2014 SGMS Meeting

Session 1
Chair: Marc J-F Suter, Eawag, Dubendorf

Detlef Günther, ETHZ, Zurich
Elemental and Isotope Analysis – high speed, high resolution, high flexibility, high information density

Session 2 
Chair: Jean-Luc Wolfender, University of Geneva

Peter C Dorrestein, University of California, La Jolla CA, USA 
A community based molecular GPS from microbes to people

Basri Gülbakan, ETHZ, Zurich
Native nano-ESI mass spectrometry studies of aptamer-ligand complexes

Manfred Heller, University of Bern
Quantitative protein measurement of circulating plasma microparticles by data-independent nanoLC-MS ²

Session 3 
Chair: Richard Knochenmuss, Tofwerk, Thun

16:15 - 16:35

Yury O Tsybin, EPFL, Lausanne
Pushing the limits of FTMS performance for improved quantitative proteomics and metabolomics

Andreas Riedo, University of Bern
High resolution chemical mapping of solid samples using a miniature LIMS system 

Blüemlisalp Dinner Buffet

Session 4 
Chair: Andreas Stämpfli, F. Hoffmann-La Roche, Basel

Laura McGregor, Markes International, Llantrisant, UK
Enhanced screening of environmental pollutants in complex matrices by GCxGC-TOFMS with variable-energy electron ionisation 

Emma Schymanski, Eawag, Dubendorf
Predicting, detecting and elucidating benzotriazole metabolites: multiple strategies enhance success

Anton Kaufmann, Kantonales Labor ZH, Zurich
Confirmation of residue findings with LC-HRMS versus LC-MS/MS 

Session 5
Chair: Michael Affolter, Nestlé Research Center, Lausanne

Bertran Gerrits, Novartis Institute for Biomedical Research, Basel
Multidimensional single cell analysis using flying rare earth metals

Laurent Bigler, University of Zurich
Elucidation of herbivore defense mechanisms in Petunia hybrida plants

Plenary Lectures

Elemental and isotope analysis – high speed, high resolution, high flexibility, high information density

Detlef Günther
Trace Element and Micro Analysis
ETHZ
Zurich
Switzerland

Laser ablation as a direct solid sampling technique in combination with ICP mass spectrometry has proved to be successfully applicable for major, minor and trace element analysis and isotope ratio determinations in a variety of samples and the number of instruments installed per year is continuously growing. However, the major applications reported in the literature are in fields where standard reference materials are available or where the task can be fulfilled using the “famous” NIST 61X series for calibration. Therefore, various attempts have been made to investigate the capabilities of LA-ICPMS when using non-matrix matched calibration materials. Therefore, plasma imaging techniques were applied to study the laser-aerosol vaporization at different ICP conditions. Effects of various gas and vapor additions to the ICP and their influence on the signal intensity will be presented. Furthermore, the sequential nature of data acquisition using quadrupole ICPMS has limited the insights into particle vaporization processes. An in-house built ICPTOFMS at 33µs temporal resolution was used to study single pulse ablations and the signal were quantified using liquid calibration introduced via micro droplet generator. 
Increased flexibility in laser ablation has always been limited due to the requirement of an airtight aerosol transport system to the ICP. A recently developed gas exchange device (GED) circumvents these requirements and makes sampling in air possible. The use of such a Q-GED system in combination with a portable laser ablation system, allowing the ablation of samples of any size and geometry, was tested for archeological samples and will be presented.
Finally, a newly designed setup for high spatial resolution imaging of tissue samples will be presented.
 

Mass spectrometry applications in atmospheric chemistry research, with special emphasis on new particle formation

Urs Baltensperger
Laboratory of Atmospheric Chemistry
Paul Scherrer Institute
Villigen
Switzerland

In recent years, mass spectrometry has seen a number of new applications, including the atmospheric pressure interface time of flight (APi-TOF) mass spectrometers. The APi-TOF measures the mass-to-charge ratio (m/z) of ions and ionic clusters of either positive or negative polarity. The sample is taken at atmospheric pressure and enters the instrument via a critical orifice. Then, the ions are accelerated by an electric field and separated by their time-of-flight, which depends on their m/z, until they are counted on a multi-channel plate detector. From the exact mass, the composition of ions and ionic clusters is determined. The addition of a chemical ionization unit in front of the APi-TOF allows also neutral clusters to be investigated.

The APi-TOF has been highly useful in the context of the CLOUD experiment at CERN. CLOUD has been designed to investigate a possible link between galactic cosmic rays and climate. Ions may enhance the stability of clusters and may in this way enhance the nucleation rate of new particles. Once these particles have grown to a diameter of 50 to 100 nm they may act as cloud condensation nuclei, meaning that they are able to form cloud droplets at the supersaturations that typically occur in the atmosphere. This change in the cloud droplet number concentration has an effect on the cloud properties and thus on climate. While the physical principles are well established we are still far away from a quantitative understanding.

The first step in the process chain is the nucleation of new particles. CLOUD, with its extremely well controlled and essentially contaminant free conditions, was able to show that indeed sulfuric acid is an important component for such new particle formation, however, for the typical temperatures encountered in the planetary boundary layer concentrations of sulfuric acid are not high enough to explain the atmospheric observations [1]. Moreover, the effect of ammonia [1], amines [2] and oxidized organic molecules [3,4] on the nucleation rate of sulfuric acid has been investigated in CLOUD so far. In all these cases, the APi-TOF was instrumental in characterizing the clusters as well as their temporal evolution, which was only possible due to the high sensitivity of this instrument.

1. J Kirkby et al, Nature, 476, 42 - 433, 2011
2. J Almeida et al Nature, 502, 359-363, 2013.
3. S Schobesberger et al , Proc Nat Acad Sci, 110, 17223-17228, 2013
4. Riccobono et al, Science, DOI 10.1126/science.1243527, 16 May 2014.

A community based molecular GPS from microbes to people

Peter C Dorrestein
Skaggs School of Pharmacy and Parmaceutical Sciences
University of California
9500 Gilman Drive
La Jolla
CA 92093-0751
USA

Multidimensional single cell analysis using flying rare earth metals

Bertran Gerrits
Developmental and Molecular Pathways - Molecular Profiling
Novartis Institute for Biomedical Research
Basel
Switzerland

The majority of biological assays capture information on a multicellular level and therefore and average of the biological process in question. To be able to understand the contribution of different cell types, or different cellular states to a particular disease, more and more technologies focus on single cells. Flow cytometry has addressed such question for the last couple of decades with increasing resolution. However the major experimental drawback is the limitation on how many paraters can be looked at for one cell

Mass Cytometry, or CyTOF is a new platform combining the sensitivity and resolving power of Inductively Coupled Plasma Mass Spectrometry with the single cell analysis capabilities of flowcytometry. While fluorescent based flow cytometry and immunostaining are constrained by limited fluorophores with the complexities of spectral overlap, Mass cytometry replaces traditional fluorescent probes with rare earth metals isotopes, allowing over 40 different mass tagged antibodies to be resolved on a cell-by-cell level. With a read time of 1000 cells per second, this is a new exciting approach to high-content cytometric analysis.

This presentation will give an comprehensive overview of the technology, application in Biomedical Research and its data analysis challenges.

