Reversible attachment of Cell-Penetrating Peptides for the efficient delivery of α-synuclein to HeLa cells and primary cortical neurons

The discovery of alpha-synuclein (a-syn) as the main component of Lewy Bodies (LBs), the pathological hallmarks of Parkinson's disease (PD), and the identification of mutations in the gene coding for a-syn in familial form of PD have reinforced the central role of a-syn in the etiology of PD. Nevertheless, the molecular mechanisms by which a-syn is involved in neurodegeneration are still not fully understood. Many studies have raised the potential role of post-translational modifications (PTMs) in regulating a-syn physiological properties but also its pathogenicity in PD development. However, a-syn can be modified simultaneously by several PTMs making it extremely challenging to understand which of those PTMs is responsible in regulating the cellular properties of a-syn. Therefore, our laboratory developed a strategy to study the role of individual PTMs using semi-synthesis, enabling the generation of site-specifically modified a-syn. However, in the absence of an efficient system to internalize these semi-synthetic proteins in cells, those proteins have been limited to biophysical and structural studies. Therefore, we proposed to use cell penetrating peptides (CPPs) to deliver site-specifically modified a-syn inside cells. In a first step, we compared CPPs to deliver a-syn inside the cells. Strikingly, only N-terminal CPP fusions were efficient in delivering it to primary cortical neurons. In order to explain the differences, we performed biophysical assays and established that C-terminal CPPs formed a hairpin with the C-terminus of a-syn, preventing penetration of the plasma membrane. N-terminal fusions displayed a more transient interaction that allowed uptake. In addition, we showed that the CPPs were enhancing the aggregation of a-syn. Among the CPPs tested, TAT fused at the N-terminus was one of the most efficient. Therefore, we evaluated its capacity to deliver a-syn in different cellular models, its subcellular localization and its stability in the cells. We started by evaluating the uptake and molecular properties of TAT- a-syn in HeLa cells and in primary cortical neurons. We found that it was rapidly and efficiently internalized and distributed in the cytosol as punctuated structures. Once internalized, TAT-a-syn was cleared from the cells through autophagy and the proteasome without promoting any cellular toxicity. Finally, TAT-a-syn phosphorylated on Serine 129 or fibrillar TAT-a-syn were also successfully delivered to cells thereby showing that our tool could be used to deliver modified a-syn in cells. However, our data suggested that TAT could influence the properties of a-syn. Thus, we aimed at improving our delivery tool by developing two reversible approaches to relieve a-syn from the influence of TAT after its uptake: 1) a photocleavable linker to release a-syn from TAT upon UV irradiation or 2) a disulfide bond which can be cleaved once in the cell. Both approaches were successful in delivering a-syn to neurons. While the photocleavable version did not allow full release, TAT-S-S-a-syn was efficiently reduced and efficient in delivering pS129 a-syn. In summary, our study allowed developing a versatile tool capable of carrying a-syn inside the cells and allowing its study in the absence of constraints originating from the presence of the CPP sequence in fusion with a-syn. We believe that this tool will be useful for the study of a-syn's PTMs, interactions and cross-talk between modifications.

Elucidating the role of post-translational modifications of alpha-synuclein using semisynthesis

