EDEM3-Congenital Disorder of Glycosylation (EDEM3-CDG; CDG2V): Research Report

Prepared: 2026-09-28 · Target: MONDO:0030423 · Category: Mendelian (autosomal recessive)

How to read this report. Six full-text or abstract records were read from the local references_cache/: PMID:34143952, 16431915, 29784879, 34698634, 35500441 and 39838427. Text in quotation marks from those PMIDs is copied exactly and can be used as a snippet. Everything else comes from web or PubMed-summary retrieval and is paraphrased; it has to be fetched with just fetch-reference before it can back a snippet. I list ontology identifiers as leads only. The repo requires every CURIE to be looked up (runoak or the term caches) before it is written into YAML, and I have marked the ones I could not source in this session.

Main limitation. The whole human clinical literature is one paper: Polla et al., 2021, Am J Hum Genet (PMID:34143952), describing 12 patients from 7 families. A PubMed search for "EDEM3" (54 hits as of today) turned up no later case series or case report. Every phenotype frequency below is therefore an n=12 figure.

1. Disease Information

Overview. EDEM3-CDG is an autosomal recessive congenital disorder of glycosylation. It is caused by biallelic loss-of-function variants in EDEM3, which encodes an ER-lumenal class I (GH47) α1,2-mannosidase that trims mannose in glycoprotein ER-associated degradation (gpERAD). The clinical picture is non-specific developmental delay or intellectual disability with speech delay, variable hypotonia and mild facial dysmorphism.

"The affected individuals present with an inherited congenital disorder of glycosylation (CDG) consisting of neurodevelopmental delay and variable facial dysmorphisms." (PMID:34143952)

"We propose to call this deficiency EDEM3-CDG." (PMID:34143952)

Identifiers (from the MONDO record via OLS, retrieved today):

Synonyms: EDEM3-CDG; CDG2V; CDG-IIv; congenital disorder of glycosylation type IIv.

Nature of the data: aggregated case-level data from one exome/GeneMatcher-assembled cohort. There is no registry-level or EHR data. The FCDGC natural-history study (NCT04199000) enrolls CDG patients generally and may include EDEM3-CDG.

2. Etiology

• Cause: biallelic germline EDEM3 variants, mostly protein-truncating.

  "we have identified seven independent families with 11 individuals with bi-allelic protein-truncating variants and one individual with a compound heterozygous missense variant in EDEM3." (PMID:34143952)

• Genetic risk factors:
• Consanguinity and founder effect in the Portuguese Romani population, where c.1859del was found in 1/96 controls. > "Because families 1 and 2 are of Portuguese Romani origin, this suggests a possible founder effect." (PMID:34143952)
• Uniparental isodisomy (family 4). > "In family 4, a bi-allelic nonsense variant, c.940A>T (p.Arg314∗), was found, resulting from maternal uniparental isodisomy of chromosome 1." (PMID:34143952)
• Environmental risk, protective or GxE factors: none reported. Not applicable to a monogenic recessive enzyme deficiency.
• Related common-variant biology (not disease risk): the low-frequency missense variant rs78444298 (p.Pro746Ser, protease-associated domain, MAF ~1.5%) is associated with about 5% lower plasma triglycerides.

  "A low-frequency EDEM3 missense variant in the protease-associated domain (rs78444298, p.Pro746Ser, minor allele frequency ∼1.5%) was associated with an approximately 5% decrease in triglyceride levels." (PMID:34143952)

• The mechanism (EDEM3 loss → higher LRP1 → faster VLDL uptake) is from Xu et al., iScience 2020, PMID:32213464, which needs fetching.
• EDEM3-CDG patients nevertheless had normal fasting triglycerides: > "In our subjects for whom fasting triglyceride levels are available (Table S1), all triglyceride measurements were within the normal range." (PMID:34143952)

3. Phenotypes

Frequencies are from PMID:34143952 (n=12). Onset is infantile or early childhood. Severity is mild to moderate and course appears static. Quality-of-life data do not exist.

