PEDS Advance Access published online on August 2, 2008
Protein Engineering Design and Selection, doi:10.1093/protein/gzn042
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Novel dominant-negative prion protein mutants identified from a randomized library
Institute of Neuropathology, University Hospital Zurich, Schmelzbergstrasse 12, CH-8091 Zurich, Switzerland
4 To whom correspondence should be addressed. E-mail: adriano.aguzzi{at}usz.ch
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Prion diseases are untreatable neurodegenerative disorders characterized by accumulation of PrPSc, an aggregated isoform of the cellular prion protein (PrPC). We generated a library of PrP variants with random mutations in the helix-3 domain and screened for dominant-negative mutants (DNMs) that would inhibit replication of prions (the Rocky Mountain Laboratory strain) in infected N2a cells. Two of the identified PrP mutants, Q167R and Q218K, were already known to counteract prion replication, thereby validating the effectiveness of this approach. In addition, novel DNMs were found efficiently to antagonize PrPSc propagation in cells. In contrast to Q167R and Q218K, the newly identified DNMs S221P and Y217C resided on the cell surface at a substantially lower level, suggesting that robust cell surface display of DNM might not be a general prerequisite for efficient prion antagonism. The newly identified DNMs point to useful target-selective therapeutic tools for the treatment of prion diseases.
Keywords: dominant-negative mutant/helix 3-library/prion/PrP/screening
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Prions are the infectious agents causing transmissible spongiform encephalopathies such as Creutzfeldt-Jakob disease in human and BSE in cattle. The central event in the pathogenesis of prion diseases is the ordered aggregation of misfolded isoform (PrPSc) of the cellular prion protein (PrPC). The structural and mechanistic basis of the aggregation remains elusive, primarily because a high-resolution structure of PrPSc is not available.
A promising approach for gaining mechanistic understanding of the conversion of PrPC into PrPSc consists in the identification of amino-acid residue substitutions that interfere with the conversion. The resulting structural perturbations might shed light onto the motifs and determinants of the transformation of the cellular protein PrPC into an infectious agent. In the present study, we present an approach based on functional screening for potent dominant-negative PrP mutants from a focussed PrP library.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Materials
Reagents were obtained from Sigma, Fluka or Merck, unless stated otherwise. TBS-Tw (20 mM Tris, pH 7.5, 150 mM NaCl, 0.1% Tween-20) was used as a buffer for the blocking solution and for the incubation with antibodies during scrapie cell-based assay (SCA) and western blot. PAA buffer [PBS/NP-40 (2%)/Tween-20 (2%)] or RIPA buffer [PBS/NP-40 (1%)/sodium deoxycholate (0.25%)] were used for cell lysis. The generation of anti-PrP antibodies POM1, POM2 and POM3 was described elsewhere (Polymenidou et al., 2005
). POM1 is directed against a discontinuous SDS-stable epitope on helix-1. POM2 and POM3 recognize, respectively, octapeptide repeats in the N-proximal region of PrP and residues 95–100 of murine PrP. Open reading frame (ORF) encoding rabbit PrP was amplified by PCR from muscle extract, and plasmid encoding canine PrP was kindly provided by Dr Martin Eiden.
Construction of the PrP mutant library
Error-prone PCR by using nucleotide analogs was performed essentially as described previously (Zaccolo et al., 1996). The helix-3 library of murine PrP was generated on the DNA level by PCR (25 cycles) with primer pair CAAGGGGGAGAACTTCACC and GTGCTGCTGGATCTTCTCCC (resulting in mutagenesis of residues Glu199–Asp 226) in the presence of Taq DNA polymerase, dNTPs (0.2 mM) and dNTP analogs: 8-oxo-dGTP and dPTP (50 µM each). Full-length ORF for PrP was assembled by conventional overlapping PCR (25 cycles) with Vent DNA polymerase and dNTPs (0.2 mM). ORF encoding the PrP library or control PrP variants was cloned into expression vector pBMN-IRES-GFP (Swift et al., 1999). The resulting pBMN-PrP-IRES-GFP plasmid encodes for PrP variant and GFP (translated due to the IRES element in the mRNA transcript). Hereafter, we refer to the pBMN-IRES- GFP vector without any PrP-ORF inserted into the multiple cloning sites as the ‘NI’ (no insert plasmid) control.
