Structural insights obtained for homodimeric full-length LRRK2/LRRK1 protein complexes and liposomal preparation as a tool to study membrane proteins under buffer gradients by Cryo-EM

Parkinson’s (PD) is one the most common neurodegenerative movement disorder. The complex etiology of the disease makes treatment difficult and although the past decades of research have substantially increased our understanding of the disease, a completely cure is still missing. Originally considered a sporadic disease, extensive genome wide studies of PD patients has identified various genes which are now linked to PD. Out of all the genes, the most prevalent is the leucine rich repeat kinase 2 (LRRK2). Mutations in LRRK2 are now believed to cause the most common familial forms, and some sporadic forms, of Parkinson’s. LRRK2 gene encodes for a large multidomain protein complex LRRK2. Structurally LRRK2 is characterized by an unique modular architecture which contains a GTPase and a kinase domain in the same complex and further surrounded by several protein-protein interaction domains. Most of the pathologically important LRRK2 mutations are clustered in the catalytic core of the protein, hinting that altered GTPase and kinase activities play a crucial role in pathogenesis. There is a need to unravel the structural mechanism that drive and modulate LRRK2 GTPase and kinase activities for a better understanding of the disease mechanism and developing advanced therapeutic strategies.
Another related but less scrutinized protein is LRRK1, the closest paralogue of LRRK2. The domain organization of LRRK1 is very similar to LRRK2 and the expression profile and cellular organization of both the proteins are also overlapping. However, irrespective of these similarities, mutations in LRRK1 have not been genetically associated with PD. This difference has stimulated various studies to understand the functional roles of LRRK2 and LRRK1 and the link between the two. Structural and functional studies on LRRK1 are not yet fully explored as LRRK2.
So far, it has been a challenge to isolate a sufficient quantity of intact, full-length LRRK2 and LRRK1 protein for structure determination. The available structural insights for LRRK2 come indirectly from the study of related proteins from the same family in lower organisms. Crystal structures for the human LRRK2 ROC domain and bacterial ROC- COR and kinase domain have been published so far. Although, these structures have advanced our understanding of LRRK2 functions but are insufficient to fully address their physiological relevance. Similarly, structural information about LRRK1 is minimal with no 3D structures reported, neither of full-length protein nor of any of its domains. In addition to continued effort to solve atomic models of individual catalytic domains of LRRK2 and LRRK1 by X-ray crystallography, there is a need to elucidate structure of full-length protein complex to delve deeper into the molecular functioning of the whole protein, given that fact that surrounding the catalytic core, LRRK2 and LRRK1 has a number of protein-interaction domains which impart high degree of conformational flexibility in order to accommodate different substrate to carry out the diverse functions.
The goal of this thesis is to solve the three dimensional structures of the homodimeric complexes formed by full-length LRRK2 and LRRK1, respectively, analyzed by cryo- electron microscopy (cryo-EM) imaging and computational single particle image anal- ysis. This will enable for the first time to unveil the tertiary structure of the protein complex. To realise the aim, the primary goal was to standardise the expression and purification of full length LRRK2 and LRRK1 to produce adequate quantity and qual- ity of proteins for structural determination. Constructs for the mammalian expression of 3xflag tagged LRRK2 and 3flag tagged LRRK1 were expressed in human embryonic kidney (HEK) 293 cells and subsequently used for affinity purification. Further, exten- sive optimization of the purified protein for cryo-EM sample preparation was carried out with the final aim to prepare homogenous sample for data collection by cryo-EM. Chapter 3 includes the methods used for the expression and purification of LRRK2, sample optimization for cryo-EM, data collection and single particle image processing. Chapter 2 gives a general introduction to Parkinson’s, its various aspects and the role of LRRK2 in Parkinson’s disease, followed by an introduction to LRRK2 and LRRK1.
In a second project, a novel method development is proposed to use liposomes as a tool to study membrane proteins under buffer gradients by cryo-EM. Methods are described on how to embed membrane proteins, such as voltage-gated potassium channels, into lipid vesicles (liposomes), while altering the buffer conditions inside and the outside of the buffer differently. This allows setting up a gradient such as pH, salt, ligands or membrane potential across the liposome bilayer membrane. Chapter 4, describes these methods to prepare liposomes, establish gradient, verify the presence of different buffers inside and outside of the liposomes. The goal of the project is the provide a proof of concept for the methods to be suitable for vitrification and image collection by cryo-EM. By optimising different lipid of protein ratios, well ordered 2D crystalline liposomes reconstituted with membrane protein were generated. These proteoliposomes can be processed by using a combination of electron crystallography and single particle processing routines. Al- ternatively, for larger membrane proteins, tomography and subtomogram averaging can also be utilised. Chapter 2 covers an introduction to electron microscopy, cryo-EM and image processing, which is the common methodological tool used in both the projects.