Native nano-ESI mass spectrometry studies of aptamer-ligand complexes

Basri Gülbakan¹, Konstantin Barylyuk¹, Petra Schneider², Max Pillong², Gisbert Scheider², Renato Zenobi¹

¹Department of Chemistry and Applied Biosciences,
²Institute of Pharmaceutical Sciences ETH Zurich, Switzerland

Introduction

Recently, a new class of molecular probes called “aptamers” has emerged. While they have been the subject of numerous studies, very few of them were on the characterization of aptamer-ligand interactions. Most of these studies were carried out by optical spectroscopy techniques which require either labelling or surface immobilization. Therefore, a direct and label-free method will be of great value. Herein we report the successful use of native nano ESI-MS to investigate different aspects of aptamer-ligand interactions such as structural selectivity, specificity and cooperativity. We validate our results by means of isothermal titration calorimetry (ITC) and circular dichroism (CD)

Methods

Adenosine (ABA) and L-argininamide-binding (LABA) aptamers were prepared in 50 mM ammonium acetate(AmAc) buffer (pH 7.4). NanoESI-MS analyses were performed with a hybrid Q-TOF mass spectrometer (Synapt G2-S, Waters, Manchester, UK) in negative ion mode. Instrument was tuned to detect ions without dissociating the non-covalent complexes. Nano ESI-MS experiments were conducted with ABA and 8 different structural isomers of adenosine: namely adenosine (1), inosine (II), 2`-deoxyadenosine (III), 3`-deoxyadenosine (IV), 5`-deoxyadenosine(V), 2-NH2-2`deoxyadenosine (VI), 3-NH2-3`-deoxyadenosine (VII), and 5-NH2 -5`-deoxyadeosine (VIII). The results revealed that binding order of different ligands with the ABA is VIII> VII >1> VI> IV=III=II. ABA was also titrated with varying concentrations of (VIII) ligand in 50 mM AmAc. Our nano-ESI results showed that K1= 110 uM and K2 = 16 uM. Native nano ESI-MS measurements of (LABA) revealed two different binding sites. One binding site (1:1 complex) bound the ligand tighter than the other one (1:2 complex). To investigate the ligand specificity, L-phenylalaninamide (P), L-tyrosinamide (T), and L-argininamide (A) ligands were mixed with LABA and measured by native ESI-MS. The determined binding strength followed the trend of A>T>P. LABA was titrated with varying concentrations of L-argininamide in 50 mM AmAc. We have found the following Kd values: K1= 23.6 uM and K2 = 213 uM. In In order to compare our results with a solution-phase reference, isothermal titration calorimetry experiments were performed in 50 mM AmAc (pH 7.4). We have found Kd=6.8 uM and Kd=27 uM for ABA / 5-NH2-5`-deoxyadenosine and LABA / L-argininamide respectively.

Figure 1. Native nano-ESI mass spectra of ABA / 5-NH2-5`-deoxyadenosine complexes (A) LABA / L-argininamide complexes (B).

Novel aspect

The first comprehensive native mass spectrometry study on aptamer-ligand interactions reveals the stoichiometry of complexes, ligand specificity, binding affinity, and cooperativity of binding.

HRMS dereplication, MS/MS spectral networks & small molecule epigenetic modifiers : tools to decipher cryptic metabolic pathways in Aspergillus sp.

Pierre-Marie Allard¹, Marija Perisic¹, Florence Mehl¹, Julien Boccard¹, Yung-Sing Wong², Katia Gindro³, Jean-Luc Wolfender¹

¹School of Pharmaceutical Sciences, EPGL, University of Geneva, University of Lausanne, Geneva, Switzerland
²Département de Pharmacochimie Moléculaire, Université Joseph Fourier, Grenoble 1, CNRS UMR 5063, CNRS ICMG FR 2607, Grenoble, France
³Swiss Federal Research Station Agroscope Changins-Wädenswil, Nyon, Switzerland

pierre-marie(dot)allard(at)unige(dot)ch

Introduction

The interest of microorganisms as a valuable source of bioactive compounds needs no more justifications. Since the discovery of Flemming’s penicillin G from Penicillium notatum at the beginnings of the XXth century to the isolation of the proteasome inhibitor salinosporamide A from Salinospora tropica in 2003, numerous valuable biologically active metabolites have been isolated from microorganisms. Recent insights from the progress in genetics have shed a new light on the biosynthesis of microbial natural products. It is now well known that under classical laboratory culture conditions, microorganisms only express a small proportion of their biosynthetic potential.[1] This phenomenon, known as gene cluster silencing, is very common and has been reported to occur in a vast range of living organisms.[2] Recently, new approaches aiming to address silenced biosynthetic pathways in eukaryotes have appeared. [3,4] One of our research interest is the de novo induction of secondary metabolites stimulated by the co-culture of diverse fungal strains [5], [6]. In the present study, inductions mechanisms are studied under a different angle and it is hoped to gain new perspectives and deeper understanding of communication between fungi at the metabolite level.

Methods

In order to explore the hidden biosynthetic potential of filamentous fungi we used small molecule epigenetic modificators (EM) of various classes (HDACi, DNAMTi) on phylogenetically diverse fungal strains. A metabolomic approach implying UHPLC-HRMS analysis, semi-automated dereplication procedures, MS/MS spectral networks generation [7] and multivariate data analysis was set up to detect the induction of novel metabolites and select promising fungal candidates for further scale-up culture.

Results

This workflow allowed us to highlight the production of various secondary metabolites not detected in control conditions. The application of HR-MS/MS networking provided valuable information regarding structures of the induced features. In particular, MS/MS networks allowed to reveal a family of closely related compounds as induced in an Aspergillus strain, thus indicating the probable unlocking of a common biosynthetic cluster.

1. Y-M Chiang et al, Nat Prod Commun, 4, 1505–10, 2009
2. H Gross, Curr Opin Drug Discov Devel, 12, 207–19, 2009
3. X Wang et al, J Nat Prod, 73, 942–8, 2010
4. RB Williams et al, Org Biomol Chem, 6, 1895–7, 2008
5. S Bertrand et al, J Nat Prod, 76, 1157–65, 2013
6. S Bertrand et al, J Chromatogr A, 1292, 219–28, 2013
7. J Watrous et al, Proc Natl Acad Sci USA, 109, E1743–52, 2012

Quantitative protein measurement of circulating plasma microparticles by data-independent nanoLC-MS²

Sophie Braga-Lagache¹, Natasha Buchs¹, Ioan Iacovache², Benoît Zuber², Christopher Jackson³, Manfred Heller¹

¹Department of Clinical Research, ²Institute of Anatomy, ³Institute of Clinical Chemistry, University of Bern, Bern, Switzerland

manfred(dot)heller(at)dkf(dot)unibe(dot)ch

Objective

Cells of the vascular system release a heterogeneous mixture of proteins, lipids, RNA, and cytoplasm components by way of spherical vesicles, e.g. microparticles (MPs) in the size range of 0.1-1µm. The shedding of MPs is induced by a variety of stress factors resulting in variable MP concentrations between health and disease. Furthermore, MPs have cell signaling and intercellular communication properties and interfere with inflammation and coagulation pathways. Therefore, MPs represent a repository of biological processes taking place in the vascular system. Today’s most used analytical technology for MP characterization, flow cytometry, is lacking sensitivity and specificity. Our goal is the development of a highly reproducible, fast, and quantitative MP assay based on targeted, multiplexed data-independent nanoLC-MS2 (MSX-DIA).

Methods

MPs were isolated from 250µL of cell-free plasma and cleaned from abundant plasma proteins by repetitive washing in PBS and centrifugation. MPs were morphologically controlled by cryo-TEM. MPs were in-solution digested with a combination of LysC and trypsin. In-solution digested MPs were subsequently analyzed with nanoLC-MS2 by data-dependent (DDA) or MSX-DIA acquisition on a QExactive instrument.

Results

MP pellets from 250µL plasma are invisible to the human eye. Staining of lipid membranes with sudan black enabled the visualization of MP pellets, as well as the training of the operator to avoid aspiration of those pellets during repetitive washing steps. Reproducible protein patterns on SDS-PAGE were achieved. MP preparations were characterized by cryo-TEM and showed the expected vesicle distribution.

