| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Division of Neuropathology, Department of Pathology, Institute of Pathology, Case Western Reserve University, Cleveland, Ohio
 |
Abstract |
|---|
 Top Abstract IntroductionExperimental ProceduresResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionReferences |
|---|
-helical content of
PrPC is diminished and the amount of ß-sheets
increases, resulting in the formation of an abnormal
PrPC isoform, called PrP scrapie or
PrPSc. PrPSc is resistant
to proteases, insoluble in nonionic detergents, and is not released
from the cell surface by
phosphatidylinositol-specific
phospholipase C (PI-PLC).10-12 Prion diseases comprise a sporadic, idiopathic form and forms that are genetically determined or transmitted by an infectious mechanism. Genetic prion diseases are linked to mutations in the gene encoding PrPC, PRNP, and are inherited as three major autosomal dominant phenotypes: familial Creutzfeldt-Jakob disease (CJD), Gerstmann-Sträussler Scheinker disease, and fatal familial insomnia.13 The most common of the human PRNP mutations occurs at codon 200 and results in the substitution of glutamic acid with lysine (E200K) in PrP.14 The E200K mutation is linked to a disease phenotype that resembles that of the typical sporadic CJD, the most common human prion disease.15 Although the presence of the PRNP E200K mutation increases the probability of developing CJD from 1:1 million, the prevalence of the sporadic form, to more than 1:1.1, the penetrance of the E200K mutation,15 the carriers of the mutation remain asymptomatic for several decades.15 Therefore, the changes caused by the E200K mutation in the mutant PrP (PrPM) make the conversion of PrPM into PrPSc almost inevitable, but the disease becomes clinically detectable only after a long incubation time. These findings raise important questions concerning the nature and, above all, the timing of the mutation-related changes that promote the conversion of PrPM into PrPSc and the beginning of the disease. In a series of studies, it has been proposed that PRNP mutations per se cause the PrPM to transform in a PrPSc-like isoform soon after its synthesis, suggesting that the long incubation time of the disease results from a slow rate of accumulation of this isoform.16-18 However, this issue remains controversial.15-19
In the present study, we investigated the effects of the PRNP E200K mutation on the metabolism of PrPM in human neuroblastoma cells and we demonstrated several abnormal features of PrPM, such as an abnormal glycosylation, an increased formation of truncated fragments, and a partial insolubility and increased resistance to digestion with proteinase K (PK). Then, we looked for these abnormal features in the PrP extracted from brains of patients affected by the E200K subtype of familial CJD. Our results demonstrate that several posttranslational changes are produced by the E200K mutation and are shared by the cell model and the E200K CJD-affected brains. However, basic characteristics of the PrPSc present in the E200K brains are not reproduced by the cell model, suggesting that although the E200K mutation renders PrPM susceptible to conversion into PrPSc, the conversion requires additional modifications of the protein to occur.
|
Experimental Procedures |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionReferences |
|---|
The following antibodies were used: anti-N, a rabbit antiserum to a synthetic peptide corresponding to human PrP residues 23 to 40 (B. Ghetti, Indiana University, Indianapolis, IN); 3F4, a monoclonal antibody that recognizes human PrP residues 109 to 112;20 anti-C, a rabbit antiserum to synthetic human PrP residues 220 to 231;21 and 8H4 a monoclonal antibody whose epitope is located within the 145 to 220 sequence.22
Cell Lines
The human neuroblastoma cell line M-17 BE(2)C (kindly provided by B. Spengler and J. Biedler, Memorial Sloan-Kettering Cancer Center, New York, NY), which does not express PrP,23 was transfected with the episomal vector Cep4ß containing the human PrP coding sequence under the control of the cytomegalovirus promoter and the hygromycin B resistance gene for selection. The PrP coding sequence, obtained from genomic human DNA, was cloned into the bacterial plasmid pVZ1 and oligonucleotide-directed mutagenesis was used to create the mutant PrP coding sequence (Bio-Rad Muta-Gene phagemid in vitro mutagenesis kit; Bio-Rad, Richmond, CA).23 . The following cell lines were used: control/129M or C, expressing normal PrP, bearing a methionine at codon 129, or mutant at codon 200 with either methionine (E200K/129M) or valine (E200K/129V) at codon 129. Moreover, cell lines with PrP mutated at codon 181 or 199, either combined or not with the E200K mutation (N181Q/129M; N181Q/129M/E200K; T199A/129M; T199A/129M/E200K), were constructed. Transfected cells were grown as bulk-selected, hygromycin-resistant cultures.23 Multiple independent transfections were used to avoid selection bias. For each experiment cells were detached with trypsin, counted, and an identical number of cells were seeded onto 10-cm plates and grown overnight to ~95% confluence.
