Gain-of-Function Variants in PLAAT4 and Spinal Predominant Demyelinating Leukodystrophy

Leukoencephalopathies are clinically and etiologically heterogeneous disorders, and establishing a specific diagnosis remains challenging.

Patients with unexplained spastic paraplegia and brain white matter hyperintensities (WMH) may be misdiagnosed as having multiple sclerosis, potentially exposing them to unnecessary immunosuppressive therapy. To define the clinical, radiological, and genetic features of spinal predominant demyelinating leukodystrophy with leg spasticity (SPDMLS).

This case series involved participants from 4 tertiary referral hospitals within the China Alliance for Rare Diseases during the period July 2019 to August 2025. The study design integrated clinical evaluations and whole-exome sequencing (WES), as well as targeted plasma lipidomics correlated with SPATAX-EUROSPA disability stages.

Variant pathogenicity was evaluated using an oligodendrocyte-specific knock-in mouse model. Participants were patients with unexplained spastic paraplegia and central nervous system (CNS) WMH and available family members, including presymptomatic carriers.

Identification of disease-associated gene and characterization of clinical, radiological, and genetic features of SPDMLS. Identification of disease-associated gene and characterization of clinical, radiological, and genetic features of SPDMLS.

Twenty-six individuals from 11 unrelated families were included; 4 patients underwent longitudinal follow-up for up to 5 years. Recurrent heterozygous PLAAT4 (phospholipase A and acyltransferase 4) variants (p.I14T and p.A96V) were associated with a novel autosomal dominant leukoencephalopathy, named SPDMLS.

Age at onset was 11 to 51 years (median [IQR], 22 [15-29] years), with patients predominantly developing spastic paraplegia. Symbol Digit Modalities Test scores ranged from 11 to 70 in 6 available patients.

Magnetic resonance imaging (MRI) demonstrated slow progression with heterogeneous brain WMH and a stereotyped pattern of symmetric, long-segment "girdle-like" spinal cord WMH accompanied by atrophy. Somatosensory- and motor-evoked potentials indicated dysfunction of the dorsal columns and corticospinal tracts, respectively.

In presymptomatic carriers, spinal cord signal alterations preceded brain involvement. An oligodendrocyte-specific knock-in mouse model provided in vivo functional evidence supporting a role of PLAAT4 gain-of-function in driving demyelination, informing SPDMLS pathogenesis.

Patient plasma lipidomics suggested accumulation of arachidonic acid-related lipids correlating with disability stage.

This study found that PLAAT4 variants delineated SPDMLS, a clinically distinct leukodystrophy that should be considered in patients with unexplained spastic paraplegia and CNS WMH.

Arachidonic acid-related lipids represent a candidate biomarker for disease progression.