Vincenza Campana

Identification of an Intracellular Site of Prion Conversion

Prion diseases are fatal, neurodegenerative disorders in humans and animals and are characterized by the accumulation of an abnormally folded isoform of the cellular prion protein (PrPC), denoted PrPSc, which represents the major component of infectious scrapie prions. Characterization of the mechanism of conversion of PrPC into PrPSc and identification of the intracellular site where it occurs are among the most important questions in prion biology. Despite numerous efforts, both of these questions remain unsolved. We have quantitatively analyzed the distribution of PrPC and PrPSc and measured PrPSc levels in different infected neuronal cell lines in which protein trafficking has been selectively impaired. Our data exclude roles for both early and late endosomes and identify the endosomal recycling compartment as the likely site of prion conversion. These findings represent a fundamental step towards understanding the cellular mechanism of prion conversion and will allow the development of new therapeutic approaches for prion diseases.

PrPC is sorted to the basolateral membrane of epithelial cells independently of its association with rafts.

PrPC is a glycosylphosphatidylinositol‐anchored protein expressed in neurons as well as in the cells of several peripheral tissues. Although the normal function of PrPC remains unknown, a conformational isoform called PrPSc (scrapie) has been proposed to be the infectious agent of transmissible spongiform encephalopathies in animals and humans. Where and how the PrPC to PrPSc conversion occurs in the cells is not yet known. Therefore, dissecting the intracellular trafficking of the wild‐type prion protein, as well as of the scrapie isoform, can be of major relevance to the pathogenesis of the diseases. In this report we have analyzed the exocytic pathway of transfected mouse PrPC in thyroid and kidney polarized epithelial cells. In contrast to the majority of glycosylphosphatidylinositol‐anchored proteins, we found that PrPC is localized mainly on the basolateral domain of the plasma membrane of both cell lines. This is reminiscent of the predominant somatodendritic localization found in neurons. However, similarly to apical glycosylphosphatidylinositol‐proteins, PrPC associates with detergent‐resistant microdomains, which have been suggested to have a role in apical sorting of glycosylphosphatidylinositol‐proteins, as well as in the conversion process of PrPC to PrPSc. In order to discriminate whether detergent‐resistant microdomains have a direct role in PrPSc conversion, or whether they are involved in the transport of the protein to the site of its conversion, we have examined the effect of disruption of detergent‐resistant microdomain association on PrPC intracellular traffic. Consistent with the unusual basolateral localization of this glycosylphosphatidylinositol‐linked protein, our data exclude a classical role for detergent‐resistant microdomains in the post‐trans‐Golgi network sorting and transport of PrPC to the plasma membrane.

Different GPI-attachment signals affect the oligomerisation of GPI-anchored proteins and their apical sorting.

To understand the mechanism involved in the apical sorting of glycosylphosphatidylinositol (GPI)-anchored proteins (GPI-APs) we fused to the C-terminus of GFP the GPI-anchor-attachment signal of the folate receptor (FR) or of the prion protein (PrP), two native GPI-anchored proteins that are sorted apically or basolaterally, respectively, in MDCK cells. We investigated the behaviour of the resulting fusion proteins GFP-FR and GFP-PrP by analysing three parameters: their association with DRMs, their oligomerisation and their apical sorting. Strikingly, we found that different GPI-attachment signals differently modulate the ability of the resulting GFP-fusion protein to oligomerise and to be apically sorted. This is probably owing to differences in the GPI anchor and/or in the surrounding lipid microenvironment. Accordingly, we show that addition of cholesterol to the cells is necessary and sufficient to drive the oligomerisation and consequent apical sorting of GFP-PrP, which under control conditions does not oligomerise and is basolaterally sorted.

Detergent-resistant membrane domains but not the proteasome are involved in the misfolding of a PrP mutant retained in the endoplasmic reticulum.

