Unraveling disease complexity with integrated spatial systems biology approach

Interrogate complex biological systems through spatially-defined quantification of numerous biomolecules.

System biology involves complex biological organization and processes in terms of molecular constituents ordered in many interconnected pathways. A holistic approach is critical to our understanding of the complexity of disease and the relationship among drug targets-disease. Changes in molecular pathways can be detected using high-throughput omics methods such as transcriptomics, proteomics, metabolomics and integration of multiple omics datasets which is becoming key to revealing novel biological insights. In this perspective, we highlight the use of an image based multi-omics integration approach to unravel information flow and mechanisms during complex biological events.

Approach

In this case study, we describe a multimodal approach by allowing the parallel spatially-defined quantification of biomolecules. Our newly developed workflow allows in-situ comprehensive molecular architecture and phenotypes analysis that have eminent potential to classify disease according to their distinct characteristics. The workflow includes the import and the overlay of a molecular image obtained from a tissue section analyzed by Quantitative Mass Spectrometry Imaging (QMSI) onto an adjacent tissue section prepared for GeoMx™ Digital Spatial Profiler (DSP) analysis with fluorescent morphological makers. The selection of the regions of interest (ROIs) on the GeoMx DSP is then based on the molecular signal provided by the QMSI analysis. This way the subsequent protein/gene expression analysis can be directly linked to a pertinent metabolic activity and allow the accurate understanding of the interplay between biomarkers within the complexity of a disease.

Case Illustration Study

In this case study, we report an example of lactate distribution derived from the analysis of a PDX tissue section by QMSI . The lactate accumulation in this tissue section showed an heterogenous distribution along the tissue section. The overlay of the molecular image with the Hematoxylin-eosin (H&E) staining on an adjacent tissue section clearly showed that the lactate mostly accumulates in the different tumor regions of the PDX model . The concentration of lactate within the different tumor regions was also heterogenous.

The QMSI-H&E overlay image was then imported into the Geomx DSP device and further aligned to the fluorescently labeled adjacent tissue section. The fluorescent morphologic markers delimitated the tumor (panCK) versus the stroma content (CD45). The alignment between the QMSI-H&E and the fluorescent image was manually performed and once both images were correctly aligned, the ROIs for protein/RNA analysis were delineated based on the QMSI signal. The drawn ROIs corresponded to the tumor regions with different lactate accumulation.

Proteins were then quantified from each selected ROIs. The expression profile obtained showed some of the protein targets within different ROIs corresponding to high lactate and low lactate concentration. The expression profile exhibited a differential protein expression that was dependent on the concentration of lactate. They were low expressing T cell markers in the high lactate regions showing that in these regions the T cell population were depleted.

Applications

Deciphering the molecular complexity at the spatial level within the tissue context remains one of the prerogatives to understand the heterogeneity of the mechanisms that give rise to various disease states and how cells interact with each other to help identifying therapeutics pathways for new treatments.

The limitations of current tools due to their inability at providing spatial information and their cumbersome workflow for understanding heterogeneity inhibit personalized medicine and the discovery of advanced therapeutics. In fighting diseases like Alzheimer‘s, diabetes and cancer, the proposed methodology by correlating the spatial information in tissue morphology with multi-omics allows to:

• Probe the more complex and transient molecular changes that underpin the course of the disease and response to treatment
• Help select the right drug target

---