Comprehensive Neuropathologic Analysis of Genetic Prion Disease Associated With the E196K Mutation in PRNP Reveals Phenotypic Heterogeneity

Transcription

1 J Neuropathol Exp Neurol Copyright Ó 2011 by the American Association of Neuropathologists, Inc. Vol. 70, No. 3 March 2011 pp. 192Y200 ORIGINAL ARTICLE Comprehensive Neuropathologic Analysis of Genetic Prion Disease Associated With the E196K Mutation in PRNP Reveals Phenotypic Heterogeneity Sabina Eigenbrod, MD, Petra Frick, PhD, Armin Giese, MD, Gabi Schelzke, MD, Inga Zerr, MD, and Hans A. Kretzschmar, MD, FRCPath Abstract The genetic forms of human transmissible spongiform encephalopathies (TSEs) are linked to mutations in the gene encoding the prion protein (PRNP) and account for 10% to 15% of human TSE cases. Some are distinct with respect to clinical signs, disease onset/ duration, and diagnostic findings, whereas others closely resemble sporadic Creutzfeldt-Jakob disease (scjd). We report a comprehensive analysis of 4 patients carrying the rare E196K (GAGYAAG) mutation who presented with clinical features of CJD. To date, information on this PRNP mutation is limited to clinical and genetic data. Consequently, the E196K mutation could not be unequivocally assigned to human prion disease. We report histopathologic and biochemical findings in addition to clinical observations, thus providing a more comprehensive analysis of this presumably genetic prion disease. Our data indicate that (i) the E196K mutation is causally linked to human prion disease, (ii) there is a complex phenotypic spectrum of this mutation that includes nonspecific symptoms at onset and features typical of scjd during disease progression, and (iii) the corresponding histologic picture comprises both cases with atypical neuropathology and cases that closely resemble subtypes of scjd corresponding to the classification of Parchi et al, with subtle modifications in hippocampal regions CA1-4. From the Center for Neuropathology and Prion Research (SE, PF, AG, HK), Ludwig Maximilians University Munich, Munich, Germany; and National TSE Reference Center (GS, IZ), Department of Neurology, Georg August University Goettingen, Goettingen, Germany. Send correspondence and reprint requests to: Hans A. Kretzschmar, MD, FRCPath, Center for Neuropathology and Prion Research, Ludwig Maximilians University Munich, Feodor-Lynen-Str. 23, Munich, Germany; hans.kretzschmar@med.uni-muenchen.de This work was supported by the Robert Koch Institute through funds of the Federal Ministry of Health (Bundesgesundheitsministerium, Grant No ). Key Words: Creutzfeldt-Jakob disease, E196K, Prion, PRNP, PrP Sc, Transmissible spongiform encephalopathy. Transmissible spongiform encephalopathies (TSEs), or prion diseases, are a group of fatal neurodegenerative disorders that are caused by a conformational conversion of the normal cellular form of the prion protein (PrP C ) to the misfolded isoform, designated PrP Sc (1Y3). Human prion diseases develop spontaneously, arise through genetic mutation (genetic/familial TSEs), or can be acquired through transmission from one host to another (acquired TSEs). Approximately 10% to 15% of human TSE cases are inherited and have been shown to segregate with mutations in the coding region of PRNP. To date, more than 30 PRNP mutations have been described. A number of these mutations are associated with distinct clinicopathologic phenotypes, including genetic Creutzfeldt-Jakob disease (CJD), Gerstmann-Sträussler- Scheinker syndrome, and fatal familial insomnia (4Y6). The phenotypic heterogeneity of familial prion diseases is not solely dependent on the PRNP mutation but seems to be influenced by the polymorphic codon 129 genotype of the mutant and wild-type alleles and the haplotype background of the mutation. For example, the presence of methionine at the polymorphic codon 129 in coupling phase with asparagine at codon 178 typically results in the distinct clinical and pathologic features of fatal familial insomnia, whereas the same mutation in combination with valine at codon 129 leads to genetic CJD (7). Mutated PrP seems to be associated with a number of different pathogenic conformers that may differ in toxicity or neuropathologic targeting (8, 9). The product of the wild-type PRNP allele may also variably contribute to the disease process (10). It has been shown that polymorphism of codon 129 can influence the physicochemical properties of proteaseresistant PrP Sc (ie, the size of the protease-resistant fragment and the ratio of the different glycoforms) extracted from the brains of affected patients. In sporadic CJD (scjd) patients, 2 major human PrP Sc types (termed type 1 and type 2) that differ in electrophoretic mobility and the unglycosylated core fragments