massNet: integrated processing and classification of spatially resolved mass spectrometry data using deep learning for rapid tumor delineation.

MOTIVATION: Mass spectrometry imaging (MSI) provides rich biochemical information in a label-free manner and therefore holds promise to substantially impact current practice in disease diagnosis. However, the complex nature of MSI data poses computational challenges in its analysis. The complexity of the data arises from its large size, high-dimensionality and spectral nonlinearity. Preprocessing, including peak picking, has been used to reduce raw data complexity; however, peak picking is sensitive to parameter selection that, perhaps prematurely, shapes the downstream analysis for tissue classification and ensuing biological interpretation. RESULTS: We propose a deep learning model, massNet, that provides the desired qualities of scalability, nonlinearity and speed in MSI data analysis. This deep learning model was used, without prior preprocessing and peak picking, to classify MSI data from a mouse brain harboring a patient-derived tumor. The massNet architecture established automatically learning of predictive features, and automated methods were incorporated to identify peaks with potential for tumor delineation. The model's performance was assessed using cross-validation, and the results demonstrate higher accuracy and a substantial gain in speed compared to the established classical machine learning method, support vector machine. AVAILABILITY AND IMPLEMENTATION: https://github.com/wabdelmoula/massNet. The data underlying this article are available in the NIH Common Fund's National Metabolomics Data Repository (NMDR) Metabolomics Workbench under project id (PR001292) with http://dx.doi.org/10.21228/M8Q70T. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Quantitative Imaging of Proteins in Tissue by Stable Isotope Labeled Mimetic Liquid Extraction Surface Analysis Mass Spectrometry.

Absolute quantification of proteins in tissue is important for numerous fields of study. Liquid chromatography-mass spectrometry (LC-MS) methods are the norm but typically involve lengthy sample preparation including tissue homogenization, which results in the loss of information relating to spatial distribution. Here, we propose liquid extraction surface analysis (LESA) mass spectrometry (MS) of stable isotope labeled mimetic tissue models for the spatially resolved quantification of intact ubiquitin in rat and mouse brain tissue. Measured ubiquitin concentrations are in agreement with values found in the literature. Images of rat and mouse brain tissue demonstrate spatial variation in the concentration of ubiquitin and demonstrate the utility of spatially resolved quantitative measurement of proteins in tissue. Although we have focused on ubiquitin, the method has the potential for broader application to the absolute quantitation of any endogenous protein or protein-based drug in tissue.

Metal Oxide Laser Ionization Mass Spectrometry Imaging (MOLI MSI) Using Cerium(IV) Oxide.

Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) is a powerful technique for spatially resolved metabolomics. A variation on MALDI, termed metal oxide laser ionization (MOLI), capitalizes on the unique property of cerium(IV) oxide (CeO2) to induce laser-catalyzed fatty acyl cleavage from lipids and has been utilized for bacterial identification. In this study, we present the development and utilization of CeO2 as an MSI catalyst. The method was developed using a MALDI TOF instrument in negative ion mode, equipped with a high frequency laser. Instrument parameters for MOLI MS fatty acid catalysis with CeO2 were optimized with phospholipid standards and fatty acid catalysis was confirmed using lipid extracts from reference bacterial strains, and sample preparation was optimized using mouse brain tissue. MOLI MSI was applied to the imaging of normal mouse brain revealing differentiable fatty acyl pools in myelinated and nonmyelinated regions. Similarly, MOLI MSI showed distinct fatty acyl composition in tumor regions of a patient derived xenograft mouse model of glioblastoma. To assess the potential of MOLI MSI to detect pathogens directly from tissue, a pseudoinfection model was prepared by spotting Escherichia coli lipid extracts on mouse brain tissue sections and imaged by MOLI MSI. The spotted regions were molecularly resolved from the supporting mouse brain tissue by the diagnostic odd-chained fatty acids and reflected control bacterial MOLI MS signatures. We describe MOLI MSI for the first time and highlight its potential for spatially resolved fatty acyl analysis, characterization of fatty acyl composition in tumors, and its potential for pathogen detection directly from tissue.

Genetically Encoded Fluorescent Proteins Enable High-Throughput Assignment of Cell Cohorts Directly from MALDI-MS Images.

Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) provides a unique in situ chemical profile that can include drugs, nucleic acids, metabolites, lipids, and proteins. MSI of individual cells (of a known cell type) affords a unique insight into normal and disease-related processes and is a prerequisite for combining the results of MSI and other single-cell modalities (e.g. mass cytometry and next-generation sequencing). Technological barriers have prevented the high-throughput assignment of MSI spectra from solid tissue preparations to their cell type. These barriers include obtaining a suitable cell-identifying image (e.g. immunohistochemistry) and obtaining sufficiently accurate registration of the cell-identifying and MALDI-MS images. This study introduces a technique that overcame these barriers by assigning cell type directly from mass spectra. We hypothesized that, in MSI from mice with a defined fluorescent protein expression pattern, the fluorescent protein's molecular ion could be used to identify cell cohorts. A method was developed for the purification of enhanced yellow fluorescent protein (EYFP) from mice. To determine EYFP's molecular mass for MSI studies, we performed intact mass analysis and characterized the protein's primary structure and post-translational modifications through various techniques. MALDI-MSI methods were developed to enhance the detection of EYFP in situ, and by extraction of EYFP's molecular ion from MALDI-MS images, automated, whole-image assignment of cell cohorts was achieved. This method was validated using a well-characterized mouse line that expresses EYFP in motor and sensory neurons and should be applicable to hundreds of commercially available mice (and other animal) strains comprising a multitude of cell-specific fluorescent labels.

Automatic 3D Nonlinear Registration of Mass Spectrometry Imaging and Magnetic Resonance Imaging Data.

Multimodal integration between mass spectrometry imaging (MSI) and radiology-established modalities such as magnetic resonance imaging (MRI) would allow the investigations of key questions in complex biological systems such as the central nervous system. Such integration would provide complementary multiscale data to bridge the gap between molecular and anatomical phenotypes, potentially revealing new insights into molecular mechanisms underlying anatomical pathologies presented on MRI. Automatic coregistration between 3D MSI/MRI is a computationally challenging process due to dimensional complexity, MSI data sparsity, lack of direct spatial-correspondences, and nonlinear tissue deformation. Here, we present a new computational approach based on stochastic neighbor embedding to nonlinearly align 3D MSI to MRI data, identify and reconstruct biologically relevant molecular patterns in 3D, and fuse the MSI datacube to the MRI space. We demonstrate our method using multimodal high-spectral resolution matrix-assisted laser desorption ionization (MALDI) 9.4 T MSI and 7 T in vivo MRI data, acquired from a patient-derived, xenograft mouse brain model of glioblastoma following administration of the EGFR inhibitor drug of Erlotinib. Results show the distribution of some identified molecular ions of the EGFR inhibitor erlotinib, a phosphatidylcholine lipid, and cholesterol, which were reconstructed in 3D and mapped to the MRI space. The registration quality was evaluated on two normal mouse brains using the Dice coefficient for the regions of brainstem, hippocampus, and cortex. The method is generic and can therefore be applied to hyperspectral images from different mass spectrometers and integrated with other established in vivo imaging modalities such as computed tomography (CT) and positron emission tomography (PET).

In Vitro Liquid Extraction Surface Analysis Mass Spectrometry (ivLESA-MS) for Direct Metabolic Analysis of Adherent Cells in Culture.

Conventional metabolomic methods include extensive sample preparation steps and long analytical run times, increasing the likelihood of processing artifacts and limiting high throughput applications. We present here in vitro liquid extraction surface analysis mass spectrometry (ivLESA-MS), a variation on LESA-MS, performed directly on adherent cells grown in 96-well cell culture plates. To accomplish this, culture medium was aspirated immediately prior to analysis, and metabolites were extracted using LESA from the cell monolayer surface, followed by nano-electrospray ionization and MS analysis in negative ion mode. We applied this platform to characterize and compare lipidomic profiles of multiple breast cancer cell lines growing in culture (MCF-7, ZR-75-1, MDA-MB-453, and MDA-MB-231) and revealed distinct and reproducible lipidomic signatures between the cell lines. Additionally, we demonstrated time-dependent processing artifacts, underscoring the importance of immediate analysis. ivLESA-MS represents a rapid in vitro metabolomic method, which precludes the need for quenching, cell harvesting, sample preparation, and chromatography, significantly shortening preparation and analysis time while minimizing processing artifacts. This method could be further adapted to test drugs in vitro in a high throughput manner.

