DGK is not the only DGK isoform that
DGKε is not the only DGK isoform that has been associated with p53. It has been shown that DGKζ binds to p53 and modulates its activity in both the cytoplasm and the nucleus . In the cytosol DGKζ promotes the degradation of p53 through the ubiquitin-proteosome system , also a likely mechanism to explain the low levels of p53 in the DGKε-WT cells. It is also reported that DGKζ-deficient Wiskostatin exhibits a high level of p53 protein , analogous to what we have shown with the DGKε−/− MEFs (Fig. 5). In addition, DGKζ can translocate to the nucleus. Inhibition of DGKζ expression in the nucleus results in the specific downregulation of the transcriptional activity of p53 . Another DGK isoform, DGKα is activated by p53 . Thus, loss of DGKε or DGKζ would result in an increase in p53, which in turn would cause the activation of DGKα. Higher activity of DGKα results in better survival rates of lung cancer patients .
Thus, we conclude that the observed increased expression and activity of GK in DGKε−/− MEFs provides a mechanism to explain the increased incorporation of glycerol into lipid that we had found in these cells. The increased GK levels in these cells can be explained by the increase in p53, a known transcription factor for the expression of this enzyme. A scheme describing these relationships is shown in Fig. 7. p53 is closely associated with cell survival and has an important role in cancer. Further studies are needed to fully understand how various isoforms of diacylglycerol kinase affect p53 signaling.
Conflict of interest
Acknowledgements We thank Dr. Fred Y. Xu for technical support. Supported by grants from the Natural Sciences and Engineering Research Council of Canada (Grant 9848, to RME) and from the Heart and Stroke Foundation of Canada (Grant G-14-0005708 to G.M.H.). G.M.H. is a Canada Research Chair in Molecular Cardiolipin Metabolism.
Solutions and Buffers
DGKθ Activity Assay The following protocol is broken down into sections to facilitate planning. The complete assay can be done in a single day once tubes are silanized and stock buffers are prepared. Where possible we have indicated useful stopping points. The protocol for purifying DGKθ can be found elsewhere (Tu-Sekine & Raben, 2012).
A surface dilution experiment not only provides data on the kinetic values of an enzyme, but also reveals information about its processivity, or scooting behavior. For example, enzymes like PLA2 are insensitive to changes in bulk DMPM, a lipid to which it binds very tightly, while a quasi-scooting enzyme like DGKθ shows a clear dependence on both surface and bulk concentrations (see Fig. 2). While this experiment may at first appear somewhat overwhelming due to the number of samples, the amount of data obtained is well worth the effort. It is extremely useful for comparative studies, such as determining the effect of point mutations or lipid composition on the affinity of the enzyme for the vesicle interface and substrate(s). Importantly, kinetic parameters can only be extracted from saturated kinetic curves, and the range of lipid has been chosen to satisfy this requirement for purified DGKθ. The following experiment covers a bulk range of lipid from 0.2 to 20mM, with surface concentrations of DOG ranging from 0% to 10%. In this experiment eight master sets of vesicles are made, each containing a different surface concentration of substrate, resulting in eight tubes of dried lipid. Once the vesicles are prepared by extrusion, each set of vesicles is sequentially diluted to produce a series of eight bulk concentrations, resulting in a total of 64 vesicle stock solutions containing sufficient volume for triplicate reactions.
Vesicle pulldown assays can be a simple and effective way to determine the binding affinity of an enzyme under specific conditions and are particularly useful when minimal amounts of purified enzyme are available for study. A binding curve can be readily constructed by serial diluting a single batch of sucrose-loaded vesicles. Spiking vesicles with a small percentage of a fluorescent lipid allows easy quantization of losses during centrifugation steps. In this experiment, DGKθ (up to 1:1 enzyme:vesicle at the highest vesicle concentration chosen) is incubated with each concentration of vesicles under standard assay conditions. The sucrose-loaded vesicles are recovered by high-speed centrifugation, and the amount of enzyme remaining in the supernatant is quantified by densitometry of silver stain or western blot images to construct a binding curve. Note that determining the amount of enzyme associated with vesicles is more direct, but high levels of lipid produce streaking and smearing during electrophoresis and can interfere with gel analysis. As always, the mass amount of enzyme necessary for detection should be confirmed in advance to guide experimental setup (e.g., 10–20ng purified DGKθ is required for silver stain, while 5–10ng is sufficient for western blot).