Creutzfeldt-Jakob Disease, Prion Protein Gene Codon 129VV, and a Novel PrPSc Type in a Young British Woman FREE

The original recognition of variant Creutzfeldt-Jakob disease (vCJD) was based on a case series of young patients with rapidly progressive dementia, a geographic and temporal association with bovine spongiform encephalopathy (BSE), and novel neuropathological findings consisting of abundant florid prion protein (PrP) plaques.¹ Molecular strain typing allowed identification of a unique type of PrP^Sc (the scrapie isoform of PrP) (type 4) in the brain that was distinct from those seen in classic (sporadic or iatrogenic) CJD and similar to that seen in BSE prion infection of cattle and other species.² Subsequent biological strain typing in both conventional and transgenic mice confirmed that vCJD and BSE were caused by the same prion strain.^{3 - 4}

Variant CJD also differs markedly from classic CJD in having prominent and consistent involvement of lymphoreticular tissue, allowing its diagnosis by tonsil biopsy findings.^{5 - 7} To date, more than 160 individuals have died of vCJD in the United Kingdom; the number infected by BSE prions and who may develop prion disease in the years ahead is unknown because human prion incubation periods may exceed 50 years.⁸

All clinical cases of vCJD studied have had a methionine-homozygous (MM) genotype at polymorphic codon 129 of the prion protein gene (PRNP).⁹ The extension of BSE prion-related disease to individuals with valine-homozygous (VV) or heterozygous (MV) genotypes at PRNP codon 129 has been predicted by comparison with other acquired human prion diseases^{10 - 11} and by transgenic mouse models.^{12 - 14} These models also predict that infection of VV and MV genotypes with BSE or vCJD prions may result in propagation of distinct prion strain types and that patients with VV or MV genotypes might present with clinical, pathological, and molecular phenotypes distinct from that of vCJD.^{12 - 14}

To date, we know of no reported cases of clinical vCJD occurring in the VV or MV genotypes. However, PrP^Sc has been reported in lymphoid tissues, but not in the brain, of a patient with PRNP 129 MV who had received blood from a person with preclinical vCJD and who died of an unrelated cause.¹⁵ In addition, abnormal PrP immunoreactivity has been reported in anonymous archived lymphoid tissue from 2 individuals with PRNP 129 VV.¹⁶ It is unknown whether the individual with the MV genotype would have gone on (or if those with VV will go on) to develop clinical disease and, if so, whether the phenotype will fit the case definition of vCJD.

Brain homogenates (10% w/v) were prepared in Dulbecco phosphate buffered saline lacking Ca²⁺ or Mg²⁺ ions. Aliquots were analyzed with or without proteinase K digestion (50 μg/mL final protease concentration, 1 hour, 37°C) by immunoblotting with anti–PrP monoclonal antibody 3F4¹⁷ as described previously.^{7 ,18} Metal ion–dependent conformations of PrP were determined as previously described.¹⁹ Genomic DNA was extracted from peripheral blood, and the entire PRNP open reading frame was amplified by polymerase chain reaction and sequenced as described previously.²⁰

A 39-year-old woman presented to an optician in January 1999 with episodes of blurred vision and photophobia, but no abnormality was found. Two months later, she noted memory impairment, diplopia, dysarthria, and an unsteady gait of fluctuating severity. Five months after onset, the gait and limb ataxia had progressed, although walking was still possible, and the memory loss became more profound. The patient then developed paranoid ideation, aggression, restless nocturnal behavior, anorexia, and mood disturbance. By 5½ months after onset, she could not walk and was unsteady sitting, and limb movements were clumsy.

Examination showed dysarthria, broken pursuit eye movements without nystagmus, impaired upgaze, and stereotyped involuntary movements of the legs. However, limb power, vibration, proprioception, tendon reflexes, and plantar responses were normal. During the ensuing 4 weeks, speech ceased and incontinence and jerky involuntary limb movements became evident. Eight months after onset, the patient was mute but could follow some commands. She was able to visually fixate and follow moving objects but also had abnormal, spontaneous horizontal roving eye movements with a supranuclear vertical gaze palsy. Her face was impassive with occasional twitching movements, brisk facial reflexes, and trismus. There were prominent jerking movements of all limbs brought out by use; power was relatively preserved and the plantar responses were extensor.

There was a strong family history of late-onset cerebellar ataxia consistent with autosomal dominant inheritance. A polyglutamine expansion in ataxin 3 associated with spinocerebellar ataxia type 3 was found in a symptomatic family member, but our patient did not share this expansion.

