Search

A central assumption of all machine learning is that the training data are an informative subset of the true distribution we want to learn. Yet, this assumption may be violated in practice. Recently, learning from the molecular structures of small molecules has moved into the focus of the machine learning community. Usually, those small molecules are of biological interest, such as metabolites or drugs. Applications include prediction of toxicity, ligand binding or retention time.We investigate how well certain large-scale datasets cover the space of all known biomolecular structures. Investigation of coverage requires a sensible distance measure between molecular structures. We use a well-known distance measure based on solving the Maximum Common Edge Subgraph (MCES) problem, which agrees well with the chemical and biochemical intuition of similarity between compounds. Unfortunately, this computational problem is NP-hard, severely restricting the use of the corresponding distance measure in large-scale studies. We introduce an exact approach that combines Integer Linear Programming and intricate heuristic bounds to ensure efficient computations and dependable results.We find that several large-scale datasets frequently used in this domain of machine learning are far from a uniform coverage of known biomolecular structures. This severely confines the predictive power of models trained on this data. Next, we propose two further approaches to check if a training dataset differs substantially from the distribution of known biomolecular structures. On the positive side, our methods may allow creators of large-scale datasets to identify regions in molecular structure space where it is advisable to provide additional training data.

The exchange of metabolites mediates algal and bacterial interactions that maintain ecosystem function. Yet, while thousands of metabolites are produced, only a few molecules have been identified in these associations. Using the ubiquitous microalgae Pseudo-nitzschia sp., as a model, we employed an untargeted metabolomics strategy to assign structural characteristics to the metabolites that distinguished specific diatom-microbiome associations. We cultured five species of Pseudo-nitzschia, including two species that produced the toxin domoic acid, and examined their microbiomes and metabolomes. A total of 4826 molecular features were detected by tandem mass spectrometry. Only 229 of these could be annotated using available mass spectral libraries, but by applying new in silico annotation tools, characterization was expanded to 2710 features. The metabolomes of the Pseudo-nitzschia-microbiome associations were distinct and distinguished by structurally diverse nitrogen compounds, ranging from simple amines and amides to cyclic compounds such as imidazoles, pyrrolidines and lactams. By illuminating the dark metabolomes, this study expands our capacity to discover new chemical targets that facilitate microbial partnerships and uncovers the chemical diversity that underpins algae-bacteria interactions.

Metabolomics using nontargeted tandem mass spectrometry can detect thousands of molecules in a biological sample. However, structural molecule annotation is limited to structures present in libraries or databases, restricting analysis and interpretation of experimental data. Here we describe CANOPUS (class assignment and ontology prediction using mass spectrometry), a computational tool for systematic compound class annotation. CANOPUS uses a deep neural network to predict 2,497 compound classes from fragmentation spectra, including all biologically relevant classes. CANOPUS explicitly targets compounds for which neither spectral nor structural reference data are available and predicts classes lacking tandem mass spectrometry training data. In evaluation using reference data, CANOPUS reached very high prediction performance (average accuracy of 99.7% in cross-validation) and outperformed four baseline methods. We demonstrate the broad utility of CANOPUS by investigating the effect of microbial colonization in the mouse digestive system, through analysis of the chemodiversity of different Euphorbia plants and regarding the discovery of a marine natural product, revealing biological insights at the compound class level. Unknown metabolites are classified from mass spectrometry data.

Interpretation of fragmentation mass spectra depends on our knowledge of collision-induced dissociation mechanisms. Computational methods for the annotation of fragmentation mechanisms operate within the boundaries of recognized fragmentation pathways. The prevalence of charge migration fragmentation (CMF) in sodiated ion fragmentation spectra, which produces nonsodiated fragment ions, is unknown. Here, we investigated the extent of CMF in the fragmentation spectra of sodiated precursors by mining the NIST17 spectral library using a diagnostic mass difference. Our results showed that a substantial amount of fragment ions in sodiated precursor spectra are derived from CMF, indicating that this fragmentation mechanism should be commonly considered by computational methods for compound annotation.

The confident high-throughput identification of small molecules is one of the most challenging tasks in mass spectrometry-based metabolomics. Annotating the molecular formula of a compound is the first step towards its structural elucidation. Yet even the annotation of molecular formulas remains highly challenging. This is particularly so for large compounds above 500 daltons, and for de novo annotations, for which we consider all chemically feasible formulas. Here we present ZODIAC, a network-based algorithm for the de novo annotation of molecular formulas. Uniquely, it enables fully automated and swift processing of complete experimental runs, providing high-quality, high-confidence molecular formula annotations. This allows us to annotate novel molecular formulas that are absent from even the largest public structure databases. Our method re-ranks molecular formula candidates by considering joint fragments and losses between fragmentation trees. We employ Bayesian statistics and Gibbs sampling. Thorough algorithm engineering ensures fast processing in practice. We evaluate ZODIAC on five datasets, producing results substantially (up to 16.5-fold) better than for several other methods, including SIRIUS, which is the state-of-the-art algorithm for molecular formula annotation at present. Finally, we report and verify several novel molecular formulas annotated by ZODIAC. To infer a previously unknown molecular formula from mass spectrometry data is a challenging, yet neglected problem. Ludwig and colleagues present a network-based approach to ranking possible formulas.

