Reading the cached Nibbeling 2017 abstract and a 2024 review that mentions PLD3, then searching the literature on the dispute and on PLD3 biology.

Pulling verbatim abstracts for the key PMIDs so the report can quote them exactly.

Checking ClinGen's record for PLD3 and the original paper's family details, then I'll write the report.

Spinocerebellar ataxia 46 (SCA46; PLD3): research report

Compiled: 2026-10-08 · Target: MONDO:0033481 · Category: Mendelian (autosomal dominant)

Summary for curation

• The disease rests on one family. SCA46 is defined by a single Dutch kindred carrying the heterozygous PLD3 missense variant c.923T>C, p.Leu308Pro (Nibbeling et al. 2017, PMID:29053796). The same kindred had been described clinically in 1995 as a dominant sensory ataxic neuropathy (van Dijk et al. 1995, PMID:8595484). No second family with segregation has been published. The only other report is a single Indian patient with a different PLD3 variant and an atypical phenotype (PMID:38059248).
• The gene–disease link is provisional and disputed. OMIM lists the entry with a question mark (?SCA46, #617770). ClinGen has no published gene–disease validity curation for PLD3. GenCC aggregators show Limited and Supportive assertions. Gonzalez et al. (2018) reported that Pld3 knockout mice show no cerebellar degeneration or ataxia, and that PLD3 expression is lowest in the adult cerebellum (PMID:30312375). The original group replied defending the variant (PMID:30312384).
• PLD3 is a lysosomal enzyme, not a phospholipase in the classic sense. It is a lysosomal 5′ exonuclease that breaks down single-stranded DNA and RNA (PMID:30111894, 34620855, 33288674). Its main physiological substrate is mitochondrial DNA (PMID:37225734). It also synthesizes the lysosomal lipid S,S-BMP (PMID:39423811). The L308P protein is held in the ER, poorly processed, and has lost exonuclease activity (Gonzalez 2018 letter). How this produces cerebellar or sensory-neuron degeneration has not been shown.
• For dismech: the mechanism should be recorded as loss of function with an inferred downstream chain. The disputed gene–disease validity should be stated openly, for example in a discussions entry of kind KNOWLEDGE_GAP or HUMAN_MODEL_MISMATCH. There is no ClinGen tier to copy into gene_disease_validity.

1. Disease information

Overview. SCA46 is a rare autosomal dominant, slowly progressive, late-onset ataxia. It mixes cerebellar ataxia with sensory axonal neuropathy in varying proportions. Dysarthria and oculomotor abnormalities are common, and cerebellar atrophy on MRI is absent or mild.

• van Dijk 1995: "We describe a Dutch family with sensory ataxia in two generations" (PMID:8595484).
• Sowmini 2023: "SCA 46 is a rare autosomal dominant ataxia initially described in a Dutch family" (PMID:38059248).

Synonyms: SCA46; "spinocerebellar ataxia, 46, autosomal dominant, with sensory axonal neuropathy"; "spinocerebellar ataxia with sensory axonal neuropathy". PLD3 gene aliases include AD19, HU-K4 and HUK4 (GTR).

Data provenance: aggregated disease-level resources (OMIM, Orphanet, MedGen) built on a single published pedigree and one case report.

2. Etiology

• Cause: germline heterozygous PLD3 c.923T>C, p.Leu308Pro (chr19:40880431, hg19). The variant lies in the second phosphodiesterase (HKD) domain. SIFT, PolyPhen-2 and MutationTaster all predict it to be damaging, and it was absent from 1000 Genomes and ExAC. It segregated with disease: present in 8 affected relatives and absent in 3 unaffected relatives. One relative with mild complaints did not carry it. Screening 96 further index patients found no additional PLD3 variants (Nibbeling 2017, full text).
• Environmental risk factors: none reported.
• Protective factors: none reported.
• Gene–environment interactions: none reported.
• Related PLD3 associations (separate diseases; do not merge with SCA46):
• Late-onset Alzheimer's disease risk. Cruchaga 2014: "A rare variant in PLD3 (phospholipase D3; Val232Met) segregated with disease status in two independent families" and "doubled risk for Alzheimer's disease in seven independent case-control series" (PMID:24336208). Later replication was inconsistent.
• A homozygous nonsense variant, p.Y62X (c.186C>G), was reported in a consanguineous family with leukoencephalopathy, hearing and vision loss, and kidney disease. Liu 2021: "homozygous mutation of PLD3 may result in a novel leukoencephalopathy syndrome" (PMID:34267643).

3. Phenotypes

Main source: the Nibbeling 2017 family (Table 2), cross-checked against the MedGen/HPO annotation list. No penetrance or frequency figures beyond this single family exist.

