Hackmann-Di Donato Syndrome (NKAP-related X-linked syndromic intellectual disability): Research Report

Prepared: 2026-10-15 · Target: MONDO:0026733 · Category: Mendelian

Sources and how far to trust them. Most of this report rests on the 18 references already cached in this worktree's references_cache/. Quotes from those are marked (abstract) or (full text) and can be copied as snippets. I also used two web sources: - the PMC full text of Fiordaliso et al. 2019 (PMC6848994), which supplies the per-feature counts; - the NCBI MedGen record (C5393302).

The per-feature counts came through a WebFetch summary, not a cached file. Check every number and quote taken from that source against the PDF before using it as a snippet.

Ontology CURIEs below come from MedGen, the stub, or the draft entry, or are leads I wrote from memory. Marked ⚠ means not yet looked up. Per the repository's term contract, run runoak or just validate-terms on each before binding.

1. Disease Information

Overview. Hackmann-Di Donato syndrome (HDDS; OMIM-style symbol MRXSHD) is an X-linked recessive neurodevelopmental disorder. It is caused by hemizygous missense variants in NKAP (Xq24), which encodes the NF-κB activating protein. NKAP is a nuclear-speckle and spliceosome-associated protein that also acts as a transcriptional corepressor with HDAC3.

The defining publication is Fiordaliso et al. 2019 (PMID:31587868, Am J Hum Genet). It reported 10 affected males from 8 families with: - developmental delay or intellectual disability; - hypotonia; - Marfanoid habitus with tall stature; - joint laxity, arachnodactyly and scoliosis; - behavioral abnormalities; - a recognizable long-face gestalt.

"we report a disorder of transcriptional regulation due to missense mutations in the X chromosome gene, NKAP. These mutations are clustered in the C-terminal region of NKAP where NKAP interacts with HDAC3 and post-catalytic spliceosomal complex proteins." — PMID:31587868 (abstract)

"The clinical features of affected individuals are highly conserved and include developmental delay, hypotonia, joint contractures, behavioral abnormalities, Marfanoid habitus, and scoliosis." — PMID:31587868 (abstract)

Eponym. The syndrome is named after Karl Hackmann and Nataliya Di Donato. Their 2016 re-examination of patients clinically diagnosed with Lujan-Fryns syndrome found an NKAP variant in one individual:

"X-exome sequencing of four individuals revealed private or non-recurrent mutations in NKAP and LAS1L in one patient each." — PMID:26358559 (abstract)

Identifiers

Synonyms - Hackmann-Di Donato syndrome; also spelled Hackman-Di Donato (PMID:41917976 uses the single-n form) - Intellectual developmental disorder, X-linked, syndromic, Hackmann-Di Donato type (MRXSHD) - Mental retardation, X-linked, with Marfanoid habitus, 2 (MedGen) - NKAP-related syndrome / NKAP mutation-related disorder (PMID:36929417, PMID:38647244)

Data provenance. Everything here comes from published case series and case reports (aggregated, disease-level). No EHR or registry data exist.

Reported individuals to date (about 14): - 10 males in Fiordaliso 2019; - 2 male fetuses in Xu 2024 (PMID:38647244); - 1 boy with an exon-4 variant of uncertain significance in Semyachkina 2026 (PMID:41917976); - 1 or more members of a family in Gérard 2026 (PMID:42481913), including "the oldest known patient".

Gérard 2026 states: "Only 12 individuals harboring pathogenic NKAP variants have been reported in the literature."

2. Etiology

Primary cause. Germline hemizygous missense variants in NKAP. Every variant classified pathogenic or likely pathogenic falls in exons 8–9, which encode the C-terminal (DUF926) region.

"All pathogenic nucleotide variants were located in exons 8 and 9, encoding the C-terminal part of NKAP." — PMID:41917976 (full text)

Genetic risk factors - Variant hotspots at p.Arg330 and p.Arg333 (see §4). - Unaffected or mildly affected carrier mothers. Three affected males in Fiordaliso 2019 inherited the variant from such mothers. Xu 2024 found carrier grandmothers and aunts in a multigenerational family. - Maternal germline mosaicism in one sibship. Two brothers carried an apparently de novo variant (PMID:41917976 full text: "In one pedigree, two siblings appeared to have inherited the mutation due to maternal germline mosaicism."). - Modifier genes: none identified.

