SEARCH

SEARCH BY CITATION

Keywords:

  • copper;
  • Creutzfeldt–Jakob disease;
  • manganese;
  • oxidative stress;
  • prion;
  • zinc

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Human prion diseases are characterized by the conversion of the normal prion protein (PrPC) into a pathogenic isomer (PrPSc). Distinct PrPSc conformers are associated with different subtypes of prion diseases. PrPC binds copper and has antioxidation activity. Changes in metal-ion occupancy can lead to significant decline of the antioxidation activity and changes in conformation of the protein. We studied the trace element status of brains from patients with sporadic Creutzfeldt–Jakob disease (sCJD). We found a decrease of up to 50% of copper and an increase in manganese of approximately 10-fold in the brain tissues from sCJD subjects. We have also studied the metal occupancy of PrP in sCJD patients. We observed striking elevation of manganese and, to a lesser extent, of zinc accompanied by significant reduction of copper bound to purified PrP in all sCJD variants, determined by the PrP genotype and PrPSc type, combined. Both zinc and manganese were undetectable in PrPC preparations from controls. Copper and manganese changes were pronounced in sCJD subjects homozygous for methionine at codon 129 and carrying PrPSc type-1. Anti-oxidation activity of purified PrP was dramatically reduced by up to 85% in the sCJD variants, and correlated with increased in oxidative stress markers in sCJD brains. These results suggest that altered metal-ion occupancy of PrP plays a pivotal role in the pathogenesis of prion diseases. Since the metal changes differed in each sCJD variants, they may contribute to the diversity of PrPSc and disease phenotype in sCJD. Finally, this study also presented two potential approaches in the diagnosis of CJD; the significant increase in brain manganese makes it potentially detectable by MRI, and the binding of manganese by PrP in sCJD might represent a novel diagnostic marker.

Abbreviations used
BHT

butylated hydroxytoulene

Ca

calcium

CJD

Creutzfeldt–Jakob disease

Cu

copper

DNPH

2,4-dinitrophenyl-hydrazine

Fe

iron

4-HNE

4-hydroxyalkenals

ICP-MS

inductively coupled plasma mass spectrometry

M

methionine

mAb

monoclonal antibody

MDA

malondialdehyde

Mg

mangesium

Mn

manganese

NBT

nitro-blue tetrazolium

PK

proteinase K

PrPC

prion protein

PrPSc

pathogenic isomer

sCJD

sporadic Creutzfeldt–Jakob disease

SOD

superoxide dismutase

V

valine

Zn

zinc.

The central event in the pathogenesis of prion diseases, which include Creutzfeldt–Jakob disease (CJD) in humans and scrapie in animals, is believed to be the post-translational conversion of the normal cellular prion protein (PrPC) into an abnormal isoform called scrapie PrP (PrPSc) that is partially resistant to proteases and is associated with transmissible disease (Prusiner 1998). Recent studies have showed that PrPC not only binds copper (Cu) within the octarepeat region located in the unstructured N-terminus (Brown et al. 1997), but under certain specific circumstances may bind along the C-terminal structured domain of protein fragments (Cereghetti et al. 2001). Furthermore, recombinant PrPC can also bind other metal ions such as manganese (Mn) (Brown et al. 2000) at both the octarepeats and the C-terminal sites (Collinge 2001). However, when Mn replaces Cu, PrP reportedly changes conformation and loses function (Brown et al. 2000). Indeed, accumulating evidence suggests that metallochemical alterations may play a role in the pathogenesis of prion diseases and other neurodegenerative diseases (Bush 2000).

Increasing evidence indicates that the phenotypic diversity of prion diseases is related to the multiple conformations that PrPSc may adopt (Monari et al. 1994; Parchi et al. 2000). Two major, but not exclusive, determinants of the disease phenotype in sporadic CJD (sCJD) are thought to be the genotype at codon 129, the site of a common methionine (M)/valine (V) polymorphism (Goldfarb et al. 1992), and the PrPSc type as determined by the size of the protease-resistant PrPSc, which in turn, is an indication of the PrPSc conformation (Monari et al. 1994; Parchi et al. 1999, 2000). Two main types and several minor types of PrPSc have been observed in CJD (Parchi et al. 1999, 2000). PrPSc type-1 and type-2 results from the cleavage of PrPSc by proteinase K (PK) at residues 82 and 97, respectively, while secondary cleavages around the two main cleavage sites generate the minor PrPSc types (Parchi et al. 2000). It has been proposed that in sCJD, the genotype at codon 129 affects the conformation of PrPSc and thus the site of protease cleavage (Parchi et al. 2000). In turn, PrPSc species with different conformations have been associated with distinct disease phenotypes (Monari et al. 1994; Telling et al. 1996; Parchi et al. 2000). Indirect evidence suggests that the level of metal-ion occupancy in PrP might also affect the conformation of PrPSc associated with human prion diseases (Wadsworth et al. 1999).

