RICERCA ITALIANA     

INTERACTION PARTNERS OF AMYLOIDOGENIC PROTEINS TO STUDY MISFOLDING AND AGGREGATION PROCESSES; POSSIBLE APPLICATIONS

Università degli Studi di Udine

Abstract

Some human diseases with a relevant socioeconomic impact are caused by the formation of insoluble protein aggregates of fibrillar nature, referred to as amyloid deposits.
The formation of these aggregates requires conformational modifications of the precursor proteins, which can be considered as a deviation from the normal protein folding process. Amyloid deposits, however, can also be formed by fragments, either structured or not, obtained by proteolytic events from larger precursor proteins. This project aims at studying proteins of either amyloidogenetic pathways. It has been shown that a large proportion of physiologically relevant amyloid deposits in tissues are made up by protein fragments derived from relatively larger protein precursors; on this basis, it is reasonable to propose the use of protein fragments as most suitable experimental systems for exploring the mechanisms of amyloid fibrils formation. On the other hand during the molecular process of fibrillar conversion of other proteins, aggregation requires a molecular destabilization phase and the formation of intermediate states that are stabilized by intramolecular interactions affecting the secondary structure. Fibrillogenesis can be modulated by the modification of chemical and physical conditions affecting protein solvation and the charge of a few key residues, but also by the interaction with natural or synthetic ligands. Metal ions appear important to this purpose because their binding can affect the molecular stability, promote the acquisition of different conformations and activate oligomerization processes. In addition to small ligands and metals, other two classes of molecules can be considered for the interaction with amylodogenic proteins, namely chaperone proteins and antibodies. The immunologic approach seems very interesting both for reliable therapeutic strategies and providing excellent specific molecular probes for analytical purposes. Based on the recent evidence in favour of the use of antibodies with amyloid forming proteins, the project aims at pursuing this immunologic approach. Recently, laboratories of this panel, have produced monoclonal antibodies against amyloidogenic proteins and others will be produced. In addition to antibody production and purification as well as protein expression and purification, the laboratories of the panel will work on the kinetic and thermodynamic characterisations of the protein-antibody complexes and classical functional biochemical assays by CD, fluorescence, UV, surface plasmon resonance, stopped flow kinetic methods. Limited proteolysis and isotope exchange followed by mass spectrometry, chromatography, capillary electrophoresis and other analytical techniques, together with AFM microscopy on the amyloid aggregates, NMR investigations on the interacting proteins, mainly through isotope filtered spectroscopy, will represent the additonal experimental tasks of the panel. <<<

