Application of a Multiomics Imaging Workflow to Explore Asparlas Treatment in Solid Tumors

Abstract

In acute lymphoblastic leukemia (ALL), hypermethylation of the asparagine synthetase (ASNS) gene promoter, leading to low levels of ASNS in tumor cells, is recognized as a prognostic biomarker, and l-asparaginase-based treatments (e.g., Asparlas) are frequently administered to these patients. In these cancers, tumor cells rely on external asparagine, and its depletion in the bloodstream results in tumor cell apoptosis. A multiomics (imaging) workflow is required to evaluate key molecular changes and characterize solid tumors to explore the potential efficacy of Asparlas in solid tumors. This study introduces a multiomics imaging workflow applicable to solid tumor specimens for the comprehensive molecular profiling of Asparlas treatment effects. The workflow integrates matrix-assisted laser desorption-ionization mass spectrometry imaging (MALDI-MSI), liquid chromatography coupled with high-resolution mass spectrometry, and histopathological staining on consecutive tumor tissue sections. It enables the detection and analysis of metabolites, lipids, and proteins. Tumor characterization was achieved through histology and clustering analysis based on lipid signatures, yielding consistent annotations. On-tissue chemical derivatization followed by MALDI-MSI was performed to assess metabolic alterations, with a focus on amino acids. ASNS distribution was mapped utilizing targeted MALDI-immunohistochemistry, followed by untargeted (spatial) proteomics on adjacent tissue sections. This study established a multiomics imaging approach and demonstrated its applicability in elucidating the metabolic changes in tumor tissue consequent to Asparlas treatment. Furthermore, it highlights the added value of multiomics imaging in pharmaceutical research and development.

1

SNU-601 induced tumor characterization via histology and molecular footprint clustering. CA derivatization was applied prior MALDI-MSI analysis to enable the simultaneous detection of metabolites (e.g., amino acids) and lipids. CA derivatization caused a yellow stain on tissue, for which histology was performed on a consecutive slide, followed by annotations. H&E-based annotations were categorized in red = tumor, black = necrosis, green = exudate, blue = stroma, yellow = immune cells. Based on acquired spatial lipidomic data, segmentation was performed, which resulted in two main clusters. Cluster 1 in yellow represents tumor tissue, and cluster 2 in red represents necrotic and stroma tissue.

2

Spatial lipidomic analysis of SNU-601 induced tumor tissue. A. Average spectrum of the obtained MALDI-MSI analysis. In blue: average spectrum of the tumor regions. In green: average spectrum of the necrotic regions. B. Volcano plot of the 59 identified lipids from the MALDI-MSI data set. In green are the lipids that were found to be significantly upregulated in tumor regions compared to the necrotic regions. C. Ion images of the 10 lipids that are primarily responsible for the segmentation clustering.

3

Metabolic profiling of Asparlas dosed SNU-601 induced tumor tissue after performing OTCD and MALDI-MSI. A. Derivatization reaction between 4-hydroxy-3-methoxycinnamaldehyde (CA) and asparagine (Asn). B. Average spectrum of MALDI-MSI generated analysis in the CA-derivatized amino acid region after root-mean-square (RMS) normalization. C. and D. MALDI-MSI analysis of amino acids, including Asn and glutamine (Gln) in SNU-601 induced tumor tissue. The distribution of CA-Asn (m/z 293) and of CA-Gln (m/z 307) was visualized after RMS normalization. The bars represent RMS normalized intensities of Asn and Gln in tumor and necrotic regions. Significance was determined by performing an ANOVA on the RMS normalized relative intensities of Asn and Gln. Significance is shown as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

4

Untargeted (spatial) proteomics analysis of SNU-601 induced tumor tissue. A. A 2D PCA plot of the spatial proteomics MALDI-MSI data set, with the axes representing PC₃ and PC₄. 95% confidence ellipses are presented for each experimental group. B. Representative peptide spectrum after root-mean-square (RMS) normalization of SNU-601 induced tumor tissue measured by MALDI-MSI. The spatial distribution of four features that were identified as peptides is presented: collagen-α1 (m/z 971.57), actin (m/z 1198.72), myosin (m/z 1521.77), and keratine-18 (m/z 1580.77).

5

Asparagine synthesis (ASNS) distribution in SNU-601 induced tumor tissue using MALDI-IHC. A. Asparagine is synthesized in an ATP-dependent manner from aspartate and glutamine by asparagine synthetase. B. MALDI-IHC was performed using a PCMT-labeled ASNS antibody (label m/z 1506.77), and the corresponding ion images is presented. Average spectrum of the control sample is shown using a root-mean-square (RMS) normalization.