generated by treatment with proteinase K (PK) have been described (11Y14). Among the PRNP mutations identified to date, the point mutation E196K has been described in 4 reports (15Y18). The patients affected were either homozygous for methionine (15) or valine (16) at codon 129 in PRNP. Only 1 case heterozygous (M/V) at codon 129 was reported to the European Creutzfeldt-Jakob disease Surveillance System (19). The pathogenic significance of the E196K mutation with respect to a causal association with symptomatic disease, although very likely, has not been established. To date, all reports on this PRNP mutation have focused on the clinical findings and have not addressed the pathologic or biochemical attributes of this rare PRNP mutation. 192 J Neuropathol Exp Neurol Volume 70, Number 3, March 2011

2 J Neuropathol Exp Neurol Volume 70, Number 3, March 2011 E196K PRNP Mutation Neuropathology We performed a comprehensive study of 4 patients carrying the PRNP mutation E196K and report for the first time the histopathologic and biochemical features associated with the E196K mutation. MATERIALS AND METHODS Selection of Cases Four patients carrying PRNP mutation E196K had been identified as suspected CJD cases to the German CJD Surveillance Unit based on current clinical criteria for CJD (20Y22); the diagnoses were verified at autopsy (Table). Clinical data include results of physical examination, electroencephalography (EEG), magnetic resonance imaging (MRI), and evaluation of biomarkers in the cerebrospinal fluid (CSF) such as protein , Tau, neuron-specific enolase (NSE), and glial protein S100b (23Y26). The investigations were performed with informed consent and following all the institutional guidelines for experimental investigations with human specimens. Mutation Analysis and Genotyping Genomic DNA was extracted from blood samples and the coding region of PRNP was amplified by polymerase chain reaction. Subsequently, the complete open reading frame of PRNP was screened for mutations by direct sequencing using the automated LICOR 4200 DNA Analyzer/ Genetic Analyzer 3130 (Applied Biosystems, Lincoln, NE), as previously described (27, 28). Briefly, the amino acid at codon 129 in coupling with the mutation was determined by cloning. The polymerase chain reaction product was cleaved with EcoRI and XmaI. Subsequently, the large fragment was purified and cloned into the appropriate cleavage sites of pbluescriptiiks(j) (Stratagene, La Jolla, CA). The mutation and the respective codon 129 were verified by sequencing of 3 independent clones. Histopathologic and Immunohistochemical Analyses Autopsy and sampling of material for histopathologic analyses of various brain regions were performed as previously described (1). In addition, frozen samples of the frontal cortex, hippocampus, and cerebellum were preserved for biochemical analysis. Control brain tissue was obtained from corresponding brain regions of scjd patients with the MM 1, VV 2, and MV 2 subtypes. Brain material for histopathologic analysis was fixed in 4% buffered formalin, inactivated with 100% formic acid for 1 hour, fixed in formalin TABLE. Clinical Features of E196K Patients Patient 1 Patient 2 Patient 3 Patient 4 Codon 129 MM VV MV MM PrP Sc type Sex F M F F Age of onset, yr Disease duration, mo Initial symptoms Gait instability, confusion, Dementia Severe headache, Tremor impaired short-term memory depression, reduced appetite Initial clinical working Brainstem infarct, internal Wernicke encephalopathy NA Parkinson disease diagnosis hydrocephalus Final clinical diagnosis Possible CJD Probable CJD Probable CJD Probable CJD Progressive dementia + + NA + Myoclonus j Visual or cerebellar alterations Ataxia Ataxia, optical hallucinations, dysarthria/anarthria Pyramidal or extrapyramidal dysfunction Ataxia, dysarthria, dysphagia Ataxia, small step gait, bradykinesia, dysarthria, apraxia NA Rigor Pyramidal signs Rigor Akinetic mutism CJD typical EEG Severe grouped dysrhythmia, (periodic sharp waves) no typical signs of CJD CSF: proteins NA Tau j NA 19,537 pg/ml pg/ml NSE 30.5 Kg/L NA ng/ml NA S100 j NA 16.2 ng/ml j Other symptoms Peripheral polyneuropathy Incontinence NA Weight loss (diabetes) CJD typical MRI (hyperintense basal ganglia or cortex) NA Y mild cerebellar and parietal atrophy Y other: multilocular leukoencephalopathy + other: white matter lesions +, present; j, not conspicuous/absent; CJD, Creutzfeldt-Jakob disease; EEG, electroencephalograph; M, methionine; MRI, magnetic resonance imaging; NA, not available; NSE, neuron-specific enolase; V, valine. Ó 2011 American Association of Neuropathologists, Inc. 193

3 Eigenbrod et al J Neuropathol Exp Neurol Volume 70, Number 3, March Ó 2011 American Association of Neuropathologists, Inc.