Raster-Mode Continuous-Flow Liquid Microjunction Mass Spectrometry Imaging of Proteins in Thin Tissue Sections.

Mass spectrometry imaging by use of continuous-flow liquid microjunction sampling at discrete locations (array mode) has previously been demonstrated. In this Letter, we demonstrate continuous-flow liquid microjunction mass spectrometry imaging of proteins from thin tissue sections in raster mode and discuss advantages (a 10-fold reduction in analysis time) and challenges (suitable solvent systems, data interpretation) of the approach. Visualization of data is nontrivial, requiring correlation of solvent-flow, mass spectral data acquisition rate, data quality, and liquid microjunction sampling area. The latter is particularly important for determining optimum pixel size. The minimum achievable pixel size is related to the scan time of the instrument used. Here we show a minimum achievable pixel size of 50 µm (x-dimension) when using an Orbitrap Elite; however a pixel size of 600 µm is recommended in order to minimize the effects of oversampling on image accuracy.

MALDI Imaging of Liquid Extraction Surface Analysis Sampled Tissue.

Combined mass spectrometry imaging methods in which two different techniques are executed on the same sample have recently been reported for a number of sample types. Such an approach can be used to examine the sampling effects of the first technique with a second, higher resolution method and also combines the advantages of each technique for a more complete analysis. In this work matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI MSI) was used to study the effects of liquid extraction surface analysis (LESA) sampling on mouse brain tissue. Complementary multivariate analysis techniques including principal component analysis, non-negative matrix factorization, and t-distributed stochastic neighbor embedding were applied to MALDI MS images acquired from tissue which had been sampled by LESA to gain a better understanding of localized tissue washing in LESA sampling. It was found that MALDI MS images could be used to visualize regions sampled by LESA. The variability in sampling area, spatial precision, and delocalization of analytes in tissue induced by LESA were assessed using both single-ion images and images provided by multivariate analysis.

Liquid Extraction Surface Analysis Mass Spectrometry Coupled with Field Asymmetric Waveform Ion Mobility Spectrometry for Analysis of Intact Proteins from Biological Substrates.

Previously we have shown that liquid extraction surface analysis (LESA) mass spectrometry is suitable for the analysis of intact proteins from a range of biological substrates. Here we show that LESA mass spectrometry may be coupled with high field asymmetric waveform ion mobility spectrometry (FAIMS) for top-down protein analysis directly from thin tissue sections (mouse liver, mouse brain) and from bacterial colonies (Escherichia coli) growing on agar. Incorporation of FAIMS results in significant improvements in signal-to-noise and reduced analysis time. Abundant protein signals are observed in single scan mass spectra. In addition, FAIMS enables gas-phase separation of molecular classes, for example, lipids and proteins, enabling improved analysis of both sets of species from a single LESA extraction.

Direct analysis of intact proteins from Escherichia coli colonies by liquid extraction surface analysis mass spectrometry.

Top-down identification of proteins by liquid extraction surface analysis (LESA) mass spectrometry has previously been reported for tissue sections and dried blood spot samples. Here, we present a modified "contact" LESA method for top-down analysis of proteins directly from living bacterial colonies grown in Petri dishes, without any sample pretreatment. It was possible to identify a number of proteins by use of collision-induced dissociation tandem mass spectrometry followed by searches of the data against an E. coli protein database. The proteins identified suggest that the method may provide insight into the bacterial response to environmental conditions. Moreover, the results show that the "contact" LESA approach results in a smaller sampling area than typical LESA, which may have implications for spatial profiling.

Multimodal platform for assessing drug distribution and response in clinical trials.

