Metabolite identification was performed using both in-house and public databases (HMDB, LIPID MAPS, MoNA). Every reference matched metabolite was manually inspected and ranked at identification level 1 to level 4 (Sumner et al., 2007).
Statistical analysis included Welch’s t-test and Cohen’s d to assess significant differences (p ≤ 0.05, |d| ≥ 0.8). Visualizations included heatmaps, PCA plots, Venn diagrams, volcano plots, and QC boxplots, created using Excel, RStudio, MetaboAnalyst 6.0 (metaboanalyst.ca), and VolcaNoseR (Goedhart & Luijsterburg, 2020).
"],"sex-based_metabolic_differentiation_protocol":["After demonstrating that the proposed extraction conditions were suitable for the five sample types, it was applied to one replicate from each individual and the pooled sample for all five sample types. One individual’s whole blood sample (E05B) was missing. The pooled plasma and blood were mixed from equal parts of individual samples.
"],"repository":["MetaboLights"],"study_status":["Public"],"ptm_modification":[""],"instrument_platform":["Liquid Chromatography MS - negative - reverse phase","Liquid Chromatography MS - positive - hilic","Liquid Chromatography MS - - reverse phase","Liquid Chromatography MS - negative - hilic"],"chromatography_protocol":["The LC-MS and data analysis workflow was modified from Klåvus et al. Briefly, we used both reverse-phase (RP) and hydrophilic interaction (HILIC) chromatography, coupled to a Q-TOF mass analyzer at both positive and negative ionization.
"],"publication":["An LC-MS untargeted metabolomic comparison between three blood microsampling devices, whole blood, and plasma. 10.1007/s11306-026-02424-6. PMID:42213273"],"submitter_name":["Andres R. Martinez Bilesio","Philippine Louail","Dennisse AVELLA"],"submitter_affiliation":["Eurac Research","Chair of Bioinformatics Jena, Faculty of Mathematics and Computer Science, Friedrich Schiller University Jena","Afekta Technologies Ltd."],"organism_part":["Solvent","Whole Blood"],"technology_type":["mass spectrometry assay"],"disease":[""],"extraction_protocol":["Extraction methods were first tested on Whatman samples using punches of three sizes (3 mm ≈ 3 µL; 6 mm ≈ 11 µL; and two 6 mm ≈ 22 µL). Extraction solvents included 40 µL of water followed by 160 µL of MeOH/ACN (50:50). The approximate concentration of the samples was 1,5% (µl blood/µl extraction mixture), 5,5%; and 11,0%. Whatman blanks were also prepared at the three concentrations. Three methods were tested: first, 2 min ultrasound with 15 min ice incubation; second, 15 min shaking at 500 rpm; and third, 2 min homogenization with 15 min ice incubation.
After the extraction, the samples were centrifuged, and supernatants were filtered before LC-MS analysis. Quality control (QC) samples were prepared by pooling aliquots from all samples. Shaking at room temperature, and a 5.5% sample concentration was selected for further use based on comparable performance and convenience.
"],"organism":["blank sample","Homo sapiens"],"full_dataset_link":["https://www.ebi.ac.uk/metabolights/MTBLS10740"],"author":["Glykeria Avgerinou. Aristotle University of Thessaloniki, School of Physical Education and Sport Science, Thessaloniki, Greece. Thessaloniki, Greece. gavgeri@phed.auth.gr.","Dennisse Avella. Afekta Technologies Ltd., Kuopio, Finland; University of Eastern Finland, School of Pharmacy, Kuopio, Finland. Microkatu 1, 70210, Kuopio, Finland. dennisse.avella@uef.fi. 3.5845E+11.","Vassilis Mougios. Aristotle University of Thessaloniki, School of Physical Education and Sport Science, Thessaloniki, Greece. Thessaloniki, Greece. mougios@auth.gr.","Soile Turunen. Afekta Technologies Ltd., Kuopio, Finland. Microkatu 1, 70210, Kuopio, Finland. soile.turunen@afekta.com.","Iman Zarei. University of Eastern Finland, School of Medicine, Kuopio, Finland. Kuopio, Finland. iman.zarei@uef.fi.","Olli Kärkkäinen. Afekta Technologies Ltd., Kuopio, Finland; University of Eastern Finland, School of Pharmacy, Kuopio, Finland. Microkatu 1, 70210, Kuopio, Finland. olli.karkkainen@uef.fi.","Kati Hanhineva. Afekta Technologies Ltd., Kuopio, Finland; University of Eastern Finland, School of Medicine, Kuopio, Finland; University of Turku, Department of Life Technologies, Turku, Finland. Kuopio, Finland. kati.hanhineva@uef.fi.","Anatoli Petridou. Aristotle University of Thessaloniki, School of Physical Education and Sport Science, Thessaloniki, Greece. Thessaloniki, Greece. apet@phed.auth.gr.","Seppo Auriola. University of Eastern Finland, School of Pharmacy, Kuopio, Finland. Kuopio, Finland. seppo.auriola@uef.fi."],"data_transformation_protocol":["Mass Hunter Qualitative (Analysis B.07.00 - Agilent Technologies) software was used to verify retention time, abundance, and system pressure consistency using a system suitability standard solution (9 compounds). Raw data were exported to MS-DIAL (version 4.9.221218), where peak detection and alignment were performed. Detection parameters included a minimum peak height of 5000 and a width of 8 scans. Mass tolerances were 0.01 (MS1) and 0.05 (MS/MS); RT tolerance was 0.1 min. Features exceeding the threshold of 5 compared to the blank solutions were neglected.
