I'm still gathering sources. The curl call needed approval, so I'm switching to WebFetch against the Europe PMC API.

I have the core primary literature. Next I'm checking candidate ontology IDs against the repo's local term caches, then I'll write the report.

OGT-Congenital Disorder of Glycosylation (OGT-CDG): research report for a dismech entry

Target: OGT-CDG / X-linked intellectual disability 106 (MRX106). MONDO:0030907 (intellectual disability, X-linked 106, confirmed against the local MONDO cache in the draft entry). Category: Mendelian. Date: 2026-10-01

Sources and how to treat this report

• Abstract quotes are verbatim. Every quote below comes from an abstract retrieved through the Europe PMC REST API. Before any of them is used as a KB snippet, run just fetch-reference PMID:… and just count-verified-snippets.
• Two review summaries were extracted by a secondary tool and can't be quoted. The patient tables and phenotype frequencies from the Pravata 2020 Eur J Hum Genet review (PMC7253464) and the Mayfield 2024 J Biol Chem review (PMC11381892) came from a model reading the full-text XML.
• Use those numbers as leads only.
• In the Mayfield extraction, the per-variant OGA and HCF1 table carried author attributions ("Makroglou 2020", "Stephen 2024") that I could not confirm. Treat that table as unverified.
• Some sources were not reached. OMIM returned HTTP 403, the PMC HTML pages served a CAPTCHA, and there is no OGT row in the local Orphanet or ClinGen caches (references_cache/ORPHA_*, CGGV_*). The OMIM, Orphanet and ClinGen identifiers below are therefore not confirmed.
• Ontology IDs are tagged by provenance.
• [cache] means the ID and label were read from this repo's cache/*/terms.csv during this session.
• [verify] means the ID was not found locally and must be looked up before binding. This follows the CLAUDE.md rule that a CURIE is never written from memory.

1. Disease information

Overview. OGT-CDG is a syndromic X-linked neurodevelopmental disorder caused mainly by missense variants in OGT (Xq13.1). OGT encodes O-GlcNAc transferase, which is the only enzyme that adds O-linked β-N-acetylglucosamine (O-GlcNAc) to serine and threonine residues of nuclear and cytosolic proteins.

• In 2020, Pravata and colleagues proposed classifying the condition as a congenital disorder of glycosylation. PMID:32080367 (review): "we compile the work from the last few years that clearly delineates a new syndromic form of ID, which we propose to classify as a novel Congenital Disorder of Glycosylation (OGT-CDG)."
• Mayfield 2024 summarises the phenotype. PMID:39059494: "This disorder primarily presents with global developmental delay and intellectual disability (ID), alongside other variable neurological features and subtle facial dysmorphisms in patients."

Identifiers

Synonyms: OGT-CDG; OGT-XLID (used in PMID:33329753 and PMID:42107645); OGT-ID (PMID:42770280); MRX106; X-linked intellectual disability 106; mental retardation, X-linked 106.

Data provenance: everything here comes from aggregated case reports and family studies plus model systems, not from EHR data.

2. Etiology

• Cause. Germline variants in OGT, mostly missense. The first OGT-CDG families were all hemizygous males, but heterozygous females with de novo catalytic-domain variants have since been reported.
• The first variant, p.L254F, was found in an X-exome screen of XLID families. PMID:25679214 (Niranjan 2015) is the screen; PMID:28302723 (Vaidyanathan 2017) characterised the variant.
• PMID:28302723: "An X-chromosome exome screen identified a missense mutation, which encodes an amino acid in the tetratricopeptide repeat, in OGT (759G>T (p.L254F)) that segregates with X-linked intellectual disability (XLID) in an affected family."
• Risk factors. The only established risk factor is being male and carrying a pathogenic OGT allele, with family history of X-linked intellectual disability as the practical marker. No environmental risk factors are known.
• Modifiers and protection.
• No modifier genes or protective alleles have been established.
• The level of OGA compensation is a candidate biochemical modifier (see Section 6).
• Skewed X-inactivation is relevant in heterozygous females (PMID:31296563).
• Gene–environment interaction. None documented. Because OGT is a nutrient sensor fed by the hexosamine biosynthetic pathway (UDP-GlcNAc), a metabolic interaction is plausible but has not been studied (reviewed in PMID:33329753).

3. Phenotypes

Frequencies. The figures come from the Pravata 2020 EJHG review summary (13 patients in 7 families). They were extracted by the secondary tool, so confirm them against the published table before use. Figures in the "~" form are approximate; denominators are the number of patients assessed.

Onset and course. Onset is congenital or in infancy and presents as developmental delay. The course is a static neurodevelopmental disorder; no regression has been reported. Severity ranges from mild to severe ID.

