HNRNPU-Related Developmental and Epileptic Encephalopathy

Mendelian MONDO:0033363 Pathograph 16 Show in embeddings browser Neurodevelopmental Disorder Genetic Disease

HNRNPU-related developmental and epileptic encephalopathy (DEE54) is an autosomal dominant neurodevelopmental disorder caused by de novo heterozygous loss-of-function variants in HNRNPU, or by 1q44 microdeletions encompassing it. HNRNPU encodes heterogeneous nuclear ribonucleoprotein U, also called scaffold attachment factor A - an abundant nuclear RNA- and DNA-binding protein that oligomerises with RNA into a dynamic nuclear mesh and so acts as a bridge between chromatin, nascent transcripts, and the nuclear matrix. It is therefore not a channel, receptor, or enzyme of any single pathway: halving its dose perturbs alternative splicing, chromatin architecture, and transcription at once, and the disease phenotype is what emerges from that diffuse perturbation rather than from one broken step. Affected individuals have developmental delay in around 95% of cases and seizures in 83-95%, with intellectual disability, craniofacial dysmorphism, hypotonia, and nonspecific brain MRI findings including ventriculomegaly and a thin corpus callosum. Seizures are commonly tonic-clonic or absence and may be refractory.

Ask OpenScientist

Ask a research question about HNRNPU-Related Developmental and Epileptic Encephalopathy. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

1
Inheritance
12
Pathophys.
13
Phenotypes
4
Gaps
16
Pathograph
1
Genes
7
Medical Actions
3
Models
1
References
1
Deep Research
👪

Inheritance

1
Autosomal dominant HP:0000006
HNRNPU-NDD is expressed in an autosomal dominant manner and is typically caused by a de novo variant, so recurrence risk to siblings is low. Presumed parental germline mosaicism has been reported in one family with two affected siblings, which is why the risk is low rather than negligible.
Autosomal dominant inheritance
Show evidence (2 references)
PMID:35274911 SUPPORT Human Clinical
"HNRNPU-NDD is expressed in an autosomal dominant manner and typically caused by a de novo HNRNPU pathogenic variant."
GeneReviews states the inheritance mode and the de novo origin.
PMID:35274911 SUPPORT Human Clinical
"Presumed parental germline mosaicism has been reported in one family with two affected sibs."
The documented exception that keeps recurrence risk above zero.
?

Discussions and Knowledge Gaps

4
Which of hnRNP U's functions - alternative splicing, chromatin architecture via the SAF-A/RNA nuclear mesh, or transcriptional regulation - is the dose-sensitive one that produces the neurodevelopmental phenotype, and are they separable at all?
KNOWLEDGE GAP OPEN gap_which_hnrnpu_function_drives_the_phenotype
Attached to
pathophysiology#Disrupted Alternative Splicing Program pathophysiology#Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture pathophysiology#Widespread Transcriptional Dysregulation
The pathograph here carries three parallel molecular arms because the protein has three documented functions, not because three have been shown to matter for disease. No study has separated them: the splicing evidence comes from a conditional truncation that removes all functions at once, and the chromatin literature explicitly says a mechanistic understanding is lacking. The domain architecture offers a route, since the DNA/RNA-binding, oligomerisation, and protein-interaction functions map to separable domains, so this is a tractable question rather than a permanent unknown. It matters because a therapy would have to target the dose-sensitive arm, and at present a screen would have no defined molecular endpoint to screen against.
Proposed experiments
Domain-selective HNRNPU separation-of-function alleles in human neurons
exp_domain_separation_of_function_alleles
Generate isogenic iPSC lines carrying alleles that selectively disable the DNA/RNA-binding domain, the ATP-dependent oligomerisation function required for mesh formation, or the RNA-binding motifs, and compare splicing changes, chromatin accessibility, transcriptional dysregulation, and progenitor viability across them, to establish which function the neurodevelopmental phenotype tracks.
Show evidence (3 references)
PMID:34823151 SUPPORT Other
"This mesh is thought to regulate nuclear and chromatin architecture, yet a mechanistic understanding is lacking."
States directly that the chromatin arm's mechanism is not understood.
PMID:34823151 SUPPORT Other
"we infer possible links between diseases emerging from SAF-A mutations and its role in chromatin organisation and regulation"
The link from the chromatin function to disease is explicitly an inference, not a demonstrated result.
PMID:35864088 SUPPORT Model Organism
"The HNRNPU protein binds to RNA, DNA, and other proteins, and these interactions facilitate its functions in organizing and stabilizing nuclear chromatin, regulating gene transcription"
Enumerates the multiple functions that the disease models cannot currently disentangle.
How much of the mouse cortical phenotype - rapid progenitor death and near-complete loss of cortical structures - reflects human HNRNPU disease, given that patients are heterozygous with nonspecific MRI findings and only 14-20% have microcephaly?
HUMAN MODEL MISMATCH OPEN mismatch_conditional_truncation_versus_human_haploinsufficiency
Attached to
pathophysiology#Neural Progenitor Cell Death pathophysiology#Loss of Cortical Structure
The mouse model is a conditional truncation that removes HNRNPU function outright, and complete knockout is embryonic lethal. Human disease is heterozygous, and the great majority of patients do not have microcephaly at all, which is hard to reconcile with a mechanism whose defining feature in the model is rapid, near-complete progenitor elimination. Either the human dose is far enough above the death threshold that the cell-death arm contributes little, or it operates at a much lower intensity than the model shows. The heterozygous human organoid does reach the same endpoint - it shows significant reductions in progenitor populations and generates significantly smaller organoids - so the mismatch is narrower than a bare dose argument suggests: progenitor loss is real at the human gene dose, and the open question is one of magnitude rather than of kind. The dose-matched heterozygous mouse now bears on this directly, and it lands on the patient side: brain size, callosal morphology, cortical thickness and hippocampal width are all unchanged, so an in vivo animal at the human genotype shows no gross structural phenotype at all. That measurement carries its own caveat, since it is gross morphometry at postnatal day 0 and the authors call for higher-resolution work, and the same mouse loses only about a fifth to a quarter of cortical hnRNP U rather than half - which may itself be why the phenotype is absent. What remains unexplained is why organoid size reduction is universal across mutant lines while microcephaly reaches only 14-20% of patients and the heterozygous mouse has none; the difference between an organoid and an intact brain, where compensation and continued growth are both available, is the obvious place to look.
Proposed experiments
Progenitor viability in heterozygous rather than truncated models
exp_heterozygous_progenitor_viability
Measure neural progenitor apoptosis and proliferation directly in Hnrnpu-heterozygous mouse cortex and in isogenic heterozygous human organoids, using the same time-lapse and marker readouts as the conditional truncation study, to establish whether the cell-death arm operates at all at the human gene dose.
Show evidence (6 references)
PMID:35864088 SUPPORT Model Organism
"HNRNPU is critical in mammalian development, and knockout mice exhibit early lethality"
Establishes that the model sits at a dose no human patient occupies.
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
Most patients do not have microcephaly, which is the observation the progenitor-death mechanism has to be reconciled with.
PMID:36594023 SUPPORT In Vitro
"we performed compositional analyses and identified significant reductions in progenitor populations"
Corrects a natural but wrong assumption: the heterozygous human organoid does report progenitor loss, not only transcriptional change.
+ 3 more references
Do human brain organoids model HNRNPU-related disorder only within a narrow embryonic developmental window, and if so what system should be used to study the postnatal seizure phenotype that dominates the clinical picture?
HUMAN MODEL MISMATCH OPEN gap_organoid_developmental_window_limits
Attached to
pathophysiology#Widespread Transcriptional Dysregulation pathophysiology#Cortical Network Hyperexcitability
The organoid study found co-dysregulation enriched against embryonic but explicitly not perinatal mouse cortex, and the authors conclude that organoids may only be suitable for certain cell types within a specific developmental window. That is a problem for this disorder specifically, because seizures affect 83-95% of patients and are a postnatal phenomenon, while the one human model system appears to lose fidelity exactly as development moves past the embryonic period. A therapeutic screen run in organoids would therefore be optimizing against an embryonic transcriptional signature while the target phenotype is postnatal epilepsy.
Proposed experiments
Extended organoid time course against postnatal cortex
exp_extended_organoid_timecourse
Profile HNRNPU-heterozygous organoids at successively later maturation stages, including sliced or vascularized long-term cultures, and test co-dysregulation against perinatal and postnatal mouse cortex, to determine whether fidelity can be recovered at later stages or whether a different system is required for the seizure phenotype.
Show evidence (2 references)
PMID:36594023 SUPPORT In Vitro
"Thus, hnRNPU deficient human organoids may only be suitable to model transcriptional dysregulation in certain cell types within a specific developmental time window."
The authors state the limitation of their own model system.
PMID:36594023 SUPPORT In Vitro
"An emerging drug discovery approach for neurodevelopmental disorders is to characterize transcriptome-wide dysregulation in an appropriate model system and screen therapeutics based on their capacity to restore functionally relevant expression patterns."
Describes the screening paradigm whose validity depends on the model matching the target developmental stage.
Why is sodium valproate reported as frequently effective in HNRNPU-related disorder, when seizures in most developmental and epileptic encephalopathies are refractory, and does that reflect a mechanistic relationship between valproate's chromatin effects and the HNRNPU lesion?
OPEN QUESTION OPEN gap_valproate_efficacy_mechanism
Attached to
pathophysiology#Recurrent Seizures pathophysiology#Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture
GeneReviews describes valproate as often used and frequently effective here, which stands out against the pharmacoresistance typical of DEEs. Valproate also has chromatin-level activity as a histone deacetylase inhibitor, and the HNRNPU lesion is partly a chromatin-architecture lesion, so a mechanistic rather than coincidental relationship is worth asking about. The question is speculative and is recorded as an open question, not a hypothesis with supporting evidence: it may simply be that this disorder's seizure types, commonly tonic-clonic and absence, are ones valproate treats well in any aetiology. Distinguishing those two explanations would need a comparison against matched seizure types in other DEEs.
Proposed experiments
Valproate response rate against seizure-type-matched DEE controls
exp_valproate_response_versus_matched_dees
Compare valproate response rates in HNRNPU-related disorder against other DEEs matched for seizure semiology, to test whether the reported efficacy exceeds what the seizure types alone predict; if it does, assay whether valproate's HDAC inhibition normalizes chromatin accessibility in HNRNPU-heterozygous human neurons.
Show evidence (2 references)
PMID:35274911 SUPPORT Human Clinical
"Standard treatment of seizures with anti-seizure medications (sodium valproate is often used and is frequently effective)"
The observation of unusual valproate efficacy that prompts the question.
PMID:35274911 SUPPORT Human Clinical
"Affected individuals are likely to experience seizures (most commonly tonic-clonic or absence) that may be refractory to treatment."
Gives the seizure semiology behind the simpler alternative explanation, and notes that refractoriness still occurs, so the efficacy is not universal.
⚙

