|Image courtesy of Wikipedia.
Coenzyme A is an essential, universally distributed, thiol-containing cofactor that works as the major acyl group carrier in all cells. This molecule is involved in hundreds of reactions and is required for the metabolism of fatty acids, carbohydrates, amino acids and ketone bodies.
CoA is a major regulator of energy metabolism, although it often is overlooked. Acetyl-CoA in particular is strategically positioned at the crossroads of energy metabolism. Just like all the roads lead to Rome, both anabolic and catabolic pathways converge at the formation of this small molecule, yet acetyl-CoA maintains order by reinforcing the partition of pyruvate between synthesis and degradation through its differential regulation of pyruvate dehydrogenase and carboxylase. Traffic control beyond this metabolic junction is exerted by acetyl- and other acyl-CoAs through both allosteric and post-translational regulation.
Several acyl-CoAs produced as metabolic intermediates are potent allosteric modulators of key enzymes, such as carnitine palmitoyltransferase I and acetyl-CoA carboxylase, and transcription factors, such as HNF4-α (1) and PPARα (2). Acetyl-CoA is used to modify enzymes, transcription factors and chromatin covalently and reversibly to govern their activities (3, 4, 5). Covalent acylation by long-chain acyl-CoAs directs proteins to membranes where substrates are activated and stimulate cell growth and proliferation in cancer (6).
These ingenious mechanisms coordinate the expression and activity of a multitude of enzymes and processes with the energy state of the cell. Thus, CoA and a few other small molecules like NAD+ and ATP can act as global regulators of cellular metabolism both together with and independent from the action of key transcription factors.
Consistent with these key functions, CoA levels are flexible in cells so that the available supply is sufficiently adaptive to metabolic challenge. But at the same time, CoA levels are maintained at threshold amounts, suggesting that an oversupply could be detrimental to function. Decades of studies have established that regulation of the CoA biosynthetic pathway occurs at the initial step catalyzed by pantothenate kinase (PanK) in bacteria and eukaryotes (7, 8).
Mammals possess four closely related PanK isoforms, PanK1α, 1β, 2 and 3, and these enzymes are regulated through feedback inhibition by CoA species and through activation by long-chain acylcarnitines and acylethanolamides. Not all the PanK isoforms are equally responsive to this allosteric regulation, and PanK2 and 3 are significantly more sensitive than PanK1α and 1β. The localization of the PanK isoforms in different subcellular compartments and their tissue-selective distribution profiles are additional features that provide combinatorial control over CoA levels in distinct cell types.
But why is there so much redundancy, and why are there so many variations on the same theme? We recently have started to get some answers from the generation of mice that lack one or more PanKs.
The single Pank1, Pank2 and Pank3 knockout mice are viable and overtly normal, with the exception of the Pank1 knockout mice that exhibit a clear metabolic phenotype. Deletion of any two Pank genes leads to either embryonic lethality (Pank1/3 and Pank2/3) or death before weaning age (Pank1/2), indicating that the isoforms can compensate for each other and that redundancy is necessary for life. The combination of isoform abundance and regulatory properties roughly correlates with the total amount of CoA in tissues and organs, so that tissues where PanK1α or 1β are most abundant (liver, heart, kidney) have higher CoA levels than tissues where PanK2 or 3 predominate (brain, skeletal muscle).
Finally, the particular localization of each PanK isoform in the cytosol, mitochondrion or nucleus may enable the response to ligands that govern activity and flux through the CoA biosynthetic pathway. The PanKs may be sensors in situ that respond to fluctuations in the local concentration of acetyl-CoA, acyl-carnitine or acylethanolamide and adjust the rate of CoA biosynthesis.
The recent characterization of mice with complete chemical inhibition of all the PanKs (9) and of mice lacking Pank1 alone or in combination with Pank2 (10, 11) has established clearly the connection between PanK expression → CoA levels → metabolism. This represents an important starting point to try and understand the complex pathology of PKAN (pantothenate kinase-associated neurodegeneration), a severe neurological disorder caused by mutations in the human PANK2 gene. The majority of these mutations are expected to decrease significantly or abolish PanK2 activity, thus suggesting that lower CoA could be the underlying cause of reduced neuronal metabolism and function in PKAN patients.
Unfortunately, Pank2 knockout mice do not reproduce the human disease, and an important future challenge will be to generate a mouse model to investigate the connection between CoA levels and neurodegeneration and, above all, to accelerate the identification of a treatment for the disease.
Given the central role of CoA in the regulation of metabolism, another important question to address will be whether metabolic diseases like diabetes are associated with dysregulated tissue CoA levels and what the importance of CoA-degrading enzymes is in the regulation of this cofactor. Clearly, research thus far has shown that cofactors such as CoA, ATP and NAD+ can limit the output of a pathway in a manner similar to reduced enzyme levels.
Perhaps CoA is regulated to prevent overactivity within a pathway, and the future research challenge will be to establish the hierarchy among those biological processes that require CoA. CoA is required for hundreds of reactions and regulates metabolism at several different levels that include 1) substrate delivery for enzymatic reactions, 2) allosteric and post-translational regulation of enzymatic activity, and 3) regulation of gene expression through reversible acetylation of histones and transcription factors.
So keep CoA in mind next time you see a metabolic phenotype: It might just happen that a pharmacological organic acid is activated by this cofactor, thereby reducing the effective concentration of CoA for normal cellular and biochemical functions.
- 1. ↵ Bogan, A.A. et al. J. Mol. Biol. 302, 831 – 851 (2000).
- 2. ↵ Schroeder, F. et al. Lipids 40, 559 – 568 (2005).
- 3. ↵ Cai, L. et al. Mol. Cell 42, 426 – 437 (2011).
- 4. ↵ Lundby, A. et al. Cell Rep. 2, 419 – 431 (2012).
- 5. ↵ Siudeja, K. et al. EMBO Mol. Med. 3, 755 – 766 (2011).
- 6. ↵ Triola, G. et al. ACS Chem. Biol. 7, 87 – 99 (2012).
- 7. ↵ Rock, C.O. et al. J. Biol. Chem. 275, 1377 – 1383 (2000).
- 8. ↵ Vallari, D. S. et al. J. Biol. Chem. 262, 2468 – 2471 (1987).
- 9. ↵ Zhang, Y.M. et al. Chem. Biol. 14, 291 – 302 (2007).
- 10. ↵ Garcia, M. et al. PLoS ONE 7, e40871 (2012).
- 11. ↵ Leonardi, R. et al. PLoS ONE 5, e11107 (2010).
Roberta Leonardi (firstname.lastname@example.org) is a scientific laboratory manager in the Department of Infectious Diseases at St. Jude Children’s Research Hospital. Later this month, she will be an assistant professor at West Virginia University. Suzanne Jackowski (email@example.com) is a faculty member at St. Jude Children’s Research Hospital and executive editor of BBA-Molecular and Cell Biology of Lipids.