BACKGROUND: Response to targeted therapy varies between patients for largely unknown reasons. Here, we developed and applied an integrative platform using mass spectrometry imaging (MSI), phosphoproteomics, and multiplexed tissue imaging for mapping drug distribution, target engagement, and adaptive response to gain insights into heterogeneous response to therapy. METHODS: Patient-derived xenograft (PDX) lines of glioblastoma were treated with adavosertib, a Wee1 inhibitor, and tissue drug distribution was measured with MALDI-MSI. Phosphoproteomics was measured in the same tumors to identify biomarkers of drug target engagement and cellular adaptive response. Multiplexed tissue imaging was performed on sister sections to evaluate spatial co-localization of drug and cellular response. The integrated platform was then applied on clinical specimens from glioblastoma patients enrolled in the phase 1 clinical trial. RESULTS: PDX tumors exposed to different doses of adavosertib revealed intra- and inter-tumoral heterogeneity of drug distribution and integration of the heterogeneous drug distribution with phosphoproteomics and multiplexed tissue imaging revealed new markers of molecular response to adavosertib. Analysis of paired clinical specimens from patients enrolled in the phase 1 clinical trial informed the translational potential of the identified biomarkers in studying patient's response to adavosertib. CONCLUSIONS: The multimodal platform identified a signature of drug efficacy and patient-specific adaptive responses applicable to preclinical and clinical drug development. The information generated by the approach may inform mechanisms of success and failure in future early phase clinical trials, providing information for optimizing clinical trial design and guiding future application into clinical practice.

Spatial Distribution of Transcytosis Relevant Phospholipids in Response to Omega-3 Dietary Deprivation.

The cell membrane of brain endothelial cells is enriched in omega-3 phospholipid species. Numerous omega-3 phospholipid species were recently proposed to be important for maintaining the low rate of transcytosis and, thus, could be important for regulating one of the mechanisms of the blood brain barrier (BBB). However, the spatial distribution of these phospholipid species within the brain was previously unknown. Here, we combined advanced mass spectrometry imaging techniques to generate a map of these phospholipids in the brain at near single cell resolution. Furthermore, we explored the effects of omega-3 dietary deprivation on both docosahexaenoic acid (DHA)-containing phospholipids and the global brain phospholipid profile. We demonstrate the unique spatial distribution of individual DHA-containing phospholipids, which may be important for the regiospecific properties of the BBB. Finally, 24 diet discriminative phospholipids were identified and showed an increase in saturated phospholipid species and ceramide containing phospholipid species under omega-3 dietary deficiency.

Peak learning of mass spectrometry imaging data using artificial neural networks.

Mass spectrometry imaging (MSI) is an emerging technology that holds potential for improving, biomarker discovery, metabolomics research, pharmaceutical applications and clinical diagnosis. Despite many solutions being developed, the large data size and high dimensional nature of MSI, especially 3D datasets, still pose computational and memory complexities that hinder accurate identification of biologically relevant molecular patterns. Moreover, the subjectivity in the selection of parameters for conventional pre-processing approaches can lead to bias. Therefore, we assess if a probabilistic generative model based on a fully connected variational autoencoder can be used for unsupervised analysis and peak learning of MSI data to uncover hidden structures. The resulting msiPL method learns and visualizes the underlying non-linear spectral manifold, revealing biologically relevant clusters of tissue anatomy in a mouse kidney and tumor heterogeneity in human prostatectomy tissue, colorectal carcinoma, and glioblastoma mouse model, with identification of underlying m/z peaks. The method is applied for the analysis of MSI datasets ranging from 3.3 to 78.9 GB, without prior pre-processing and peak picking, and acquired using different mass spectrometers at different centers.

Interim clinical trial analysis of intraoperative mass spectrometry for breast cancer surgery.