It has been demonstrated that both recombinant and brain-derived PrP have superoxide dismutase (SOD)-like activity when Cu is bound to the octarepeat region resulting in conformational changes to the protein (Brown et al. 1999, 2001; Wong et al. 2000b). When Cu is replaced with Mn in recombinant PrP, it loses the SOD-like activity (Brown et al. 2000). At the same time, there is increasing evidence linking oxidative stress to several neurodegenerative diseases (Bush 2000), including animal prion disease (Guentchev et al. 2000). All these motivate us to investigate (i) whether, like in Alzheimer's disease (Deibel et al. 1996), metal binding is altered in sCJD, the most common human prion disease, (ii) whether these alterations correlate with disease phenotype such as PrPSc type and PrP genotype at codon 129, and (iii) whether metal imbalances also correlate with PrP loses antioxidant function. These studies were carried out in brain tissues and affinity purified PrP preparations (i.e. PrPC, PrPSc and possibly other abnormal PrP species) obtained from four major variants of sCJD identified according to the genotype at codon 129 of the PrP gene and the PrPSc type as established by Parchi et al. (1999). We report that Cu binding to PrP purified from sCJD was significantly decreased while the binding of Mn and Zn was markedly increased. The diminution of bound Cu was especially severe in PrP preparations containing PrPSc type-1, while bound Mn showed a more pronounced increase in PrP preparations from sCJD subjects homozygous for methionine (MM) at codon 129. SOD-like activity was reduced by approximately 85% in each of the sCJD variants examined.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Unless described, all chemicals and enzymes were purchased from Sigma (St Louis, MO, USA) and Roche Diagnostic (Indianapolis, IN, USA), respectively.

Tissues

Brain tissues were obtained with informed consent at autopsy from the frontal cortex of nine subjects with clinically and pathologically proven sCJD in the USA, and they were deposited at the National Prion Disease Pathology Surveillance Center in Cleveland, OH, USA. The genotype for codon 129 of the PrP gene (PRNP), which encodes a methionine (M) or valine (V) polymorphism, in each sCJD sample was determined by sequencing and restriction digests of the amplified PRNP fragment as described (Goldfarb et al. 1992). The sCJD cases were classified according to the criteria proposed by Parchi et al. (1999): (i) Four cases carried PrPSc type-1 which include the MM type-1 variant (n = 3) and the clinically rare VV type-1 variant (n = 1) (Parchi et al. 1999); (ii) five cases carried PrPSc type-2 including the VV type-2 variant (n = 3) and the infrequent MM type-2 variant (n = 2) (Parchi et al. 1999); in addition, (iii) three cases in which the presence of prion diseases had been ruled out were used as controls. Brain homogenate was prepared and total protein concentrations measured as described (Parchi et al. 2000; Wong et al. 2001).

Antibodies

The monoclonal anti-PrP 8H4 recognized an epitope between residue 175–185 along human PrP (Zanusso et al. 1998), and it has been demonstrated to react equally well with both normal and abnormal PrPs (Zanusso et al. 1998; Manson et al. 1999). Purified goat polyclonal antibody to actin was purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA), while sheep polyclonal antibodies against Cu/Zn-SOD and Mn-SOD were purchased from Calbiochem (San Diego, CA, USA).

PrP purification and immunodetection

The purification of PrP from brain homogenates was performed as described (Brown et al. 2001). Briefly, anti-PrP 8H4 antibody was coupled to cyanogen bromide activated Sepharose beads (Amersham, Piscataway, NJ, USA) before mixing with the brain extracts overnight at 4°C. After extensive washing, the protein was eluted from the beads with glycine (pH 4) and neutralized with Tris–HCl (pH 8) followed by assessing the eluted protein by immunoblotting and silver staining (Pierce, Rockford, IL, USA). The purified protein was quantified by spectrophotometry (Brown et al. 2001) to ensure identical amounts of protein were taken and boiled in loading buffer before separation by electrophoresis (Novex, Los Angeles, CA, USA), transferred to nitrocellulose, probed with primary and horseradish peroxidase (HRP)-conjugated secondary antibodies, respectively, and analysed for immunoreactivity using the POD chemiluminescence kit (Roche Diagnostic, Indianapolis, IN, USA) (Wong et al. 2001).

ICP-MS analysis

Trace element contents of both brain homogenate and immunoprecipitated protein samples were determined using inductively coupled plasma mass spectrometry (ICP-MS) (Beauchemin and Kisilevsky 1998). PrP was affinity purified from brain homogenate as described (Brown et al. 2001) using 8H4-coupled Sepharose beads (Amersham). The concentrations of total protein in the brain homogenate and of immuno-purified PrP were determined by spectrophotometry (Brown et al. 2001). Before analysis, all lab-ware was acid-leached overnight under pressure to minimize residual metal contaminants (Beauchemin and Kisilevsky 1998). Samples were hydrolysed overnight in nitric acid (1 mL SpA grade per 100 µL sample) at room temperature (24°C). They were then sealed and heated under pressure and temperature control to 120 psi and at least 140°C for 10 min. The resulting cleared hydrolysed samples were diluted to 4% nitric acid before analysing for Cu, Mn and Zn contents. Quantitative calibrations were made using standards prepared from certified single elements (1000 ppm standard solutions). The concentrations used were 0, 2, 4, 6, 8, 10, 20, 40, 60, 80 and 100 ppb and each solution contained 100 ppb internal standard to monitor the analysis. Each sample was analysed at least three times, washed with 2% nitric acid for 110 s between solutions. A 10 ppb standard check solution was analysed after each set of samples to monitor the analysis. The dilution factors were recorded and accounted for in the calculation of the results, quoted in gram wet weight or µg/mg protein. Background measurements without homogenate or with 8H4-coupled Sepharose beads not mixed with brain homogenate were determined and subtracted from appropriate readings.