Patients and Tissues
Four patients carrying the E200K mutation were studied. All patients were clinically affected and died after a duration of symptoms ranging from 4 to 18 months. Tissue was obtained at autopsy in three patients and from a biopsy in the fourth.
Frozen tissue from the cerebral cortex and cerebellum was used for the biochemical studies. A semiquantitative evaluation of spongiosis, neuronal loss, and gliosis was carried out in the same brain regions sampled for the biochemical studies.24 The histopathology was rated as follows: a, minimal where only minimal gliosis was present; b, intermediate where spongiosis and gliosis were mild to moderate; and c, severe where the spongiosis and astrogliosis were moderate to severe and neuronal loss was visually detectable.
Preparation of Samples
Whole Cell Proteins
Cells were washed three times with cold phosphate-buffered saline and lysed in ice-cold lysis buffer (100 mmol/L NaCl, 10 mmol/L ethylenediaminetetraacetic acid, 0.5% Nonidet P-40, 0.5% Na-deoxycholate, 10 mmol/L Tris, pH 7.4, 1 mmol/L phenylmethyl sulfonyl fluoride, and 10 mg/ml each of leupeptin, antipain, pepstatin). Nuclei and large debris were removed by centrifugation at 690 x g for 10 minutes at 4°C. The supernatant was precipitated with 4 volumes of methanol at -20°C overnight.
Surface Proteins (Released by PI-PLC)
Cells were washed twice and then incubated in serum-free Opti-MEM (Life Technologies, Inc., Grand Island, NY) containing 59 ng/ml PI-PLC14 for 30 minutes at 37°C. The medium was removed, centrifuged at 290 x g at 4°C for 10 minutes and methanol precipitated.
Brain Tissue
Gray matter brain samples were obtained from frozen brains of E200K-affected patients and age-related controls.24 From each brain sample ~100 mg of tissue was homogenized in 9 volumes of lysis buffer and aliquots equivalent to 0.3 mg of wet tissue were used for PK digestion.24 All tissue preparations were carried out at 4°C.
Western Blots
Protein samples (brain tissue equivalent to 0.3 mg of wet tissue or lysate from ~25,000 cells, double quantity for surface PrP) were resuspended in sample buffer (6% sodium dodecyl sulfate [SDS], 5% ß-mercaptoethanol, 4 mmol/L ethylenediaminetetraacetic acid, 20% glycerol, 125 mmol/L Tris, pH 6.8) and boiled for 10 minutes before loading. Protein samples were separated in 12, 14, or 16% SDS-polyacrylamide gel (37.5:1 acrylamide: bis-acrylamide) or in 10% Tris-16.5% Tricine gels.25 Proteins were transferred to Immobilon P (Millipore Corp., Bedford, MA) for 2 hours at 60 V, blocked with 10% nonfat milk in Tris-buffered saline, pH 7.5, and probed with the appropriate antibody. The immunoreactivity was visualized by enhanced chemiluminescence (ECL; Amersham, Arlington Heights, IL) on Kodak X-Omat film (Eastman Kodak, Rochester, NY) and quantified using a computer-assisted densitometric scanner.24 Data analysis was performed using Excel 5 (Microsoft).
Pulse Chase
Cells were washed and pre-incubated for 30 minutes at 37°C with methionine-deficient MEM (def MEM; ICN Biomedicals, Irvine, CA). A pulse with 0.5 mCi 35S-translabel (ICN) in 3 ml of def MEM was followed by washing with Opti-MEM and incubation at 37°C in the same media for the different chase points. When indicated PI-PLC treatment was performed by incubating the cells in Opti-MEM + PI-PLC for the last 30 minutes of chase at 37°C. Medium was collected and cells were lysed at different time points.
Pulse Chase with Inhibitors
Plated cells were pre-incubated for 30 minutes with the inhibitor, then pulsed and chased as above in the presence of the drug. For each inhibitor the lowest effective concentration was empirically determined. Inhibitors’ concentrations were as follows: 2 mg/ml tunicamycin (Boehringer Mannheim, Mannheim, Germany), 2 mmol/L dithiothreitol, 50 mmol/L Swainsonine (Oxford Glycosystem).