Onset. Late adult onset. The mean age was 53.5 years, with a range of 35 to almost 70 (Nibbeling 2017). The 1995 report gives "late onset of symptoms (over the age of 40 years) and slow progression" (PMID:8595484).

Atypical case (PLD3 p.Ile26Thr, c.77T>C; zygosity and segregation not reported; PMID:38059248). - A 20-year-old woman with bilateral sensorineural hearing loss from age 13, then ataxia, dysarthria and early optic atrophy. - She had severe sensorimotor axonal polyneuropathy with distal weakness and diffuse cerebellar atrophy, and became wheelchair-bound. - Treat this as weak, unconfirmed evidence. Do not add hearing loss or optic atrophy as core SCA46 features without a notes caveat.

Quality of life. No published SARA, ICARS, EQ-5D or SF-36 data specific to SCA46.

4. Genetic and molecular information

• Gene: PLD3 (phospholipase D family member 3), 19q13.2, hgnc:17158, NCBI Gene 23646, UniProt Q8IV08.
• Pathogenic variant: NM_012268 c.923T>C, p.Leu308Pro. UniProt annotates it as of uncertain significance, with reduced lysosomal localization; induces retention in the ER; reduction of proteolytic cleavage; loss of exonuclease activity (UniProt). No ClinVar or ACMG classification was retrieved, and the gnomAD allele frequency was not retrieved.
• Origin: germline.
• Functional class: loss of function (LoF).
• Nibbeling 2017 (COS-7 lysates): "PLD3-L308P exhibited significantly reduced activity compared to PLD3-WT". The values were WT 2.2 ± 0.2 and L308P 1.5 ± 0.06, against a GFP mock of 1.7 ± 0.1. The L308P result is therefore close to background.
• Nibbeling 2017 found no difference from wild type in localization, stability or ER stress.
• Gonzalez 2018 (untagged protein): ER retention, impaired lysosomal delivery and processing, and loss of 5′ exonuclease activity (PMID:30312375).
• A dominant-negative mechanism has not been tested. Haploinsufficiency is the default assumption, but Pld3 heterozygous and homozygous mice show no ataxia.
• ClinGen / GenCC: ClinGen: "ClinGen has not yet published curations for PLD3". GenCC: SCA46 AD at Supportive (Orphanet) and Limited (genebe). Per CLAUDE.md, record only the tiers you copy directly from a GenCC or Orphanet source. Orphanet submits everything at Supportive.
• Modifier genes, epigenetics, chromosomal abnormalities: none reported for SCA46. PLD3 hippocampal methylation changes have been reported in Alzheimer's disease (PMID:30208929), which is not relevant to SCA46.
• Network context: Nibbeling 2017 placed PLD3 in the synaptic transmission cluster with FAT2, KIF26B, PDYN, FGF14, PRKCG and CACNA1A. The abstract states: "Our work links spinocerebellar ataxia to alterations in synaptic transmission and transcription regulation" (PMID:29053796; cached). This is a computational, guilt-by-association result.

5. Environmental information

No environmental, lifestyle or infectious contributors are reported. Not applicable.

6. Mechanism and pathophysiology

Ordered causal chain (★ = demonstrated; ◇ = inferred)

1. ★ A heterozygous germline PLD3 p.Leu308Pro variant in the luminal HKD catalytic domain leads to misfolding and retention of the mutant protein in the ER, with failed ESCRT-dependent delivery to lysosomes and failed proteolytic processing. Evidence is from overexpressing cells only (PMID:30312375; the trafficking route is described in PMID:29386126: "PLD3 reaches lysosomes as a type II transmembrane protein").
2. ★ This results in loss of lysosomal acid 5′ exonuclease activity from the mutant allele.
3. "PLD3 as the principal acid 5' exonuclease in HeLa cells" (PMID:33288674).
4. "they are 5' exonucleases, probably identical to spleen phosphodiesterase, that break down TLR9 ligands" (PMID:30111894).
5. In heterozygotes the expected residual activity is about 50%. Whether the mutant has a dominant-negative effect is untested ◇.
6. ◇ This leads to slower lysosomal breakdown of single-stranded nucleic acids, especially mitochondrial DNA.
7. "We identified mitochondrial DNA (mtDNA) as a major physiological substrate" (PMID:37225734).
8. "The deficiency of PLD3 leads to the slowed degradation of nucleic acids in lysosomes" (PMID:37994783).
9. Branch A ◇ (shown in PLD3-deficient neurons and cells, not in SCA46 patients): mtDNA leaks from lysosomes into the cytosol and activates cGAS–STING, which leads to increased autophagy, APP-CTF and cholesterol accumulation, and lysosomal stress.
10. "Lysosomal leakage of mtDNA to the cytosol activates cGAS-STING signaling" (PMID:37225734).
11. Endosomal TLR7/TLR9 sensing of unbroken nucleic acids is a parallel innate-immune route (PMID:34620855, 38697119). It is mainly shown in myeloid cells.
12. Branch B ◇: loss of PLD3 synthase activity reduces lysosomal S,S-BMP, which impairs lysosomal lipid and ganglioside breakdown.
13. "Deletion of PLD3 or PLD4 markedly reduced BMP levels in cells or in murine tissues"; disease-associated PLD3 mutants have reduced activity (PMID:39423811).
14. Whether L308P was tested specifically was not confirmed from the abstract.
15. ◇ Lysosomal dysfunction in long-lived neurons (cerebellar Purkinje cells and/or large-fibre dorsal root ganglion sensory neurons) leads to slow neurodegeneration and axonal degeneration. This step has not been demonstrated for PLD3. Indirect support: Zfp212 knockout mice show Purkinje cell death accompanied by Pld3 downregulation, and restoring ZNF212 rescues it. "Pld3 alone was tightly regulated by Flag-tagged ZNF212 overexpression" (PMID:34815492).
16. ★ Clinically, sensory axonal neuropathy (large-fibre loss) plus mild cerebellar dysfunction results in mixed sensory–cerebellar gait ataxia, dysarthria and oculomotor signs (PMID:8595484, 29053796).