Superoxide dismutase assay

SOD activity of affinity-purified PrP was determined as described (Brown et al. 1999, 2000) employing superoxide production from xanthine oxidase and detection of a coloured formazan product formed from nitro-blue tetrazolium (NBT) at 560 nm. The SOD activity was expressed as percentage inhibition of formazan produced where 100% formazan product formation is the amount of NBT reduced by xanthine oxidase-formed radicals in control reactions without brain extracts or immuno-purified PrP. For Mn SOD activity measurement, PrP-depleted brain extracts were incubated at room temperature in 4 mm KCN for 20 min before assaying. Cu/Zn SOD activity was calculated by subtracting Mn SOD activity from total SOD activity in the PrP-depleted brain homogenates. All assays were performed in triplicate.

Reactive carbonyl assay of oxidized proteins

Protein oxidation in brain homogenates was measured by assaying the amount of carbonyl groups on proteins using the OxyBlot™ Protein Oxidation Detection Kit (Intergen, Boston, MA, USA). The procedure was performed as described (Wong et al. 2001), according to manufacturer's recommendation, by derivatizing brain homogenates with 2,4-dinitrophenyl-hydrazine (DNPH) before loading onto a reducing 4–20% linear gradient sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) (Novex). The proteins were then electrotransferred, probed with a rabbit antiserum against the derivatized carbonyl groups and visualized using the POD™ chemiluminescence kit (Roche). No immunoreactivity was detected in the non-DNPH derivatized brain lysates (data not shown). Each sample was analysed at least three times.

Lipid peroxidation assay

Lipid peroxidation, via the levels of malondialdehyde (MDA) and 4-hydroxyalkenals (4-HNE), in brain homogenate was measured using the N-methyl-2-phenylindole-based LPO-586™ lipid peroxidation kit (Oxis International, Portland, OR, USA) as described (Wong et al. 2001). Standard curves of both MDA and 4-HNE were established. To minimize non-specific oxidation during sample preparation, 5 mm butylated hydroxytoulene (BHT) (dissolved in acetonitrile) was added to the extraction buffer. The assay was performed in triplicates according to manufacturer's recommendation in which identical amount of total brain lysates were mixed with 0.5 m BHT and 0.1 mmN-methyl-2-phenylindole. For MDA measurement, 150 µL of reagent grade HCl was added, whereas for 4-HNE measurement 150 µL of 15.4 m methanesulfonic acid was added. The sample was mixed, incubated for 1 h at 45°C, supernatant extracted after brief centrifugation before reading at 586 nm.

Statistical analysis

The data obtained in the metals analysis was analysed by two-tailed distribution Student's t-test. The significance of the differences for the superoxide dismutase activity and lipid peroxidation analysis was assessed using paired sample one-tailed distribution Student's t-test.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

PrPSc species

The main distinction between PrPC and PrPSc is not the primary sequence but has been postulated to be their protein conformation (Prusiner 1998). In prion-infected brains, PrPSc together with PrPC and other abnormal forms of PrP (include the suggested intermediary between PrPC conversion to PrPSc, termed PrP*) are present, in contrast to the preponderant of PrPC in non-diseased situation (Caughey 2000; Parchi et al. 2000). The conformation adopted by PrPSc allegedly makes the protein partially resistant to protease digestion, which resulted in generating a truncated PrP fragment designated as PrP27-30 (Prusiner 1998). This proteolytic process is the current mean of discriminating between PrPC and PrPSc, but the cleavage removed the main metal-binding octarepeat region (Parchi et al. 2000). Paradoxically, the protease-sensitive PrP* has also been implicated in certain prion transmissions (Lasmezas et al. 1997; Caughey 2000). As a result of this, we decided to examine total PrP (including PrPC, PrPSc and other abnormal forms of PrP) in the brain homogenates obtained from four groups of sCJD cases which were homozygous for methionine (MM) or valine (VV) at codon 129 of the PrP gene and carried PrPSc type-1 or type-2 (Parchi et al. 1997, 1999). Cases heterozygous at codon 129 were not analysed because following PK treatment the PrPSc species present in sCJD cases that are MV heterozygous at position 129 generate fragments similar in size to those generated in VV homozygous cases suggesting that PrPSc in MV and VV cases have comparable conformations (Parchi et al. 2000). PrP preparations from the control and four sCJD groups untreated with PK showed similar reactivity for the monoclonal antibody (mAb) to PrP 8H4 (Fig. 1a, left panel). Following treatment with PK, all the PrPSc species in sCJD samples migrated faster on the gel but, as expected, PrPSc type-2 migrated faster than type-1 (Monari et al. 1994; Parchi et al. 2000): the unglycosylated form of PrPSc type-1 migrated at ∼21 kDa while that of type-2 migrated at ∼19 kDa (Fig. 1a, right panel). PrP preparation from control was totally digested by proteinase K, confirming they contain only the PrPC isoform.

image

Figure 1.  Prion protein and its associated superoxide dismutase (SOD)-like activity from the frontal cortex of control and sporadic CJD (sCJD) variants brain homogenates. (a) Detection of PrP in 30 µg of brain homogenates (left panel), and PrPSc in sporadic CJD samples (right panel) after proteinase K treatment (50 µg/mL for 1 h at 37°C) (Parchi et al. 1999) detected with anti-PrP 8H4 which has been shown to react equally well with both normal and abnormal PrP (Zanusso et al. 1998; Manson et al. 1999). Note the different gel mobility of the protease-resistant PrPSc fragment in the type-1 and 2 preparations (Parchi et al. 2000). (b) Affinity-purified PrP using 8H4-coupled Sepharose beads were assessed by (left panel) immunoblotting using anti-PrP 8H4 (Zanusso et al. 1998; Li et al. 2000) and (right panel) silver staining. The molecular weight markers are indicated in kDa. (c) The level of SOD activity using 1 µg of affinity purified PrP in each sample (Brown et al. 2001). There was no detectable background SOD reading from the 8H4-coupled Sepharose beads not incubated with brain homogenate. *SOD values from sporadic CJD variants, classified by PrP genotype (white bars) of MM (n = 5) and VV (n = 4), or PrPSc type-1 (n = 4) and 2 (n = 5) (grey bars) significantly different from control values (black bar, n = 3) (Student's t-test, p < 0.001). VV values significantly different from MM values, and type-1-values differed significantly from type-2-values (Student's t-test, p < 0.03). All values represent the mean ± SEM (the positive and negative error bars are of the same value) of three assays for each sample.