SCA used in the screening was modified, based on the original protocol initially developed for sensitive detection of scrapie prions in a cell-based bioassay (Klohn et al., 2003). Neuroblastoma N2a cells [the subclone N2aPK1 (Klohn et al., 2003)], chronically infected with the the Rocky Mountain Laboratory (RML) scrapie prion strain, were seeded in growth medium (OptiMEM with 10% FCS, Pen/Strep/GlutaMax) on 96-well cell culture plate at density 1.2 x 104 cells per well, and transiently transfected with pBMN-PrP-IRES-GFP plasmids 2 h later. Mock-infected N2aPK1 cells (treated with brain homogenate from non-inoculated mouse) were used as a control. Per one well, we routinely used 0.2 µg plasmid DNA and 0.5 µl Lipofectamine 2000 (Invitrogen), each diluted in 25 µl of OptiMEM. The total volume of the growth medium was 200 µl per well. Each PrP variant was tested in a triplet on a cell culture 96-well plate. Four days later, the growth medium was removed, and the cells were harvested in 200 µl PBS/EDTA (2 mM). One aliquot of the cell suspension (80 µl) was used for the determination of cell viability by using the MTS assay (CellTiter 96 Aqueous One Solution Cell Proliferation Assay Kit; Promega; absorbance readout at 490 nm; background signal was defined as the reading of samples devoid of cells), and another aliquot was used for determination of the PrPSc accumulation in transfected cells as follows. The cell suspension (80 µl) was mixed with 80 µl proteinase K (PK) solution (60 µg/ml) in the PAA buffer, and incubated for 1 h at 37°C. All following incubation steps were carried out at room temperature. The digested cell lysate was transferred into the Immobilon-P plate (hydrophobic, high protein binding white plate; Millipore) pre-rinsed with 70% ethanol and PBS, and passed through the filter by applying vacuum. The plate was washed with PBS, and residual PK activity was stopped by incubation with PMSF (1 mM in PBS) for 10 min. PrPSc adherent to the filters was then denatured with guanidine thiocyanate (3 M in PBS) for 10 min. Wells were washed with PBS and incubated sequentially with blocking solution (Top-block, Sigma) for 1 h, the POM1 anti-PrP antibody (0.4 µg/ml), and eventually with anti-murine IgG1-HRP conjugate (0.2 µg/ml; Zymed, Cat. No. 61-0120). Luminescence signal [in relative luminescence units (RLU)] was recorded by using ‘SuperSignal ELISA Pico chemiluminescent substrate’ (Pierce; 100 µl per well). In pilot experiments, we found that the luminescence signal in the modified SCA reflected the percentage of the RML-infected cells in a sample containing a mixture of RML-infected and non-infected N2a cells (data not shown). We conclude that the luminescence signal linearly reflected the amount of PrPSc in the samples. The accumulation of PrPSc (luminescence signal) was normalized to the cell viability (A490 nm; see above) resulting in RLU' values for each well. Average RLU'-values (ØRLU') were determined from three or more technical replicates on 96-well microtiter plates. An example of raw data is depicted in Supplementary figure available at PEDS online, Fig. S1. The PrPSc accumulation for each tested library member (i) was corrected for residual PrPC-signal, and related to the corrected PrPSc signal from RML-infected cells transfected with a ‘dummy’ vector (NI control) as follows:
|
|
The SCA assay was repeated in at least two additional independent experiments for each library members that caused a decrease of the PrPSc accumulation in the cells by more than 50% of the NI control.
