Western blot analysis

MS-based quantitative proteomics has quantified a number of proteins expressed differentially in 14N-HAB/15N-HAB, 14N-HAB/15N-NAB, and 14N-LAB/15N-NAB comparisons. Relative protein levels of several selected proteins were analyzed by Western blot. Protein mixtures with equal protein content (10-30µg) were first resolved by SDS-PAGE. Subsequently, the separated proteins were transferred onto polyvinylidene fluoride (PVDF) membranes. Western blot analysis was performed with several selected antibodies. The membranes were then incubated with HRP-conjugated secondary antibody. ECL system and film were used for membrane visualization. ECL images were quantified by

QuantityOne software (BioRad).

3.2.7

Metabolomics

The metabolomic analyses shown in this thesis were processed at the ‘Metabolomics Core’ of the University of California, Davis, CA, USA (Dr. Vladimir Tolstikov).

3.2.7.1

Sample preparation

Six mice from each animal line were employed in metabolic studies.

The plasma samples were prepared by using the method described previously (Fiehn and Kind, 2007). Briefly, the plasma proteins were precipitated, and the metabolite extraction was obtained. An aliquot of plasma extract was dried down, and the other aliquots were frozen for recording purposes. The plasma was derivatized by first adding methoxyamine in an aprotic basic solvent and then adding a trimethylsilylating agent. The derivatized sample was analyzed by direct thermodesorption GC-TOF.

The brain tissue samples were prepared as follows: First, the extraction solution was prepared by mixing acetonitrile, isopropanol, and water in the volume proportion 3:3:2. The pH of acetonitrile and isopropanol (pH7) was checked using wetted pH paper. The extraction solution mix was rinsed with small bubbles of argon for 5 min. The Argon line

Material and methods

was flushed out of air before being used for degassing the extraction solvent solution. The extraction solution was pre-cooled at -18°Cto -22°C. Eppendorf tubes with two metal balls (3 mm diameter) containing frozen mouse brain samples were placed in a freezer pre-chilled to -80°C or with liquid nitrogen Eppendorf-holder of the grinder. Immediately afterwards the Eppendorf tubes were put back in liquid nitrogen. 10-50 mg of frozen mouse brain (hippocampus) was homogenized with 500-2.500 μl (or aliquot according to sample aliquot) extraction solution mix for 45 s in 25 ml conical polypropylene tubes in the homogenizer Tissue Master 125. The homogenate was centrifuged at 2500 rpm for 5 min. An aliquoted 250 or 500 μl of supernatant was evaporated in the Labconco Centrivap cold trap concentrator to complete dryness. The dried residue was then re-suspended with 500 μl of acetonitrile : water (1:1 v/v) mixture and centrifuged for 2 min at 14000 rcf in the centrifuge Eppendorf 5415 D. The clear supernatant was for GC-TOF-MS analysis. Samples were dried in the Labconco Centrivap cold trap concentrator to complete dryness and, once dried, stored in darkness under argon. The dried samples were then derivatized for GC/MS profiling. Methyl oxime derivatives were produced by dissolving the dry extracts in 20 μL freshly prepared omethylhydroxylamine·HCl (40 mg/mL in pyridine) and incubated at 30°C for 90 min while being shaken continuously. Subsequent trimethyl silylation was achieved by adding 80 μL of N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA), followed by continuous shaking for 30 min at 37°C.

3.2.7.2

GC-MS data acquisition

GC-TOF-MS analysis was performed by using an Agilent 6890 N gas chromatograph (Palo Alto, CA, USA) interfaced to a time-of-flight (TOF) Pegasus III mass spectrometer (Leco, St. Joseph, MI, USA). The mass spectrometer first was tuned according to the manufacturer’s manuals to achieve optimal parameters for ion lenses, detector voltage, and other settings. Automated injections were performed with a programmable robotic Gerstel MPS2 multipurpose sampler (Mülheim an der Ruhr, Germany). The gas chromatograph (GC) was fitted with both an Agilent injector and a Gerstel temperature-programmed injector, a cooled injection system (model CIS 4) with a Peltier cooling source. An automated liner exchange (ALEX), designed by Gerstel, was used to eliminate cross-contamination from the sample matrix between sample runs. Multiple baffled liners for the GC inlet were deactivated with 1 μL injections of MSTFA. One microliter of each sample was injected in splitless mode, depending on the metabolite concentrations and eventual signal-to-noise ratios in the GC-MS profiles. The Agilent injector temperature was held constant at 250°C

increased at a rate of 10°C/s to a final temperature of 330°C, hold time 10 min). Injections of 1 μL were made in split (1:5) mode (purge time 120 s, purge flow 40 ml/min). Chromatography was performed on an Rtx-5Sil MS column (30 m × 0.25 mm i.d., 0.25 μm film thickness) with an Integra-Guard column (Restek, Bellefonte, PA, USA). Helium carrier gas was used at a constant flow of 1 mL/min. The GC oven temperature program had an initial temperature of 50°C, with a 1 min hold time, and was ramped at 20°C/min to a final temperature of 330°C with a 5 min hold time. Both the transfer line and source temperatures were 250°C. The Pegasus III TOF (Leco, St. Joseph, MI, USA) mass spectrometer ion source operated at -70 kV filament voltage with ion source. After a solvent delay of 350 s, mass spectra were acquired at 20 scans per second with a mass range of 50 to 500 m/z.

3.2.7.3

Data analysis

The data were processed according to the methods described previously (Zou and Tolstikov, 2008). The Xconvert program included in Xcalibur was used to convert the Xcalibur (*.raw) files to netCDF (*.cdf) format. Automatic peak finding, deconvolution, and alignment were performed using XCMS running on the open statistical platform R. Preliminary data were explored by unsupervised methods such as principle component analysis (PCA) and clustering. For PCA, a scree plot (to show the optimal number of eigenvalues), a score plot (to show the most important principal components and visually detect clusters), and a loading plot (to show positive and negative correlations of components) were included for each analysis by using the R package pcaMethods in the Bioconductor project. Cluster analysis of the PCA scores was performed with partitioning methods such as K-means using the function kmeans() in R package stats; hierarchical agglomerative methods such as Ward's method using the function hclust() in R package stats; multiscale bootstrap resampling using R package pvclust; and the model-based clustering approach using R package mclust, which assumes a variety of data models. Maximum likelihood estimation and Bayes criteria were applied to identify the most likely model and number of clusters. All calculations were performed in an R integrated development environment (IDE), RKWard, under Kubuntu 7.10, a Debian Linux operating system, on a quad core Dell OptiPlex 755 workstation (4 x 3.0 GHz CPU speed, 2 x 4 MB L2 cache, 8 GB RAM). The current versions of Kubuntu, R, Bioconductor, XCMS, pcaMethods, stats, pvclust, mclust, GALGO, and RKWard are free open source softwares (FOSS).

Material and methods

The MarkerView 1.1 Software (Applied Biosystems/MDS Sciex, Concord, Ontario, Canada) allows data from several samples to be compared so that differences can be identified. Typical applications include metabolomics, biomarker discovery, metabolite identification, impurity profiling, etc. In the current study, this software was used for data analysis in conjunction with the techniques described above. The program uses multivariate analysis techniques to compare samples and provides both supervised and unsupervised methods. Supervised methods use prior knowledge of the sample groups (for example, affected/altered vs. control) to determine the variables that distinguish the groups. In contrast, unsupervised methods allow the structure within the data to be determined and visualized. The two approaches can be combined, i.e. unsupervised methods can be used to determine the groups, and then supervised methods can be used to confirm the important variables.