Supporting quotes (PMID:34143952):

"All affected individuals presented with developmental delay and/or intellectual disability (ID) and speech delay (Table S1). Hypotonia was present in six out of 12 persons."

"Additionally, gastroesophageal reflux was observed in three persons, and two individuals had early feeding difficulties requiring a nasogastric tube; of these, one individual needed a percutaneous endoscopic gastrostomy placement."

"Brain magnetic resonance imaging (MRI) of affected individuals from families 2, 3, and 5 did not detect structural abnormalities or myelination defects."

Additional features listed by CDG Hub (secondary source summarizing Table S1; each needs a primary-source check against the Polla supplement before curation): anosmia, apnea, delayed bone age, muscle atrophy, Poland sequence (one case), astigmatism, strabismus, failure to thrive, hyperactivity, anxiety and attention deficit.

4. Genetic / Molecular Information

Gene: EDEM3, located at 1q25.3. - HGNC:16787. Write it as hgnc:16787 in YAML, and confirm with just validate-terms. - NCBI Gene 80267 and UniProt Q9BZQ6. Both are from memory; verify them. - OMIM 610214. - Transcript NM_025191.3.

Protein: a 931-aa soluble ER-lumenal protein (mouse). It has an N-terminal GH47 α-mannosidase homology domain and a C-terminal protease-associated (PA) domain.

"EDEM3 consists of 931 amino acids and has all the signature motifs of Class I alpha-mannosidases (glycosyl hydrolase family 47) in its N-terminal domain and a protease-associated motif in its C-terminal region." (PMID:16431915)

Reported pathogenic variants (PMID:34143952; NM_025191.3). All were germline, with carrier parents. None has an ACMG classification in the paper; check ClinVar.

• Allele frequency: p.Asp61Gly is in 6/218,508 gnomAD alleles (rs777353823). p.Asp456Asn is absent from gnomAD.
• Functional consequence: loss of function via nonsense-mediated decay and absent protein.

  "These demonstrated the absence of EDEM3 in individual IV-4 (family 1) and individual II-1 (family 3) consistent with loss of function of EDEM3 (Figure 3C)." (PMID:34143952)

• Modifier genes: none established.
• EDEM1 is a candidate functional backup, but it is not upregulated in patient cells: "EDEM1 levels were at 97% of normal levels (p = 0.9373) in fibroblast cell lines from affected individuals" (PMID:34143952).
• TXNL4/ERp46 (ERp46 = TXNDC5) is required for EDEM3 activity (PMID:29784879). It is a plausible modifier, but this is unproven.
• Epigenetic and chromosomal findings: none, apart from the UPD1 mechanism above.

5. Environmental Information

Not applicable. No toxins, lifestyle factors or infectious triggers are described. Tunicamycin is used only as an experimental ER-stress inducer. This section should be left empty rather than filled.

6. Mechanism / Pathophysiology

Ordered causal chain

1. Biallelic truncating or missense EDEM3 variants lead to NMD of mRNA (about 17–18% residual) and absent EDEM3 protein. [Demonstrated: patient LCLs and fibroblasts, PMID:34143952]
2. Loss of ER α1,2-mannosidase activity (GH47; needs a Cys82–Cys441 disulfide and is activated by ERp46) leads to failure to trim Man8GlcNAc2 isomer B to Man7/6/5. [Demonstrated: patient fibroblasts, patient plasma and Edem3-KO mouse plasma/brain (PMID:34143952); biochemistry (PMID:34698634, 29784879, 16431915)]
3. Branch 2a, global N-glycome change (secreted and cellular glycoproteins): lower M3–M7, relatively preserved or raised M8/M9, ↑M9:M3 and ↓M3:M4. This is the diagnostic biomarker. [Demonstrated]
4. Branch 2b, M5 → M4 trimming deficit with Glc1Man5 accumulation. The authors speculate that lipid-linked oligosaccharide (LLO) synthesis is impaired. [Observation demonstrated; LLO mechanism speculative]
5. Lack of the α1,6-mannose-exposing glycan degron (M7A/M6/M5) leads to reduced recognition by OS-9/XTP3-B and slower gpERAD of misfolded glycoproteins. [Inferred for patients. Demonstrated in EDEM1/3 double-KO HCT116 cells (PMID:34698634); patient-cell ERAD-substrate kinetics not measured]
6. Altered ER proteostasis and unfolded protein response (UPR): patient LCLs show blunted induction of PERK (EIF2AK3) by tunicamycin. [Demonstrated in vitro, n=3 vs n=3; direction is ambiguous — "impaired UPR" or "increased capacity to eliminate misfolded proteins"]
7. Branch 4b (contrasting model): EDEM3-KO HepaRG hepatic cells show increased BiP/PERK/p-eIF2α, rising apoptosis and loss of viability with passage (PMID:39838427, a cancer cell line). The UPR direction therefore appears to depend on cell context.
8. Neurodevelopmental dysfunction leads to DD/ID, speech delay and hypotonia, and dysmorphism arises through an undetermined craniofacial developmental route. [Not demonstrated. No neural cell or organoid model of EDEM3-CDG exists; MRI is normal]