Quality of life. No QoL instrument data have been published. The functional burden comes from lifelong ID, language impairment and behavioural comorbidity.

4. Genetics and molecular information

Gene. OGT (Xq13.1) has an N-terminal tetratricopeptide-repeat (TPR) domain that handles substrate and partner recognition, and a C-terminal glycosyltransferase domain split into two catalytic lobes. OGT also proteolytically matures HCF1 (HCFC1, hgnc:4839 [cache]), which is itself an XLID gene. OGT's counter-enzyme is OGA (gene OGA, formerly MGEA5; HGNC ID [verify]).

Reported variants (from the primary abstracts and the EJHG review table)

The 2026 OGT/OGA variant catalogue (PMID:42107645, oglcnac.mcw.edu/ogtoga/ogt/) now lists 101 pathogenic OGT-XLID variants curated with clinicians: "we partnered directly with clinicians and researchers to curate the most comprehensive and up-to-date collection of pathogenic OGT-XLID variants (n = 101)."

Variant class and functional effect - Variants are mostly missense, with one splice variant. They are hypomorphic (partial loss of function). Null alleles are expected to be lethal because OGT is essential for vertebrate embryogenesis (PMID:31296563). - TPR variants cause modest instability. PMID:29769320: "modest declines in thermodynamic stability and/or activities." - Catalytic-domain variants cause loss of activity. PMID:31296563: "decreased OGT stability and disruption of the substrate binding site, resulting in loss of catalytic activity."

Other genetic features - Variants are germline; none are somatic. - Population frequency: pathogenic variants are absent or ultra-rare. gnomAD OGT variants trigger a stronger homeostatic feedback response than OGT-CDG variants (PMID:39706180). - Curation status: Genomics England PanelApp lists OGT Green (high evidence) for intellectual disability. No ClinGen gene–disease validity record is in the local cache. Because no classification was retrieved, gene_disease_validity should be left empty. - Epigenetics: none in patients. Mechanistic links run through OGT–TET complexes (PMID:41033462) and through chromatin and Polycomb roles (reviewed in PMID:33329753). - No chromosomal abnormalities are associated with the disorder.

5. Environmental information

None known. There is no infectious agent, toxin or lifestyle factor. For a dismech entry, leave environmental: empty rather than inventing a hexosamine or nutrient link.

6. Mechanism and pathophysiology

Causal chain

1. A germline hypomorphic OGT variant in the TPR or catalytic domain leads to reduced OGT protein stability, reduced catalytic activity, or altered substrate and partner recognition (PMID:28302723, PMID:28584052, PMID:29606577, PMID:31296563).
2. Branch A, catalytic variants (N567K, N648Y, C921Y): catalytic activity is lost, which results in global hypo-O-GlcNAcylation in stem cells and in brain (PMID:31627256, PMID:37334838, PMID:38566589).
3. Branch B, TPR variants (L254F, R284P, A259T, E339G): global O-GlcNAc is near normal. L254F distorts the TPR helix (PMID:29606577). R284P also impairs HCF1 proteolysis (PMID:28584052).
4. Reduced OGT function triggers homeostatic compensation: OGA expression falls, through an OGT/mSin3A-HDAC1 repressor complex at the OGA promoter. This keeps global O-GlcNAc roughly stable, but imperfectly (PMID:28302723, PMID:28584052).
5. PMID:28302723: "lymphoblastoids from affected individuals displayed a marked decrease in steady-state OGA protein and mRNA levels."
6. PMID:28584052: "Our data suggest that mutant cells attempt to maintain global O-GlcNAcylation by down-regulating O-GlcNAcase expression."
7. The step from compensation to disease is supported by a screen. PMID:39706180 found reduced feedback across OGT-CDG variants, which "points to reduced disruption of O-GlcNAc homeostasis as a common mechanism." In the paper's terms, that means a blunted homeostatic response.
8. This O-GlcNAc dyshomeostasis leads to altered gene expression in pluripotent and early neural cells. Reported changes include:
9. cell-fate and LXR/RXR signalling genes (PMID:29769320);
10. reduced Oct4/Sox2 and self-renewal (PMID:37334838);
11. Zscan4 upregulation through the OGT–TET complex (PMID:41033462);
12. delayed neuronal differentiation (PMID:31296563);
13. altered neural stem cell morphology (PMID:38759397).
14. The link from these changes to the human brain phenotype is inferred.
15. Disrupted early neurodevelopment results in microcephaly, thinner cortex, superficial-layer cortical dysplasia and corpus callosum hypoplasia. This is shown in mouse (PMID:42770280, PMID:38566589).
16. Together with synaptic and neuronal circuit defects, this leads to ID, learning deficits and behavioural phenotypes.
17. Flies show impaired NMJ synaptogenesis and habituation learning, plus sleep instability (PMID:35500025, PMID:39535175).
18. Mice show hyperactivity and impulsivity (PMID:42770280).
19. A separate branch acts on growth: lower body weight and fat mass and short stature (mouse, PMID:38566589). This is the counterpart of short stature and low birth weight in patients. The underlying mechanism is not established.