Environmental risk factors. None reported. In Xu 2024 the mother of the CHD fetuses reported "no history of illness, medication usage or exposure to either radiation or harmful substances during her pregnancy" (PMID:38647244 full text).

Protective factors and gene–environment interactions. None reported.

Sex-specific protection in females. The plausible explanation is skewed X-inactivation, but it is only partly demonstrated: - Mouse: heterozygous females with a deleterious Nkap allele showed 83–97% inactivation of the mutant X, explained as secondary selection against cells expressing the mutant allele (PMID:41828556). - Human: one report of a nonsense NKAP variant in a hemophilia A carrier with extreme skewing. It is cited as ref. 10 in PMID:41828556, but its PMID is not cached; fetch it before citing. - Not measured in HDDS carrier mothers. Fiordaliso 2019 reports none: "Mothers carrying NKAP mutations were unaffected or much less severely affected." This is a curatable knowledge gap.

3. Phenotypes

Frequencies are from Fiordaliso 2019 Table 1 (n=10 males, ages 4–21 years). Verify them against the PDF before encoding. Suggested HPO terms come from MedGen's HPO list where available; ⚠ marks a term to look up.

Core neurodevelopmental features

Growth and habitus

Musculoskeletal

Craniofacial

Cardiovascular

Fiordaliso 2019 reports roughly 6/10 with a cardiac finding. Xu 2024 counts 8/12 across all reported cases, including the two fetuses (PMID:38647244 Table 3).

Genotype–phenotype note:

"Particularly, all four individuals carrying mutation at p.R330 had cardiac defects, strongly suggesting a correlation between NKAP p.R330 mutation and cardiac development." — PMID:38647244 (full text)

Genitourinary

Prenatal findings (PMID:38647244)

• Increased nuchal translucency of 3.0 mm at 12 weeks (HP:0010880 ⚠).
• Mild lateral ventriculomegaly.
• Choroid plexus cysts (HP:0002190 ⚠).
• Fetal tachycardia.

"These prenatal anomalies include cardiac defect and thickened NT, which could be observed as early as 12 weeks gestation" — PMID:38647244 (full text)

Phenotype the draft entry includes but the sources do not support

Macrocephaly is in the draft YAML description. It does not appear in Fiordaliso 2019 (per the full-text read), MedGen's HPO list, or any cached reference. Remove it from the description unless a source turns up.

Quality of life

No QoL instrument data exist (no EQ-5D, SF-36 or PROMIS). The main inferred burden is lifelong ID with behavioral problems, connective-tissue complications (scoliosis, joint instability), and cardiac follow-up.

4. Genetic / Molecular Information

Causal gene: NKAP (hgnc:29873; NCBI Gene 79576; OMIM 300766; Xq24). The transcript is NM_024528.4, encoding a 415-amino-acid protein of about 47 kDa. Its domains are: - an N-terminal RS (arginine/serine-rich) domain that mediates nuclear localization; - a central basic region; - a C-terminal DUF926 domain that binds HDAC3 and spliceosomal proteins. Shapiro 2019 (PMID:30804042) places the minimal HDAC3-binding site around F347/Y352.

Variants reported

Xu 2024 Table 3 prints c.988G>A for p.R333Q in some rows. Fiordaliso's own text gives c.998G>A, which is the correct HGVS. Use c.998G>A.

• Variant type: all are missense. No truncating, splice or copy-number variant has been reported in an affected male. This is consistent with complete loss of NKAP being lethal (see the model data below).
• Population frequency: the hotspot alleles are absent from gnomAD (PM2 cited in PMID:38647244).
• Origin: germline. There is no somatic role in HDDS. Somatic NKAP loss in sarcoma (PMID:27694884) is a separate cancer-biology observation and not part of this disease.