MP protein profiles of 12 healthy volunteers were recorded twice with a digestion protocol using either a protein clean-up step by acetone precipitation, or without clean-up. A total of 946 gene products were identified in at least half of the samples by DDA analysis (combined search results of MaxQuant and Easyprot at 1% PSM FDR), with more proteins detected preparing samples without a clean-up step. All categories of cellular proteins were identified and quantified, including ECM, and 36% are plasma membrane proteins. This data set is very rich in cell markers, e.g. we detected at least 46 cluster of differentiation proteins enabling the generation of a cell profile of MP origin.

We have also implemented a MSX-DIA LC-MS2 method with Skyline software. The MSX-DIA method uses only ? of machine time compared with the DDA approach, provides an electronic sample archive for post-acquisition interrogation of protein quantities, and enables comparative protein profiling between samples spanning at least 3-4 orders of magnitude in protein concentration with CV’s <6%. For instance, we were able to detect and quantify CCL5 (RANTES) present in human plasma at a concentration of ~3ng/mL in the healthy population. In comparison, serum albumin was the most abundant protein with a more than 25’000 times higher intensity.

Conclusions

The protein profiles of 12 healthy volunteers were very similar without differences between gender or age. Removal of abundant plasma proteins is key for in-depth MP proteome analysis, enabling the measurement of targets like plasma membrane proteins, which indicate MP origin. Together with intracellular proteins, for instance from the cytoskeleton, these proteins can be used to quantify and characterize MPs. Levels of organelle-derived proteins might point to causes of tissue damage. MP analysis by MSX-DIA LC-MS2 instead of flow cytometry will improve the diagnostic use of MPs in cardiovascular and pneumological diseases.

Pushing the limits of FTMS performance for improved quantitative proteomics and metabolomics

Yury O. Tsybin, Konstantin O. Nagornov, Anton N. Kozhinov

Ecole Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

yury(dot)tsybin(at)epfl(dot)ch

Fourier transform mass spectrometry (FTMS) greatly benefits life sciences applications by providing the highest resolution and mass accuracy among all MS techniques. Nevertheless, novel challenges in the depth and accuracy of proteomics and metabolomics applications require achieving high resolution and mass accuracy in a shorter ion detection period. Particularly, the growing use of isotopic mass defects in quantitative proteomics and metabolomics renders FTMS platforms insufficiently slow or even incapable of reaching the required performance level. On the other hand, many molecular quantitation methods employ the a priori knowledge of the target and reference molecules mass values. Tandem mass spectrometry-based quantitation methods, e.g., TMT and iTRAQ, provide structural information on the reporter ions. However, the employed FT signal processing in standard FTMS does not use this information. We implemented super-resolution signal processing methods, including least-squares fitting (LSF) and filter diagonalization method (FDM), for FTMS to make use of the provided information on molecular targets and to speed-up the quantitation experiments both in proteomics and metabolomics. For example, based on LSF, we present a method tailored for increased scan rate in FTMS-based quantitative proteomics with 10-plex TMT reagents. Both LSF and FDM show their utility for NeuCode SILAC experiments performed on the advanced FTMS platforms, e.g., Orbitrap Elite and Fusion. Furthermore, the developed methods enable resolution-demanding experiments on entry-level FTMS platforms, e.g., Exactive series Orbitraps. Finally, combination of the advanced signal processing methods and novel mass analyzers in ion cyclotron resonance (ICR) FTMS allow to significantly speed-up this traditionally slow but extremely capable FTMS technology.

High resolution chemical mapping of solid samples using a miniature LIMS system

Andreas Riedo¹, Valentine Grimaudo², Pavel Moreno-García², Maike Brigitte Neuland¹, Marek Tulej¹, Peter Broekmann², Peter Wurz¹

¹Physics Institute, Space Research and Planetary Sciences Division, University of Bern, Switzerland
²Department of Chemistry and Biochemistry, Interfacial Electrochemistry Group, University of Bern, Switzerland

andreas(dot)riedo(at)space(dot)unibe(dot)ch

Sensitive mass spectrometric investigations with high lateral and high depth resolution are of considerable interest and importance in various fields of applications, ranging from semiconductor industry, to academic research on highly heterogeneous materials, such as meteorites or samples from sample return space missions. Chemical mapping of inhomogeneous materials with high lateral resolution allows studies of micro structures without a prior sample preparation in terms of powdering the sample material and analyzing a mean average afterwards. Moreover, accurate information of the chemical composition of the sample material with increasing depth is of special interest, for instance, for studies of the chemical composition of our Sun by analyzing the solar wind implanted in solid samples (Genesis mission), or to study the incorporation of impurities during the electrochemical copper deposition process in the microchip industry.

Since more than a decade our team is further developing a miniature (160mm x Ø 60mm) reflectron-type time-of-flight LIMS system, which was originally designed for in situ space exploration [1-3]. With a detection sensitivity of element/isotope concentration in sample material of about 10 ppb, high dynamic range of more than eight orders of magnitude, a sufficient high mass resolution m/?m of up to 1000, the LIMS system is suitable for many different applications in various fields where the quantitative information of the chemical (elements and isotopes) composition of solid materials is required. Current studies are conducted with an ultra-short pulsed laser ablation ion source (775 nm, 190 fs, <5 TW/cm2) where the laser pulses are focused onto the sample surface to a spot size of about Ø~10µm. A xyz-micro translational stage allows for a position accuracy and precision of about 2 µm and resulting in high spatial resolution studies of heterogeneous materials. The measurement procedure and the performance of the current instrumental setup will be presented in detail based on high spatial resolution 3D measurements conducted on various samples, including wafers, natural samples, and meteorites. Recent studies on Si-wafers showed that our LIMS system can be used for chemical analysis with a mean ablation rate of sample material in the sub-nm region. The high depth profiling resolution opens new perspectives for sensitive, accurate and precise monitoring and controlling of processes applied for production of new materials, e.g. by state-of the-art semiconductor industry and nanotechnology.

1. Riedo et al, J Anal At Spectrom, 28, 1256 – 1269, 2013
2. Neuland et al, Planet Space Sci, 2014, 101, 196 – 209, 2014
3. Riedo et al, Planet Space Sci, 87, 1 – 13, 2013

Enhanced screening of environmental pollutants in complex matrices by GCxGC-TOFMS with variable-energy electron ionisation

L. McGregor¹, A. Gravell², P. Kutty², S. Smith¹, N. Bukowski¹, I. Allan³, G. Mills⁴

¹Markes International, Gwaun Elai Medi-Science Campus, Llantrisant, RCT, Wales, UK
²National Resources Wales (NRW), Llanelli, Wales, UK
³ Norwegian Institute for Water Research (NIVA), Gaustadalleen 21, Oslo, Norway
⁴ University of Portsmouth, School of Pharmacy and Biomedical Sciences, White Swan Road, Portsmouth, UK

Two-dimensional gas chromatography with time-of-flight mass spectrometry (GCxGC-TOFMS) can provide highly sensitive detection and confident mass spectral identification of pollutants within complex environmental extracts. Nevertheless, the identification of individual compounds may be hindered by weak molecular ions or when similar mass spectral characteristics are evident across entire chemical classes. Select-eV ion source technology aims to combat this problem by allowing both hard and soft electron ionisation with no inherent loss in sensitivity. Select-eV provides enhanced molecular ions whilst retaining structurally-significant fragment ions, delivering both confident compound identification and increased selectivity. We show the potential of this technology for the analysis of both target pollutants and unknown chemicals in two different, complex extracts.

The routine monitoring of water quality is now a requirement of environmental legislation, such as the EU’s Water Framework Directive. Often the cause of a poor water quality status is unknown and extensive investigative monitoring is needed to determine what chemical maybe responsible. Passive sampling devices (e.g. semi-permeable membrane devices (SPMD), LDPE and silicone rubber) are often used for this purpose. The samplers were deployed for several weeks in a polluted river course in the UK to effectively sequester large volumes of water and provide a concentrated, representative extract for analysis by GCxGC-TOFMS with Select-eV.