Parallel hypotheses that are not resolved. These include TPR-interactome disruption (PMID:33356293, PMID:39059494), hypo-O-GlcNAcylation of specific neuronal substrates (22 candidates involved in Ras/MAPK signalling, translational repression, cytoskeleton and chromatin; PMID:35863433), and HCF1 misprocessing, which affects only some variants (PMID:28584052, PMID:31296563). These map naturally to mechanistic_hypotheses with status: EMERGING.

Detail

• Pathways: O-GlcNAc cycling; hexosamine-pathway nutrient sensing; HCF1 maturation (the HCF-1:OGT axis in neurogenesis, PMID:41651253); TET-dependent DNA demethylation; Ras/MAPK.
• GO terms:
• GO:7770074 protein O-linked glycosylation via N-acetylglucosamine [cache]
• GO:0006493 protein O-linked glycosylation [cache]
• GO:0030182 neuron differentiation [cache]
• GO:0007420 brain development [cache]
• GO:0007416 synapse assembly [cache]
• GO:0019827 stem cell population maintenance [cache]
• GO:0006338 chromatin remodeling [cache]
• Molecular function "protein O-GlcNAc transferase activity" [verify]
• Cell types:
• CL:0002322 embryonic stem cell [cache]
• CL:0000047 neural stem cell [cache]
• CL:0011020 neural progenitor cell [cache]
• CL:0000540 neuron [cache]
• Cellular compartments: nucleus and cytosol (GO CC IDs [verify]).
• Structural biology: L254F TPR helix distortion (PMID:29606577); TPR α-solenoid dynamics (PMID:31628985); OGT dimer cryo-EM structure (PMID:34764280).
• Omics:
• hESC transcriptomes (PMID:29769320)
• mESC quantitative proteomics (PMID:41033462)
• mouse brain proteomics (PMID:42770280)
• TPR BioID interactome, MassIVE MSV000085626 (PMID:33356293)
• No single-cell or spatial studies were found.

7. Anatomical structures affected

• Primary: brain, especially the cerebral cortex (superficial layers of the cingulate cortex in mouse) and the corpus callosum.
• Other: eye (refractive errors, amblyopia), skeleton and growth (short stature, clinodactyly), face, external genitalia, ear.
• UBERON IDs [verify] for: brain, cerebral cortex, corpus callosum, eye.
• Subcellular: nucleus and cytoplasm, where OGT and its substrates are located.
• Lateralization: not applicable.

8. Temporal development

• Onset: congenital, with prenatal growth effects (low birth weight). Developmental delay is evident in infancy.
• Course: a static, lifelong neurodevelopmental disorder. No stages and no remission.
• Critical period: stem cell and mouse data point to embryonic and early neurodevelopmental windows (PMID:37334838, PMID:38759397, PMID:41033462). Intervention in adulthood partially rescues fly synaptic and sleep phenotypes (PMID:39535175), which suggests some postnatal plasticity. The human relevance is inferred.

9. Inheritance and population

• Inheritance: X-linked, with HP:0001419 X-linked recessive inheritance [cache] for the familial male cases.
• Heterozygous females are usually unaffected carriers (FCDGC).
• Exceptions are de novo catalytic variants in affected females with skewed X-inactivation (PMID:31296563). A second inheritance block or a note may be warranted. HP:0001423 X-linked dominant inheritance [cache] should only be used if the de novo female cases are curated as such.
• Prevalence: unknown. About 101 pathogenic variants have been catalogued by 2026 (PMID:42107645). Use measure_type: CASES_IN_LITERATURE with prevalence_class: ULTRA_RARE or RARE; do not invent a rate.
• Other features:
• Sex ratio: strongly male-predominant.
• Penetrance: complete in hemizygous males as reported. Expressivity is variable.
• No anticipation and no founder effects.
• Consanguinity is not relevant.
• Carrier frequency is unknown.