Functional consequence. This is best described as partial or altered loss of function confined to the C-terminal interaction domain, not null: - NKAP mRNA and protein levels were normal in patient lymphoblastoid cell lines (LCLs), and HDAC3 levels and distribution were unchanged (Fiordaliso, full text). - p.R330C reduced NKAP–HDAC3 co-IP from 44.9% to 16.7% and shifted NKAP to the nuclear periphery in HEK293T/HeLa overexpression experiments (PMID:38647244). - In mice, complete Nkap deletion in hematopoietic cells is lethal (PMID:20610818). In zebrafish, a C-terminal truncation is lethal by 4 dpf (PMID:31587868).

Suggested functional_impact_category: PARTIAL_LOSS_OF_FUNCTION, or leave it unset with a note. Dominant-negative and gain-of-function have not been tested; do not assert either.

Genotype–phenotype. p.Arg330 variants are associated with congenital heart disease (4/4 carriers). No other correlation is established.

Epigenetics. No disease-specific methylation signature (episignature) has been reported. Skewed X-inactivation in carrier females has been shown only in the mouse (PMID:41828556).

Chromosomal abnormalities. None. This is a point-mutation disorder, and karyotype and chromosomal microarray were normal in the fetal proband (PMID:38647244).

ClinGen. No gene–disease validity curation for NKAP–HDDS was found. Leave gene_disease_validity absent.

5. Environmental Information

There are no environmental, lifestyle or infectious contributors. The environmental: section should stay empty.

6. Mechanism / Pathophysiology

Ordered causal chain

1. Germline hemizygous missense variant in the NKAP C-terminal (DUF926) domain (exons 8–9; hotspots R330, R333). Demonstrated in patients (PMID:31587868).
2. The variant alters NKAP protein interactions without lowering NKAP abundance.
3. Patient LCLs: NKAP mRNA and protein are normal (PMID:31587868 full text).
4. 2a. Reduced NKAP–HDAC3 association. Demonstrated in vitro for p.R330C only (PMID:38647244). Inferred for other variants, which sit in the HDAC3-binding region. The importance of this interaction is supported by the Y352A HDAC3-binding-null mouse (PMID:30804042, PMID:31387873).
5. 2b. Possible perturbation of NKAP's role in the post-catalytic spliceosome (C*/P complex, exon ligation). Inferred from the structural location of the variants (PMID:30705154: "CACTIN, SDE2, and NKAP-factors implicated in alternative splicing-further stabilize the catalytic conformation of the spliceosome during exon ligation").
6. 2c. Altered intranuclear (nuclear speckle) distribution. In vitro, p.R330C only (PMID:38647244). The nuclear-speckle framing is from the review PMID:36929417.
7. Transcriptional dysregulation. Loss of NKAP–HDAC3 corepressor activity, including on Notch targets, leads to a distinctive transcriptome signature.
8. Patient LCLs show 455 upregulated and 721 downregulated genes, with HES1 and JAG1 up (PMID:31587868 full text).
9. 3a. Long genes with many exons are selectively downregulated. Demonstrated in patient LCLs (PMID:31587868 abstract: "transcriptome signature, which is characterized by the downregulation of long genes with higher exon numbers").
10. 3b. Notch target genes are de-repressed. Demonstrated in mouse T cells (PMID:19409814: Hes1, Deltex1 8–20-fold higher with NKAP loss). Patient LCLs show HES1/JAG1 changes. That this occurs in neural or mesenchymal tissue is inferred.
11. Splicing changes in patient cells were modest: only 39 events, with some increase in intron retention. This is weak evidence that mis-splicing is the main driver.
12. Branching tissue consequences (all inferred; none shown in human tissue):
13. 4a. Brain: disrupted neurodevelopmental gene programs lead to ID/DD, hypotonia and behavioral abnormality. Supporting context: NKAP is highly expressed in neural progenitors and migrating neuroblasts (mouse SVZ/RMS, PMID:24583034), and long, exon-rich genes are enriched among neuronal and synaptic genes. That second point is my inference, not a statement from the cited sources.
14. 4b. Connective tissue / skeleton: dysregulated extracellular-matrix gene expression (ECM terms enriched in the LCL differentially expressed genes) leads to Marfanoid habitus, joint laxity, arachnodactyly and scoliosis. Inferred; the mechanism is unknown.
15. 4c. Heart: altered expression of cardiac morphogenesis genes (JAG1 and DNAH11 up, DHRS3 down with p.R330C in HEK293T) leads to septal defects, outflow-tract and valve anomalies (PMID:38647244). In vitro inference. The authors themselves note that the direction of change for DHRS3 conflicts with the LCL data.
16. Clinical syndrome: ID, hypotonia, Marfanoid habitus, connective-tissue features and variable CHD.