Metabolites provide a direct functional signature of cellular state. Untargeted metabolomics usually relies on mass spectrometry, a technology capable of detecting thousands of compounds in a biological sample. Metabolite annotation is executed using tandem mass spectrometry. Spectral library search is far from comprehensive, and numerous compounds remain unannotated. So-called in silico methods allow us to overcome the restrictions of spectral libraries, by searching in much larger molecular structure databases. Yet, after more than a decade of method development, in silico methods still do not reach the correct annotation rates that users would wish for. Here, we present a novel computational method called Mad Hatter for this task. Mad Hatter combines CSI:FingerID results with information from the searched structure database via a metascore. Compound information includes the melting point, and the number of words in the compound description starting with the letter ‘u’. We then show that Mad Hatter reaches a stunning 97.6% correct annotations when searching PubChem, one of the largest and most comprehensive molecular structure databases. Unfortunately, Mad Hatter is not a real method. Rather, we developed Mad Hatter solely for the purpose of demonstrating common issues in computational method development and evaluation. We explain what evaluation glitches were necessary for Mad Hatter to reach this annotation level, what is wrong with similar metascores in general, and why metascores may screw up not only method evaluations but also the analysis of biological experiments. This paper may serve as an example of problems in the development and evaluation of machine learning models for metabolite annotation.

Untargeted metabolomics experiments rely on spectral libraries for structure annotation, but these libraries are vastly incomplete; in silico methods search in structure databases but cannot distinguish between correct and incorrect annotations. As biological interpretation relies on accurate structure annotations, the ability to assign confidence to such annotations is a key outstanding problem. We introduce the COSMIC workflow that combines structure database generation, in silico annotation, and a confidence score consisting of kernel density p-value estimation and a Support Vector Machine with enforced directionality of features. In evaluation, COSMIC annotates a substantial number of hits at small false discovery rates, and outperforms spectral library search for this purpose. To demonstrate that COSMIC can annotate structures never reported before, we annotated twelve novel bile acid conjugates; nine structures were confirmed by manual evaluation and two structures using synthetic standards. Second, we annotated and manually evaluated 315 molecular structures in human samples currently absent from the Human Metabolome Database. Third, we applied COSMIC to 17,400 experimental runs and annotated 1,715 structures with high confidence that were absent from spectral libraries.

Untargeted metabolomics experiments rely on spectral libraries for structure annotation, but, typically, only a small fraction of spectra can be matched. Previous in silico methods search in structure databases but cannot distinguish between correct and incorrect annotations. Here we introduce the COSMIC workflow that combines in silico structure database generation and annotation with a confidence score consisting of kernel density P value estimation and a support vector machine with enforced directionality of features. On diverse datasets, COSMIC annotates a substantial number of hits at low false discovery rates and outperforms spectral library search. To demonstrate that COSMIC can annotate structures never reported before, we annotated 12 natural bile acids. The annotation of nine structures was confirmed by manual evaluation and two structures using synthetic standards. In human samples, we annotated and manually validated 315 molecular structures currently absent from the Human Metabolome Database. Application of COSMIC to data from 17,400 metabolomics experiments led to 1,715 high-confidence structural annotations that were absent from spectral libraries.

Untargeted mass spectrometry is employed to detect small molecules in complex biospecimens, generating data that are difficult to interpret. We developed Qemistree, a data exploration strategy based on the hierarchical organization of molecular fingerprints predicted from fragmentation spectra. Qemistree allows mass spectrometry data to be represented in the context of sample metadata and chemical ontologies. By expressing molecular relationships as a tree, we can apply ecological tools that are designed to analyze and visualize the relatedness of DNA sequences to metabolomics data. Here we demonstrate the use of tree-guided data exploration tools to compare metabolomics samples across different experimental conditions such as chromatographic shifts. Additionally, we leverage a tree representation to visualize chemical diversity in a heterogeneous collection of samples. The Qemistree software pipeline is freely available to the microbiome and metabolomics communities in the form of a QIIME2 plugin, and a global natural products social molecular networking workflow. [Figure not available: see fulltext.]