Counter-evidence to attach to steps 6 and 7

• Pld3 knockout mice: "These data do not reveal a major degenerative phenotype of the cerebellum in PLD3 KO mice."
• PLD3 expression is "lowest… in the adult cerebellum compared to all other brain regions" (PMID:30312375).
• Encode this as HUMAN_MODEL_MISMATCH, not as a refutation of the variant. Mice may need a longer lifespan, a second hit, or compensation by PLD4.

Suggested ontology anchors (leads only; resolve each CURIE before binding)

• GO biological processes: lysosomal DNA catabolic process / DNA catabolic process; 5′-3′ exonuclease activity (molecular function); lysosome organization; cGAS/STING signaling (positive regulation of type I interferon production); protein retention in ER lumen; lysosomal transport.
• GO cellular components: lysosomal lumen and lysosomal membrane; endoplasmic reticulum (mutant retention); multivesicular body.
• CL cell types: Purkinje cell; sensory neuron of dorsal root ganglion (large-fibre / proprioceptive); neuron.
• Biological scale: step 1 MOLECULAR; steps 2–5 CELLULAR; step 6 TISSUE; step 7 ORGANISM.

Molecular profiling and advanced technologies

• No transcriptomic, proteomic, metabolomic or single-cell data exist for SCA46.
• Structures of PLD3 are available: a crystal structure (PMID:37994783, 39325669, 39517096) and cryo-EM structures with bound substrate (PMID:41381514, 38537643). These could support modelling of L308P, but none of the abstracts retrieved mention L308P.

7. Anatomical structures affected

• Primary: cerebellum (mild or absent atrophy); peripheral sensory nerves (sural nerve axonal loss); dorsal root ganglia and dorsal columns (inferred from large-fibre sensory loss).
• Secondary: oculomotor control pathways in the cerebellum and brainstem.
• Subcellular: lysosome (normal site of action); ER (where the mutant is retained).
• Laterality: bilateral and symmetric, distal-predominant neuropathy.
• Suggested UBERON terms (verify): cerebellum, sural nerve, dorsal root ganglion, peripheral nerve.

8. Temporal development

• Onset: adult or late adult, mean 53.5 years (range 35 to about 70). Insidious.
• Course: slowly progressive and lifelong.
• Stages, remission, critical windows: not described. There is no natural-history study.

9. Inheritance and population

• Inheritance: autosomal dominant (heterozygous). Penetrance appears high in the family: 8 of 8 carriers were affected. One mildly symptomatic relative was a non-carrier, which could be a phenocopy. Expressivity is variable between sensory-predominant and cerebellar-predominant forms. No anticipation, since this is not a repeat expansion.
• Prevalence: unknown. Only one family is confirmed (CASES_IN_LITERATURE). Orphanet's prevalence class was not retrieved because the page was blocked; check just structured-rebuild-orphanet --id 589522.
• Population: Dutch, with a possible additional Indian case. There is no founder data, no sex bias (both generations were affected), and no consanguinity in the dominant family.