Optimal resection of breast tumors requires removing cancer with a rim of normal tissue while preserving uninvolved regions of the breast. Surgical and pathological techniques that permit rapid molecular characterization of tissue could facilitate such resections. Mass spectrometry (MS) is increasingly used in the research setting to detect and classify tumors and has the potential to detect cancer at surgical margins. Here, we describe the ex vivo intraoperative clinical application of MS using a liquid micro-junction surface sample probe (LMJ-SSP) to assess breast cancer margins. In a midpoint analysis of a registered clinical trial, surgical specimens from 21 women with treatment naïve invasive breast cancer were prospectively collected and analyzed at the time of surgery with subsequent histopathological determination. Normal and tumor breast specimens from the lumpectomy resected by the surgeon were smeared onto glass slides for rapid analysis. Lipidomic profiles were acquired from these specimens using LMJ-SSP MS in negative ionization mode within the operating suite and post-surgery analysis of the data revealed five candidate ions separating tumor from healthy tissue in this limited dataset. More data is required before considering the ions as candidate markers. Here, we present an application of ambient MS within the operating room to analyze breast cancer tissue and surgical margins. Lessons learned from these initial promising studies are being used to further evaluate the five candidate biomarkers and to further refine and optimize intraoperative MS as a tool for surgical guidance in breast cancer.

PHD3 Loss Promotes Exercise Capacity and Fat Oxidation in Skeletal Muscle.

Rapid alterations in cellular metabolism allow tissues to maintain homeostasis during changes in energy availability. The central metabolic regulator acetyl-CoA carboxylase 2 (ACC2) is robustly phosphorylated during cellular energy stress by AMP-activated protein kinase (AMPK) to relieve its suppression of fat oxidation. While ACC2 can also be hydroxylated by prolyl hydroxylase 3 (PHD3), the physiological consequence thereof is poorly understood. We find that ACC2 phosphorylation and hydroxylation occur in an inverse fashion. ACC2 hydroxylation occurs in conditions of high energy and represses fatty acid oxidation. PHD3-null mice demonstrate loss of ACC2 hydroxylation in heart and skeletal muscle and display elevated fatty acid oxidation. Whole body or skeletal muscle-specific PHD3 loss enhances exercise capacity during an endurance exercise challenge. In sum, these data identify an unexpected link between AMPK and PHD3, and a role for PHD3 in acute exercise endurance capacity and skeletal muscle metabolism.

Localized Metabolomic Gradients in Patient-Derived Xenograft Models of Glioblastoma.

Glioblastoma (GBM) is increasingly recognized as a disease involving dysfunctional cellular metabolism. GBMs are known to be complex heterogeneous systems containing multiple distinct cell populations and are supported by an aberrant network of blood vessels. A better understanding of GBM metabolism, its variation with respect to the tumor microenvironment, and resulting regional changes in chemical composition is required. This may shed light on the observed heterogeneous drug distribution, which cannot be fully described by limited or uneven disruption of the blood-brain barrier. In this work, we used mass spectrometry imaging (MSI) to map metabolites and lipids in patient-derived xenograft models of GBM. A data analysis workflow revealed that distinctive spectral signatures were detected from different regions of the intracranial tumor model. A series of long-chain acylcarnitines were identified and detected with increased intensity at the tumor edge. A 3D MSI dataset demonstrated that these molecules were observed throughout the entire tumor/normal interface and were not confined to a single plane. mRNA sequencing demonstrated that hallmark genes related to fatty acid metabolism were highly expressed in samples with higher acylcarnitine content. These data suggest that cells in the core and the edge of the tumor undergo different fatty acid metabolism, resulting in different chemical environments within the tumor. This may influence drug distribution through changes in tissue drug affinity or transport and constitute an important consideration for therapeutic strategies in the treatment of GBM. SIGNIFICANCE: GBM tumors exhibit a metabolic gradient that should be taken into consideration when designing therapeutic strategies for treatment.See related commentary by Tan and Weljie, p. 1231.

Pre- and Postoperative Neratinib for HER2-Positive Breast Cancer Brain Metastases: Translational Breast Cancer Research Consortium 022.