Metabolomics using nontargeted tandem mass spectrometry can detect thousands of molecules in a biological sample. However, structural molecule annotation is limited to structures present in libraries or databases, restricting analysis and interpretation of experimental data. Here we describe CANOPUS (class assignment and ontology prediction using mass spectrometry), a computational tool for systematic compound class annotation. CANOPUS uses a deep neural network to predict 2,497 compound classes from fragmentation spectra, including all biologically relevant classes. CANOPUS explicitly targets compounds for which neither spectral nor structural reference data are available and predicts classes lacking tandem mass spectrometry training data. In evaluation using reference data, CANOPUS reached very high prediction performance (average accuracy of 99.7% in cross-validation) and outperformed four baseline methods. We demonstrate the broad utility of CANOPUS by investigating the effect of microbial colonization in the mouse digestive system, through analysis of the chemodiversity of different Euphorbia plants and regarding the discovery of a marine natural product, revealing biological insights at the compound class level.

SIRIUS 4 is the best-in-class computational tool for metabolite identification from high-resolution tandem mass spectrometry data. It offers de novo molecular formula annotation with outstanding accuracy. When searching fragmentation spectra in a structure database, it reaches over 70% correct identifications. A predicted fingerprint, which indicates the presence or absence of thousands of molecular properties, helps to deduce information about the compound of interest even if it is not contained in any structure database. Here, we present best practices and describe how to leverage the full potential of SIRIUS 4, how to incorporate it into your own workflow, and how it adds value to the analysis of mass spectrometry data beyond spectral library search.

SIRIUS 4 is the best-in-class computational tool for metabolite identification from high-resolution tandem mass spectrometry data. It offers de novo molecular formula annotation with outstanding accuracy. When searching fragmentation spectra in a structure database, it reaches over 70% correct identifications. A predicted fingerprint, which indicates the presence or absence of thousands of molecular properties, helps to deduce information about the compound of interest even if it is not contained in any structure database. Here, we present best practices and describe how to leverage the full potential of SIRIUS 4, how to incorporate it into your own workflow, and how it adds value to the analysis of mass spectrometry data beyond spectral library search.

Cyanobacterial regulation of gene expression must contend with a genome organization that lacks apparent functional context, as the majority of cellular processes and metabolic pathways are encoded by genes found at disparate locations across the genome and relatively few transcription factors exist. In this study, global transcript abundance data from the model cyanobacterium Synechococcus sp. PCC 7002 grown under 42 different conditions was analyzed using Context-Likelihood of Relatedness (CLR). The resulting network, organized into 11 modules, provided insight into transcriptional network topology as well as grouping genes by function and linking their response to specific environmental variables. When used in conjunction with genome sequences, the network allowed identification and expansion of novel potential targets of both DNA binding proteins and sRNA regulators. These results offer a new perspective into the multi-level regulation that governs cellular adaptations of the fast-growing physiologically robust cyanobacterium Synechococcus sp. PCC 7002 to changing environmental variables. It also provides a methodological high-throughput approach to studying multi-scale regulatory mechanisms that operate in cyanobacteria. Finally, it provides valuable context for integrating systems-level data to enhance gene grouping based on annotated function, especially in organisms where traditional context analyses cannot be implemented due to lack of operon-based functional organization.

Motivation Metabolites, small molecules that are involved in cellular reactions, provide a direct functional signature of cellular state. Untargeted metabolomics experiments usually rely on tandem mass spectrometry to identify the thousands of compounds in a biological sample. Recently, we presented CSI:FingerID for searching in molecular structure databases using tandem mass spectrometry data. CSI:FingerID predicts a molecular fingerprint that encodes the structure of the query compound, then uses this to search a molecular structure database such as PubChem. Scoring of the predicted query fingerprint and deterministic target fingerprints is carried out assuming independence between the molecular properties constituting the fingerprint. Results We present a scoring that takes into account dependencies between molecular properties. As before, we predict posterior probabilities of molecular properties using machine learning. Dependencies between molecular properties are modeled as a Bayesian tree network; the tree structure is estimated on the fly from the instance data. For each edge, we also estimate the expected covariance between the two random variables. For fixed marginal probabilities, we then estimate conditional probabilities using the known covariance. Now, the corrected posterior probability of each candidate can be computed, and candidates are ranked by this score. Modeling dependencies improves identification rates of CSI:FingerID by 2.85 percentage points. Availability and implementation The new scoring Bayesian (fixed tree) is integrated into SIRIUS 4.0 (https://bio.informatik.uni-jena.de/software/sirius/).