4 J Neuropathol Exp Neurol Volume 70, Number 3, March 2011 and subsequently paraffin-embedded, and finally subjected to routine staining on a Benchmark staining machine with a standard 3,3-diaminobenzidine detection system, according to the manufacturer s instructions (Ventana Medical Systems, Tucson, AZ). Immunohistochemistry and paraffin-embedded tissue (PET) section blots were performed as previously described (1, 29), using PrP-specific antibody L42 (30). For the detection of hyperphosphorylated Tau protein, specific antibody AT8 (1:200; Perbio, Frankfurt, Germany) was used. Detection of >-synuclein was performed using antibody 15G7 (1:600; Novocastra, Newcastle upon Tyne, UK). Histologic sections were evaluated blinded and scored semiquantitatively for spongiosis, gliosis, and neuronal loss independently by 2 neuropathologists. PrP Sc levels in hippocampal regions CA1-4 and subiculum were quantified by densitometry using AIDA image analyzer (Raytest, Straubenhardt, Germany). Immunoblot Analyses Tissues from the frontal cortex, hippocampus, and the cerebellum were homogenized in 9 volumes (wt/vol) of lysis buffer containing 100 mmol/l sodium chloride, 10 mmol/l ethylenediaminetetraacetic acid, 0.5% sodium deoxycholate, 0.5% octylphenoxypolyethoxyethanol (Nonidet P-40), 100 mmol/l tris(hydroxymethyl)aminomethane, ph 6.9, at 37-C. After PK digestion (100 Kg/mL, 1 hour at 37-C), homogenates were subjected to sodium dodecylsulfateypolyacrylamide gel electrophoresis using precast 12% Bis-Tris gels (NuPAGE; Invitrogen, Karlsruhe, Germany). For classification and comparison of PrP Sc types, we used brain homogenates from scjd patients with PrP Sc types 1 (129 MM) and 2 (129 VV and 129 MV), respectively. The monoclonal antibody 3F4 (1: 3.000; DAKO, Hamburg, Germany), recognizing the epitope 109Y112 of human PrP, was used for immunodetection. Immunoreactivity was visualized by enhanced chemiluminescence (CDP-Star; Roche, Mannheim, Germany). PrP Sc glycoform ratios were evaluated densitometrically using AIDA image analyzer (Raytest). RESULTS Clinical Patient 1 A 77-year-old woman from the North of Germany was admitted to hospital with the working diagnosis of brainstem infarction/internal hydrocephalus by her general practitioner. At that point, the patient had a 6-week history of gait instability, confusion, and impaired short-term memory. She also had a history of diabetes mellitus accompanied by peripheral polyneuropathy. In the following 2 weeks, she showed progressive dementia and finally akinetic mutism. The EEG revealed severe grouped dysrhythmia, but periodic sharpwave complexes typical of CJD were absent. Examination of her CSF revealed slightly elevated NSE (30.5 Kg/L). There E196K PRNP Mutation Neuropathology was no record of other measurements of other CSF biomarkers associated with CJD. The patient was classified as possible CJD according to clinical criteria (20Y22). The family history was negative for dementia or a neurodegenerative disease. The patient s mother had died at the age of 83 with pneumonia, and her father had died at the age of 70 after a stroke. The patient died after a disease duration of 2 months. Direct sequencing of the PRNP gene open reading frame revealed the E196K mutation and homozygosity for methionine at codon 129. Patient 2 A 72-year-old man from Southern Germany was admitted to the hospital for etiologic clarification of dementia. The past medical history revealed chronic alcohol abuse. Initially, the patient was diagnosed with Wernicke encephalopathy. Two months later, his wife reported worsening of symptoms that included visual hallucinations, speech, and walking difficulties and incontinence. On physical examination, the patient was drowsy but awake. Neurologic examination demonstrated ataxia, rigor, dysarthria/anarthria, and myoclonus. Electroencephalography revealed typical periodic sharp-wave complexes. Cerebrospinal fluid proteins were detected, but there were no other relevant CSF protein measurements recorded. Magnetic resonance imaging showed mild cerebellar and parietal atrophy but no typical signs of CJD. The patient was classified as probable CJD according to clinical data. He developed akinetic mutism and died after a disease duration of 2.5 months. Direct sequencing of the PRNP gene open reading frame revealed the E196K and homozygosity for valine at codon 129. The family history was negative for dementia or a neurodegenerative disease. Patient 3 A 74-year-old woman from Southwest Germany first developed severe bifrontal headache paralleled by depression