Inherited prion diseases are neurodegenerative pathologies related to genetic mutations in the prion protein (PrP) gene, which favour the conversion of PrPC into a conformationally altered pathogenic form, PrPSc. The molecular basis of PrPC/PrPSc conversion, the intracellular compartment where it occurs and how this process leads to neurological dysfunction are not yet known. We have studied the intracellular synthesis, degradation and localization of a PrP mutant associated with a genetic form of Creutzfeldt-Jakob disease (CJD), PrPT182A, in transfected FRT cells. PrPT182A is retained in the endoplasmic reticulum (ER), is mainly associated with detergent-resistant microdomains (DRMs) and is partially resistant to proteinase K digestion. Although an untranslocated form of this mutant is polyubiquitylated and undergoes ER-associated degradation, the proteasome is not responsible for the degradation of its misfolded form, suggesting that it does not have a role in the pathogenesis of inherited diseases. On the contrary, impairment of PrPT182A association with DRMs by cholesterol depletion leads to its accumulation in the ER and substantially increases its misfolding. These data support the previous hypothesis that DRMs are important for the correct folding of PrP and suggest that they might have a protective role in pathological scrapie-like conversion of PrP mutants.

Lipid Rafts and Clathrin Cooperate in the Internalization of PrPC in Epithelial FRT Cells

Background The cellular prion protein (PrPC) plays a key role in the pathogenesis of Transmissible Spongiform Encephalopathies in which the protein undergoes post-translational conversion to the infectious form (PrPSc). Although endocytosis appears to be required for this conversion, the mechanism of PrPC internalization is still debated, as caveolae/raft- and clathrin-dependent processes have all been reported to be involved. Methodology/Principal Findings We have investigated the mechanism of PrPC endocytosis in Fischer Rat Thyroid (FRT) cells, which lack caveolin-1 (cav-1) and caveolae, and in FRT/cav-1 cells which form functional caveolae. We show that PrPC internalization requires activated Cdc-42 and is sensitive to cholesterol depletion but not to cav-1 expression suggesting a role for rafts but not for caveolae in PrPC endocytosis. PrPC internalization is also affected by knock down of clathrin and by the expression of dominant negative Eps15 and Dynamin 2 mutants, indicating the involvement of a clathrin-dependent pathway. Notably, PrPC co-immunoprecipitates with clathrin and remains associated with detergent-insoluble microdomains during internalization thus indicating that PrPC can enter the cell via multiple pathways and that rafts and clathrin cooperate in its internalization. Conclusions/Significance These findings are of particular interest if we consider that the internalization route/s undertaken by PrPC can be crucial for the ability of different prion strains to infect and to replicate in different cell lines.

Characterization of the Properties and Trafficking of an Anchorless Form of the Prion Protein

Conversion of PrPC into PrPSc is the central event in the pathogenesis of transmissible prion diseases. Although the molecular basis of this event and the intracellular compartment where it occurs are not yet understood, the association of PrP with cellular membranes and in particular its presence in detergent-resistant microdomains appears to be of critical importance. In addition it appears that scrapie conversion requires membrane-bound glycosylphosphatidylinositol (GPI)-linked PrP. The GPI anchor may affect either the conformation, the intracellular localization, or the association of the prion protein with specific membrane domains. However, how this occurs is not known. To understand the relevance of the GPI anchor for the cellular behavior of PrP, we have studied the biosynthesis and localization of a PrP version which lacks the GPI anchor attachment signal (PrPΔGPI). We found that PrPΔGPI is tethered to cell membranes and associates to membrane detergent-resistant microdomains but does not assume a transmembrane topology. Differently to PrPC, this protein does not localize at the cell surface but is mainly released in the culture media in a fully glycosylated soluble form. The cellular behavior of anchorless PrP explains why PrPΔGPI Tg mice can be infected but do not show the classical signs of the disorder, thus indicating that the plasma membrane localization of PrPC and/or of the converted scrapie form might be necessary for the development of a symptomatic disease.

Development of antibody fragments for immunotherapy of prion diseases

Prions are infectious proteins responsible for a group of fatal neurodegenerative diseases called TSEs (transmissible spongiform encephalopathies) or prion diseases. In mammals, prions reproduce themselves by recruiting the normal cellular protein PrP(C) and inducing its conversion into the disease-causing isoform denominated PrP(Sc). Recently, anti-prion antibodies have been shown to permanently cure prion-infected cells. However, the inability of full-length antibodies and proteins to cross the BBB (blood-brain barrier) hampers their use in the therapy of TSEs in vivo. Alternatively, brain delivery of prion-specific scFv (single-chain variable fragment) by AAV (adeno-associated virus) transfer delays the onset of the disease in infected mice, although protection is not complete. We investigated the anti-prion effects of a recombinant anti-PrP (D18) scFv by direct addition to scrapie-infected cell cultures or by infection with both lentivirus and AAV-transducing vectors. We show that recombinant anti-PrP scFv is able to reduce proteinase K-resistant PrP content in infected cells. In addition, we demonstrate that lentiviruses are more efficient than AAV in gene transfer of the anti-PrP scFv gene and in reducing PrP(Sc) content in infected neuronal cell lines. Finally, we have used a bioinformatic approach to construct a structural model of the D18scFv-PrP(C) complex. Interestingly, according to the docking results, Arg(PrP)(151) (Arg(151) from prion protein) is the key residue for the interactions with D18scFv, anchoring the PrP(C) to the cavity of the antibody. Taken together, these results indicate that combined passive and active immunotherapy targeting PrP might be promising strategies for therapeutic intervention in prion diseases.

