Unit: Protein Folding and Modeling - CNRS URA 2185
Director: GOLDBERG Michel
The unit combines cellular biology, genetics, biochemistry and physical-chemistry to study various problems related to protein structure, function and integration in a variety of cellular processes such as the acquisition of their functional conformation, the mechanisms of molecular transmital of the abnormal conformation of a non-conventional pathogen (the "prion" protein) and the cellular mechanisms responsible for the propagation of this pathogen.
One group of the unit deals with the conformation of the PrP protein inside the prion. A second one is interested in the way prions propagate in the infected organism to reach the brains of the animal and induce the neurodegenerative disease. A third group has a long term interest in the molecular mechanisms involved in the cellular detection of misfolded bacterial proteins and, besides these fundamental studies, develops cloning strategies for recombinant protein production.
1- Immunochemical approach of the PrP protein conformation inside the "Prion" (M.-C. Blom-Potar, P. Bittoun, M. Cardona, P. Falanga, M. Goldberg and M. Hontebeyrie)
The prion is the infectious agent responsible for spongiform encephalopathies such as the sheep scrappie, the Mad Cow Disease, or the human Creutzfeldt-Jakob disease. The prion is made of a protein, the PrP, that exists in healthy as well as diseased individuals. However, the conformation (shape) of the PrP of healthy individuals differs from that of PrP in the prion. The "Prion" group aims at characterizing this conformation difference by means of specific antibodies. Besides its purely scientific interest, this research aims at improving the sensitivity of prion detection in potentialy infected individuals or animals.
After having set up all the technical and biological tools required for our studies, and having isolated several cellular clones that produce high levels of the PrP protein, we have shown that these cell lines are able, after infection with a defined strain of prions, to produce large amounts of PrP under a form that is protease resistant (like the prion) and infectious. The kinetics of production of PrPSc (the pathogenic form characterized by its protease resistance) have been studied, and the amounts of total PrP and protease-resistant PrP produced by one of these cell lines have been measured by means of a quantitative test set up in the laboratory. We thus could show that the cell line we isolated and characterized produces about ten fold more PrPc and PrPSc than the initial cell line. This makes it possible to obtain from cell cultures of moderate size milligram amounts of the "pathogenic" PrP, which is enough for undertaking biochemical studies of this for of the PrP protein.
The study we planned to undertake, dealing with the reactivity of the prion towards monoclonal antibodies directed against the structured domain of the PrPSc protein, could not be conducted on the "natural" PrPSc obtained either from infected animals or from cell cultures. Indeed, in addition to its "proteic" part, natural PrP carries two large, complex molecular structures made of sugar polymers. For a variety of reasons, we infered that, in the prion, the sugars form a shell that covers the protein and prevents it from interacting with antibodies. We therefore attempted to get rid of these sugars so as to be able to perform our planned immunochemical study of the prion conformation. To this effect, we already had set up cell culture conditions in the presence of a glycosylation inhibitor that enabled us to obtain unglycosylated PrPSc. This year, we adapted these conditions to our PrP-hyperproducing cell line, which enabled us to obtain significant amounts of prion amyloid fibers (SAFs) predominantly unglycosylated, and thus well adapted for biochemical and immunochemical studies of the PrP within the prion.
The infectivity of various PrP preparations obtained either from animals or in cell culture has been analyzed.
Furthermore, we characterized several moinoclonal antibodies that we produced, using as immunogens peptides derived from the sequence of the structured, protease-resistant core of the PrP.