Pathophysiology

12
Heterozygous HNRNPU Loss-of-Function Lesion
A de novo heterozygous loss-of-function point variant in HNRNPU, or a 1q44 microdeletion encompassing the gene, is present. Both routes reduce HNRNPU to a single functional copy, which is the starting point this node records - but they are not the same lesion, because a 1q44 deletion also removes AKT3 and ZBTB18 (see the genetic section). This node records the HNRNPU lesion only, not its consequences.
neuron CL:0000540 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves neuron (CL:0000540). CL:0000540 is a cell type from the Cell Ontology.
Downstream
Reduced hnRNP U Protein Dose One functional allele produces roughly half the normal protein.
Show evidence (2 references)
PMID:35864088 SUPPORT Human Clinical
"Microdeletions in the 1q44 locus encompassing HNRNPU and other genes and point mutations in HNRNPU cause brain disorders, including early-onset seizures and severe intellectual disability."
Establishes both lesion types as causes of the same clinical picture.
PMID:36594023 SUPPORT Human Clinical
"De novo loss-of-function variants in HNRNPU have been found to cause a neurodevelopmental syndrome"
Confirms the de novo loss-of-function mechanism.
Reduced hnRNP U Protein Dose
Haploinsufficiency for an abundant, ubiquitously expressed nuclear protein. Because hnRNP U is required for mammalian development - complete knockout mice die early - the disease is specifically a dose problem rather than an absence problem, and the relevant question is which of its several functions are most dose-sensitive.
Downstream
Disrupted Alternative Splicing Program hnRNP U participates directly in spliceosomal RNA processing.
Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture hnRNP U oligomerises with RNA to form the nuclear mesh that organizes chromatin; this is a separate function from splicing.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"HNRNPU is critical in mammalian development, and knockout mice exhibit early lethality"
Establishes that complete loss is lethal, so human disease is a partial-dose state.
Disrupted Alternative Splicing Program
Expression and alternative splicing of many genes are altered, with the affected transcripts concentrated in cell survival, cell motility, and synapse formation. This is the post-transcriptional arm.
regulation of alternative mRNA splicing, via spliceosome GO:0000381 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves abnormal regulation of alternative mRNA splicing, via spliceosome (GO:0000381). GO:0000381 is a biological process from the Gene Ontology. ⚠ ABNORMAL
Downstream
Neural Progenitor Cell Death Mis-splicing of cell-survival regulators precedes the death of progenitors, and the pharmacological rescue data support a causal rather than incidental link.
Impaired Radial Neuronal Migration Mis-splicing of cell-motility regulators accompanies the migration defect.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"expression and alternative splicing of multiple genes involved in cell survival, cell motility, and synapse formation are affected following Hnrnpu's conditional truncation"
Names the splicing consequence and the functional categories affected.
Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture
hnRNP U oligomerises in an ATP-dependent manner and binds RNA to form a dynamic nuclear mesh that regulates chromatin organization in interphase and mitosis. Reduced protein destabilizes that mesh. Nothing in a splicing assay measures this arm, which is why it is modeled separately.
chromatin organization GO:0006325 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves abnormal chromatin organization (GO:0006325). GO:0006325 is a biological process from the Gene Ontology. ⚠ ABNORMAL
Downstream
Widespread Transcriptional Dysregulation Altered chromatin architecture is one proposed route to the observed transcriptional changes, though the mechanism is not established.
Show evidence (1 reference)
PMID:34823151 SUPPORT Other
"SAF-A can oligomerise in an ATP-dependent manner and interact with RNA to form a dynamic nuclear mesh."
States the oligomerisation and mesh-formation function that reduced protein dose compromises.
Widespread Transcriptional Dysregulation
Transcriptome-wide changes across functionally relevant pathways. The dysregulation is notable for being broad but individually modest rather than a small number of large effects, which is why no single downstream gene explains the phenotype.
regulation of DNA-templated transcription GO:0006355 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves abnormal regulation of DNA-templated transcription (GO:0006355). GO:0006355 is a biological process from the Gene Ontology. ⚠ ABNORMAL
Experimental models
Isogenic HNRNPU heterozygous human iPSC-derived brain organoid ORGANOID The only human model of the transcriptional consequence, and the one that establishes cross-species convergence with embryonic mouse cortex.
Downstream
Impaired Synapse Formation and Circuit Maturation Dysregulation of synaptic transcripts is proposed to degrade circuit assembly.
Show evidence (1 reference)
PMID:36594023 SUPPORT In Vitro
"We identified widespread dysregulation in functionally relevant pathways"
Documents the breadth of transcriptional dysregulation in a human model.
Neural Progenitor Cell Death
Loss of HNRNPU function causes rapid cell death of both postmitotic neurons and neural progenitors, with progenitors apparently the more sensitive population. This differential vulnerability matters: it means the lesion subtracts cells that would each have produced many neurons, not just the cells directly lost.
neural progenitor cell CL:0011020 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves neural progenitor cell (CL:0011020). CL:0011020 is a cell type from the Cell Ontology.
apoptotic process GO:0006915 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased apoptotic process (GO:0006915). GO:0006915 is a biological process from the Gene Ontology. ↑ INCREASED
Experimental models
Isogenic HNRNPU heterozygous human iPSC-derived brain organoid ORGANOID Reaches the progenitor endpoint at the heterozygous human gene dose, which the mouse conditional truncation cannot do.
Animal models
Hnrnpu conditional cortical knockout mouse Mouse recapitulates Provides the direct evidence for progenitor-selective cell death and, through the rescue experiments, for its causal role in the cortical phenotype.
Downstream
Loss of Cortical Structure Depleting the progenitor pool removes the cells that build the cortex.
Show evidence (4 references)
PMID:35864088 SUPPORT Model Organism
"HNRNPU loss of function leads to rapid cell death of both postmitotic neurons and neural progenitors, with an apparent higher sensitivity of the latter"
Reports the cell death and the progenitor-selective vulnerability.
PMID:35864088 SUPPORT Model Organism
"We suppressed cell death by pan-Caspase inhibitors, p53 inhibitors, and necroptosis inhibitors, demonstrating the involvement of p53-dependent canonical and noncanonical death mechanisms in neural progenitors"
Identifies the death mechanism as p53-dependent through both canonical and non-canonical routes, established by pharmacological suppression rather than correlation.
PMID:35864088 SUPPORT Model Organism
"We showed that reducing the levels of the splicing factor SRSF3 opposes HNRNPU’s loss of function effects and improves neural progenitors’ viability and neuronal migration"
Lowering the competing splicing factor SRSF3 rescues progenitor viability and migration, which ties the death back to the splicing arm specifically.
+ 1 more reference
Impaired Radial Neuronal Migration
Neurons that survive migrate abnormally along the radial axis. This is a positioning defect distinct from the cell-death arm, and it is separately rescuable in the model.
neuron CL:0000540 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves neuron (CL:0000540). CL:0000540 is a cell type from the Cell Ontology.
Downstream
Loss of Cortical Structure Mispositioned neurons contribute to disturbed cortical architecture.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"ameliorate radial neuronal migration defects"
The rescue experiments identify radial migration as a separately measurable and separately correctable defect.
Loss of Cortical Structure
Cortical structures are lost in the mutant embryonic brain. In the mouse conditional knockout this is severe; the human heterozygous condition produces the much milder nonspecific findings seen on MRI, so this node represents the same process at a different dose.
cell population proliferation GO:0008283 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves decreased cell population proliferation (GO:0008283). GO:0008283 is a biological process from the Gene Ontology. ↓ DECREASED
Animal models
Hnrnpu+/113DEL heterozygous mouse Mouse fails to recapitulate No gross cortical phenotype appears at the human gene dose. Brain size, corpus callosum morphology, cortical thickness and hippocampal width are all unchanged, in sharp contrast to the near-complete cortical loss the conditional truncation produces.
Downstream
Impaired Synapse Formation and Circuit Maturation A structurally compromised cortex assembles circuits abnormally.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"we identified pharmaceutical and genetic agents that can partially reverse the loss of cortical structures in Hnrnpu mutated embryonic brains"
Names the loss of cortical structures as the tissue-level outcome.
Impaired Synapse Formation and Circuit Maturation
Synapse formation is among the functional categories whose genes are mis-spliced, and the resulting circuits mature abnormally. This is the synaptic-level convergence point for the splicing, transcriptional, and structural arms.
neuron CL:0000540 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves neuron (CL:0000540). CL:0000540 is a cell type from the Cell Ontology.
Downstream
Cortical Network Hyperexcitability Abnormally assembled circuits are epileptogenic, though the specific excitatory-inhibitory imbalance has not been characterized in this disorder.
Developmental Delay and Intellectual Disability Disturbed circuit maturation underlies the cognitive phenotype.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"expression and alternative splicing of multiple genes involved in cell survival, cell motility, and synapse formation are affected"
Names synapse formation among the affected functional categories.
Cortical Network Hyperexcitability
Cortical networks become seizure-prone. In patients this is inferred from the clinical seizure frequency rather than measured directly; no electrophysiological characterization of the human excitatory-inhibitory balance in this disorder has been published. The direct excitability measurement comes instead from the dose-matched heterozygous mouse, in which the threshold for electrically evoked seizures is significantly reduced.
neuron CL:0000540 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves neuron (CL:0000540). CL:0000540 is a cell type from the Cell Ontology.
Animal models
Hnrnpu+/113DEL heterozygous mouse Mouse recapitulates Supplies the only direct excitability measurement at the human gene dose. The lowered electroconvulsive threshold is a measurement of network excitability rather than an inference from seizure counts, which is what the patient evidence on this node otherwise rests on.
Downstream
Recurrent Seizures A hyperexcitable cortex generates recurrent seizures.
Show evidence (2 references)
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
Establishes the very high seizure frequency that this node accounts for.
PMID:37782669 SUPPORT Model Organism
"Hnrnpu+/113DEL mice demonstrated a significantly lower threshold for induction of maximal tonic hindlimb extension seizures, consistent with a greater seizure predisposition"
The one direct excitability measurement available at the human gene dose: a lowered evoked-seizure threshold is the operational definition of network hyperexcitability this node asserts.
Recurrent Seizures
Seizures affect 83-95% of individuals, most commonly tonic-clonic or absence, and may be refractory to treatment.
Animal models
Hnrnpu+/113DEL heterozygous mouse Mouse fails to recapitulate The clinical endpoint does not appear. Patients have seizures in 83-95% of cases, but the dose-matched mouse never develops spontaneous ones - only a lowered threshold for evoked seizures. The model reproduces seizure susceptibility, not epilepsy.
Downstream
Developmental Delay and Intellectual Disability Seizure burden plausibly compounds the developmental phenotype, which the splicing and structural arms already produce independently.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Affected individuals are likely to experience seizures (most commonly tonic-clonic or absence) that may be refractory to treatment."
GeneReviews gives the seizure semiology and refractoriness.
Developmental Delay and Intellectual Disability
Developmental delay in around 95% of individuals and intellectual disability - typically moderate to severe - with speech and language delay or absent speech. Two upstream routes converge here, the developmental lesion and the seizure burden, and the published data do not separate their contributions.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"HNRNPU-related neurodevelopmental disorder (HNRNPU-NDD) is characterized by developmental delay and intellectual disability"
GeneReviews defines the core developmental phenotype.
⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.
Pathograph: causal mechanism network for HNRNPU-Related Developmental and Epileptic Encephalopathy Interactive directed graph showing how pathophysiology mechanisms, phenotypes, genetic factors and variants, experimental models, environmental triggers, and treatments relate through causal and linked edges.
●

Phenotypes

13
Digestive 1
Feeding Difficulties HP:0011968 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Feeding difficulties (HP:0011968). HP:0011968 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"There may be feeding difficulties during the neonatal period as well as hypotonia"
GeneReviews documents neonatal feeding difficulties.
Eye 1
Extraneurological Features Strabismus HP:0000486 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Strabismus (HP:0000486). HP:0000486 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Less common findings include cardiac abnormalities, strabismus, undescended testes in males, renal anomalies, and skeletal features, including joint laxity, polydactyly, and scoliosis"
GeneReviews names the extraneurological cluster, including strabismus, which this node is bound to as its most frequently reported member.
Head and Neck 2
Craniofacial Dysmorphism FREQUENT Abnormal facial shape HP:0001999 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Abnormal facial shape (HP:0001999). HP:0001999 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
The reported range 54-95% straddles the 30-79% frequent and 80-99% very frequent bands; the lower bound is taken so the band is not over-claimed.
PMID:35274911 SUPPORT Human Clinical
"Dysmorphic features have been described but they are nonspecific."
GeneReviews qualifies the diagnostic value of the dysmorphism.
Microcephaly OCCASIONAL HP:0000252 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Microcephaly (HP:0000252). HP:0000252 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
14-20% falls entirely within the 5-29% occasional band.
PMID:35864088 SUPPORT Human Clinical
"lower penetrance microcephaly, a thin corpus callosum, dysmorphic facial features, and hypotonia"
Independently describes microcephaly as a lower-penetrance feature.
Musculoskeletal 1
Hypotonia HP:0001252 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Hypotonia (HP:0001252), qualified as temporality chronic. HP:0001252 is a phenotype from the Human Phenotype Ontology.
Temporal: CHRONIC
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"There may be feeding difficulties during the neonatal period as well as hypotonia, which often remains lifelong."
GeneReviews describes hypotonia and its persistence.
Nervous System 6
Seizures VERY_FREQUENT HP:0001250 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Seizure (HP:0001250). HP:0001250 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
A reported range of 83-95% sits entirely within the 80-99% very frequent band.
PMID:35274911 SUPPORT Human Clinical
"Affected individuals are likely to experience seizures (most commonly tonic-clonic or absence) that may be refractory to treatment."
GeneReviews corroborates the seizure phenotype and its semiology.
Global Developmental Delay VERY_FREQUENT HP:0001263 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Global developmental delay (HP:0001263). HP:0001263 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
The reported figure of approximately 95% falls in the 80-99% very frequent band.
Intellectual Disability FREQUENT HP:0001249 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Intellectual disability (HP:0001249). HP:0001249 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:36594023 SUPPORT Human Clinical
"with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
52% falls in the 30-79% frequent band.
PMID:35274911 SUPPORT Human Clinical
"developmental delay and intellectual disability"
GeneReviews corroborates intellectual disability as a core feature; its moderate-to-severe severity is stated in the same chapter.
Absent or Delayed Speech Absent speech HP:0001344 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Absent speech (HP:0001344). HP:0001344 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"with speech and language delay and/or absent speech"
GeneReviews names absent speech as a core feature.
Ventriculomegaly HP:0002119 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Ventriculomegaly (HP:0002119). HP:0002119 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Nonspecific brain MRI findings include ventriculomegaly and thinning of the corpus callosum."
GeneReviews names ventriculomegaly.
Autistic Features Autistic behavior HP:0000729 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Autistic behavior (HP:0000729). HP:0000729 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Affected individuals may also display autistic features."
GeneReviews records autistic features as a possible manifestation.
Respiratory 1
Abnormal Breathing Pattern VERY_RARE Apnea HP:0002104 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Apnea (HP:0002104). HP:0002104 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Rarely, abnormal breathing patterns, including hyperventilation and apnea, may be present and can lead to sleep disturbance."
GeneReviews describes this as rare, which maps to the 1-4% very rare band.
Other 1
Thin Corpus Callosum HP:0033725 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Thin corpus callosum (HP:0033725). HP:0033725 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Nonspecific brain MRI findings include ventriculomegaly and thinning of the corpus callosum."
GeneReviews names the callosal finding.
🧬

Genetic Associations

1
HNRNPU (De novo heterozygous loss-of-function variants)
Gene: HNRNPU hgnc:5048 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is HNRNPU (hgnc:5048). hgnc:5048 is a gene from the HUGO Gene Nomenclature Committee. relationship_type: CAUSATIVE variant_origin: DE_NOVO
Show evidence (6 references)
PMID:36594023 SUPPORT Human Clinical
"De novo loss-of-function variants in HNRNPU have been found to cause a neurodevelopmental syndrome"
Establishes HNRNPU as the causal gene and the variant class.
PMID:35864088 SUPPORT Human Clinical
"Microdeletions in the 1q44 locus encompassing HNRNPU and other genes and point mutations in HNRNPU cause brain disorders"
Documents the two lesion routes to the same disorder.
PMID:28283832 SUPPORT Human Clinical
"Our study demonstrates that AKT3 haploinsufficiency is the main driver for microcephaly, whereas HNRNPU alteration mostly drives epilepsy and determines the degree of intellectual disability"
Apportions the 1q43q44 phenotype across contiguous genes, showing HNRNPU drives the epilepsy and the degree of intellectual disability while AKT3 drives the microcephaly.
+ 3 more references
💊

Medical Actions

7
Anti-Seizure Medication
Action: PharmacotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Pharmacotherapy (NCIT:C15986). NCIT:C15986 is a clinical intervention from the NCI Thesaurus. NCIT:C15986
Agent: valproic acid CHEBI:39867 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses valproic acid (CHEBI:39867). CHEBI:39867 is a therapeutic agent from Chemical Entities of Biological Interest.
Standard anti-seizure medication, with sodium valproate often used and frequently effective. Newer-generation agents or the ketogenic diet are considered in refractory cases.
Mechanism Target:
INHIBITS Recurrent Seizures — Sodium valproate is described as frequently effective for the seizures in this disorder.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Standard treatment of seizures with anti-seizure medications (sodium valproate is often used and is frequently effective)"
GeneReviews reports valproate as frequently effective here, which is notable given how often DEE seizures are refractory.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Standard treatment of seizures with anti-seizure medications (sodium valproate is often used and is frequently effective)"
GeneReviews management recommendation.
Ketogenic Diet
Action: Dietary InterventionNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Dietary Intervention (NCIT:C15447). NCIT:C15447 is a clinical intervention from the NCI Thesaurus. NCIT:C15447
Considered together with newer-generation anti-seizure medications for those with refractory seizures.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"consider instituting the ketogenic diet and/or newer generation anti-seizure medications in those with refractory seizures"
GeneReviews recommends the diet for the refractory subgroup.
Respiratory Support for Sleep Apnea
Action: respiratory therapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is respiratory therapy (NCIT:C15322). NCIT:C15322 is a clinical intervention from the NCI Thesaurus. Ontology label: Respiratory Therapy NCIT:C15322
Supplemental oxygen, CPAP, or BiPAP for those with sleep apnea arising from the abnormal breathing pattern.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Consider supplemental oxygen, CPAP, or BiPAP in those with sleep apnea."
GeneReviews management recommendation for the respiratory phenotype.
Feeding Therapy and Enteral Support
Action: nutritional supportNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is nutritional support (NCIT:C15433). NCIT:C15433 is a clinical intervention from the NCI Thesaurus. Ontology label: Nutritional Support NCIT:C15433
Feeding therapy, with a temporary or permanent feeding tube considered for persistent feeding problems.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Feeding therapy; consider a temporary or permanent feeding tube for those with persistent feeding issues."
GeneReviews management recommendation.
Avoidance of Seizure-Inducing Agents and Activities
Action: preventive interventionNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is preventive intervention (NCIT:C15843). NCIT:C15843 is a clinical intervention from the NCI Thesaurus. Ontology label: Preventive Intervention NCIT:C15843
GeneReviews lists activities and agents that may induce seizures under agents and circumstances to avoid.
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Agents/circumstances to avoid: Activities and agents that may induce seizures."
The GeneReviews agents-to-avoid guidance.
Routine Surveillance
Action: Supportive CareNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Supportive Care (NCIT:C15747). NCIT:C15747 is a clinical intervention from the NCI Thesaurus. NCIT:C15747
At each visit: growth parameters, nutritional status and safety of oral intake, monitoring for constipation, new seizures, hyperventilation, apnea, and changes in tone, and assessment of developmental progress and behavior. Annually or as clinically indicated: ophthalmologic and audiologic evaluations - which is also how the strabismus in the extraneurological cluster is expected to be detected.
Show evidence (2 references)
PMID:35274911 SUPPORT Human Clinical
"At each visit: measurement of growth parameters; evaluation of nutritional status and safety of oral intake; monitor for evidence of constipation, new seizures, hyperventilation, apnea, and changes in tone; assessment of developmental progress and behavior"
The GeneReviews per-visit surveillance schedule.
PMID:35274911 SUPPORT Human Clinical
"Annually or as clinically indicated: ophthalmologic and audiologic evaluations"
The GeneReviews annual surveillance schedule.
Genetic Counseling
Action: Genetic CounselingNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Genetic Counseling (NCIT:C15240). NCIT:C15240 is a clinical intervention from the NCI Thesaurus. NCIT:C15240
Counseling for an autosomal dominant, typically de novo condition. Recurrence risk to siblings is hypothesized to be low, with the caveat of a reported presumed germline mosaicism family.
Show evidence (2 references)
PMID:35274911 SUPPORT Human Clinical
"The risk to other family members is hypothesized to be low."
GeneReviews genetic counseling guidance.
PMID:35274911 SUPPORT Human Clinical
"Once the HNRNPU pathogenic variant has been identified in an affected family member, prenatal testing and preimplantation genetic testing are possible"
Records the reproductive options available once the familial variant is known.
🔬