and loss of appetite. In the following months, she developed ataxia associated with dysarthria and dysphagia and, later, myoclonus and akinetic mutism. Because of her markedly impaired communication ability, dementia could not be assessed. The EEG indicated the appearance of periodic sharpwave complexes, and CSF levels of , Tau (19,537 pg/ml), NSE (113.8 ng/ml), and S100b (16.2 ng/ml) were elevated. Magnetic resonance imaging showed white matter lesions termed multilocular leukoencephalopathy and a general atrophy but no typical features of CJD. On the basis of clinical data, probable CJD was diagnosed before the genetic analysis was available. Direct sequencing of PRNP revealed heterozygosity for methionine and valine (MV) at codon 129. Subsequent cloning experiments demonstrated coupling of the E196K mutation with methionine at codon 129. The patient died after a disease duration of 4 months. There was no family history of a neurologic or psychiatric disorder. FIGURE 1. Neuropathologic features of Creutzfeldt-Jakob disease in the affected E196K subjects. Spongiform changes in hematoxylin and eosin (H&E)Ystained frontal cortex (scale bar = 100 Km), cerebellar (scale bar = 50 Km), and hippocampal sections (scale bar = 100 Km) and detection of PrP Sc deposits using PrP-specific antibody L42. Left column indicates Patient 1; middle column, Patient 2; right column, Patient 4. Ó 2011 American Association of Neuropathologists, Inc. 195

5 Eigenbrod et al J Neuropathol Exp Neurol Volume 70, Number 3, March 2011 Patient 4 A 69-year-old woman from Eastern Germany had a 6-month history of tremor accompanied by signs of dementia for 3 weeks before she was hospitalized. On admission, she additionally showed rigor, bradykinesia, cogwheel phenomenon, and small step gait and was first diagnosed with Parkinson disease. The patient reported a lack of concentration and considerable weight loss without reduced appetite in the last months. The EEG showed triphasic waves predominantly on the right. Magnetic resonance imaging revealed signal hyperintensities in the frontoparieto-occipital cortex and in the basal ganglia, and white matter lesions and general atrophy were also observed. In the CSF, the proteins were detectable, and Tau was elevated (91200 pg/ml); S100 protein level was unremarkable. In the course of disease, she developed ataxia, dysarthria, and apraxia and finally showed the picture of akinetic mutism, with pyramidal and extrapyramidal signs as well as myoclonus. The clinical diagnosis was probable CJD. The patient died after a disease duration of 8 months. Direct sequencing of the PRNP gene open reading frame revealed the E196K mutation and homozygosity for methionine at codon 129. Importantly, there was a positive family history: the patient s brother was diagnosed with probable genetic CJD (E196K mutation) and died at age 50 years. Detailed clinical data are not available for this patient, and no autopsy was performed. Nevertheless, these data suggest autosomal dominant inheritance of this putative pathologic mutation. Neuropathologic and Immunohistochemical Findings Histologic examination of all 4 cases demonstrated morphologic and immunohistochemical features of CJD. The findings in Patients 1 (129 MM) and 3 (129 MV) resembled those of the scjd MM/MV 1 type with mild to severe spongiform degeneration, astrocytic gliosis, and focal loss of neurons, mainly affecting the cerebral cortex, basal ganglia, thalamus, and cerebellum (Fig. 1, left column); the brainstem was unaffected. In contrast, the hippocampal regions CA1-4, which are usually spared in the MM/MV 1 type (1), were also affected by mild spongiform changes (Fig. 1, left column). Immunohistochemistry revealed a synaptic pattern of PrP Sc deposition in the affected areas including regions CA1-4 of the hippocampus (Fig. 1, left). The latter was even more obvious on PET blot sections (Fig. 2A). Patient 3 additionally showed a distinct microvasculopathy in the white matter, but the neuropathologic findings were similar to those in Patient 1 (not shown). Patient 2 (129 VV) had mild to moderate spongiform degeneration, astrocytic gliosis, and neuronal loss with laminar distribution in the deep cortical layers as well as prominent involvement of basal ganglia, thalamus, cerebellum, and brainstem nuclei (Fig. 1, middle column). The hippocampus, which is characteristically involved in the VV 2 subtype (1), was severely affected (Fig. 1, middle column). Immunostaining revealed synaptic and perineuronal PrP Sc deposits in the affected areas, predominantly in the