Introduction

Human prion protein contains 253 amino acids encoded by the prion protein gene (PRNP), which is located on chromosome 20 in humans.1 PRNP plays an important role in conferring susceptibility to prion disease. A number of point and insertion mutations of PRNP have been linked to familial Creutzfeldt–Jakob disease (CJD), Gerstmann–Sträussler–Scheinker disease (GSS), and fatal familial insomnia (FFI).2, 3, 4 Moreover, polymorphisms of PRNP appear to be able to influence expression of prion disease in sporadic and iatrogenic CJD.5, 6, 7

The downstream prion-like protein (doppel or Dpl) shares significant biochemical and structural homology with the cellular prion protein, PrPC The PrP-like protein gene (PRND) is located at approximately 27 kb downstream of the human PRNP.8 In the PRND, the codons 25, 56, and 174 and the 3′ untranslated region (UTR) +28 site have been shown to be polymorphic.9, 10 The PRND polymorphism at 3′ UTR +28 is a noncoding thymidine (T)–cytidine (C) change. Recently, polymorphisms in the open reading frame (ORF) of human PRND have been analysed in relation to the occurrence of prion diseases and other neurodegenerative disorders.9, 10, 11, 12, 13, 14 However, an association between PRND polymorphism at 3′ UTR +28 and sporadic CJD as well as genotype and allele distributions of this polymorphism in the normal population have not been reported thus far.

Therefore, we have examined PRND polymorphism at 3′ UTR +28 in 102 healthy Koreans and 110 sporadic CJD patients to investigate whether the genotype and allele frequencies of this PRND polymorphism is associated with sporadic CJD in a Korean population.

Materials and methods

CJD patient population

We used previously established criteria15 to enhance the level of accuracy of diagnosis of sporadic CJD in Korea. Neuropathologically confirmed patients and/or patients with immunochemical detection of PrPSc in the brain were classified as definite CJD. Patients were classified as probable CJD if they exhibited a rapidly progressive dementia, periodical sharp wave complexes (PSWC) on electroencephalography (EEG), 14-3-3 protein in CSF, a duration of dementia <2 years and two of the following: myoclonus, visual and cerebellar symptoms or both, pyramidal or extrapyramidal signs or both, or akinetic mutism. The sporadic CJD cases, which were diagnosed by immunohistochemistry and DNA sequencing, have been reported previously in Korea.16 Of the 110 suspected CJD cases, 17 cases were classified as definite CJD and 93 cases were classified as probable CJD. Blood samples were collected from 102 healthy volunteers and 110 sporadic CJD patients in Korea. The study was approved by the Ethical Committee of Chunchon Sacred Heart Hospital.

Laboratory analyses

Genomic DNA was extracted from 200 μl of blood using the QIAamp DNA blood mini kit (Qiagen, Valencia, CA, USA) following the supplier's instructions. For polymerase chain reaction (PCR) amplification of PRND polymorphism at 3′ UTR +28, we designed the Hdpl-1 (5′-GTTTCTCTGGCAGGTTCTG-3′) and Hdpl-2 (5′-GAGAAGAGCTGGGTCACTT-3′) primers producing a 623 bp fragment. The PCR reagents contained 50 pmol of each primer, 5 μl of 10 × Taq DNA polymerase buffer, 1.5 mM MgCl2, 0.2 mM of each dNTP mixture, and 2.5 units of Taq DNA polymerase (Promega, Madison, WI, USA). The PCR conditions were as follows: an initial denaturation step of 95°C for 2 min, 30 cycles of 94°C for 40 s, 56°C for 45 s, and 72°C for 90 s, and final extension at 72°C for 10 min. The purification of PCR products for sequencing was preformed using a QIAquick gel extraction kit (Qiagen, Valencia, CA, USA). The DNA sequencing was preformed on an ABI 377 automatic sequencer using a Taq dideoxy terminator cycle sequencing kit (ABI, Foster City, CA, USA). Nucleic acid sequences were assembled and edited using a combination of the ABI 377 DNA Sequencer Data Analysis Program and Sequence Navigator Software.

Statistical analyses

χ2 tests were used to determine whether PRND 3′ UTR +28 were in Hardy–Weinberg equilibrium (HWE) in the Korean population. Differences in genotype and allelic frequencies between healthy Koreans and patients with sporadic CJD were compared by χ2 tests using the SAS 8.1 Software (SAS Institute Inc., Cary, NC, USA). We also examined Lewontin's D′ (∣D′∣) and linkage disequilibrium coefficient r2 between three single-nucleotide polymorphisms (SNPs) of PRND and PRNP.17, 18