Diagnosis

3
Molecular Genetic Testing for a Heterozygous HNRNPU Variant (Positive in affected individuals)
The diagnosis is established in a proband with suggestive findings and a heterozygous pathogenic HNRNPU variant. Testing must be able to detect a 1q44 microdeletion as well as a point variant, since either causes the disorder.
genetic testing NCIT:C15709 NCI Thesaurus (NCIT)
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"The diagnosis of HNRNPU-NDD is established in a proband with suggestive findings and a heterozygous pathogenic variant in HNRNPU identified by molecular genetic testing."
GeneReviews states the diagnostic criterion directly.
DNA Methylation Episignature
A genome-wide DNA methylation signature specific to HNRNPU-related disorder, developed as a diagnostic biomarker on Infinium Methylation EPIC arrays and useful for reclassifying variants of uncertain significance.
DNA methylation profiling NCIT:C15709 NCI Thesaurus (NCIT)
Show evidence (1 reference)
PMID:37120726 SUPPORT Human Clinical
"We performed genome-wide DNA methylation (DNAm) analysis in a cohort of individuals to develop a diagnostic biomarker and gain functional insights into the molecular pathophysiology of HNRNPU-related disorder"
States the purpose and method of the episignature analysis.
Brain MRI
MRI shows nonspecific findings - ventriculomegaly and thinning of the corpus callosum. These support but do not establish the diagnosis.
brain magnetic resonance imaging NCIT:C16809 NCI Thesaurus (NCIT)
Show evidence (1 reference)
PMID:35274911 SUPPORT Human Clinical
"Nonspecific brain MRI findings include ventriculomegaly and thinning of the corpus callosum."
GeneReviews describes the imaging findings and their nonspecificity.
🧫

Experimental Models

1
Isogenic HNRNPU heterozygous human iPSC-derived brain organoid ORGANOID
Human iPSC-derived brain organoids carrying a heterozygous HNRNPU loss-of-function allele, profiled transcriptome-wide and compared against embryonic and perinatal mouse cortex.
Publication
PMID:36594023
Pathograph links
Neural Progenitor Cell Death recapitulates fidelity: moderate Reaches the progenitor endpoint at the heterozygous human gene dose, which the mouse conditional truncation cannot do.
Limitations: The reduction is measured as population composition and organoid size rather than as direct time-lapse cell death, so the mode of loss is inferred; and the size phenotype is universal across mutant lines while microcephaly affects only 14-20% of patients.
Readouts
Progenitor population composition and organoid size
Direction: DECREASED
Interpretation: Heterozygous HNRNPU loss reduces progenitor populations and organoid size in human tissue.
Show evidence (2 references)
PMID:36594023 SUPPORT In Vitro
"we performed compositional analyses and identified significant reductions in progenitor populations"
Reports the compositional reduction in progenitor populations.
+ 1 more reference
Show evidence (1 reference)
PMID:36594023 SUPPORT In Vitro
"both HNRNPU+/−mutant lines generated significantly smaller organoids, with a size phenotype emerging as early as DIV 10"
Establishes that the heterozygous human system reproduces a growth phenotype.
Widespread Transcriptional Dysregulation recapitulates fidelity: moderate The only human model of the transcriptional consequence, and the one that establishes cross-species convergence with embryonic mouse cortex.
Limitations: Co-dysregulation was enriched against embryonic but explicitly NOT perinatal mouse cortex, so 45-day organoids appear to model only a restricted developmental window and set of cell types; the authors state this limitation themselves.
Readouts
Transcriptome-wide dysregulation in 45-day organoids
Direction: ALTERED
Interpretation: Heterozygous HNRNPU loss produces broad transcriptional dysregulation in human tissue.
Show evidence (1 reference)
PMID:36594023 SUPPORT In Vitro
"We identified widespread dysregulation in functionally relevant pathways"
Reports the transcriptomic measurement behind this readout.
Show evidence (1 reference)
PMID:36594023 SUPPORT In Vitro
"we find enrichment of co-dysregulation between 45-day-old human organoids and embryonic, but not perinatal, mice from distinct models of HNRNPU-related disorder"
Establishes both the cross-species convergence and its developmental limits.
🐁

Animal Models

2
Hnrnpu conditional cortical knockout mouse
Conditional truncation of Hnrnpu in the developing cortex causes rapid death of neural progenitors and postmitotic neurons, radial migration defects, and loss of cortical structures, each partially reversible with pharmacological or genetic intervention.
Species
Mouse
Genotype
Hnrnpu conditional truncation in embryonic cortical neuroepithelium
Publication
PMID:35864088
Pathograph links
Neural Progenitor Cell Death recapitulates fidelity: moderate Provides the direct evidence for progenitor-selective cell death and, through the rescue experiments, for its causal role in the cortical phenotype.
Limitations: This is a conditional truncation removing HNRNPU function outright, whereas human disease is heterozygous loss of function; complete Hnrnpu knockout is embryonic lethal in mice, so the model sits at a far more severe point on the dose-response curve than any human genotype and produces near-complete cortical loss rather than the nonspecific MRI findings seen in patients.
Readouts
Neural progenitor viability after Cre-mediated truncation
Direction: DECREASED
Interpretation: Progenitors die rapidly and more readily than postmitotic neurons.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"HNRNPU loss of function leads to rapid cell death of both postmitotic neurons and neural progenitors, with an apparent higher sensitivity of the latter"
Reports the viability measurement behind this readout.
Cortical structure after rescue intervention
Direction: RESTORED
Interpretation: Pharmacological and genetic agents partially reverse the cortical loss, supporting cell death as its cause rather than a correlate.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"we identified pharmaceutical and genetic agents that can partially reverse the loss of cortical structures in Hnrnpu mutated embryonic brains, ameliorate radial neuronal migration defects and rescue cultured neural progenitors' cell death"
Reports the rescue result behind this readout.
Show evidence (1 reference)
PMID:35864088 SUPPORT Model Organism
"We aimed to understand HNRNPU's roles in the developing brain."
Establishes the model's purpose as characterizing HNRNPU in cortical development.
Hnrnpu+/113DEL heterozygous mouse
A CRISPR-induced constitutive heterozygous truncating deletion in exon 1 of mouse Hnrnpu, targeting the region where at least five human patient truncating variants fall. This is the dose-matched counterpart of the conditional cortical knockout above: it models the human genotype rather than complete loss of function, and it is informative as much for what it fails to reproduce as for what it does.
Species
Mouse
Genotype
Hnrnpu heterozygous 113-bp out-of-frame exon 1 deletion (Hnrnpu+/113DEL)
Publication
PMID:37782669
Pathograph links
Cortical Network Hyperexcitability recapitulates fidelity: moderate Supplies the only direct excitability measurement at the human gene dose. The lowered electroconvulsive threshold is a measurement of network excitability rather than an inference from seizure counts, which is what the patient evidence on this node otherwise rests on.
Limitations: Excitability is measured as susceptibility to an electrically evoked seizure, not as spontaneous epileptiform activity, so it establishes a lowered seizure threshold rather than the ongoing hyperexcitable state patients are presumed to carry. The model also under-represents the human lesion: cortical hnRNP U falls by only about 20-26% rather than the expected half, which the authors read as evidence of compensation, so the measured effect is likely a floor.
Readouts
Maximal seizure electroconvulsive threshold
Direction: DECREASED
Interpretation: Less current is required to evoke a maximal tonic hindlimb extension seizure in heterozygous mutants than in wild-type littermates.
Show evidence (2 references)
PMID:37782669 SUPPORT Model Organism
"Hnrnpu+/113DEL mice demonstrated a significantly lower threshold for induction of maximal tonic hindlimb extension seizures, consistent with a greater seizure predisposition"
Reports the threshold measurement behind this readout.
+ 1 more reference
Show evidence (1 reference)
PMID:37782669 SUPPORT Model Organism
"Heterozygous mutants demonstrated global developmental delay, impaired ultrasonic vocalizations, cognitive dysfunction and increased seizure susceptibility, thus modeling aspects of the human disease."
Establishes the model as informative for the seizure arm, in the authors' own deliberately partial phrasing.
Recurrent Seizures fails to recapitulate fidelity: low The clinical endpoint does not appear. Patients have seizures in 83-95% of cases, but the dose-matched mouse never develops spontaneous ones - only a lowered threshold for evoked seizures. The model reproduces seizure susceptibility, not epilepsy.
Limitations: This is a substantive negative result, not an unexamined gap: over 300 hours of video EEG in adults found no spontaneous epileptiform activity, and no seizure-like behaviour was seen in routine handling at any age. The likeliest reasons are the shortfall in Hnrnpu reduction noted above and species differences in seizure threshold, but neither has been tested, so this link should not be read as evidence against the human mechanism.
Readouts
Spontaneous epileptiform activity on video EEG
Direction: UNCHANGED
Interpretation: No spontaneous generalized epileptiform activity was detected in heterozygous adults despite extended recording.
Show evidence (2 references)
PMID:37782669 SUPPORT Model Organism
"Despite over 300 total hours of video EEG recordings among Hnrnpu+/113DEL adults, there was no evidence of spontaneous generalized epileptiform activity"
The negative EEG result behind this readout, and the reason the link is typed as a failure to recapitulate.
+ 1 more reference
Show evidence (1 reference)
PMID:37782669 SUPPORT Model Organism
"Despite over 300 total hours of video EEG recordings among Hnrnpu+/113DEL adults, there was no evidence of spontaneous generalized epileptiform activity"
Substantiates treating this model as uninformative for the spontaneous-seizure endpoint.
Loss of Cortical Structure fails to recapitulate fidelity: low No gross cortical phenotype appears at the human gene dose. Brain size, corpus callosum morphology, cortical thickness and hippocampal width are all unchanged, in sharp contrast to the near-complete cortical loss the conditional truncation produces.
Limitations: The measurements are gross morphometry at postnatal day 0, and the authors say explicitly that higher-resolution studies able to detect subtler abnormalities are still warranted - so this is evidence against a gross structural phenotype, not against a structural contribution as such. It is nonetheless the most directly relevant datum available for the dose question, since it is the only structural measurement made at the human genotype in vivo.
Readouts
Gross brain morphometry at postnatal day 0
Direction: UNCHANGED
Interpretation: Neither brain size nor callosal morphology differs from wild type at birth.
Show evidence (1 reference)
PMID:37782669 SUPPORT Model Organism
"we did not observe differences in brain size and corpus callosum morphology compared to WT at postnatal day 0"
The negative morphometric result behind this readout.
Cortical thickness and hippocampal width
Direction: UNCHANGED
Interpretation: The two regional measures most relevant to a progenitor-depletion mechanism are also unaffected.
Show evidence (1 reference)
PMID:37782669 SUPPORT Model Organism
"Further examination of Hnrnpu+/113DEL brains also showed no significant change in cortical thickness and hippocampal width"
The negative regional measurement behind this readout.
Show evidence (1 reference)
PMID:37782669 SUPPORT Model Organism
"Further examination of Hnrnpu+/113DEL brains also showed no significant change in cortical thickness and hippocampal width"
Substantiates the failure to recapitulate the cortical-structure node at the heterozygous dose.
{ }

Source YAML

click to show 
name: HNRNPU-Related Developmental and Epileptic Encephalopathy
creation_date: "2026-08-19T00:00:00Z"
category: Mendelian
synonyms:
- DEE54
- EIEE54
- HNRNPU-related neurodevelopmental disorder
- HNRNPU-NDD
- developmental and epileptic encephalopathy 54
description: >-
  HNRNPU-related developmental and epileptic encephalopathy (DEE54) is an
  autosomal dominant neurodevelopmental disorder caused by de novo heterozygous
  loss-of-function variants in HNRNPU, or by 1q44 microdeletions encompassing it.
  HNRNPU encodes heterogeneous nuclear ribonucleoprotein U, also called scaffold
  attachment factor A - an abundant nuclear RNA- and DNA-binding protein that
  oligomerises with RNA into a dynamic nuclear mesh and so acts as a bridge
  between chromatin, nascent transcripts, and the nuclear matrix. It is therefore
  not a channel, receptor, or enzyme of any single pathway: halving its dose
  perturbs alternative splicing, chromatin architecture, and transcription at
  once, and the disease phenotype is what emerges from that diffuse perturbation
  rather than from one broken step. Affected individuals have developmental delay
  in around 95% of cases and seizures in 83-95%, with intellectual disability,
  craniofacial dysmorphism, hypotonia, and nonspecific brain MRI findings
  including ventriculomegaly and a thin corpus callosum. Seizures are commonly
  tonic-clonic or absence and may be refractory.
disease_term:
  preferred_term: developmental and epileptic encephalopathy, 54
  term:
    id: MONDO:0033363
    label: developmental and epileptic encephalopathy, 54
parents:
- Neurodevelopmental Disorder
- Genetic Disease

references:
- reference: PMID:35274911
  title: "HNRNPU-Related Neurodevelopmental Disorder."
  tags:
  - GeneReviews

inheritance:
- name: Autosomal dominant
  description: >-
    HNRNPU-NDD is expressed in an autosomal dominant manner and is typically caused
    by a de novo variant, so recurrence risk to siblings is low. Presumed parental
    germline mosaicism has been reported in one family with two affected siblings,
    which is why the risk is low rather than negligible.
  inheritance_term:
    preferred_term: Autosomal dominant inheritance
    term:
      id: HP:0000006
      label: Autosomal dominant inheritance
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "HNRNPU-NDD is expressed in an autosomal dominant manner and typically caused by a de novo HNRNPU pathogenic variant."
    explanation: >-
      GeneReviews states the inheritance mode and the de novo origin.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Presumed parental germline mosaicism has been reported in one family with two affected sibs."
    explanation: >-
      The documented exception that keeps recurrence risk above zero.

pathophysiology:
- name: Heterozygous HNRNPU Loss-of-Function Lesion
  biological_scale: MOLECULAR
  description: >-
    A de novo heterozygous loss-of-function point variant in HNRNPU, or a 1q44
    microdeletion encompassing the gene, is present. Both routes reduce HNRNPU to a
    single functional copy, which is the starting point this node records - but they
    are not the same lesion, because a 1q44 deletion also removes AKT3 and ZBTB18
    (see the genetic section). This node records the HNRNPU lesion only, not its
    consequences.
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Microdeletions in the 1q44 locus encompassing HNRNPU and other genes and point mutations in HNRNPU cause brain disorders, including early-onset seizures and severe intellectual disability."
    explanation: >-
      Establishes both lesion types as causes of the same clinical picture.
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "De novo loss-of-function variants in HNRNPU have been found to cause a neurodevelopmental syndrome"
    explanation: >-
      Confirms the de novo loss-of-function mechanism.
  downstream:
  - target: Reduced hnRNP U Protein Dose
    causal_link_type: DIRECT
    description: >-
      One functional allele produces roughly half the normal protein.

- name: Reduced hnRNP U Protein Dose
  biological_scale: MOLECULAR
  description: >-
    Haploinsufficiency for an abundant, ubiquitously expressed nuclear protein.
    Because hnRNP U is required for mammalian development - complete knockout mice
    die early - the disease is specifically a dose problem rather than an absence
    problem, and the relevant question is which of its several functions are most
    dose-sensitive.
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "HNRNPU is critical in mammalian development, and knockout mice exhibit early lethality"
    explanation: >-
      Establishes that complete loss is lethal, so human disease is a partial-dose
      state.
  downstream:
  - target: Disrupted Alternative Splicing Program
    causal_link_type: DIRECT
    description: >-
      hnRNP U participates directly in spliceosomal RNA processing.
  - target: Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture
    causal_link_type: DIRECT
    description: >-
      hnRNP U oligomerises with RNA to form the nuclear mesh that organizes
      chromatin; this is a separate function from splicing.

- name: Disrupted Alternative Splicing Program
  biological_scale: MOLECULAR
  description: >-
    Expression and alternative splicing of many genes are altered, with the
    affected transcripts concentrated in cell survival, cell motility, and synapse
    formation. This is the post-transcriptional arm.
  biological_processes:
  - preferred_term: regulation of alternative mRNA splicing, via spliceosome
    term:
      id: GO:0000381
      label: regulation of alternative mRNA splicing, via spliceosome
    modifier: ABNORMAL
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "expression and alternative splicing of multiple genes involved in cell survival, cell motility, and synapse formation are affected following Hnrnpu's conditional truncation"
    explanation: >-
      Names the splicing consequence and the functional categories affected.
  downstream:
  - target: Neural Progenitor Cell Death
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Mis-splicing of cell-survival regulators precedes the death of progenitors,
      and the pharmacological rescue data support a causal rather than incidental
      link.
  - target: Impaired Radial Neuronal Migration
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Mis-splicing of cell-motility regulators accompanies the migration defect.