Download figure to PowerPoint

Direct measurement of metal binding by PrP

Analysis of short peptides encompassing the octarepeat region suggests Cu bind with a surprising low dissociation constant (µm range) (Whittal et al. 2000). However, other studies indicate that both the octarepeat region at the N-terminal and the binding sites along the C-terminal in human PrP can binds Cu with a significantly higher affinity (Kd) of 10−14 M (Collinge 2001). These observations suggest that studies using synthetic peptides do not truly reflect the interaction between full-length PrP and metal ions. Furthermore, no detectable loss of the bound metal was observed after affinity purifying PrP (Brown et al. 2001). Besides Cu, PrP can also binds Mn and Zn, but not Mg, Fe or Ca in vitro (Brown et al. 2000; Collinge 2001).

We therefore decided to measure the amount of Cu, Mn and Zn bound to affinity purified PrP. The preparation was assessed by immunoblotting using anti-PrP 8H4 (Fig. 1b, left panel) and silver staining (Fig. 1b, right panel). The metal content of PrP from both control and sCJD brains was assessed using inductively coupled plasma mass spectrometry (ICP-MS) (Beauchemin and Kisilevsky 1998). Affinity purified PrP from all sCJD cases combined indicated a significantly decrease in bound Cu cation of between 70 and 90% as compared with PrPC purified from controls (Table 1). Furthermore, PrP purified from sCJD cases carried significant amounts of Mn and Zn whereas these metals were undetectable in PrPC from control brains(Table 1). When the sCJD cases were grouped either according to the PrPSc type or the 129 genotype, we observed levels of bound Cu significantly lower in the preparation containing PrPSc type-1 than in that containing PrPSc type-2 (Table 1). A significant difference was also found between the VV and MM sCJD groups (Table 1). Mn level was especially high in the MM sCJD group, and was significantly different from the VV group (Table 1). Similar difference in Mn level between preparation containing PrPSc type-1 and those containing PrPSc type-2 was also observed. Zn level was significantly increased in all groups as compared with the controls, but when the individual sCJD groups were compared, only the PrPSc type-2 group showed significantly higher Zn levels than the PrPSc type-1 group.

Table 1.   Copper, zinc and manganese divalent cations associated with affinity purified PrP from the frontal cortex of control and sporadic CJD (sCJD) variants as determined by the prion protein genotype (MM, VV) and by the scrapie prion protein type-1and 2
 CuZnMn
  1. Values determined by ICP-MS are given as µg cation per mg of protein. All labware was acid-leached overnight under pressure to minimise residual metal contaminants. Cu, Zn and Mn cations were undetectable on 8H4-coupled Sepharose beads not mixed with brain homogenates. Data are mean ± SEM of three independent measurements for each sample. Zn and Mn cations were undetectable on immuno-purified PrP from control preparation. a,dValues significantly different from control (Student's t-test, dp < 0.01; ap < 0.006); b,cType-1 values significantly different from type-2 values, and MM values significantly different from VV values (Student's t-test, cp < 0.05; bp < 0.01).

Control (n = 3)13.53 ± 0.68UndetectableUndetectable
sCJD (n = 9)2.41 ± 0.51a0.64 ± 0.08a4.80 ± 0.42a
Type-1 (MM and VV) (n = 4)1.23 ± 0.46a,b0.47 ± 0.03a,c5.63 ± 0.99c,d
Type-2 (MM and VV) (n = 5)3.35 ± 0.53a0.77 ± 0.12a4.15 ± 0.86a
VV (type-1 and -2) (n = 4)3.38 ± 0.77a0.61 ± 0.11d3.15 ± 0.32a
MM (type-1 and -2) (n = 5)1.63 ± 0.51a,c0.66 ± 0.14d6.13 ± 0.54a,b

In summary, Cu bound to PrP was most severely reduced in the sCJD associated with PrPSc type-1. In contrast, Zn was most increased in the sCJD associated with PrPSc type-2 and the level of bound Mn was the highest in the sCJD MM homozygous cases.

Metal concentrations in brain homogenates

We next determine whether the altered metal binding by purified PrP in sCJD variants was associated with perturbations in the metals concentration in the frontal cortex. ICP-MS analysis showed that Cu cations were significantly decreased by up to 50% while Mn cations were markedly increased by up to 10-fold in the brain homogenates from all sCJD cases combined as compared with controls (Table 2). When all the sCJD cases were grouped according to the 129 genotype regardless of the PrPSc type, the MM cases showed a significantly more severe decrease in Cu level and a more prominent increase in Mn level than the VV cases (Table 2). Both groups were also significantly different from controls (Table 2). In contrast, when the sCJD cases were grouped according to the PrPSc type, no significant difference in Cu and Mn levels was observed between the PrPSc type-1 and type-2 cases, although again both groups significantly differed from controls in the level of these metals (Table 2). No significant difference in the Zn levels was observed between all sCJD cases combined and controls. The Zn levels were significantly different between the sCJD cases type-1 and 2 but not between the MM and VV cases.