Mass spectrometry is a predominant experimental technique in metabolomics and related fields, but metabolite structural elucidation remains highly challenging. We report SIRIUS 4 (https://bio.informatik.uni-jena.de/sirius/), which provides a fast computational approach for molecular structure identification. SIRIUS 4 integrates CSI:FingerID for searching in molecular structure databases. Using SIRIUS 4, we achieved identification rates of more than 70% on challenging metabolomics datasets. SIRIUS 4 is a fast and highly accurate tool for molecular structure interpretation from mass-spectrometry-based metabolomics data.

Filamentous anoxygenic phototrophs (FAPs) are abundant members of microbial mat communities inhabiting neutral and alkaline geothermal springs. Natural populations of FAPs related to Chloroflexus spp. and Roseiflexus spp. have been well characterized in Mushroom Spring, where they occur with unicellular cyanobacteria related to Synechococcus spp. strains A and B′. Metatranscriptomic sequencing was applied to the microbial community to determine how FAPs regulate their gene expression in response to fluctuating environmental conditions and resource availability over a diel period. Transcripts for genes involved in the biosynthesis of bacteriochlorophylls (BChls) and photosynthetic reaction centers were much more abundant at night. Both Roseiflexus spp. and Chloroflexus spp. expressed key genes involved in the 3-hydroxypropionate (3-OHP) carbon dioxide fixation bi-cycle during the day, when these FAPs have been thought to perform primarily photoheterotrophic and/or aerobic chemoorganotrophic metabolism. The expression of genes for the synthesis and degradation of storage polymers, including glycogen, polyhydroxyalkanoates and wax esters, suggests that FAPs produce and utilize these compounds at different times during the diel cycle. We summarize these results in a proposed conceptual model for temporal changes in central carbon metabolism and energy production of FAPs living in a natural environment. The model proposes that, at night, Chloroflexus spp. and Roseiflexus spp. synthesize BChl, components of the photosynthetic apparatus, polyhydroxyalkanoates and wax esters in concert with fermentation of glycogen. It further proposes that, in daytime, polyhydroxyalkanoates and wax esters are degraded and used as carbon and electron reserves to support photomixotrophy via the 3-OHP bi-cycle.

Glycogen and compatible solutes are the major polymeric and soluble carbohydrates in cyanobacteria and function as energy reserves and osmoprotectants, respectively. Glycogen synthase null mutants (glgA-I glgA-II) were constructed in the cyanobacterium Synechococcus sp. strain PCC 7002. Under standard conditions the double mutant produced no glycogen and more soluble sugars. When grown under hypersaline conditions, the glgA-I glgA-II mutant accumulated 1.8-fold more soluble sugars (sucrose and glucosylglycer-(ol/ate)) than WT, and these cells spontaneously excreted soluble sugars into the medium at high levels without the need for additional transporters. An average of 27% more soluble sugars was released from the glgA-I glgA-II mutant than WT by hypo-osmotic shock. Extracellular vesicles budding from the outer membrane were observed by transmission electron microscopy in glgA-I glgA-II cells grown under hypersaline conditions. The glgA-I glgA-II mutant serves as a starting point for developing cell factories for photosynthetic production and excretion of sugars.

Synechococcus sp. strain PCC 7002 has been gaining significance as both a model system for photosynthesis research and for industrial applications. Until recently, the genetic toolbox for this model cyanobacterium was rather limited and relied primarily on tools that only allowed constitutive gene expression. This work describes a two-plasmid, Zn 2+ -inducible expression platform that is coupled with a zurA mutation, providing enhanced Zn 2+ uptake. The control elements are based on the metal homeostasis system of a class II metallothionein gene ( smtA 7942 ) and its cognate SmtB 7942 repressor from Synechococcus elongatus strain PCC 7942. Under optimal induction conditions, yellow fluorescent protein (YFP) levels were about half of those obtained with the strong, constitutive phycocyanin ( cpcBA 6803 ) promoter of Synechocystis sp. strain PCC 6803. This metal-inducible expression system in Synechococcus sp. strain PCC 7002 allowed the titratable gene expression of YFP that was up to 19-fold greater than the background level. This system was utilized successfully to control the expression of the Drosophila melanogaster β-carotene 15,15′-dioxygenase, NinaB, which is toxic when constitutively expressed from a strong promoter in Synechococcus sp. strain PCC 7002. Together, these properties establish this metal-inducible system as an additional useful tool that is capable of controlling gene expression for applications ranging from basic research to synthetic biology in Synechococcus sp. strain PCC 7002. IMPORTANCE This is the first metal-responsive expression system in cyanobacteria, to our knowledge, that does not exhibit low sensitivity for induction, which is one of the major hurdles for utilizing this class of genetic tools. In addition, high levels of expression can be generated that approximate those of established constitutive systems, with the added advantage of titratable control. Together, these properties establish this Zn 2+ -inducible system, which is based on the smtA 7942 operator/promoter and smtB 7942 repressor, as a versatile gene expression platform that expands the genetic toolbox of Synechococcus sp. strain PCC 7002.