   "Further functional studies are necessary to determine the precise pathophysiological mechanism of EDEM3-CDG." (PMID:34143952)

Key supporting quotes

• Enzymatic step:

  "Experiments in human fibroblast cell lines, human plasma, and mouse plasma and brain tissue demonstrated decreased trimming of Man8GlcNAc2 isomer B to Man7GlcNAc2, consistent with loss of EDEM3 enzymatic activity." (PMID:34143952)

• M5 trimming:

  "In human cells, Man5GlcNAc2 to Man4GlcNAc2 conversion is also diminished with an increase of Glc1Man5GlcNAc2." (PMID:34143952)

• UPR:

  "Furthermore, analysis of the unfolded protein response showed a reduced increase in EIF2AK3 (PERK) expression upon stimulation with tunicamycin as compared to controls, suggesting an impaired unfolded protein response." (PMID:34143952)

• EDEM3 as the main second-step enzyme:

  "Thus, EDEM3 is a major α1,2-mannosidase for the second step from M8B." (PMID:34698634, IN_VITRO)

• Disulfide requirement:

  "Results showed that the mutations C160A and C529A of EDEM1 as well as C82A and C441A of EDEM3 indeed inactivated EDEM1 and EDEM3, respectively, in gpERAD" (PMID:34698634)

• ERp46 partner:

  "the mannose-trimming activity of EDEM3 toward the model misfolded substrate, the glycoprotein T-cell receptor α locus (TCRα), was reconstituted only when ERp46 had established a covalent interaction with EDEM3." (PMID:29784879)

• ERAD enhancement:

  "EDEM3 accelerates glycoprotein ERAD in transfected HEK293 cells, as shown by increased degradation of misfolded alpha1-antitrypsin variant (null (Hong Kong)) and of TCRalpha." (PMID:16431915)

• Trimming beyond misfolded proteins:

  "Overexpression of EDEM3 also greatly stimulates mannose trimming not only from misfolded alpha1-AT null (Hong Kong) but also from total glycoproteins" (PMID:16431915)

• M9 substrate (in vitro):

  "EDEM3 can convert an asparagine-linked M9 glycan to M8 and M7 glycans in contrast to glycine-linked M9 glycan, and the activity is enhanced in the presence of ERp46" (PMID:35500441)

• Loss in hepatic cells:

  "Conversely, cell depletion of EDEM3 resulted in significant ER stress inducing pro-apoptotic mechanisms and cell death." (PMID:39838427, IN_VITRO, HCC context)

Ontology suggestions (all require lookup)