Immunoprecipitation
Medium, PI-PLC-released proteins, and cell lysates were prepared as described above. The postcentrifugation supernatant was immunoprecipitated with the appropriate antibody in 1% bovine serum albumin, 0.1% N-lauryl sarcosine, 0.1 mmol/L phenylmethyl sulfonyl fluoride by rocking at 4°C overnight. Protein-antibody complexes were bound to protein A Sepharose beads. The beads were washed 6 times in 1 ml of ice-cold wash buffer (150 mmol/L NaCl, 10 mmol/L Tris, pH 7.8, 0.1% N-lauryl sarcosine with 0.1 mmol/L phenylmethyl sulfonyl fluoride), resuspended in sample buffer, and boiled to release the bound proteins. After protein separation by SDS-polyacrylamide gel electrophoresis, the gels were fixed by soaking in methanol:acetic-acid:water (40:10:50) for 15 minutes, dehydrated in dimethylsulphoxide for 1 hour, and enhanced by rocking the gels in 2,5-diphenyloxazole/dimethylsulphoxide (22%) for 90 minutes, followed by precipitation of the 2,5-diphenyloxazole with water. Gels were dried, exposed to film, and analyzed as for Western blots (see above).
Endoproteinase Lys-C Digestion
35S-labeled PrP was extracted from SDS gels, denatured in 6 mol/L guanidine hydrochloride in 50 mmol/L Tris-HCl, pH 8, reduced with 2 mmol/L dithiothreitol, carboxy-methylated with 6 mmol/L Na-iodoacetate, and precipitated with 10 volumes of ethanol at -20°C. The pellet was resuspended in 0.01% SDS, 1 mmol/L ethylenediaminetetraacetic acid, 25 mmol/L Tris-HCl, pH 8.5, and digested overnight at 37°C.21
PNGase-F, Endoglycosidase-H Digestion
Proteins were precipitated in 4 volumes of methanol, resuspended in denaturing buffer (0.5% SDS, 1% ß-mercaptoethanol), boiled for 10 minutes, and treated with PNGase-F (New England Biolabs, Beverly, MA) in 1% Nonidet P-40, 50 mmol/L sodium-citrate, pH 7.5, or with endoglycosidase-H (New England Biolabs) in 50 mmol/L sodium-citrate, pH 5.5, overnight at 37°C.
Detergent Solubility Test
To determine detergent solubility the tissues were lysed in 9 volumes of lysis buffer and spun at 690 x g for 10 minutes at 4°C. The supernatants were centrifuged at 100,000 x g for 1 hour to obtain the detergent-soluble -S2- (supernatant) and detergent-insoluble -P2- (pellet) fractions. Both fractions were methanol-precipitated and resuspended in the same volume of buffer.
PK Digestion
Brain homogenates were digested with 100 µg/ml PK for 1 hour at 37°C.24 Cells lysates were digested with 3.3 or 5 µg/ml of PK (Boehringer-Mannheim) for 10 minutes at 37°C. The reaction was terminated by the addition of phenylmethyl sulfonyl fluoride to a final concentration of 3 mmol/L.
|
Results |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionReferences |
|---|
After a 3-minute pulse and immunoprecipitation with the 3F4
antibody, both PrPM and
PrPC migrated as three well-defined bands (Figure 1A
, lanes 1 and 2). The upper two bands
have been shown to be the precursors (PH, PI) of the diglycosylated or
high (H) and of the monoglycosylated or intermediate (I) mature PrP
forms, whereas the lowest band contains the unglycosylated (U)
form.23,26
No difference was detected between
PrPM and PrPC preparations.
Therefore, early stages of PrP synthesis and posttranslational
modification seem to be primarily unaffected by the mutation. However,
at increasing chase times, during which in PrPC
the H and I precursors undergo processing of the glycans and attain the
mature migration pattern, the PrPM H form
differed in gel mobility from HPrPC (Figure 1A
,
lanes 5 and 6). HPrPM migrated as an ill-defined
smear of 31 to 40 kd, as opposed to
HPrPC that migrated at 33 to 42 kd, suggesting
that the maturation of the glycans is abnormal in
PrPM (Figure 1A)
.
|
. The UPrPM released after a 4-hour
chase accounted for 1 ± 0.2% of the total
UPrPM form at the 0 chase time, whereas
UPrPC accounted for 4 ± 0.6%
(P < 0.01, n = 3).
