW PGGSGDGIFS
   241  PGGAISNMYA MMIARFKMFP EVKEKGMAAL PRLIAFTSEH SHFSLKKGAA ALGIGTDSVI
   301  LIKCDERGKM IPSDLERRIL EAKQKGFVPF LVSATAGTTV YGAFDPLLAV ADICKKYKIW
   361  MHVDAAWGGG LLMSRKHKWK LSGVERANSV TWNPHKMMGV PLQCSALLVR EEGLMQNCNQ
   421  MHASYLFQQD KHYDLSYDTG DKALQCGRHV DVFKLWLMWR AKGTTGFEAH VDKCLELAEY
   481  LYNIIKNREG YEMVFDGKPQ HTNVCFWYIP PSLRTLEDNE ERMSRLSKVA PVIKARMMEY
   541  GTTMVSYQPL GDKVNFFRMV ISNPAATHQD IDFLIEEIER LGQDL

LocalizationUniProt · AlphaFold · HPA

Whether an antibody against GAD2 can act on the native protein depends on physical access: surface and secreted proteins are reachable by circulating antibodies, intracellular proteins usually are not.

Antibody reachability
Cell surface
Secreted
No
Transmembrane segments
0
Mean surface accessibility (rSASA)
0.29
Highest tissue expression
65 nTPM

Expression across tissuesHPA

Tissue

  • basal ganglia: 65 nTPM
  • hypothalamus: 33 nTPM
  • cerebral cortex: 27 nTPM
  • amygdala: 14 nTPM
  • cerebellum: 9.8 nTPM
  • hippocampal formation: 9.6 nTPM

Single-cell type

  • pancreatic islet cells: 547 nCPM
  • brain inhibitory neurons: 385 nCPM
  • retinal amacrine cells: 179 nCPM
  • other brain neurons: 146 nCPM
  • oocytes: 111 nCPM
  • retinal horizontal cells: 20 nCPM

Immune cell

  • basophil: 0 nTPM
  • classical monocyte: 0 nTPM
  • eosinophil: 0 nTPM
  • gdT-cell: 0 nTPM
  • intermediate monocyte: 0 nTPM
  • MAIT T-cell: 0 nTPM

Brain region

  • hypothalamus: 101 nTPM
  • basal ganglia: 92 nTPM
  • midbrain: 44 nTPM
  • cerebral cortex: 42 nTPM
  • thalamus: 40 nTPM
  • amygdala: 37 nTPM

DiseaseUniProt · ClinVar · IEDB · PubMed

Four sources answering four different questions about GAD2.

Disease | ImmuneIEDB

Conditions an epitope on GAD2 was assayed in.

Disease | AutoantibodyPubMed

Conditions in which antibodies against GAD2 are reported. Each links to that disease's full target list.

Showing 20 of 72 — disease pages carrying at least 10 antigens.

ReferencesPubMed · IEDB

Publications for GAD2 from three distinct lines of evidence, kept separate because they answer different questions: whether antibodies are directed at the protein, whether a B-cell epitope has been mapped on it, and whether a T-cell epitope has. Each is labelled with its source.

Reference: AutoantibodyPubMed

1,518 publications

Show 20 more of 1,518 total

Reference: B cellIEDB

25 publications

Show 20 more

Reference: T cellIEDB

81 publications

Show 20 more of 81 total

Sources: PubMed — antigen-level antibody evidence from a custom retrieval. Records matching a controlled set of autoantibody terms (the MeSH descriptors Autoantibodies and Autoantigens, with title and abstract term variants) were obtained through NCBI E-utilities, and their titles and abstracts parsed for constructions that direct an antibody at a named protein rather than for co-occurrence. Captured names were resolved against UniProt nomenclature and each antigen adjudicated individually against the source text. IEDB — curated epitope assays from the Immune Epitope Database (Vita et al., Nucleic Acids Research 2019). Bibliographic records from PubMed and MeSH, U.S. National Library of Medicine; citation metrics from NIH iCite (Hutchins et al., PLoS Biology 2016). Titles link to PubMed; abstracts are not reproduced here. The NLM does not endorse this analysis.

Genetic constraint and essentialitygnomAD · DepMap

Does the body need this protein intact? Low LOEUF or a strong DepMap dependency means loss or blockade of the protein is likely to be felt.

gnomAD LOEUF (loss-of-function intolerance)
0.37
gnomAD pLI
0.75
gnomAD missense Z
1.53
DepMap mean gene effect
0.09
DepMap dependency class
none

Cancer expressionTCGA

Across TCGA tumor cohorts, this protein is over-expressed in roughly 2% of surveyed tumor types (aggregate summary; per-cohort expression, alteration, and survival load in the interactive view).

OntologyGO

Biological processes

Molecular functions

Cellular components

Protein domainsUniProt · Pfam · InterPro

KeywordsUniProt

Antibody and autoantibody relevanceSeroatlas analysis

Seroatlas reads GAD2 as an antibody target. Whether an autoantibody or antibody against GAD2 could matter depends on whether native GAD2 is physically reachable, whether the body needs it intact, and whether it acts in a disease-relevant tissue.

GAD2 is annotated at the cell surface, where native GAD2 is exposed to circulating antibodies and is a prime autoantibody target that could block, deplete, or overstimulate it.

Source-annotated serology context

The source annotations explicitly mention antibody, autoantibody, autoantigen, or autoimmune context. This is biological context, not study-specific reactivity.

  • This gene encodes one of several forms of glutamic acid decarboxylase, identified as a major autoantigen in insulin-dependent diabetes.
  • A pathogenic role for this enzyme has been identified in the human pancreas since it has been identified as an autoantibody and an autoreactive T cell target in insulin-dependent diabetes.

Canonical record: https://seroatlas.com/gene/GAD2. Study-independent annotations aggregated from UniProt, Human Protein Atlas, PubMed, IEDB, Pfam, InterPro, Gene Ontology, AlphaFold, gnomAD, DepMap, ClinVar, TCGA. Catalog release seroatlas-reviewed-human-uniprot-20260313.

Seroatlas is the reference for exploring autoantibody and antibody serology at the human-protein level: the autoreactome and human serome, multiplex serology (HuProt, HuScan, VirScan, PhIP-Seq).

© 2026 · Seroatlas, Discover Oregon LLC · Terms of use

Loading the interactive Seroatlas protein explorer...