PI3K-Like Protein Kinases
What is the DNA-dependent protein kinase, DNA-PK?
Dr Susan Lees-Miller - University of Calgary, Calgary, Canada
The DNA-dependent protein kinase is composed of two parts, a DNA binding subunit that’s the Ku70/Ku80 subunit and the catalytic subunit that we call DNA-PKCS, that’s DNA dependent protein kinase catalytic subunit. It functions in the repair of radiation induced DNA double strand breaks through the non-homologous end-joining pathway. What we think it does there is it detects the breaks and then undergoes an autophosphorylation dependent reaction to then pass the ends of the DNA on to end processing enzymes that then remove non-ligatable endgroups and then the DNA ligase 4 complex religates the ends together. DNA-PKCS also functions in VDJ recombination where it promotes the endonuclease activity of artemis, it’s a nuclease that opens DNA hairpins, and also there’s recent evidence that it’s involved in other processes such as transcription and mitosis.
Is the name slightly misleading?
I think the name has served it well for the last 25 years. I think that in most cases it does act as a DNA dependent protein kinase. There are some exceptions, certainly in vitro we’ve shown that by changing the metal ion from magnesium to manganese you can alleviate the need for Ku and its requirement for DNA. There are also experiments from Kathy Meek showing that if you immobilise DNA-PKCS with an antibody that can lead to its activation in the absence of Ku. So maybe things that cause a conformational change in DNA-PKCS can activate it in the absence of DNA. But for the most part I think it is still most of the time a DNA activated protein kinase, maybe with these new roles that have been uncovered in recent years, maybe that will change. But at the moment the name still stands.
Given its abundance in human cells, what are the implications?
Yes, the abundance, that’s a very interesting question. It’s been estimated there are maybe hundreds of thousands of molecules of DNA-PKCS in the nucleus of a human cell line. It’s interesting that there are a lot less in mouse cells and it’s not detected or it’s not found in C. elegans or drosophila or in yeast. So certainly its function is associated more with vertebrate cells. The fact that there is so much of the protein there does suggest that it’s doing more than responding just to DNA double strand breaks that occur endogenously and so maybe the abundance is related to its roles in transcription or mitosis or maybe other roles that we haven’t uncovered yet.
Do PI3 kinases work independently of each other?
No, I think that the different family members interact a lot more than people originally realised. Certainly in the DNA damage response we’ve found that ATM can phosphorylate a number of proteins in the non-homologous end-joining pathway and our ongoing work in mitosis shows that there’s definitely co-operativity between DNA-PK and ATM. I think one of the things that has come out of this meeting is that there is a lot of cross-talk between the different family members.
What are the therapeutic implications?
One of the areas that we’re exploring at the moment is the use of ATM inhibitors and PARP inhibitors which we’ve found are specifically toxic in cells with mutant p53. Because many human cancers have mutation or deletion in p53 this might be another avenue to target those cancers.