The three full-length PrP glycoforms are known to have
N-terminal-truncated forms that are generated by cleavage at residues
111 to 112 after re-internalization from the plasma
membrane.28,29
We examined these forms by sequential
double-immunoprecipitation using the 3F4 antibody to eliminate the
full-length forms followed by recovery of the N-terminally truncated
fragments by using the anti-C antibody. The truncated
PrPM and PrPC glycoforms
were visualized as 25- to 30-kd and 28- to 33-kd bands for the H forms
and as 20- to 23-kd and 20- to 25-kd bands for the I forms,
respectively, whereas the U form migrated at 18 kd in both preparations
(Figure 1C)
. The truncated HPrPM showed a faster
migration than the truncated HPrPC, as was found
in the corresponding full-length forms. In addition, the unglycosylated
18-kd PrPM peptide was preferentially
underrepresented (Figure 1C)
. Additional PrP fragments were seen after
immunoprecipitation (see below). 3F4 revealed a 20-kd fragment, whereas
the anti-C antibody detected 20-kd and 12-kd fragments. The 12-kd
fragment became detectable in the intracellular compartment only after
a 3-hour chase and was more abundant in the E200K cell preparations
than in the controls (Figure 1C)
. In conclusion, analyses of
PrPM by metabolic labeling show three major and
consistent changes: 1) presence of abnormal glycans in the H form; 2)
underrepresentation of the U form at the cell surface in both the
full-length and truncated forms; and 3) increase in quantity of the
PrPM fragments. We examined these three changes
in more detail.
Abnormal Glycosylation of the PrPM H Form
After removal of the glycans with PNGase-F,
PrPM and PrPC display
similar gel mobility, confirming that the glycans are the cause of the
difference between the two H forms (Figure 2A)
. We then examined whether the change
affects the glycans at one or both glycosylation sites. First, we
treated the PrP preparations with the endoproteinase Lys-C, which
generates a fragment containing only the 181-glycosylation site, and
found no difference between PrPM and
PrPC (Figure 2B)
. Second, we used N181Q and T199A
glycosylation knock-out mutants with or without the E200K substitution,
and demonstrated a difference in mobility between the
PrPM and PrPC I forms only
in the 181-glycan knock-out (N181Q) mutant (Figure 2C)
. Therefore, only
the glycan attached to residue 197 is aberrant.
|
.30
The blockade of the
3-6 mannosidase II by Swainsonine, which prevents the removal of
3-6 mannose residues after the addition of the first
N-acetylglucosamine in the medial region of the Golgi
apparatus, was the first step at which the two H forms differed (Figure 2, E and D
; step 3). Thus in PrPM, the 197 glycan
begins to differ from the corresponding PrPC
glycan at the step in which normally one N-acetylglucosamine
molecule is added (Figure 2D
, step 3). Underrepresentation of the U Form of PrPM at the Cell Surface
After treatment with tunicamycin, which prevents glycosylation,
all of the PrP produced by the cell is in the U form.31
After a 2-hour chase, we observed that, of the total PrP produced at
time 0, the amount of PrPC that is left exceeded
that of PrPM by 32% intracellularly and by 46%
at the cell surface (data not shown), indicating that the preferential
decrease of UPrPM at the cell surface is not
because of hyperglycosylation. Other mechanisms that may account for
this decrease are the preferential degradation of the
UPrPM or its preferential aggregation, with
consequent epitope masking and inefficient immunoprecipitation. In the
pulse-chase experiments, the UPrPM obtained
with immunoprecipitation with the 3F4 antibody is reduced compared to
UPrPC (Figure 1, B and C)
. In contrast, by
immunoblotting the supernatant after immunoprecipitation, we detected a
higher amount of residual PrPM, especially of the
U form, compared to PrPC (Figure 3)
. Because comparable amounts of PrP
were detected in immunoblots of the cell lysate (Figures 5C and 6D)
, a
finding that suggests that the antibody does not have a different
affinity for PrPM or PrPC,
it seems that aggregation is the immediate cause of the
UPrPM being underrepresented in the
immunoprecipitate.32
However, the aggregated
UPrPM is not increased compared to
UPrPC in immunoblots of total cell lysates, hence
the UPrPM does not accumulate in intracellular
compartments (Figures 5C and 6D)
.