In a second study, explanted silicone breast prostheses obtained from patients over a wide age range were collected. Silicone oils in the prosthesis extracts were removed using a multi-step extraction procedure and the resultant extracts were analysed by GCxGC-TOFMS with Select-eV. This novel approach aims to better estimate the overall body burden of bio-accumulative substances and how this changes over time of exposure.

This presentation shows the suitability of this novel analytical platform for environmental investigations, using both target-focused studies as well as non-targeted routines for screening for the presence of emerging contaminants.

Predicting, Detecting and Elucidating Benzotriazole Metabolites: Multiple Strategies Enhance Success

Emma L Schymanskia¹, Sebastian Huntschaa², Michael Gerlich³, Steffen Neumann³, Heinz P Singer¹, Juliane Hollender^1,4

¹Eawag: Swiss Federal Institute of Aquatic Science and Technology, Dübendorf, Switzerland
²Agroscope, Wädenswil, Switzerland
³IPB: Leibniz Institute for Plant Biochemistry, Germany
⁴ETH Zurich, Switzerland.

emma(dot)schymanski(at)eawag(dot)ch

Benzotriazoles (BTs) are widely used as corrosion inhibitors and anti-icing agents, such that they are now amongst the highest concentration micropollutants in wastewaters [1,2!. Investigations into the biological degradation of 1H-BT as well as 4- and 5-methyl-BT in the microbial communities in activated sludge from wastewater plants was performed to elucidate the main BT metabolites [3]. Sample analysis was performed using liquid chromatography coupled to high resolution, accurate mass tandem mass spectrometry (LC-HRAM/MS/MS). While the parent compounds were target analytes, the metabolites were identified using a combination of "suspect" (looking for predicted metabolites) and "non-target" approaches (time trend analysis) [4]. The rule-based EAWAG-BBD pathway prediction system was used for prediction, enhanced manually to include the monohydroxylation reactions. The corresponding masses were then added to "inclusion lists" for data-dependent MS/MS acquisition. Non-target masses were selected by investigating the time trends of masses from t=0 and t>0 samples. Full structure elucidation was attempted for all masses of interest.

Very few of the predicted metabolites were present in the large public compound databases ChemSpider or PubChem and even fewer were available as reference standards. Thus, structure elucidation was performed using the structure generator MOLGEN [5], which enabled the incorporation of detailed substructure information obtained via manual interpretation of the MS/MS spectra and experimental context. This also allowed rejection of some tentatively identified suspects. This dataset was then used to add multiple substructure functionality to the structure retrieval in MetFusion [6].

The relevance of some of the major metabolites was proven by the detection of hydroxylated and carboxylated-BTs in wastewater effluents (OH-BTs up to ug/L levels), while 1-methyl-BT was detected in the River Thur [3]. All MS/MS spectra available of confirmed, tentatively identified and unknown metabolites are available at  www.massbank.eu.

1. S Weiss et al, Environ Sci Technol, 40, 7193-7139, 2006
2. EL Schymanski et al, Environ Sci Technol, 48, 1811-1819, 2014
3. S Huntscha et al, Environ Sci Technol, 48, 4435–4443, 2014
4. DE Helbling et al, Environ Sci Technol, 44, 6621-6627, 2010
5. A Kerber et al, Mathematical Chemistry and Chemoinformatics: Structure Generation, Elucidation and Quantitative Structure-Property Relationships. De Gruyter, Berlin, Germany, 2013
6. M Gerlich, S Neumann, J Mass Spectrom, 48, 291-298, 2013

Confirmation of residue findings with LC-HRMS versus LC-MS/MS

Anton Kaufmann

Kantonales Labor Zürich, Switzerland

anton(dot)kaufmann(at)klzh(dot)ch

Confirmation is the important final step before a found and quantified compound (residue) can be reported. The process of confirmation is intended to ensure that the compound responsible for the observed detector signal is truly identical with the finally reported analyte. This important task has for a long time been the domain of tandem triple quadrupole mass spectrometry (QqQ). The currently most widely used confirmation strategy relies on the use of two selected reaction monitoring signals (SRM). The details of this confirmation procedure are described in the Commission Decision 93/256/EC (CD).

High resolution mass spectrometry (HRMS) is now increasingly used for multi-residue methods. Hence, there is an interest to use this technology for confirmation as well. It was the attempt of this work to develop a HRMS based confirmation criteria which results in an equal, or lower likelihood of false positive and false negative findings than the established CD.

The rate of false positives was compared by measuring so called dummy elemental compositions (CxHyNzOdCle). A random generator was used to calculate the elemental compositions of a precursor ion and two derived product ions. Some “chemical logic” was employed (e.g. ratio between carbon and hydrogen atoms in precursor and corresponding neutral losses) to mimic elemental compositions of small molecule drugs. A blank bovine liver extract was chromatographed and 50 dummy precursor & product ion sets were measured by extracting the calculated masses. The 100 transitions measured with QqQ and the 50 accurate precursor and 100 corresponding product ion masses measured with HRMS produced a large number of chromatographic peaks upon applying collision energy. The number of observed chromatographic peaks and the integrated peak areas were used to evaluate the likelihood of such false detects.

On the other hand, a comparison of false negatives was achieved by spiking a blank bovine liver extract with some 100 veterinary drugs at different concentration levels. The ability of detecting & confirming these truly present residues was compared.

A HRMS criterion was derived, which produced a lower number of false positive and false negative than conventional QqQ.

Finally, it was observed that matrix compounds frequently produce a signal in one of the QqQ transitions used to confirm a particular analyte. As a consequence, a truly present analyte would be measured with an ion ratio which lies beyond the acceptance window. QqQ instrument sensitivity has been dramatically improved, while the selectivity of the technology has remained stagnant. This development leads to the situation that truly present compounds can be detected, but due to potential matrix effects, confirmation may fail.

Targeted isolation of biomarkers highlighted by MS-based metabolomics in fungal co-cultures

Nadine Bohni^1,4, Olivier Schumpp², Florence Mehl¹, Sylvain Schnee², Samuel Bertrand¹, Michel Monod³, Katia Gindro², Jean-Luc Wolfender¹

¹School of Pharmaceutical Sciences, University of Lausanne, University of Geneva, Geneva, Switzerland
²Mycology group, Agroscope Changins ACW, Nyon, Switzerland
³Departement of Dermatology and Venereology, Laboratory of Mycology, CHUV, Lausanne, Switzerland
⁴Department of Chemistry, University of Basel, Basel, Switzerland

nadine(dot)bohni(at)unibas(dot)ch

The prevalence of Fusarium spp. as causative agent of onychomycoses is rising and Fusarium spp. as well as other non-dermatophyte fungi appear to be insensitive to systemic standard treatment [1]. Hence, new antifungal agents active against Fusarium spp. are needed. The concept of microorganism co-culture [2], was chosen to uncover possible anti-Fusarium compounds from the Basidiomycete Hohenbuehelia reniformiscultivated with human pathogenic Fusarium solani.

The co-culture of both fungi on the same Petri dish strongly induced the release of red pigments in the growth medium and a distinct distance repulsion could be observed. A metabolomics study using UHPLC-TOFMS analyses proved to be efficient to detect upregulation of several metabolites. On the other hand, upregulation of pigments could only be confirmed using a targeted UHPLC-UV analysis.

Still, isolation of biomarkers (significant loadings highlighted by statistical analysis (OPLS-DA) of the UHPLC-TOFMS datasets) revealed to be difficult. In many cases, very minor constituents with good ionizability in MS were highlighted by multivariate data analysis but were difficult or impossible to isolate. Thus, ELSD – evaporative light scattering detection, a universal and quantitative detector for LC – was chosen to devise an isolation strategy to purify the biomarkers that are potential antifungal compounds that may be responsible for the repulsion observed when both H. reniformis and Fusarium spp. were co-cultivated.

