Automated, high-throughput protein pipeline in place at PNNL
Pacific Northwest National Laboratory researchers have developed a high-throughput automated pipeline that can produce and purify dozens to hundreds of proteins per week, at levels up to 50 mg per individual protein. In addition, PNNL's protein pipeline includes the high-throughput isolation of protein complexes in vitro and the automated elucidation of protein-protein interactions.
The technology is scalable to up to hundreds of unique protein complexes per week, providing thousands of protein interactions for bioinformatic analyses per year. It is universally applicable for drug or biomarker discovery, proteome-wide functional analysis of proteins, and network and pathway analysis in systems biology. It is also generic—different organisms can simply be "plugged" into the workflow as long as the protein expression system is capable of producing correctly folded bait proteins.
The automated protein pipeline is a result of a research challenge raised by the Department of Energy's Genomics: Genomes to Life (GTL) program (http://www.genomestolife.org/) to understand how the information held in a microbe's DNA sequence transforms into the myriad molecular processes that allow an organism to function. The pipeline is part of the joint Oak Ridge National Laboratory/PNNL Center for Molecular and Cellular Systems (http://www.doegenomestolife.org/research/ornl.shtml) pilot project on protein complexes. The Center is working toward the Genomics: GTL program goal of developing a technology capable of characterizing 80% of the protein complexes in a microbe in 1 year (see http://www.doegenomestolife.org/program/goal1.shtml).
The pipeline team at PNNL includes Vladimir Kery, Gordon Anderson, Deanna Auberry, Kenneth Auberry, William R. (Bill) Cannon, Yuri Gorby, Eric Hill, Brian Hooker, Gary Kiebel, Chiann-Tso Lin, Eric Livesay, Priscilla Moore, Ronald Moore, Elena Peterson, Heidi Sofia, Kristin Victry, Richard D. (Dick) Smith, and Steven Wiley.
Protein complexes—"the molecular machines" of the cell—are responsible for molecular processes within and between cells, and they dictate how a cell or organism interacts with its environment. Thus, identifying and characterizing these protein complexes is the key step towards defining biological function from the molecular level.
96 different genes of Shewanella oneidensis were expressed with hexahistidine affinity tags in E. coli and purified using the Beckman Biomek FX robot and MagneHis beads. The purified proteins were analyzed by SDS-PAGE. Average yield of 40 proteins purified to >90% homogeneity was 0.25 mg. Full Image.
The process involves expressing cloned probe proteins and affinity purifying them. The purified probes are then immobilized and exposed to host cell lysate. Protein complexes are formed exogenously as the large amount of the probe protein drives protein complex formation. Protein interactors are then eluted and automatically digested, and the peptides of the individual components are identified by automated liquid chromatography (LQ) tandem mass spectrometry (MS).
Protein interaction network for several important cellular processes in Shewanella oneidensis. Full Image.
Pipeline Data and Validation
PNNL has demonstrated the automated pipeline on several well-known, well-characterized bacterial protein complexes. Validated protein interaction networks are being put into a database of protein interactions to be publicly available via the Internet. All data from the pipeline are statistically analyzed to obtain probabilities for individual protein interactions. The protein interaction networks are then visualized.
PNNL researchers are developing complex bioinformatics analyses of the experimentally obtained protein interaction networks to ultimately combine multiple sets of biological and bioinformatics data (gene profiling, nearest gene neighbor analysis, DNA arrays, protein expression, whole proteome analysis, protein interaction networks) into one "snapshot" about specific interactors in an organism of interest.