Ron Wehrens

LC-MS based global metabolite profiling of grapes: solvent extraction protocol optimisation

Optimal solvent conditions for grape sample preparation were investigated for the purpose of metabolite profiling studies, with the aim of obtaining as many features as possible with the best analytical repeatability. Mixtures of water, methanol and chloroform in different combinations were studied as solvents for the extraction of ground grapes. The experimental design used a two stage study to find the optimum extraction medium. The extracts obtained were further purified using solid phase extraction and analysed using a UPLC full scan TOF MS with both reversed phase and hydrophilic interaction chromatography. The data obtained were processed using data extraction algorithms and advanced statistical software for data mining. The results obtained indicated that a fairly broad optimal area for solvent composition could be identified, containing approximately equal amounts of methanol and chloroform and up to 20% water. Since the water content of the samples was variable, the robustness of the optimal conditions suggests these are appropriate for large scale profiling studies for characterisation of the grape metabolome.

Stability-based biomarker selection.

Biomarker identification, i.e., finding those variables that indicate true differences between two or more populations, is an ever more important topic in the omics sciences. In most cases, the number of variables far exceeds the number of samples, making biomarker identification extremely difficult. We present a strategy based on the stability of putative biomarkers under perturbation of the data, and show that in several cases important gains can be achieved. The strategy is very general and can be applied with all common biomarker identification methods; it also has the advantage that it does not rely on error estimates from crossvalidation, that in this setting tend to be highly variable.

metaMS: An open-source pipeline for GC–MS-based untargeted metabolomics

Untargeted metabolomics are rapidly becoming an important tool for studying complex biological samples. Gas chromatography-mass spectrometry (GC-MS) is the most widely used analytical technology for metabolomic analysis of compounds that are volatile or can be chemically derivatised into volatile compounds. Unfortunately, data processing and analysis are not straightforward and the field is dominated by vendor-supplied software that does not always allow easy integration for large laboratories with different instruments. This paper presents an open-source pipeline for high-throughput GC-MS data processing, written in the R language and available as package metaMS. It features rapid annotation using in-house databases, and also provides support for building and validating such databases. The results are presented in simple-to-use tables, summarising the relative concentrations of identified compounds and unknowns in all samples. The use of the pipeline is illustrated using three experimental data sets.

Carbon, hydrogen and oxygen stable isotope ratios of whole wood, cellulose and lignin methoxyl groups of Picea abies as climate proxies

RATIONALE Carbon, hydrogen and oxygen (C, H and O) stable isotope ratios of whole wood and components are commonly used as paleoclimate proxies. In this work we consider eight different proxies in order to discover the most suitable wood component and stable isotope ratio to provide the strongest climate signal in Picea abies in a southeastern Alpine region (Trentino, Italy). METHODS δ(13)C, δ(18)O and δ(2)H values in whole wood and cellulose, and δ(13)C and δ(2)H values in lignin methoxyl groups were measured. Analysis was performed using an Isotopic Ratio Mass Spectrometer coupled with an Elemental Analyser for measuring (13)C/(12)C and a Pyrolyser for measuring (2)H/(1)H and (18)O/(16)O. The data were evaluated by Principal Component Analysis, and a simple Pearsons correlation between isotope chronologies and climatic features, and multiple linear regression were performed to evaluate the data. RESULTS Each stable isotope ratio in cellulose and lignin methoxyl differs significantly from the same stable isotope ratio in whole wood, the values begin higher in cellulose and lignin except for the lignin δ(2)H values. Significant correlations were found between the whole wood and the cellulose fractions for each isotope ratio. Overall, the highest correlations with temperature were found with the δ(18)O and δ(2)H values in whole wood, whereas no significant correlations were found between isotope proxies and precipitation. CONCLUSIONS δ(18)O and δ(2)H values in whole wood provide the best temperature signals in Picea abies in the northern Italian study area. Extraction of cellulose and lignin and analysis of other isotopic ratios do not seem to be necessary.

A benchmark spike-in data set for biomarker identification in metabolomics

The development and the validation of innovative approaches for biomarker selection are of paramount importance in many ‐omics technologies. Unfortunately, the actual testing of new methods on real data is difficult, because in real data sets, one can never be sure about the “true” biomarkers. In this paper, we present a publicly available metabolomic ultra performance liquid chromatography–mass spectrometry spike‐in data set for apples. The data set consists of 10 control samples and three spiked sets of the same size, where naturally occurring compounds are added in different concentrations. In this sense, the data set can serve as a test bed to assess the performance of new algorithms and compare them with previously published results.

High-throughput carotenoid profiling using multivariate curve resolution

We present automated data analysis of high-throughput high-performance liquid chromatography with diode array detection (HPLC-DAD) data using multivariate curve resolution. This technique provides spectra and elution profiles of all UV-Vis active compounds present in the mixture. The specifics of using this method in noninteractive fashion are discussed. A case study on the stability of isoprenoids in grape extracts under two different experimental regimes serves to illustrate the potential of the method: quantitative results clearly show that the addition of triethylamine is beneficial in that carotenoid, chlorophyll, and tocopherol compounds are much more stable and in this way can be kept up to at least 30 days without any sign of degradation.