Prions: protein only or something more? Overview of potential prion cofactors.

Transmissible spongiform encephalopathies (TSEs) in humans and animals are attributed to protein-only infectious agents, called prions. Prions have been proposed to arise from the conformational conversion of the cellular protein PrP(C) into a misfolded form (e.g., PrP(Sc) for scrapie), which precipitates into aggregates and fibrils. It has been proposed that the conversion process is triggered by the interaction of the infectious form (PrP(Sc)) with the cellular form (PrP(C)) or might result from a mutation in the gene for PrP(C). However, until recently, all efforts to reproduce this process in vitro had failed, suggesting that host factors are necessary for prion replication. In this review we discuss recent findings such as the cellular factors that might be involved in the conformational conversion of prion proteins and the potential mechanisms by which they could operate.

Coexpression of Wild-type and Mutant Prion Proteins Alters Their Cellular Localization and Partitioning into Detergent-resistant Membranes

Transmissible spongiform encephalopathies (TSEs) are a group of diseases of infectious, sporadic and genetic origin, found in higher organisms and caused by the pathological form of the prion protein. The inheritable subgroup of TSEs is linked to insertional or point mutations in the prion gene prnp, which favour its misfolding and are passed on to offspring in an autosomal‐dominant fashion. The large majority of patients with these diseases are heterozygous for the prnp gene, leading to the coexpression of the wild‐type (wt) (PrPC) and the mutant forms (PrPmut) in the carriers of these mutations. To mimic this situation in vitro, we produced Fischer rat thyroid cells coexpressing PrPwt alongside mutant versions of mouse PrP including A117V, E200K and T182A relevant to the human TSE diseases Gestmann–Sträussler–Scheinker (GSS) disease and familial Creutzfeldt–Jakob disease (fCJD). We found that coexpression of mutant PrP with wt proteins does not affect the glycosylation pattern or the biochemical characteristics of either protein. However, FRET and co‐immunoprecipitation experiments suggest an interaction occurring between the wt and mutant proteins. Furthermore, by comparing the intracellular localization and detergent‐resistant membrane (DRM) association in single‐ and double‐expressing clones, we found changes in the intracellular/surface ratio and an increased sequestration of both proteins in DRMs, a site believed to be involved in the pathological conversion (or protection thereof) of the prion protein. We, therefore, propose that the mutant forms alter the subcellular localization and the membrane environment of the wt protein in co‐transfected cells. These effects may play a role in the development of these diseases.

Gene expression profile of quinacrine-cured prion-infected mouse neuronal cells

Prion diseases are transmissible fatal neurodegenerative diseases of humans and animals, characterised by the presence of an abnormal isoform (scrapie prion protein; PrPSc) of the endogenous cellular prion protein (PrPC). The pathological mechanisms at the basis of prion diseases remain elusive, although the accumulation of PrPSc has been linked to neurodegeneration. Different genomic approaches have been applied to carry out large‐scale expression analysis in prion‐infected brains and cell lines, in order to define factors potentially involved in pathogenesis. However, the general lack of overlap between the genes found in these studies prompted us to carry an analysis of gene expression using an alternative approach. Specifically, in order to avoid the complexities of shifting gene expression in a heterogeneous cell population, we used a single clone of GT1 cells that was de novo infected with mouse prion‐infected brain homogenate and then treated with quinacrine to clear PrPSc. By comparing the gene expression profiles of about 15 000 genes in quinacrine‐cured and not cured prion‐infected GT1 cells, we investigated the influence of the presence or the absence of PrPSc. By real‐time PCR, we confirmed that the gene encoding for laminin was down‐regulated as a consequence of the elimination of PrPSc by the quinacrine treatment. Thus, we speculate that this protein could be a specific candidate for further analysis of its role in prion infection and pathogenesis.