- name: Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture
  biological_scale: MOLECULAR
  description: >-
    hnRNP U oligomerises in an ATP-dependent manner and binds RNA to form a dynamic
    nuclear mesh that regulates chromatin organization in interphase and mitosis.
    Reduced protein destabilizes that mesh. Nothing in a splicing assay measures
    this arm, which is why it is modeled separately.
  biological_processes:
  - preferred_term: chromatin organization
    term:
      id: GO:0006325
      label: chromatin organization
    modifier: ABNORMAL
  evidence:
  - reference: PMID:34823151
    reference_title: "The role of SAF-A/hnRNP U in regulating chromatin structure."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "SAF-A can oligomerise in an ATP-dependent manner and interact with RNA to form a dynamic nuclear mesh."
    explanation: >-
      States the oligomerisation and mesh-formation function that reduced protein
      dose compromises.
  downstream:
  - target: Widespread Transcriptional Dysregulation
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Altered chromatin architecture is one proposed route to the observed
      transcriptional changes, though the mechanism is not established.

- name: Widespread Transcriptional Dysregulation
  biological_scale: CELLULAR
  description: >-
    Transcriptome-wide changes across functionally relevant pathways. The
    dysregulation is notable for being broad but individually modest rather than
    a small number of large effects, which is why no single downstream gene
    explains the phenotype.
  biological_processes:
  - preferred_term: regulation of DNA-templated transcription
    term:
      id: GO:0006355
      label: regulation of DNA-templated transcription
    modifier: ABNORMAL
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "We identified widespread dysregulation in functionally relevant pathways"
    explanation: >-
      Documents the breadth of transcriptional dysregulation in a human model.
  downstream:
  - target: Impaired Synapse Formation and Circuit Maturation
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Dysregulation of synaptic transcripts is proposed to degrade circuit assembly.

- name: Neural Progenitor Cell Death
  biological_scale: CELLULAR
  description: >-
    Loss of HNRNPU function causes rapid cell death of both postmitotic neurons and
    neural progenitors, with progenitors apparently the more sensitive population.
    This differential vulnerability matters: it means the lesion subtracts cells
    that would each have produced many neurons, not just the cells directly lost.
  cell_types:
  - preferred_term: neural progenitor cell
    term:
      id: CL:0011020
      label: neural progenitor cell
  biological_processes:
  - preferred_term: apoptotic process
    term:
      id: GO:0006915
      label: apoptotic process
    modifier: INCREASED
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "HNRNPU loss of function leads to rapid cell death of both postmitotic neurons and neural progenitors, with an apparent higher sensitivity of the latter"
    explanation: >-
      Reports the cell death and the progenitor-selective vulnerability.
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "We suppressed cell death by pan-Caspase inhibitors, p53 inhibitors, and necroptosis inhibitors, demonstrating the involvement of p53-dependent canonical and noncanonical death mechanisms in neural progenitors"
    explanation: >-
      Identifies the death mechanism as p53-dependent through both canonical and
      non-canonical routes, established by pharmacological suppression rather than
      correlation.
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "We showed that reducing the levels of the splicing factor SRSF3 opposes HNRNPU’s loss of function effects and improves neural progenitors’ viability and neuronal migration"
    explanation: >-
      Lowering the competing splicing factor SRSF3 rescues progenitor viability and
      migration, which ties the death back to the splicing arm specifically.
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we performed compositional analyses and identified significant reductions in progenitor populations"
    explanation: >-
      The same endpoint measured in heterozygous human organoids rather than in the
      mouse truncation.
  downstream:
  - target: Loss of Cortical Structure
    causal_link_type: DIRECT
    description: >-
      Depleting the progenitor pool removes the cells that build the cortex.

- name: Impaired Radial Neuronal Migration
  biological_scale: CELLULAR
  description: >-
    Neurons that survive migrate abnormally along the radial axis. This is a
    positioning defect distinct from the cell-death arm, and it is separately
    rescuable in the model.
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "ameliorate radial neuronal migration defects"
    explanation: >-
      The rescue experiments identify radial migration as a separately measurable
      and separately correctable defect.
  downstream:
  - target: Loss of Cortical Structure
    causal_link_type: DIRECT
    description: >-
      Mispositioned neurons contribute to disturbed cortical architecture.

- name: Loss of Cortical Structure
  biological_scale: TISSUE
  description: >-
    Cortical structures are lost in the mutant embryonic brain. In the mouse
    conditional knockout this is severe; the human heterozygous condition produces
    the much milder nonspecific findings seen on MRI, so this node represents the
    same process at a different dose.
  biological_processes:
  - preferred_term: cell population proliferation
    term:
      id: GO:0008283
      label: cell population proliferation
    modifier: DECREASED
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "we identified pharmaceutical and genetic agents that can partially reverse the loss of cortical structures in Hnrnpu mutated embryonic brains"
    explanation: >-
      Names the loss of cortical structures as the tissue-level outcome.
  downstream:
  - target: Impaired Synapse Formation and Circuit Maturation
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      A structurally compromised cortex assembles circuits abnormally.

- name: Impaired Synapse Formation and Circuit Maturation
  biological_scale: TISSUE
  conforms_to: "epilepsy_excitation_inhibition_imbalance#Ion Channel and Synaptic Dysfunction"
  description: >-
    Synapse formation is among the functional categories whose genes are
    mis-spliced, and the resulting circuits mature abnormally. This is the
    synaptic-level convergence point for the splicing, transcriptional, and
    structural arms.
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "expression and alternative splicing of multiple genes involved in cell survival, cell motility, and synapse formation are affected"
    explanation: >-
      Names synapse formation among the affected functional categories.
  downstream:
  - target: Cortical Network Hyperexcitability
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Abnormally assembled circuits are epileptogenic, though the specific
      excitatory-inhibitory imbalance has not been characterized in this disorder.
  - target: Developmental Delay and Intellectual Disability
    causal_link_type: DIRECT
    description: >-
      Disturbed circuit maturation underlies the cognitive phenotype.

- name: Cortical Network Hyperexcitability
  biological_scale: TISSUE
  conforms_to: "epilepsy_excitation_inhibition_imbalance#Neuronal Hyperexcitability and Hypersynchrony"
  description: >-
    Cortical networks become seizure-prone. In patients this is inferred from the
    clinical seizure frequency rather than measured directly; no electrophysiological
    characterization of the human excitatory-inhibitory balance in this disorder has
    been published. The direct excitability measurement comes instead from the
    dose-matched heterozygous mouse, in which the threshold for electrically evoked
    seizures is significantly reduced.
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      Establishes the very high seizure frequency that this node accounts for.
  - reference: PMID:37782669
    reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Hnrnpu+/113DEL mice demonstrated a significantly lower threshold for induction of maximal tonic hindlimb extension seizures, consistent with a greater seizure predisposition"
    explanation: >-
      The one direct excitability measurement available at the human gene dose: a
      lowered evoked-seizure threshold is the operational definition of network
      hyperexcitability this node asserts.
  downstream:
  - target: Recurrent Seizures
    causal_link_type: DIRECT
    description: >-
      A hyperexcitable cortex generates recurrent seizures.

- name: Recurrent Seizures
  biological_scale: ORGANISM
  conforms_to: "epilepsy_excitation_inhibition_imbalance#Recurrent Unprovoked Seizures"
  description: >-
    Seizures affect 83-95% of individuals, most commonly tonic-clonic or absence,
    and may be refractory to treatment.
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Affected individuals are likely to experience seizures (most commonly tonic-clonic or absence) that may be refractory to treatment."
    explanation: >-
      GeneReviews gives the seizure semiology and refractoriness.
  downstream:
  - target: Developmental Delay and Intellectual Disability
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Seizure burden plausibly compounds the developmental phenotype, which the
      splicing and structural arms already produce independently.

- name: Developmental Delay and Intellectual Disability
  biological_scale: ORGANISM
  description: >-
    Developmental delay in around 95% of individuals and intellectual disability -
    typically moderate to severe - with speech and language delay or absent speech.
    Two upstream routes converge here, the developmental lesion and the seizure
    burden, and the published data do not separate their contributions.
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "HNRNPU-related neurodevelopmental disorder (HNRNPU-NDD) is characterized by developmental delay and intellectual disability"
    explanation: >-
      GeneReviews defines the core developmental phenotype.

phenotypes:
- category: Neurological
  name: Seizures
  description: >-
    Seizures in 83-95% of individuals, most commonly tonic-clonic or absence, and
    potentially refractory to treatment.
  phenotype_term:
    preferred_term: Seizure
    term:
      id: HP:0001250
      label: Seizure
  frequency: VERY_FREQUENT
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      A reported range of 83-95% sits entirely within the 80-99% very frequent band.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Affected individuals are likely to experience seizures (most commonly tonic-clonic or absence) that may be refractory to treatment."
    explanation: >-
      GeneReviews corroborates the seizure phenotype and its semiology.
- category: Neurological
  name: Global Developmental Delay
  description: >-
    Developmental delay in approximately 95% of individuals, the most consistent
    feature of the disorder.
  phenotype_term:
    preferred_term: Global developmental delay
    term:
      id: HP:0001263
      label: Global developmental delay
  frequency: VERY_FREQUENT
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      The reported figure of approximately 95% falls in the 80-99% very frequent band.
- category: Neurological
  name: Intellectual Disability
  description: >-
    Intellectual disability, typically moderate to severe, reported in 52% of
    individuals.
  phenotype_term:
    preferred_term: Intellectual disability
    term:
      id: HP:0001249
      label: Intellectual disability
  frequency: FREQUENT
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      52% falls in the 30-79% frequent band.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "developmental delay and intellectual disability"
    explanation: >-
      GeneReviews corroborates intellectual disability as a core feature; its
      moderate-to-severe severity is stated in the same chapter.
- category: Neurological
  name: Absent or Delayed Speech
  description: >-
    Speech and language delay, or absent speech, is part of the core GeneReviews
    description.
  phenotype_term:
    preferred_term: Absent speech
    term:
      id: HP:0001344
      label: Absent speech
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with speech and language delay and/or absent speech"
    explanation: >-
      GeneReviews names absent speech as a core feature.
- category: Craniofacial
  name: Craniofacial Dysmorphism
  description: >-
    Dysmorphic features are reported in 54-95% of individuals but are described as
    nonspecific, so they support the diagnosis without defining it.
  phenotype_term:
    preferred_term: Abnormal facial shape
    term:
      id: HP:0001999
      label: Abnormal facial shape
  frequency: FREQUENT
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      The reported range 54-95% straddles the 30-79% frequent and 80-99% very
      frequent bands; the lower bound is taken so the band is not over-claimed.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Dysmorphic features have been described but they are nonspecific."
    explanation: >-
      GeneReviews qualifies the diagnostic value of the dysmorphism.
- category: Neurological
  name: Microcephaly
  description: >-
    Microcephaly in 14-20% of individuals, described as a lower-penetrance feature.
  phenotype_term:
    preferred_term: Microcephaly
    term:
      id: HP:0000252
      label: Microcephaly
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      14-20% falls entirely within the 5-29% occasional band.
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "lower penetrance microcephaly, a thin corpus callosum, dysmorphic facial features, and hypotonia"
    explanation: >-
      Independently describes microcephaly as a lower-penetrance feature.
- category: Neurological
  name: Hypotonia
  description: >-
    Hypotonia is present from the neonatal period and often remains lifelong.
  phenotype_term:
    preferred_term: Hypotonia
    term:
      id: HP:0001252
      label: Hypotonia
    temporality: CHRONIC
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "There may be feeding difficulties during the neonatal period as well as hypotonia, which often remains lifelong."
    explanation: >-
      GeneReviews describes hypotonia and its persistence.
- category: Neurological
  name: Thin Corpus Callosum
  description: >-
    Thinning of the corpus callosum is a recurrent but nonspecific brain MRI finding.
  phenotype_term:
    preferred_term: Thin corpus callosum
    term:
      id: HP:0033725
      label: Thin corpus callosum
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Nonspecific brain MRI findings include ventriculomegaly and thinning of the corpus callosum."
    explanation: >-
      GeneReviews names the callosal finding.
- category: Neurological
  name: Ventriculomegaly
  description: >-
    Ventriculomegaly is the other recurrent nonspecific MRI finding.
  phenotype_term:
    preferred_term: Ventriculomegaly
    term:
      id: HP:0002119
      label: Ventriculomegaly
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Nonspecific brain MRI findings include ventriculomegaly and thinning of the corpus callosum."
    explanation: >-
      GeneReviews names ventriculomegaly.
- category: Behavioral
  name: Autistic Features
  description: >-
    Autistic features are reported in a proportion of affected individuals.
  phenotype_term:
    preferred_term: Autistic behavior
    term:
      id: HP:0000729
      label: Autistic behavior
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Affected individuals may also display autistic features."
    explanation: >-
      GeneReviews records autistic features as a possible manifestation.
- category: Multisystem
  name: Extraneurological Features
  description: >-
    Beyond the neurological core, less common findings include cardiac
    abnormalities, strabismus, undescended testes in males, renal anomalies, and
    skeletal features including joint laxity, polydactyly, and scoliosis. These are
    grouped here rather than split into separate nodes because the GeneReviews
    source names them in a single sentence without per-feature frequencies, and the
    per-feature percentages available elsewhere come from a research summary whose
    underlying cohort citations could not be resolved.
  phenotype_term:
    preferred_term: Strabismus
    term:
      id: HP:0000486
      label: Strabismus
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Less common findings include cardiac abnormalities, strabismus, undescended testes in males, renal anomalies, and skeletal features, including joint laxity, polydactyly, and scoliosis"
    explanation: >-
      GeneReviews names the extraneurological cluster, including strabismus, which
      this node is bound to as its most frequently reported member.
- category: Gastrointestinal
  name: Feeding Difficulties
  description: >-
    Feeding difficulties in the neonatal period, sometimes requiring a temporary or
    permanent feeding tube.
  phenotype_term:
    preferred_term: Feeding difficulties
    term:
      id: HP:0011968
      label: Feeding difficulties
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "There may be feeding difficulties during the neonatal period as well as hypotonia"
    explanation: >-
      GeneReviews documents neonatal feeding difficulties.
- category: Respiratory
  name: Abnormal Breathing Pattern
  description: >-
    Rarely, hyperventilation and apnea occur and can cause sleep disturbance. This
    is uncommon but worth recognizing because it is treatable.
  phenotype_term:
    preferred_term: Apnea
    term:
      id: HP:0002104
      label: Apnea
  frequency: VERY_RARE
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Rarely, abnormal breathing patterns, including hyperventilation and apnea, may be present and can lead to sleep disturbance."
    explanation: >-
      GeneReviews describes this as rare, which maps to the 1-4% very rare band.