Stable isotope ratios of H, C, N and O in Italian citrus juices†

Stable isotope ratios (SIRs) of C, N, H and O have been exensively used in fruit juices quality control (ENV and AOAC methods) to detect added sugar and the watering down of concentrated juice, practices prohibited by European legislation (EU Directive 2012/12). The European Fruit Juice Association (AIJN) set some reference guidelines in order to allow the judging of the genuiness of a juice. Moreover, various studies have been carried out to determine the natural variability of SIRs in fruit juices, but none of these has investigated SIRs extensively in authentic citrus juices from Italy. In this work, about 500 citrus juice samples were officially collected in Italy by the Italian Ministry of Agricultural and Forestry Policies from 1998 onwards. (D/H)(I) and (D/H)(II) in ethanol and δ(13) C(ethanol), δ(13) C(pulp), δ(13) C(sugars), δ(18) O(vegetalwater), δ(15) N(pulp), and δ(18) O(pulp) were determined using Site-Specific Natural Isotope Fractionation-Nuclear Magnetic Resonance and Isotope Ratio Mass Spectrometry, respectively. The characteristic ranges of variability in SIRs in genuine Italian citrus juice samples are here presented as well as their relationships and compliance with the limits indicated by the AIJN and others proposed in the literature. In particular, the Italian range of values was found to be not completely in agreement with AIJN guidelines, with the risk that genuine juices could be judged as not genuine. Variety seems not to influence SIRs, whereas harvest year and region of origin have some influence on the different ratios, although their data distribution shows overlapping when principal component analysis is applied.

Metabolite profiling in LC-DAD using multivariate curve resolution: the alsace package for R

For those chemical compounds absorbing in the UV–Vis region and not readily applicable to routine mass spectrometry ionisation methods, liquid chromatography coupled to diode array detection is a convenient platform to perform metabolite profiling. Data processing by hand is labour-intensive and error prone. In the present study a strategy based on multivariate curve resolution, and its implementation in an R package called alsace are described. The final result of an analysis is a table containing peak heights or peak areas for all features of the individual injections. The capabilities of the software, providing elements such as splitting the data into separate, possibly overlapping time windows, merging the results of the individual time windows, and parametric time warping to align features, are illustrated using a cassava-derived data set.

Metabolic Biomarker Identification with Few Samples

Biomarker selection represents a key step in bioinformatic data processing pipelines; examples range from DNA microarrays (Tusher et al., 2001; Yousef et al., 2009) to proteomics (Araki et al., 2010; Oh et al., 2011) to metabolomics (Chadeau-Hyam et al., 2010). Meaningful biological interpretation is greatly aided by identification of a “short-list” of features – biomarkers – characterizing themain differences between several states in a biological system. In a two-class setting the biomarkers are those variables (metabolites, proteins, genes ...) that allow discrimination between the classes. A class or group tag can be used to distinguish many situations: it can be used to discriminate between treated and non-treated samples, to mark different varieties of the same organism, etcetera. In the following, we will – for clarity – restrict the discussion to metabolomics, and the variables will constitute concentration levels of metabolites, but similar arguments hold mutatis mutandis for other -omics sciences, such as proteomics and transcriptomics, where the variables correspond to protein levels or expression levels, respectively.

Oxygen and Hydrogen Stable Isotope Ratios of Bulk Needles Reveal the Geographic Origin of Norway Spruce in the European Alps

Background Tracking timber is necessary in order to prevent illegal logging and protect local timber production, but there is as yet no suitable analytical traceability method. Stable isotope ratios in plants are known to reflect geographical variations. In this study we analysed four stable isotope ratios in order to develop a model able to identify the geographic origin of Norway spruce in the European Alps. Methodology and Principal Findings δ18O, δ2H, δ13C and δ15N were measured in bulk needles of Picea abies sampled in 20 sites in and around the European Alps. Environmental and spatial variables were found to be related to the measured isotope ratios. An ordinary least squares regression was used to identify the most important factor in stable isotope variability in bulk needles. Spatial autocorrelation was tested for all isotope ratios by means of Moran’s I. δ18O, δ2H and δ15N values differed significantly between sites. Distance from the coast had the greatest influence on δ2H, while latitude and longitude were strongly related to δ18O. δ13C values did not appear to have any relationship with geographical position, while δ15N values were influenced by distance from the motorway. The regression model improved the explanatory power of the spatial and environmental variables. Positive spatial autocorrelations were found for δ18O and δ2H values. Conclusions The δ 18O, δ2H and δ15N values in P. abies bulk needles are a suitable proxy to identify geographic origin as they vary according to geographical position. Although the regression model showed the explanatory variables to have significant power and stability, we conclude that our model might be improved by multivariate spatial interpolation of the δ 18O and δ2H values.