Mass spectrometry based metabolomics for Pseudomonas fluorescens: Investigating effects of anti-sigma factor MucA and carbon sources on metabolism

Abstract

Pseudomonas is a bacterial genus containing species of both industrial and medical relevance due to their metabolic diversity and ability to colonize a wide variety of ecological niches including soil, water, insects, plants and animals. The biosynthetic diversity of Pseudomonas strains includes production of the polysaccharide alginate. Alginate has many commercial applications, but it is a complication to the human host during P. aeruginosa infections. In P. aeruginosa alginate production is often initiated by inactivation of the pleotropic anti-sigma factor MucA. The non-pathogenic strain P. fluorescens SBW25 used in this study does not produce alginate, but alginate production can also in this strain be initiated by MucA inactivation.

In a metabolome study the effects of inactivation of anti-sigma factor MucA was investigated, both in the presence and absence of alginate production, and when using different carbon sources. The investigation was conducted using nitrogen-limited fructose and glycerol chemostat cultivations. The strains used were P. fluorescens SBW25 wild type, an alginate producing mucA- strain and two alginate non-producing mucA- strains (a mucA- ΔalgC strain for fructose cultivations and a mucA- TTalgD strain for glycerol cultivations). Cultivation data from these chemostats showed that all mucA- mutants had an about 40% decreased fructose uptake rate, and an about 20% decreased glycerol uptake rate compared to the wild type (correcting for the carbon source fraction shuttled to alginate synthesis for the alginate producing mucA- strain). The metabolome of the various strains on the two carbon sources were characterized by preparing metabolite extracts, and analyzing them by gas chromatography – mass spectrometry (GC-MS) and liquid chromatography – tandem mass spectrometry (LCMS/ MS). The metabolome datasets showed that the different strains had distinct metabolite compositions depending on the carbon source utilized. However, amino acids and organic acids had relatively similar concentrations for all strains. Similar amino acid pools are not unexpected as all strains were cultivated at the same growth rate in nitrogen-limited chemostats. The metabolome study had two striking findings. The first finding was that for both carbon sources, the guanine nucleotide pools for the mucA- strains (the mucA- strain, the mucA- ΔalgC strain and the mucA- TTalgD strain) differed from the wild type in an alginate production dependent manner. The alginate production dependent change in guanine nucleotides for the mucA- strains was not surprising, as GTP is utilized in alginate biosynthesis. However, what causes the specific differences in pool sizes for the alginate producing and non-producing mucAstrains is not clear. The second finding was that for both carbon sources, the adenine nucleotide pools for the mucA- strains (the mucA- strain, the mucA- ΔalgC strain and the mucA- TTalgD strain), differed from the wild type in an alginate production nondependent manner. This resulted in a decreased energy charge (EC) for all mucA- strains compared to the wild type, an indication of the pleiotropic effects of MucA inactivation.

A fluxome study was performed to complement the metabolome study, and to further elucidate the effect of MucA inactivation on the metabolism of P. fluorescens. In this study 13C-labeled fructose was used in nitrogen-limited chemostat cultivations of P. fluorescens SBW25 and the non-alginate producing mucA- ΔalgC strain. Metabolite extract from the cultivations were analyzed by GC-MS/MS and LC-MS/MS producing metabolite mass isotopomer datasets. These datasets were then used in the simulation software 13CFLUX2 to determine intracellular fluxes in central carbon metabolism for P. fluorescens in the presence and absence of an active MucA. The flux distribution results revealed that the net fluxes of central carbon metabolism proceed in the same direction for both wild type and the mucA- ΔalgC strain, and that they both utilize the same main route for fructose uptake. The study also revealed that there are distinct differences between the two strains at important branch points in primary metabolism, and that such differences often coincide with changes in metabolite concentrations at, or close to, the specific branch points. The fluxome study also had two striking findings. The first finding was that compared to the wild type, the mucA- ΔalgC strain has an increased flux through the pentose phosphate pathway (PPP), and a decreased flux through the Entner – Doudoroff pathway (EDP). The second finding was that the mucA- ΔalgC strain utilizes the glyoxylate shunt, at the expense of the tricarboxylic acid cycle (TCA), whilst the wild type does not. This overall flux distribution of the mucA- ΔalgC strain causes it to produce less NADH than the wild type, whilst ATP and NADPH production is similar. The reduced NADH production for the mucA- ΔalgC strain is probably a factor contributing to its reduced EC, as less NADH is available for oxidative phosphorylation in this strain. The results from the metabolome study and the fluxome study have revealed new aspect of P. fluorescens metabolism in connection to MucA inactivation and the results can also act as a basis for future studies.

An integral part of this doctoral work was development of sample preparation protocols for metabolite extracts, and development of MS-methods for sample analysis. For the metabolome study, a targeted and a non-targeted GC-MS method for alkylated metabolites was developed, and a targeted reversed phase ion-pairing LC-MS/MS method was implemented. For the fluxome study a targeted GC-MS/MS method for alkylated metabolites was developed for detection of mass isotopomers, whilst the LCMS/ MS method from the metabolome study was expanded and quality tested for detection of mass isotopomers. In addition an isotope coded derivatization (ICD) GCMS/ MS method that improves precision for analyses of samples of silylated metabolites was developed. The silylation ICD GC-MS/MS method can complement more established normalization techniques such as isotope dilution mass spectrometry (IDMS) in the field of mass spectrometry based metabolomics.

This PhD project has contributed to the development of mass spectrometry methodology for metabolomics and fluxomics, and has employed these methods to elucidate the metabolic consequences of MucA inactivation in P. fluorescens.