Finally, we attempted to deglycosylate natural PrP, either the normal or the PrPSc pathogenic form, by treating them with exoglycosidases that, in principle, shouyld be able to cleave the sugar bonds present in the glycosidic motives of PrP. Yet, even in its PrPc (natural) conformation, PrP turns out to be highly glycosidase resistant and we could not obtain deglycosylated PrP. However, this study enabled us to discover a proteolytic activity leading to the loss of the N-terminal, unstructured part of the PrP polypeptide chain. This proteolytic activty is detected only after treatment of a brain extract by α-fucosidase. Several fucosidases of different origins showed the same effect, suggesting that the protease is likely not a contaminant of the fucosidase prepartions. This was confirmed by the fact that native recombinant PrP prepared in the laboratory was not cleaved by the various preparations of fucosidase tested. Moreover, the recombinant PrP turned out to be protease resistant even when mixed with a brain extract in which the endogenous PrP was cleaved by the fucosidase-dependent protease. The native recombinant Prp and the the natural endogenous PrP thus show distinct protease sensitivities. The molecular origin of this difference remains unexplained: is it a "conformational difference" (in which case tne 3D structure of natural PrP would be different from that of the recombinant PrP which was determined by NMR and X-Ray crystallography ) ? Is it due to the fact that recombinant PrP is soluble while natural PrP is linked to the membrane via a GPI anchor? Is it because the protease is active only on a partially glycosylated PrP (recombinant PrP is unglycosylated) ? These are important questions, which our group will however not be able to answer since it will soon stop its activities.
2. Study of the neuro-invasion by prions (C. Barnier, C. Cuche, C. Jacquemot, C. Vadrot and F. Lazarini)
One of the major points to elucidate in prion diseases deals with the mechanisms of propagation of prions by peripheral route. The focus of our research is on the mechanisms of propagation of prion infectivity from the periphery to the central nervous system such as it occurred in the Creutzfeldt-Jakob cases that surged after the spongiform bovine encephalopathy epidemics and those resulting from growth hormone contamination. Using a murine model we study, among the immune cells, the targets of this infectious agent, in particular dendritic cells and macrophages. We are also exploring whether repeated peripheral infection with subinfectious prion doses can have a cumulative effect and eventually induce this deadly disease. Finally to validate new disinfection procedures eliminating the risk of transmission of prions by medical and surgical instruments, we analyse the effects of several physical and chemical treatments against prion infectivity of the inoculum in a scrapie murine model.
3- Protein folding in the bacterial periplasm (N. Sassoon-Clavier, M. Miot and J.-M. Betton)
The cell envelope of Gram-negative bacteria, such as E. coli, comprises an extracytoplasmic compartment, the periplasm that separates both inner and outer membranes. This cellular compartimentalization and the lack of ATP in the periplasm imply the existence of specific mechanisms monitoring protein folding via the coordinated action of chaperone and protease in the envelope.
To study protein folding in the bacterial envelope, we use a model system, MalE31, a variant of the maltose-binding protein, the periplasmic receptor for the high affinity transport of maltose in E. coli. The defective folding pathway of MalE31 leads, at high levels of production, to the aggregation of folding intermediates and to the formation of inclusion bodies. By combining genetic and biochemical approaches, we previously identified the periplasmic heat-shock chaperone FkpA, which exhibits peptidyl-prolyl isomerase activity. In collaboration with G. Bentley (Unit of Structural Immunology), we solved its tridimensionnal structure at 2 Å resolution. FkpA is a dimeric molecule in which the 245-residue subunit is divided into two domains. The N-terminal domain includes three helices that are interlaced with those of the other subunit to provide all inter-subunit contacts maintaining the dimeric species. The C-terminal domain, which belongs to the FK506-binding protein (FKBP) family, binds the FK506 ligand. The overall form of the dimer is V-shaped, and a long helice connecting the two domains provides flexibility to the molecule. In the light of this structure, we define a deletion mutant, comprising the N-terminal domain only. This domain exists in solution as a mixture of monomeric and dimeric species, and exhibits chaperone activity. By contrast, a deletion mutant comprising the C-terminal domain only is monomeric, and although it shows peptidyl-prolyl isomerase activity, it is devoid of chaperone function. These results show that the chaperone and catalytic activities are distinct, and reside in the N- and C- terminal domains, respectively. We are currently researching the physiologic substrates of FkpA.
