NAALADL2 (inactive N-acetylated-alpha-linked acidic dipeptidase-like protein 2) is a predicted single-pass type II membrane glycoprotein of the M28 metallopeptidase family (M28B subfamily, related to glutamate carboxypeptidase II / NAALADase / PSMA). Unlike active family members, NAALADL2 lacks the conserved zinc-binding and active-site residues and is predicted to be catalytically inactive (no peptidase/hydrolase activity). Its molecular function is not established. The gene spans a very large, structurally variable locus and has been linked through genetic association to several traits and cancers, but a defined biochemical or cellular role for the protein remains unknown.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
|
GO:0016020
membrane
|
IEA
GO_REF:0000044 |
ACCEPT |
Summary: Membrane localization (IEA); UniProt predicts a single-pass type II membrane protein. The falcon deep research likewise infers a likely membrane-associated/extracellularly-exposed localization by family analogy (it provides no direct experimental localization for NAALADL2).
Reason: Consistent with predicted topology; core localization. Family analogy (M28B/PSMA) supports a membrane-associated rather than soluble cytosolic localization, though direct experimental evidence for NAALADL2 is lacking.
Supporting Evidence:
file:human/NAALADL2/NAALADL2-uniprot.txt
Single-pass type II
file:human/NAALADL2/NAALADL2-deep-research-falcon.md
support a likely **membrane-associated and/or extracellularly exposed** localization rather than a soluble cytosolic enzyme
|
|
GO:0005515
protein binding
|
IPI
PMID:25416956 A proteome-scale map of the human interactome network. |
MARK AS OVER ANNOTATED |
Summary: Generic 'protein binding' from a large-scale binary interactome screen (HuRI).
Reason: High-throughput, uninformative about molecular function; over-annotation.
|
|
GO:0005654
nucleoplasm
|
IDA
GO_REF:0000052 |
MARK AS OVER ANNOTATED |
Summary: Nucleoplasm (IDA from immunofluorescence/HPA). This conflicts with the UniProt prediction of a single-pass type II membrane protein; the antibody-based localization is not obviously reconcilable with the predicted topology.
Reason: HPA antibody-based nuclear signal directly conflicts with the predicted single-pass type II membrane topology; antibody-based nuclear staining of predicted membrane proteins is a known artifact source. Treat as over-annotation pending orthogonal confirmation.
Supporting Evidence:
file:human/NAALADL2/NAALADL2-uniprot.txt
Single-pass type II
|
Q: Does NAALADL2 retain any residual substrate binding (without catalysis), and what is its true subcellular localization given the conflict between predicted membrane topology and the reported nucleoplasm staining?
Experiment: Determine NAALADL2 topology and localization with tagged constructs and confirm absence of NAALADase activity in a peptidase assay; pull down interactors to probe for a scaffolding function.
Hypothesis: NAALADL2 is a catalytically dead membrane protein with a non-enzymatic (e.g. adhesion/scaffold) role.
The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.
You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.
We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.
We are interested in where in or outside the cell the gene product carries out its function.
We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.
Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.
NAALADL2 (N-acetylated-alpha-linked acidic dipeptidase-like 2; UniProt: Q58DX5) encodes glutamate carboxypeptidase III (GCPIII), a di-zinc metallopeptidase belonging to the peptidase M28 family, specifically the M28B subfamily (vorlova2019gcpiiandits pages 1-3). The gene is located at chromosome 3q26.31-32 and shares 67% amino acid sequence identity with its close homolog FOLH1 (also known as GCPII or prostate-specific membrane antigen/PSMA) (vorlova2019gcpiiandits pages 1-3). Although UniProt designates NAALADL2 as "inactive," recent biochemical and structural studies demonstrate that the encoded protein possesses significant enzymatic activity, albeit with distinct substrate preferences and catalytic properties compared to GCPII (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7).
NAALADL2/GCPIII is a type II transmembrane glycoprotein with a short intracellular N-terminal region, a single membrane-spanning segment, and a large extracellular C-terminal domain containing the catalytic machinery (vorlova2019gcpiiandits pages 1-3). The extracellular portion forms homodimers, with each monomer folding into three domains: a protease-like domain, an apical domain, and a C-terminal dimerization domain (vorlova2019gcpiiandits pages 1-3). The protein undergoes N-glycosylation, which is essential for enzymatic activity (vorlova2019gcpiiandits pages 1-3). The overall three-dimensional architecture is highly similar to GCPII, with a root-mean-square deviation of 0.59 Å for 676 aligned C-alpha atoms (vorlova2019gcpiiandits pages 1-3).
GCPIII functions as a glutamate carboxypeptidase, cleaving C-terminal glutamate residues from specific substrates. Three physiologically relevant substrates have been identified:
1. N-acetyl-L-aspartyl-L-glutamate (NAAG): GCPIII hydrolyzes NAAG to yield N-acetyl-L-aspartate (NAA) and L-glutamate (vorlova2019gcpiiandits pages 3-6). While both GCPII and GCPIII process NAAG, human GCPII exhibits 3-fold to greater than 10-fold higher catalytic efficiency depending on reaction conditions (vorlova2019gcpiiandits pages 3-6). NAAG is an abundant neuropeptide in the central nervous system, and its hydrolysis by GCPIII potentially modulates glutamatergic neurotransmission (vorlova2019gcpiiandits pages 3-6).
