No abstract

Summary The structure of genetic interaction networks predicts that, analogous to synthetic lethal interactions between non-essential genes, combinations of compounds with latent activities may exhibit potent synergism. To test this hypothesis, we generated a chemical-genetic matrix of 195 diverse yeast deletion strains treated with 4915 compounds. This approach uncovered 1221 genotype-specific inhibitors, which we termed cryptagens. Synergism between 8128 structurally disparate cryptagen pairs was assessed experimentally and used to benchmark predictive algorithms. A model based on the chemical-genetic matrix and the global genetic interaction network failed to accurately predict synergism. However, a combined random forest and Naïve Bayesian learner that associated chemical structural features with genotype-specific growth inhibition had strong predictive power. This approach identified previously unknown compound combinations that exhibited species-selective toxicity towards human fungal pathogens. This work demonstrates that machine learning methods trained on unbiased chemical-genetic interaction data may be widely applicable for the discovery of synergistic combinations in different species.

No abstract

No abstract

BackgroundTo connect gene expression with cellular physiology, we need to follow levels of proteins over time. Experiments typically use variants of Green Fluorescent Protein (GFP), and time-series measurements require specialist expertise if single cells are to be followed. Fluorescence plate readers, however, a standard in many laboratories, can in principle provide similar data, albeit at a mean, population level. Nevertheless, extracting the average fluorescence per cell is challenging because autofluorescence can be substantial.ResultsHere we propose a general method for correcting plate reader measurements of fluorescent proteins that uses spectral unmixing and determines both the fluorescence per cell and the errors on that fluorescence. Combined with strain collections, such as the GFP fusion collection for budding yeast, our methodology allows quantitative measurements of protein levels of up to hundreds of genes and therefore provides complementary data to high throughput studies of transcription. We illustrate the method by following the induction of the GAL genes in Saccharomyces cerevisiae for over 20 hours in different sugars and argue that the order of appearance of the Leloir enzymes may be to reduce build-up of the toxic intermediate galactose-1-phosphate. Further, we quantify protein levels of over 40 genes, again over 20 hours, after cells experience a change in carbon source (from glycerol to glucose).ConclusionsOur methodology is sensitive, scalable, and should be applicable to other organisms. By allowing quantitative measurements on a per cell basis over tens of hours and over hundreds of genes, it should increase our understanding of the dynamic changes that drive cellular behaviour.

The butyrate-producing anaerobe Fusobacterium varium is an integral constituent of human gut microflora. Unlike many gut microorganisms, F. varium is capable of fermenting both amino acids and glucose. Although F. varium has been implicated in beneficial as well as pathological bacterium-host interactions, its genome has not been sequenced. To obtain a better understanding of the metabolic processes associated with amino acid fermentation by F. varium, we used a gel-based proteomic approach to examine the changes in the soluble proteome accompanying the utilization of eight different growth substrates: glucose, L-and D-glutamate, L-histidine, L-and D-lysine, and L-and D-serine. Using LC-MS/MS to analyze ,25% of the detected protein spots, we were able to identify 47 distinct proteins. While the intracellular concentrations of enzymes characteristic of a catabolic pathway for a specific amino acid were selectively increased in response to the presence of an excess of that amino acid in the growth medium, the concentrations of the core acetate-butyrate pathway enzymes remained relatively constant. Our analysis revealed (i) high intracellular concentrations of glutamate mutase and b-methylaspartate ammonia-lyase under all growth conditions, underscoring the importance of the methylaspartate pathway of glutamate catabolism in F. varium (ii) the presence of two enzymes of the hydroxyglutarate pathway of glutamate degradation in the proteome of F. varium ((R)-2-hydroxyglutaryl-CoA dehydratase and NAD-specific glutamate dehydrogenase) specifically when Lglutamate was the main energy source (iii) the presence of genes in the genome of F. varium encoding each of the enzymes of the hydroxyglutarate pathway (iv) the presence of both L-and Dserine ammonia-lyases (dehydratases) which permit F. varium to thrive on either L-or D-serine, respectively, and (v) the presence of aspartate-semialdehyde dehydrogenase and dihydrodipicolinate synthase, consistent with the ability of F. varium to synthesize meso-2,6-diaminopimelic acid as a component of its peptidoglycan. Proteins involved in other cellular processes, including oxidation-reduction reactions, protein synthesis and turnover, and transport were also identified.

No abstract

Fragmentation pathways have been studied on the monoanions formed during electrospray ionization of a wide range of aliphatic dicarboxylic acids and their monoesters. All negative ion spectra were obtained from alcoholic or aqueous methanolic solutions without buffers or adjustment of pH, using either a Finnigan LCQ ion trap or a VG-Micromass Quattro triple quadrupole mass spectrometer. Fragmentation pathways were studied using collision-induced dissociation and isotopic-labelling techniques. Two primary fragmentation pathways of the dicarboxylic acid monoanions were observed, namely decarboxylation of the non-ionized carboxyl group and loss of water from this group. The fragmentations were strongly dependent on the chain lengths of the diacids. In the case of a monoester anion, loss of a molecule of alcohol paralleled the loss of water from the diacid monoanion. Losses of water or alcohol were shown to lead to formation of reactive ynolate anions (HOOC(CH2)xC≡CO–, x = 3–9), which in the ion trap spectrometer engaged in complex ion – molecule reactions consistent with the chemistry of these anions. For the longer chains (x > 6), the interactions between the ionized and non-ionized carboxyl groups led to readily formed ion–neutral complexes, which are described as a neutral molecule (ROH, R = H or alkyl) held by a pair of molecular tweezers.Key words: ESI-MS/MS on negative ions, fragmentation pathways of acyclic carboxylic acid monoanions, ion–molecule reactions in an ion trap mass spectrometer, hydrogen–deuterium exchange in a gas-phase anion–neutral complex.