genetic:
- name: HNRNPU
  association: De novo heterozygous loss-of-function variants
  presence: Positive
  relationship_type: CAUSATIVE
  variant_origin: DE_NOVO
  notes: >-
    HNRNPU encodes heterogeneous nuclear ribonucleoprotein U, also known as scaffold
    attachment factor A. The protein binds RNA, DNA, and other proteins, and those
    interactions support its roles in organizing and stabilizing nuclear chromatin,
    regulating transcription, and RNA splicing and stability. Disease arises from de
    novo heterozygous loss-of-function point variants or from 1q44 microdeletions
    encompassing the gene; both converge on haploinsufficiency for HNRNPU itself, but
    they are NOT interchangeable at the phenotype level: in the 1q43q44 contiguous-gene
    region, AKT3 haploinsufficiency is the main driver of microcephaly and ZBTB18 of
    corpus callosum anomalies, while HNRNPU alteration mostly drives the epilepsy and
    sets the degree of intellectual disability. A deletion patient's microcephaly or
    callosal anomaly may therefore be attributable to a neighbouring gene rather than
    to HNRNPU. Complete loss is embryonic lethal in mice, so no null human genotype
    exists. A DNA methylation episignature has been developed for this disorder and
    can serve as a diagnostic biomarker, including for variant reclassification.
  gene_term:
    preferred_term: HNRNPU
    term:
      id: hgnc:5048
      label: HNRNPU
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "De novo loss-of-function variants in HNRNPU have been found to cause a neurodevelopmental syndrome"
    explanation: >-
      Establishes HNRNPU as the causal gene and the variant class.
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Microdeletions in the 1q44 locus encompassing HNRNPU and other genes and point mutations in HNRNPU cause brain disorders"
    explanation: >-
      Documents the two lesion routes to the same disorder.
  - reference: PMID:28283832
    reference_title: "Genetic and phenotypic dissection of 1q43q44 microdeletion syndrome and neurodevelopmental phenotypes associated with mutations in ZBTB18 and HNRNPU."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Our study demonstrates that AKT3 haploinsufficiency is the main driver for microcephaly, whereas HNRNPU alteration mostly drives epilepsy and determines the degree of intellectual disability"
    explanation: >-
      Apportions the 1q43q44 phenotype across contiguous genes, showing HNRNPU drives
      the epilepsy and the degree of intellectual disability while AKT3 drives the
      microcephaly.
  - reference: PMID:28283832
    reference_title: "Genetic and phenotypic dissection of 1q43q44 microdeletion syndrome and neurodevelopmental phenotypes associated with mutations in ZBTB18 and HNRNPU."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "ZBTB18 may also contribute to microcephaly and HNRNPU to thin corpus callosum, but with a lower penetrance"
    explanation: >-
      Qualifies the split: HNRNPU does contribute to thin corpus callosum, but with
      lower penetrance than the primary drivers.
  - reference: PMID:37120726
    reference_title: "DNA methylation episignature and comparative epigenomic profiling of HNRNPU-related neurodevelopmental disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We performed genome-wide DNA methylation (DNAm) analysis in a cohort of individuals to develop a diagnostic biomarker and gain functional insights into the molecular pathophysiology of HNRNPU-related disorder"
    explanation: >-
      Establishes the DNA methylation episignature as a diagnostic biomarker for the
      disorder.
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "The HNRNPU protein binds to RNA, DNA, and other proteins, and these interactions facilitate its functions in organizing and stabilizing nuclear chromatin, regulating gene transcription"
    explanation: >-
      Describes the multifunctional protein whose dose is halved.

diagnosis:
- name: Molecular Genetic Testing for a Heterozygous HNRNPU Variant
  description: >-
    The diagnosis is established in a proband with suggestive findings and a
    heterozygous pathogenic HNRNPU variant. Testing must be able to detect a 1q44
    microdeletion as well as a point variant, since either causes the disorder.
  diagnosis_term:
    preferred_term: genetic testing
    term:
      id: NCIT:C15709
      label: Genetic Testing
  presence: Positive in affected individuals
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The diagnosis of HNRNPU-NDD is established in a proband with suggestive findings and a heterozygous pathogenic variant in HNRNPU identified by molecular genetic testing."
    explanation: >-
      GeneReviews states the diagnostic criterion directly.
- name: DNA Methylation Episignature
  description: >-
    A genome-wide DNA methylation signature specific to HNRNPU-related disorder,
    developed as a diagnostic biomarker on Infinium Methylation EPIC arrays and
    useful for reclassifying variants of uncertain significance.
  diagnosis_term:
    preferred_term: DNA methylation profiling
    term:
      id: NCIT:C15709
      label: Genetic Testing
  evidence:
  - reference: PMID:37120726
    reference_title: "DNA methylation episignature and comparative epigenomic profiling of HNRNPU-related neurodevelopmental disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We performed genome-wide DNA methylation (DNAm) analysis in a cohort of individuals to develop a diagnostic biomarker and gain functional insights into the molecular pathophysiology of HNRNPU-related disorder"
    explanation: >-
      States the purpose and method of the episignature analysis.
- name: Brain MRI
  description: >-
    MRI shows nonspecific findings - ventriculomegaly and thinning of the corpus
    callosum. These support but do not establish the diagnosis.
  diagnosis_term:
    preferred_term: brain magnetic resonance imaging
    term:
      id: NCIT:C16809
      label: Magnetic Resonance Imaging
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Nonspecific brain MRI findings include ventriculomegaly and thinning of the corpus callosum."
    explanation: >-
      GeneReviews describes the imaging findings and their nonspecificity.

treatments:
- name: Anti-Seizure Medication
  description: >-
    Standard anti-seizure medication, with sodium valproate often used and
    frequently effective. Newer-generation agents or the ketogenic diet are
    considered in refractory cases.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: valproic acid
      term:
        id: CHEBI:39867
        label: valproic acid
  target_mechanisms:
  - target: Recurrent Seizures
    treatment_effect: INHIBITS
    description: >-
      Sodium valproate is described as frequently effective for the seizures in this
      disorder.
    evidence:
    - reference: PMID:35274911
      reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: "Standard treatment of seizures with anti-seizure medications (sodium valproate is often used and is frequently effective)"
      explanation: >-
        GeneReviews reports valproate as frequently effective here, which is notable
        given how often DEE seizures are refractory.
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Standard treatment of seizures with anti-seizure medications (sodium valproate is often used and is frequently effective)"
    explanation: >-
      GeneReviews management recommendation.
- name: Ketogenic Diet
  description: >-
    Considered together with newer-generation anti-seizure medications for those
    with refractory seizures.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: Dietary Intervention
    term:
      id: NCIT:C15447
      label: Dietary Intervention
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "consider instituting the ketogenic diet and/or newer generation anti-seizure medications in those with refractory seizures"
    explanation: >-
      GeneReviews recommends the diet for the refractory subgroup.
- name: Respiratory Support for Sleep Apnea
  description: >-
    Supplemental oxygen, CPAP, or BiPAP for those with sleep apnea arising from the
    abnormal breathing pattern.
  therapeutic_modality: DEVICE
  treatment_term:
    preferred_term: respiratory therapy
    term:
      id: NCIT:C15322
      label: Respiratory Therapy
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Consider supplemental oxygen, CPAP, or BiPAP in those with sleep apnea."
    explanation: >-
      GeneReviews management recommendation for the respiratory phenotype.
- name: Feeding Therapy and Enteral Support
  description: >-
    Feeding therapy, with a temporary or permanent feeding tube considered for
    persistent feeding problems.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: nutritional support
    term:
      id: NCIT:C15433
      label: Nutritional Support
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Feeding therapy; consider a temporary or permanent feeding tube for those with persistent feeding issues."
    explanation: >-
      GeneReviews management recommendation.
- name: Avoidance of Seizure-Inducing Agents and Activities
  description: >-
    GeneReviews lists activities and agents that may induce seizures under
    agents and circumstances to avoid.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: preventive intervention
    term:
      id: NCIT:C15843
      label: Preventive Intervention
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Agents/circumstances to avoid: Activities and agents that may induce seizures."
    explanation: >-
      The GeneReviews agents-to-avoid guidance.
- name: Routine Surveillance
  description: >-
    At each visit: growth parameters, nutritional status and safety of oral intake,
    monitoring for constipation, new seizures, hyperventilation, apnea, and changes
    in tone, and assessment of developmental progress and behavior. Annually or as
    clinically indicated: ophthalmologic and audiologic evaluations - which is also
    how the strabismus in the extraneurological cluster is expected to be detected.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: Supportive Care
    term:
      id: NCIT:C15747
      label: Supportive Care
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "At each visit: measurement of growth parameters; evaluation of nutritional status and safety of oral intake; monitor for evidence of constipation, new seizures, hyperventilation, apnea, and changes in tone; assessment of developmental progress and behavior"
    explanation: >-
      The GeneReviews per-visit surveillance schedule.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Annually or as clinically indicated: ophthalmologic and audiologic evaluations"
    explanation: >-
      The GeneReviews annual surveillance schedule.
- name: Genetic Counseling
  description: >-
    Counseling for an autosomal dominant, typically de novo condition. Recurrence
    risk to siblings is hypothesized to be low, with the caveat of a reported
    presumed germline mosaicism family.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: Genetic Counseling
    term:
      id: NCIT:C15240
      label: Genetic Counseling
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The risk to other family members is hypothesized to be low."
    explanation: >-
      GeneReviews genetic counseling guidance.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Once the HNRNPU pathogenic variant has been identified in an affected family member, prenatal testing and preimplantation genetic testing are possible"
    explanation: >-
      Records the reproductive options available once the familial variant is known.

animal_models:
- name: Hnrnpu conditional cortical knockout mouse
  species: Mouse
  genotype: Hnrnpu conditional truncation in embryonic cortical neuroepithelium
  publication: PMID:35864088
  description: >-
    Conditional truncation of Hnrnpu in the developing cortex causes rapid death of
    neural progenitors and postmitotic neurons, radial migration defects, and loss
    of cortical structures, each partially reversible with pharmacological or
    genetic intervention.
  modeled_mechanisms:
  - target: Neural Progenitor Cell Death
    relationship: RECAPITULATES
    fidelity: MODERATE
    description: >-
      Provides the direct evidence for progenitor-selective cell death and, through
      the rescue experiments, for its causal role in the cortical phenotype.
    limitations: >-
      This is a conditional truncation removing HNRNPU function outright, whereas
      human disease is heterozygous loss of function; complete Hnrnpu knockout is
      embryonic lethal in mice, so the model sits at a far more severe point on the
      dose-response curve than any human genotype and produces near-complete cortical
      loss rather than the nonspecific MRI findings seen in patients.
    readouts:
    - name: Neural progenitor viability after Cre-mediated truncation
      target: Neural Progenitor Cell Death
      direction: DECREASED
      interpretation: >-
        Progenitors die rapidly and more readily than postmitotic neurons.
      evidence:
      - reference: PMID:35864088
        reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "HNRNPU loss of function leads to rapid cell death of both postmitotic neurons and neural progenitors, with an apparent higher sensitivity of the latter"
        explanation: >-
          Reports the viability measurement behind this readout.
    - name: Cortical structure after rescue intervention
      target: Neural Progenitor Cell Death
      direction: RESTORED
      interpretation: >-
        Pharmacological and genetic agents partially reverse the cortical loss,
        supporting cell death as its cause rather than a correlate.
      evidence:
      - reference: PMID:35864088
        reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "we identified pharmaceutical and genetic agents that can partially reverse the loss of cortical structures in Hnrnpu mutated embryonic brains, ameliorate radial neuronal migration defects and rescue cultured neural progenitors' cell death"
        explanation: >-
          Reports the rescue result behind this readout.
    evidence:
    - reference: PMID:35864088
      reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: "We aimed to understand HNRNPU's roles in the developing brain."
      explanation: >-
        Establishes the model's purpose as characterizing HNRNPU in cortical
        development.

- name: Hnrnpu+/113DEL heterozygous mouse
  species: Mouse
  genotype: Hnrnpu heterozygous 113-bp out-of-frame exon 1 deletion (Hnrnpu+/113DEL)
  publication: PMID:37782669
  description: >-
    A CRISPR-induced constitutive heterozygous truncating deletion in exon 1 of mouse
    Hnrnpu, targeting the region where at least five human patient truncating variants
    fall. This is the dose-matched counterpart of the conditional cortical knockout
    above: it models the human genotype rather than complete loss of function, and it
    is informative as much for what it fails to reproduce as for what it does.
  modeled_mechanisms:
  - target: Cortical Network Hyperexcitability
    relationship: RECAPITULATES
    fidelity: MODERATE
    description: >-
      Supplies the only direct excitability measurement at the human gene dose. The
      lowered electroconvulsive threshold is a measurement of network excitability
      rather than an inference from seizure counts, which is what the patient
      evidence on this node otherwise rests on.
    limitations: >-
      Excitability is measured as susceptibility to an electrically evoked seizure,
      not as spontaneous epileptiform activity, so it establishes a lowered seizure
      threshold rather than the ongoing hyperexcitable state patients are presumed to
      carry. The model also under-represents the human lesion: cortical hnRNP U falls
      by only about 20-26% rather than the expected half, which the authors read as
      evidence of compensation, so the measured effect is likely a floor.
    readouts:
    - name: Maximal seizure electroconvulsive threshold
      target: Cortical Network Hyperexcitability
      direction: DECREASED
      interpretation: >-
        Less current is required to evoke a maximal tonic hindlimb extension seizure
        in heterozygous mutants than in wild-type littermates.
      evidence:
      - reference: PMID:37782669
        reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "Hnrnpu+/113DEL mice demonstrated a significantly lower threshold for induction of maximal tonic hindlimb extension seizures, consistent with a greater seizure predisposition"
        explanation: >-
          Reports the threshold measurement behind this readout.
      - reference: PMID:37782669
        reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "In epilepsy mouse models, neuronal hyperexcitability can also manifest as greater susceptibility to induced seizure events"
        explanation: >-
          States the assay's interpretive basis, which is why a threshold shift is
          read here as network hyperexcitability.
    evidence:
    - reference: PMID:37782669
      reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: "Heterozygous mutants demonstrated global developmental delay, impaired ultrasonic vocalizations, cognitive dysfunction and increased seizure susceptibility, thus modeling aspects of the human disease."
      explanation: >-
        Establishes the model as informative for the seizure arm, in the authors' own
        deliberately partial phrasing.
  - target: Recurrent Seizures
    relationship: FAILS_TO_RECAPITULATE
    fidelity: LOW
    description: >-
      The clinical endpoint does not appear. Patients have seizures in 83-95% of
      cases, but the dose-matched mouse never develops spontaneous ones - only a
      lowered threshold for evoked seizures. The model reproduces seizure
      susceptibility, not epilepsy.
    limitations: >-
      This is a substantive negative result, not an unexamined gap: over 300 hours of
      video EEG in adults found no spontaneous epileptiform activity, and no
      seizure-like behaviour was seen in routine handling at any age. The likeliest
      reasons are the shortfall in Hnrnpu reduction noted above and species
      differences in seizure threshold, but neither has been tested, so this link
      should not be read as evidence against the human mechanism.
    readouts:
    - name: Spontaneous epileptiform activity on video EEG
      target: Recurrent Seizures
      direction: UNCHANGED
      interpretation: >-
        No spontaneous generalized epileptiform activity was detected in heterozygous
        adults despite extended recording.
      evidence:
      - reference: PMID:37782669
        reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "Despite over 300 total hours of video EEG recordings among Hnrnpu+/113DEL adults, there was no evidence of spontaneous generalized epileptiform activity"
        explanation: >-
          The negative EEG result behind this readout, and the reason the link is
          typed as a failure to recapitulate.
      - reference: PMID:37782669
        reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "no spontaneous seizure-like behaviors or sudden death were observed following routine handling of this mouse line at any age"
        explanation: >-
          Confirms the negative result behaviourally as well as electrographically,
          and across the lifespan rather than only in the recorded adults.
    evidence:
    - reference: PMID:37782669
      reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: "Despite over 300 total hours of video EEG recordings among Hnrnpu+/113DEL adults, there was no evidence of spontaneous generalized epileptiform activity"
      explanation: >-
        Substantiates treating this model as uninformative for the spontaneous-seizure
        endpoint.
  - target: Loss of Cortical Structure
    relationship: FAILS_TO_RECAPITULATE
    fidelity: LOW
    description: >-
      No gross cortical phenotype appears at the human gene dose. Brain size, corpus
      callosum morphology, cortical thickness and hippocampal width are all
      unchanged, in sharp contrast to the near-complete cortical loss the conditional
      truncation produces.
    limitations: >-
      The measurements are gross morphometry at postnatal day 0, and the authors say
      explicitly that higher-resolution studies able to detect subtler abnormalities
      are still warranted - so this is evidence against a gross structural phenotype,
      not against a structural contribution as such. It is nonetheless the most
      directly relevant datum available for the dose question, since it is the only
      structural measurement made at the human genotype in vivo.
    readouts:
    - name: Gross brain morphometry at postnatal day 0
      target: Loss of Cortical Structure
      direction: UNCHANGED
      interpretation: >-
        Neither brain size nor callosal morphology differs from wild type at birth.
      evidence:
      - reference: PMID:37782669
        reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "we did not observe differences in brain size and corpus callosum morphology compared to WT at postnatal day 0"
        explanation: >-
          The negative morphometric result behind this readout.
    - name: Cortical thickness and hippocampal width
      target: Loss of Cortical Structure
      direction: UNCHANGED
      interpretation: >-
        The two regional measures most relevant to a progenitor-depletion mechanism
        are also unaffected.
      evidence:
      - reference: PMID:37782669
        reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "Further examination of Hnrnpu+/113DEL brains also showed no significant change in cortical thickness and hippocampal width"
        explanation: >-
          The negative regional measurement behind this readout.
    evidence:
    - reference: PMID:37782669
      reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: "Further examination of Hnrnpu+/113DEL brains also showed no significant change in cortical thickness and hippocampal width"
      explanation: >-
        Substantiates the failure to recapitulate the cortical-structure node at the
        heterozygous dose.