Most biological nitrogen (N 2 ) fixation results from the activity of a molybdenum-dependent nitrogenase, a complex iron-sulfur enzyme found associated with a diversity of bacteria and some methanogenic archaea. Azotobacter vinelandii , an obligate aerobe, fixes nitrogen via the oxygen-sensitive Mo nitrogenase but is also able to fix nitrogen through the activities of genetically distinct alternative forms of nitrogenase designated the Vnf and Anf systems when Mo is limiting. The Vnf system appears to replace Mo with V, and the Anf system is thought to contain Fe as the only transition metal within the respective active site metallocofactors. Prior genetic analyses suggest that a number of nif -encoded components are involved in the Vnf and Anf systems. Genome-wide transcription profiling of A. vinelandii cultured under nitrogen-fixing conditions under various metal amendments (e.g., Mo or V) revealed the discrete complement of genes associated with each nitrogenase system and the extent of cross talk between the systems. In addition, changes in transcript levels of genes not directly involved in N 2 fixation provided insight into the integration of central metabolic processes and the oxygen-sensitive process of N 2 fixation in this obligate aerobe. The results underscored significant differences between Mo-dependent and Mo-independent diazotrophic growth that highlight the significant advantages of diazotrophic growth in the presence of Mo.

Synechococcus sp. strain PCC 7002 is a robust, genetically tractable cyanobacterium that produces six different xanthophyll carotenoids (zeaxanthin, cryptoxanthin, myxoxanthophyll (myxol-2'-fucoside), echinenone, 3'-hydroxyechinenone, and synechoxanthin) and tolerates many environmental stresses, including high light intensities. Targeted mutations were introduced to block the branches of the carotenoid biosynthetic pathway leading to specific xanthophylls, and a mutant lacking all xanthophylls was constructed. Some of the mutants showed severe growth defects at high light intensities, and multi-locus mutants had somewhat lower chlorophyll contents and lower photosystem I levels. The results suggested that xanthophylls, particularly zeaxanthin and echinenone, might play regulatory roles in thylakoid biogenesis. Measurements of reactive oxygen (ROS) and nitrogen (RNS) species in the mutants showed that all xanthophylls participate in preventing ROS/RNS accumulation and that a mutant lacking all xanthophylls accumulated very high levels of ROS/RNS. Results from transcription profiling showed that mRNA levels for most genes encoding the enzymes of carotenogenesis are significantly more abundant after exposure to high light. These studies indicated that all xanthophylls contribute to protection against photo-oxidative stress.

[NiFe] hydrogenases are widespread among microorganisms and catalyze the reversible cleavage of molecular hydrogen. However, only a few bacteria, such as Ralstonia eutropha H16 (Re), synthesize [NiFe] hydrogenases that perform H(2) cycling in the presence of O(2). These enzymes are of special interest for biotechnological applications. To gain further insight into the mechanism(s) responsible for the remarkable O(2) tolerance, we employ FTIR and EPR spectroscopy to study mutant variants of the membrane-bound hydrogenase (MBH) of Re-carrying substitutions of a particular cysteine residue in the vicinity of the [NiFe] active site that is characteristic of O(2)-tolerant membrane-bound [NiFe] hydrogenases. We demonstrate that these MBH variants, despite minor changes in the electronic structure and in the interaction behavior with the embedding protein matrix, display all relevant catalytic and noncatalytic states of the wild-type enzyme, as long as they are still located in the cytoplasmic membrane. Notably, in the oxidized Ni(r)-B state and the fully reduced forms, the CO stretching frequency increases with increasing polarity of the respective amino acid residue at the specific position of the cysteine residue. We purified the MBH mutant protein with a cysteine-to-alanine exchange to apparent homogeneity as dimeric enzyme after detergent solubilization from the membrane. This purified version displays increased oxygen sensitivity, which is reflected by detection of the oxygen-inhibited Ni(u)-A state, an irreversible inactive redox state, and the light-induced Ni(a)-L state even at room temperature.