PURPOSE: This pilot study was performed to test our ability to administer neratinib monotherapy before clinically recommended craniotomy in patients with HER2-positive metastatic breast cancer to the central nervous system, to examine neratinib's central nervous system penetration at craniotomy, and to examine postoperative neratinib maintenance. PATIENTS AND METHODS: Patients with HER2-positive brain metastases undergoing clinically indicated cranial resection of a parenchymal tumor received neratinib 240 mg orally once a day for 7 to 21 days preoperatively, and resumed therapy postoperatively in 28-day cycles. Exploratory evaluations of time to disease progression, survival, and correlative tissue, cerebrospinal fluid (CSF), and blood-based analyses examining neratinib concentrations were planned. The study was registered at ClinicalTrials.gov under number NCT01494662. RESULTS: We enrolled 5 patients between May 22, 2013, and October 18, 2016. As of March 1, 2019, patients had remained on the study protocol for 1 to 75+ postoperative cycles pf therapy. Two patients had grade 3 diarrhea. Evaluation of the CSF showed low concentrations of neratinib; nonetheless, 2 patients continued to receive therapy without disease progression for at least 13 cycles, with one on-study treatment lasting for nearly 6 years. Neratinib distribution in surgical tissue was variable for 1 patient, while specimens from 2 others did not produce conclusive results as a result of limited available samples. CONCLUSION: Neratinib resulted in expected rates of diarrhea in this small cohort, with 2 of 5 patients receiving the study treatment for durable periods. Although logistically challenging, we were able to test a limited number of CSF- and parenchymal-based neratinib concentrations. Our findings from resected tumor tissue in one patient revealed heterogeneity in drug distribution and tumor histopathology.

Rapid MALDI mass spectrometry imaging for surgical pathology.

Matrix assisted laser desorption ionization mass spectrometry imaging (MALDI MSI) is an emerging analytical technique, which generates spatially resolved proteomic and metabolomic images from tissue specimens. Conventional MALDI MSI processing and data acquisition can take over 30 min, limiting its clinical utility for intraoperative diagnostics. We present a rapid MALDI MSI method, completed under 5 min, including sample preparation and analysis, providing a workflow compatible with the clinical frozen section procedure.

Molecular Characterization of Prostate Cancer with Associated Gleason Score Using Mass Spectrometry Imaging.

Diagnosis of prostate cancer is based on histologic evaluation of tumor architecture using a system known as the "Gleason score." This diagnostic paradigm, while the standard of care, is time-consuming, shows intraobserver variability, and provides no information about the altered metabolic pathways, which result in altered tissue architecture. Characterization of the molecular composition of prostate cancer and how it changes with respect to the Gleason score (GS) could enable a more objective and faster diagnosis. It may also aid in our understanding of disease onset and progression. In this work, we present mass spectrometry imaging for identification and mapping of lipids and metabolites in prostate tissue from patients with known prostate cancer with GS from 6 to 9. A gradient of changes in the intensity of various lipids was observed, which correlated with increasing GS. Interestingly, these changes were identified in both regions of high tumor cell density, and in regions of tissue that appeared histologically benign, possibly suggestive of precancerous metabolomic changes. A total of 31 lipids, including several phosphatidylcholines, phosphatidic acids, phosphatidylserines, phosphatidylinositols, and cardiolipins were detected with higher intensity in GS (4+3) compared with GS (3+4), suggesting they may be markers of prostate cancer aggression. Results obtained through mass spectrometry imaging studies were subsequently correlated with a fast, ambient mass spectrometry method for potential use as a clinical tool to support image-guided prostate biopsy. IMPLICATIONS: In this study, we suggest that metabolomic differences between prostate cancers with different Gleason scores can be detected by mass spectrometry imaging.

Integrated mapping of pharmacokinetics and pharmacodynamics in a patient-derived xenograft model of glioblastoma.

Therapeutic options for the treatment of glioblastoma remain inadequate despite concerted research efforts in drug development. Therapeutic failure can result from poor permeability of the blood-brain barrier, heterogeneous drug distribution, and development of resistance. Elucidation of relationships among such parameters could enable the development of predictive models of drug response in patients and inform drug development. Complementary analyses were applied to a glioblastoma patient-derived xenograft model in order to quantitatively map distribution and resulting cellular response to the EGFR inhibitor erlotinib. Mass spectrometry images of erlotinib were registered to histology and magnetic resonance images in order to correlate drug distribution with tumor characteristics. Phosphoproteomics and immunohistochemistry were used to assess protein signaling in response to drug, and integrated with transcriptional response using mRNA sequencing. This comprehensive dataset provides simultaneous insight into pharmacokinetics and pharmacodynamics and indicates that erlotinib delivery to intracranial tumors is insufficient to inhibit EGFR tyrosine kinase signaling.