Principal Investigator

Paolo VIGLINO Università degli Studi di UDINE

Research Objectives

Aim of the project is to investigate three different topics regarding the aberrant assembly of proteins or protein fragments which lead to the formation of fibrils and plaques of amyloidogenic diseases.
Six research units will collaborate to the realization of the project; three research units (Pavia, Firenze, Padova) are mainly involved in the expression or production of biological material (proteins, proteins fragments, antybodies etc), but also contribute to the characterization of thermodinamical, biochemical and kinetic parameters which control oligomerization and fibrillogenesis; the three other research units (Udine, Genova, Napoli) contribution is mainly due to the use of sophisticated techniques such as Atomic Force Microscopy, Multidimensional NMR spectroscopy and limited proteolysis, H/D exchange and chemical crosslinking coupled to Mass Spectroscopy.
- Studies on the aggregation of protein fragments. Amyloid deposits can be formed by fragments, either structured or not, obtained by proteolytic events from larger precursor proteins. Protein fragments can adopt in isolation partly folded states that expose hydrophobic patches that are usually hidden in the interior of a native intact protein. Therefore, fragments are capable of strong intermolecular hydrophobic interactions, thus leading to their association to form soluble oligomers, which constitute a critical nucleus of the overall process of fibrillogenesis towards the final, well-ordered mature fibrils. Proteolysis of proteins coupled with Mass Spectroscopy can be used for the identification of protein regions or fragments that are more prone to aggregation. These fragments, normally hidden in the interior of the protein molecule, once excised from the rest of the protein structure easily can aggregate. Clearly, protein fragments can be used as suitable and simplified experimental systems for studying the difficult problem of protein aggregation. Indeed, these studies are already underway and the initial results obtained with fragments of lysozyme have been positive and very encouraging.
- Interaction with metal ions. The conformational stability of amyloidogenic proteins in their non-amyloidogenic state is a crucial aspect to design new strategies directed to prevent the fibrillogenic process. Interactions with metal ions, are known to play a central role in affecting the delicate balance between protein folding and misfolding that normally favours the former and whose alteration is responsible for the appearance of aggregates. Rapidly accumulating data show that the interactions of some metal ions of d-block (Fe, Cu, Zn and Mn) might be a common denominator in affecting conformational diseases. A metal-protein association favouring protein aggregation and fibril formation has been observed in the case of alpha-synuclein, beta2-microglobulin, and acylphosphatase. The application of structural and functional analysis will aim to obtain a number of parameters, both kinetic and thermodynamic, which will be used to classify the interaction with the metal, the relative stability of the adducts, the structural features of the interaction. NMR spectroscopy in conjunction with mass spectrometry will be employed. By these techniques many type of information on the specific b2-m variants can be gained as a function of the metal concentration. These include the binding constants, the occurrence of conformational transitions, the isotope exchange parameters, size, dimension and mobility of the protein and protein-metal adducts, the changes in exposed surface and solvent accessibility pattern. Additional information may be obtained also from the observation of the metal effects on a model protein of the AcP superfamily, i.e. the sequence from S. solfataricus. NMR analysis of the system will be carried out, based on the structure determination work that is being performed. NMR is ideally suited for monitoring paramagnetic ion-protein interactions from the protein viewpoint due to the selective broadening induced by paramagnetic relaxation. Either mass spectrometry and NMR analysis are expected to provide structural information on the aggregation and metal-interaction processes, in terms of precise identification of involved sites and residues.
- antibodies interactions The aim of this approach is to prepare specific immunological probes to detect pre-fibrillar amyloid aggregates or their mature fibrils of three model proteins, alpha-synuclein (associated with Parkinson's disease and other synucleinopathies) beta2-microglobulin and its N-truncated fragment (associated with dialysis-related amyloidoses) and the N-terminal domain of the prokaryotic hydrogenase maturation factor HypF (HypF-N), not associated with any amyloid disease or other pathology. The potential capability of anti beta2-microglobulin antibodies in modifying the misfolding process responsible of the amyloid fibrils formation will be studied. These antibodies will be used in a series of basic studies in which we should verify the effect of the local interaction between the antibody and the specific epitope on the global folding stability of b2-m. The same antibodies should also have broad diagnostic and therapeutic exploitations. Antibodies raised against DN6B2m species, truncated at the sixth N-terminal residue and particularly susceptible to fibril aggregation, will permit to identify the fraction of this species of b2-m present in plasma or in periarticular fluids. This research strategy is well integrated in new proposals for possible immunological approaches to the therapy of amyloid diseases. This new therapeutic strategies can become more and more feasible along the progress of the technology of production of very specific antibodies fragments presenting high effectiveness and minimal side effects. Monoclonal antibodies we have previously produced against b2-microglobulin will be characterised in the light of most recent data. New antibodies specific for the highly amyloidogenic N-terminal truncated species will be prepared and an scFv library will be screened. At present, the scFv technology seems to be the best available to select specific molecular probes since in the biopanning methodology used to select the scFv phages, the ligand is maintained in a conformational state suitable to be specifically recognised. This step is of fundamental importance to select a specific probe.
The data we expect to obtain will be useful to shed light on the process of misfolding and aggregation of a number of proteins or their derivatives either natively unfolded, partially unfolded or fully folded, associated or not with amyloid disease. The research will also provide information on the potential of these antibody fragments for practical applications as anti-amyloid agents; hopefully, it will also lead to get molecules able to recognize epitopes displayed specifically by toxic forms of the supramolecular aggregates of two proteins associated with important protein deposition diseases such as Parkinson's disease and dialysis related amyloidosis for which no immunotherapeutic approach has since now been reported. <<<