|
|
|
Immunoblot analysis confirmed the presence of 20-kd and 12-kd PrP
fragments, in addition to the 18-kd peptide (Figure 4)
. The 20-kd band, present in both
mutant and control cell lines, was formed by two PrP fragments similar
in length but truncated at different sites. The first corresponded to
the N-terminal 20-kd band that was previously described.23
This peptide, which reacts with the anti-N antibody, and, therefore,
lacks the C-terminus, was equally represented in the mutant and control
cell lysates at 0 chase time (Figure 2)
. In contrast, the second 20-kd
fragment, recognized by the anti-C, 8H4, and 3F4 antibodies, but not by
the anti-N antibody, was overrepresented in PrPM
preparations compared to PrPC (10.1 ± 4
versus 4.5 ± 2.2, P < 0.01,
n = 4) (Figure 4)
. This C-terminal fragment displayed a
glycoform ratio similar to that of full-length surface PrP (Figure 1C)
and appeared after a 1-hour chase, suggesting that it is
generated after re-internalization. Immunoblotting with the
anti-C antibody confirmed the increased amount of the 12-kd fragment in
the mutant compared to the control cells (Figure 4)
.
|
E200K PrPM has been reported to have some of the properties of the PrPSc, namely insolubility in nonionic detergents as well as partial resistance to PI-PLC and protease treatments.16-18 Thus, we assessed these properties in our PrPM preparations.
PrPM was recovered in significantly higher
amounts than PrPC in the detergent insoluble
fraction, P2, (22 ± 7% of total
PrPM versus 8 ± 2% of total
PrPC, P < 0.01,
n = 4) (Figure 5A)
. All
PrPM glycoforms were represented in the
aggregated fraction, but the U form was relatively overrepresented,
accounting for ~40% of the total aggregated
PrPM (Figure 5A)
. Except for the 18-kd fragment,
all fragments were more highly represented in the mutant cells. This
feature was especially pronounced in the 12-kd
PrPM fragment, of which ~20% was insoluble,
whereas no insoluble 12-kd fragment was present in control cells
(Figure 5B)
.
PI-PLC was significantly less effective in cleaving the anchor in the E200K than in the control cells (56% ± 6 of PrPM cleaved versus 71% ± 7 of PrPC, P < 0.01, n = 3) (data not shown).
The sensitivity to proteases was examined, as in previous
studies,19
by treating the cell lysates with 3.3 µg/ml
of PK at 37°C for 10 minutes (Figure 5C)
. Several fragments were
observed. After immunoreaction with the 3F4 antibody, a fragment
corresponding to the 20-kd C-terminal peptide described above was the
main isoform present. This fragment is close, in size, to the so-called
PrP27–30 generated after digestion with PK (50 to 100 µg/ml) of
affected brains, and was significantly more abundant in
PrPM than PrPC preparations
(25 ± 4.5% versus 2.8 ± 2.5%,
n = 3, P < 0.001) (Figure 5C)
.
Immunoblotting with the anti-C antibody showed the presence of the
12-kd fragment described above only in PrPM
preparations. Moreover, the 8H4, as well as the anti-C antibody (data
not shown for the last), showed in both PrPM and
PrPC preparations substantial amounts of an 18-kd
fragment, which is known to be generated by a cleavage at residues
111/112 (Figure 5, B and C)
.29
To test whether the
increased PrPM resistance to PK digestion was
simply the consequence of the increase in the aggregated form,
preparations from the E200K and control cells were normalized for the
content of the detergent insoluble PrP, and digested with PK (Figure 5D)
. Although under this condition, a higher PK concentration was
required for PrPC to be digested, the resistance
of PrPM to digestion remained significantly
higher. Therefore, the PrPM resistance to PK
digestion in not simply because of PrPM-increased
aggregation, but it is likely to be because of newly acquired
properties of the mutant protein.
PrP Properties in E200K CJD Brains
To assess the relevance of the alterations observed in the cell
model to the corresponding human disease, we compared the cellular
PrPM with the total PrP and with
PrPSc extracted from brains of patients affected
by the E200K subtype of CJD. Although the direct comparison between the
two systems is limited by the heterozygosity of the E200K mutation,
which results in the presence of both PrPM and
PrPC in the brain samples, we observed several
similarities (Table 1)
.
|
. This finding is
likely to result from both heterogeneity in gel mobility and reduction
in quantity of the U forms in the E200K brain. Indeed, the U form in
the total brain homogenate from an area with minimal pathology was
decreased by ~40% compared to the control brain (Figure 6A)
. In the
cell model, PrPM shows a slightly slower gel
migration of the PrPM U form
(UPrPM) compared to the control (Figures 5D
or
6D). Therefore, the brain findings concerning UPrP are consistent with
the results of the cell studies, which show that
UPrPM has reduced gel mobility and, especially at
the cell surface, is lower in amount than UPrPC.