Table 2.   Copper, zinc and manganese cations in the frontal cortex from brains of control and variants of sporadic CJD (sCJD) as determined by prion protein genotype (MM, VV) and scrapie prion protein type-1and 2
 CuZnMn
  1. Values determined by ICP-MS are given as µg cation per g of brain, wet weight. All labware was acid-leached overnight under pressure to minimize residual metal contaminants. Background measurements without homogenate were determined and subtracted from actual readings. Data are mean ± SEM of three independent measurements for each sample. a,cValues significantly different from control (Student's t-test, cp < 0.03; ap < 0.003); bType-1 values significantly different from type-2 values, and MM values significantly different from VV cases values (Student's t-test, p < 0.003).

Control (n = 3)6.44 ± 0.1814.41 ± 0.150.51 ± 0.06
sCJD (n = 9)4.23 ± 0.33a14.23 ± 0.924.08 ± 0.43a
Type-1 (MM and VV) (n = 4)3.71 ± 0.33a11.50 ± 0.22a,b4.70 ± 0.40a
Type-2 (MM and VV) (n = 5)4.65 ± 0.18c16.41 ± 0.15a3.58 ± 0.70c
MM (type-1 and -2) (n = 5)3.49 ± 0.10a,b13.43 ± 1.665.06 ± 0.10a,b
VV (type-1 and -2) (n = 4)5.16 ± 0.26c15.23 ± 1.072.85 ± 0.31a

Anti-oxidation activity of PrP

Since the metal perturbations that we have observed in sCJD may impair the antioxidation function associated with PrPC (Brown et al. 1999, 2000), we therefore measured the SOD-like activity of affinity purified PrP (Brown et al. 2001) from the sCJD preparations grouped according either to the PrP genotype or PrPSc type using the NBT-based SOD assay (Brown et al. 1999). The SOD activity was markedly reduced in all PrP species purified from sCJD and represented only approximately 15% of the control values (Fig. 1c). The observed SOD-like activity was not due to the presence of either Cu/Zn-SOD or Mn-SOD as they were not immunodetected in the purified PrP preparations (data not shown). Significant difference was observed between the MM and VV cases, and also between preparations containing PrPSc type-1 and those containing PrPSc type-2. The SOD activity of PrP purified from sCJD cases associated with VV homozygousity and PrPSc type-2 are the most severely reduced.

Cu/Zn-SOD and Mn-SOD activity

We also assessed the activities of Cu/Zn-SOD and Mn-SOD, two other related antioxidation enzymes that are not associated with PrP, in the brain homogenates immunodepleted of PrP. The efficiency of the immunodepletion process has been demonstrated previously (Wong et al. 2000a). In the PrP-immunodepleted brain homogenates, there was no significant difference in the levels of Cu/Zn-SOD and Mn-SOD activities among the sCJD cases grouped according to the PrPSc type (Fig. 2a). When the sCJD cases were grouped by their codon 129 genotype, only PrP-depleted brain homogenates from the MM sCJD cases showed a significant reduction of the Cu/Zn-SOD activity as compared with the VV sCJD cases (Fig. 2a). In contrast, the Cu/Zn-SOD and Mn-SOD activity from sCJD samples were significantly different from those of controls. Cu/Zn-SOD activity was significantly reduced in the sCJD groups, while Mn-SOD activity was more than twofold higher in the sCJD groups as compared with control. These variations in activities were not due to increase expression of Cu/Zn-SOD or Mn-SOD, which was present in similar amounts in both sCJD and control brains (Fig. 2b).

image

Figure 2.  Copper/zinc- and manganese-superoxide dismutase activities and protein amounts in control and sporadic CJD (sCJD) variants brain homogenates after immunodepletion with anti-PrP. (a) Activities and (b) protein amounts of Cu/Zn-SOD and Mn-SOD in PrP-depleted brain homogenates using anti-PrP 8H4. (a) SOD activity was measured using 20 µg of brain homogenates. *Values for which there is a significant (Student's t-test, *p < 0.04, p < 0.005) difference between the respective sCJD variants, as determined by the prion protein genotype (white bars) of MM (n = 5) and VV (n = 4), and also by the scrapie prion protein type-1 (n = 4) and 2 (n = 5) (grey bars), with the control value (black bar, n = 3). There are also significant different between the two codon 129 genotypes of sCJD variants (Student's t-test, p < 0.001). All values represent the mean ± SEM (the positive and negative error bars are of the same value) of three assays for each sample. (b) Immunodetection was performed using 30 µg and 80 µg of brain homogenates for Cu/Zn-SOD and Mn-SOD, respectively. All samples were separated and immunoblotted on the same gel. Actin was immunoblotted to ensure similar gel loading of the starting preparation.