This study provides the first spectroscopic characterization of the membrane-bound oxygen-tolerant [NiFe] hydrogenase (MBH) from Ralstonia eutropha H16 in its natural environment, the cytoplasmic membrane. The H2-converting MBH is composed of a large subunit, harboring the [NiFe] active site, and a small subunit, capable in coordinating one [3Fe4S] and two [4Fe4S] clusters. The hydrogenase dimer is electronically connected to a membrane-integral cytochrome b. EPR and Fourier transform infrared spectroscopy revealed a strong similarity of the MBH active site with known [NiFe] centers from strictly anaerobic hydrogenases. Most redox states characteristic for anaerobic [NiFe] hydrogenases were identified except for one remarkable difference. The formation of the oxygen-inhibited Niu-A state was never observed. Furthermore, EPR data showed the presence of an additional paramagnetic center at high redox potential (+290 mV), which couples magnetically to the [3Fe4S] center and indicates a structural and/or redox modification at or near the proximal [4Fe4S] cluster. Additionally, significant differences regarding the magnetic coupling between the Nia-C state and [4Fe4S] clusters were observed in the reduced form of the MBH. The spectroscopic properties are discussed with regard to the unusual oxygen tolerance of this hydrogenase and in comparison with those of the solubilized, dimeric form of the MBH.

[NiFe] hydrogenases catalyze the reversible conversion of H2 into protons and electrons. The reaction takes place at the active site, which is composed of a nickel and an iron atom and three diatomic ligands, two cyanides and one carbon monoxide, bound to the iron. The NiFe(CN-)2CO cofactor is synthesized by an intricate posttranslational maturation process, which is mediated by a set of six conserved Hyp proteins. Depending on the cellular location and the physiological function, additional auxiliary proteins are involved in hydrogenase biosynthesis. Here we present evidence that the auxiliary proteins HoxL and HoxV assist in assembly of the Fe(CN-)2CO moiety. This unit was identified as a cofactor intermediate of the oxygen-tolerant membrane-bound [NiFe] hydrogenase (MBH) in the beta-proteobacterium Ralstonia eutropha H16. Both HoxL and HoxV proved to be essential for H2-oxidizing activity and MBH-driven growth on H2. Copurification studies revealed that HoxL and HoxV directly interact with the hydrogenase apoprotein. HoxV forms complexes with HoxL and HypC, a HoxL paralogue that is essential for cofactor assembly. These observations suggest that HoxL acts as a specific chaperone assisting the transfer of the Fe(CN-)2CO cofactor intermediate from the Hyp machinery to the MBH. This shuttle also involves the scaffold protein HoxV. Indeed, infrared spectroscopy and metal analysis identified for the first time a non-redox-active Fe(CN-)2CO intermediate coordinated to HoxV.

Knallgas bacteria such as certain Ralstonia spp. are able to obtain metabolic energy by oxidizing trace levels of H2 using O2 as the terminal electron acceptor. The [NiFe] hydrogenases produced by these organisms are unusual in their ability to oxidize H2 in the presence of O2, which is a potent inactivator of most hydrogenases through attack at the active site. To probe the origin of this unusual O2 tolerance, we conducted a study on the membrane-bound hydrogenase from Ralstonia eutropha H16 and that of the closely related organism Ralstonia metallidurans CH34, which was purified using a new heterologous overproduction system. Direct electrochemical methods were used to determine apparent inhibition constants for O2 inhibition of H2 oxidation (K I(app)O2) for each enzyme. These values were at least 2 orders of magnitude higher than those of "standard" [NiFe] hydrogenases. Amino acids close to the active site were exchanged in the membrane-bound hydrogenase of R. eutropha H16 for those from standard hydrogenases to probe the role of individual residues in conferring O2 sensitivity. Michaelis constants for H2 (K M H2) were determined, and for some mutants these were increased more than 20-fold relative to the wild type. Mutations resulting in membrane-bound hydrogenase enzymes with increased K M H2 or decreased K I(app)O2 values were associated with impaired lithoautotrophic growth in the presence of high O2 concentrations.

Northern analysis was employed to investigate mRNA produced by mutant strains of Azotobacter vinelandii with defined deletions in the nif structural genes and in the intergenic noncoding regions. The results indicate that intergenic RNA secondary structures effect the differential accumulation of transcripts, supporting the high Fe protein-to-MoFe protein ratio required for optimal diazotrophic growth.

