Unraveling the unique biochemistry of brain metabolism
Some of the most interesting, enigmatic and understudied cells in metabolic biochemistry are those of the nervous system. The brain has unique metabolic requirements and expresses unique metabolic enzymes, many of which remain poorly characterized (1). Given that neurons have an exceedingly limited capacity for renewal, understanding neuronal metabolic responses to environmental, nutritional and pharmacological interventions is made all the more important. Determining the basic metabolic biochemistry of the nervous system has the potential to affect translational medicine directly.
Experimental manipulation of fatty acid metabolism in the brain has led to some of the most surprising recent work in neurometabolism. Genetic or pharmacological manipulation of brain fatty acid and lipoprotein metabolism causes dramatic changes in energy balance-related behavior and physiology (2 – 5). Although these are important foundational experiments, they highlight our need to understand more fully lipid metabolism in the nervous system. Predicting the results from these experiments would have been difficult or impossible given our current understanding of the interaction between the brain and circulating or de novo produced lipids. Do a subset of neurons or glia require exogenous fatty acids to sense and respond to dietary cues? Surprisingly, some of the basic dogmas in neurometabolism are not based on strong direct experimental evidence, which hampers our ability to build detailed biochemical models. Biochemistry textbooks state “fatty acids do not serve as fuel for the brain,” (6) but what is meant by “brain” is quite subjective, as some cell types in the brain can use fatty acids for fuel. Also, does a subset of neurons or glia require fatty acid beta-oxidation? To date, experimental data cannot stringently answer the question, largely due to the lack of tools.
The lack of strong metabolic data for the mammalian brain is mainly because of significant experimental challenges to metabolic biochemists that require innovative new methodology to overcome. A formidable obstacle to understanding neurometabolism is the heterogeneous nature of cells in the brain coupled with the diversity of neurons themselves. Heterogeneity is a significant confounder for even the most advanced targeted or lipidomic analysis, and, unfortunately, sorting or culturing cells irrevocably alters their metabolism. Genetically encoded metabolite sensors, based largely on fluorescence resonance energy transfer reporters, are useful in revealing cellular and subcellular metabolite changes in situ. Several mammalian reporters, which are adapted largely from bacterial proteins, now exist for various metabolites (7 – 9). However, there are no genetically encoded biosensors for lipid metabolites, and FRET reporters may not be ideal for analyzing brain metabolism in vivo due to their limited dynamic range. For questions of tissue heterogeneity, it would be more advantageous to couple lipid metabolite concentrations to a more direct and dynamic measurement of reporter activity (e.g., short-lived GFP fluorescence).
The challenges of studying brain lipid metabolism are compounded by the relative lack of experimental tools and by the use of inadequate and often nonspecific pharmacologic inhibitors. Although knockout mice are invaluable for determining the requirements of enzymes in vivo, they are not able to tell the whole story. Ideally, one would combine the quick kinetics, dose responsiveness and reversibility of small-molecule pharmacology with the specificity of targeted knockouts. There has been considerable progress in the development of small stabilizing or destabilizing protein domains that interact with well-defined inert small molecules (10, 11). To manipulate fatty-acid metabolism acutely in vivo, we combined small-molecule inducible protein stabilization with genetically tractable recombination-mediated transgene expression (12). This technique allowed us to manipulate fatty-acid metabolism in a tissue-specific, dose-dependent and reversible manner in live mice. Since the small molecule interacts only with a user-engineered protein, wildtype mice can be used to control for off-target effects. In this way, one can annotate the function of metabolic pathways in a cell-specific manner in vivo while mitigating changes in compensatory pathways.
Despite significant challenges, the study of lipid metabolism in the nervous system is an area ripe for discovery. Combining molecular genetics and biochemistry, we can answer some fundamental mechanistic questions that are relevant for human health and disease. The specialized nature of the nervous system suggests that there are sure to be many unique and surprising roles for lipid metabolism that have yet to be uncovered.
- 1. Wolfgang, M.J. et al. Proc. Natl. Acad. Sci. USA 103, 7282 – 7287 (2006).
- 2. Loftus, T.M. et al. Science 288, 2379 – 2381 (2000).
- 3. Chakravarthy, M. V. et al. J. Clin. Invest. 117, 2539 – 2552 (2007).
- 4. He, W. et al. Nat. Neurosci. 9, 227 – 233 (2006).
- 5. Wang, H. et al. Cell Metabolism 13, 105 – 113 (2011).
- 6. Berg, J.M. et al. Biochemistry, 4th ed., W.H. Freeman and Company, New York (1995).
- 7. Okumoto, S. et al. Proc. Natl. Acad. Sci. USA 102, 8740 – 8745 (2005).
- 8. Berg, J. et al. Nat. Methods 6, 161 – 166 (2009).
- 9. Ewald, J.C. et al. PLoS One 6, e28245 (2011).
- 10. Banaszynski, L.A. et al. Cell 126, 995 – 1004 (2006).
- 11. Bonger, K.M. et al. Nat. Chem. Biol. 7, 531 – 537 (2011).
- 12. Rodriguez, S., and Wolfgang, M.J. Chem. Biol. 19, 391 – 398 (2012).
Jessica M. Ellis (firstname.lastname@example.org) is a postdoctoral fellow in the department of biological chemistry at the Johns Hopkins University School of Medicine. Michael J. Wolfgang (email@example.com) is an assistant professor in the department of biological chemistry at the Johns Hopkins University School of Medicine.