Parallel branch to model as a hypothesis, not a demonstrated disease mechanism

Genome-stability functions of NKAP: - SUMOylated NKAP anchors CENP-E at kinetochores for chromosome alignment (PMID:27694884). - NKAP–HDAC3 prevents R-loop accumulation and DNA damage (PMID:37322264). - Mutant NKAP mouse embryonic fibroblasts proliferate 3.4-fold less, become senescent, and upregulate p27/p57, IL-1β and IL-6 (PMID:41828556).

None of these has been tested with HDDS patient variants. Their relevance to HDDS is a knowledge gap. The SUMOylation sites lie in amino acids 277–415, which overlaps the patient-variant region and makes this a reasonable hypothesis.

NKAP biology by category

• Molecular pathways:
• Notch signaling (NKAP is a corepressor with CIR/HDAC3): PMID:19409814, PMID:24583034.
• NF-κB activation (original description, overexpression in 293 cells): PMID:14550261.
• Pre-mRNA splicing at exon ligation: PMID:30705154.
• Cellular processes: transcriptional repression; mitotic chromosome alignment; R-loop suppression and genome stability; hematopoietic stem cell maintenance (PMID:20610818, PMID:30804042); T/iNKT/Treg maturation (PMID:21624937, PMID:23481390, PMID:31387873). The immune and hematopoietic phenotypes are mouse conditional-knockout findings with no reported human HDDS counterpart. Do not add immune phenotypes.
• Protein dysfunction: altered binding (HDAC3) and altered localization. Not misfolding or aggregation.
• Metabolic and immune involvement: none reported in patients.
• Molecular profiling: LCL RNA-seq in Fiordaliso 2019. No GEO accession is quoted in the summary I have; check the paper's data availability statement before running just discover-datasets.

Suggested ontology terms (⚠ verify all)

GO, biological processes - GO:0000122 negative regulation of transcription by RNA polymerase II - GO:0045746 negative regulation of Notch signaling pathway - GO:0000398 mRNA splicing, via spliceosome - GO:0051310 metaphase chromosome alignment - GO:0007399 nervous system development - GO:0007507 heart development - GO:0030198 extracellular matrix organization

GO, cellular components - GO:0016607 nuclear speck - GO:0071013 catalytic step 2 spliceosome - GO:0000776 kinetochore

CL, cell types - CL:0000540 neuron - CL:0011020 neural progenitor cell - CL:0000057 fibroblast - CL:0000746 cardiac muscle cell (for the CHD branch) - CL:0000037 hematopoietic stem cell (model-only context)

Interacting gene: HDAC3 (HGNC ID ⚠ — look it up; do not write it from memory).

7. Anatomical Structures Affected

Primary systems - Central nervous system: brain, UBERON:0000955 ⚠. - Musculoskeletal and connective tissue: vertebral column UBERON:0001130 ⚠ (scoliosis); joints; digits. - Heart: UBERON:0000948 ⚠. Specifically the interventricular and interatrial septa, mitral valve and aortic root.

Secondary systems - Testis / cryptorchidism (urogenital). - Cerebral white matter (one family).

Subcellular: nucleus, nuclear speckles, spliceosome, and kinetochores during mitosis.

Lateralization: systemic and bilateral. Scoliosis is asymmetric by nature.

8. Temporal Development

• Onset: congenital or prenatal.
• Cardiac defects and increased nuchal translucency are detectable from 12 weeks' gestation (PMID:38647244).
• Hypotonia and developmental delay appear in infancy.
• Course: a static neurodevelopmental disorder, with no regression reported.
• Skeletal features such as scoliosis and Marfanoid proportions become more evident during childhood growth.
• The white matter lesions in the Gérard family were "stable".
• Duration: lifelong.
• Ages reported: 4–21 years in the original cohort. Gérard 2026 reports the oldest known patient; the exact age is not in the abstract.
• Stages, remission, critical periods: not defined. Early intervention during infancy and childhood is the main window for developmental therapies (general principle; not disease-specific evidence).