This work discusses how to link results obtained from MS-based metabolomics (identification of biomarkers as MS features) and the targeted isolation of such biomarkers for de novo structure elucidation by NMR and assessment of biological activity.

1. F Baudraz-Rosselet et al, Dermatology, 220, 164-168, 2010
2. S Bertrand et al, Biotechnology Advances, 32, 1180-1204, 2014

Acknowledgment: This work was supported by the Swiss National Science Foundation Sinergia Grant CRSII3_127187, which was awarded to J-LW, MM and KG

Automated Bligh and Dyer Extraction in Combination with Dual-ColumnUHPLC-SWATH/MS Separations Performed on an Integrated Metabolic Platform for the Analyses of Tissues and Cells

Sandra Jahn¹, Emmanuel Varesio¹, Guenter Boehm², Renzo Picenoni², Sandrine Cudré Correia De Almeida¹, Gerard Hopfgartner¹

¹University of Geneva, Geneva, Switzerland
²CTC Analytics AG, Zwingen, Switzerland

Sample preparation workflows for metabolomic studies of tissues or cells often require extraction procedures, like the Bligh and Dyer (B&D) approach, where the aqueous fraction (containing polar endogenous metabolites) is separated from the organic fraction (containing lipidic compounds). Proteins remain pelleted at the solvent interface. We present an integrated platform for performing B&D extraction in an automated manner on a robotic system coupled to a dual-column UHPLC-MS set-up for the metabolomic analysis of tissues or cells. For mass spectrometric detection a data independent (DIA) acquisition workflow based on SWATH/MS using variable Q1 windows was applied. Aqueous fractions are analyzed sequentially at two different mobile phase pH values, whereas lipidic fractions are analyzed alternately with an extended gradient. The integration of sample preparation and analysis enables a highly reproducible workflow and reduces the risk of time-dependent degradation processes to a minimum as samples are always freshly prepared prior to injection. This is especially important in metabolomics research since each systematic variation can lead to falsified results.

For the automated B&D extraction and column-switching a PAL RTC system (CTC Analytics) was used. Two quaternary low-pressure Nexera LC30AD UHPLC pumps (Shimadzu) served for dual-column separation. Detection was carried out on a TripleTOF 5600 (AB Sciex) operated in positive or negative ESI and MS/MS data was acquired in SWATH mode applying 24 variable Q1 windows.

After transfer of a previously lyzed and solubilized algae sample (Chlamydomonas reinhardtii) into a clear glass vial, the automated B&D extraction started by further solvent addition to the solution. Vortex-mixing and centrifugation followed, until the extraction finished by aspiration and transfer of the upper aqueous, and the lower organic phase into a dedicated new glass vial each. An aliquot of the aqueous phase was then diluted in acidic or basic mobile phase and injected onto a first analytical column (C18). In parallel, the lower organic phase was prepared by evaporation and reconstitution. During its analysis on a second reversed-phase column (C8), washing and re-conditioning steps of the first column with acidic or alkaline

mobile phase, respectively, were performed to analyze a second aliquot of the aqueous fraction at the corresponding conditions. High resolution MS2 spectra were recorded in SWATH mode, a novel DIA technique that uses fixed or variable Q1 windows for fragmentation. It allows for the generation of qualitative and quantitative information at once (QUAL/QUAN), without the need for multiple injections since adequate fragment ions can specifically be chosen post-acquisition for data treatment. Subsequent search against an in-house-built, SWATH-based spectral library, enabled fast identification of unknown low molecular weight compounds.

Elucidation of Herbivore Defense Mechanisms in Petunia hybrida Plants

Laurent Bigler¹, Sven Avak¹, Joëlle Sasse¹, Enrico Martinoia¹

¹Department of Chemistry, University of Zurich, Switzerland
² Institute of Plant Biology, University of Zurich, Switzerland

laurent(dot)bigler(at)chem(dot)uzh(dot)ch

Several Pleiotropic Drug Resistance (PDR)-type ABC transporters have been described to be involved in transport of secondary metabolites and be localized to trichomes. Furthermore, these membrane proteins play an essential role in plant defense mechanisms [1]. Objective of this project was to investigate the function of PDR transporters with regard to cellular export of deterrents toxic to herbivores. We performed our experiments with Petunia plants, because their leaves are highly toxic to generalist herbivores and solanaceae specialists [2]. Petunia is also a solanaceae model species, a family including important crop plants like e.g. potato, tomato, and eggplants.

Here, we report on the function of Petunia hybrida PDR2, a protein localized to glandular trichomes of leaf and stem tissue. PDR2 is involved in herbivore defense, as PDR2 silenced (pdr2) lines are less toxic to larvae of the generalist caterpillar Spodoptera littoralis. The secondary metabolites of interest were identified with an approach based on non-targeted metabolomics. Toxic wild-type plants were compared with less toxic pdr2 lines by measuring trichome extracts with UHPLC-ESI-HR-MS. Data was preprocessed (incl. non-linear retention time alignment and grouping), analyzed with statistics procedures, and the structures of the interesting metabolites were finally assigned.

This approach revealed the presence of several conspicuous sterol derivatives in the trichomes of Petunia hybrida, which have been reported to be potent insecticides [3]. It could be further demonstrated that the concentration of several steroids is reduced in the trichomes of pdr2 lines, implying that they cause the toxicity of wild-type Petunia plants.

1. J Crouzet et al , FEBS Lett 580,1123–1130, 2006
2. CA Elliger et al, J Chem Soc, Perkin Trans 1 143–149. doi: 10.1039/p19890000143, 1989
3. CA Elliger et al, J Chem Soc, Perkin Trans 1 711–717. doi: 10.1039/p19880000711, 1988

 Posters (Size: 146 high x 118 wide)

Identification of new micropollutant transformation products formed during wastewater treatment using LC-HR-MS/MS

Sven Avak^1,2, Jennifer Schollée^1,3, Emma Schymanski¹, Juliane Hollender^1,3

¹Eawag - Swiss Federal Institute of Aquatic Science and Technology, Dubendorf, Switzerland
²University of Zurich, Switzerland
³ETH Zurich, Switzerland.

sven(dot)avak(at)eawag(dot)ch

Micropollutant transformation products (TPs) pose a risk to the environment and many of these TPs can be formed during wastewater treatment processes such as biological treatment with activated sludge [1]. However, as the identities of many TPs are not yet known, assessing the risk to downstream aquatic communities or to humans is therefore not possible. Thus, there is a need to identify new TPs that are being formed during wastewater treatment.
State of the art techniques such as liquid chromatography coupled to high-resolution tandem mass spectrometry (LC-HR-MS/MS) allow for screening of polar organic compounds including TPs of pharmaceuticals, pesticides, and personal care products as shown by Schymanski et al [2].
The objective of this work is to identify TPs caused by biological treatment. Samples were collected from the influent and effluent of a wastewater treatment plant in Switzerland and were measured with LC-HR-MS using electrospray ionization (ESI). Resulting data were processed using an untargeted screening approach that included principal component analysis (PCA) for the selection of peaks characteristic of the different sample types. Peaks characteristic of influent samples were classified as potential parent compounds, while peaks characteristic of the effluent samples were classified as potential TPs. A linkage analysis was done which calculated expected TP masses from the list of potential parent compounds based on a set of expected transformation reactions and matched these to the potential TPs list. The links were further prioritized based on Mann Whitney significance testing and visual inspection of the extracted ion chromatographs. Data processing was carried out with R statistical software, including the R packages ‘enviPick’ [3] and ‘nontarget’ [4]. Identification through structure elucidation of selected linkages is currently performed using LC-HR-MS/MS data and openly accessible software and databases such as MassBank, MetFrag, and ChemSpider.