Mirva Hejjaoui

Alpha-synuclein (α-syn) is a natively unfolded protein that is closely linked to Parkinson’s disease (PD) by genetic, neuropathologic and biochemical evidence. Aggregated and fibrillar forms of α-syn are the main components of intracellular protein inclusions found in PD patients’ brains, termed Lewy Bodies (LB). Both in animal models and in vitro, α-syn forms fibrillar aggregates that resemble those observed in PD brain tissues. Although disease-associated mutations have been shown to promote the fibrillization of α-syn, the exact mechanisms responsible for triggering α-syn aggregation and toxicity in sporadic PD remain unknown. Addressing this gap of knowledge is crucial for understanding the molecular basis of the disease and developing effective therapies for the treatment of PD and other synucleinopathies. This project was initiated on the basis of the working hypothesis that post-translational modifications (PTM) may play important roles in modulating α-syn function and/or regulating its aggregation and toxicity. More specifically, α-syn is ubiquitously N-terminally acetylated, and phosphorylated (serines 87 and S129), ubiquitinated (lysines 12, 21 and 23) and truncated forms of α-syn have been observed in association with wild-type α-syn in LB and in brain tissues from PD patients and transgenic animals. Other modifications, such as phosphorylation at tyrosine 125 (Y125), were significantly reduced in diseased brains. Despite the discovery of candidate enzymes that mediate α-syn phosphorylation, ubiquitination and truncation, little is known about how each of these modifications alters α-syn structure, function, aggregation and toxicity in vivo. This is primarily due to the lack of tools that allow site-specific introduction of these modifications and the lack of natural mutations that can mimic the effect of these modifications, including phosphorylation. The primary focus of this thesis was to develop strategies to overcome these limitations so as to elucidate the effect of phosphorylation at Y125 and monoubiquitination at lysine 6 (K6) on α-syn structure, fibril formation, membrane binding and subcellular localization. Towards this goal, we developed two semisynthetic strategies that allow site-specific introduction of PTM in α-syn based on the ligation of synthetic peptides containing the desired modified amino acid with recombinantly expressed proteins using Expressed Protein Ligation. This approach enables the introduction of single or multiple PTM in the N- or C-terminal regions of α-syn, and the preparation of modified α-syn in milligram quantities. Using these approaches, we were able to show for the first time that ubiquitination stabilizes the monomeric form of the protein and inhibits, rather than promotes, α-syn aggregation, while phosphorylation at Y125 does not significantly change the structure and aggregation propensity of α-syn. With the semisynthetic pY125 α-syn in our hands, we were also able to investigate for the first time the sub-cellular localization of pY125 α-syn through its microinjection into primary neurons. This was not previously possible due to technical limitations related to the absence of appropriate antibodies against pY125 α-syn for immunocytochemical studies and difficulties in generating sitespecifically modified α-syn. Furthermore, we were able to investigate the effect of α-syn phosphorylation on its interactions with other proteins, by probing the effect of pS129 and pY125 on the binding to a nanobody that was specifically selected by phage-display to tightly bind to the C-terminal domain of α-syn. Accordingly, we demonstrated that phosphorylation at a single residue is capable of disrupting the binding of full-length α-syn to another protein. These results have wide-ranging implications for the potential role of phosphorylation and other PTM in regulating α-syn’s function(s). We also exploited our ability to combine semisynthetic and enzymatic approaches to investigate potential cross-talk between different PTM, namely phosphorylation at Y125 and S129 and monoubiquitination at K6. These advances would eventually allow investigating the effect of cross-talk between other N and C-terminal modifications on α-syn’s properties and to investigate PTM-dependent protein-protein and protein-ligand interactions in vitro and in cellular models of synucleinopathies. Together, our semisynthetic approaches provide novel means for the introduction of site-specific modifications in the N- and C-termini of α-syn. Current efforts in our laboratory are focused on extending these approaches to investigate the dynamics of PTM through the use of photocaged modified amino acids and to prepare novel fluorescently labeled α-syn variants to investigate its folding at the single-molecule level in living cells. The results presented in this thesis and preliminary studies from our group demonstrate that semisynthetic α-syn can provide unique opportunities to investigate structure-function of α-syn and the role of PTM in the biology of α-syn in health and disease.

EPFL2012

Exploring the role of post-translational modifications in regulating α-synuclein interactions by studying the effects of phosphorylation on nanobody binding

Anass Chiki, Farah El Turk, Bruno Claude Daniel Fauvet, Juan Kiwi, Hilal Lashuel

Intracellular deposits of α-synuclein in the form of Lewy bodies are major hallmarks of Parkinson’s disease (PD) and a range of related neurodegenerative disorders. Post-translational modifications (PTMs) of α-synuclein are increasingly thought to be major modulators of its structure, function, degradation and toxicity. Among these PTMs, phosphorylation near the C-terminus at S129 has emerged as a dominant pathogenic modification as it is consistently observed to occur within the brain and cerebrospinal fluid (CSF) of post-mortem PD patients, and its level appears to correlate with disease progression. Phosphorylation at the neighboring tyrosine residue Y125 has also been shown to protect against α-synuclein toxicity in a Drosophila model of PD. In the present study we address the potential roles of C-terminal phosphorylation in modulating the interaction of α-synuclein with other protein partners, using a single domain antibody fragment (NbSyn87) that binds to the C-terminal region of α-synuclein with nanomolar affinity. The results reveal that phosphorylation at S129 has negligible effect on the binding affinity of NbSyn87 to α-synuclein while phosphorylation at Y125, only four residues away, decreases the binding affinity by a factor of 400. These findings show that, despite the fact that α-synuclein is intrinsically disordered in solution, selective phosphorylation can modulate significantly its interactions with other molecules, and suggest how this particular form of modification could play a key role in regulating the normal and aberrant function of α-synuclein.