Quantification of PrPSc accumulation in transfected N2a/RML cells by western blot
N2a cells were chronically infected with the RML prions and were seeded in 25-cm2 cell culture T-flasks. Cells were then transiently transfected as described above (a scale-up factor 30, compared with the 96-well plate). Three days later, the cells were harvested, and the pellet was resuspended in 120 µl of the RIPA buffer. The cell lysate was cleared by centrifugation at 2000 g for 3 min, adjusted to protein concentration 5 mg/ml (determined by BCA, Pierce), and incubated with PK (20 µg/ml) at 37°C for 30 min. Non-digested aliquot was kept on ice. The samples were then incubated with NuPAGE sample buffer (Invitrogen; supplemented with 100 mM DTT) for 5 min at 95°C. Proteins in the sample (25 µg per lane) were separated on a pre-cast NuPAGE 12% Bis–Tris gel, and blotted onto nitrocellulose membrane. We used anti-PrP antibody POM3 (5 µg/ml) for detection of PrPSc because (i) its epitope (residues 95-100 in murine PrP) remains intact during the PK-digest and (ii) recognition of PrP by the POM1 antibody used for the primary screening was found later to be affected in mutants containing mutation Y217C (data not shown). Sample aliquots not digested with PK were probed by western blot by using anti-GFP antibody (100 ng/ml; Roche Applied Science) or anti-β-actin antibody (100 ng/ml; Chemicon) in order to assess transfection efficiency in N2a cells and protein loading on the gel, respectively. Rabbit anti-murine IgG1-HRP conjugate was used as a secondary antibody. Light development using ‘SuperSignal West Pico chemiluminescent substrate’ (Pierce) was detected by a CCD camera, and the intensity of protein bands was quantified by using the TINA software (Raytest).
Quantification of PrP expression in HpL3–4 cells by western blot
Immortalized HpL3–4 cells derived from hippocampus of Prnp–/– mouse (Kuwahara et al., 1999) were seeded on 12-well cell culture plates at density 1.8 x 105 cells per well and transiently transfected as described above (a scale-up factor 5, compared with the 96-well plate). Twenty-four hours later, the cells were harvested, resuspended in 50 µl of the PAA buffer supplemented with protease inhibitor cocktail (0.5%; Sigma, Cat. No. P-8340), and cleared by centrifugation at 2000 g for 3 min. An aliquot of the cell lysate was used for protein determination by the BCA. The cell lysates were mixed with the NuPAGE sample buffer containing DTT (100 mM) and incubated for 5 min at 95°C. The western blot was performed as described above. Here, anti-PrP antibody POM2 (0.4 µg/ml) was used as a primary antibody for detection of full-length PrP because POM2 has a very high avidity due to its multiple binding of octapeptide repeats in the N-proximal region of PrP. For quantification, signals corresponding to the amount of detected PrP and GFP (co-expressed due to the IRES-element) were normalized to the signal corresponding to β-actin, and then the PrP signal was normalized to the GFP signal (compensation for variation in transfection efficiency among the cell samples). Expression level of dominant-negative mutants (DNMs) was related to PrP wild-type (WT).
Quantification of cell surface expression of PrP in HpL3–4 cells by flow cytometry
HpL3–4 cells were seeded on 12-well cell culture plates and transiently transfected as described above. Twenty-four hours after the transfection, the cells were harvested and stained with the conjugate of anti-PrP antibody POM2 with Cy5. PBS/FCS (1%)/EDTA (10 mM)/NaN3 (0.1%) was used for the antibody dilution and subsequent washes. The incubation with the POM2–Cy5 and washes were carried out at 4°C in order to prevent internalization of PrP in the cells. Data were acquired on an instrument ‘BD FACS Calibur’. We gated on live and GFP-positive cells, and the mean Cy5 fluorescence was used for quantification of the PrP expression level. Signal from the cells transfected with the NI control was used for offset of the background.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Design and generation of a focussed PrP library
We set out to generate a library of PrP mutants and screen it for DNMs, which might interfere with the conversion of WT PrP. Random mutagenesis is an established tool for extensive exploration of the sequence space in proteins (Hermes et al., 1990; Drummond et al., 2005). Therefore, we opted to produce a library of mutants randomized in the ORF of murine Prnp by error-prone PCR. A number of methods are available for in vitro conversion of PrPC into a PrPSc-like state (Kocisko et al., 1994; Saborio et al., 2001) yet none of these methods has been validated for the discovery of DNMs. Although more cumbersome and not easily amenable to forward genetics, cell culture assays were deemed more suitable for this purpose because of their sensitivity and precise quantitative responses (Klohn et al., 2003). Therefore, we decided to screen a focussed small-size PrP library for DNMs in a functional, albeit low-throughput screening in cells.
