“Evaluation of biological variability in the proteomic characterization of a hydrogen-producing mutant of Rhodopseudomonas palustris”.

Introduction Rhodopseudomonas palustris is a purple nonsulfur anoxygenic phototrophic bacterium that is ubiquitous in the environment. R. palustris is of great interest due to its high metabolic diversity. R. palustris is capable of producing hydrogen gas, making it a potential biofuel producer. We have created a mutant variant (strain 2044) which is derepressed for nitrogenase biosynthesis in the presence of ammonia, enabling copious quantities of hydrogen gas to be produced. The wild-type strain was incapable of producing hydrogen under similar conditions. The biological goal of this project was to characterize and compare the proteomes of wild-type versus strain 2044 under photoheterotrophic conditions. This study enabled comparison of technical and biological replicates with label free quantitative “shotgun” proteomics. Methods Triplicate batches of R. palustris Strain 2044 and CGA010 wild-type were grown to log phase under photoheterotrophic conditions with ammonia present. Cells from each biological replicate were harvested, washed with buffer, and disrupted with sonication and fractionated by centrifugation. Proteome fractions were denatured, reduced, digested with trypsin, desalted, and concentrated. All samples were analyzed via a two-dimensional nano-LC-ES-MS/MS (2-dimensional linear ion trap mass spectrometer operating in data-dependent MS/MS mode). All MS/MS spectra were searched with SEQUEST and filtered/compared with DTASelect/Contrast. The three biological replicates were analyzed with duplicate technical replicates for each fraction. Label free quantitation was preformed by comparing relative protein abundance by sequence coverage, spectral count, and peptide count between WT and Strain 2044 proteins by statistical methods. Preliminary results R. palustris Strain 2044 was constructed by knocking out the two RuBisCO genes. Initially this double mutant did not grow photoheterotrophically; however after a lag of about 21 days; growth was obtained under these conditions. This culture was then streaked to isolate single clones capable of growth under photoheterotrophic conditions without a lag, which yielded strain 2044. This strain 2044 was subsequently shown to produce copious quantities of molecular hydrogen under photoheterotrophic growth in the presence of high levels of ammonia. Our goal was to determine up and down-regulation of proteins important to this new mutant and its potential for hydrogen gas production. We prepared individual peptide solutions from each replicate proteome for duplicate technical replicates by LC/LC-MS/MS on a linear ion trap. Our technical goal was to compare the variability between instrumental technical replicates and the true biological variability in living cell cultures. From the first biological replicate we identified 2,385 proteins with at least two unique peptides with conservative filtering parameters. From the WT strain 2045 proteins were identified with 80% reproducibility. From the mutant strain 2032 proteins were identified with 78% reproducibility. Over 150 proteins showed dramatic up or down expression. Many proteins involved in nitrogen fixation, such as NifX, NifK, NifD, FixA, FixB, FixC and FixX, were highly expressed in the mutant strain, even though fixed nitrogen was readily available in the media. The biological replicates are currently being analyzed with technical duplicates. Initial results have indicated that biological variability appears to be more substantial than technical variability. Our newly developed statistical package for label free quantitation (Wang et al. submitted JPR, Jan. 2005) was used to compare results across all biological and technical replicates. Novel Aspect : First systematic characterization of biological versus technical variability in a hydrogen-producing derepressed mutant of R. palustris.