A cell is a bustling factory carrying out hundreds if not thousands of chemical reactions concurrently. These reactions are compartmentalized into different organelles within the eukaryotic cell and generate specific chemical products with unique functions. Collectively, these chemical products are called metabolites and are direct indicators and vital drivers of cellular activities. In the past decade, advances in chromatography/mass spectrometry–based and nuclear magnetic resonance spectroscopy–based metabolomics have not only been rapidly expanding the list of metabolites but have also enabled analytical quantification of metabolites in different biological samples and linked changes in their levels to a variety of diseases and health issues. Despite the ever increasing power to identify more and more metabolites biochemically, visualizing metabolites at a subcellular level and understanding
spatial dynamics of metabolite actions remain unresolved challenges in cell biology.
Spatial partitioning of metabolic functions is a phenomenon fundamental to life, and these partitions constitute a hierarchy ranging from organs to cells to subcellular organelles. “Where you are is who you are” definitely applies to biological molecules, including metabolites. Many of these metabolites exhibit unique distribution patterns among different organelles in the cell, and their spatial specificity contributes to their functional specificity. First of all, the spatially restricted production of metabolites specifies organelle functions, and is in turn highly sensitive to changes in these functions. As a result, specific metabolites derived from different organelles serve as direct mediators of organelle activities. Second, exchanging metabolites between organelles is a crucial means of metabolic coordination and a key facet of organelle interplay. Last, metabolites can traffic away from their production site and cooperate with other factors to drive signal transduction and transcription responses. Mislocalization of metabolites can disrupt cellular homeostasis, leading to functional decline and deficits. Thus, resolving the spatial distribution of metabolites in vivo is essential for understanding cellular mechanisms governed by metabolites and for revealing pathological mechanisms underlying metabolic disorders.
Richard Feynman said “to answer many of these fundamental biological questions, you just look at the thing!” It is true that advances in microscopic techniques have been transforming the way we study biology, especially cell biology. Nowadays, we can image specific proteins at unprecedented resolution, and the Cell Atlas has mapped over 12,000 proteins at a single-cell level to subcellular structures. Metabolite imaging, on the other hand, is lagging far behind. Direct visualization using fluorescence microscopy is applicable to only a handful of metabolites that intrinsically emit fluorescence, like NADH/NADPH. For non-fluorescent metabolites, one way to detect them is to use fluorescent protein and RNA sensors. These genetically encoded sensors can bind to specific metabolites and undergo conformational changes to generate fluorescent signals. The labeled metabolites can then be visualized at a subcellular level using fluorescence microscopy. Although fluorescent sensors are powerful tools for revealing the spatial distribution of metabolites in living cells, only a small number of sensors have been generated and their selectivity is influenced by a highly complex cellular environment. On the other hand, Raman chemical imaging techniques do not require labeling and can directly visualize metabolites in living cells at a subcellular level. This technique targets metabolites based on their chemical structures, and offers high detection specificity independently of cellular environment. However, when metabolites lack characteristic chemical features, distinguishing them using Raman imaging can be difficult. Mass spectrometry imaging (MSI) is another way to visualize metabolites directly. Although not applicable to living cells, MSI can detect many types of metabolites in a single scan with unprecedented specificity and sensitivity. By reducing the laser spot size using a pinhole or laser-focusing optics, the spatial resolution of MSI can now reach down to ~5 µm, about
the size of a red blood cell. But this resolution is incapable of localizing metabolites to subcellular structures.
Facing this challenge, others and we are working to improve the capacity, selectivity, sensitivity, and resolution of metabolite imaging with current technologies. At the same time, new techniques may emerge that will tackle this problem from a very different angle, or as Sydney Brenner said, “Progress in science depends on new techniques, new discoveries, and new ideas, probably in that order.” Making invisible metabolites visible and mapping a cell atlas of the metabolome will open new avenues to understand the active role these molecules play in organelle interactions, cellular signaling, and epigenetic regulation, and their vital contributions to health and diseases.
About the Author
Meng Wang is Howard Hughes Medical Institute Investigator and holds the Robert C. Fyfe Endowed Chair on Aging in the Department of Molecular and Human Genetics and the Huffington Center on Aging at Baylor College of Medicine.