It has been observed that N-terminal truncations of PrP (deletion of residues 23–88) did not prevent PrPSc propagation in cell culture (Rogers et al., 1993) and in mice (Fischer et al., 1996), suggesting that the N-proximal region does not play a crucial role in the PrP conversion. In contrast, PrP-mutations, identified due to naturally occurring ‘prion-resistant’ polymorphism in sheep and human, were found to possess potent dominant-negative properties. In particular, the substitutions Q167R in the loop connecting β-sheet 2 and helix-2 (loop ‘L1’, residues 165–173) and Q218K in the helix-3 (residues 199–226) were found strongly to inhibit PrPSc propagation in cell culture and in transgenic mice (Kaneko et al., 1997b; Perrier et al., 2002).
We decided to analyze a series of random mutations within the helix-3 domain. We predicted that the replacement of many of its residues might result in disruption of stabilizing intrahelical and intramolecular hydrogen bonds and salt bridges, thereby bringing about structural perturbations in the PrP molecule that may be a prerequisite for interference with PrPSc aggregation. The length of helix-3 (28 residues) was expected to allow for satisfactory residue substitution coverage by screening 
100 members of a library with a mutational load of approximately one to two mutations per clone. Hence, the above strategy would result in testing approximately five substitutions of each residue within the helix-3.
As an important design consideration, we opted to avoid introducing an epitope tag for the 3F4 antibody (Kascsak et al., 1987), which recognizes murine PrP with mutations L108M/V111M. This epitope tag is often used to distinguish endogenous and heterologous PrP in murine neuroblastoma cells. It has been reported that the mutant of murine PrP L108M/V111M, when expressed in N2a cells, is not converted into PrPSc and can even antagonize accumulation of WT PrPSc in the N2a cells (Priola et al., 1994). In agreement with these reports, we had observed that PrP/L108M/V111M did not contribute to the generation of PrPSc when expressed in chronically infected N2a cells (data not shown). We therefore reasoned that the introduction of the L108M/V111M mutations into the sequence of PrP randomized in the helix-3 would considerably weaken the interpretation of any dominant- negative effects.
The library of PrP randomized in the helix-3 has been generated on the DNA level by error-prone PCR, as described in Materials and Methods. We analyzed sequences of 40 randomly picked clones in order to characterize the quality of the library. As expected, the vast majority of nucleotide substitutions consisted of the transitions A
G and TC. The rate of mutations was 3% at the level of DNA, and the mutational load was 2 ± 2 mutations per clone. Most of the analyzed clones (97%) encoded full-length PrP. The vast majority (90%) of all analyzed mutations were found within the gene region that encodes the helix-3. We conclude that the quality of the generated library was suitable for our screening purpose.
Screening of the PrP library by SCA
The neuroblastoma N2a cells (Race et al., 1987) provide an attractive cell culture model of the chronic propagation of scrapie-derived prions in vitro and have been exploited in many paradigms for high-throughput screens of small molecular weight anti-prion compounds (Kocisko and Caughey, 2006). Prion propagation in the infected cells is usually stable for several months and does not seem to affect cell viability. We decided to use SCA (Klohn et al., 2003) suitable for handling of N2a cells in the 96-well format. N2a cells (subclone PK1) chronically infected with RML prions (hereafter N2a/RML) were transfected with plasmids encoding individual members of the PrP library, and several days later the amount of PrPSc accumulated in the cells was assessed.
We modified SCA conditions in order to ensure efficient removal of PrPC which interferes with the detection of PrPSc. This is an important consideration in our set-up because the transfected N2a cells express endogenous PrP under the control of the authentic murine PrP promoter and also a PrP variant under the control of a viral promoter present in the expression vector. In particular, we found that it was crucial to increase the concentration of PK from 0.5 to 30 µg/ml (final concentration; data not shown for the optimization of SCA). Under these optimized conditions, signals yielded by N2a/RML cells transfected with the NI control plasmid were ca. 10-fold higher than those of mock-infected N2a cells transfected with plasmid encoding WT PrP. This indicates that the detection of PrPSc was not compromised by the overexpression of PrPC in transfected cells.