The 20-kd C-terminal fragment that was increased in the mutant cells
was also increased in the affected brains (Figure 6B)
, where it
corresponded to the so-called "PrP27–30" fragment that is also
formed in vivo.29
In addition, the 12-kd
peptide that was increased in the mutant cells, accounted for ~13%
of the total PrP in the E200K CJD brains whereas it was not detected in
control brains. The presence of these truncated fragments in
preparations from brain biopsies excluded the possibility that they are
simply postmortem artifacts (Figure 6B)
.
It has been shown that in CJD and other human prion diseases, on
treatment with PK, PrPSc generates either one of
two major fragments, which have a relative molecular mass
(Mr) of 21 kd and 19 kd (Figure 6C)
and have been
designated type 1 and type 2, respectively.33-35
Although
the CJD patients carrying the more common E200K-129M haplotype examined
in this study form PrPSc type 1, those carrying
the rare E200K-129V haplotype form PrPSc type
2.36
Therefore, we generated a mutant cell line carrying
the E200K mutation coupled with valine rather than methionine at codon
129 (E200K-129V). The PrPM expressed by both the
E200K cell lines was similarly resistant to PK, and in both cell lines
it generated a PK-resistant PrPM fragment that
co-migrated at ~20 kd (Figure 6D)
. Therefore, the E200K cell models
do not reproduce the PrPSc dualism found in the
corresponding human diseases (Figure 6C)
.
|
Discussion |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionReferences |
|---|
Amino acid substitutions in flanking or adjacent regions are known to influence both the efficiency and type of glycosylation.37,38 However, this study shows that the E200K mutation does not affect glycosylation efficiency but rather selectively interferes with the modifications of the glycan chain at residue 197 which results in an enhanced gel migration of the highly glycosylated PrPM form. The mutation-related glycan change is first detected at the stage of N-acetylglucosamine addition, which takes place in the medial Golgi. A possible explanation for the higher gel mobility of the PrPM 197 glycan is that at this stage an increased number of glycans receive a bisecting N-acetylglucosamine molecule that cannot be extended further and consequently migrates faster on gel. The HPrPSc present in the affected E200K brains and the HPrPM recovered from fibroblasts of affected patients also show an increase in gel mobility that is comparable to what is observed in the cell model.39 These findings argue that in the corresponding human disease only the 197 PrPM glycan is changed in a manner similar to that of the cell model.
Although glycans are known to contribute to protein stability, it is
unlikely that the abnormal glycan attached to residue 197 significantly
increases the susceptibility of PrPM to convert
into PrPSc. In contrast, this may be because of
the other changes associated with the E200K mutation, such as the
increased aggregation and resistance to PI-PLC and PK treatments, which
may all be related to the misfolding and destabilizing effect of the
mutation on PrPM. A reasonable cascade of events,
which, according to our data, applies to both the cell model and the
brain, is that the increased instability and aggregation make
PrPM more dependent on the presence of the
glycans to remain soluble and reach the cell surface (Table 1)
. This
mechanism easily explains the underrepresentation, especially at the
cell surface, of the PrPM U form, which is the
least soluble and most likely to be degraded before reaching its
destination. The underrepresentation of UPrPM in
the cell model is a feature shared by the Q217R and the D178N
mutations23
and is common to the E200K and D178N familial
variants of CJD and to Fatal Familial
Insomnia.21,24
The present study strongly argues
that the underrepresentation of the U form in the E200K
PrPSc results from the effect of the E200K
mutation on PrPM before, not after, the
conversion of PrPM into
PrPSc occurs.
After limited digestion with PK, the cellular
PrPM generates a C-terminal fragment, which is
similar in size to the most common PK resistant fragment of
PrPSc. The PK resistance of the cellular fragment
is at least two orders of magnitude lower than that of
PrPSc. However it is significantly increased in
the mutant protein compared to PrPC. It has been
proposed that PrPM expressed in cell and animal
models has the essential properties of PrPSc and
that the lower level of PK resistance is because of the shorter time
available for PrPM conversion and accumulation in
these models.16-18
The present study does not support
this conclusion. We confirm that the C-terminal region of E200K
PrPM has an increased resistance to PK digestion.
Furthermore, by correcting for the amount of the aggregated form, we
show, for the first time, that the increased PK resistance is not
simply because of the higher aggregation of PrPM,
but is likely to result from an intrinsic change in the structure of
PrPM. However, the present data also show that
the PK-resistant PrPM expressed in the cell model
is qualitatively different from PrPSc (Table 1)
.