Normal results of the following investigations were found: complete blood cell count, erythrocyte sedimentation rate, C-reactive protein, electrolytes, liver function, thyroid function, enzyme-linked immunosorbent assay for syphilis, vitamin B₁₂, folate, ferritin, vitamin E, and serum ceruloplasmin. Tests for antinuclear, antineuronal, anti–Purkinje cell, and antiganglioside antibodies were negative. Nerve conduction studies showed no evidence of a peripheral neuropathy. The electroencephalogram 6 months after onset was reported as normal, but at 7 and 8 months electroencephalograms showed diffuse slow-wave activity, without epileptiform changes or periodic discharges typical of CJD. Cerebrospinal fluid examination showed a normal cell count, protein level, and glucose level, and oligoclonal immunoglobulin bands were absent. The protein S100b level of 4.39 ng/mL (reference cutoff, < 0.38 ng/mL), neuron-specific enolase level of 98 ng/mL (reference cutoff, < 20 ng/mL), and 14-3-3 protein were all abnormal values.

A magnetic resonance image of the brain (Figure 1) showed diffuse cerebellar atrophy and diffuse increased signal within both caudate nuclei and each putamen. Tonsil biopsy was not possible because of a previous tonsillectomy from which little tissue remained. Genetic testing for mutations associated with spinocerebellar ataxia 1, 2, 3, 6, and 7 and Friedreich ataxia gave negative results. Sequencing of the PRNP open reading frame was normal on 2 separate occasions. A polymerase chain reaction performed with primers designed to amplify the octapeptide repeat region of PRNP did not demonstrate an insertion mutation. The codon 129 polymorphism was homozygous for valine.

Fourteen months after onset, the patient died and an autopsy was performed.

Histopathologic examination was limited to the brain and spinal cord (Figure 2). The findings were atypical of sporadic CJD in the severity of white matter degeneration and the extensive nature of PrP deposition in the cortex and white matter. The frontal cortex showed extremely severe neuronal loss with striking astrocytosis and prominent spongiform vacuolation. There was severe overall loss of white matter, in places reminiscent of infarction. Deposition of PrP was extensive throughout the cortex and white matter. In places this was a diffuse punctate deposition similar to the recognized synaptic pattern. Occasionally, individual cells, mainly pyramidal neurons, were outlined by PrP deposition and had a fine granular intracellular deposition. More dense deposits, similar to plaques, were seen in the cortex. Also in the white matter, PrP deposits were seen ranging from a couple of micrometers to much larger plaquelike deposits, although these were not florid.

Temporal, parietal, and occipital lobes showed histologic features similar to those described in the frontal lobe, the occipital lobe being most severe. The hippocampus was relatively well preserved. In the caudate, putamen, and amygdala there was neuronal loss, astrocytosis, and microglial activation. The thalamus, midbrain, and substantia nigra showed mild to moderate spongiform change, neuronal loss, and astrocytosis with intraneuronal and extracellular punctate deposits. The pons and medulla were less severely affected than the midbrain with punctate PrP deposits. The cerebral peduncles were severely affected, with nearly complete loss of myelin. The cerebellum was very severely affected, with a dramatic loss of Purkinje and granule cells accompanied by vacuolation and astrocytosis. The cerebellar white matter showed severe white matter loss similar to incipient infarcts. Deposition of PrP in the cerebellum was marked with accumulation of punctate deposits resembling plaques, most commonly in the granule cell layer. In the white matter the deposits were denser still, occasionally plaquelike or forming irregular linear deposits.

Western blot analysis was performed on fresh frozen cerebellar tissue from the patient. Identical results were obtained from separately analyzed tissue samples from opposite poles of the cerebellum. The glycoform ratio and fragment sizes resembled PrP^Sc type 4 seen in vCJD (Figure 3). The nonglycosylated band was seen as a doublet, as is seen for PrP^Sc in the cerebellum in vCJD (Figure 4). The effect of adding the metal ion chelator EDTA to the cerebellum homogenate before proteinase K cleavage was to reduce the apparent molecular weight of PrP^Sc fragments. This reflects the involvement of metal ions (most likely copper and zinc) in the conformation of PrP and determination of accessible protease cleavage sites.¹⁹ This deduction was verified by showing that application of zinc ions to EDTA-treated samples before proteolysis resulted in preservation of the original PrP^Sc fragment size (Figure 4C). Although similar dependence on metal ions is observed for some PrP^Sc conformers associated with sporadic CJD,^{19 ,21} this is not observed with PrP^Sc type 4 propagated in vCJD^{19 ,21} (Figure 4). Therefore, these findings reflect a novel PrP^Sc type when compared with the diversity we and others have so far documented.^{21 - 22}