Download figure to PowerPoint

Oxidative stress in the brains

Since oxidative stress has been shown to be a pivotal event in the brains of scrapie-infected animals (Guentchev et al. 2000), we looked for similar evidence in the frontal cortex of sCJD subjects using two established markers for cellular oxidative damage: protein oxidation (Berlett and Stadtman 1997; Halliwell and Gutteridge 1999; Butterfield and Kanski 2001) and lipid peroxidation (Halliwell and Gutteridge 1999). The immunoreactivity of carbonyl groups was markedly increased in all sCJD samples compared with the controls (Fig. 3). The levels of two products of lipid peroxidation, malonaldehyde (MDA) (Fig. 4a) and 4-hydroxyalkenals (4-HNE) (Fig. 4b) assayed in each of the sCJD groups were double those of the controls. These results indicate that sCJD brains undergo oxidative stress, which is likely to be related to the marked decline of the antioxidation activity by the PrP (Fig. 1c).

image

Figure 3.  Increased protein oxidation in sporadic CJD. Reactive carbonyl groups present in the brain homogenates from control and sCJD variants. Reactive carbonyl groups were immunoblotted using a rabbit polyclonal antibody after derivatizing the samples with 2,4-dinitrophenylhydrazine (DNPH). Actin was immunoblotted to ensure similar gel loading of the starting preparation.

Download figure to PowerPoint

image

Figure 4.  Increased generation of lipid peroxidation end products in sporadic CJD.The lipid peroxidation end products (a) malonaldehyde (MDA) and (b) 4-hydrozyalkenals (4-HNE) were assay as described in Materials and methods using 10 mg of brain homogenates per reaction from control (black bar, n = 3) and sporadic CJD (sCJD) variants classified by the PrP genotype (white bars) MM (n = 5) and VV (n = 4), or PrPSc type-1 (n = 4) and 2 (n = 5) (grey bars). Data are mean ± SEM (the positive and negative error bar are of the same value) of three independent experiments for each sample. *Values for which there is a significant (Student's t-test, *p < 0.005, p < 0.007) difference between the respective CJD variants and corresponding control value.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Recent advances add new complexity to the pathogenesis of prion diseases. As originally proposed, the prion hypothesis postulates that a change in conformation leads to the conversion of PrPC to PrPSc, which then causes the disease (Prusiner 1998). Now it appears that metal-ions such as Cu and Mn, may contribute to the pathogenesis of prion diseases. It is well established that PrPC binds Cu (Brown et al. 1997; Cereghetti et al. 2001), but recent data suggest that PrPC binding of Cu is needed to induce conformational changes (Brown et al. 1999, 2001; Wong et al. 2000b) that enable PrPC to function as an antioxidant (Brown et al. 1999). On the other hand, the replacement of Cu by other metals such as Mn, not only causes the loss of this antioxidant activity but also increases PrP β-sheet content and resistance to proteases suggesting that these aberrations in metal binding change the conformation of PrP (Brown et al. 2000). According to Wadsworth et al. (1999) the conformation of some of the PrPSc species associated with specific CJD phenotypes could be affected by the level of metal binding in vitro. Therefore, these results collectively argue that abnormal metal binding by PrP may promote PrPSc formation. Furthermore, the level of metal occupancy might contribute to distinct conformations of PrPSc and the expression of different disease phenotypes. Since these altered PrP species lack SOD-like activity they are likely to expose brain cells to oxidative stress facilitating neurodegeneration.

Previous studies on the effects of metal binding on PrPC or PrPSc-like species have been carried out either in vitro (Brown et al. 2000) or are based on an indirect approach (Wadsworth et al. 1999). In this report, we directly measured and compared with controls the quantities of Cu, Mn and Zn cations as well as the SOD-like activities that are present in brain tissues or are directly associated with affinity purified PrP (PrPC, PrPSc and possibly other abnormal PrP species) obtained from human subjects affected by the major phenotypic variants of sCJD. We found a striking decrease of bound Cu and quantities of Mn cations exceeding those of Cu in the purified PrP preparations from sCJD subjects as compared with controls, while Mn cations were undetectable in PrPC preparations from controls. Similarly, Zn cations were detectable only on PrP preparations isolated from sCJD subjects. These changes were also accompanied by striking alterations in the brain metals concentrations; a significant decrease of Cu cations and a dramatic increase of Mn cations in brain tissues from sCJD subjects. These metal binding aberrations were associated with an ∼ 85% diminution of the SOD-like activity of PrP present in sCJD. The present data suggest that in sCJD, Mn and perhaps Zn replace Cu in binding to the total PrP population (PrPC, PrPSc and related abnormal PrP species), and that this replacement leads to the loss of the SOD-like activity associated with PrPC found in non-CJD. They also support the suggestion that metal binding aberrations may play a role in the pathogenesis of prion diseases by causing the loss of the antioxidation function associated with PrPC. The sizable increase of markers associated with oxidative damage, such as protein oxidation and lipid peroxidation end products, also argue for a failure of antioxidation activities in sCJD. Currently, oxidative stress is thought to be a pivotal event in several neurodegenerative diseases although it is likely that the mechanisms leading to it are different in the various diseases (Bush 2000; Butterfield and Kanski 2001). Indeed, to our knowledge the elevation in Mn that we showed to be so prominent in brain tissue and PrP preparations from subjects with sCJD has not been reported in other neurodegenerative diseases. Alterations in brain Cu and Zn (Deibel et al. 1996) but not Mn (Markesbery et al. 1984) have been observed in Alzheimer's disease. Moreover, the level of Mn does not change significantly during aging (Markesbery et al. 1984).