First Results

In phase Ia several structural, thermodynamic and kinetic aspects of the fibrillogenic process of selected amyloidogenic proteins will be established. One of the predicted results is represented by the characterisation of intermediates of the folding/misfolding pathway stabilised by metal ions.
In phase Ib the results are the antibodies that target b2microglobulin, Hypf and sinuclein. For each antibody exaustive characterisation of binding properties, and definition of structural properties of the epitope will be provided.The results will consist in the identification of antibodies and their fragments able to interfere with the fibrillogenic process and in the possible elucidation of mechanism of correction of fibrillogenic propensity of the protein model under investigation. Once the functional (anti-fibrillogenic) antibodies will be availabe and characterised, it will become affordable the definition of favourable modifications induced by the antibodies, regarding the tertiary structure, the surface topology and thermodynamic stability. <<<

Timescale

24 months

National and international background

Some human pathologies with a great socioeconomic impact are caused by the formation of insoluble protein aggregates of fibrillar nature, referred to as amyloid deposits.
The formation of these aggregates requires conformational modifications of the precursor proteins, in analogy to the onset of intermediate states along the normal protein folding process. Formation of such intermediates appears to be critical for the onset of fibril formation, since these species are capable of strong intermolecular interactions given by the exposure of the polypeptide main-chain and hydrophobic side-chains that are otherwise buried in the overall fold of the native protein [Dobson 2003]. This view has led to the proposal that all polypeptide chains in principle can form amyloid aggregates under appropriate experimental conditions [Dobson 1999]. However, since all fibrils derived from different proteins share a common structural motif, the cross beta-structure, it is clear that major conformational rearrangements should take place in order to produce the well-ordered protein aggregates. Nowadays, there is a common belief that partly folded or molten globule states of proteins can be key intermediates in protein aggregation and fibrillogenesis (Fink 1998). Amyloid deposits, however, can also be formed by fragments, either structured or not, obtained by proteolytic events from larger precursor proteins. This project aims at studying proteins of either amyloidogenetic pathways.
A large proportion of physiologically relevant amyloid deposits in tissues are made up by protein fragments derived from relatively larger protein precursors (Dobson, 1999, 2003; Sacchettini and Kelly, 2002; Fink,1998) . For example, the fibrils associated with Alzheimer's disease are formed by peptide fragment(s) deriving from limited proteolysis of the amyloid precursor protein (APP) by secretases. Other examples of amyloid deposits given by protein fragments include serum amyloid A protein, gelsolin, apolipoprotein A1, immunoglobulin monoclonal light chains and fibrinogen among others. Protein fragments derived by limited proteolysis of protein precursors are particularly prone to aggregation, since they can usually only adopt, at most, partially folded states and cannot establish the long-range interactions present in the intact native protein. In particular, a common property of protein fragments is that they may contain hydrophobic patches or clusters of residues that can trigger protein aggregation by intermolecular hydrophobic interactions. Therefore, protein fragmentation by limited proteolysis appears to be an important phenomenon underlying a substantial proportion of amyloid diseases and perhaps even a causative mechanism of them. On this basis, it is reasonable to propose the use of protein fragments as most suitable experimental systems for exploring the mechanisms of amyloid fibrils formation.
On the other hand in the molecular process of fibrillar conversion of globular proteins such as lysozyme, beta2-microglobulin, acylphosphatase, synuclein, aggregation requires a molecular destabilization phase and the formation of intermediate states that are stabilized by intramolecular interactions affecting the secondary structure and priming the oligomerization process.
The laboratories of this panel have long been involved in the study of the folding dynamics and in the structural characterization of lysozyme, beta2-microglobulin, acylphosphatase and other proteins, in order to investigate the molecular events which support the fibrillogenesis process. A number of results have been obtained.( Esposito et al 2000,2003, Bellotti et al 1998, Chiti et al 2001a, 2001b, Corazza et.al. 2004, Monti et al 2002, Polverino de Laureto et al 2003, etc )
Fibrillogenesis can be modulated by the modification of chemical and physical conditions affecting protein solvation and the charge of a few key residues, but also by the interaction with natural or synthetic ligands. Metal ions appear uniquely effective to this purpose because their binding can affect the molecular stability, promote the acquisition of different conformations and activate polymerization processes. Infact it has been demonstrated that the interactions with some metal ions of d-block (Fe, Cu, Zn and Mn) can play a role in the formation of amyloid fibrils and could be a common denominator of some conformational diseases.( Morgan et.al. 2001) )