deep cortical layers and the hippocampus. The cerebellum showed intense plaquelike PrP Sc deposits not only in the molecular and granular 196 FIGURE 2. Paraffin-embedded tissue (PET) blot analysis and quantification of PrP Sc deposits in the hippocampus of E196K patients and corresponding sporadic Creutzfeldt-Jakob disease (scjd) controls. (A) PrP Sc deposits in the hippocampus including regions CA1-4 compared with scjd subtype MM 1 and VV 2 controls. The black line in the scjd MM 1 control indicates the typical border between strong PrP Sc deposits in the subiculum (S) in contrast to the almost spared regions CA1-4. (B) Densitometric quantification of PrP Sc deposits in regions CA1-4 of MM/MV subjects normalized to the intensity of the subiculum (S). (C) Corresponding quantification of PrP Sc deposits in regions CA1-4 of VV subjects. Black bars indicate E196K cases; empty bars, scjd controls. cell layers but also in the white matter (Fig. 1, middle column). Involvement of the hippocampus was also demonstrated on PET blot sections (Fig. 2A). Patient 4 (129 MM) showed a different pattern: spongiform changes, gliosis, and neuronal loss were more severe but in general followed the same morphology and distribution as in Patients 1 and 3 (Fig. 1, right column). There was also mild microvasculopathy in the white matter. Immunohistochemistry displayed clear synaptic PrP Sc staining in the affected areas including the CA1-4 regions of the hippocampus, additionally combined with plaquelike PrP Sc deposits in the cortex and cerebellum (Fig. 1, right column). Large confluent vacuoles with perivacuolar PrP Sc staining were absent in all cortical layers. In contrast, the cerebellum had minor synaptic PrP Sc deposits in the molecular and granular cell layers but also showed a small number of focal plaquelike PrP Sc deposits in the molecular layer. Involvement of the hippocampal regions CA1-4 was confirmed in the PET blot (Fig. 2A). Densitometric quantification of PrP Sc in the hippocampus revealed markedly increased PrP Sc deposits in MM/MV patients carrying the E196K mutation (Patients 1, 3, and 4) versus MM/MV scjd control cases (Fig. 2B). This was also seen in Patient 2 (Fig. 2C); however, for codon Ó 2011 American Association of Neuropathologists, Inc.

6 J Neuropathol Exp Neurol Volume 70, Number 3, March 2011 E196K PRNP Mutation Neuropathology 129 VV patients, the CA1-2 region is also more affected in scjd (1). All cases were additionally examined for possible coexistent synucleinopathies and/or tauopathies. Immunohistochemistry using an antiysynuclein antibody was negative for pathologic deposits in all patients, No hyperphosphorylated tau-positive structures other than very few neurofibrillary tangles in the transentorhinal region of Patients 2 and 4 (corresponding to Braak stage I [31]) were detected (not shown). Biochemical Analysis of the E196K Mutation in PRNP The presence of PrP Sc was demonstrated by Western blot analysis in 3 different brain regions (frontal cortex, cerebellum, and hippocampus) in the 4 patients (Figs. 3 and 4). According to the Western blot classification of PrP Sc types by Parchi et al (11, 12), the PK-resistant PrP (PrP27-30) found in Patients 1 (129 MM) and 3 (129 MV) was PrP Sc type 1 (Figs. 3A, C), whereas Patient 2 (129 VV) presented with PrP Sc type 2 (Fig. 3B). For Patient 1, Western blot analysis in the cerebellum consistently resulted in poor band quality, which we attributed to tissue preservation because this case was otherwise indistinguishable from Case 3. Notably, Western blot analysis of the neuropathologically atypical Patient 4 (129 MM) revealed PrP Sc type 2 in all samples and, in addition, PrP Sc type 1 in the frontal cortex (Fig. 3D, lane 4) and the occipital lobe (not shown). Importantly, this patient does not resemble the scjd cases with so-called mixed phenotype and coexistence of PrP Sc types 1 and 2 (13), which typically have PrP Sc type 1 in the cerebellum and perivacuolar deposits associated with confluent vacuoles in the cortex, whereas this patient had PrP Sc type 2 in the cerebellum FIGURE 3. Western blot analysis of proteinase K (PK)Ytreated homogenates extracted from 3 different brain regions from the 4 affected E196K subjects. (AYD) Frontal cortex homogenates from sporadic Creutzfeldt-Jakob disease (scjd) subtypes classified as MM 1 (lane 1), VV 2 (lane 2), and MV 2 (lane 3) were used as controls. PK-resistant PrP from Patient 1 (A), Patient 2 (B), Patient 3 (C), and Patient 4 (D). Positions of the molecular mass markers are indicated (kda). Cb indicates