Does the PrP^Sc typing suggest a BSE-related cause, or can our findings be accommodated by the spectrum seen in sporadic CJD cases worldwide? The molecular strain typing of the patient's brain material demonstrated a novel PrP^Sc type when compared with our archived cases.²¹ There is as yet no internationally agreed-on classification of PrP^Sc type. Parchi and colleagues²³ identified 2 PrP^Sc types in sporadic CJD. However, Hill et al²¹ described 3 PrP^Sc types associated with sporadic and iatrogenic CJD (types 1-3) and PrP^Sc type 4 associated with vCJD. The PrP^Sc type 5 has, to our knowledge, been observed only in mice expressing human PrP 129V inoculated with vCJD.^{3 ,12} Hill et al²¹ recently described a novel PrP^Sc type 6 in sporadic CJD.

The PrP^Sc type from our case has features similar to PrP^Sc type 4 (vCJD) in the predominance of the diglycosylated band; however, it is distinct from PrP^Sc type 4 in the dependence of the protease cleavage pattern of PrP^Sc on metal ions, suggesting a distinct PrP^Sc conformation. Unfortunately, only cerebellum was available for Western blotting in this case, although in vCJD cases from which whole brain was available we have not found evidence of any regional variation in PrP^Sc type. Others have reported coexistence of Gambetti PrP^Sc type 1 in the brain from patients with vCJD as a minority component.²⁴ It would also have been interesting to look for peripheral lymphoreticular PrP deposition because this is prominent in vCJD, but that tissue was not available for analysis. Transmission of BSE isolates to transgenic mice expressing human PrP 129 valine results in clinical prion disease with undetectable PrP^Sc; however, transmission of vCJD isolates to the same mice produces PrP^Sc type 5 that shares the same predominance of diglycosylated PrP^Sc to that of PrP^Sc type 4, and these data suggest that the molecular signature of BSE may be preserved after BSE transmission to PRNP codon 129 VV humans.^{3 ,12} Transmission studies of the current case in transgenic mice are now being undertaken to investigate transmission characteristics.

We have described a novel PrP^Sc type that would be designated type 7 by our classification. A firm connection between novel PrP^Sc types and BSE cannot be made on the basis of a single case, and it will be important to see whether other similar cases occur in the United Kingdom and other BSE-exposed countries but not elsewhere and to perform detailed transmission studies of prions from this patient into transgenic and conventional mice to compare with BSE-derived isolates from cattle and other species. Two other cases of prion disease with valine homozygosity and atypical features have been reported in the United Kingdom and the Netherlands. One of these cases was atypical because of very young onset and a protracted psychiatric history²⁵ ; the other was notable because certain clinical and molecular features of the case overlapped with those of vCJD, including Western blot analysis of autopsied brain showing a predominance of a diglycosylated PrP^Sc isoform.²⁶

We recommend keeping an open mind about the etiology of such cases during the ensuing years. These cases emphasize the importance both of continued surveillance of prion disease and the further development and refinement of molecular classification of prion diseases of humans and animals. It will also be important to assess lymphoreticular involvement in subsequent cases either at diagnostic tonsil biopsy or at autopsy.

Correspondence: John Collinge, FRS, MRC Prion Unit and Department of Neurodegenerative Disease, Institute of Neurology, University College London, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, England (j.collinge@prion.ucl.ac.uk).

Accepted for Publication: February 22, 2006.

Author Contributions:Study concept and design: Mead and Collinge. Acquisition of data: Mead, Joiner, Desbruslais, O’Donoghue, Lantos, Wadsworth, and Collinge. Analysis and interpretation of data: Mead, Joiner, Desbruslais, Beck, Wadsworth, and Collinge. Drafting of the manuscript: Mead, Desbruslais, Beck, Lantos, Wadsworth, and Collinge. Critical revision of the manuscript for important intellectual content: Mead, Joiner, O’Donoghue, Wadsworth, and Collinge. Statistical analysis: Mead. Obtained funding: Collinge. Administrative, technical, and material support: Desbruslais, Beck, Lantos, and Collinge. Study supervision: Wadsworth and Collinge.

Funding/Support: This work was funded by the MRC and undertaken at the University College London Hospitals and University College London, who received a proportion of funding from the Department of Health's National Institute for Health Research Biomedical Research Centres funding scheme.

Additional Contributions: Ray Young assisted with figure design. The MRC London Neurodegenerative Diseases Brain Bank (Institute of Psychiatry) provided pathological material. We also acknowledge the many clinicians involved in the care of this patient.

Financial Disclosure: None reported.