9. Inheritance and Population

• Inheritance: X-linked recessive. HP:0001419 is used in the draft; the OMIM/MedGen designation agrees.
• Penetrance in hemizygous males: apparently complete for C-terminal variants, with variable expressivity, especially for cardiac involvement.
• Heterozygous females:
• Carrier mothers were "unaffected or much less severely affected".
• In Xu 2024, female carriers (mother, aunt, grandmother) were healthy.
• An uncle in that family had developmental delay and spinal instability and died at age 1; he was not genotyped.
• De novo rate: 6/10 in the original cohort, as stated by Xu 2024 (Fiordaliso reports 4 apparently de novo individuals plus 2 brothers from germline mosaicism).
• Germline mosaicism: yes, one maternal case.
• Anticipation, founder effects, consanguinity: none.
• Epidemiology:
• Prevalence is unknown.
• About 14 individuals have been reported worldwide.
• Populations: European (German, Czech, Italian, UK), US, Japanese, Korean, Chinese (fetuses) and Russian/Slavic.
• Suggested prevalence record: measure_type: CASES_IN_LITERATURE, prevalence_class: ULTRA_RARE or NOT_YET_DOCUMENTED, citing PMID:42481913 ("Only 12 individuals …").
• Sex ratio: affected individuals are male.

10. Diagnostics

Genetic testing (the main route) - Exome or genome sequencing, ideally trio, is the diagnostic test. Trio exome was used in PMID:38647244 and PMID:42481913, and genome sequencing in PMID:41917976. - NKAP is on the Genomics England PanelApp intellectual disability panel (PanelApp). - X-exome sequencing first detected it in a Lujan-Fryns-like patient (PMID:26358559). - Chromosomal microarray and karyotype are normal; they are useful only for excluding differentials.

Supporting work-up - Echocardiography for septal, valve and aortic-root disease. - Spine radiographs for scoliosis. - Brain MRI: white matter lesions in one family. Gérard 2026 suggests systematic neuroimaging. - EEG if seizures occur. - Prenatal: nuchal translucency and fetal echocardiography, with prenatal exome sequencing in an at-risk family.

Biomarkers: none validated. The patient-LCL transcriptome signature (downregulated long, exon-rich genes) is research-only.

Clinical criteria: none formalized. Hackmann 2016 argues that "the combination of ID and marfanoid habitus in a male patient is not sufficient for the diagnosis of LFS" (PMID:26358559).

Differential diagnosis (all Marfanoid habitus with ID): - Lujan-Fryns syndrome (MED12; also UPF3B, ZDHHC9): nasal speech and specific facies. - Marfan syndrome (FBN1): ectopia lentis and aortic dilatation, with normal intellect. - Loeys-Dietz syndrome. - Homocystinuria: run plasma homocysteine. - Stickler syndrome. - Other nuclear-speckleopathies (ZTTK/SON) (PMID:36929417).

11. Outcome / Prognosis

• Survival: no data. Postnatal patients have been followed into adulthood (21 years in the original cohort; older in the Gérard family).
• Pregnancy outcomes: both p.R330C fetuses were terminated after a CHD diagnosis.
• Morbidity: lifelong intellectual disability; behavioral problems; risk of scoliosis and joint dislocation; cardiac valve and septal disease, plus potential aortic-root dilatation needing surveillance.
• Prognostic factors: p.Arg330 variants predict CHD (based on 4/4 carriers). Nothing else is known.
• QoL measures: none published.

12. Treatment

There is no disease-specific or disease-modifying therapy, and no clinical trials. ClinicalTrials.gov and ICTRP searches are not indicated; none was found. Management is supportive and follows standard care for ID and connective-tissue disorders. These recommendations are by analogy; no HDDS-specific guideline exists.

Evidence gap. No cached reference describes treatment outcomes. Any treatment rows will need evidence: cite the case reports' management descriptions, or record each row as a general-care recommendation with a notes: explanation.