experimental_models:
- name: Isogenic HNRNPU heterozygous human iPSC-derived brain organoid
  experimental_model_type: ORGANOID
  description: >-
    Human iPSC-derived brain organoids carrying a heterozygous HNRNPU
    loss-of-function allele, profiled transcriptome-wide and compared against
    embryonic and perinatal mouse cortex.
  publication: PMID:36594023
  modeled_mechanisms:
  - target: Neural Progenitor Cell Death
    relationship: RECAPITULATES
    fidelity: MODERATE
    description: >-
      Reaches the progenitor endpoint at the heterozygous human gene dose, which the
      mouse conditional truncation cannot do.
    limitations: >-
      The reduction is measured as population composition and organoid size rather
      than as direct time-lapse cell death, so the mode of loss is inferred; and the
      size phenotype is universal across mutant lines while microcephaly affects only
      14-20% of patients.
    readouts:
    - name: Progenitor population composition and organoid size
      target: Neural Progenitor Cell Death
      direction: DECREASED
      interpretation: >-
        Heterozygous HNRNPU loss reduces progenitor populations and organoid size in
        human tissue.
      evidence:
      - reference: PMID:36594023
        reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
        supports: SUPPORT
        evidence_source: IN_VITRO
        snippet: "we performed compositional analyses and identified significant reductions in progenitor populations"
        explanation: >-
          Reports the compositional reduction in progenitor populations.
      - reference: PMID:36594023
        reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
        supports: SUPPORT
        evidence_source: IN_VITRO
        snippet: "both HNRNPU+/−mutant lines generated significantly smaller organoids, with a size phenotype emerging as early as DIV 10"
        explanation: >-
          Reports the organoid size phenotype and its early onset.
    evidence:
    - reference: PMID:36594023
      reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
      supports: SUPPORT
      evidence_source: IN_VITRO
      snippet: "both HNRNPU+/−mutant lines generated significantly smaller organoids, with a size phenotype emerging as early as DIV 10"
      explanation: >-
        Establishes that the heterozygous human system reproduces a growth phenotype.
  - target: Widespread Transcriptional Dysregulation
    relationship: RECAPITULATES
    fidelity: MODERATE
    description: >-
      The only human model of the transcriptional consequence, and the one that
      establishes cross-species convergence with embryonic mouse cortex.
    limitations: >-
      Co-dysregulation was enriched against embryonic but explicitly NOT perinatal
      mouse cortex, so 45-day organoids appear to model only a restricted
      developmental window and set of cell types; the authors state this limitation
      themselves.
    readouts:
    - name: Transcriptome-wide dysregulation in 45-day organoids
      target: Widespread Transcriptional Dysregulation
      direction: ALTERED
      interpretation: >-
        Heterozygous HNRNPU loss produces broad transcriptional dysregulation in
        human tissue.
      evidence:
      - reference: PMID:36594023
        reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
        supports: SUPPORT
        evidence_source: IN_VITRO
        snippet: "We identified widespread dysregulation in functionally relevant pathways"
        explanation: >-
          Reports the transcriptomic measurement behind this readout.
    evidence:
    - reference: PMID:36594023
      reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
      supports: SUPPORT
      evidence_source: IN_VITRO
      snippet: "we find enrichment of co-dysregulation between 45-day-old human organoids and embryonic, but not perinatal, mice from distinct models of HNRNPU-related disorder"
      explanation: >-
        Establishes both the cross-species convergence and its developmental limits.

discussions:
- discussion_id: gap_which_hnrnpu_function_drives_the_phenotype
  prompt: >-
    Which of hnRNP U's functions - alternative splicing, chromatin architecture via
    the SAF-A/RNA nuclear mesh, or transcriptional regulation - is the dose-sensitive
    one that produces the neurodevelopmental phenotype, and are they separable at all?
  kind: KNOWLEDGE_GAP
  status: OPEN
  attaches_to:
  - pathophysiology#Disrupted Alternative Splicing Program
  - pathophysiology#Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture
  - pathophysiology#Widespread Transcriptional Dysregulation
  rationale: >-
    The pathograph here carries three parallel molecular arms because the protein has
    three documented functions, not because three have been shown to matter for
    disease. No study has separated them: the splicing evidence comes from a
    conditional truncation that removes all functions at once, and the chromatin
    literature explicitly says a mechanistic understanding is lacking. The domain
    architecture offers a route, since the DNA/RNA-binding, oligomerisation, and
    protein-interaction functions map to separable domains, so this is a tractable
    question rather than a permanent unknown. It matters because a therapy would have
    to target the dose-sensitive arm, and at present a screen would have no defined
    molecular endpoint to screen against.
  evidence:
  - reference: PMID:34823151
    reference_title: "The role of SAF-A/hnRNP U in regulating chromatin structure."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "This mesh is thought to regulate nuclear and chromatin architecture, yet a mechanistic understanding is lacking."
    explanation: >-
      States directly that the chromatin arm's mechanism is not understood.
  - reference: PMID:34823151
    reference_title: "The role of SAF-A/hnRNP U in regulating chromatin structure."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "we infer possible links between diseases emerging from SAF-A mutations and its role in chromatin organisation and regulation"
    explanation: >-
      The link from the chromatin function to disease is explicitly an inference,
      not a demonstrated result.
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "The HNRNPU protein binds to RNA, DNA, and other proteins, and these interactions facilitate its functions in organizing and stabilizing nuclear chromatin, regulating gene transcription"
    explanation: >-
      Enumerates the multiple functions that the disease models cannot currently
      disentangle.
  proposed_experiments:
  - experiment_id: exp_domain_separation_of_function_alleles
    name: Domain-selective HNRNPU separation-of-function alleles in human neurons
    description: >-
      Generate isogenic iPSC lines carrying alleles that selectively disable the
      DNA/RNA-binding domain, the ATP-dependent oligomerisation function required for
      mesh formation, or the RNA-binding motifs, and compare splicing changes,
      chromatin accessibility, transcriptional dysregulation, and progenitor
      viability across them, to establish which function the neurodevelopmental
      phenotype tracks.

- discussion_id: mismatch_conditional_truncation_versus_human_haploinsufficiency
  prompt: >-
    How much of the mouse cortical phenotype - rapid progenitor death and
    near-complete loss of cortical structures - reflects human HNRNPU disease, given
    that patients are heterozygous with nonspecific MRI findings and only 14-20%
    have microcephaly?
  kind: HUMAN_MODEL_MISMATCH
  status: OPEN
  attaches_to:
  - pathophysiology#Neural Progenitor Cell Death
  - pathophysiology#Loss of Cortical Structure
  rationale: >-
    The mouse model is a conditional truncation that removes HNRNPU function
    outright, and complete knockout is embryonic lethal. Human disease is
    heterozygous, and the great majority of patients do not have microcephaly at all,
    which is hard to reconcile with a mechanism whose defining feature in the model
    is rapid, near-complete progenitor elimination. Either the human dose is far
    enough above the death threshold that the cell-death arm contributes little, or
    it operates at a much lower intensity than the model shows. The heterozygous human organoid does
    reach the same endpoint - it shows significant reductions in progenitor
    populations and generates significantly smaller organoids - so the mismatch is
    narrower than a bare dose argument suggests: progenitor loss is real at the human
    gene dose, and the open question is one of magnitude rather than of kind. The
    dose-matched heterozygous mouse now bears on this directly, and it lands on the
    patient side: brain size, callosal morphology, cortical thickness and hippocampal
    width are all unchanged, so an in vivo animal at the human genotype shows no
    gross structural phenotype at all. That measurement carries its own caveat, since
    it is gross morphometry at postnatal day 0 and the authors call for
    higher-resolution work, and the same mouse loses only about a fifth to a quarter
    of cortical hnRNP U rather than half - which may itself be why the phenotype is
    absent. What remains unexplained is why organoid size reduction is universal
    across mutant lines while microcephaly reaches only 14-20% of patients and the
    heterozygous mouse has none; the difference between an organoid and an intact
    brain, where compensation and continued growth are both available, is the obvious
    place to look.
  evidence:
  - reference: PMID:35864088
    reference_title: "Heterogeneous nuclear ribonucleoprotein U (HNRNPU) safeguards the developing mouse cortex."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "HNRNPU is critical in mammalian development, and knockout mice exhibit early lethality"
    explanation: >-
      Establishes that the model sits at a dose no human patient occupies.
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "with patients exhibiting developmental delay (∼95%), intellectual disability (52%), craniofacial dysmorphism (54–95%) including microcephaly (14–20%) and seizures (83–95%)"
    explanation: >-
      Most patients do not have microcephaly, which is the observation the
      progenitor-death mechanism has to be reconciled with.
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we performed compositional analyses and identified significant reductions in progenitor populations"
    explanation: >-
      Corrects a natural but wrong assumption: the heterozygous human organoid does
      report progenitor loss, not only transcriptional change.
  - reference: PMID:37782669
    reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
    supports: REFUTE
    evidence_source: MODEL_ORGANISM
    snippet: "Further examination of Hnrnpu+/113DEL brains also showed no significant change in cortical thickness and hippocampal width"
    explanation: >-
      The dose-matched mouse shows no gross cortical phenotype, which is the
      strongest in vivo evidence that the truncation model's structural severity does
      not transfer to the human genotype.
  - reference: PMID:37782669
    reference_title: "Neurodevelopmental deficits and cell-type-specific transcriptomic perturbations in a mouse model of HNRNPU haploinsufficiency."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "revealed a decrease in Hnrnpu expression of approximately 20–25% in Hnrnpu+/113DEL mice"
    explanation: >-
      Qualifies the negative result above: the heterozygous mouse does not achieve a
      full halving of hnRNP U, so an absent phenotype there is not conclusive for the
      human dose.
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "The reduction in size of hCOs may phenocopy the subset of human patients with microcephaly"
    explanation: >-
      The organoid authors themselves connect the size reduction to the microcephalic
      subset of patients, which is the link the dose argument has to account for.
  proposed_experiments:
  - experiment_id: exp_heterozygous_progenitor_viability
    name: Progenitor viability in heterozygous rather than truncated models
    description: >-
      Measure neural progenitor apoptosis and proliferation directly in
      Hnrnpu-heterozygous mouse cortex and in isogenic heterozygous human organoids,
      using the same time-lapse and marker readouts as the conditional truncation
      study, to establish whether the cell-death arm operates at all at the human
      gene dose.

- discussion_id: gap_organoid_developmental_window_limits
  prompt: >-
    Do human brain organoids model HNRNPU-related disorder only within a narrow
    embryonic developmental window, and if so what system should be used to study
    the postnatal seizure phenotype that dominates the clinical picture?
  kind: HUMAN_MODEL_MISMATCH
  status: OPEN
  attaches_to:
  - pathophysiology#Widespread Transcriptional Dysregulation
  - pathophysiology#Cortical Network Hyperexcitability
  rationale: >-
    The organoid study found co-dysregulation enriched against embryonic but
    explicitly not perinatal mouse cortex, and the authors conclude that organoids may
    only be suitable for certain cell types within a specific developmental window.
    That is a problem for this disorder specifically, because seizures affect 83-95%
    of patients and are a postnatal phenomenon, while the one human model system
    appears to lose fidelity exactly as development moves past the embryonic period.
    A therapeutic screen run in organoids would therefore be optimizing against an
    embryonic transcriptional signature while the target phenotype is postnatal
    epilepsy.
  evidence:
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Thus, hnRNPU deficient human organoids may only be suitable to model transcriptional dysregulation in certain cell types within a specific developmental time window."
    explanation: >-
      The authors state the limitation of their own model system.
  - reference: PMID:36594023
    reference_title: "Evidence of shared transcriptomic dysregulation of HNRNPU-related disorder between human organoids and embryonic mice."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "An emerging drug discovery approach for neurodevelopmental disorders is to characterize transcriptome-wide dysregulation in an appropriate model system and screen therapeutics based on their capacity to restore functionally relevant expression patterns."
    explanation: >-
      Describes the screening paradigm whose validity depends on the model matching
      the target developmental stage.
  proposed_experiments:
  - experiment_id: exp_extended_organoid_timecourse
    name: Extended organoid time course against postnatal cortex
    description: >-
      Profile HNRNPU-heterozygous organoids at successively later maturation stages,
      including sliced or vascularized long-term cultures, and test co-dysregulation
      against perinatal and postnatal mouse cortex, to determine whether fidelity can
      be recovered at later stages or whether a different system is required for the
      seizure phenotype.

- discussion_id: gap_valproate_efficacy_mechanism
  prompt: >-
    Why is sodium valproate reported as frequently effective in HNRNPU-related
    disorder, when seizures in most developmental and epileptic encephalopathies are
    refractory, and does that reflect a mechanistic relationship between valproate's
    chromatin effects and the HNRNPU lesion?
  kind: OPEN_QUESTION
  status: OPEN
  attaches_to:
  - pathophysiology#Recurrent Seizures
  - pathophysiology#Destabilized SAF-A/RNA Nuclear Mesh and Chromatin Architecture
  rationale: >-
    GeneReviews describes valproate as often used and frequently effective here,
    which stands out against the pharmacoresistance typical of DEEs. Valproate also
    has chromatin-level activity as a histone deacetylase inhibitor, and the HNRNPU
    lesion is partly a chromatin-architecture lesion, so a mechanistic rather than
    coincidental relationship is worth asking about. The question is speculative and
    is recorded as an open question, not a hypothesis with supporting evidence: it may
    simply be that this disorder's seizure types, commonly tonic-clonic and absence,
    are ones valproate treats well in any aetiology. Distinguishing those two
    explanations would need a comparison against matched seizure types in other DEEs.
  evidence:
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Standard treatment of seizures with anti-seizure medications (sodium valproate is often used and is frequently effective)"
    explanation: >-
      The observation of unusual valproate efficacy that prompts the question.
  - reference: PMID:35274911
    reference_title: "HNRNPU-Related Neurodevelopmental Disorder."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Affected individuals are likely to experience seizures (most commonly tonic-clonic or absence) that may be refractory to treatment."
    explanation: >-
      Gives the seizure semiology behind the simpler alternative explanation, and
      notes that refractoriness still occurs, so the efficacy is not universal.
  proposed_experiments:
  - experiment_id: exp_valproate_response_versus_matched_dees
    name: Valproate response rate against seizure-type-matched DEE controls
    description: >-
      Compare valproate response rates in HNRNPU-related disorder against other DEEs
      matched for seizure semiology, to test whether the reported efficacy exceeds
      what the seizure types alone predict; if it does, assay whether valproate's
      HDAC inhibition normalizes chromatin accessibility in HNRNPU-heterozygous human
      neurons.

datasets: []
📚

References & Deep Research

References

1
PMID:35274911
HNRNPU-Related Neurodevelopmental Disorder.
No top-level findings curated for this source.

Deep Research

1
Claude Code ▸
HNRNPU-Related Developmental and Epileptic Encephalopathy: Comprehensive Research Report
claude-haiku-4-5-20251001, claude-sonnet-5 24 citations 2026-08-19T13:26:26.259637

HNRNPU-Related Developmental and Epileptic Encephalopathy: Comprehensive Research Report

1. Disease Information

Overview. HNRNPU-related developmental and epileptic encephalopathy (also called HNRNPU-related neurodevelopmental disorder, HNRNPU-NDD; historically "early infantile epileptic encephalopathy 54," EIEE54) is a severe, typically de novo, autosomal dominant neurodevelopmental disorder caused by heterozygous loss-of-function variants in HNRNPU on chromosome 1q44. It is characterized by global developmental delay, moderate-to-severe intellectual disability, early-onset seizures (often the presenting feature), profound speech impairment, hypotonia, and variable dysmorphic, cardiac, and renal features (GeneReviews, NBK578573; OMIM #617391).