We screened 140 library members using the optimized SCA. The raw results of a representative SCA are shown in Supplementary figure available at PEDS online, Fig. S1. Since the screen was aimed at identifying potent DNMs that would significantly inhibit PrPSc propagation in the cells, we focussed on analysis of PrP mutants which caused more than 50% decrease of PrPSc accumulation in N2a/RML cells when compared with the NI control (Fig. 1). The mutations within the sequence of the helix-3 domain in these DNMs were distributed across the entire domain. In addition to the positive control (PrP/Q218K), we identified three other single mutations among the DNMs: PrP/Q167R, PrP/S221P and PrP/Y225C. Q167 does not reside in the helix-3, but in the L1 loop which had not been intentionally mutagenized. Hence, the Q167R mutation was incidentally introduced during assembly of the PrP gene by PCR. All other DNMs contained multiple mutations. We introduced mutations K203R, T215A, Y225H or Y217C that were frequently found in several multiple DNMs (Fig. 1) into the PrP gene and evaluated the effect of these single mutants in SCA. We found that PrP/Y217C is a very potent DNM, whereas the inhibitory effects of the mutants PrP/Y225H and PrP/T215A were more moderate. The mutant PrP/K203R (with a basic residue substituted for another basic residue) behaved in SCA like PrP WT (Fig. 1).
|
We tested in the SCA also rabbit and canine PrP, which may be particularly resilient to conversion into PrPSc given that rabbits and dogs are resistant to various prion strains (Loftus and Rogers, 1997
; Lysek et al., 2004). PrPSc propagation in N2a/RML cells transfected with these PrP variants was similar to that observed in the NI control cells (85 ± 15% and 93 ± 5% for rabbit and canine PrP, respectively; n = 2), suggesting that rabbit PrP and canine PrP were indeed not converted into PrPSc form, but on the other hand indicating that these PrP variants do not antagonize propagation of RML prions.
In order to confirm the above results from the screening, we analyzed the effect of the most potent DNMs on clearance of RML in the cells by western blot. In agreement with the SCA results, transfection of N2a/RML cells with PrP/Q167R, PrP/Y217C, PrP/S221P, PrP/Y217C/Y225H/D226G or PrP/M212L/Q218R/S221P resulted in substantial (70–90%) reduction of PrPSc, whereas transfection of the N2a/RML cells with a plasmid encoding PrP WT resulted in 3-fold increase in PrPSc accumulation (Fig. 2). Notably, the prevalence of transfected N2a/mock or N2a/RML cells was only 30–50%. This suggests that DNM expression, in addition to clearing prions from transfected cells, interferes with prion propagation also in non-transfected N2a/RML cells by a mechanism which remains to be elucidated.
|
Expression of selected DNMs in HpL3–4 cells
The biogenesis of the most potent DNMs was characterized in PrP-deficient HpL3–4 cells, which were originally derived from the hippocampus of a Prnp–/– mouse (Kuwahara et al., 1999). Under these conditions, the folding and cellular trafficking of the tested PrP variants cannot be modified by the co-expression of WT PrPC, as may be the case in N2a cells. Western blot analyses by using the anti-PrP antibody POM2, which binds octapeptide repeats in the N-proximal region of PrP (Polymenidou et al., 2005), showed that the overall expression level and expression profile (mobility of PrP glycoforms) of PrP/Q167R, PrP/Y217C, PrP/Q218K, PrP/S221P, PrP/Y217C/Y225H/D226G and PrP/M212L/Q218R/S221P were similar to that of transfected PrP WT (Fig. 3A). We detected a tendency for PrP/Q218K to be expressed at somewhat higher levels than PrP WT; however, expression levels of PrP WT and tested DNMs did not differ significantly (Fig. 3B).
|
Mature PrP molecules are normally attached to the plasma membrane via glycosylphosphatidylinositol (GPI) anchor, and localized in lipid rafts (Stahl et al., 1987
). We therefore wondered whether the selected DNMs stably reside on the cell surface. In order to address this question, we opted to quantify PrP localized on the cell surface by flow cytometry. HpL3–4 cells were transfected with DNM-encoding plasmids and stained the anti-PrP-antibody POM2 coupled to Cy5 fluorophore. Staining was carried out at 4°C in order to prevent internalization of the antibody prior to flow cytometry. The data reported in Fig. 4 suggest that PrP/Q218K and PrP/Q167R reside on the cell surface at a similar level as PrP WT. In contrast, the surface presence of other DNMs (PrP/Y217C, PrP/S221P, PrP/Y217C/Y225H/D226G and PrP/M212L/Q218R/S221P) was much lower compared with the PrP WT.
