Automated 96-well bacterial protein purification:

The Harvard Institute of Proteomics demonstrated for the first time, the methodology of high-throughput purification of proteins in a 96-well format (Braun et al, 2002). In order to meet our ongoing need for several hundreds of microgram levels of purified proteins, we developed an automated bacterial high-throughput protein purification pipeline. The basic elements of this 96-well format pipeline include bacterial protein expression, walk-away automated purification and automated analysis to determine the size, quantity and purify of the protein. Briefly, protein expression is performed in 1 ml. cultures and the expressed recombinant proteins are purified robotically followed by automated analysis. The purification pipeline has a throughput capacity of >500 proteins per day and yields 1-40mg/ml culture of protein based on the expression levels. Several conditions in the pipeline including protein expression, media, strains, buffers etc were extensively optimized. In our recent experiments, we have successfully generated over a thousand protein expression clones and expressed several hundreds of proteins from different species, currently being used for antigen discovery. In general, an expression success rate of 90% for the cytosolic proteins and 40% for membrane proteins was observed. Some of the membrane proteins, which failed to express as full-length, were successfully expressed upon deletion of the hydrophobic amino acid regions. Various families/sets of proteins purified on this platform have been used in biochemical and immunological assays. Novel high-throughput protein assay development methodologies in harmony with the purification pipeline have also been established (to be published). Families/sets of proteins purified on this pipeline have been used in various assays in collaboration with other laboratories (to be published).

High-throughput bacterial cell-free system for rapid protein production:

Bacterial cell-free protein synthesis is a simple process where extraneously added DNA is transcribed and translated in vitro to produce protein. Efforts from different laboratories in the past few years led to design of protocols to generate highly synthetic bacterial cell extracts capable of producing hundreds of micrograms of protein in batch reactions. However, the short lifetime of the extract in batch reactions, consequently leading to low yield of protein is a limitation of the cell-free translation systems. Nevertheless the cell-free protein synthesis has several advantages over cell-based systems particularly in the expression of toxic proteins, labeling of amino acids for structural studies and expression of mutants of a protein for rapid analysis. Cell-free protein synthesis enables addition of detergents, chaperones and appropriate ligands during the process of protein synthesis, which may aid in proper folding of the proteins. Most of the genes cloned into bacterial expression vectors with T7 promoter, can also serve as templates for bacterial cell-free expression, obviating the need for sub-cloning. Cell-free protein synthesis requires several ingredients such as tRNA, amino acids, nucleotides, components of energy regenerating system, small molecules and T7 RNA polymerase in optimum proportions. Use of this complex mixture requires extensive optimization to produce proteins in a reproducible manner. Commercial extracts for protein synthesis are highly expensive, not practical for high-throughput studies and are not amenable to modifications, as the composition is not disclosed. We have adopted the strategies of Yokoyama’s group in Japan, for preparation of bacterial cell extracts for protein synthesis. With a few modifications, we standardized production of bacterial cell extract, optimized conditions for protein synthesis and tested the efficacy of our extracts with appropriate control proteins. 145 proteins belonging to different species were expressed in the cell-free system. We have reported the high-throughput method of cell-free expression and affinity purification of 63 proteins (Murthy et al., 2004).

Braun, P., Hu, Y., Shen, B., Halleck, A., Koundinya, M., Harlow, E. and LaBaer, J. Proteome-scale purification of human proteins from bacteria. Proc. Natl. Acad. Sci. USA. 2002 Mar 5;99(5):2654-9.

Murthy, T. V., Wu, W., Qiu, Q. Q., Shi, Z., LaBaer, J. and Brizuela, L. Bacterial cell-free system for high-throughput protein expression and a comparative analysis of Escherichia coli cell-free and whole cell expression systems. Protein Expr. Purif. 2004 Aug;36(2):217-25. -->PDF

Pascal Braun

To develop a generalized method for high-throughput human protein expression and purification, we have examined polypeptide purification tags for robust chemistry with favorable effects on protein yield and purity. We utilized a test set of 32 human genes, encoding proteins of varying sizes and activities. This test set was transferred into four expression vectors, each containing a different purification tag: poly-histidine (His_6-), calmodulin-binding peptide (CBP-), glutathione-S-transferase (GST-), and maltose-binding protein (MBP-). For each of these tags, we developed a simple 96-well formatted purification method that can be completed in less than two hours. We characterized the 128 fusion-proteins for total expression, yield, protein purity and which steps entailed large losses. Although the small tags were not suited to non-denaturing conditions, the GST- and MBP-tags allowed us to purify 26/32 and 28/32 proteins, respectively, to yields of >1ug purified protein per 1ml of culture. We confirmed that our proteins are active in two independent functional assays.

The developed methods were applied to a larger set of 336 randomly selected cDNAs. Sixty percent of the corresponding proteins were successfully purified under denaturing conditions and 82% of these under non-denaturing conditions. A relational database, FLEXProt, was built to compare properties of successfully purified and failed proteins. We found that some pfam domains were almost exclusively found in proteins that were successfully purified and thus may have predictive character.

Currently, we are expanding this test set, refine the bioinformatic analysis and analyze the results for statistical significance.
PDF