The alterations in metal binding we observed were not uniformly distributed among the different phenotypic variants of sCJD as determined by the genotype at codon 129 of the PrP gene and the PrPSc type. In brain tissue, the changes in Cu and Mn cations were more prominent in the variants associated with MM homozygosity than with VV homozygous variants while no definite difference was detected according to the PrPSc type. In contrast, Cu was decreased significantly more in purified preparations containing PrPSc type-1 than in those with PrPSc type-2. However, Mn increase was more prominent in MM homozygous cases. Combined, our data show that MM homozygosity at codon 129 and the presence of type-1 PrPSc are the two conditions associated with the most pronounced changes in Cu and Mn. Accordingly, the sCJD variant MM type-1, which corresponds to the classical and by far the most common CJD subtype (Parchi et al. 1999), has the most prominent Cu and Mn variations. Zn cations appear to distribute differently, since in brain tissue they are significantly decreased in cases carrying PrPSc type-1 and increased in the PrPSc type-2 cases, while in PrP preparations they are increased in all preparations but significantly more in PrPSc type-2 than type-1 preparations.

Wadsworth et al. (1999) reported that the gel migration characteristics of certain PrPSc species, hence their conformation, depend on their Cu and Zn ion occupancy. This was based on the observation that chemically chelating for these two metal-ions affected the conformation of PrPSc, hence their susceptibility toward proteinase K digestion. Although it is not easy to compare our data with those by Wadsworth et al. (1999) because of differences in our classification of human CJD (Parchi et al. 1997), data from both groups collectively point to variations in metal-ions occupancy as a possible determinant of PrP and phenotypic diversities in prion diseases.