10. Diagnostics

• Genetic testing is the diagnosis. Exome or genome sequencing, or ID/XLID gene panels including OGT (PanelApp Green). FCDGC: "OGT-CDG is ultimately diagnosed through genetic testing in blood."
• Glycosylation screening does not help. Standard CDG screening (transferrin isoelectric focusing) assesses N-glycans, so it is not expected to detect O-GlcNAc defects. I could not find a source that says this explicitly; it is my inference and needs one before it goes in the entry.
• Functional confirmation:
• patient fibroblast western blot for O-GlcNAc and OGA levels (PMID:28584052, PMID:41030119);
• recombinant enzyme activity assays (PMID:41030119);
• an mESC double-fluorescence feedback assay that separates pathogenic from benign VUS (PMID:39706180).
• A blood Ogt/Oga mRNA ratio has been proposed as a monitoring biomarker (mouse, PMID:42209020).
• Imaging: brain MRI (corpus callosum hypoplasia, microcephaly).
• Differential diagnosis: other syndromic XLID, including HCFC1-related cblX and XLID, and DDX3X; storage disorders because of the coarse facies and drooling; other CDGs.
• Screening: no newborn screening. Carrier and cascade testing of mothers and female relatives; prenatal or preimplantation testing once the familial variant is known.

11. Outcome and prognosis

No natural-history data exist. FCDGC: "Due to the rarity of the disease, there is no clear prognosis." Life expectancy is not reported to be shortened. Morbidity is lifelong ID, language impairment, and behavioural and visual issues. A single report of hepatoblastoma (PMID:41030119) raises an unproven question about cancer risk.

12. Treatment

• There is no disease-specific therapy. FCDGC: "There are no known treatments for OGT-CDG."
• Supportive and rehabilitative care. Suggested NCIT terms, all [verify]:
• Supportive Care (CLAUDE.md lists NCIT:C15747)
• Physical Therapy (NCIT:C15302)
• Speech-language therapy (NCIT:C159273)
• Occupational therapy (NCIT:C121351)
• Genetic Counseling (NCIT:C15240)
• Corrective lenses or ophthalmology care, and antiseizure medication where seizures occur.
• Experimental: OGA inhibition (preclinical only).
• Drosophila: PMID:35500025: "The habituation deficit can be corrected by blocking O-GlcNAc hydrolysis… blocking O-GlcNAc hydrolysis is a potential strategy to treat OGT-CDG." PMID:39535175: phenotypes "can be partially rescued by genetically or chemically targeting OGA."
• Mouse: crossing with catalytically dead OGA "partially restored O-GlcNAc homeostasis in brain and blood" (PMID:42209020).
• Brain-penetrant OGA inhibitors such as Thiamet-G, and clinical-stage compounds developed for tauopathy, are candidate agents (review PMID:39509538).
• No OGT-CDG clinical trial (NCT) was found.

13. Prevention

• Primary prevention: genetic counselling, carrier testing of at-risk females, and prenatal or preimplantation diagnosis.
• Tertiary prevention: early developmental intervention, eye surveillance, and seizure monitoring.
• No vaccines or environmental interventions apply.

14. Other species and natural disease

• No naturally occurring animal disease has been reported (no OMIA search was done).
• OGT is highly conserved:
• Drosophila super sex combs (sxc, a Polycomb-group gene);
• C. elegans ogt-1, which modulates seizure susceptibility (PMID:34797853);
• zebrafish ortholog listed on ZFIN (ZDB-GENE-030131-9631).

15. Model organisms

Limitations. - Mouse models reproduce growth and structural brain features, but cognitive testing in rodents only approximates human ID. - Fly models lack cortical architecture. - Stem cell compensation may mask phenotypes in early stages (PMID:38759397). - TPR-variant mouse models have not been reported.

For dismech: curate the mouse models under animal_models: and the cell models under experimental_models:. Use modeled_mechanisms targeted at nodes such as "O-GlcNAc Dyshomeostasis", "Impaired Neuronal Differentiation" and "Cortical Malformation", with PARTIALLY_RECAPITULATES where appropriate.

Key references

Web sources: FCDGC OGT-CDG page · LOVD MRX106 · PanelApp OGT · DOID:0080240 · Mayfield 2024, JBC · Czajewski 2024, eLife · Europe PMC REST API

Reference Validation

Checked with linkml-reference-validator 0.3.0rc3.

Unresolved references

These identifiers did not resolve to a record and may be fabricated. A lookup that failed for transport reasons is indistinguishable from one that failed because the record does not exist, so spot-check before acting on them:

• PMID:41651253 (1 mention) - Identifier did not resolve to a record

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

• DOID:0080240 (4 mentions) - the report calls it "DOID"; DOID calls it non-syndromic X-linked intellectual disability 106

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

• MONDO:0030907 (2 mentions) - the report calls it "intellectual disability, X-linked 106`, confirmed against the local MONDO cache in the draft entry"; MONDO calls it intellectual disability, X-linked 106