Experiments in proteomics are often described using the terms “bottom-up” or “top-down.” These terms refer to two different approaches in sample preparation and the type of information about proteins or protein complexes that is being sought. Bottom-up experiments involve a large amount of sample manipulation, often the digestion of a protein to short tryptic peptides. This differs from top-down style experiments that look at intact proteins or protein structures with modifications and native mass intact. Each strategy presents a unique set of limitations as well as a complementary set of advantages. In top-down-style experiments, the intact mass of a protein can assist in the identification of post- or co-translational modifications, as well as the identification of proteins in complexes, although correct assignment of high molecular weight complexes can be difficult. Bottom-up proteomics is useful for the identification of an unknown protein but can often be based on a low percentage of sequence coverage and incomplete data concerning large regions of a protein. There is a large difference between the mass of a tryptic peptide and an intact protein. In Wu et al., a hybrid strategy is developed using the proteolysis of proteins by outer membrane protease OmpT from Escherichia coli in combination with size-dependent fractionation attempts to combine the strengths of both approaches.1 The so-called “middle-down” proteomics approach described uses larger fragments of proteins for accurate mass measurements and a greater coverage of protein in total and identification of intact posttranslational modifications.
OmpT is a non-energy-dependent outer membrane protease found in Escherichia coli and is associated with some pathogenic strains.2 The active site is on the outer leaflet of the membrane and interacts with extracellular proteins. The fold of OmpT is called a beta-barrel and is similar to other outer membrane proteins. This fold makes OmpT more stable or resistant to denaturants used in the digestion process, such as detergents or urea. The recognition sequence for cutting by OmpT is two consecutive basic residues (arginine-arginine, lysine-lysine, lysine-arginine, or arginine-lysine) and, to a lesser extent, a basic residue followed by an alanine residue. Since this combination is rarer than one basic residue, the fragments generated by OmpT digestion of a protein mixture are larger and the subsequent mixture of peptides is less complex. The protein fragments generated are typically between 3 and 30 kDa, which is ideal for accurate mass assignment. Accurate mass assignment makes for more accurate prediction of sites and identities of post-translational modifications as well as differentiation between two different isoforms of the same protein.
Applying this strategy to an entire proteome of HeLa cells, Wu et al. were able to isolate and identify 3697 peptides from 1038 unique proteins using a linear ion-trap Fourier-Transform Ion Cyclotron Resonance Mass Spectrometer (LTQ-FT Ultra) (Thermo Scientific). The average sequence coverage for proteins in this survey was 26%, which is favorably compared to other shotgun proteomics studies.
While this does not solve all the problems associated with classical bottom-up and top-down proteomics experiments, this new strategy does provide a way to identify many proteins and peptides in a less complex mixture than would be generated by a trypic digest without moving into large mass ranges where charge state and large shifts makes accurate assignment of spectra an issue. The utility of this approach for the determination of isoforms or the presence of posttranslational modifications will likely prove useful.
1. Wu, C., et al. (2012) ‘A protease for ‘middle-down’ proteomics‘, Nature Methods, 9 (8), (pp. 822-824)
2. Thomassin, J., et al. (2012) ‘OmpT outer membrane proteases of enterohemorrhagic and enteropathogenic Escherichia coli contribute differently to the degradation of human LL-37‘, infectious Immunology, 80 (2), (pp. 483-492)