The high accuracy measurements of m/z values (average mass error lower than 1 ppm) allowed for MS1-based putative identifications of the metabolites of interest with a mass tolerance of 3 ppm and a false discovery rate <10%. For this, centroided MSI datasets were uploaded to MetaSpace 25. The metabolite annotation was performed using the FDR-controlled method26 delivering annotations 'at Level 2’ (putatively annotated compounds) according to the Metabolomics Standard Initiative 27. Adjacent tissue sections, originating from the same organs (identical processing, storage time and sectioning conditions) as those used for the DESI MSI, were prepared for direct injection MSMS and ion pairing LCMS.
"],"repository":["MetaboLights"],"study_status":["Public"],"ptm_modification":[""],"instrument_platform":["MS Imaging - negative"],"publication":["Revealing Hidden Variables in DESI-based Spatial Metabolomics: Solvent Composition and Tissue Type as Critical Drivers."],"submitter_affiliation":["VIB -KU Leuven"],"submitter_name":["Marco Giampà"],"organism_part":["kidney","mouse","liver","brain","heart"],"technology_type":["mass spectrometry assay"],"disease":[""],"organism":["Sprague Dawley","C57 black mouse"],"full_dataset_link":["https://www.ebi.ac.uk/metabolights/MTBLS14771"],"author":["Marco Giampà. VIB -KU Leuven. marco.giampa@kuleuven.be.","Bart Ghesquière. KU Leuven, Department of Cellular and Molecular Medicine, Laboratory of Applied Mass Spectrometry (LAMaS). bart.ghesquiere@kuleuven.be."],"data_transformation_protocol":["The raw data were converted to the h5 format for the import in Mozaic (Version 2025.11.0.b2; Spectroswiss), where mass accuracy analysis, recalibration, and peak picking were performed. Centroided data were exported in imzML format. A subset of the centroided dataset was generated using the function peakAlign(), in Cardinal’s R package28,29 considering a target peak list with a tolerance of 5 ppm. The target peak list for subsequent analyses included metabolites from glycolysis, the TCA cycle, as well as abundant amino acids and fatty acids. The full target peak list is shown in Table S3 in Supporting information. The subset of the centroided dataset was normalized by TIC. Selection of the acquisition region is restricted by the DESI source control software to rectangular forms. In order to perform the data analysis purely on the tissue region, a spatial clustering analysis was performed with the glass region removed.
"],"study_factor":["Organism"],"submitter_email":["marco.giampa@kuleuven.be"],"sample_collection_protocol":["Mouse tissues (heart, liver, brain and kidney) were obtained from wild type C57 black 6 mice, wrapped in aluminum foil and snap-frozen in liquid nitrogen and stored immediately at -80°C as approved by the KU Leuven ethical committee for animal experimentation (M018/2024). For the replication batch, brain tissues were collected from a knockout mouse.
Male Sprague–Dawley rats (250 g; 6-8-weeks-old) were housed at the KU Leuven animal facility. After at least 4 days of acclimatization, they were anesthetized using isoflurane (5% for induction, 1.5% for maintenance). Analgesics were administered before the start of the surgery, including Vetergesic, Ecuphar 0.1 mg/kg, and Naropin (AstraZeneca) 5mg/kg. Then both renal pedicles were dissected free through a midline abdominal incision. The right kidney pedicle was clamped with a microaneurysm clamp for 60 min to induce ischemia. Reperfusion was initiated by removal of the clamp. After 3 hours of reperfusion, the rat was anesthetized using isoflurane (5% for induction, 1.5% for maintenance). Kidneys were collected and immediately snap-frozen in liquid nitrogen and then stored at -80°C. The rats were sacrificed under anesthesia by a lethal dose of sodium pentobarbital (Dolethal, Vetoquinol). All protocols had been approved by the KU Leuven ethical committee for animal experimentation (M018/2024).
"],"omics_type":["Metabolomics"],"histology_protocol":["After DESI MSI, rat kidney tissue sections were stained in hematoxylin/eosin (H&E) using Leica Autostainer XL (ST5010). Briefly, the slides goes through the following washes: milliQWater (4 min), Hematoxylin (MHS32-1L, Sigma) (5 min), running tap water (5 min), Eosine Y, aqueous (HT110232-1L, Sigma) (2 min), running tap water (40 sec), 95% Ethanol (2 min), 100% Ethanol (4 min), 100% Ethanol (5 min), Xylene (2x for 10 min). The stained slides exit in xylene and were mounted with DPX Mountant for Histology (06522-500mL, Sigma). The stained tissue slides were scanned with high resolution of 40x (0.25µm/pixel) with Vectra PolarisTM. Rat kidney histology annotations were performed using QuPath as digital pathology software.
"],"preparation_protocol":["All organs were sectioned into 20 μm thick sections (Cryostar NX70 cryostat, ThermoFisher Scientific, Massachusetts, USA) employing a fresh disposable blade and tissue temperature parameters for each organ (see Table S 1). Enough material was sectioned to allow for 4 technical replicates for each organ and for each solvent system (one technical replicate as representative batch and three technical replicates reproducibility batch) For a direct comparison between the two solvents, two subsequent sections were mounted on two different glass slides.