The aligned data were pre-processed with Notame in RStudio (2023.12.1 version 3 of the GNU). Features were log-transformed, drift-corrected using cubic splines, and filtered based on reproducibility and detection thresholds. Features with >70% presence in QCs, >50% in at least one group, RSD <20%, and D-ratio <40% were retained (High-quality features) and imputed using random forest for missing values (Klåvus et al., 2020).
"],"study_factor":["Extraction method","Sex","Age","BmicroS device"],"submitter_email":["a.martinez@uni-jena.de","plouail@eurac.edu","dennisse.avella@afekta.com"],"sample_collection_protocol":["Samples were collected from 10 healthy adult volunteers (5 males, 5 females) with informed consent and ethical approval from the Aristotle University of Thessaloniki. A pooled 11th sample was prepared from equal aliquots of the 10 individual samples. All participants fasted overnight before venous blood collection.
The venous blood samples were divided into five types: whole blood (B), plasma (P), and microsamples Mitra (M), Capitainer (C), and Whatman (W). Samples were prepared as follows: plasma was obtained by centrifugation, Mitra tips were dipped in the whole blood, and Capitainer and Whatman cards were prepared using pipetted volumes. Samples were frozen at −80 °C (whole blood and plasma) or −20 °C (microsamples), shipped on dry ice, and stored at −80 °C before analysis.
"],"extraction_across_devices_protocol":["The optimized extraction protocol was applied to whole blood, plasma, Mitra, Capitainer, and Whatman samples (three replicates each, prepared from the pooled sample). 20 µl samples were collected from each sample type and placed in 2 mL Eppendorf tubes. For all sample types, 80 µL of water and 320 µL of MeOH/ACN (40:40) were added, followed by vortexing for 10 sec, shaking for 15 min at 520 rpm, centrifugation (10 min at 17,000×g, 4 °C), and filtration (0.2 µm PTFE). The sample preparation and analysis order were randomized using the Wranglr tool (antonvsdata.shinyapps.io/wranglr). Pooled QC samples and blanks were included.
"],"omics_type":["Metabolomics"],"study_design":["Human","Metabolomics","Plasma","Volumetric Absorptive MicroSampling (VAMS)","Microsampling","blood metabolome","Dried Blood Spot","Capitainer","liquid chromatography-tandem mass spectrometry","untargeted metabolites","Mitra"],"curator_keywords":["Human","Metabolomics","Plasma","Volumetric Absorptive MicroSampling (VAMS)","Microsampling","blood metabolome","Dried Blood Spot","Capitainer","liquid chromatography-tandem mass spectrometry","untargeted metabolites","Mitra"],"mass_spectrometry_protocol":["Untargeted metabolomics analysis was performed using reversed-phase (RP) and HILIC chromatography, coupled to a Q-TOF mass spectrometer, in both positive and negative ionization modes. Full details are available in previous publications (Klåvus et al., 2020).