cerebellum; Fx, frontal cortex; Hip, hippocampus. FIGURE 4. Relative abundance of the 3 glycoforms of proteinase K(PK)Yresistant PrP in the 4 patients carrying the E196K mutation versus sporadic Creutzfeldt-Jakob disease (scjd) controls (Co). The relative abundance of the diglycosylated (upper) band is depicted in white, the monoglycosylated (middle) band is shown in gray, and the nonglycosylated (lower) band is shown in black. Proportions of nonglycosylated bands were analyzed using Student t test. ***, p G 0.001; *, p G (Fig. 3D) and plaquelike deposits in the cortex (Fig. 5). Densitometric analysis of the PK-resistant bands showed glycoform ratios different from those found in scjd cases with underrepresentation of nonglycosylated forms in all brain regions in all 4 E196K patients irrespective of the PrP Sc type and neuropathologic phenotype (Figs. 3 and 4). Statistical analysis revealed a significant underrepresentation of nonglycosylated forms in all brain regions in all 4 patients compared with the respective scjd controls (Patients 1, 2, and 3: p G 0.001, Patient 4: p G 0.05). DISCUSSION Little information has been published regarding the E196K mutation, and in previous reports, only clinical data were available (15Y18). Thus, a definite diagnosis of CJD, which requires autopsied brain material, was not feasible. The cases reported previously were thus classified as probable CJD according to available clinical data. Consequently, the E196K mutation could not be unequivocally assigned to human prion disease. According to the Diagnostic Criteria Guidance from the Advisory Committee on Dangerous Pathogens TSE Working Group ( consum_dh/groups/dh_digitalassets/@dh/@ab/documents/ digitalasset/dh_ pdf), Definite genetic TSE requires a neuropathological confirmation of TSE, plus either definite TSE in a first degree relative (i.e., a parent, child or sibling), or a pathogenic prion protein gene (PRNP) mutation. We are the first to provide a neuropathologic confirmation of TSE in patients with the E196K mutation (based on histopathologic, immunohistochemical, and biochemical results); thus, our data confirm that E196K results in genetic prion disease. Generally, the clinical picture in our 4 patients was initially characterized by nonspecific symptoms including behavioral abnormalities at disease onset that led to initial diagnoses of Parkinson disease, Wernicke encephalopathy, Ó 2011 American Association of Neuropathologists, Inc. 197

7 Eigenbrod et al J Neuropathol Exp Neurol Volume 70, Number 3, March 2011 FIGURE 5. Immunohistochemistry for pathological PrP deposition in Case 4. (AYC) PrP deposits in the frontal cortex. A predominantly synaptic deposition pattern in the gray matter can be seen with focal plaquelike deposits. Panels (B) and (C) show larger magnifications of the areas indicated in (A) with a lined box (B) and dashed box (C), respectively. (DYF) Focal synaptic and small plaquelike deposits in the cerebellar cortex. Scalebar: (A, D) = 200 Km; scalebar: (BYC, F) = 20 Km, scalebar: (E) = 50 Km. and brainstem infarction/internal hydrocephalus. This nonspecific initial picture is consistent with the previous reports, but nonspecific symptoms at onset are common in scjd and sometimes lead to alternative clinical diagnoses. Although the number of cases we analyzed is small, frequent occurrence of alternative initial clinical diagnoses seems unusual to us; as the disease progressed, the patients developed typical features of CJD. Information on additional patients without neuropathologic investigations is provided elsewhere (Schelzke et al, unpublished data). Immunohistochemical analysis ruled out an underlying >-synucleinopathy that could potentially influence the initial clinical picture, which is of particular interest with respect to Patient 4 who showed symptoms suggestive of Parkinson disease. Similarly, there was only minor tau pathology in Patients 2 and 4, making it unlikely that there was a contribution of these changes to the clinical manifestations in these patients. These observations suggest that the nonspecific clinical symptoms at the disease onset were not secondary to other neurodegenerative disease processes. However, Patient 3 had signs suggestive of microvasculopathy, and Patient 4 additionally had white matter lesions on MRI; therefore, we cannot exclude the possibility that the initial clinical syndrome, especially the parkinsonism in Patient 4 (32), was at least partially related to vascular changes in these patients. 198 Ó 2011 American Association of Neuropathologists, Inc.