|
PrPC is known to be constitutively internalized (Kaneko et al., 1997a
; Prado et al., 2004). It is plausible that the DNMs, which seem to be not abundant on the cell surface, are properly targeted to the cell surface; however, due to their destabilized structure, they are internalized in a substantially faster rate than WT PrPC. An analogous mechanism has been previously demonstrated for several mutated G protein-coupled receptors that also reside in lipid rafts (Ott et al., 2004). Taken together, it seems that stable localization on the cell surface of the selected DNMs is not a prerequisite for the prion antagonism. In fact, fast internalization into early endosomes may enhance prion antagonism for, at least, some of the selected DNMs since it was suggested that PrP is likely to be converted under the low pH conditions found in endosomes (Borchelt et al., 1990; Kaneko et al., 1997a). |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
In this study, we screened for PrP mutants that would inhibit prion propagation in prion-infected N2a cells. Antagonistic effects of heterologous (mutated or chimeric) PrP molecules on prion propagation in cell culture had been initially reported by Priola et al. (1994
). Subsequent mutagenesis studies analyzed the effects of naturally occurring protective polymorphisms (Kaneko et al., 1997b; Zulianello et al., 2000; Crozet et al., 2004), glutamine residue substitutions (Atarashi et al., 2006) or reversal of charged residues in the helix-1 (Norstrom and Mastrianni, 2006) on PrP conversion. Here, we reasoned that the structural determinants of prion propagation are very poorly understood, and therefore rational predictions about the impact of individual mutations might be difficult if not impossible. For these reasons, we opted to undertake an unbiased effort by randomizing the sequence of a defined region of PrP, helix-3 within the carboxy-proximal globular domain of PrP.
We found a relatively broad distribution and various kinds of amino-acid residue substitution in the selected DNMs. Some of these individual mutations, in particular Q167R, Y217C, Q218K and S221P, had a pronounced effect on prion propagation. Although PrP/Q167R and PrP/Q218K had been already identified as potent DNMs based on the natural ‘prion-resistant’ polymorphism in ovine and human PrP, respectively (Kaneko et al., 1997b), the Y217C and S221P mutations point to hitherto unreported hot spots that may be crucial for the process of PrP conversion. It is conceivable that the additional cysteine residue in the PrP/Y217C mutant, possibly affecting the formation of the conserved S–S bond during the folding of PrP, and the S221P replacement within the helix-3, which can cause a break in the helix, have effect on PrP misfolding and aggregation resulting in the interference with the PrP conversion in the presence of WT PrPC in the cells.
How do DNMs exert their anti-prion effects? The nucleation hypothesis of prion propagation (Jarrett and Lansbury, 1993; Aguzzi and Weissmann, 1997) provides a very simple model for their action. In this framework, prions consist essentially of highly ordered, oligomeric aggregates of ca. 15–20 PrPSc molecules (Silveira et al., 2005). In many similar paradigms of nucleation, such aggregates grow appositionally at one end by addition of individual monomers. Here, DNMs may yield ‘defective’ monomers that can be recruited by existing seeds, yet do not have reactive interface necessary for the recruitment of further monomers. Hence, DNMs might effectively ‘cap’ seeds and quench their infectious potential (Horiuchi et al., 2000).
Testable structural models of DNMs, derived, for example, from nuclear magnetic resonance studies and/or from folding simulations, are likely to assist in the elucidation of the mechanism of the PrP conversion at the atomic level, and may help devising even more powerful DNMs based on rational predictions rather than random mutagenesis.
However, several alternative models of conversion inhibition are thinkable and would be compatible with the results reported above. For example, it is possible that DNMs affect prion propagation by competing with PrPC for a putative cofactor. The extraordinarily inefficient generation of prion infectivity in vitro from chemically defined constituents (Legname et al., 2004) suggests the existence of such cofactors in cells and in tissues. Circumstantial evidence suggests that the site of PrP conversion, and possibly of any hypothetical auxiliary factor (Kaneko et al., 1997b) may be congruent with caveolae-like domains of the cell surface or endosomes (Kaneko et al., 1997a; Nunziante et al., 2003). However, other cellular localizations are by no means excluded, and the auxiliary factors may encompass also non-proteinaceous polyanionic molecules such as nucleic acids (Deleault et al., 2007).