1. B Escher et al, Environ Sci Technol, 45, 3835–3847, 2011
2. E Schymanski et al, Environ Sci Technol, 48, 1811–1818, 2014
3. R package enviPick: Partitioning, EIC clustering and peak detection for baseline-corrected and centroided LC-HRMS data in mzXML format.  http://cran.r-project.org/web/packages/enviPick/index.html (accessed 22/09/2014).
4. R package nontarget: Detecting, Combining and Filtering Isotope, Adduct and Homologue Series Relations in High-Resolution Mass Spectrometry (HRMS) Data.  http://cran.r-project.org/web/packages/nontarget/index.html (accessed 22/09/2014).

MS-based metabolomic strategy for the search of de novo metabolite induction in mycobiome interactions through multi well fungal co-culture

Antonio Azzollini¹ ,Samuel Bertrand¹, Olivier Schumpp², Nadine Bohni¹, Michel Monod³, Katia Gindro², Jean-Luc Wolfender¹

¹School of Pharmaceutical Sciences, EPGL, University of Geneva, University of Lausanne, Geneva, Switzerland
²Swiss Federal Research Station Agroscope Changins-Wädenswil, Nyon, Switzerland
³Department of Dermatology and Venereology, Laboratory of Mycology, CHUV, Lausanne, Switzerland

In the field of natural product (NP) research, finding new sources of bioactive compounds is of primary importance. In this respect microorganisms have provided a large number of biologically active molecules. Recently, the use of fungal co-cultures for the induction of new NPs has emerged as a promising field in drug discovery [1,2].

A key point for the success of such studies is the development of co-culture experiments that provide high reproducibility of metabolite induction patterns and that are compatible with high throughput analytical procedures.

To tackle this issue, a method based on 12-well-plate miniaturized Petri dishes compatible with high throughput UHPLC-TOF-MS metabolomics [3] has been developed. This strategy was used to screen for metabolite induction in co-cultures of Aspergillus and Fusarium species.

This approach provided a satisfactory reproducibility and was used for the identification of induced biomarkers. This study demonstrates the consistent induction of new metabolites through co-culture. Moreover, the developed strategy is generic and can be applied to other types of microorganisms that can grow on solid media and that are part of the myco- or microbiome.

This 12 well-plate approach and the adapted data mining strategy were validated by the untargeted metabolomic study of a model co-culture (Eutypa lata versus Botryosphaeria obtusa responsible for confrontation zone lines in Vitis wood [1]). This procedure is currently used for screening novel metabolite induction in various fungal co-cultures.

1. G Glauser et al, J Agr Food Chem, 57, 1127, 2009
2. S Bertrand et al, Biotechnology Advances, 32, 1180–1204, 2014
3. S Bertrand et al, J Chromatogr A, 1292, 219-228, 2013

Full validation of a UPLC-MS/MS method for determination of an anti-allergic indolinone derivative and application to brain drug permeability studies

Evelyn A Jähne¹, Daniela E Eigenmann¹, Maxime Culot², Romeo Cecchelli², Fruzsina R Walter³, Maria A Deli³, Matthias Hamburger¹, Mouhssin Oufir¹

¹University of Basel, Switzerland
²Université Lille Nord de France, Lens, France
³Hungarian Academy of Sciences, Szeged, Hungary

d(dot)eigenmann(at)unibas(dot)ch

Natural products are an important and unique source of lead compounds in drug discovery. In a previous study, we identified (E,Z)-3-(4-hydroxy-3,5-dimethoxybenzylidene)indolin-2-one (indolinone) from woad (Isatis tinctoria L., Brassicaceae) as a compound possessing histamine release inhibitory and anti-inflammatory properties [1]. Evaluation of the pharmacokinetic properties of the compound [2], and in particular, of its ability to cross the blood-brain barrier (BBB), is a prerequisite. However, quantification methods used to determine brain permeability of lead compounds in BBB models have typically not been validated. Hence, the reliability of such data has to be questioned. To overcome these shortcomings, we developed and validated a quantitative LC-MS/MS method for indolinone according to current industry guidelines [3,4]. The method was used to assess brain drug permeability of indolinone in several in vitro cell-based human and animal BBB models. The calibration curve of indolinone in Ringer HEPES buffer in the range between 30.0 and 3000 ng/mL was quadratic, and a weighting factor of 1/X2 was applied. Diluting samples up to 100-fold did not affect precision and accuracy. The carry-over did not impact the results. Indolinone proved to be stable for 3 hours at room temperature, and for 3 successive freeze/thaw cycles. The processed samples could be stored in the autosampler at 10°C for at least 28 hours. Moreover, indolinone was stable for at least 16 days in Ringer HEPES buffer when stored below -65°C. In the in vitro BBB models, indolinone showed apparent permeability coefficient (Papp) values higher than 20 x 10-6 cm/s. The validation study demonstrates that our LC-MS/MS quantification method for indolinone is specific, selective, precise, accurate, and capable of producing reliable results. In the human [5] and animal in vitro BBB models, indolinone showed a high potential of brain permeation [6].

1. Kiefer S et al, Eur J Pharm Sci, 40, 143-147, 2010
2. Oufir M et al, J Chromatogr B, 902, 27-34, 2012
3. Guidance for Industry: Bioanalytical Method Validation, US Food and Drug Administration (FDA), Center for Drug Evaluation and Research, May 2001
4. Guideline on bioanalytical method validation. European Medicines Agency (EMEA/CHMP/EWP/192217/2009), London, 21 July 2011
5. Eigenmann DE et al, Fluids Barriers CNS, 10, 33-50, 2013
6. Jähne EA et al, J Pharmaceut Biomed, 98, 235-246, 2014

Steroid and lipid analysis by high resolution ion mobility-TOF MS

Michael Groessl, Benoit Plet, Stephan Graf, Richard Knochenmuss

Tofwerk AG, Thun, Switzerland.

Typically, GC-MS, LC-MS or direct infusion ultra-high resolution MS are used to investigate steroids and lipids but usually fail to resolve multiple isomers. Alternatively, ion mobility (IM) can be employed for separation of isomers without increasing cycle times and with minimal added experimental complexity. High resolution IM-MS was investigated for analysis of closely related isomeric steroids and their metabolites which are commonly included in clinicalurine steroid profiles. We demonstrate that the technique is able to separate isomerssuch as hydroxyprogesterones or cortisone metabolitesbased on the differences in collision-cross sections.
Additionally, we employed IM-MS for the direct analysis of lipids. 
Baseline resolution was achieved for phospholipids that only differed in the position of a double bond, e.g. 18:1 (delta6-cis) and (delta9-cis) PC as well as regioisomers, e.g. 18:0-16:1 PC and 16:1-18:0 PC. We also demonstrate that elevated pressures in the drift cell can be used to further improve IM resolution, leading to partial separation of cisand transisomers of PCs. In combination with post-processing of the raw data, ion mobility resolutions >200 are reached. This is achieved by multiplexed IM-MS, which also improves ion transmission over 200 times and signal-to-noise ratios 10 times compared to conventional pulsed mode.

Mixture of chemical pollutants at European legislation – effects of “safe” concentrations towards freshwater alga Chlamydomonas reinhardtii.

Daniel Jancula¹, Smitha Pillai¹, Rene Schonenberger¹, Kristin Schirmer^1,2,3, Marc J.-F. Suter^1,2

¹Eawag - Swiss Federal Institute of Aquatic Science and Technology, Dubendorf, Switzerland
²ETH Zurich, Switzerland
³EPF Lausanne, Switzerland

The potential risks associated with the presence of low concentrations of chemical compounds in aquatic ecosystems is increasingly debated. In this study we investigated the effects of a mixture of 14 substances (pesticides, polyaromatic hydrocarbons, heavy metals and plasticizer), at concentrations equivalent to the Annual Average Environmental Quality Standard (European legislation quality standards). Here we present the effects of the mixture at the transcriptome and proteome level, including the effects on growth and photosynthetic activity of freshwater green alga Chlamydomonas reinhardtii. Whereas growth inhibition of alga was not detected, significant changes in photosynthetic yields, the proteome, and the trascriptome occurred. We demonstrated that the proteome response as well as other physiological and biochemical endpoints may serve as a useful tool for ecotoxicological assessment of chemical mixtures in green algae.