Results

The genotype frequencies at PRND 3′ UTR +28 were in HWE in Korean control group (P=0.373) (Table 1). We analysed genotype and allele distributions of 3′ UTR +28 polymorphism in a Korean normal population. Of the 102 normal samples, 24 (23.5%) were homozygous for T, 20 (19.6%) were homozygous for C, and 58 (56.9%) were heterozygous at 3′ UTR +28, with allele frequency of 0.52:0.48 T:C (Table 1). To investigate whether there is an association between this polymorphism and sporadic CJD, we analysed the genotype and allele frequencies of Korean sporadic CJD cases. Of the 110 sporadic CJD cases, 48 (43.6%) were homozygous for T, 21 (19.1%) were homozygous for C, and 41 (37.3%) were heterozygous at 3′ UTR +28, with allele frequency of 0.623:0.377 T:C (Table 1). Significant differences were found between sporadic CJD and normal controls in genotype distribution (P=0.005) and in allele frequency (P=0.032) at 3′ UTR +28. Furthermore, as shown in Table 2, neither age nor gender had any influences on the PRND polymorphism at 3′ UTR +28. We have also examined whether there are significant differences in genotype frequency of this polymorphism between definite cases or probable cases and controls. There was significant difference in genotype frequencies between normal controls and definite sporadic CJD (P=0.036) as well as a significant difference between controls and probable sporadic CJD (P=0.015) (Table 2). PRNP polymorphisms at codon 129 and 219 are thought to play an important role in susceptibility to sporadic CJD.6, 7 These polymorphisms in Korean are also associated with sporadic CJD (data not shown). To examine whether strong linkage disequilibrium is observed between the SNP of PRND at 3′ UTR +28 and the two SNPs of PRNP, linkage disequilibrium coefficients (∣D′∣) and r2 between three SNPs were calculated in normal Korean controls (Table 3). Absolute linkage disequilibrium (∣D′∣=1 and r2=1), complete linkage disequilibrium (∣D′∣=1 and r2≠1), and strong linkage disequilibrium (∣D′∣>0.9) were not observed in three SNPs. An SNP at PRND 3′ UTR +28 was in no linkage disequilibrium with two SNPs at PRNP codons 129 and 219.

Table 1 Genotype and allele frequencies of the polymorphism at 3′ UTR +28 of the PRND in samples of the normal Korean population and sporadic CJD patients
Full size table
Table 2 Genotypes at 3′ UTR +28 of the PRND, age, and sex in Korean normal population and definite/probable cases with sporadic CJD
Full size table
Table 3 Pairwise linkage disequilibrium coefficients between three SNPs of PRND and PRNP in Korean control samples
Full size table

Discussion

Polymorphisms in the PRND have been reported in several European countries, whereas analysis of PRND sequence in Asian countries has not been examined previously. In this study, we studied polymorphism at 3′ UTR +28 of PRND in a sampling of the Korean population. We have found evidence that a specific genotype (TT) or allele (T) at 3′ UTR +28 of the PRND is a risk factor for sporadic CJD.

In a previous study, it has been shown that the genotype frequencies of the ORF of PRNP at codon 219 as well as codon 129 are associated with susceptibility to sporadic CJD.6, 7 The doppel protein has 25% homology to the C-terminal part of the prion protein and the overexpression of doppel is associated with cerebellar neurodegeneration of PrP-knockout mice.8 These studies strongly suggest that polymorphisms of the PRND could play a role in prion diseases.

The major PRND polymorphism reported in several European countries was substitution of threonine (Thr) with methionine (Met) at codon 174. In these studies, codon 174 polymorphism in British and French populations was not associated with sporadic CJD or variant CJD,9, 10 nor was this polymorphism in Spanish people associated with sporadic Alzheimer's disease (AD).12 In contrast, in German and Dutch populations codon 174 polymorphism was associated with sporadic CJD and/or AD.11, 13 In the German control group, genotype frequencies were not in HWE proportions (P=0.002). This prevents the exact interpretation of the study since HWE may be deviated by bias in sample selection or genotyping errors.11 The polymorphisms at codons 26 and 56 of PRND were not associated with susceptibility to sporadic CJD.9, 10 The PRND polymorphism at 3′ UTR +28 has been reported in two studies.9, 19 The allele frequency for PRND polymorphism at 3′ UTR +28 in the normal Korean population (52% T, 48% C) is not significantly different from that previously reported for healthy French controls (51% T, 49% C) (χ2=0.0579, df=2, P=0.810).19 The genotype frequency of this PRND polymorphism is not related to those of polymorphisms at codons 129 and 219 of PRNP, as the genotype frequencies for the codons 129 and 219 in Korean are quite different from those in French. It has been known that SNPs in 3′ UTR of genes can affect gene expression.20 It is possible that the 3′ UTR +28 polymorphism is involved in either doppel gene expression or gene stability and that this may play a role in susceptibility to sporadic CJD. Recently, it was found that Dpl levels in brain tissue of sporadic CJD patients were similar to the levels in normal controls.21 This result suggests that expression levels of Dpl may not be influenced in prion disease.21 Unexpectedly, genotype frequency of T/C heterozygotes in Korean sporadic CJD patients (37.3%) was lower than that in normal controls (56.9%). There may be various explanations for this result. One explanation is that this polymorphism may be in linkage disequilibrium with a nearby locus involved in sporadic CJD. Undetected polymorphisms in strong linkage disequilibrium with this polymorphism may cause a spurious association. Finally, similar to our results, an association between SNP at 3′ UTR and the incidence of other diseases has been reported.22, 23 In the current study, comparing the total percentage of TT vs the combined TC/CC for CJD patients yielded 43.6% TT and 56.4% TC/CC compared to 23.5% TT and 76.5% TT/CC for controls; this was significantly different (χ2=9.5403, df=1, P=0.002).

Our results are the first genetic association study of the PRND noncoding region with sporadic CJD. Recently, we reported that the distributions of codons 129 and 219 genotypes of PRNP in a Korean population differ significantly from those reported for other ethnic groups.24 Thus, further investigations in different ethnic groups including Europeans will be necessary to assess association between sporadic CJD and the PRND 3′ UTR +28 polymorphism. Furthermore, since it is unknown whether this polymorphism affects mRNA stability or gene expression of PRND, further experiments should be conducted to clarify the role of this polymorphism in PRND function.