As a next step, it will be of interest to ascertain the potential of the newly discovered DNMs to inhibit PrPSc propagation in vivo. For any therapeutic application to be successful, it will be crucial that prion antagonists efficiently diffuse within the CNS and exert their effects non-cell- autonomously. Encouragingly, recombinant non-glycosylated PrP/Q218K administered in a soluble form (without the GPI anchor) was found to antagonize PrPSc propagation in prion-infected N2a cells (Kishida et al., 2004). However, another study found that cell-autonomous expression and tethering by a GPI anchor is important for efficient PrPSc antagonism by PrP/Q218K in prion-infected N2a cells (Zulianello et al., 2000). We made similar observations when selected DNMs were expressed in a soluble form, without the GPI anchor, in N2a/RML cells (data not shown). Maybe the GPI anchor is important for efficient targeting of DNM into endosomes, where PrPC is presumed to be converted into PrPSc. If so, this requirement could be fulfilled by the use of molecular targeting signals for transport of DNMs into endosomes. Efficient targeting signals may be, for instance, derived from the internalization domain of bacterial or plant toxins (Sandvig and van Deurs, 2002; Medina-Kauwe, 2007).
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
A.A. is supported by EU grants LSHB-CT-2005-018805 and FOOD-CT-2004-023144, the Swiss National Science Foundation, the Stammbach Foundation, the Ernst-Jung Foundation and the NCCR on ‘Neural Plasticity and Repair’. D.O. was supported by the Forschungskredit der Universität Zürich and the Theodor und Ida Herzog-Egli Foundation. The funding institutes and foundations had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
|
Footnotes |
|---|
1 Present address: MorphoSys AG, Lena-Christ-Strasse 48, D-82152 Martinsried/Planegg, Germany
2 Present address: Klinische Immunologie, University Hospital Zürich, Moussonstrasse 13, CH-8044 Zürich, Switzerland
3 These authors contributed equally.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
We thank Christian Julius and Gino Miele for discussions, Jan Kranich for Cy5-labeled POM2, Garry Nolan (Stanford University; CA, USA) for the pBMN-IRES-GFP plasmid, Martin Eiden (Friedrich-Loeffler-Institut, Insel Riems, Germany) for plasmid encoding canine PrP and Amedeo Caflisch and François Marchand (University of Zurich, Switzerland) for calculations of PrP aggregation propensity.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Aguzzi A., Weissmann C. Nature (1997) 389:795–798.[CrossRef][Web of Science][Medline]
Atarashi R., Sim V.L., Nishida N., Caughey B., Katamine S. J. Virol. (2006) 80:7854–7862.
Borchelt D.R., Scott M., Taraboulos A., Stahl N., Prusiner S.B. J. Cell. Biol. (1990) 110:743–752.
Crozet C., Lin Y.L., Mettling C., Mourton-Gilles C., Corbeau P., Lehmann S., Perrier V. J. Cell. Sci. (2004) 117:5591–5597.
Deleault N.R., Harris B.T., Rees J.R., Supattapone S. Proc. Natl. Acad. Sci. USA (2007) 104:9741–9746.
Drummond D.A., Iverson B.L., Georgiou G., Arnold F.H. J. Mol. Biol. (2005) 350:806–816.[CrossRef][Web of Science][Medline]
Fischer M., Rülicke T., Raeber A., Sailer A., Moser M., Oesch B., Brandner S., Aguzzi A., Weissmann C. EMBO J. (1996) 15:1255–1264.[Web of Science][Medline]
Hermes J.D., Blacklow S.C., Knowles J.R. Proc. Natl. Acad. Sci. USA (1990) 87:696–700.
Horiuchi M., Priola S.A., Chabry J., Caughey B. Proc. Natl. Acad. Sci. USA (2000) 97:5836–5841.
Jarrett J.T., Lansbury P.T. Jr. Cell (1993) 73:1055–1058.[CrossRef][Web of Science][Medline]
Kaneko K., Vey M., Scott M., Pilkuhn S., Cohen F.E., Prusiner S.B. Proc. Natl. Acad. Sci. USA (1997) a 94:2333–2338.
Kaneko K., Zulianello L., Scott M., Cooper C.M., Wallace A.C., James T.L., Cohen F.E., Prusiner S.B. Proc. Natl. Acad. Sci. USA (1997) b 94:10069–10074.