13. Prevention

• Primary prevention: not applicable.
• Genetic counseling:
• A carrier mother has a 50% risk of an affected son with each male pregnancy.
• An apparently de novo case still carries recurrence risk because of possible germline mosaicism, which has been documented.
• Testing options in a known family: carrier testing of female relatives, prenatal diagnosis (shown in PMID:38647244), and preimplantation genetic testing.
• Secondary prevention: cardiac and spinal surveillance after diagnosis.
• Newborn screening: not applicable.

14. Other Species / Natural Disease

• No naturally occurring animal disease is known, and nothing is listed in OMIA (not searched exhaustively).
• Orthologs:
• Mouse Nkap (on the X chromosome).
• Zebrafish nkap.
• Rat to human conservation of the C-terminal lysines (PMID:27694884).
• R330 is "highly conserved among various species" (PMID:38647244).
• NCBI Gene IDs for the orthologs were not retrieved; look them up before encoding.

15. Model Organisms

Suggested model links (modeled_mechanisms): - Zebrafish → "Dysregulated Transcription": PARTIALLY_RECAPITULATES, fidelity: LOW. - Mouse Δ123bp → the hypotonia phenotype: PARTIALLY_RECAPITULATES, fidelity: LOW. Add a BOUNDARY_OMISSION divergence (no Marfanoid habitus) and a PROXY_QUANTITY divergence (grip strength standing in for hypotonia).

Knowledge gap worth recording (HUMAN_MODEL_MISMATCH): no model carries a patient C-terminal missense allele, and no iPSC-derived neuron or organoid model exists.

Notes for curating kb/disorders/Hackmann-Di_Donato_Syndrome.yaml

1. File name. The stub's proposed_name is Intellectual_Developmental_Disorder_X-linked_Syndromic_Hackmann-Di_Donato_Type, but the draft uses Hackmann-Di_Donato_Syndrome. Either is fine. Delete stubs/Intellectual_Developmental_Disorder_X-linked_Syndromic_Hackmann-Di_Donato_Type.yaml in the PR.
2. Remove "macrocephaly" from the draft description; no source supports it. "Joint hypermobility" (9/10 joint laxity) and "scoliosis" are supported.
3. Build out the pathograph. Add the HDAC3-binding and intranuclear-localization nodes, split "Dysregulated Transcription" into its Notch-target and long-gene branches, and wire edges to the neurodevelopmental, connective-tissue and cardiac phenotypes. Without these the phenotypes stay disconnected (just list-disconnected-phenotypes).
4. Discussions. Record XCI in carrier mothers and the genome-stability functions as KNOWLEDGE_GAP discussions.
5. Exon-4 variant. Keep p.Arg204Ser out of the causative variant list or mark it VUS (clinical_significance), citing PMID:41917976.
6. No GeneReviews chapter is expected. Confirm with just check-genereviews.
7. Every snippet quoted above from an "(abstract)" or cached "(full text)" source should be checked with just count-verified-snippets. Numbers from the PMC6848994 WebFetch summary are not in the local cache: that file (PMID_31587868.md) is abstract-only, so quote only its abstract.

Sources

Cached references - PMID:31587868 · PMID:26358559 · PMID:38647244 · PMID:41917976 · PMID:42481913 - PMID:41828556 · PMID:36929417 · PMID:30705154 · PMID:19409814 · PMID:20610818 - PMID:21624937 · PMID:23481390 · PMID:30804042 · PMID:31387873 · PMID:27694884 - PMID:37322264 · PMID:24583034 · PMID:14550261

Web sources - Fiordaliso et al. 2019, PMC full text - MedGen C5393302 - Xu et al. 2024, PMC - Genomics England PanelApp – NKAP - CHOP news release, 2019 - ZFIN publication ZDB-PUB-191008-1

Reference Validation

Checked with linkml-reference-validator 0.3.0.

All extracted references resolved successfully.

Term Validation

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

Terms the report names something else

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

• NCIT:C15986 (2 mentions) - the report calls it "Antiseizure medication when epilepsy is present"; NCIT calls it Pharmacotherapy

Terms whose name is worth a second look

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

• NCBITaxon:10090 (1 mention) - the report calls it "Mouse, germline"; NCBITaxon calls it Mus musculus, and lists "mouse" among its other names