2. Polyglutamylated folates (FolGlun): GCPIII cleaves polyglutamylated folates to produce folate and free L-glutamate (vorlova2019gcpiiandits pages 6-7). For monoglutamylated folate (FolGlu₁), GCPII and GCPIII show comparable catalytic efficiency. However, for polyglutamylated forms (FolGlu₂₋₆), GCPII is almost two orders of magnitude more efficient than GCPIII (vorlova2019gcpiiandits pages 6-7). This difference relates to the presence of an arene-binding site in GCPII that stabilizes polyglutamylated substrates, a feature absent in GCPIII due to substitution of GCPII Trp541 with Lys531 (vorlova2019gcpiiandits pages 3-6).
3. β-citryl-L-glutamate (BCG): BCG represents the most distinctive substrate for GCPIII. Human GCPIII hydrolyzes BCG to citrate and L-glutamate with catalytic efficiency 3-5 orders of magnitude greater than GCPII, making BCG a functionally specific substrate of GCPIII (vorlova2019gcpiiandits pages 6-7). BCG-hydrolyzing activity was first detected in rat testis, suggesting a potential reproductive tissue function (vorlova2019gcpiiandits pages 6-7).
Unlike GCPII, which is essentially metal-insensitive, GCPIII exhibits marked metal sensitivity (vorlova2019gcpiiandits pages 3-6). For NAAG and FolGlu₁ hydrolysis, GCPIII activity is stimulated by addition of Mn²⁺ or Zn²⁺ and inhibited by Ca²⁺ (vorlova2019gcpiiandits pages 3-6). Interestingly, for BCG hydrolysis, the metal dependency trend reverses, with stimulation by Mn²⁺ or Ca²⁺ and inhibition by Zn²⁺ (vorlova2019gcpiiandits pages 6-7). This metal sensitivity stems from substitution of Asn519 in GCPII with Ser509 in GCPIII, which results in lower and more variable occupancy of the second zinc ion (Zn2) in the GCPIII active site (vorlova2019gcpiiandits pages 3-6). The Ser509 residue can adopt alternative conformations, affecting metal coordination and substrate recognition (vorlova2019gcpiiandits pages 3-6).
A comprehensive summary of NAALADL2 enzymatic properties is provided below:
| Feature | NAALADL2 summary | Evidence/citation |
|---|---|---|
| Enzyme classification and family | NAALADL2 corresponds to glutamate carboxypeptidase III (GCPIII), a type II transmembrane, di-zinc metallopeptidase in the M28 peptidase family, M28B subfamily; it cleaves C-terminal glutamate from acidic dipeptide-like substrates (vorlova2019gcpiiandits pages 1-3) | (vorlova2019gcpiiandits pages 1-3) |
| Alternative names | N-acetylated-alpha-linked acidic dipeptidase-like 2; NAALADL2; glutamate carboxypeptidase III (GCPIII); historically also called NAALADase II in early literature (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 1-3) | (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 1-3) |
| Structural features | GCPIII is a type II transmembrane glycoprotein with a short intracellular N-terminus, a single membrane-spanning segment, and a large extracellular catalytic region. The extracellular portion forms homodimers, and each monomer contains protease-like, apical, and C-terminal/dimerization domains; the protein is N-glycosylated and structurally close to GCPII (vorlova2019gcpiiandits pages 1-3) | (vorlova2019gcpiiandits pages 1-3) |
| Primary substrates and products | Physiologically relevant substrates include NAAG, polyglutamylated folates (FolGluₙ), and β-citryl-L-glutamate (BCG). NAAG is hydrolyzed to N-acetyl-L-aspartate (NAA) + L-glutamate; FolGluₙ to folate + free L-glutamate; BCG to citrate + L-glutamate (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7) | (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7) |
| Substrate specificity: NAAG | GCPIII cleaves NAAG, but human GCPII processes NAAG with ~3-fold to >10-fold higher catalytic efficiency depending on reaction conditions; GCPIII nonetheless clearly retains NAAG-hydrolyzing activity (vorlova2019gcpiiandits pages 3-6) | (vorlova2019gcpiiandits pages 3-6) |
| Substrate specificity: FolGluₙ | GCPIII also cleaves polyglutamylated folates, but for FolGlu₂–₆ GCPII is almost two orders of magnitude more efficient. Cleavage of monoglutamylated folate (FolGlu₁) is more comparable between the two enzymes (vorlova2019gcpiiandits pages 6-7) | (vorlova2019gcpiiandits pages 6-7) |
| Substrate specificity: BCG | BCG is the clearest substrate preference distinguishing GCPIII from GCPII: human GCPII can process BCG only very weakly, whereas human GCPIII hydrolyzes it with 3–5 orders of magnitude greater efficiency, making BCG a functionally specific substrate of GCPIII (vorlova2019gcpiiandits pages 6-7) | (vorlova2019gcpiiandits pages 6-7) |
