PRNP P105L: research

What phenotype has been described?

Itoh and colleagues (Journal of the Neurological Sciences 1994) proposed a five-part phenotype from a clinicopathological study of one man together with the pedigrees known then: onset at age 38 to 48, duration of 7 to 11 years, prominent spastic paraparesis with dementia and ataxia, numerous cortical amyloid plaques, and amorphous PrP deposits with deep cortical layer neuronal loss and only minor cerebellar change. Their own patient began at 42 with right hand clumsiness and died at 53. Myoclonus and periodic synchronous discharges were absent.

Later reports show the phenotype is not uniform within a family. Ishizawa and colleagues (Brain and Behavior 2018) autopsied three relatives. Two sisters had onset at 38 and 44 with spastic paraparesis, with later dysarthria, emotional disturbance and dementia, and durations of 6 and 12 years. The son of the second sister had onset at 47 with stereotyped behaviour and speech rather than spastic paraparesis, then clumsy walking, tremor and myoclonus, with a duration of about 2 years.

Kawai and colleagues (Prion 2023) describe a woman whose symptoms began around 41 with progressive spastic gait, depression and emotional lability, with cerebellar ataxia, gaze-evoked nystagmus, spastic paraparesis and extrapyramidal signs on examination, and no myoclonus. She was bedridden and without meaningful speech three years later. Her father had similar symptoms at 45 and died at 62.

What does the neuropathology show?

Numerous PrP plaques in cerebral cortex are the consistent finding. Ishizawa and colleagues describe compact, well defined plaques in superficial and middle cortical layers and blurry, ill defined, confluent deposits in deep layers, with plaques also in basal ganglia, hippocampus and thalamus. Cerebellar plaques were abundant in one of the three cases and absent in the other two. Severe pyramidal tract myelin loss was present in the two cases with spastic paraparesis and absent in the third, whose cortex was least degenerated.

Phosphorylated tau was present in all three, in and around PrP plaques, as neurofibrillary tangles, dystrophic neurites and neuropil threads, with both 3R and 4R tau associated with plaques. Amyloid-beta was absent in two cases. In the third it was deposited in cortex, mostly colocalised with PrP plaques and often with a PrP-positive core. The authors found PrP without amyloid-beta in some areas and never amyloid-beta without PrP, from which they suggest PrP deposition precedes amyloid-beta deposition. That inference rests on three autopsies in one family.

Kawai and colleagues state, citing prior reports, that spongiform change is absent in P105L, and contrast P105L with P102L, where diffusion-weighted MRI hyperintensity, a mosaic-like pattern of reduced cerebral blood flow on SPECT and spongiform change are more often seen.

Which codon 129 allele carries the variant?

In the three relatives reported by Ishizawa and colleagues, P105L sat on the valine 129 allele, with a codon 129 Met/Val genotype. The 1994 patient was also codon 129 Met/Val. The patient reported by Kawai and colleagues was Met/Val at codon 129 and glutamic acid homozygous at codon 219. Whether the cis allele determines phenotype in P105L, as it does for D178N, has not been tested in a series of sufficient size.

Do CSF and imaging tests detect it?

In the case reported by Kawai and colleagues, CSF 14-3-3 and total tau were not elevated and RT-QuIC for PrPSc was negative. EEG showed no periodic sharp wave complexes. MRI at 11 and 52 months showed bilateral frontal-dominant atrophy with no diffusion-weighted hyperintensity at either point. SPECT showed non-specific mild hypoperfusion at 11 months and a marked reduction in regional cerebral blood flow in the bilateral, right-dominant frontal lobes at 52 months, more extensive than the atrophy alone would explain. The authors attribute the perfusion deficit to cortical PrPSc plaque load and note that SPECT data in this variant are limited by the very small number of patients studied.

The 302-case, 11-centre biomarker cohort of Schmitz and colleagues (Brain 2022) covered 36 PRNP variants and reports a P105T group of 3 cases, with no P105L group. Readers should not transfer that paper’s P102L-specific CSF cut-offs to P105L.

How common is it?

Japanese national surveillance over ten years (Nozaki and colleagues, Brain 2010) classified 1,222 patients as having prion disease, of whom 216, or 17.7 percent, had genetic prion disease. The four most frequent variants in that group were V180I at 41.2 percent, P102L at 18.1 percent, E200K at 17.1 percent and M232R at 15.3 percent, leaving about 8.3 percent for all remaining variants, P105L among them. The paper’s abstract does not break that remainder down.

Itoh and colleagues record the variant in four Japanese pedigrees as of 1994. Ishizawa and colleagues cite further Japanese reports, including a second family with two affected siblings and families presenting with atypical parkinsonism, without giving a worldwide total. A reliable count of reported P105L families has not been published in the sources opened here.

Related pages: protective variants, penetrance, the variants hub and the plain-language P105L explained.