2018

Folding and Structure of the Pore Forming Toxin Aerolysin

Mircea Ioan Iacovache

The first obstacle encountered by a bacterial pathogen once inside the host is the plasma membrane surrounding the target cells. Throughout evolution bacteria has acquired and maintained genes that upon stimulation express proteins capable of damaging the membrane of other cells. Among these proteins pore forming toxins (PFTs) are a major class of bacterial effectors that are upregulated and secreted during bacterial infections. As their name suggests, pore forming toxins are proteins capable of inserting transmembrane pores in the membranes of the target cells which in turn leads to the lysis of the cell and release of nutrients. The mechanism by which the PFTs function during a bacterial attack has been the subject of extensive research over the years. In most cases PFTs are produced by the bacteria as soluble proteins that require the help of specialized secretion mechanisms to arrive as functional proteins in the external milieu. Once secreted by the producing bacteria these proteins diffuse towards the target cell and bind to the target membrane. Once bound to the plasma membrane of target cells they are capable of initiating a series of structural changes that will eventually lead to the conversion of the water-soluble PFT to a membrane inserted channel. The series of events and the characterization of the different structural changes required for a PFT to convert from a water-soluble protein to a membrane inserted channel is the subject of this thesis. Aerolysin, a PFT produced by Aeromonas hydrophilla, is one of the best candidates for a research into the details of the mode of action of bacterial PFTs. This particular PFT is produced by the bacterium as a soluble periplasmic protein and then secreted outside of the bacterium as a fully folded protein with the help of a type II secretion system. Binding to the target cell is achieved through two high affinity binding sites that recognize sugar modifications which are absent in A. hydrophila, a mechanism that insures that the producing cell is not damaged by its own PFT. Once bound to the target cell aerolysin requires proteolytic activation, a step which cleaves a C-terminal peptide (CTP). Activation is achieved using proteases present on the target cell and the removal of the CTP is thought to initiate the sequence of events leading to pore formation. Following activation aerolysin is able to oligomerize forming heptameric ring-like structures which spontaneously rearrange forming a transmembrane beta-barrel through the membrane. My thesis project, focused on the structural changes required in the mode of action of aerolysin, set off trying to identify the aminoacid sequence involved in the formation of the transmembrane beta-barrel. It was long thought that aerolysin would cross the membrane in a porin like fashin, forming a beta-barrel through the plasma membrane, primarily due to the lack of a hydrophobic patch of aminoacids in its sequence. An initial model proposed in the early '90s postulated that the only region that could form the transmembrane beta-barrel was the Domain 4 of the protein. In this model the removal of the CTP in the activation process would unravel the hydrophobic residues required for the beta-barrel formation and insertion. We and others were able to show however that the fourth domain of the protein is not involved directly in the formation of the pore and we identified a conserved loop in the third domain of the protein which is responsible for the formation of the beta-barrel. This loop presents an alternating pattern of hydrophobic and hydrophilic residues, a requirement for the formation of a transmembrane pore with a hydrophobic exterior and a hydrophilic cavity. Our research led us to propose a sequence of events upon insertion of the aerolysin pore in which a rearrangement of the DIII-loops of the seven monomers in the oligomer forms the initial beta-barrel and generates a hydrophobic tip which drives insertion of the structure through the membrane. Once the bilayer has been crossed the hydrophobic tips folds back on the membrane in a rivet like fashion, anchoring the pore. Following the identification of the DIII-loop as the region that forms the transmembrane pore my researched focused on the structural changes leading to the conversion of a water-soluble protein to a membrane inserted oligomer. While removal of the CTP is the key requirement for this conversion, the role of the CTP in aerolysin mode of action and the sequence of events triggered by its removal is not fully understood. Using a combination of in vivo, in vitro and in silico approaches we were able to show that the CTP plays a wider role in the aerolysin mode of action than previously thought. Indeed our research shows that the CTP is initially required for the correct folding of the soluble protein inside the bacterium, acting as a intramolecular chaperone during the folding of aerolysin. Following folding the CTP binds tightly to a hydrophobic pocket in the fourth domain of the protein locking the PFT in its soluble conformation, a role resembling C-terminal intramolecular chaperones previously described for tail spikes of bacteriophages or fiber forming collagen. This research will be continued with a study on the structural changes triggered by the removal of the CTP and their role in oligomerization and pore formation. The main focus of my thesis project has been however the determination of the structure of the oligomeric form of aerolysin. This part of the project is still ongoing and will be discussed in the final chapter of my thesis. Using 2D and 3D crystallography, AFM and modeling we hope to be able to improve our current understanding of the aerolysin heptameric form and the structural changes required in its formation.