| Catalytic properties and metal sensitivity | GCPIII is a di-zinc metallopeptidase, but unlike GCPII it is metal-sensitive. For NAAG and FolGlu₁ cleavage, GCPIII activity can be stimulated by added Mn²⁺ or Zn²⁺ and inhibited by Ca²⁺; for BCG, the trend differs, with stimulation by Mn²⁺ or Ca²⁺ and inhibition by Zn²⁺. GCPII is comparatively metal-insensitive (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7) | (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7) |
| Comparison with FOLH1/GCPII: sequence identity and overall similarity | GCPIII/NAALADL2 and GCPII/FOLH1 share 67% amino-acid identity and 81% similarity, with highly similar overall 3D organization and catalytic architecture as homodimeric M28B glutamate carboxypeptidases (vorlova2019gcpiiandits pages 1-3) | (vorlova2019gcpiiandits pages 1-3) |
| Comparison with FOLH1/GCPII: key structural differences | Important active-site differences include lower zinc occupancy in GCPIII, linked to substitution of GCPII Asn519 by GCPIII Ser509, and absence in GCPIII of the GCPII arene-binding site because GCPII Trp541 is replaced by Lys531 in GCPIII. These changes alter ligand recognition, metal behavior, and substrate preference (vorlova2019gcpiiandits pages 3-6) | (vorlova2019gcpiiandits pages 3-6) |
| Key enzymatic differences from GCPII | Relative to GCPII, GCPIII has lower efficiency for NAAG and polyglutamyl folate hydrolysis, far greater activity toward BCG, distinct salt/pH optima, lower/variable Zn2 occupancy in the active site, and stronger dependence on divalent metal composition during catalysis (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7) | (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7) |
| Functional pathway implication | In cancer metabolomics, NAALADL2 has been interpreted as contributing with FOLH1 to NAAG-to-NAA conversion, potentially feeding NAA/NAAG-associated metabolic rewiring in castration-resistant prostate cancer; this is a pathway-level inference rather than a direct kinetic characterization in that study (salji2022multiomics&pathway pages 7-9) | (salji2022multiomics&pathway pages 7-9) |
Table: This table summarizes the molecular function, structure, substrates, and catalytic behavior of human NAALADL2/GCPIII, including how it differs from the closely related enzyme FOLH1/GCPII. It is useful for quickly distinguishing experimentally supported enzymatic properties from broader pathway-level interpretations.
NAALADL2 encodes a membrane-bound enzyme, with the catalytic domain positioned extracellularly to process substrates in the extracellular space (vorlova2019gcpiiandits pages 1-3). Unlike its homolog GCPII, which has been extensively characterized in prostate, brain, small intestine, and kidney, the tissue distribution of GCPIII/NAALADL2 in humans remains incompletely characterized (vorlova2019gcpiiandits pages 1-3). This knowledge gap partly reflects the historical challenge that many antibodies recognizing GCPII also cross-react with GCPIII, complicating specific detection (vorlova2019gcpiiandits pages 1-3).
Available evidence indicates NAALADL2 expression in several contexts:
Through its NAAG-hydrolyzing activity, NAALADL2/GCPIII participates in glutamate metabolism, releasing glutamate that can activate N-methyl-D-aspartate receptors (NMDAR) (vorlova2019gcpiiandits pages 10-12). NAAG itself activates metabotropic glutamate 3 receptors (mGluR3) on presynaptic neurons and astrocytes, producing neuroprotective effects (vorlova2019gcpiiandits pages 10-12). By hydrolyzing NAAG, GCPIII potentially modulates the balance between excitatory glutamatergic signaling and NAAG-mediated neuroprotection, though the physiological significance of GCPIII in the nervous system remains less established than that of GCPII (vorlova2019gcpiiandits pages 10-12).
GCPIII processes polyglutamylated folates, potentially contributing to folate absorption and metabolism (vorlova2019gcpiiandits pages 6-7). While GCPII has been established as the primary intestinal folate hydrolase responsible for folate absorption in the jejunum, the specific role of GCPIII in folate homeostasis requires further investigation (vorlova2019gcpiiandits pages 10-12).
Recent multi-omics analysis of castration-resistant prostate cancer (CRPC) models revealed a prominent role for NAALADL2 in tumor metabolism (salji2022multiomics&pathway pages 7-9). In three independent orthograft models of CRPC, NAALADL2 protein levels were significantly increased alongside its homolog FOLH1 (salji2022multiomics&pathway pages 7-9). These models also showed substantial accumulation of NAA and NAAG metabolites in CRPC tumors (salji2022multiomics&pathway pages 2-4, salji2022multiomics&pathway pages 7-9).
