Quantitative Mass Spectrometry Imaging Using Multivariate Curve Resolution and Deep Learning: A Case Study

In the present contribution, a novel approach based on multivariate curve resolution and deep learning (DL) is proposed for quantitative mass spectrometry imaging (MSI) as a potent technique for identifying different compounds and creating their distribution maps in biological tissues without need for sample preparation. As a case study, chlordecone as a carcinogenic pesticide was quantitatively determined in mouse liver using matrix-assisted laser desorption ionization-MSI (MALDI-MSI). For this purpose, data from seven standard spots containing 0 to 20 picomoles of chlordecone and four unknown tissues from the mouse livers infected with chlordecone for 1, 5, and 10 days were analyzed using a convolutional neural network (CNN). To solve the lack of sufficient data for CNN model training, each pixel was considered as a sample, the designed CNN models were trained by pixels in training sets, and their corresponding amounts of chlordecone were obtained by multivariate curve resolution-alternating least-squares (MCR-ALS). The trained models were then externally evaluated using calibration pixels in test sets for 1, 5, and 10 days of exposure, respectively. Prediction R² for all three data sets ranged from 0.93 to 0.96, which was superior to support vector machine (SVM) and partial least-squares (PLS). The trained CNN models were finally used to predict the amount of chlordecone in mouse liver tissues, and their results were compared with MALDI-MSI and GC-MS methods, which were comparable. Inspection of the results confirmed the validity of the proposed method.

Visualizing the distribution of curcumin in the root of Curcuma longa via VUV-postionization mass spectrometric imaging

Curcumin is a dietary spice and coloring agent widely used in food and herbal medicine. Herein, we visualized the distribution of curcumin in fresh Curcuma longa (turmeric) root sections using the state-of-the-art vacuum-ultraviolet (VUV, 118 nm) single photon-postionization mass spectrometric imaging method. Compared with other mass spectrometric imaging methods, the proposed method does not require any sample pre-treatment. The proposed approach could be more conducive to in situ detection of small molecules. The mass spectroscopic imaging (MSI) images of curcumin sections with a lateral resolution of 100 μm indicated that the concentrations of curcumin decreased from the phloem to the xylem of the root. We also show MS imaging of curcumin in the turmeric root at different maturity periods, revealing the transformation of this endogenous species. The result of quantitative analysis indicates that the total curcumin content of the mature turmeric root is estimated to be 3.43%, which is consistent with the previous report that the content of curcumin in the turmeric root is estimated between 3% and 5%. The report indicated that the proposed method of VUV single photon postionization MSI can be used to explore the metabolic process of plants, which is critical for herbal farming, harvest, and its ingredient extraction.

A review on quantitation-related factors and quantitation strategies in mass spectrometry imaging of small biomolecules

Accurate quantitative information of the analytes in mass spectrometry imaging (MSI) is fundamental for determining the accurate spatial distribution, which can provide additional insight into the living processes, disease progression or the pharmacokinetic-pharmacodynamic mechanisms. However, performing a quantitative analysis in MSI is still challenging. This review focuses on the quantitation-related factors and recent advances in the strategies of quantitative MSI (q-MSI) of small molecules. The main quantitation-related factors are discussed according to the new investigations in recent years, including the regionally varied extraction efficiencies and ionization efficiencies, signal-concentration regression functions, and the repeatability of surface sampling/ionization methods. Newly developed quantitation strategies in MSI based on aforementioned factors are introduced, including new techniques in standard curve calibration with normalization to an internal standard, kinetic calibration, and chemometric methods. Different strategies for validating q-MSI methods are discussed. Finally, the future perspectives to q-MSI are proposed.

Per-pixel absolute quantitation for mass spectrometry imaging of endogenous lipidomes by model prediction of mass transfer kinetics in single-probe-based ambient liquid extraction