Acute and chronic effects of tralopyril exposure on the proteome of the mussel Mytilus galloprovincialis

Isabel B Oliveira¹, Ksenia J Groh^2,3, Rene Schönenberger², Carlos Barroso¹, Kevin Thomas⁴, Marc J-F Suter^2,3

¹Biology Department & CESAM, University of Aveiro, Aveiro, Portugal
²Eawag - Swiss Federal Institute of Aquatic Science and Technology, Dubendorf, Switzerland
³ETHZ, Zurich, Switzerland
⁴Norwegian Institute for Water Research (NIVA),Oslo, Norway

isabeloliveira(at)ua(dot)pt

The growth of unwanted organisms on submerged surfaces leads to faster degradation of underwater equipment, increasing the maintenance costs. Antifouling (AF) paints are commonly used to prevent this. Although this approach is economically relevant, a special attention should be given to the risk assessment of booster biocides used to increase paint efficacy. Tralopyril has been recently approved under the Biocidal Product Regulation (BPR, Regulation EU No. 528/2012) to be used as an AF product. However, the knowledge on its toxicity towards aquatic organisms and its mode of action is still insufficient. We used global proteomics analysis aiming to i) characterize the acute and chronic effects of tralopyril on the proteome of the mussel Mytilus galloprovincialis and ii) understand if a recovery occurs after depuration. For this, mussels were exposed to tralopyril (1µg/L) or solvent control (5x10-5 % DMSO; 14.8µl into 30L of seawater) during 30 days and then transferred to clean water where they were kept for 10 days. Gills were collected from three individuals per condition (control, solvent control and tralopyril) at different time points (after 2 (T2) and 30 (T30) days of exposure, as well as after the depuration period (T40)). For protein extraction, gills from three individuals per condition and time point were pooled together. Proteins were digested with trypsin and the resulting peptide mixture was analysed using mass spectrometry-based Multidimensional Protein Identification Technology (MudPIT). Differentially expressed proteins were identified using a label-free approach based on spectral counting and G-testing. Our results show that acute exposure to tralopyril but also solvent control affected proteins involved in oxidative stress, immune response and active efflux.. Chronic exposure (T30) caused less proteome changes than the acute one (T2). After the depuration period (T40), protein alterations were still observed. This study contributes to the understanding of molecular mechanisms of tralopyril toxicity and may result in the identification of new biomarkers of exposure. It also questions the use of the DMSO as an innocuous solvent.

Novel data independent analysis methods for comprehensive and reproducible characterization of a biological pathway

Madalina Oppermann¹, Myriam Demant¹

Thermo Fisher Scientific,¹Stockholm, Sweden,²Reinach, Switzerland

Transitioning from a discovery experiment intended to characterize what proteins are in a sample, to obtaining the relative abundance of those proteins using highly sensitive targeted quantitation methods (e.g., PRM) can present a substantial obstacle to many research pipelines. Data Independent Analysis (DIA) bridges protein identification and targeted quantitation by acting as a screening technique to efficiently survey a sample and direct future targeted analysis. DIA creates a comprehensive record of a sample by sequentially scanning all ions across a wide mass-to-charge range ensuring that all detectable parent ions and fragments are examined. This technique is entirely untargeted: there is no scheduling of transitions and thus no unselected peptides, no mistimed transitions.

Nevertheless, not all DIA methods are created equal, and particular adaptations of the DIA method to the specific sample will further enhance analytical outcomes: speed, sensitivity, and selectivity form an ideal limit on how much information can be extracted from a DIA experiment.

This presentation will cover building high performance methods on the Orbitrap Fusion Tribrid and the Q Exactive HF mass spectrometers to take advantage of analyzer parallelization with WiSIM and precursor multiplexing with msxDIA. The content will include guidance on how to generate a high quality spectral library and use it to reveal important expression changes in the pathway analysis of interest, and even how to go back in time and use a DIA file’s record observable proteome to retrospectively interrogate data for new target peptides and proteins. This presentation will demonstrate how DIA can be used to select appropriate targets for further more sensitive SRM/PRM analyses and how reducing sample complexity with immunoprecipitation can significantly enhance the sensitivity of DIA supporting its use as a satisfactory alternative to targeted methods.

Steroid hormone metabolite analysis by two-dimensional gas chromatography time-of-flight mass spectrometry (GCxGC-TOF)

Georgios Papadopoulos¹, Michael Groessl¹, Bernhard Dick², Bruno Vogt², Richard Knochenmuss¹

¹Tofwerk AG, Thun, Switzerland
²Department of Nephrology and Hypertension, Bern University Hospital (Inselspital), Bern, Switzerland

papadopoulos(at)tofwerk(dot)com

Quantification of endogenous steroid hormones and their metabolites in urine samples (androgens, estrogens, metabolites of cortisol, corticosterone, progesterone and aldosterone) is routinely used in clinical assays for the detection of several diseases such as congenital adrenal hyperplasia, hyper- and hypo- aldosteronism and Cushing's disease. Moreover, the screening of these analytes in humans and rats with different diseases provides insight into the basis of physiological regulation and allows for the study of different aspects of hypertention, salt retention, aging and inborn errors of metabolism.
In clinical laboratories, the standard method for the qualification and quantification of these analytes is GC-Quadrupole MS. However, GC-QMS suffers from major limitations. Many of the metabolites of interest are closely related isomers having overlapping elution times and similar or identical fragmentation patterns. Moreover, mass identification/separation of analytes is limited by the mass resolving power of quadrupole instruments. These limitations may be overcome by adding more dimensions in which analytes can disperse, and by using faster mass spectrometers with higher mass resolving power.
Here we present the use of two-dimensional gas chromatography time-of-flight mass spectrometry (GCxGC-TOF) for the analysis of urinary steroids and progesterone metabolites. Eluent from a first GC column is frozen periodically and then rapidly re-mobilized and injected into a second GC column for further separation. Detection is by TOF mass spectrometry. This allows for two dimensional separation of analytes, and many analytes that would co-elute in 1D GC become baseline separated. In this way, identification of an analyte is more precise, since the mass spectra that are recorded belong to single analytes and not to co-eluting species. In addition, one can use the most intense fragments (or even all the fragments) for quantification. This leads to a decrease in the limits of detection. Moreover, the high TOF mass resolving power adds effectively one more dimension in the separation, as accurate mass and isotopic distribution can help identify an eluted analyte.