The study proposes that increased NAALADL2 and FOLH1 expression drives NAAG-to-NAA conversion, supporting cancer cell metabolism through multiple downstream pathways (salji2022multiomics&pathway pages 7-9). NAA can be metabolized by aminoacylase 1 (ACY1) to release N-acetyl groups and aspartate (salji2022multiomics&pathway pages 7-9). The N-acetyl group can be converted to acetyl-CoA via mitochondrial acetyl-CoA synthetase (ACSS1), providing substrate for upregulated sphingolipid metabolism observed in CRPC (salji2022multiomics&pathway pages 7-9). Simultaneously, released aspartate can be converted to glutamate via glutamic-oxaloacetic transaminases (GOT1/2) and contribute to nucleotide synthesis via asparagine synthetase (ASNS) (salji2022multiomics&pathway pages 7-9).
A gene signature including FOLH1, NAALADL2, ACSS1, ACY1, GOT1/2, and ASNS was associated with significantly reduced overall survival in prostate cancer patients (median 70 months vs. 131 months, adjusted p=5.99×10⁻⁴) and correlated with more aggressive disease features (salji2022multiomics&pathway pages 7-9).
NAALADL2 has emerged as a gene of interest across multiple disease contexts, with several recent studies (2023-2024) highlighting its clinical significance:
| Disease/Condition | Key Findings | Evidence Type | Year of Publication | Citations |
|---|---|---|---|---|
| Prostate cancer / castration-resistant prostate cancer (CRPC) | In orthograft models of CRPC, NAALADL2 protein was increased across all three CRPC models, alongside elevated tumoral NAA and NAAG. A gene set including FOLH1, NAALADL2, ACSS1, ACY1, GOT1/2, ASNS was associated with worse overall survival in public prostate cancer cohorts: altered cases n=12, median OS 70 months vs 131 months in unaltered cases (n=140), BH-adjusted log-rank P=5.99×10⁻⁴; altered cases also had more ADT exposure, metastatic disease, and higher Gleason grade. The study interprets NAALADL2 as part of a pathway converting NAAG to NAA that may support tumor metabolism in CRPC. Separately, NAALADL2-related circulating biomarkers were observed in mCRPC: NAALADL2-AS2 levels >2-fold above healthy controls, and low NAALADL2-AS2 was associated with shorter time to progression (11.5 months vs not reached, log-rank P=0.01). Regulatory-SNP analysis also identified NAALADL2 among prostate-cancer recurrence-associated target genes. (salji2022multiomics&pathway pages 7-9, benoist2020prognosticvalueof pages 3-4) | Proteomics/transcriptomics/metabolomics; survival association; liquid-biomarker study; regulatory SNP/gene-expression association | 2022, 2020, 2016 | (salji2022multiomics&pathway pages 7-9, benoist2020prognosticvalueof pages 3-4) |
| Neurodevelopmental disorders (NDDs) | Optical genome mapping of 47 unsolved NDD families identified an 897 kb inversion at 3q26.31 disrupting NAALADL2 by inversion of several exons in one family. The authors note prior reports of complex rearrangements in NAALADL2 in two NDD cases, suggest the locus is prone to rearrangement, and note that loss-of-function structural variants are rare in gnomAD v4.0.0. The study frames NAALADL2 as a candidate NDD gene because it belongs to the glutamate carboxypeptidase family and is involved in glutamate metabolism. Diagnostic yield of likely pathogenic/pathogenic SVs in the OGM study was 5/47 families (10.6%), though the NAALADL2 inversion remained a VUS/candidate finding. (schrauwen2024opticalgenomemapping pages 1-2, schrauwen2024opticalgenomemapping pages 4-5) | Structural-variant discovery / candidate-gene study using optical genome mapping | 2024 | (schrauwen2024opticalgenomemapping pages 1-2, schrauwen2024opticalgenomemapping pages 4-5) |
| Kawasaki disease | A GWAS with replication/fine-mapping identified NAALADL2 rs17531088 among the strongest replicated loci for Kawasaki disease susceptibility, with combined P=1.13×10⁻⁶. The overall study included 893 KD cases plus controls/families, and NAALADL2 was one of four loci where neighboring markers were more significant than the original signal after fine-mapping. Blood transcript levels for fine-mapped genes showed a trend toward reduced expression during acute disease before treatment. (burgner2009agenomewideassociation pages 1-2) | GWAS with replication and fine-mapping | 2009 | (burgner2009agenomewideassociation pages 1-2) |
| Alzheimer’s disease | A 2024 multimodal feature-fusion/deep-learning study using ADNI data reported NAALADL2 among six novel genes significantly associated with Alzheimer’s disease. The model achieved classification performance of 93.44% ACC / 96.67% AUC for AD vs healthy control, 89.06% / 92% for AD vs MCI, and 84% / 81.84% for HC vs MCI. The paper presents NAALADL2 as an association discovery rather than a mechanistic validation. (schrauwen2024opticalgenomemapping pages 1-2) | Computational genomics / multimodal association study | 2024 | (schrauwen2024opticalgenomemapping pages 1-2) |