Key identifiers: - OMIM: #617391 — Developmental and Epileptic Encephalopathy 54 (DEE54) - Gene: HNRNPU, HGNC:5048, chromosome 1q44 - MONDO: MONDO:0033363 (developmental and epileptic encephalopathy 54); a broader complex neurodevelopmental disorder mapping (MONDO:0100038) is also used in some resources - GeneReviews: NBK578573 ("HNRNPU-Related Neurodevelopmental Disorder") - ClinGen gene-disease validity: Definitive (Epilepsy and Intellectual Disability/Developmental Delay Gene Curation Expert Panels) (thegencc.org/genes/HGNC:5048)

Synonyms: HNRNPU-related neurodevelopmental disorder (HNRNPU-NDD/HNRNPU-RNDD); Developmental and Epileptic Encephalopathy 54 (DEE54); Early Infantile Epileptic Encephalopathy 54 (EIEE54); heterogeneous nuclear ribonucleoprotein U deficiency; SAF-A (scaffold attachment factor A)-related disorder.

Source of information. The evidence base is aggregated-disease-level (case series, cohort/registry studies, GeneReviews, OMIM) rather than raw EHR data, supplemented by a growing international patient registry (Bain Lab/Columbia; >140 confirmed individuals as of the 2025 phenotype-expansion review) and model-organism studies (Hodgson et al. 2025, PMID:39976380).

2. Etiology

Disease causal factors. DEE54 is caused by heterozygous, predominantly de novo, loss-of-function variants in HNRNPU — nonsense, frameshift, canonical splice-site variants, small intragenic deletions/duplications, and a minority of missense variants — or by microdeletions of 1q43q44 that encompass HNRNPU. A study of clinical exome cohorts concluded that "haploinsufficiency was the main mechanism of pathogenicity" (GeneReviews).

Genetic risk factors: - Causal variants: Truncating (nonsense, frameshift, splice) variants predominate; missense variants also reported (Bramswig et al. 2017 found "three non-sense and two missense variants, one small intragenic deletion, and one duplication" in a 7-patient cohort, PMID:28393272). - No genotype-phenotype correlation has been established — variant type does not reliably predict severity (GeneReviews). - Contiguous-gene 1q43q44 microdeletions: When HNRNPU is co-deleted with neighboring genes (AKT3, ZBTB18), phenotype is more severe/complex; deletion mapping studies show HNRNPU alteration specifically "drives epilepsy and determines the degree of intellectual disability," while AKT3 haploinsufficiency drives microcephaly and ZBTB18 loss drives corpus callosum anomalies with incomplete penetrance — with additive effects when multiple genes are co-deleted (Depienne et al., PMID:28283832). - Germline mosaicism: Presumed parental germline mosaicism has been reported in a family with two affected siblings, and a 2025 report documented the first vertically transmitted familial case (parent-to-child), demonstrating a milder end of the phenotypic spectrum is compatible with reproduction (Hodgson et al. 2025, PMID:39976380).

Environmental risk factors: None established; this is a monogenic disorder.

Protective factors: None identified at genetic or environmental levels.

Gene-environment interactions: Not reported; the disorder's severity appears to be driven by the genetic lesion itself (variant location relative to functional domains, and whether contiguous genes are co-deleted) rather than by environmental modifiers.

3. Phenotypes

Frequencies below are drawn primarily from GeneReviews' synthesis of the published cohort (~83–140+ individuals) and the original Bramswig/Thevenon-type cohort studies.

Phenotype Frequency HPO term (suggested)
Developmental delay (global) ~100% HP:0001263
Intellectual disability (moderate–severe) 84% HP:0002342 / HP:0010864
Speech delay/absent speech 80% HP:0000750 / HP:0001344
Seizures 95% HP:0001250
— Tonic-clonic seizures ~60% of seizure cases HP:0002069
— Absence seizures ~44% of seizure cases HP:0002121
— Seizure onset before 24 months ~90% of seizure cases HP:0011097 (infantile onset)
Hypotonia (often lifelong, may progress to spasticity) 79% HP:0001252
Dysmorphic craniofacial features (nonspecific) 97% HP:0001999
Feeding difficulties (sometimes requiring gastrostomy) 57% HP:0011968
Behavioral abnormalities ~50% HP:0000708
Autism spectrum disorder ~33% HP:0000729
Short stature ~50% HP:0004322
Strabismus 36% HP:0000486
Abnormal brain MRI (ventriculomegaly most common, then thin corpus callosum) 61% (of 62 imaged cases) HP:0002119 (ventriculomegaly); HP:0002079 (thin corpus callosum)
Congenital heart defects (ASD most common, then VSD) 30% HP:0006695 (ASD)
Undescended testis (males) ~20% HP:0000028
Renal anomalies (agenesis, multicystic dysplastic kidney, pelvic ectasia) 8% HP:0000107
Sensorineural hearing loss rare (2 cases) HP:0000407
Joint hyperlaxity 8 individuals reported HP:0001382
Scoliosis 3 individuals reported HP:0002650
Sleep apnea/abnormal breathing (hyperventilation) uncommon but recurrent HP:0002104

Source: GeneReviews NBK578573; confirmatory cohort data in Bramswig et al. 2017, PMID:28393272 (seizures 6/7, severe ID 6/6, severe speech impairment 6/6, hypotonia 6/7, CNS 5/6, cardiac 4/6, renal 3/4) and Yates/Durkin et al. 2017, PMID:28815871 (all four patients had seizures, developmental delay, ID, neurologic regression, behavioral issues, dysmorphism).

Phenotype characteristics: - Onset: Neonatal (feeding difficulties, hypotonia) to infantile (seizure onset typically <24 months, often the presenting symptom alongside developmental delay). - Severity: Variable but generally moderate-to-severe for ID; a milder end of spectrum has been increasingly recognized in 2024–2025 case reports (including the first familial, vertically transmitted case). - Progression: Hypotonia may evolve into hypertonia/spasticity over time; "data on possible progression of behavior abnormalities or neurologic findings are still emerging" (GeneReviews). - Quality of life impact: Substantial — most individuals are minimally or nonverbally communicative, many require lifelong caregiver support, gastrostomy feeding in severe cases, and educational/behavioral supports (IEP, ABA); no disease-specific QoL instrument data identified in the literature.

4. Genetic/Molecular Information

Causal gene: HNRNPU (HGNC:5048), encoding heterogeneous nuclear ribonucleoprotein U, also known as scaffold attachment factor A (SAF-A). Located at 1q44.

Pathogenic variant spectrum: Nonsense, frameshift, canonical splice-site variants, small intragenic deletions/duplications, and missense variants. Representative variants from Yates/Durkin cohort: c.651_660del (p.Gly218Alafs118), c.1089G>A (p.Trp363), c.1714C>T (p.Arg572), c.2270_2271del (p.Pro757Argfs7) (PMID:28815871).

Variant classification: Per ACMG/AMP, predicted/confirmed loss-of-function variants (nonsense, frameshift, splice-disrupting) are classified pathogenic/likely pathogenic given established haploinsufficiency mechanism; missense variants require careful case-by-case evaluation given absence of clear genotype-phenotype correlation.

Functional consequence: Loss of function / haploinsufficiency — "de novo loss-of-function variants in HNRNPU can lead to a disease phenotype... haploinsufficiency was the main mechanism of pathogenicity" (GeneReviews).

Somatic vs. germline: Germline, virtually always de novo; presumed germline mosaicism reported in one sibling pair; a familial (inherited) case documented in 2025 (PMID:39976380).

Modifier genes: No specific modifier genes identified for isolated HNRNPU variants; however, within 1q43q44 contiguous deletions, co-deletion of AKT3 and ZBTB18 has additive phenotypic effects, effectively modifying overall severity (microcephaly, corpus callosum defects) (PMID:28283832).

Epigenetic information: A robust, reproducible DNA methylation episignature has been identified in blood from individuals with pathogenic HNRNPU variants using Infinium EPIC arrays, distinct from — but partially overlapping — 56 other neurodevelopmental-disorder episignatures. This episignature has clinical utility for reclassifying HNRNPU variants of uncertain significance (VUS) via the EpiSign platform (Genetics in Medicine, 2023, 25(8):100871; related: germline HNRNPU variants and blood methylome alterations, Nature-EJHG; EpiSignature VUS reclassification case report, PMC12688365). This directly implicates HNRNPU's chromatin-regulatory role in disease pathogenesis.

Chromosomal abnormalities: 1q43q44 microdeletions encompassing HNRNPU (along with AKT3, ZBTB18) produce a related but broader contiguous-gene syndrome (microcephaly, corpus callosum abnormalities, epilepsy, short stature) — see Section 2 (PMID:28283832; case report of 163kb 1q44 microdeletion, PMID:22975012).

gnomAD constraint: HNRNPU is highly constrained against loss-of-function variation (consistent with a haploinsufficiency mechanism and near-complete de novo occurrence in patients), in keeping with its essential, ubiquitously expressed roles in RNA processing and chromatin organization (general constraint framework via gnomAD; exact pLI/LOEUF values were not independently retrievable in this search pass and should be confirmed directly against the gnomAD browser at curation time).

5. Environmental Information

No environmental, lifestyle, toxin, or infectious risk factors have been identified as contributing to HNRNPU-related DEE54 — this is a purely monogenic disorder arising from de novo germline variants. The GeneReviews management section notes only a general precaution that "activities and agents that may induce seizures" should be avoided as a secondary, symptom-driven consideration (not an etiologic factor) (GeneReviews).

6. Mechanism / Pathophysiology

Molecular function of HNRNPU/SAF-A: HNRNPU is a highly abundant, ubiquitously expressed nuclear RNA/DNA-binding protein with a multidomain architecture: an N-terminal SAP domain (direct DNA/RNA binding), a SPRY/B30.2 protein-protein interaction domain, a central AAA+ ATPase oligomerization domain, and a C-terminal domain with RGG/RG motifs for RNA binding. It functions as a dynamic bridge between chromatin, nascent RNA, and the nuclear matrix, regulating: (1) pre-mRNA processing and alternative splicing, (2) higher-order chromatin architecture (3D genome organization), (3) transcriptional regulation, and (4) X-chromosome inactivation (ScienceDirect review, PMID:34823151; PLOS Genetics 2025, PMC12176297).

X-chromosome inactivation role: SAF-A/HNRNPU localizes to the inactive X chromosome and interacts directly with XIST lncRNA, and is required for proper XIST RNA territorial localization and XIST-dependent heterochromatin/histone modification — an SAP-domain-dependent function (serines S14/S26 critical) (PLOS Genetics). This is a molecular function distinct from, but potentially contributing to, the neurodevelopmental phenotype's dosage sensitivity.

Causal chain — cortical development (mouse model, Sapir et al. 2022, Nature Communications, PMID:35864088): 1. Trigger: Hnrnpu loss of function in embryonic cortical neuroepithelium. 2. Molecular: Dysregulated alternative splicing of >850 genes, notably Mdm2 (exon 3 skipping → reduced p53 inhibition), Dcc, Siva1 (migration/apoptosis regulators), and cytoskeletal/synaptic transcripts. 3. Cellular: Elevated Tp53 target gene expression → activation of canonical and non-canonical (including necroptotic) p53-dependent cell death. Neural progenitors show markedly higher vulnerability than postmitotic neurons — time-lapse imaging showed progenitor death within 1–5 hours of Cre-mediated excision, versus attenuated death dynamics in postmitotic cells. 4. Tissue: Rapid, near-complete elimination of cortical structures in conditional knockouts. 5. Rescue evidence (mechanistic validation): Pan-caspase inhibitors (Z-VAD-fmk, Q-VD-OPH), p53 inhibitor pifithrin-μ, and necroptosis inhibitor Nec-1 each partially rescued progenitor viability; genetic Tp53 co-deletion "enabled cortical formation, increased progenitors proliferating" (incomplete rescue); co-deletion of the competing splicing factor Srsf3 rescued neurosphere phenotypes and migration defects to near-control levels.

Causal chain — postnatal circuit dysfunction (mouse model, PLOS Genetics 2023, PMC10569524): 1. Heterozygous Hnrnpu+/− mice show global developmental delay, impaired ultrasonic vocalizations, cognitive dysfunction (increased gamma oscillations), and lowered electroconvulsive seizure threshold (p<1×10⁻⁴) despite no spontaneous seizures on prolonged video-EEG. 2. Single-cell RNA-seq of hippocampus/neocortex reveals widespread but modest transcriptional dysregulation (hippocampus: 955 DE events, 73% downregulated; neocortex: 454 DE events, 51% downregulated). 3. Subiculum excitatory neurons carry the highest DEG burden of any cell type examined, with striking enrichment for developmental-delay, epilepsy, and autism-associated genes. 4. Mef2c (a well-known NDD gene) is the most downregulated transcript (50% reduction, log2FC=−1.11, FDR=8×10⁻³⁷), with hnRNP U binding sites at its locus exceeding 99% of other examined genes — directly implicating Mef2c dysregulation as a candidate downstream driver of the epilepsy/cognitive phenotype.

Cross-model convergence: An iScience 2022 study (PMID:36594023) comparing isogenic HNRNPU+/− human iPSC-derived brain organoids to embryonic/perinatal mouse cortex found significant enrichment of shared, co-dysregulated transcripts, supporting a conserved developmental transcriptomic signature across species and model systems (PMC9804147).

Suggested GO terms: GO:0006397 (mRNA processing), GO:0000381 (regulation of alternative mRNA splicing), GO:0006355 (regulation of transcription, DNA-templated), GO:0006974 (DNA damage response), GO:0097191 (extrinsic apoptotic signaling pathway), GO:0070182 (DNA polymerase binding — chromatin structural role), GO:0008380 (RNA splicing).

Suggested CL terms: CL:0000047 (neural stem cell) / CL:0002608 (neural progenitor cell, radial glia), CL:0000679 (glutamatergic neuron, subiculum excitatory), CL:0000540 (neuron, general).

Suggested UBERON terms: UBERON:0001950 (neocortex), UBERON:0002421 (hippocampal formation)/UBERON:0003881 (subiculum), UBERON:0002336 (corpus callosum).

7. Anatomical Structures Affected

Organ level: - Primary: Central nervous system (cerebral cortex, hippocampus/subiculum) — the dominant site of disease. - Secondary/variable involvement: Cardiovascular system (septal defects), renal system (agenesis, dysplasia), eyes (strabismus), ears (rare sensorineural hearing loss), musculoskeletal system (joint laxity, scoliosis), male reproductive system (undescended testis). - Body systems: Nervous system (primary), cardiovascular, renal/urologic, musculoskeletal, ophthalmologic, otologic.

Tissue/cell level: Neuroepithelium/radial glia and neural progenitor cells (most vulnerable population per mouse studies), postmitotic cortical and hippocampal excitatory neurons (subiculum especially implicated), cardiac septal tissue, renal parenchyma.

Subcellular level: Nucleus (chromatin/nuclear matrix — HNRNPU's primary site of action), specifically associated with the inactive X chromosome territory in female cells; splicing machinery/spliceosome-associated nuclear speckles. GO Cellular Component: GO:0005654 (nucleoplasm), GO:0016607 (nuclear speck), GO:0000785 (chromatin).

Localization: Bilateral, diffuse cortical/subcortical involvement (no lateralization reported); brain MRI abnormalities (ventriculomegaly, thin corpus callosum) are typically symmetric/generalized rather than focal.

8. Temporal Development

Onset: Congenital/neonatal manifestations (hypotonia, feeding difficulty) evident from birth or early infancy; developmental delay recognized in first year; seizure onset typically before 24 months of age (~90% of those with seizures), sometimes triggered initially by fever before becoming afebrile.

Onset pattern: Insidious/progressive developmental delay from infancy, punctuated by acute seizure onset events.