"],"metabolite_name":["LPE 18:0 (18:0_0:0)","LPC P-16:0","L-Phenylalanine","Hippuric acid","LPC 22:6 (22:6/0:0)","p-Cresol sulfate","LPC 18:0 (18:0/0:0)","PC 38:3 (18:0_20:3)","LPE 16:0 (16:0/0:0)","PE O-40:5 (P-18:0/22:4)","LPE 18:1 (18:1/0:0)","3-Carboxy-4-methyl-5-propyl-2-furanpropionic acid","SM 36:1;O2 (d18:0_18:1)","PC 36:4 (16:0_20:4)","Testosterone sulfate","LPC 20:4 (20:4/0:0)","PC 38:6 (16:0_22:6)","PC 34:2 (16:0_18:2)","Hydroxyisovaleric acid","PC 34:0 (16:0_18:0)","PC 32:2 (14:0_18:2)","L-Tryptophan","LPC 16:0 (16:0/0:0)","LPC 14:0 (14:0/0:0)","LPC O-16:0","PE 34:1 (16:0/18:1)","LPC 16:1 (16:1/0:0)","LPC 15:0","LPC 18:1 (18:1/0:0)","LPE 18:2 (18:2/0:0)","PC 32:0 (16:0/16:0)","Indoxyl sulfate","PC 34:3 (16:1_18:2)","LPC 17:0","PC 38:4 (18:0_20:4)","PC 36:5 (16:0_20:5)","PC 36:2 (18:1/18:1)","Pyrocatechol sulfate","PC(18:0/HODE)","PC 36:4 (18:2/18:2)","PC 36:1 (18:0_18:1)","4-Hydroxybenzaldehyde"],"additional_accession":[]},"is_claimable":false,"name":"An LC-MS untargeted metabolomic comparison between three blood microsampling devices, whole blood, and plasma","description":"Blood microsampling (BµS) devices collect less than 100 µL of blood, offering a less invasive and more cost-effective alternative to venipuncture. However, its metabolomic comparability to conventional samples remains unclear, and standardized BµS metabolomic workflows are lacking. This study compared BµS extraction methods and assessed the metabolite coverage of three BµS devices (Mitra®, Capitainer®, and Whatman™ 903) as an alternative to conventional samples (plasma and whole blood) for human biomonitoring. Venous blood from 10 adults (5 males, 5 females) was sampled onto the three devices. First, three extraction techniques (ultrasound, shaker, and homogenizer) were evaluated at three blood concentrations (1.5%, 5.5%, and 11%). The optimized method was then used to compare the metabolite profiles between BµS devices, whole blood, and plasma. Reverse-phase and hydrophilic-interaction chromatography, in positive and negative ionization modes, were combined for Liquid Chromatography–Mass Spectrometry (LC-MS) analysis. All extraction techniques and concentrations proved suitable for BµS untargeted metabolomics. Combining different analytical modes and fragmentation ranges proved useful for maximizing metabolite coverage. BµS-derived metabolite profiles aligned more closely with whole blood than plasma. Some metabolites were more characteristic of a sample type, whereas others were common across sample types. All sample types enabled sex-based differentiation, with metabolites such as testosterone sulfate and hydroxyisovaleric acid driving the separation. These findings enhance our understanding of BµS metabolite coverage and highlight its potential in human biomonitoring. The choice of device depends on the application and the metabolites of interest, offering flexibility for clinical use and research.
","dates":{"publication":"2026-06-01","submission":"2025-07-10"},"accession":"MTBLS10740","cross_references":{"HMDB":["HMDB0000001","HMDB0000479","HMDB0001161","HMDB0240732","HMDB0000034","HMDB0000191 - HMDB0006483","HMDB0000651","HMDB0000201","HMDB0000824","HMDB0002013","HMDB0013128","HMDB0000791","HMDB0000904","HMDB0000064","HMDB0000562","HMDB0004437","HMDB0000148","HMDB0000125","HMDB0000086","HMDB0000043","HMDB0000870","HMDB0029422","HMDB0000157","HMDB0000172","HMDB0000517","HMDB0000168","HMDB0000062","HMDB0061115","HMDB0001645","HMDB0000641","HMDB0000177","HMDB0000182","HMDB0000159","HMDB0000162","HMDB0000167","HMDB0000929","HMDB0000158","HMDB0000883","HMDB0001325","HMDB0001406","HMDB0003152","HMDB0094696","HMDB0005765","HMDB0000214","HMDB0004827","HMDB0012089","HMDB0001257","HMDB0000251","HMDB0000875","HMDB0000008 - HMDB0341410","HMDB0011718","HMDB0000176","HMDB0000407","HMDB0000190","HMDB0000289","HMDB0011749","HMDB0061112","HMDB0001496","HMDB0000501","HMDB0001893","HMDB0001847","HMDB0002250","HMDB0005066","HMDB0002014","HMDB0000222","HMDB0000848","HMDB0005065","HMDB0006469","HMDB0000705","HMDB0000067","HMDB0251427","HMDB0039317","HMDB0000714","HMDB0029737","HMDB0000684","HMDB0010379","HMDB0010381","HMDB0010382","HMDB0010383","HMDB0012108","HMDB0011128","HMDB0010384","HMDB0062651","HMDB0010385","HMDB0010386","HMDB0010393","HMDB0061699","HMDB0010395","HMDB0010404","HMDB0010407","HMDB0011503","HMDB0011506","HMDB0011507","HMDB0011488","HMDB0001860","HMDB0000564","HMDB0007874","HMDB0007940","HMDB0008132","HMDB0007970","HMDB0007971","HMDB0007880","HMDB0007973","HMDB0008134","HMDB0007881","HMDB0008037","HMDB0000593","HMDB0008072","HMDB0008136","HMDB0007983","HMDB0008138","HMDB0007890","HMDB0007984","HMDB0008046","HMDB0008048","HMDB0007991","HMDB0008926","HMDB0008937","HMDB0009093","HMDB0008946","HMDB0008148","HMDB0029377","HMDB0248405","HMDB0002003","HMDB0002825","HMDB0001889","HMDB0000682","HMDB0011130","HMDB0011635","HMDB0059724","HMDB0011391","HMDB0002833"],"pubmed":["42213273"],"PubChem":["CID 71751093","CID 134782884"],"LIPID MAPS":["LMGP01060010"]}}