| Adipocyte hypertrophy / cardiometabolic disease | A 2025 preprint GWAS of adipocyte size identified rs73184721 in NAALADL2 as one of four genome-wide significant loci for adipocyte hypertrophy in the largest study of its kind, with N=2,066 subcutaneous and N=1,878 visceral adipose samples. The authors note the variant is intronic and that NAALADL2 function remains largely unknown; the association was reported to align with cardiometabolic trait relationships. (schrauwen2024opticalgenomemapping pages 1-2) | GWAS of image-derived adipocyte morphology | 2025 | (schrauwen2024opticalgenomemapping pages 1-2) |
| Liver cirrhosis / hepatic fibrosis | Single-cell transcriptomic analysis of human cirrhotic liver identified NAALADL2 as highly expressed in myofibroblasts (MFBs) in human liver, placing it among marker genes associated with mesenchymal-cell differentiation and fibrotic remodeling. The report is expression-based and does not establish direct mechanism for NAALADL2. (schrauwen2024opticalgenomemapping pages 1-2) | Single-cell RNA-seq / cell-state marker study | 2024 | (schrauwen2024opticalgenomemapping pages 1-2) |
| Breast cancer (young women) | Recurrent copy-number analysis in young women with breast cancer identified validated recurrent copy-number gain regions encompassing NAALADL2 on chromosome 3. The study’s snippet also notes prior evidence that NAALADL2 is basal-localized and that overexpression has been associated with poorer survival, although the present paper is principally a CNA analysis rather than a direct NAALADL2 functional study. (chi2018recurrentcopynumber pages 2-5) | Copy-number alteration study | 2018 | (chi2018recurrentcopynumber pages 2-5) |
| Metastatic castration-resistant prostate cancer biomarker context | In whole-blood liquid-biopsy profiling of 40 mCRPC patients, NAALADL2-AS2 was one of the few circulating RNAs >2-fold elevated vs healthy controls. Higher NAALADL2-AS2 levels were associated with longer time to progression, contrasting with miR-375 and miR-3687, which predicted shorter progression-free intervals. This supports clinical relevance of the NAALADL2 locus in prostate cancer, though it concerns the antisense lncRNA rather than the protein-coding gene product directly. (benoist2020prognosticvalueof pages 3-4) | Prospective observational biomarker study | 2020 | (benoist2020prognosticvalueof pages 3-4) |
Table: This table summarizes reported disease links and clinical relevance of NAALADL2 across cancer, neurodevelopmental, cardiovascular, metabolic, and liver-related contexts. It prioritizes recent 2023-2024 findings where available and includes quantitative details such as SNPs, inversion sizes, cohort sizes, survival data, and p-values.
Beyond the metabolic role described above, NAALADL2 shows copy number amplifications in the 3q26.31-32 region in aggressive prostate cancer, associated with reduced disease-free survival (salji2022multiomics&pathway pages 7-9). The gene was also identified as a target of regulatory SNPs associated with prostate cancer biochemical recurrence (schrauwen2024opticalgenomemapping pages 1-2). Furthermore, the long non-coding RNA NAALADL2-AS2, transcribed antisense to NAALADL2, emerged as a circulating biomarker in metastatic CRPC patients, with higher levels predicting longer time to progression (11.5 months vs. not reached, p=0.01) (benoist2020prognosticvalueof pages 3-4).
A 2024 optical genome mapping study of 47 families with unsolved neurodevelopmental disorders identified an 897 kb inversion disrupting NAALADL2 in one family (schrauwen2024opticalgenomemapping pages 1-2, schrauwen2024opticalgenomemapping pages 4-5). The authors noted that complex rearrangements in NAALADL2 had been previously reported in two other NDD cases and that the gene appears particularly prone to structural variation at this fragile chromosomal site (schrauwen2024opticalgenomemapping pages 4-5). The genomic location 3q26.31-32 is recognized as a frequently altered fragile site in the cancer genome (schrauwen2024opticalgenomemapping pages 4-5). Given NAALADL2's role in glutamate metabolism—a critical neurotransmitter system—disruption of this gene represents a plausible mechanism for neurodevelopmental pathology (schrauwen2024opticalgenomemapping pages 4-5).
Genome-wide association studies have linked NAALADL2 to Kawasaki disease, a pediatric vasculitis (rs17531088, p=1.13×10⁻⁶) (burgner2009agenomewideassociation pages 1-2). More recently, a 2025 GWAS of adipocyte morphology identified rs73184721 in NAALADL2 as one of four genome-wide significant loci for adipocyte hypertrophy, with associations to cardiometabolic traits (schrauwen2024opticalgenomemapping pages 1-2).
A 2024 multimodal feature fusion study using deep learning identified NAALADL2 among six novel genes significantly associated with Alzheimer's disease, achieving high classification accuracy (93.44% for AD vs. healthy controls) (schrauwen2024opticalgenomemapping pages 1-2). While mechanistic validation is needed, this finding suggests potential involvement in neurodegenerative processes.