8 J Neuropathol Exp Neurol Volume 70, Number 3, March 2011 The median onset of disease in the affected E196K subjects was 72 T 3.4 years (mean T SD), which is approximately 10 years later than reported for other point mutations in the PRNP gene, including the most common mutation E200K (19). This finding is consistent with previous reports (15Y18). The mean duration of disease was 4 T 2.7 months, which is shorter than that reported for other genetic TSEs ranging from 5 to 40 months (19). Thus, the affected E196K subjects had a later disease onset along with shorter disease duration. The extensive assessment of genetic and neuropathologic findings in combination with the biochemically determined PrP Sc type in scjd patients has led to a comprehensive classification that includes 6 scjd PrP Sc subtypes (1) and also accounts for mixed phenotypes in an updated classification (13). According to this widely accepted classification of scjd cases, we found that Patients 1 and 3 were closely related with the scjd MM/MV 1, Patient 2 was closely related with the scjd VV 2 subtype, whereas Patient 4 had an atypical phenotype. Densitometric analysis of the PK-resistant bands revealed a significant underrepresentation of the nonglycosylated form in all brain regions in all 4 patients carrying the E196K mutation, irrespective of the PrP Sc type and neuropathologic phenotype. Such a marked underrepresentation of the nonglycosylated form relative to the diglycosylated and monoglycosylated form is not pathognomonic for the E196K mutation as it has previously been observed in subjects with a D178N mutation coupled with methionine codon at position 129 in PRNP (33). More precisely, the classic scjd VV 2 subtype usually presents with spongiform changes and plaquelike/perineuronal PrP Sc deposits predominantly in the deep cortical layers, hippocampal region CA4, and the cerebellum, whereas the scjd MM/MV 1 subtype usually shows spongiform changes throughout all cortical layers as well as synaptic PrP Sc staining. Patients 1, 2, and 3 closely resemble the picture typically seen in the scjd subtypes of the corresponding codon 129 genotype (1), indicating that the phenotype associated with the E196K mutation in these 3 patients is determined by the polymorphic codon 129. Codon 129 polymorphism can influence the disease phenotype as best described for the D178N mutation presenting as fatal familial insomnia when coupled with methionine and as genetic CJD when coupled with valine at codon 129 in PRNP (7). In addition to this overall similarity to scjd, the present study indicates that the E196K mutation severely affects the hippocampal CA1-4 regions. In summary, our findings indicate that the histopathology associated with this rare PRNP mutation comprises both cases with atypical neuropathology and cases that closely resemble subtypes of scjd corresponding to the classification of Parchi et al (1) with subtle modifications in the hippocampal regions CA1-4. ACKNOWLEDGMENTS The authors thank Eva Grasbon-Frodl and Rüdiger Schmalzbauer for providing the technical basis regarding sequencing and initial summary of the relevant literature, E196K PRNP Mutation Neuropathology respectively. Elzbieta Staniszewski, Ursula Jung, and Gertrud Kwiatkowski are gratefully acknowledged for expert technical assistance. REFERENCES 1. Parchi P, Giese A, Capellari S, et al. Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects. Ann Neurol 1999;46:224Y33 2. Horwich AL, Weissman JS. Deadly conformationsvprotein misfolding in prion disease. Cell 1997;98:499Y Prusiner SB. Prions. Proc Natl Acad Sci U S A 1998;95:13363Y83 4. Lugaresi E, Medori R, Montagna P, et al. Fatal familial insomnia and dysautonomia with selective degeneration of thalamic nuclei. N Engl J Med 1986;315:997Y Brown P, Cervenáková L, Boellaard JW, et al. Identification of a PRNP gene mutation in Jakob s original Creutzfeldt-Jakob disease family. Lancet 1994;344:130Y31 6. Hainfellner JA, Brantner-Inthaler S, Cervenáková L, et al. The original Gerstmann-Sträussler-Scheinker family of Austria: Divergent clinicopathological phenotypes but constant PrP genotype. Brain Pathol 1995; 5:20 7. Goldfarb LG, Petersen RB, Tabaton M, et al. Fatal familial insomnia and familial Creutzfeldt-Jakob disease: Disease phenotype determined by a DNA polymorphism. Science 