Multiple Genome-wide Alignments are a first crucial step to compare genomes. Gain and loss of genes, duplications and genomic rearrangements are challenging problems that aggravate with increasing phylogenetic distances. We describe a multiple genome-wide alignment tool for bacteria, called POMAGO, which is based on orthologous genes and their syntenic information determined by Proteinortho.This strategy enables POMAGO to efficiently define anchor points even across wide phylogenetic distances and outperform existing approaches in this field of application. The given set of orthologous genes is enhanced by several cleaning and completion steps, including the addition of previously undetected orthologous genes. Protein-coding genes are aligned on nucleotide and protein level, whereas intergenic regions are aligned on nucleotide level only. We tested and compared our program at three very different sets of bacteria that exhibit different degrees of phylogenetic distances: 1) 15 closely related, well examined and described E. coli species, 2) six more divergent Aquificales, as putative basal bacteria, and 3) a set of eight extreme divergent species, distributed among the whole phylogenetic tree of bacteria. POMAGO is written in a modular way which allows extending or even exchanging algorithms in different stages of the alignment process. Intergenic regions might for instance be aligned using an RNA secondary structure aware algorithm rather than to rely on sequence data alone. The software is freely available from http://www.rna.uni-jena.de/supplements/pomago

Metabolomics complements investigation of the genome, transcriptome, and proteome of an organism. Today, the vast majority of metabolites remain unknown, in particular for non-model organisms. Mass spectrometry is one of the predominant techniques for analyzing small molecules such as metabolites. A fundamental step for identifying a small molecule is to determine its molecular formula.

An uncultured member of the phylum Chlorobi, provisionally named 'Candidatus Thermochlorobacter aerophilum', occurs in the microbial mats of alkaline siliceous hot springs at the Yellowstone National Park. 'Ca. T. aerophilum' was investigated through metagenomic and metatranscriptomic approaches. 'Ca. T. aerophilum' is a member of a novel, family-level lineage of Chlorobi, a chlorophototroph that synthesizes type-1 reaction centers and chlorosomes similar to cultivated relatives among the green sulfur bacteria, but is otherwise very different physiologically. 'Ca. T. aerophilum' is proposed to be an aerobic photoheterotroph that cannot oxidize sulfur compounds, cannot fix N(2), and does not fix CO(2) autotrophically. Metagenomic analyses suggest that 'Ca. T. aerophilum' depends on other mat organisms for fixed carbon and nitrogen, several amino acids, and other important nutrients. The failure to detect bchU suggests that 'Ca. T. aerophilum' synthesizes bacteriochlorophyll (BChl) d, and thus it occupies a different ecological niche than other chlorosome-containing chlorophototrophs in the mat. Transcription profiling throughout a diel cycle revealed distinctive gene expression patterns. Although 'Ca. T. aerophilum' probably photoassimilates organic carbon sources and synthesizes most of its cell materials during the day, it mainly transcribes genes for BChl synthesis during late afternoon and early morning, and it synthesizes and assembles its photosynthetic apparatus during the night.

The genome of the unicellular, euryhaline cyanobacterium Synechococcus sp. PCC 7002 encodes about 3200 proteins. Transcripts were detected for nearly all annotated open reading frames by a global transcriptomic analysis by Next-Generation (SOLiDTM) sequencing of cDNA. In the cDNA samples sequenced, ~90% of the mapped sequences were derived from the 16S and 23S ribosomal RNAs and ~10% of the sequences were derived from mRNAs. In cells grown photoautotrophically under standard conditions (38 °C, 1% (v/v) CO2 in air, 250 µmol photons m-2 s-1), the highest transcript levels (up to 2% of the total mRNA for the most abundantly transcribed genes (e. g., cpcAB, psbA, psaA)) were generally derived from genes encoding structural components of the photosynthetic apparatus. High light exposure for one hour caused changes in transcript levels for genes encoding proteins of the photosynthetic apparatus, Type-1 NADH dehydrogenase complex and ATP synthase, whereas dark incubation for one hour resulted in a global decrease in transcript levels for photosynthesis-related genes and an increase in transcript levels for genes involved in carbohydrate degradation. Transcript levels for pyruvate kinase and the pyruvate dehydrogenase complex decreased sharply in cells incubated in the dark. Under dark anoxic (fermentative) conditions, transcript changes indicated a global decrease in transcripts for respiratory proteins and suggested that cells employ an alternative phosphoenolpyruvate degradation pathway via phosphoenolpyruvate synthase (ppsA) and the pyruvate:ferredoxin oxidoreductase (nifJ). Finally, the data suggested that an apparent operon involved in tetrapyrrole biosynthesis and fatty acid desaturation, acsF2-ho2-hemN2-desF, may be regulated by oxygen concentration.