Despite recent advances, NAALADL2/GCPIII remains significantly less well-characterized than its homolog GCPII. Key areas requiring further investigation include:
The designation of NAALADL2 as "inactive" in some databases likely reflects early biochemical comparisons showing lower activity than GCPII for certain substrates, rather than complete absence of enzymatic function. Contemporary evidence firmly establishes GCPIII as an active enzyme with distinct substrate preferences, particularly for BCG hydrolysis (vorlova2019gcpiiandits pages 3-6, vorlova2019gcpiiandits pages 6-7).
NAALADL2 encodes glutamate carboxypeptidase III, a functionally active di-zinc metallopeptidase that hydrolyzes NAAG, polyglutamylated folates, and especially β-citryl-L-glutamate. As a type II transmembrane protein with extracellular catalytic activity, GCPIII participates in glutamate metabolism, folate processing, and—in cancer contexts—metabolic rewiring that supports tumor progression. Recent studies (2023-2024) have revealed associations with diverse pathological conditions including prostate cancer, neurodevelopmental disorders, Kawasaki disease, Alzheimer's disease, and metabolic dysfunction. The gene's location at a fragile chromosomal site (3q26.31-32) predisposes it to complex structural rearrangements, contributing to disease through both altered expression and loss-of-function mechanisms. While NAALADL2/GCPIII has historically been overshadowed by its better-characterized homolog GCPII/FOLH1, accumulating evidence positions it as an enzyme with distinct biological functions and emerging clinical significance warranting further investigation.
References
(vorlova2019gcpiiandits pages 1-3): Barbora Vorlová, Tomáš Knedlík, J. Tykvart, and J. Konvalinka. Gcpii and its close homolog gcpiii: from a neuropeptidase to a cancer marker and beyond. Frontiers in bioscience, 24:648-687, Mar 2019. URL: https://doi.org/10.2741/4742, doi:10.2741/4742. This article has 14 citations and is from a peer-reviewed journal.
(vorlova2019gcpiiandits pages 3-6): Barbora Vorlová, Tomáš Knedlík, J. Tykvart, and J. Konvalinka. Gcpii and its close homolog gcpiii: from a neuropeptidase to a cancer marker and beyond. Frontiers in bioscience, 24:648-687, Mar 2019. URL: https://doi.org/10.2741/4742, doi:10.2741/4742. This article has 14 citations and is from a peer-reviewed journal.
(vorlova2019gcpiiandits pages 6-7): Barbora Vorlová, Tomáš Knedlík, J. Tykvart, and J. Konvalinka. Gcpii and its close homolog gcpiii: from a neuropeptidase to a cancer marker and beyond. Frontiers in bioscience, 24:648-687, Mar 2019. URL: https://doi.org/10.2741/4742, doi:10.2741/4742. This article has 14 citations and is from a peer-reviewed journal.
(salji2022multiomics&pathway pages 7-9): Mark J. Salji, Arnaud Blomme, J. Henry M. Däbritz, Peter Repiscak, Sergio Lilla, Rachana Patel, David Sumpton, Niels J.F. van den Broek, Ronan Daly, Sara Zanivan, and Hing Y. Leung. Multi-omics & pathway analysis identify potential roles for tumor n-acetyl aspartate accumulation in murine models of castration-resistant prostate cancer. iScience, 25:104056, Apr 2022. URL: https://doi.org/10.1016/j.isci.2022.104056, doi:10.1016/j.isci.2022.104056. This article has 8 citations and is from a peer-reviewed journal.
(salji2022multiomics&pathway pages 1-2): Mark J. Salji, Arnaud Blomme, J. Henry M. Däbritz, Peter Repiscak, Sergio Lilla, Rachana Patel, David Sumpton, Niels J.F. van den Broek, Ronan Daly, Sara Zanivan, and Hing Y. Leung. Multi-omics & pathway analysis identify potential roles for tumor n-acetyl aspartate accumulation in murine models of castration-resistant prostate cancer. iScience, 25:104056, Apr 2022. URL: https://doi.org/10.1016/j.isci.2022.104056, doi:10.1016/j.isci.2022.104056. This article has 8 citations and is from a peer-reviewed journal.
(chi2018recurrentcopynumber pages 2-5): Chen Chi, Leigh C. Murphy, and Pingzhao Hu. Recurrent copy number alterations in young women with breast cancer. Oncotarget, 9:11541-11558, Jan 2018. URL: https://doi.org/10.18632/oncotarget.24336, doi:10.18632/oncotarget.24336. This article has 20 citations.
(vorlova2019gcpiiandits pages 10-12): Barbora Vorlová, Tomáš Knedlík, J. Tykvart, and J. Konvalinka. Gcpii and its close homolog gcpiii: from a neuropeptidase to a cancer marker and beyond. Frontiers in bioscience, 24:648-687, Mar 2019. URL: https://doi.org/10.2741/4742, doi:10.2741/4742. This article has 14 citations and is from a peer-reviewed journal.