• GO biological process:
• ERAD pathway GO:0036503
• protein alpha-1,2-demannosylation GO:0036508
• protein N-linked glycosylation GO:0006487
• response to endoplasmic reticulum stress GO:0034976
• PERK-mediated unfolded protein response GO:0036499
• ER mannose trimming involved in glycoprotein ERAD pathway (check whether a specific term exists)
• GO molecular function: mannosyl-oligosaccharide 1,2-alpha-mannosidase activity GO:0004571.
• GO cellular component: endoplasmic reticulum lumen GO:0005788.
• CL: fibroblast CL:0000057 (skin fibroblasts studied); B cell–derived lymphoblastoid lines (bind to B cell, CL:0000236, with a note); neuron CL:0000540 (inferred target tissue only).
• Biological scale: step 1 MOLECULAR; step 2 MOLECULAR; step 3 CELLULAR; step 4 CELLULAR; step 5 ORGANISM.
• Candidate modules: check just list-modules for an ERAD/UPR or glycosylation module before creating any conforms_to.

7. Anatomical Structures Affected

• Organ/system: central nervous system (functional, with no structural MRI lesion), craniofacial region and GI tract (reflux, feeding).
• Subcellular: ER lumen, where EDEM3 localizes with PDI:

  "showing the expected EDEM3 localization within the ER compartment, as revealed by significant overlapping with the ER marker, PDI" (PMID:39838427)

• UBERON suggestions (verify): brain UBERON:0000955, face UBERON:0001456, esophagus UBERON:0001043.
• Lateralization: not applicable; the facial features are symmetric. The single Poland sequence case is unilateral by definition and unverified.

8. Temporal Development

• Onset: congenital or infantile. Feeding difficulty appears in infancy, and delays are recognized in early childhood.
• Course: apparently static neurodevelopmental disorder. No regression is reported. The oldest reported patient was 33 years old (CDG Hub).
• No staging, no remission, no longitudinal natural-history data. The critical period is early childhood, for developmental therapy.

9. Inheritance and Population

• Inheritance: autosomal recessive (HP:0000007), with carrier parents confirmed by Sanger sequencing:

  "The unaffected parents were all heterozygous carriers." (PMID:34143952)

• Penetrance and expressivity: apparently complete for DD/ID. Dysmorphism and hypotonia vary (about 50%).
• Founder effect: c.1859del in the Portuguese Romani population (1/96 control carrier; shared 3.16 Mb homozygous haplotype).
• Prevalence: unknown. There are 12 reported cases. Use measure_type: CASES_IN_LITERATURE with prevalence_class: ULTRA_RARE or NOT_YET_DOCUMENTED.
• Sex ratio, anticipation, mosaicism: not reported, not applicable and not reported, respectively.
• Carrier frequency: not estimated. gnomAD pLoF counts could be queried.

10. Diagnostics

• Genetic: exome or genome sequencing is how every case was found (exome + GeneMatcher). EDEM3 is included on some CDG panels. Chromosomal microarray and SNP array can reveal UPD1 or runs of homozygosity.
• Biochemical (key): semiquantitative plasma N-glycan profiling (MALDI/LC-MS).

  "M9:M3 was increased in all 12 affected individuals." (PMID:34143952)

"M6:M9 ratio provided the highest discrimination between tested obligate heterozygotes (parents, n = 4) and affected individuals (n = 12; Table S3)." (PMID:34143952)

"Therefore, the combination of high M9:M3 and low M3:M4 ratios might also provide diagnostic clues for EDEM3-CDG when M5:M9 and M6:M9 ratios are normal." (PMID:34143952) - Reference ratios reported: M3:M4 normal 0.39–0.56; M9:M3 normal 1.16–2.92. - The standard CDG screen misses it: "Of note, human transferrin was normally glycosylated in the common clinical screening test for CDG in the three affected individuals from family 3." (PMID:34143952) - Variant validation: "The aberrant plasma N-glycan profile provides a quick, clinically available test for validating variants of uncertain significance that may be identified by molecular genetic testing." (PMID:34143952) - Research assays: fibroblast [2-³H]mannose pulse-chase glycan analysis; EDEM3 immunoblot; qPCR. - Imaging: brain MRI is normal, which usefully separates it from many CDGs with cerebellar hypoplasia (e.g., PMM2-CDG). - Differential diagnosis: - Other type II/ERAD-related CDGs with normal transferrin: MAN1B1-CDG (which has a transferrin abnormality), MOGS-CDG, and PMM2-/ALG-CDG (distinguished by ↑M3/M4). - Non-specific ID syndromes. - Screening: no newborn screening. Carrier and prenatal testing are possible once familial variants are known. Targeted carrier testing is a consideration in the Portuguese Romani community.