4- ATP-dependent Clp proteases from Mycobacterium tuberculosis (N. Benaroudj and E. Johnson)
Clp proteins are ATP-dependent autocompartimentalized proteases which are widely recognized as essential factors for cell survival and involved in the virulence of many bacteria. In E. coli, Clp proteases are composed of the proteolytic subunit, ClpP, and the ATPase subunits, CpA or ClpX. The ClpP subunits are organized into two superimposed heptameric rings, which form a central chamber. In M. tuberculosis, these proteases have not been studied, but genomic sequencing revealed the presence of two genes coding for ClpP proteins (ClpP1 and ClpP2). We have produced in E. coli, purified and started the characterization of ClpP1 and ClpP2 of M. tuberculosis. Preliminary results indicate that both proteins, separately produced, are unable to hydrolyze substrate peptides, and form single heptameric rings under experimental conditions tested. We wish to determine whether the different ClpP peptidases work independently or in concert in protein breakdown and how they could be linked to the virulence of M. tuberculosis. A more long term goal will be to study new chaperones that do not belong to the DnaK or GroEL families and whose functions are still unknown.
5- Cell-free recombinant protein production (J. Rogé and J.-M. Betton)
With the recent improvements of cell-free transcription-translation systems, we have developed a new cloning strategy, based on the possibility to express, in vitro, a gene from a PCR product. This strategy, which is amenable to automation for high throughput studies, allows a rapid optimization of expression parameters, and was successfully applied for several proteins from M. leprae, selected inside targets of the structural genomic program at Pasteur Institute. In collaboration with A. Ghazi (Orsay University), we have investigated the possibility of cell-free synthesis of membrane proteins in the absence of membrane and in the presence of detergent. We used the bacterial mechanosensitive channel MscL, a homopentamer, as a model protein. A wide range of nonionic or zwitterionic detergents, Triton X-100, Tween 20, Brij 58p, n-dodecyl beta-D-maltoside, and CHAPS, were compatible with cell-free synthesis, while n-octyl beta-D-glucoside and deoxycholate had an inhibitory effect. In vitro synthesis in the presence of Triton X-100 yielded milligram amounts of MscL per milliliter of lysate. Cross-linking experiments showed that the protein was able to oligomerize in detergents. When the purified protein was reconstituted in liposomes and studied by the patch-clamp technique, its activity at the single-molecule level was similar to that of the recombinant protein produced in E. coli. Cell-free synthesis of membrane proteins should prove a valuable tool for the production of membrane proteins whose overexpression in heterologous systems is difficult. Finally, several proteins or protein domains of SARS coronavirus were produced in vitro, and purified in order to develop serologic diagnostic tests.
6- Production and characterization of a recombinant protein, candidate vaccine against paludism (A. Chaffotte and A.-G. Planson)
The C-terminal fragment F19 of MSP1 (Merozoite Surface Protein) from Plasmodium falciparum is a vaccine candidate against malaria. Our aim is to design the production of this fragment in large amount at low cost by expression in the bacterium E. coli. We had shown previously that this fragment can be obtained in its native conformation after specific, controlled proteolytic cleavage of a recombinant protein comprising the MBP (Maltose Binding Protein) fused to the F19 fragment expressed in the bacterial periplasm. While studying the in vitro folding assisting effect of MBP (Maltose Binding Protein) within the MBP-F19 fusion protein, we designed particular oxidative refolding conditions for the complete refolding of isolated denatured reduced F19. This property of autonomous oxidative refolding of F19 opens the way to the production at low cost of large amounts of this fragment after its cytoplasmic expression in E. coli.
7- Contribution to the technical platform of Biological Macromolecular Biophysics (A. Chaffotte)
The unit is in charge of the technical aspects and the scientific supervision (help in the conception and interpretation) of experiments in circular dichroism and Fourier transform infrared spectroscopy performed within the framework of the technical platform of Biological Macromolecular Biophysics. It houses and takes care of the maintenance of the corresponding equipments and has actively contributed to the choice, acquisition and setting up of a new spectrodichrograph.
8- Teaching and education
The Unit is in charge of organizing the "Protein Biochemistry" laboratory course (co-Director: A. Chaffotte and J.M. Betton) of the Institut Pasteur, which is associated to the Master's program of the Paris 6, Paris 7 and Orsay Universities, the Ecole Normale Supérieure, the Ecole Polytechnique, and the CEA and members of this Unit have contributed to the laboratory sessions of the 2004 course.
Keywords: Prion, deglycosylation, molecular chaperone, folding, stress response