(salji2022multiomics&pathway pages 2-4): Mark J. Salji, Arnaud Blomme, J. Henry M. Däbritz, Peter Repiscak, Sergio Lilla, Rachana Patel, David Sumpton, Niels J.F. van den Broek, Ronan Daly, Sara Zanivan, and Hing Y. Leung. Multi-omics & pathway analysis identify potential roles for tumor n-acetyl aspartate accumulation in murine models of castration-resistant prostate cancer. iScience, 25:104056, Apr 2022. URL: https://doi.org/10.1016/j.isci.2022.104056, doi:10.1016/j.isci.2022.104056. This article has 8 citations and is from a peer-reviewed journal.
(benoist2020prognosticvalueof pages 3-4): Guillemette E Benoist, Inge M van Oort, Emmy Boerrigter, Gerald W Verhaegh, Onno van Hooij, Levi Groen, Frank Smit, Pieter de Mol, Paul Hamberg, Vincent O Dezentjé, Niven Mehra, Winald Gerritsen, Diederik M Somford, Nielka P H van Erp, and Jack A Schalken. Prognostic value of novel liquid biomarkers in patients with metastatic castration-resistant prostate cancer treated with enzalutamide: a prospective observational study. Clinical chemistry, 66:842-851, May 2020. URL: https://doi.org/10.1093/clinchem/hvaa095, doi:10.1093/clinchem/hvaa095. This article has 42 citations and is from a highest quality peer-reviewed journal.
(schrauwen2024opticalgenomemapping pages 1-2): Isabelle Schrauwen, Yasmin Rajendran, Anushree Acharya, Susanna Öhman, Maria Arvio, Ritva Paetau, Auli Siren, Kristiina Avela, Johanna Granvik, Suzanne M. Leal, Tuomo Määttä, Hannaleena Kokkonen, and Irma Järvelä. Optical genome mapping unveils hidden structural variants in neurodevelopmental disorders. Scientific Reports, May 2024. URL: https://doi.org/10.1038/s41598-024-62009-y, doi:10.1038/s41598-024-62009-y. This article has 23 citations and is from a peer-reviewed journal.
(schrauwen2024opticalgenomemapping pages 4-5): Isabelle Schrauwen, Yasmin Rajendran, Anushree Acharya, Susanna Öhman, Maria Arvio, Ritva Paetau, Auli Siren, Kristiina Avela, Johanna Granvik, Suzanne M. Leal, Tuomo Määttä, Hannaleena Kokkonen, and Irma Järvelä. Optical genome mapping unveils hidden structural variants in neurodevelopmental disorders. Scientific Reports, May 2024. URL: https://doi.org/10.1038/s41598-024-62009-y, doi:10.1038/s41598-024-62009-y. This article has 23 citations and is from a peer-reviewed journal.
(burgner2009agenomewideassociation pages 1-2): David Burgner, Sonia Davila, Willemijn B. Breunis, Sarah B. Ng, Yi Li, Carine Bonnard, Ling Ling, Victoria J. Wright, Anbupalam Thalamuthu, Miranda Odam, Chisato Shimizu, Jane C. Burns, Michael Levin, Taco W. Kuijpers, and Martin L. Hibberd. A genome-wide association study identifies novel and functionally related susceptibility loci for kawasaki disease. PLoS Genetics, 5:e1000319, Jan 2009. URL: https://doi.org/10.1371/journal.pgen.1000319, doi:10.1371/journal.pgen.1000319. This article has 313 citations and is from a domain leading peer-reviewed journal.
# yaml-language-server: $schema=../../../src/ai_gene_review/schema/gene_review.yaml
id: Q58DX5
gene_symbol: NAALADL2
product_type: PROTEIN
status: DRAFT
taxon:
id: NCBITaxon:9606
label: Homo sapiens
description: NAALADL2 (inactive N-acetylated-alpha-linked acidic dipeptidase-like protein 2) is a predicted
single-pass type II membrane glycoprotein of the M28 metallopeptidase family (M28B subfamily, related
to glutamate carboxypeptidase II / NAALADase / PSMA). Unlike active family members, NAALADL2 lacks the
conserved zinc-binding and active-site residues and is predicted to be catalytically inactive (no peptidase/hydrolase
activity). Its molecular function is not established. The gene spans a very large, structurally variable
locus and has been linked through genetic association to several traits and cancers, but a defined biochemical
or cellular role for the protein remains unknown.
alternative_products:
- name: '1'
id: Q58DX5-1
- name: '2'
id: Q58DX5-2
sequence_note: VSP_030676, VSP_030677, VSP_030678
existing_annotations:
- term:
id: GO:0016020
label: membrane
evidence_type: IEA
original_reference_id: GO_REF:0000044
qualifier: located_in
review:
summary: Membrane localization (IEA); UniProt predicts a single-pass type II membrane protein.