In conclusion, we have shown that major alterations in the distribution and PrP association of metal-ions occur in sCJD; these changes may play a pivotal role in the pathogenesis of prion diseases as they lead to the loss of antioxidant function associated with PrPC and may promote the diversity of PrPSc and other abnormal PrP species. Whether the metal imbalance is a contributory cause or a mere consequence of PrPC conversion to abnormal isoforms remains to be determined. Finally, this study also presented two potential approaches in the diagnosis of human CJD; the significant increase of Mn in the frontal cortex makes it potentially detectable by clinical magnetic resonance imaging because of its paramagnetic property, and the binding of Mn by PrP in sCJD, but not in the control, might represent a novel diagnostic marker.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank physicians across United States for depositing brain samples to the National Prion Disease Pathology Surveillance Center. We also like to thank Dr Lawrence Sayre (Chemistry, CWRU) for enlightening discussion and Dr Neil Greenspan (Pathology, CWRU) for critical reading of the manuscript. DRB was supported by a fellowship from BBSRC. This work was supported in part by grants from National Institutes of Health, Center for Disease Controls, Britton Fund and UK BBSRC.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Beauchemin D. & Kisilevsky R. (1998) A method based on ICP-MS for the analysis of Alzheimer's amyloid plaques. Anal. Chem. 70, 10261029.DOI: 10.1021/ac970783f
  • Berlett B. S. & Stadtman E. R. (1997) Protein oxidation in aging, disease, and oxidative stress. J. Biol. Chem. 272, 2031320316.
  • Brown D. R., Wong B. S., Hafiz F., Clive C., Haswell S., Jones I. M. (1999) Normal prion protein has an activity like that of superoxide dismutase. Biochem. J. 344, 15.DOI: 10.1042/0264-6021:3440001
  • Brown D. R., Qin K., Herms J. W., Madlung A., Manson J., Strome R., Fraser P. E., Kruck T., Von Bohlen A., Schulz-Schaeffer W., Giese A., Westaway D., Kretzschmar H. (1997) The cellular prion protein binds copper in vivo. Nature 390, 684687.DOI: 10.1038/37783
  • Brown D. R., Hafiz F., Glasssmith L. L., Wong B. S., Jones I. M., Clive C., Haswell S. J. (2000) Consequences of manganese replacement of copper for prion protein function and proteinase resistance. EMBO J. 19, 11801186.
  • Brown D. R., Clive C., Haswell S. J. (2001) Anti-oxidant activity related to copper binding of native prion protein. J. Neurochem. 76, 6976.DOI: 10.1046/j.1471-4159.2001.00009.x
  • Bush A. I. (2000) Metals and neuroscience. Curr. Opin. Chem. Biol. 4, 184191.
  • Butterfield D. A. & Kanski J. (2001) Brain protein oxidation in age-related neurodegenerative disorders that are associated with aggregated proteins. Mech. Ageing Dev. 122, 945962.
  • Caughey B. (2000) Transmissible spongiform encephalopathies, amyloidoses and yeast prions: common threads? Nat. Med. 6, 751754.DOI: 10.1038/77476
  • Cereghetti G. M., Schweiger A., Glockshuber R., Van Doorslaer S. (2001) EPR evidence for binding of Cu2+ to the C-terminal domain of the murine prion protein. Biophys. J. 81, 516521.
  • Collinge J. (2001) Prion diseases of humans and animals: their causes and molecular basis. Annu. Rev. Neurosci. 24, 519550.
  • Deibel M. A., Ehmann W. D., Markesbery W. R. (1996) Copper, iron, and zinc imbalances in severely degenerated brain regions in Alzheimer's disease: possible relation to oxidative stress. J. Neurol. Sci. 143, 137142.
  • Goldfarb L. G., Petersen R. B., Tabaton M., Brown P., LeBlanc A. C., Montagna P., Cortelli P., Julien J., Vital C., Pendelbury W. W. and et al. (1992) Fatal familial insomnia and familial Creutzfeldt-Jakob disease: disease phenotype determined by a DNA polymorphism. Science 258, 806808.
  • Guentchev M., Voigtlander T., Haberler C., Groschup M. H., Budka H. (2000) Evidence for oxidative stress in experimental prion disease. Neurobiol. Dis. 7, 270273.DOI: 10.1006/nbdi.2000.0290
  • Halliwell B. & Gutteridge J. M. C. (1999). Free Radicals in Biology and Medicine. Oxford University Press, New York.
  • Lasmezas C. I., Deslys J. P., Robain O., Jaegly A., Beringue V., Peyrin J. M., Fournier J. G., Hauw J. J., Rossier J., Dormont D. (1997Transmission of the BSE agent to mice in the absence of detectable abnormal prion protein. Science 275, 402405.DOI: 10.1126/science.275.5298.402
  • Li R., Liu T., Wong B. S., Pan T., Morillas M., Swietnicki W., O'Rourke K., Gambetti P., Surewicz W. K., Sy M. S. (2000) Identification of an epitope in the C terminus of normal prion protein whose expression is modulated by binding events in the N terminus. J. Mol Biol. 301, 567573.DOI: 10.1006/jmbi.2000.3986
  • Manson J. C., Jamieson E., Baybutt H., Tuzi N. L., Barron R., McConnell I., Somerville R., Ironside J., Will R., Sy M. S., Melton D. W., Hope J., Bostock C. (1999) A single amino acid alteration (101L) introduced into murine PrP dramatically alters incubation time of transmissible spongiform encephalopathy. EMBO J. 18, 68556864.
  • Markesbery W. R., Ehmann W. D., Hossain T. I., Alauddin M. (1984) Brain manganese concentrations in human aging and Alzheimer's disease. Neurotoxicology 5, 4957.
  • Monari L., Chen S. G., Brown P., Parchi P., Petersen R. B., Mikol J., Gray F., Cortelli P., Montagna P., Ghetti B. and et al. (1994) Fatal familial insomnia and familial Creutzfeldt–Jakob disease: different prion proteins determined by a DNA polymorphism. Proc. Natl Acad. Sci. USA 91, 28392842.
  • Parchi P., Capellari S., Chen S. G., Petersen R. B., Gambetti P., Kopp N., Brown P., Kitamoto T., Tateishi J., Giese A., Kretzschmar H. (1997) Typing prion isoforms. Nature 386, 232234.
  • Parchi P., Giese A., Capellari S., Brown P., Schulz-Schaeffer W., Windl O., Zerr I., Budka H., Kopp N., Piccardo P., Poser S., Rojiani A., Streichemberger N., Julien J., Vital C., Ghetti B., Gambetti P., Kretzschmar H. (1999) Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects. Ann. Neurol. 46, 224233.DOI: 10.1002/1531-8249(199908)46:2<224::aid-ana12>3.3.co;2-n
  • Parchi P., Zou W., Wang W., Brown P., Capellari S., Ghetti B., Kopp N., Schulz-Schaeffer W. J., Kretzschmar H. A., Head M. W., Ironside J. W., Gambetti P., Chen S. G. (2000) Genetic influence on the structural variations of the abnormal prion protein. Proc. Natl Acad. Sci. USA 97, 1016810172.
  • Prusiner S. B. (1998) Prions. Proc. Natl Acad. Sci. USA 95, 1336313383.
  • Telling G. C., Parchi P., DeArmond S. J., Cortelli P., Montagna P., Gabizon R., Mastrianni J., Lugaresi E., Gambetti P., Prusiner S. B. (1996) Evidence for the conformation of the pathologic isoform of the prion protein enciphering and propagating prion diversity. Science 274, 20792082.DOI: 10.1126/science.274.5295.2079
  • Wadsworth J. D. F., Hill A. F., Joiner S., Jackson G. S., Clarke A. R., Collinge J. (1999) Strain-specific prion-protein conformation determined by metal ions. Nat. Cell Biol. 1, 5559.DOI: 10.1038/9030
  • Whittal R. M., Ball H. L., Cohen F. E., Burlingame A. L., Prusiner S. B., Baldwin M. A. (2000) Copper binding to octarepeat peptides of the prion protein monitored by mass spectrometry. Protein Sci. 9, 332343.
  • Wong B. S., Pan T., Liu T., Li R. L., Gambetti P., Sy M. S. (2000a) Differential contribution of superoxide dismutase activity by prion protein in vivo. Biochem. Biophys. Res. Commun. 273, 136139.DOI: 10.1006/bbrc.2000.2911
  • Wong B. S., Venien-Bryan C., Williamson R. A., Burton D. R., Gambetti P., Sy M. S., Brown D. R., Jones I. M. (2000b) Copper refolding of prion protein. Biochem. Biophys. Res. Commun. 276, 12171224.DOI: 10.1006/bbrc.2000.3604
  • Wong B. S., Liu T., Li R. L., Pan T., Petersen R. B., Smith M. A., Gambetti P., Perry G., Manson J. C., Brown D. R., Sy M. S. (2001) Increased levels of oxidative stress markers detected in the brains of prion knock-out mice. J. Neurochem. 76, 565572.DOI: 10.1046/j.1471-4159.2001.00028.x
  • Zanusso G., Liu D., Ferrari S., Hegyi I., Yin X., Aguzzi A., Hornemann S., Liemann S., Glockshuber R., Manson J. C., Brown P., Petersen R. B., Gambetti P., Sy M. S. (1998) Prion protein expression in different species: analysis with a panel of new mAbs. Proc. Natl Acad. Sci. USA 95, 88128816.
Footnotes
  1. 1Present address of M. Colucci is the Department of Neurological Sciences and Vision, University of Genoa, Genoa, Italy.