Cyanobacterium Synechococcus sp. PCC 7002 contains a single gene (glbN) coding for GlbN, a protein of the 2/2 hemoglobin lineage. The precise function of GlbN is not known, but comparison to similar 2/2 hemoglobins suggests that reversible dioxygen binding is not its main activity. In this report, the results of in vitro and in vivo experiments probing the role of GlbN are presented. Transcription profiling indicated that glbN is not strongly regulated under any of a large number of growth conditions and that the gene is probably constitutively expressed. High levels of nitrate, used as the sole source of nitrogen, and exposure to nitric oxide were tolerated better by the wild-type strain than a glbN null mutant, whereas overproduction of GlbN in the null mutant background restored the wild-type growth. The cellular contents of reactive oxygen/nitrogen species were elevated in the null mutant under all conditions and were highest under NO challenge or in the presence of high nitrate concentrations. GlbN overproduction attenuated these contents significantly under the latter conditions. The analysis of cell extracts revealed that the heme of GlbN was covalently bound to overproduced GlbN apoprotein in cells grown under microoxic conditions. A peroxidase assay showed that purified GlbN does not possess significant hydrogen peroxidase activity. It was concluded that GlbN protects cells from reactive nitrogen species that could be encountered naturally during growth on nitrate or under denitrifying conditions. The solution structure of covalently modified GlbN was determined and used to rationalize some of its chemical properties.

[NiFe]-hydrogenases catalyze the oxidation of H(2) to protons and electrons. This reversible reaction is based on a complex interplay of metal cofactors including the Ni-Fe active site and several [Fe-S] clusters. H(2) catalysis of most [NiFe]-hydrogenases is sensitive to dioxygen. However, some bacteria contain hydrogenases that activate H(2) even in the presence of O(2). There is now compelling evidence that O(2) affects hydrogenase on three levels: 1) H(2) catalysis, 2) hydrogenase maturation, and 3) H(2)-mediated signal transduction. Herein, we summarize the genetic, biochemical, electrochemical, and spectroscopic properties related to the O(2) tolerance of hydrogenases resident in the facultative chemolithoautotroph Ralstonia eutropha H16. A focus is given to the membrane-bound [NiFe]-hydogenase, which currently represents the best-characterized member of O(2)-tolerant hydrogenases.

Knallgas bacteria such as certain Ralstonia spp. are able to obtain metabolic energy by oxidizing trace levels of H2 using O2 as the terminal electron acceptor. The [NiFe] hydrogenases produced by these organisms are unusual in their ability to oxidize H2 in the presence of O2, which is a potent inactivator of most hydrogenases through attack at the active site. To probe the origin of this unusual O2 tolerance, we conducted a study on the membrane-bound hydrogenase from Ralstonia eutropha H16 and that of the closely related organism Ralstonia metallidurans CH34, which was purified using a new heterologous overproduction system. Direct electrochemical methods were used to determine apparent inhibition constants for O2 inhibition of H2 oxidation (K I(app)O2) for each enzyme. These values were at least 2 orders of magnitude higher than those of "standard" [NiFe] hydrogenases. Amino acids close to the active site were exchanged in the membrane-bound hydrogenase of R. eutropha H16 for those from standard hydrogenases to probe the role of individual residues in conferring O2 sensitivity. Michaelis constants for H2 (K M H2) were determined, and for some mutants these were increased more than 20-fold relative to the wild type. Mutations resulting in membrane-bound hydrogenase enzymes with increased K M H2 or decreased K I(app)O2 values were associated with impaired lithoautotrophic growth in the presence of high O2 concentrations.

Hydrogenases are important enzymes in the energy metabolism of microorganisms. Therefore, they are widespread in prokaryotes. We analyzed the occurrence of hydrogenases in cyanobacteria and deduced a FeFe-hydrogenase in three different heliobacterial strains. This allowed the first phylogenetic analysis of the hydrogenases of all five major groups of photosynthetic bacteria (heliobacteria, green nonsulfur bacteria, green sulfur bacteria, photosynthetic proteobacteria, and cyanobacteria). In the case of both hydrogenases found in cyanobacteria (uptake and bidirectional), the green nonsulfur bacterium Chloroflexus aurantiacus was found to be the closest ancestor. Apart from a close relation between the archaebacterial and the green sulfur bacterial sulfhydrogenase, we could not find any evidence for horizontal gene transfer. Therefore, it would be most parsimonious if a Chloroflexus-like bacterium was the ancestor of Chloroflexus aurantiacus and cyanobacteria. After having transmitted both hydrogenase genes vertically to the different cyanobacterial species, either no, one, or both enzymes were lost, thus producing the current distribution. Our data and the available data from the literature on the occurrence of cyanobacterial hydrogenases show that the cyanobacterial uptake hydrogenase is strictly linked to the occurrence of the nitrogenase. Nevertheless, we did identify a nitrogen-fixing Synechococcus strain without an uptake hydrogenase. Since we could not find genes of a FeFe-hydrogenase in any of the tested cyanobacteria, although strains performing anoxygenic photosynthesis were also included in the analysis, a cyanobacterial origin of the contemporary FeFe-hydrogenase of algal plastids seems unlikely.