Introduction
Our project provides human recombinant proteins to our partners in the SMP Protein and to collaboration partners in the NGFN. Proteins are expressed in bacterial and insect cell systems, purified via chromatography and characterised by biophysical assays and optical spectroscopy. We rely on the methods and know-how developed during NGFN-1 in the protein platform and in the BMBF Leitprojekt Protein Structure Factory [1, 2]. The Protein Structure Factory (PSF) is part of the international structural genomics initiative [3] and aims at the determination of human protein structures by Xray diffraction methods and NMR spectroscopy using standardised
high-throughput procedures. A complete pipeline has been established for this purpose that comprises cloning, protein expression in small and large scale, biophysical protein characterisation, crystallisation, X-ray diffraction and structure calculation. 537 different human proteins have been studied by the PSF. 139 proteins (18%) could be expressed and purified in soluble form [4]. The structures of ten full length proteins have been solved by crystallography and several domains were solved by NMR (Table 1).
Target proteins
The SMP Protein is focussed on protein-protein interactions, which are studied by a variety of methods. The structural biology project of the platform aims at understanding the mechanism of protein-protein interaction at the atomic level. We will solve structures of protein complexes identified by our partners in the SMP Protein. Interacting proteins are produced recombinantly in the expression systems E. coli and insect cells/baculovirus. Standard affinity peptide tags and specific protease cleavage sites allow to purify proteins by standard affinity chromatography and to remove affinity tags by proteolysis prior to crystallisation. Protein complexes are assembled in vitro or by co-expression (see below). Protein-protein interactions are verified by gel filtration and dynamic light scattering. Isothermal titration calorimetry is performed in collaboration with Prof. K.P. Hofmann, Charité Berlin.
Protein production for structural biology
The full-length protein is not always the ideal starting point for crystallisation experiments. Many eukaryotic proteins are large and contain several domains. Some also contain sequence stretches which are not folded in the cell but adopt a random, flexible conformation. Protein crystallisation is often complicated by the presence of such sequences. Consider the determination of the APEG-1 protein structure as an example. APEG-1, aortic preferentially expressed gene, is a marker for vascular smooth muscle cell differentiation. Although the protein is small and can be expressed well in E. coli cells, no crystals of sufficient quality could be obtained. Sequence analysis revealed that the N-terminal 16 amino acids have a high probability of lacking an ordered structure. A new expression construct was designed without the Nterminal residues. The resulting protein could be crystallised and the structure solved (PDB accession 1U2H) [5]. Based on sequence analysis, we design up to ten expression constructs for a given target protein to increase the chances of successful protein crystallisation. Each target protein is analysed by a variety of different sequence analysis tools (Table 2). The domain structure is determined with SMART and detailed domain information is obtained from PFAM. Functional annotation and protein-protein interaction data is retrieved from OMIM, the NCBI Genes database, KEGG, iPATH and STRING. All target sequences are compared to sequences of known structures from the Protein Data Bank (PDB) with BLAST. Disordered regions are predicted with DISOPRED2 and DISEMBL. Splice variants for target genes are obtained from the ENSEMBL database. Fully sequenced cDNA clones as templates for cloning experiments are obtained from the German Resource Center (RZPD). An interactive web interface (NGFN target check) was created that displays sequence and clone information and annotations and allows performing all relevant sequence analyses easily by following links in a web browser (Fig. 1).
Cloning Strategy
The Gateway cloning system of Invitrogen allows shuttling a cDNA expression construct into a variety of expression vector by a standardised recombination step. This avoids the risk of sequence errors introduced by PCR amplification and standard cloning procedures. First, an entry clone is created by standard cloning (Fig. 2). A cleavage site of tobacco etch virus (TEV) protease is included into the constructs, which allows to remove any N-terminal affinity tags after protein expression and purification. When the entry clone has been created and is proven to be free of errors by DNA sequencing, it is used to prepare a set of expression clones by simple
recombination reactions. Thereby, the same expression constructs can be studied with a variety of experimental approaches. For protein production, the optimal expression system can be quickly determined. All clones and protein expression experiments are stored in
the project database which is connected to the central database of the Protein Structure Factory. The central PSF database is accessible via web interfaces and integrates data from all experimental steps, from cloning to final structures.
Co-Expression
We have developed a set of novel expression vectors for protein co-expression in E. coli. Two or more different proteins can be expressed in the same cell. The vectors have His- or GST-affinity tags and are compatible to the Gateway cloning system. An example is shown in Figure 3.
Conclusion
High throughput methods for cloning, protein production and crystallisation are applied systematically for the study of protein-protein interactions in the SMP Protein. The close collaboration between the platform partners and the careful selection of target proteins and complexes ensures a high scientific value of structures solved by the platform.
Lit.: 1. http://www.proteinstrukturfabrik.de. 2. Heinemann, U., et al., Facilities and methods for the high-throughput crystal structure analysis of human proteins. Acc. Chem. Res., 2003. 36(3): p. 157-163. 3. Zhang, C., et al., Overview of structural genomics: from structure to function. Curr. Opin. Chem. Biol., 2003. 7(1): p. 28-32. 4. Büssow, K., et al., Structural Genomics of human proteins - target selection and generation of a public catalog of expression clones. Microbial Cell Factories, 2005. 4: p. 21. 5. Manjasetty, B.A., et al., Xray structure of engineered human Aortic Preferentially Expressed Protein-1 (APEG-1). submitted for publication, 2005.