For the rat hypoxia pilot study, three animals were sacrificed and sections of hypoxic and normoxic kidney were mounted on the same slide for each animal. All slides were vacuum dried immediately after mounting and further stored at -80 °C until analysis, in order to maximally reduce metabolite delocalization. The sections from -80 °C were vacuum dried again immediately before the MSI analysis.
"],"study_design":["Metabolomics","liver","Mass spectrometry imaging","DESI","brain","C57 black mouse","heart","semi-targeted analysis","experimental sample","kidney","mouse","Sprague Dawley","MRT SELECT Series"],"curator_keywords":["Metabolomics","liver","Mass spectrometry imaging","DESI","brain","C57 black mouse","heart","semi-targeted analysis","experimental sample","kidney","mouse","Sprague Dawley","MRT SELECT Series"],"mass_spectrometry_protocol":["A Waters ACQUITY M-Class uBSM pump was used to continuously deliver the solvent (90% MeOH or ACN in H2O; 9:1 v/v) at 2 uL/min to the Waters DESI XS source coupled with a hybrid quadrupole-multireflecting time-of-flight Select Series MRT mass spectrometer (Waters, Massachusetts, USA). The solvent line was implemented using the low-flow kit (Waters) to increase backpressure (between 1400-1500 psi) and improve spray stability. All DESI MSI analyses were done in negative ionization mode. Initial mass calibration was performed with polyalanine (5 mg/mL in 95:5 v/v MeOH/H2O). Oleic acid (ubiquitous in tissues; m/z= 281.2489) was used as internal lock mass.
Capillary voltage was set at 0.45-0.50 kV, cone voltage at 35 V, ion source temperature 95 C, API gas at 0.7 bar and the heated transfer line temperature at 100 C. The MRT mass spectrometer was tuned in order to best transmit metabolites with m/z values <500.
Ion transfer was facilitated using a StepWave ion guide with RF voltage set to 100 V. The T-Wave ion guide operated at a velocity of 300 m/s and a pulse height of 0.5 V. The entrance, bias, and DC offset voltages were 2.0 V, 2.0 V, and -4.0 V, respectively. The exit voltage was 0.0 V (readback: -76.5 V).
The MS profile was set to Manual at m/z 50, 150 and 200, dwell time (% scan time) 80 and 5, ramp time 5 and 10 (% scan time). The collision energy was set to 5 V, and the collision gas (nitrogen) flow rate was 0.9 mL/min. The transfer lens acceleration voltage was 20 V, and the detector voltage was maintained at 1.96 kV. The polarity was set to negative mode. The scan time was set to 1.5 s and 1 s per pixel for a pixel resolution of 100 um and 50 um, respectively. Based on lower concentrations of metabolites in tissue compared with other classes of analytes, such as lipids, we have chosen extended scan times of 1-1.5 s allowing for higher intensities due to longer microextraction effects for the investigated analytes in sample sections. This corresponds to a DESI stage motion speed of 66.6 um/s and 50 um/s respectively. Spray stability was assessed through multiple tests, including solvent acquisition, ink measurement, and a 30-minute acquisition on test tissue, followed by evaluation of the TIC heatmap. The ink intensity exceeded 5 x 10^6, passing the sensitivity control.
"],"additional_accession":[]},"is_claimable":false,"name":"Revealing Hidden Variables in DESI-based Spatial Metabolomics: Solvent Composition and Tissue Type as Critical Drivers","description":"In the development of a desorption electrospray ionization (DESI) workflow for spatial metabolomics, we investigated the impact of two commonly used solvent systems, 90% acetonitrile (ACN) and 90% methanol (MeOH), on the spatial metabolomic profiling of multiple murine tissues. The performance of both solvents was evaluated across several metabolite classes (central carbon metabolites, amino acids, and fatty acids). While the ACN-based solvent system led to higher signal intensities of small polar metabolites involved in glycolysis, the tricarboxylic acid (TCA) cycle, and amino acid metabolism, the MeOH-based solvent system provided superior signal intensities of fatty acids. These findings demonstrate that solvent composition differentially influences metabolite extraction and ionization processes in DESI and should be carefully matched to the biological question and metabolite classes of interest. To illustrate the utility of the optimized workflow, the ACN solvent system was applied to a rat model of renal ischemic injury. Spatial mapping of metabolites across distinct kidney regions (cortex, outer medulla, and inner medulla) revealed pronounced region-specific metabolic changes between normoxic and ischemic conditions. Together, these results demonstrate the importance of solvent selection in DESI-based spatial metabolomics and showcase the ability of this approach to uncover spatially resolved metabolic adaptations associated with tissue injury.","dates":{"publication":"2026-09-11","submission":"2026-06-16"},"accession":"MTBLS14771","cross_references":{}}