11. Outcome / Prognosis

• Survival, mortality and QoL data: none.
• Survival into adulthood is documented (age 33).
• Morbidity comes from ID, speech impairment and feeding problems; one patient needed a gastrostomy.
• No prognostic biomarkers.

12. Treatment

There is no disease-specific therapy; care is supportive and multidisciplinary (CDG Hub, FCDGC). NCIT candidates below come from the CLAUDE.md list; confirm each by lookup.

• Experimental: NCT04199000, the FCDGC natural-history study. It is an observational CDG study (dietary interventions are explored generally) and not EDEM3-specific. Fetch it before citing.
• No gene, RNA or targeted therapy exists.
• Caution on pharmacology: kifunensine inhibits ER α1,2-mannosidases, so it would phenocopy the defect rather than treat it. EDEM3 inhibition has been proposed for hypertriglyceridemia (PMID:32213464) and cancer (PMID:39838427). That therapeutic vector runs opposite to this disease.

13. Prevention

• Primary prevention is limited to genetic counseling, cascade carrier testing, and prenatal or preimplantation diagnosis for known familial variants.
• Tertiary prevention covers early developmental intervention and feeding and reflux management.
• No vaccination, public-health or environmental measures apply.

14. Other Species / Natural Disease

• No naturally occurring animal disease has been reported (OMIA not checked in this session).
• Orthologs: mouse Edem3; yeast HTM1/MNL1, whose activity likewise needs a PDI partner (Pdi1p), paralleling ERp46 (PMID:29784879); C. elegans edem-3.
• The mechanism is conserved across eukaryotes:

  "This mechanism is conserved among eukaryotes, and mannose trimming from N-glycans is crucial for the degradation of glycoproteins by ER-associated degradation (ERAD)" (PMID:29784879)

• Not zoonotic.

15. Model Organisms and Experimental Models

Model gap: there is no iPSC-derived neuronal or organoid model of EDEM3-CDG.

Evidence-source summary for curation

Gaps for the entry: - No GeneReviews chapter (confirm with just check-genereviews). - No natural-history, prevalence, prognosis or treatment data. - The neural mechanism is unmodeled. - The direction of the UPR change is unresolved: blunted in patient LCLs, activated in KO hepatic cells. This is a good KNOWLEDGE_GAP discussion.

Sources: - Local references_cache/ records: PMID_34143952, PMID_16431915, PMID_29784879, PMID_34698634, PMID_35500441, PMID_39838427 - CDG Hub – EDEM3-CDG (CDG-IIv) - FCDGC – EDEM3-CDG - OLS – MONDO:0030423 - EDEM3 Modulates Plasma Triglyceride Level through LRP1 (PMC7093811 / PMID:32213464) - Exome-wide association study of plasma lipids, Nat Genet 2017 - CDG state of the art 2022, Orphanet J Rare Dis - PubMed E-utilities search for "EDEM3"

Reference Validation

Checked with linkml-reference-validator 0.3.0rc3.

All extracted references resolved successfully.

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:0051050 (1 mention) - the report calls it "DOID"; DOID calls it congenital disorder of glycosylation type IIv
• CL:0000540 (1 mention) - the report calls it "inferred target tissue only"; CL calls it neuron
• NCIT:C15747 (1 mention) - the report calls it "a gastrostomy procedure term needs lookup"; NCIT calls it Supportive Care
• NCIT:C15986 (1 mention) - the report calls it "agent unspecified in the source"; NCIT calls it Pharmacotherapy

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:

• CL:0000057 (1 mention) - the report calls it "skin fibroblasts studied"; CL calls it fibroblast

Prefixes with no resolver

Terms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: ORPHA.