Metabolomics—often referred to as the youngest of the omics—provides key insight into phenotype. However, bulk metabolomics requires the homogenization of the sample and is thus unable to discern metabolic differences at a cellular level. This leads to the question of whether our readers are likely to encounter spatial metabolomics in the near future. Will it soon be relatively easy to apply spatial metabolomics at the scale of single cells and even organelles? Or are the complications insurmountable for most research groups?
Metabolomics as a snapshot of phenotype
The goal of metabolomics is to identify and quantify metabolites. A metabolite in this context can be defined as any small molecule (with a molecular weight under 2,000 daltons) found within or produced by a biological system. These metabolites encompass a vast array of compounds—including fuel substrates like glucose, fatty acids, phospholipids, signaling molecules, steroids, drugs, and environmental toxins.
Search Mass Spectrometry Search Now Search our directory to find mass spec tools for your research needs.
Metabolomics can be thought of as a chemical readout of a biological phenotype. Erica Forsberg, Ph.D., works with Bruker Scientific’s Life Science Mass Spectrometry team as a Market Manager for Metabolomics Americas, while Kate Stumpo, Ph.D., is a business Development Manager for Bruker Scientific’s SpatialOMx. As Forsberg and Stumpo note, “When we look at the biological cascade from genes to proteins to metabolites, the small molecule metabolites tend to be closest to the expressing phenotype. This has a major impact on understanding upstream biological processes and developing therapeutics.”
The emergence of mass spectrometry imaging
Although multiple approaches toward spatial metabolomics are possible—including fluorescence lifetime imaging (FLIM), Fourier-transform infrared spectroscopy (FTIR), and Raman spectroscopy—mass spectrometry imaging (MSI) has gained the most momentum. With MSI, the surface of a sample (often a tissue section) is divided into a series of grids or pixels. Molecules are desorbed from each pixel using an ablation method and then analyzed to create a mass spectrum of each individual pixel. As Zach Pitluk, Ph.D., Vice President of Life Science Business Development for Paradigm4, explains, “Basically, you have a laser beam that’s scanning along and vaporizing a layer of tissue. And it splatters. It’s like a microwave. It’s a huge mess. But they are able to pull out the signals for the small molecules.”
Different MSI methods vary based on the source type—the mechanism by which molecules are ablated from a sample. On a fundamental level, Pitluk explains that spatial metabolomics will require “any instrument that has the beam small enough so you can accommodate the cell.” Matrix-assisted laser desorption/ionization (MALDI) and desorption electrospray ionization (DESI) are the two most commonly employed MSI techniques due to their sensitivity and reliability. David Heywood, Senior Manager of Bioresearch Global Marketing with Waters Corporation, explains that both methods “favor slightly different sets of molecules, with DESI being slightly more suitable for some metabolites such as amino acids.” Forsberg and Stumpo highlight the benefits of MALDI: “MALDI imaging provides researchers with the ability to specifically visualize where changes are occurring in a tissue that can impact how disease-based research is interpreted.”
Major challenges associated with spatial metabolomics
According to Amir Hossen, Applications Scientist at Shimadzu Scientific Instruments, the area of ablation in MSI makes very few ions. “The sensitivity of the instrument with high mass accuracy is still a barrier in spatial metabolomics analysis.” He also points out the technical difficulties associated with trying to integrate a microscopic camera with MSI.
Unlike genomics and proteomics, in which the molecules being analyzed are structurally similar, metabolites are chemically diverse compounds with a broad range of polarity, hydrophobicity, and size. And while there are likely hundreds of thousands of proteins that can be analyzed with proteomics, the number of potential metabolites is likely in the millions. As Pitluk explains, “You are doing spectral library matching. You see this with proteomics. Metabolomics is 100 times harder."
Only about 1% of molecules can be annotated out of millions of metabolites. As Pitluk notes, the researchers must “figure out how to aggregate the fragments with the molecules. Some of them break down. You have half a molecule here, half a molecule there. It is going to be a massive computation problem.” To help overcome these issues, hundreds of researchers have already used METASPACE, a new open-source platform for metabolite annotation of MSI data.
In addition to the technical challenges associated with MSI, Heywood notes that “Reproducibility is one of the biggest challenges for metabolomics, spatial metabolomics, and discovery science in general. This is especially true when considering the variety of phenotypes in human studies.”
Recent advances in spatial metabolomics
Although spatial metabolomics is still in its infancy, a growing body of work has already been published. As of 2022, key papers have used spatial metabolomics to elucidate metabolic pathways in breast cancer, esophageal cancer, glioblastoma, and lung cancer. Other recent studies have investigated spatial metabolic profiles in insects, plants, and the human body (the kidney, the heart, the brain, the liver, the lung, and cytotoxic T cells).
In particular, MSI-mediated spatial metabolomics has recently been harnessed to study the alteration of metabolites in people with cancer. Spatial resolution is key because the metabolic heterogeneity of cancer contributes significantly to its poor treatment outcomes. As Heywood explains, “A tumor consists of a very heterogeneous collection of cells. Isolating specific cancer-related biological processes becomes more feasible at single-cell levels.”
Despite the technical difficulties, spatial metabolomics has already yielded some fascinating results. A 2019 study used MALDI-MSI to elucidate where HIV drugs become localized within the brain. Another group has documented changes in lipid compositions within liver cells for an in vitro model of lipid metabolic disorders. Finally, several recent studies have employed spatial metabolomics at an organelle scale to understand metabolic changes in vesicles within neural cells and lysosomes. At this rate, it seems likely that many of our readers will indeed be encountering spatial metabolomics at some point in the near future!