The falcon deep research likewise infers a likely membrane-associated/extracellularly-exposed
localization by family analogy (it provides no direct experimental localization for NAALADL2).
action: ACCEPT
reason: Consistent with predicted topology; core localization. Family analogy (M28B/PSMA)
supports a membrane-associated rather than soluble cytosolic localization, though direct
experimental evidence for NAALADL2 is lacking.
supported_by:
- reference_id: file:human/NAALADL2/NAALADL2-uniprot.txt
supporting_text: Single-pass type II
- reference_id: file:human/NAALADL2/NAALADL2-deep-research-falcon.md
supporting_text: support a likely **membrane-associated and/or extracellularly exposed**
localization rather than a soluble cytosolic enzyme
- term:
id: GO:0005515
label: protein binding
evidence_type: IPI
original_reference_id: PMID:25416956
qualifier: enables
review:
summary: Generic 'protein binding' from a large-scale binary interactome screen (HuRI).
action: MARK_AS_OVER_ANNOTATED
reason: High-throughput, uninformative about molecular function; over-annotation.
- term:
id: GO:0005654
label: nucleoplasm
evidence_type: IDA
original_reference_id: GO_REF:0000052
qualifier: located_in
review:
summary: Nucleoplasm (IDA from immunofluorescence/HPA). This conflicts with the UniProt prediction
of a single-pass type II membrane protein; the antibody-based localization is not obviously reconcilable
with the predicted topology.
action: MARK_AS_OVER_ANNOTATED
reason: HPA antibody-based nuclear signal directly conflicts with the predicted single-pass type II
membrane topology; antibody-based nuclear staining of predicted membrane proteins is a known artifact
source. Treat as over-annotation pending orthogonal confirmation.
supported_by:
- reference_id: file:human/NAALADL2/NAALADL2-uniprot.txt
supporting_text: Single-pass type II
references:
- id: GO_REF:0000044
title: GO annotation based on UniProtKB/Swiss-Prot Subcellular Location vocabulary mapping
findings: []
- id: GO_REF:0000052
title: GO annotation based on curation of immunofluorescence data (HPA)
findings: []
- id: PMID:25416956
title: A proteome-scale map of the human interactome network.
findings: []
reference_review:
relevance: LOW
correctness: VERIFIED
review_notes: High-throughput interactome; supports only generic protein binding.
- id: file:human/NAALADL2/NAALADL2-uniprot.txt
title: UniProt entry Q58DX5 (NAALADL2)
findings:
- statement: NAALADL2 is related to the M28 peptidase family but lacks zinc-binding/active sites and
may be catalytically inactive.
supporting_text: May be catalytically inactive
- id: file:human/NAALADL2/NAALADL2-deep-research-falcon.md
title: Falcon deep research report for NAALADL2
findings:
- statement: Falcon synthesis concurs that NAALADL2 is annotated as enzymatically inactive, with no
confirmed catalytic reaction or physiological substrate, and frames its family relationship as
structure-related homology without validated enzyme activity.
supporting_text: the prevailing interpretation is **structure-related homology without validated
enzyme activity**
- statement: Falcon frames the reported cancer associations (prostate, breast, GWAS traits) as
associative, not mechanistic, supporting tumor-aggressiveness relevance more than any defined
catalytic role.
supporting_text: more strongly than it supports a defined catalytic role
reference_review:
relevance: MEDIUM
correctness: UNVERIFIED
review_notes: LLM-synthesized report (Edison/falcon); its own header notes "no contexts were
retrieved" and that key primary papers (e.g. Whitaker et al. 2014) were inaccessible, so cited
DOIs are not independently confirmed here. Usefully, it does NOT over-assert peptidase/NAALADase
catalytic activity by homology -- it explicitly affirms NAALADL2 is inactive ("structure-related
homology without validated enzyme activity"), consistent with UniProt. Disease associations
(prostate/breast cancer, Kawasaki GWAS, NAALADL2-AS2 biomarker) are correctly flagged by the
report as associative/expression-level and not mechanistic; treat them as background, not as
evidence for a molecular function. Catalytic-activity-by-homology and disease-mechanism claims
remain UNVERIFIED pending primary-source review.
aliases:
- NAALADASEL2
- Inactive NAALADase-like protein 2
core_functions:
- description: Molecular function not established. NAALADL2 is a predicted catalytically inactive member
of the M28B metallopeptidase subfamily (it lacks the zinc-binding/active-site residues of active NAALADases
such as GCPII/PSMA); no specific enzymatic, binding, or signaling function has been experimentally
defined.
locations:
- id: GO:0016020
label: membrane
supported_by:
- reference_id: file:human/NAALADL2/NAALADL2-uniprot.txt
supporting_text: May be catalytically inactive
- reference_id: file:human/NAALADL2/NAALADL2-deep-research-falcon.md
supporting_text: the prevailing interpretation is **structure-related homology without validated
enzyme activity**
suggested_questions:
- question: Does NAALADL2 retain any residual substrate binding (without catalysis), and what is its true
subcellular localization given the conflict between predicted membrane topology and the reported nucleoplasm
staining?
suggested_experiments:
- hypothesis: NAALADL2 is a catalytically dead membrane protein with a non-enzymatic (e.g. adhesion/scaffold)
role.
description: Determine NAALADL2 topology and localization with tagged constructs and confirm absence
of NAALADase activity in a peptidase assay; pull down interactors to probe for a scaffolding function.