10. Diagnostics

• Clinical work-up:
• Nerve conduction studies show a sensory axonal neuropathy.
• Sural nerve biopsy shows loss of myelinated fibres. 1995 report: "Clinical, electrophysiological and sural nerve biopsy findings revealed a sensory polyneuropathy".
• Brain MRI shows normal or mildly atrophic cerebellum.
• Eye-movement examination.
• Genetic testing:
• Ataxia or neuropathy NGS panels, or exome or genome sequencing (exome found the original variant).
• GTR lists PLD3 sequence analysis (4 clinical tests), deletion/duplication analysis (4) and targeted variant analysis (1).
• First exclude the common repeat-expansion SCAs (SCA1/2/3/6/7/17, DRPLA), RFC1 CANVAS and FGF14-GAA (SCA27B). These are relevant because the sensory-plus-cerebellar picture overlaps heavily with CANVAS.
• Biomarkers: none.
• Differential diagnosis:
• RFC1-related CANVAS, which has sensory neuronopathy and vestibular areflexia.
• Friedreich ataxia, which has earlier onset, GAA expansion and cardiomyopathy.
• SCA4, which also features sensory axonal neuropathy.
• SANDO / POLG disorders.
• Vitamin E deficiency.
• Acquired sensory ganglionopathies.
• Screening: cascade testing of relatives once a familial variant is known.

11. Outcome and prognosis

• No mortality or survival data. Slow progression suggests a near-normal lifespan, but this is not documented.
• Expected disability: impaired gait, possible need for walking aids, and falls. The atypical young patient became wheelchair-bound (PMID:38059248).
• No prognostic biomarkers.

12. Treatment

There is no disease-modifying therapy and no clinical trials (ClinicalTrials.gov not searched in depth). Management is supportive.

• Pharmacogenomics: not applicable.
• Experimental options: none specific. Small-molecule modulators of PLD3/PLD4 exist as research tools (PMID:34332037, 38560090). STING inhibition normalized APP-CTF in PLD3-deficient models: "STING inhibition largely normalizes APP-CTF levels" (PMID:37225734). This is preclinical and not related to ataxia.

13. Prevention

• Primary prevention: not applicable.
• Reproductive options: genetic counseling, with prenatal or preimplantation testing available once the familial variant is known (50% recurrence risk).
• Tertiary prevention: fall prevention and foot care for the sensory neuropathy.

14. Other species and natural disease

• No naturally occurring PLD3 ataxia is listed in OMIA (not checked exhaustively).
• Orthologs: mouse Pld3 and zebrafish pld3 (IDs not retrieved).

15. Model organisms

Main gap: there is no L308P knock-in model and no patient iPSC-derived Purkinje cells or sensory neurons.

Evidence quick-reference (verbatim abstract fragments)

Caveats on this table: - Except for PMID:29053796, the fragments came from a web-fetch tool and not from references_cache. Run just fetch-reference PMID:<id> and re-copy each snippet from the cache before using it as evidence. - The Gonzalez (30312375) and Ma & Verbeek (30312384) letters have no PubMed abstract. Snippets from them need full-text access. - Full-text quotes from Nibbeling 2017 (onset 53.5 years, segregation in 8 affected, the activity values) are not in the abstract-only cache.

Open issues to record

1. Gene–disease validity is unresolved. There is one family, the OMIM entry is provisional, ClinGen has no curation, and the knockout mouse is negative. Do not set relationship_type: CAUSATIVE without a note; consider whether the evidence supports it.
2. Haploinsufficiency versus dominant-negative effect is untested.
3. The cell type is unknown. It is not established whether the primary lesion is in Purkinje cells, sensory ganglion neurons, or both. The mixed sensory–cerebellar phenotype is consistent with both.
4. The Indian p.Ile26Thr case needs a segregation and zygosity check before it is counted as SCA46.

Sources

• Nibbeling et al. 2017, Brain — PMID:29053796; full text
• van Dijk et al. 1995, Brain — PMID:8595484
• Gonzalez et al. 2018, Brain letter — PMC6202572 (PMID:30312375)
• Ma & Verbeek 2018 reply — Brain e79 (PMID:30312384)
• Sowmini et al. 2023 — J Neurosci Rural Pract (PMID:38059248)
• Gavin et al. 2018, Nat Immunol — PMC6105523 (PMID:30111894)
• Gavin et al. 2021, Nat Commun — PMC8497607 (PMID:34620855)
• Van Acker et al. 2023, Nat Commun — PMC10209153 (PMID:37225734)
• Gonzalez et al. 2018 Cell Rep (PMID:29386126); Cappel et al. 2021 JBC (PMID:33288674); Singh et al. 2024 Cell (PMID:39423811); Yuan et al. 2022 Nature (PMID:36450991); Khang et al. 2021 Sci Rep (PMID:34815492); Liu et al. 2021 Front Aging Neurosci (PMID:34267643); Cruchaga et al. 2014 Nature (PMID:24336208); PLD3 structures (PMID:37994783, 39517096, 41381514)
• MedGen 1624251 · GTR C4540404 · ClinGen PLD3 · UniProt Q8IV08 · LOVD SCA46 · genebe PLD3 (GenCC) · MalaCards · Frontiers Liu 2021

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:

• DOID:0080288 (3 mentions) - the report calls it "DOID"; DOID calls it spinocerebellar ataxia 46

Prefixes with no resolver

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