Metabolite monitoring in fed batch cell cultures using MALDI TOF MS

Robert Steinhoff¹, Thomas Villiger², Jasmin Krismer¹, Miroslav Soos², Martin Pabst¹, Renato Zenobi¹

¹Laboratory of Organic Chemistry, ETHZ, Zurich, Switzerland
²Institute for Chemical and Bioengineering, ETHZ, Switzerland

steinhoff(at)org(dot)chem(dot)ethz(dot)ch,  zenobi(at)org(dot)chem(dot)ethz(dot)ch

The analysis of intracellular metabolites is becoming an important task to routinely monitor biotechnological fed batch process reproducibility and performance. Traditionally, these metabolite levels are followed using liquid chromatography (LC) combined with UV detection. However the long LC-UV runtimes compromise the possibilities to regulate a running process via feedback. We are presenting a novel MALDI-MS method that enormously reduces the time to analyze intracellular metabolites, and provides excellent robustness. A commercial MALDI TOF instrument (5800, ABSciex, Germany) and a microarray sample target for mass spectrometry were used in the study. The sample target consists of a coated and micro structured ITO glass slide. Hydrophilic spots on the otherwise omniphobic chip allow for fast, automated aliquoting and focusing of any organic solvent. Furthermore, a short extraction protocol was developed and used to monitor fed batch reactors. Adenosine-­-5´triphosphate was detected and quantified using an isotopically labeled internal standard (13C15N-Adenosine-5´-triphosphate). The appearance of ATP is in good agreement with literature data. The di- and monophosphates of adenosine as well as guanisine, cytidine, uridine were also monitored. Any methodologically induced analyte fragmentation or hydrolysis was corrected for. Moreover, the recorded metabolite profiles were analyzed using statistical tools, e.g. principal component analysis. The presented method has high-throughput capabilities and can potentially be applied in an industrial environment. The detected metabolite profiles were cross-validated using HPLC-UV. Development and implementation of a microarray for metabolite analysis of fed batch cultures using MALDI-MS. Fast and effective analyte extraction protocol combined with an analysis method including isotopically labeled internal standard.

RMassBank: a workflow for automated processing, recalibration and annotation of high-resolution mass spectra for the MassBank spectral library

Michael A. Stravs¹, Emma Schymanski¹, Erik Müller², Steffen Neumann³, Tobias Schulze², Heinz Singer¹, Juliane Hollender^1,4

¹Eawag - Swiss Federal Institute of Aquatic Science and Technology, Dubendorf, Switzerland
²UFZ - Helmholtz-Zentrum für Umweltforschung, Leipzig, Germany
³ IPB - Leibniz Institute for Plant Biochemistry, Halle, Germany
⁴ ETHZ, Zurich, Switzerland

stravsmi(at)eawag(dot)ch

Today, accurate mass high-resolution mass spectrometry (HRMS) enables the identification of unknown chemical compounds in environmental analytical chemistry and other fields (e.g. metabolomics, forensics). Successful identification is often dependent on the availability of comprehensive spectra databases. However, databases of LC-MS/MS and particularly LC-HRMS/MS spectra do not yet reach the broad coverage of GC-MS databases, and are often dominated by metabolites.
The MassBank project ( http://www.massbank.jp) [1] established a freely accessible, open database for MS spectra contributed by users. However, the acquisition, extraction, processing and annotation of spectra is time-consuming and laborious if conducted by hand. Furthermore, a high quality of the spectra is important for good performance of the database. Non-commercial contributors often lack time resources for extensive manual processing and curation of spectra. 
In the NORMAN network ( http://www.norman-network.net), the NORMAN MassBank ( http://www.massbank.eu) acts as a second MassBank server with the aim to improve the identification of organic micropollutants in the environment by establishing a comprehensive HRMS/MS database of environmentally relevant chemicals. Members are encouraged to deposit their spectra to enhance exchange between institutes. RMassBank was developed as a workflow which automates the creation of high-quality library spectra for MassBank from HRMS raw data files. In the data processing workflow, spectra are automatically extracted from the raw files. Accurate mass data is exploited for automated fragment formula assignment. This information is used in a first step for recalibration and in a second step for noise filtering, to yield cleaned, recalibrated spectra. From a minimum of one structure identifier, annotation data is retrieved from Internet resources (CACTUS Chemical Identifier Resolver,  http://cactus.nci.nih.gov/chemical/structure; Chemical Translation Service,  http://cts.fiehnlab.ucdavis.edu/). Finally, the cleaned spectra are combined with the annotation information and exported ready for upload into a MassBank library. 
The workflow was developed and demonstrated with a set of 70 pesticide spectra recorded on an LTQ Orbitrap XL (Thermo Scientific) mass spectrometer [2]. Currently, 8864 RMassBank-processed spectra of 580 environmental pollutants have been published on MassBank and more are in preparation. The original workflow has recently been expanded with a graphical user interface (GUI) for enhanced ease of use and more flexible handling of different instrument and data types.

1. H Horai et al, J Mass Spectrom, 45, 703, 2010
2. M Stravs et al ,J Mass Spectrom, 48, 89, 2013

Human biofluid based metabolomics enabled by recent QTOF technology

Dirk Wunderlich

Bruker Daltonik GmbH, Fahrenheitstrasse 4, 28359 Bremen, Germany

dw(at)bdal(dot)com

Modern LC-QTOF-MS/MS instruments enable “one shot acquisitions”, providing qualitative and quantitative results for large sequences of complex metabolomics samples. Recent improvements in QTOF instrumentation like increased scan speed and new digitizer technology for higher dynamic range and resolution allow the in depth analysis of very complex samples. These hardware features are of relevance for in-depth characterization of metabolomics samples containing analytes in different concentrations over >4-5 orders of magnitude. The increased dynamic range of QTOF mass spectrometer helps to identify metabolites based on mass accuracy and isotopic pattern reliability for low and high intense signals.

Examples will be presented highlighting combined non-targeted and pathway driven targeted metabolomics making use of the same high resolution LC-MS data set. Identification workflows using accurate mass and isotopic pattern information from MS and MS/MS spectra are illustrated in a Diabetic/non Diabetic study.Combined with a novel human metabolome MS/MS library this leads the way to seamlessly detect and identify potential biomarkers.

Influence of the target plate material and sample layer thickness on LDI ionization efficiency for C60

Guido P. Zeegers, Renato Zenobi

ETHZ, Zurich, Switzerland

guido(dot)zeegers(at)org(dot)chem(dot)ethz(dot)ch

In order to establish the validity of models thus far proposed for matrix-assisted laser desorption/ionization (MALDI), such as: disproportionation ionization [1], the “lucky survivor” model [2], cluster formation [3] and exciton pooling [4], a basic LDI approach, based on electrospray deposition of C60 (stable, clear fragmentation pattern and ionizable in both negative and positive mode) with different layer thicknesses on a range of different target plate materials, was chosen as a starting point. The ion yield was monitored by a commercial MALDI-TOF MS instrument (Bruker Ultraflex III) for both a broad laser fluence and ion extraction delay time range. The presence of one or more ion yield maxima at different time intervals could be an indication for the number of ionization mechanisms contributing to LDI and the time frames in which they operate. Varying the target plate material and sample layer thickness could shed light on the role of the sample itself and the influence of target plate material interactions during ion formation in both negative and positive ionization mode.

An automated high-throughput electrospray sample deposition setup was developed suited for MALDI target plate insets made of a variety of different materials, among them a range of metals, alloys and insulators. Standard MALDI target plates were milled out and subsequently fitted with these insets.

The deposited C60 was analyzed over a laser fluence (below ionization threshold-maximum) and extraction delay time (0-750 ns) range in positive and negative reflectron mode. The resulting spectra were processed in an automated fashion (MATLAB): integrating the spectra over a 360-725 m/z range and averaging the resulting AUCs for repeating experiments, finally yielding a signal intensity profile accompanied by a standard deviation plot.

The results obtained (Ultraflex III) show that the ion yield in positive mode is a fourfold higher than its negative counterpart for all sample insets tested and ion yield profiles are markedly different for different metals and alloys. All trends observed remain to be verified on an alternative commercial MALDI-TOF MS instrument (AB Sciex 5800), but are a strong indication that substrate material influence on ion yield should not be underestimated and that careful substrate material selection could prove beneficial for ion yield enhancement.

1. BH Liu et al, J of Phys Chem B, 114, 10853-10859, 2010
2. M Karas et al, J of Mass Spectrom 35, 1-12, 2000
3. M Karas, R Kruger, Chemical Reviews 103, 427-440, 2003
4. R Knochenmuss, VL Zhigilei, J of Phys Chem B 109, 22947-22957, 2005