In most human prion diseases, including virtually all of the sporadic
and familial variants of CJD, PK treatment generates either one of two
major protease-resistant fragments of PrPSc: a
fragment of ~21 kd called type 1 and one of 19 kd called type
2.33,34
The difference in Mr of the two
PrPSc fragments is because of the different site
of PK cleavage which are most commonly located at residue 82 for
PrPSc type 1 and 97 for
PrPSc type 2, respectively.40
The
different cleavage site, in turn, is likely to result from the
different conformation of the two PrPSc isoforms,
or from PrPSc binding to different ligands. In
inherited prion diseases, the presence of either of the two
PrPSc is determined primarily by the 129 codon
coupled with the mutation on PRNP.33,34
Thus,
in addition to the most common familial CJD in which the E200K mutation
is coupled to the codon 129 expressing methionine and
PrPSc type 1 is present in brain, there also is a
E200K-129 valine familial CJD associated with
PrPSc type 2.36
When we modeled
these two diseases in cells, the E200K-129 methionine and E200K-129
valine cell lines failed to form the PK-resistant
PrPM fragments of 21 kd and 19 kd, respectively,
but both lines formed only a PK-resistant PrPM
isoform of ~20 kd. These findings strongly argue that whereas in the
human disease PrPM destabilized by the mutation
is eventually refolded into a specific PrPSc
isoform, in the cell model PrPM fails to reach
this stage. Therefore, although in the cell model the
PrPM reproduces the changes associated with the
E200K mutation and makes PrPM susceptible to
convert into the PrPSc form, the cellular
PrPM does not undergo this conversion and remains
different from PrPSc.
An unexpected finding of this study is the formation, after limited PK digestion, not only of the 20-kd fragment that is increased in the PrPM, but also of a much greater amount of an 18-kd C-terminal fragment in both PrPM and PrPC. The weak PK resistance of the 20-kd fragment in the PrPM preparations and of the 18-kd fragment in both PrPM and PrPC may be interpreted in view of recent nuclear magnetic resonance data of recombinant PrP. The structural data have demonstrated that PrP comprised a highly ordered region encompassing the C-terminus of PrPC approximately from residue 113, whereas the remaining N-terminal region is primarily unstructured.41-43 Thus, we propose that the E200K mutation (and other PRNP mutations having a similar effect) cause the tertiary structure to extend toward the N-terminal region to include the unstructured 112 to 90 segment. These findings point to a conformational alteration of the region between 90 to 112 of PrPM, as the underlying factor in the pathogenic process. In the E200K affected brains, as well as in other prion diseases, this extended C-terminal region is likely to be the site of major conformational changes during the conversion of PrPC to PrPSc.1,44 Furthermore, because this region includes the cleavage sites for the generation of the 20-kd and 18-kd fragments, the altered structure of PrPM results in the incorrect cleavage of the mutant protein during the recycling. This would explain the increased formation of the 20-kd fragment in the mutant cells. In scrapie-infected cells, a C-terminal PrPC fragment equivalent to the 20 kd has been shown to be directly converted into the PrPSc conformer.45 Moreover the 20-kd fragment might also be inherently pathogenic because it retains the 106 to 126 region that has been shown to be toxic in vitro,46 while it is cleaved to generate the 18-kd fragment.29
In conclusion, our data show that multiple alterations in PrPM are driven by the E200K mutation. Although some of these alterations, such as the assembly of abnormal glycans in the site flanking the mutation, are probably not critical for the pathogenic process, others, like the structural changes in the N-terminal region and the generation of potentially harmful fragments, are more likely to play a role in PrPM susceptibility to conversion into PrPSc. However other events possibly related to aging and not occurring in cell models are probably required for PrPSc formation and the clinical onset of the disease. The identification of these events and of their timing is needed to elucidate the pathogenetic mechanism and establish a preventive treatment in inherited prion diseases.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by National Institutes of Health grants AG08155 and AG08992 and by the Britton Fund.
