Emma Jakobsson
PI: Kleywegt (UU)
Project: Crystallographic studies of monoamine oxidase, a drug target in the treatment of clinical depression and Parkinson's disease
Funding ends: 31 December, 2002
Research progress
* Structural studies of monoamine oxidase (MAO)
[Emma Jakobsson, Kristina Bäckbro, Torsten Unge, Gerard Kleywegt; collaborators: Jean Shih, Creed Abell]
Monoamine oxidase (MAO) is important in the metabolism of aromatic amines, including neurotransmitters such as serotonin, adrenaline, histamine and dopamine. The first modern antidepressants were irreversible and unspecific inhibitors of MAO, but due to their adverse side effects they fell into disuse. After the finding that the enzyme occurs in two isoforms, MAO-A and MAO-B, the interest in the development of isoform-specific MAO inhibitors (MAOIs) has evolved. The therapeutic effects of MAOIs fall into two categories. MAO-A inhibitors have been used mostly in the treatment of mental disorders, in particular depression. The only MAO-B inhibitor that has been extensively tested in human disorders is L-deprenyl. This drug has some positive effects in the treatment of both neurological disorders, e.g. Parkinson's and Alzheimer's diseases, and clinical depression. There is still a need for further development of these inhibitors. Therefore, knowledge of the three-dimensional structures of MAO-A and B could have a significant impact on the rational (structure-based) design of novel inhibitors, potentially with therapeutic applications.
We have worked with different systems to try and obtain large quantities of protein of a quality suitable for crystallisation. MAO-B was purified from bovine liver mitochondria using published protocols. It is possible to get large quantities, but not with adequate quality due to aggregation problems. In collaboration with Professor J.C. Shih and her group at USC, Los Angeles, USA, we expressed a full-length and a C-terminal-truncated human MAO-A construct in Sf21 cells using the baculovirus system. However, these recombinant proteins also aggregate. In the absence of a crystal structure, we have previously constructed a homology model of MAO. This model has been used to design mutants with the goal to get a more soluble but still active protein. Mutants have been cloned and expressed in E. coli. The construct was made such that it could easily be transferred to a vector for expression in Pichia pastoris. No large amounts of soluble and pure protein for crystallisation have been obtained. Since Binda et al. have recently solved the structure of MAO-B this project will be discontinued.
* eIF5A
[Emma Jakobsson, Kristina Bäckbro, Gerard Kleywegt; collaborator: Hans Johansson]
Eukaryotic initiation factor 5A (eIF5A) is a protein essential for survival of the eukaryotic cell. eIF5A is a small (17 kDa) protein which is involved in the first step of peptide-bond formation in translation and it also takes part in the cell-cycle regulation.
In many eukaryotic organisms there are two different genes for eIF5A. Recently, a second human gene for eIF5A was identified by our collaborators (Hans Johansson and co-workers). The two variants, called eIF5A I and eIF5A II, share 84% sequence identity and differ mainly in the C-terminal part.
The eIF5A I gene has been moved into a suitable plasmid for over-expression in E. coli. Pure protein has been produced and crystallisation trials are underway. eIF5A II will be moved to the same plasmid as eIF5A I so as to be able to over-express the protein and purify it for crystallisation. Previously, a homology model of eIF5A I has been calculated using IF5A from Pyrobaculum aerophilum (PDB code 1BKB) as a template. This model has been used to inspect putative metal-binding residues and as a guide for mutation studies.
* EgDf1 and EgDf2 - fatty-acid-binding proteins from Echinococcus granulosus
[Emma Jakobsson, Terese Bergfors, Gerard Kleywegt, Alwyn Jones; collaborator: Adriana Esteves]
EgDf is a fatty-acid-binding protein from the tapeworm Echinococcus granulosus, a parasite that causes hydatide disease. EgDf is developmentally regulated and could play a crucial role in the parasite life cycle since this organism is unable to synthesise most of its own lipids de novo. There are two isoforms called EgDf1 and EgDf2. These small proteins, 15 kDa, belong to a multigenic family of structurally related low-molecular-weight cytosolic proteins that includes fatty-acid-binding proteins, cellular retinoid-transport proteins, and myelin P2 protein.
Diffracting crystals of EgDf1 have been obtained and data to 1.6 Å were collected. The structure has been solved by Molecular Replacement and refined (R-value 0.176, free R-value 0.220).
Improvements of the purification procedure and crystallisation trials are underway for EgDf2.
* "Protein X"
[Emma Jakobsson, Gerard Kleywegt; collaborators: Joakim Nilsson, Derek Ogg, Martin Norin]
A collaboration has been initiated with Biovitrum (Stockholm) with the goal of solving the structure of a human protein of pharmaceutical and biological interest. Diffracting crystals had been obtained previously by Dr Ogg, but not of sufficient quality to collect a complete dataset and solve the structure. Due to lack of pure protein, further experiments were not feasible at the time. Recently, we have produced at least 60 mg of pure protein, and crystallisation trials will be initiated in the beginning of 2002.
* Substrate-assisted catalysis in the reaction mechanism of the glutathione transferases ?
[Emma Jakobsson, Gerard Kleywegt; collaborators: Ann Gustafsson, Birgit Olin, Bengt Mannervik]
Glutathione is an abundant thiol that naturally serves in the protection of cells against reactive oxygen species and their secondary toxic products as well as other electrophilic chemical species occurring in the environment or produced in living organisms. Glutathione S-transferases (GSTs) are detoxication enzymes that catalyse the inactivation of a very broad range of such noxious chemical species. Many of the GST substrates are genotoxic and cause mutations and cancer. GSTs consequently have an anti-carcinogenic function and the lack of adequate GST activity in an organism may predispose the individual to development of cancer.
By studying the activity with different glutathione analogues it was found that in addition to the sulfhydryl group of the substrate glutathione, the [[gamma]]-glutamyl-[[alpha]]-carboxylate is crucial for activity. However, some of the activity can be regained if the [[gamma]]-glutamyl-[[alpha]]-carboxylate of glutathione is replaced by a carboxylate group in the enzyme. In the mutant T68E the carboxylate group is introduced by mutation of threonine 68, which interacts with the [[gamma]]-glutamyl-[[alpha]]-carboxylate of glutathione in the wild-type,enzyme, to a glutamate.
We are currently refining the structures of the wild-type enzyme with glutathione and an analogue, GABA-Cys-Gly, lacking the [[gamma]]-glutamyl-[[alpha]]-carboxylate and the mutant without ligand and with GABA-Cys-Gly. The resolution is between 1.8 and 2.0 Å in all cases.
Miscellaneous events during 2001
* January/February: practicals in the Molecular Bioinformatics course (course organiser: Lars Liljas)
* March: lectures and practicals in the Medical Bioinformatics course (course organiser: Peter Gustavsson)
* April: Annual ICM Meeting, Högbo Bruk, Sweden
* April: RapiData 2001, Brookhaven National Laboratory, USA
* May: Bioorganic catalysis, 5 p., Uppsala University
* E. Jakobsson, T. Bergfors, A. Esteves & G.J. Kleywegt, "EgDf1 from Echinococcus granulosus", poster at the Fifth Annual Conference of the Structural Biology Network (SBNet), Tällberg, June 2001.
* August: practicals in the EMBO Practical Course "Advanced bioinformatic techniques in genomics, transcriptomics and proteomics", Linnaeus Centre for Bioinformatics, Uppsala University
Latest update at 29 March, 2001.