In addition to small ligands and metals, other two classes of interactors of amylodogenic proteins are particularly important, namely chaperone proteins and antibodies. Their relevance stems from either possible therapeutic/diagnostic application or fine analytical tool properties.
The interaction with other proteins can affect the folding dynamics and the aggregation process of the amyloid-forming proteins. Chaperone proteins can play an important role in stabilizing the folded state and preventing the aggregation. However chaperone proteins have normally low specificity and, as a consequence, it is difficult to think of their application for therapeutical approaches aimed at preventing protein misfolding, except when the latter affects the very release of the protein from endoplasmatic reticulum. Moreover if soluble non-fibrillar oligomers, the so-called nuclei, are the first responsible of the amyloid disease, the usage of chaperone molecules may even be detrimental because favouring rather than preventing the formation of pathologic adducts. With amyloidogenic proteins or proteolytic fragments, instead, chaperones may be ideally suited as analytical tools. The immunologic approach can be more interesting for reliable therapeutic strategies, besides providing excellent specific molecular probes also for analytical purposes. The interest in the use of antibodies or their fragments to study and contrast protein deposition diseases has increasingly raised since 1999, when it was first reported that immunization with the Abeta peptide dramatically reduced amyloid load in the brains of transgenic mice miming Alzheimer's disease (Schenk et.al 1999). Accordingly, in the last few years many studies have been focused on the use of antibodies or their fragments as possible therapeutic tools in the treatments of varying amyloidoses, such as Alzheimer's or prion diseases (White and Hawke 2003). Although experimental evidence has been reported that the immune response obtained by rising antibodies against Alzheimer plaques can remove amyloid fibrils, serious complications were observed due to brain barrier crossing and subsequent meningo-enkephalitis inflammation (Munch and Robinson 2002). However the high effectiveness of plaque inhibition of those trials suggest that antibodies may be a viable approach, provided passive immunotherapy protocols will be adopted. Similar results have been reported also for application against prion disease (Kayed et al. 2003). Polyclonal/monoclonal antibodies or their fragments are also a valuable tool to study the features of protein folding, misfolding and aggregation. A recent AFM study has investigated the differential ability of specific anti-Abeta monoclonal antibodies to directly interact with Abeta and to inhibit fibrillogenesis, thus correlating the different anti-aggregating efficiencies to the different epitope targeted (Legleiter et. al. 2004 ).
Based on the sound evidence in favour of the use of antibodies with amyloid forming proteins, this project aims at pursuing this immunologic approach. One lab of the panel has already produced (Pavia ) monoclonal antibodies against different beta2-microglobulin regions and other will be produced against the wild type and variants of beta2-microglobulin as well as the other proteins of interest, in order to study the effect on the aggregation process of the targeted protein. The use of a scFv library, The Nissim library (Nissim et al.1994), has been made available by the RU of Firenze. The library consists of differing repertoires of Vh genes rearranged in vitro from a bank of 50 cloned human Vh gene segments and a random nucleotide sequence encoding Vh-CDR3 of 4-12 amino acid residues. The resulting rearranged variable heavy chain repertoires are assembled with a constant Vlamba3 light chain affording a single chain variable fragment (scFv) repertoire. In addition, because of the specific role of beta2-microglobulin in the primary immune response, where the molecule is present as non-polymorphic component of MHCI, the interaction with the polymorphic heavy chain natural ligand will be addressed to monitor aggregation.
In addition to antibody production and purification as well as protein expression and purification, the laboratories of the panel will work on the kinetic and thermodynamic characterisations of the protein-antibody complexes and classical functional biochemical assays by CD, fluorescence, UV, surface plasmon resonance, stopped flow kinetic methods. Other amyloidogenic protein interactors, i.e. metal ions and chaperones, will also be addressed in comparison. Limited proteolysis and isotope exchange followed by mass spectrometry, chemical crosslinking followed by enzymatic digestion and mass spectrometric identification of the cross-linked peptides, chromatography, capillary electrophoresis and other analytical techniques, together with AFM microscopy on the amyloid aggregates, NMR investigations on the interacting proteins, mainly through isotope filtered spectroscopy, will represent the additonal experimental tasks of the panel. <<<