With the development of mass spectrometry imaging (MSI), techniques providing quantitative information on the spatial distribution have attracted more attentions recent years. However, for MSI of endogenous compounds in bio-samples, the uncertainty of locally varied sampling efficiencies always hinders accurate absolute quantitation. Here single-probe was used for ambient liquid extraction MSI in rat cerebellum, and standards of phosphatidylcholines (PCs) and cerebrosides (CBs) were doped in extraction solvent. The extraction kinetic curves of endogenous lipids in the ambient liquid extraction during probe parking in single pixel of tissue were investigated. From the results, the extraction kinetic curves were varied between different lipid species in different brain regions, resulting in variations of extraction efficiencies between imaging pixels, and calibration with standards deposited in tissue could not compensate for the variations. In our approach, the theoretical kinetic model of ambient liquid extraction was established, and original concentrations of endogenous lipids in each pixel of tissue were predicted by fitting the experimental extraction kinetic curve in each imaging pixel to the model. The experimental data was demonstrated to be well fitted to the kinetic model with R² > 0.86, and only with 18-s extraction in each pixel, the original lipid concentrations were predicted accurately with relative errors <23%. With the new method, totally 157 lipids and small metabolites were imaged, and per-pixel quantitation was achieved for 19 PCs and 4 CBs. Compared with conventional quantitative MSI (q-MSI) method, the new q-MSI method had better reproducibility and wider linear range, and produced better contrast in the quantitative images of lipids in brain tissue with less hot spots and noises. The absolute quantitation results by the new method were verified by quantitative LC-MS method with Pearson'r > 0.9 and the slope of the linear fitting line of the correlation plot near 1.

Applications of mass spectrometry imaging in virus research

Mass spectrometry imaging (MSI) is a label-free molecular imaging technique allowing an untargeted detection of a broad range of biomolecules and xenobiotics. MSI enables imaging of the spatial distribution of proteins, peptides, lipids and metabolites from a wide range of samples. To date, this technique is commonly applied to tissue sections in cancer diagnostics and biomarker development, but also molecular histology in general. Advances in the methodology and bioinformatics improved the resolution of MS images below the single cell level and increased the flexibility of the workflow. However, MSI-based research in virology is just starting to gain momentum and its full potential has not been exploited yet. In this review, we discuss the main applications of MSI in virology. We review important aspects of matrix-assisted laser desorption/ionization (MALDI) MSI, the most widely used MSI technique in virology. In addition, we summarize relevant literature on MSI studies that aim to unravel virus-host interactions and virus pathogenesis, to elucidate antiviral drug kinetics and to improve current viral disease diagnostics. Collectively, these studies strongly improve our general understanding of virus-induced changes in the proteome, metabolome and metabolite distribution in host tissues of humans, animals and plants upon infection. Furthermore, latest MSI research provided important insights into the drug distribution and distribution kinetics, especially in antiretroviral research. Finally, MSI-based investigations of oncogenic viruses greatly increased our knowledge on tumor mass signatures and facilitated the identification of cancer biomarkers.

Multivariate Calibration for Carbon Nanotubes in the Environment Using the Microwave Induced Heating Method

The goal of the present paper is to develop chemometrics-based multivariate calibration approaches for simultaneously determining quantity of individual carbon nanotubes (CNTs) in a multicomponent environmental matrix using a microwave induced heating method. A multifactor and multilevel experiment design was used to create 4 separate calibration datasets. Each calibration dataset contained 25 orthogonal CNTs with 2 or 3 factors (CNTs: single-walled CNTs (SWCNTs)/multi-walled CNTs (MWCNTs)/carboxylated MWCNTs (MWCNT-COOH)) and 5 levels (CNTs mass). The temperature rise (ΔT) spectral information was obtained for each sample by exposing to varying microwave conditions. This study showed the potential and applicability of partial least square regression (PLS), least square-support vector machine (LS-SVM) and artificial neural networks (ANN) in predicting quantities of SWCNTs, MWCNTs and MWCNT-COOH in environmental matrices with microwave induced temperature rises data. Our results revealed that the developed LS-SVM model presented higher R2 and lower root mean square error of prediction (RMSEP) (R2 = 0.74-0.93, RMSEP =0.0251 mg to 0.0328 mg in 2-component systems and R2 = 0.64-0.95, RMSEP = 0.0243 mg to 0.0410 mg in 3-component systems), while the ANN model was only accurate in estimating mass of SWCNT and MWCNT in a 2-component mixture (R2 = 0.77-0.89, RMSEP = 0.0322 mg to 0.0503 mg). The PLS model was found not effectively interpret relationship between microwave induced temperature rises data and mass of CNTs, indicated by small R2 (0.20-0.87) and large RMSEP (0.0209 mg -0.1021 mg).