Kascsak R.J., Rubenstein R., Merz P.A., Tonna-DeMasi M., Fersko R., Carp R.I., Wisniewski H.M., Diringer H. J. Virol. (1987) 61:3688–3693.
Kishida H., Sakasegawa Y., Watanabe K., Yamakawa Y., Nishijima M., Kuroiwa Y., Hachiya N.S., Kaneko K. Amyloid (2004) 11:14–20.[Web of Science][Medline]
Klohn P.C., Stoltze L., Flechsig E., Enari M., Weissmann C. Proc. Natl. Acad. Sci. USA (2003) 100:11666–11671.
Kocisko D.A., Caughey B. Methods Enzymol. (2006) 412:223–234.[Web of Science][Medline]
Kocisko D.A., Come J.H., Priola S.A., Chesebro B., Raymond G.J., Lansbury P.T., Caughey B. Nature (1994) 370:471–474.[CrossRef][Web of Science][Medline]
Kuwahara C., et al. Nature (1999) 400:225–226.[CrossRef][Web of Science][Medline]
Legname G., Baskakov I.V., Nguyen H.O., Riesner D., Cohen F.E., DeArmond S.J., Prusiner S.B. Science (2004) 305:673–676.
Loftus B., Rogers M. Gene (1997) 184:215–219.[CrossRef][Web of Science][Medline]
Lysek D.A., Nivon L.G., Wüthrich K. Gene (2004) 341:249–253.[CrossRef][Web of Science][Medline]
Medina-Kauwe L.K. Adv. Drug Deliv. Rev. (2007) 59:798–809.[CrossRef][Web of Science][Medline]
Norstrom E.M., Mastrianni J.A. J. Virol. (2006) 80:8521–8529.
Nunziante M., Gilch S., Schätzl H.M. Chembiochem (2003) 4:1268–1284.[CrossRef][Web of Science][Medline]
Ott D., Frischknecht R., Plückthun A. Protein Eng. Des. Sel. (2004) 17:37–48.
Perrier V., Kaneko K., Safar J., Vergara J., Tremblay P., DeArmond S.J., Cohen F.E., Prusiner S.B., Wallace A.C. Proc. Natl. Acad. Sci. USA (2002) 99:13079–13084.
Polymenidou M., Stoeck K., Glatzel M., Vey M., Bellon A., Aguzzi A. Lancet Neurol. (2005) 4:805–814.[CrossRef][Web of Science][Medline]
Prado M.A., Alves-Silva J., Magalhaes A.C., Prado V.F., Linden R., Martins V.R., Brentani R.R. J. Neurochem. (2004) 88:769–781.[Web of Science][Medline]
Priola S.A., Caughey B., Race R.E., Chesebro B. J. Virol. (1994) 68:4873–4878.
Race R.E., Fadness L.H., Chesebro B. J. Gen. Virol. (1987) 68:1391–1399.
Rogers M., Yehiely F., Scott M., Prusiner S.B. Proc. Natl. Acad. Sci. USA (1993) 90:3182–3186.
Saborio G.P., Permanne B., Soto C. Nature (2001) 411:810–813.[CrossRef][Web of Science][Medline]
Sandvig K., van Deurs B. FEBS Lett. (2002) 529:49–53.[CrossRef][Web of Science][Medline]
Silveira J.R., Raymond G.J., Hughson A.G., Race R.E., Sim V.L., Hayes S.F., Caughey B. Nature (2005) 437:257–261.[CrossRef][Web of Science][Medline]
Stahl N., Borchelt D.R., Hsiao K., Prusiner S.B. Cell (1987) 51:229–240.[CrossRef][Web of Science][Medline]
Swift S., Lorens J., Achacoso P., Nolan G.P. Curr. Protoc. Immunol. (1999) 10.17.14–10.17.29.
Zaccolo M., Williams D.M., Brown D.M., Gherardi E. J. Mol. Biol. (1996) 255:589–603.[CrossRef][Web of Science][Medline]
Zulianello L., Kaneko K., Scott M., Erpel S., Han D., Cohen F.E., Prusiner S.B. J. Virol. (2000) 74:4351–4360.
Received January 7, 2008; revised June 15, 2008; accepted June 27, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Aguzzi and A. M. Calella Prions: Protein Aggregation and Infectious Diseases Physiol Rev, October 1, 2009; 89(4): 1105 - 1152. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||