Progression: - Developmental trajectory: continued, if slow, developmental gains reported in many individuals rather than regression, though at least one case series (Yates/Durkin, PMID:28815871) reported "neurologic regression" in some patients — indicating phenotypic heterogeneity in course. - Muscle tone: hypotonia in infancy may transition to hypertonia/spasticity later in childhood. - Seizures: generally become more manageable with age and appropriate anti-seizure regimens in most reported individuals, though refractory epilepsy occurs in a subset. - Disease duration: chronic, lifelong; based on current data, "life span is not significantly limited by this condition, as several adults have been reported" (GeneReviews).

Patterns: No formal remission is described; seizure control with medication is common but not universal. No defined "critical intervention window" has been established in the literature to date, though early developmental intervention (0–3 years) is recommended per standard NDD management.

9. Inheritance and Population

Epidemiology: Ultra-rare; exact prevalence/incidence unknown. GeneReviews states "the prevalence of this condition is unknown. To date, approximately 83 individuals with HNRNPU-NDD have been reported" (as of the March 2022 GeneReviews update); by 2025, an international patient registry (Bain Lab, Columbia) had grown to encompass a substantially larger cohort (a 2025 phenotype-expansion paper added 17 previously unpublished patients, and unrelated registry sources describe well over 100 confirmed individuals) (Hodgson et al. 2025, PMID:39976380).

Inheritance pattern: Autosomal dominant, virtually always de novo. Rare instances of parental germline mosaicism and, as of 2025, the first documented familial (parent-to-child) transmission.

Penetrance: Appears complete/high for the core developmental delay and seizure phenotype among reported carriers, though ascertainment bias (severe cases more likely to be sequenced) may inflate apparent penetrance; the emerging recognition of milder cases suggests a broader phenotypic spectrum than initially appreciated.

Expressivity: Highly variable — ranging from the "classic" severe DEE54 presentation to milder cases now increasingly reported (2024–2025 case series), without clear genotype-phenotype correlation.

Genetic anticipation: Not described (not applicable to a haploinsufficiency-mechanism, non-repeat-expansion disorder).

Founder effects / consanguinity: Not applicable — disorder is dominant and de novo, not associated with consanguinity or population founder mutations.

Carrier frequency: Not applicable (de novo dominant disorder, not a recessive carrier state).

Population demographics: - No ethnic or geographic predilection reported; cases have been described from multiple continents (US, Europe, various case series). - Sex ratio: No clear sex bias reported in the literature reviewed; both males and females affected (male-specific phenotype item: undescended testis in ~20% of affected males). - Age distribution: Predominantly diagnosed in infancy/childhood via exome/genome sequencing for developmental delay and epilepsy; adults are increasingly recognized as historically underdiagnosed due to lack of earlier genetic testing availability.

10. Diagnostics

Establishing the diagnosis: Requires "a proband with suggestive findings and a heterozygous pathogenic variant in HNRNPU identified by molecular genetic testing" (GeneReviews).

Molecular testing approaches, by yield: - Sequence analysis (exome/genome or NDD/epilepsy gene panel including HNRNPU): ~98% detection rate. - Gene-targeted deletion/duplication analysis (or chromosomal microarray for 1q43q44 deletions): ~2% of cases (contiguous gene deletion presentations). - Single-gene sequential testing of HNRNPU alone is "rarely useful and typically NOT recommended" given the nonspecific phenotype — multigene panel or exome/genome sequencing is preferred.

Genetic testing modalities: - Multigene intellectual disability/epileptic encephalopathy panel including HNRNPU. - Whole exome sequencing (WES) — most commonly used diagnostic route in published cohorts. - Whole genome sequencing (WGS) — increasingly used, captures structural variants and intronic/regulatory changes missed by exome. - Chromosomal microarray (CMA) — detects 1q43q44 microdeletions/duplications encompassing HNRNPU. - DNA methylation episignature (EpiSign) testing — emerging clinical tool for reclassifying HNRNPU VUS, given the validated disease-specific episignature (Genetics in Medicine 2023, 25(8):100871).

Clinical/laboratory tests: No disease-specific biomarker or metabolic screening test exists; diagnosis is genetic. Brain MRI is used to characterize (not diagnose) the condition — ventriculomegaly and thin corpus callosum are the most frequent findings (61% of imaged cases abnormal). EEG is used to characterize seizure semiology, not to establish diagnosis.

Differential diagnosis: Broad — "all disorders with epileptic encephalopathy and intellectual disability without other distinctive findings should be considered," referencing the OMIM Developmental and Epileptic Encephalopathy Phenotypic Series for systematic comparison (GeneReviews). Specific considerations include other DEE-causing genes, and — for patients with contiguous deletions — the broader 1q43q44 microdeletion syndrome (distinguishing isolated HNRNPU variants from deletions also involving AKT3/ZBTB18, which add microcephaly and corpus callosum anomalies).

Screening: No population or newborn screening applicable (ultra-rare, not detectable by biochemical newborn screening); prenatal testing/preimplantation genetic testing available for known familial variants (relevant given the newly documented familial transmission case).

11. Outcome/Prognosis

Survival/mortality: Life expectancy does not appear significantly reduced; "several adults have been reported" and formal life-span data, while incomplete, does not suggest premature mortality as a defining feature (GeneReviews). No disease-specific mortality statistics are available given the ultra-rare, only-recently-delineated nature of the condition.

Morbidity/function: Substantial lifelong disability — most affected individuals have moderate-to-severe intellectual disability, are minimally/nonverbally communicative, and require ongoing multidisciplinary support (PT/OT/speech, special education, seizure management). No validated disease-specific quality-of-life instrument has been applied in the literature reviewed.

Complications: Refractory epilepsy in a subset; feeding/nutritional complications (sometimes requiring gastrostomy); sleep apnea requiring respiratory support; secondary orthopedic complications of hypertonia/spasticity (contractures) or hyperlaxity (scoliosis).

Prognostic factors: No genotype-phenotype correlation identified to date, so variant type does not predict severity. Severity appears greater when HNRNPU loss occurs in the context of a larger 1q43q44 contiguous gene deletion (additional microcephaly, corpus callosum defects from co-deleted AKT3/ZBTB18).

Recovery potential: Developmental gains continue in many individuals with early, sustained intervention, though the underlying intellectual disability persists lifelong; seizure control is achievable in most patients with standard or combination anti-seizure therapy.

12. Treatment

Pharmacotherapy (seizures): - Sodium valproate — "the most commonly used & effective medication" per GeneReviews for first-line seizure control (NCIT:C15986 Pharmacotherapy; therapeutic agent CHEBI valproate). - Newer-generation anti-seizure medications for refractory cases (unspecified beyond class in the literature reviewed). - Ketogenic diet — recommended for refractory seizures; broader epilepsy literature supports its efficacy and safety, including in combination with valproate, for drug-resistant epilepsy generally (NCIT:C15447 Dietary Intervention). - Combined pharmacotherapy for associated symptoms has been documented in individual cases (acetazolamide, alprazolam, aripiprazole) — indication-specific, not seizure-first-line.

Advanced/experimental therapeutics: - Antisense oligonucleotide (ASO) therapy: In 2024, the Bain Lab (Columbia University) launched the first precision-therapeutics trial for the related HNRNP-family disorder H2-RNDD (HNRNPH2), administering individualized ASO therapy to 8 patients via the n-Lorem Foundation (n-of-1 model). While this specific program targets HNRNPH2 rather than HNRNPU, it establishes proof-of-concept for RNA-targeted precision therapeutics in the HNRNP gene family and signals a plausible future therapeutic direction for HNRNPU-NDD given the shared haploinsufficiency mechanism and RNA-binding-protein biology (NCIT:C15238 Gene Therapy category; therapeutic_modality: ANTISENSE_OLIGONUCLEOTIDE). No HNRNPU-specific ASO clinical trial has yet been identified in this search. - Preclinical ASO work in the HNRNPH2 mouse model (Science Translational Medicine 2025/2026) demonstrates feasibility of splice/expression-modulating ASO rescue in this gene family, informing potential translational strategies for HNRNPU.

Surgical/interventional: No disease-specific surgery; organ-specific procedures as needed (cardiac septal defect repair, orchiopexy for undescended testis, orthopedic procedures/Botox for severe spasticity or scoliosis).

Supportive/rehabilitative care: - Early intervention programs (0–3 years), developmental preschool (3–5 years), individualized education plans. - Physical therapy (gross motor/tone), occupational therapy (fine motor/adaptive function), speech-language pathology with augmentative/alternative communication (AAC) evaluation (NCIT:C15302 Physical Therapy). - Feeding therapy for dysphagia; nasogastric or gastrostomy tube placement for persistent feeding dysfunction. - Respiratory support (supplemental O2, CPAP/BiPAP) for sleep apnea. - Tone management: baclofen, tizanidine, botulinum toxin, or orthopedic procedures for hypertonia (physical medicine & rehabilitation involvement). - Behavioral intervention: applied behavior analysis (ABA) for autism-related features; pediatric psychiatry for severe aggressive/destructive behaviors.

Organ-specific surveillance/management: Cardiology (congenital heart defects), ophthalmology (strabismus), audiology (hearing loss), nephrology (renal anomalies), urology (undescended testes).

Treatment strategy: No formal published treatment algorithm specific to HNRNPU-NDD beyond the GeneReviews management/surveillance framework; management is symptom-directed and multidisciplinary, following general DEE/intellectual disability care pathways.

Personalized medicine: DNA methylation episignature profiling offers a genotype-informed diagnostic refinement tool (VUS reclassification); no pharmacogenomic guidance specific to HNRNPU-NDD identified.

13. Prevention

Primary prevention: Not applicable in the traditional sense — this is a de novo genetic disorder with no known modifiable environmental trigger to avoid.

Secondary prevention: Early diagnosis via exome/genome sequencing in infants presenting with early-onset seizures and developmental delay allows earlier initiation of supportive therapies and seizure management, potentially reducing secondary complications (status epilepticus, feeding-related morbidity).

Genetic counseling: Recurrence risk to siblings of an affected proband is low but not zero (given documented germline mosaicism cases), warranting parental testing to confirm de novo status. For an affected individual, the risk of transmission to offspring is 50% per pregnancy (autosomal dominant); prenatal testing and preimplantation genetic testing (PGT) are available for families with a known familial variant, an option that gained clinical relevance following the first reported vertical transmission case in 2025.

Screening: No population-level newborn or carrier screening applicable (ultra-rare, dominant, de novo disorder without ethnic founder effects).

Public health/behavioral interventions: Not applicable — no known modifiable population-level risk factor.

Prophylaxis: General epilepsy-safety precautions (avoiding known seizure triggers) apply once a seizure phenotype is established, as with any epilepsy syndrome.

14. Other Species / Natural Disease

Taxonomy: Naturally occurring HNRNPU-related disease has not been reported in non-human species; the gene is highly conserved (near-universal expression and function across vertebrates).

Model organism gene: Mouse Hnrnpu (MGI ortholog); human HNRNPU NCBI Gene ID 3192.

Natural disease in other species: Not reported in OMIA or veterinary literature reviewed — this is not a recognized naturally occurring veterinary disease.

Comparative biology: HNRNPU/SAF-A is highly conserved across mammals given its essential, ubiquitous roles in chromatin organization and RNA processing; the core molecular mechanism (splicing regulation, p53-dependent progenitor apoptosis upon loss) is conserved between mouse and human, as demonstrated by the organoid/mouse cross-species transcriptomic convergence study (PMID:36594023).

Zoonotic potential/cross-species susceptibility: Not applicable — this is a genetic, non-infectious disorder.

15. Model Organisms

Mouse models (genetic, knockout/conditional): 1. Constitutive/germline heterozygous Hnrnpu+/− mouse (PLOS Genetics 2023, PMID likely associated with PMC10569524): Models the human haploinsufficiency state. Recapitulates global developmental delay, impaired ultrasonic vocalizations, cognitive dysfunction (elevated wakeful gamma oscillations), and lowered electroconvulsive seizure threshold (p<1×10⁻⁴), though without spontaneous seizures on extensive video-EEG (300+ hours) — a partial phenotype recapitulation (increased seizure susceptibility rather than overt epilepsy). Single-cell RNA-seq identifies subiculum excitatory neurons as the most transcriptomically perturbed cell type, with Mef2c as the most robustly downregulated candidate driver gene. 2. Conditional cortical Hnrnpu knockout mouse (Nature Communications 2022, PMID:35864088, Sapir/Reiner lab): Demonstrates near-complete cortical structure elimination via p53-dependent apoptosis of neural progenitors (more vulnerable) and postmitotic neurons (less vulnerable, delayed death). Provides mechanistic and pharmacological rescue data (caspase inhibitors, p53 inhibitor pifithrin-μ, necroptosis inhibitor Nec-1; genetic Tp53 or Srsf3 co-deletion). This model captures the severe, embryonic-lethal-if-homozygous end of the mechanistic spectrum rather than the milder heterozygous human phenotype.

In vitro/organoid models: - Isogenic human iPSC-derived brain organoids (PGP1 line) with CRISPR-engineered HNRNPU+/− frameshift variants (1bp duplication, 10bp deletion), showing ~25% reduction in HNRNPU protein and reduced mRNA. Used for cross-species transcriptomic comparison, confirming conserved dysregulated gene modules between 45-day human organoids and embryonic mouse cortex (PMID:36594023; PMC9804147).

Model limitations: - Mouse heterozygous models show increased seizure susceptibility rather than the spontaneous, often treatment-refractory epilepsy seen in human patients — a translational gap (species-scale/circuit-maturation difference) worth flagging as a HUMAN_MODEL_MISMATCH if curated into a pathophysiology module. - The severe conditional cortical-knockout mouse models complete loss of function in a defined lineage/timepoint, which is more mechanistically informative for the p53/splicing pathway than directly representative of the human heterozygous dosage state. - No zebrafish, Drosophila, or C. elegans HNRNPU ortholog disease model was identified in this search (in contrast to the sister gene HNRNPH2, for which zebrafish/mouse models are more developed).

Applications: These models collectively support (1) the mechanistic causal chain from HNRNPU loss → splicing dysregulation (Mdm2, Mef2c, and hundreds of other targets) → p53-dependent progenitor death / circuit-level excitatory neuron dysfunction → cortical malformation and seizure susceptibility, and (2) proof-of-concept pharmacological (caspase/p53/necroptosis inhibitors) and genetic (Tp53, Srsf3 co-deletion) rescue strategies that could inform future therapeutic development, complementing the ASO precision-medicine approach being piloted in the related HNRNPH2 disorder.

Resources: MGI (Mouse Genome Informatics) for Hnrnpu allele records; no dedicated IMPC/KOMP full-line data specifically confirmed in this search pass — recommend direct IMPC query at curation time for standardized phenotyping data.

Summary of Key Ontology Term Suggestions

Category Suggested terms
MONDO MONDO:0033363 (DEE54)
OMIM #617391
HGNC HGNC:5048 (HNRNPU)
HPO (phenotypes) HP:0001263, HP:0002342, HP:0001344, HP:0001250, HP:0002069, HP:0002121, HP:0001252, HP:0001999, HP:0011968, HP:0000729, HP:0004322, HP:0000486, HP:0002119, HP:0002079, HP:0006695, HP:0000028, HP:0000107, HP:0000407
GO (biological process) GO:0006397, GO:0000381, GO:0006355, GO:0006974, GO:0097191, GO:0008380
GO (cellular component) GO:0005654, GO:0016607, GO:0000785
CL CL:0002608 (neural progenitor/radial glia), CL:0000679 (glutamatergic neuron)
UBERON UBERON:0001950 (neocortex), UBERON:0003881 (subiculum), UBERON:0002336 (corpus callosum)
NCIT (treatment) NCIT:C15986 (Pharmacotherapy), NCIT:C15447 (Dietary Intervention/ketogenic diet), NCIT:C15302 (Physical Therapy), NCIT:C15238 (Gene Therapy/ASO)
CHEBI valproate/valproic acid

Sources

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Outcome Count
References checked 15
Resolved 15
Unresolved (possible confabulation) 0
Unverifiable 0
References weighed for topical relevance 15
On topic 11
Off topic 0

All extracted references resolved successfully.