Accepted for publication May 4, 2000.
|
References |
|---|
|
TopAbstractIntroductionExperimental ProceduresResultsDiscussionReferences |
|---|
-helices into ß-sheets features in the formation of the scrapie prion proteins. Proc Natl Acad Sci USA 1993, 90:10962-10966This article has been cited by other articles:
 |
|
 |
|
  M. W. van der Kamp and V. Daggett The consequences of pathogenic mutations to the human prion protein Protein Eng. Des. Sel., August 1, 2009; 22(8): 461 - 468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Notari, R. Strammiello, S. Capellari, A. Giese, M. Cescatti, J. Grassi, B. Ghetti, J. P. M. Langeveld, W.-Q. Zou, P. Gambetti, et al. Characterization of Truncated Forms of Abnormal Prion Protein in Creutzfeldt-Jakob Disease J. Biol. Chem., November 7, 2008; 283(45): 30557 - 30565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Life, E.-G. Lee, S. W. Eastman, and M. L. Linial Mutations in the Amino Terminus of Foamy Virus Gag Disrupt Morphology and Infectivity but Do Not Target Assembly J. Virol., July 1, 2008; 82(13): 6109 - 6119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Fioriti, S. Dossena, L. R. Stewart, R. S. Stewart, D. A. Harris, G. Forloni, and R. Chiesa Cytosolic Prion Protein (PrP) Is Not Toxic in N2a Cells and Primary Neurons Expressing Pathogenic PrP Mutations J. Biol. Chem., March 25, 2005; 280(12): 11320 - 11328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kikuchi, T. Kakeya, A. Sakai, K. Takatori, N. Nakamura, H. Matsuda, T. Yamazaki, K.-i. Tanamoto, and J.-i. Sawada Propagation of a protease-resistant form of prion protein in long-term cultured human glioblastoma cell line T98G J. Gen. Virol., November 1, 2004; 85(11): 3449 - 3457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Zanusso, A. Farinazzo, F. Prelli, M. Fiorini, M. Gelati, S. Ferrari, P. G. Righetti, N. Rizzuto, B. Frangione, and S. Monaco Identification of Distinct N-terminal Truncated Forms of Prion Protein in Different Creutzfeldt-Jakob Disease Subtypes J. Biol. Chem., September 10, 2004; 279(37): 38936 - 38942. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. A. Lawson, S. A. Priola, K. Meade-White, M. Lawson, and B. Chesebro Flexible N-terminal Region of Prion Protein Influences Conformation of Protease-resistant Prion Protein Isoforms Associated with Cross-species Scrapie Infection in Vivo and in Vitro J. Biol. Chem., April 2, 2004; 279(14): 13689 - 13695. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Satoh, T. Muramoto, T. Tanaka, N. Kitamoto, J. W. Ironside, K. Nagashima, M. Yamada, T. Sato, S. Mohri, and T. Kitamoto Association of an 11-12 kDa protease-resistant prion protein fragment with subtypes of dura graft-associated Creutzfeldt-Jakob disease and other prion diseases J. Gen. Virol., October 1, 2003; 84(10): 2885 - 2893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Drisaldi, R. S. Stewart, C. Adles, L. R. Stewart, E. Quaglio, E. Biasini, L. Fioriti, R. Chiesa, and D. A. Harris Mutant PrP Is Delayed in Its Exit from the Endoplasmic Reticulum, but Neither Wild-type nor Mutant PrP Undergoes Retrotranslocation Prior to Proteasomal Degradation J. Biol. Chem., June 6, 2003; 278(24): 21732 - 21743. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A Harris Trafficking, turnover and membrane topology of PrP: Protein function in prion disease Br. Med. Bull., June 1, 2003; 66(1): 71 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gauczynski, S. Krasemann, W. Bodemer, and S. Weiss Recombinant human prion protein mutants huPrP D178N/M129 (FFI) and huPrP+9OR (fCJD) reveal proteinase K resistance J. Cell Sci., November 1, 2002; 115(21): 4025 - 4036. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Lorenz, O. Windl, and H. A. Kretzschmar Cellular Phenotyping of Secretory and Nuclear Prion Proteins Associated with Inherited Prion Diseases J. Biol. Chem., March 1, 2002; 277(10): 8508 - 8516. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ivanova, S. Barmada, T. Kummer, and D. A. Harris Mutant Prion Proteins Are Partially Retained in the Endoplasmic Reticulum J. Biol. Chem., November 2, 2001; 276(45): 42409 - 42421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. A. Lawson, S. A. Priola, K. Wehrly, and B. Chesebro N-terminal Truncation of Prion Protein Affects Both Formation and Conformation of Abnormal Protease-resistant Prion Protein Generated in Vitro J. Biol. Chem., September 14, 2001; 276(38): 35265 - 35271. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Glucose molecule; •, mannose molecule; and 
,
N-acetylglucosamine molecule. E: Cells were
treated with Swainsonine to inhibit 