1992;258:806Y8 8. Piccardo P, Dlouhy SR, Lievens PM, et al. Phenotypic variability of Gerstmann-Sträussler-Scheinker disease is associated with prion protein heterogeneity. J Neuropathol Exp Neurol 1998;57:979Y88 9. Parchi P, Chen SG, Brown P, et al. Different patterns of truncated prion protein fragments correlate with distinct phenotypes in P102L Gerstmann-Sträussler-Scheinker disease. Proc Natl Acad Sci U S A 1998;95:8322Y Chen SG, Parchi P, Brown P, et al. Allelic origin of the abnormal prion protein isoform in familial prion diseases. Nat Med 1997;3:1009Y Parchi P, Castellani R, Capellari S, et al. Molecular basis of phenotypic variability in sporadic Creutzfeldt-Jakob disease. Ann Neurol 1996;39: 767Y Parchi P, Capellari S, Chen SG, et al. Typing prion isoforms. Nature 1997;386:232Y Parchi P, Strammiello R, Notari S, et al. Incidence and spectrum of sporadic Creutzfeldt-Jakob disease variants with mixed phenotype and co-occurrence of PrP Sc types: An updated classification. Acta Neuropathol 2009;118:659Y Notari S, Capellari S, Giese A, et al. Effects of different experimental conditions on the PrP Sc core generated by protease digestion: Implications for strain typing and molecular classification of CJD. J Biol Chem 2004;279:16797Y Peoc h K, Manivet P, Beaudry P, et al. Identification of three novel mutations (E196K, V203I, E211Q) in the prion protein gene (PRNP) in inherited prion diseases with Creutzfeldt-Jakob disease phenotype. Hum Mutat 2000;15: Tumani H, Windl O, Kretzschmar HA, et al. Clinically atypical CJD: Diagnostic relevance of cerebrospinal fluid markers and molecular genetic analysis? [in German] Dtsch Med Wochenschr 2002;127:318Y Clerici F, Elia A, Girotti F, et al. Atypical presentation of Creutzfeldt- Jakob disease: The first Italian case associated with E196K mutation in the PRNP gene. J Neurol Sci 2008;275:145Y Béjot Y, Osseby GV, Caillier M, et al. Rare E196K mutation in the PRNP gene of a patient exhibiting behavioral abnormalities. Clin Neurol Neurosurg 2010;12:244Y Kovács GG, Puopolo M, Ladogana A, et al. EUROCJD. Genetic prion disease: The EUROCJD experience. Hum Genet 2005;118:166Y WHO. Human transmissible spongiform encephalopathies. Wkly Epidemiol Rec 1998;73:361Y Zerr I, Pocchiari M, Collins S, et al. Analysis of EEG and CSF proteins as aids to the diagnosis of Creutzfeldt-Jakob disease. Neurology 2000;55:811Y Zerr I, Kallenberg K, Summers DM, et al. Updated clinical diagnostic criteria for sporadic Creutzfeldt-Jakob disease. Brain 2009;132: 2659Y68 Ó 2011 American Association of Neuropathologists, Inc. 199

9 Eigenbrod et al J Neuropathol Exp Neurol Volume 70, Number 3, March Mokuno K, Kato K, Kawai K, et al. Neuron specific enolase and S-100 protein levels in the cerebrospinal fluid of the patients with various neurological diseases. J Neurol Sci 1983;60:443Y Hsich G, Kenney K, Gibbs CJ, et al. The brain protein in cerebrospinal fluid as a marker for transmissible spongiform encephalopathies. N Engl J Med 1996;335:924Y Zerr I, Bodemer M, Weber T. The brain protein and transmissible spongiform encephalopathy. N Engl J Med 1997;336: Zerr I, Bodemer M, Gefeller O, et al. Detection of protein in the cerebrospinal fluid supports the diagnosis of Creutzfeldt-Jakob disease. Ann Neurol 1998;43:32Y Windl O, Dempster M, Estibeiro JP, et al. Genetic basis of Creutzfeldt- Jakob disease in the United Kingdom: A systematic analysis of predisposing mutations and allelic variation in the PRNP gene. Hum Genet 1996; 98:259Y Grasbon-Frodl E, Lorenz H, Mann U, et al. Loss of glycosylation associated with the T183A mutation in human prion disease. Acta Neuropathol 2004;108:476Y Schulz-Schaeffer WJ, Tschoke S, Kranefuss N, et al. The paraffinembedded tissue blot detects PrP(Sc) early in the incubation time in prion diseases. Am J Pathol 2000;156:51Y Vorberg I, Buschmann, A, Harmeyer S, et al. A novel epitope for the specific detection of exogenous prion proteins in transgenic mice and transfected murine cell lines. Virology 1999;255:26Y Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 1991;82:239Y Sibon I, Tison F. Vascular parkinsonism. Review. Curr Opin Neurol 2004;17:49Y Parchi P, Capellari S, Chin S, et al. A subtype of sporadic prion disease mimicking fatal familial insomnia. Neurology 1999;52:1757Y Ó 2011 American Association of Neuropathologists, Inc.