Need help from the experts in kinases please! - (Jul/09/2006 )

1) Can a phosphorylation mutant (i.e. pseudophosphorylation mutant, exchanging Serine for glutamate) of a protein kinase able to phosphorylate other proteins??
Then, if it can not phosphorylate other proteins, how a constitutively active kinase able to activate a signal cascade (i.e. MAPK pathway) ?

2) Does anybody knows of any protein that can phosphorylate other proteins without having a kinase domain, but just the ability to autophosphorylate in vitro??

Please, anybody in the kinase field help me!!!

Thank you very much

I'm sorry, I think I don't understand what you want.
If your kinase is still able to phosphorylate a protein, while you mutated the kinase domain, then it means you didn't mutate the kinase domain ! or the mutation was not deleterious for the function of the kinase domain.
even for autophosphorylation, you need a functional kinase domain.

Mutating Serine to Glutamate will mimic a phosphorylation event. So it depends whether the phosphorylation of that residue causes activation or deactivation of the kinase. I think most kinases are activated by phosphorylation so it should produce a constitutively activated mutant of the kinase. If you combine it with specific mutations in the kinase domain you could produce a substrate trapping mutant which would be a dominant negative.

Ceri

What I meant to say was if a pseudophosphorylated kinase which mimics phosphorylation can phosphorylate other proteins, what I thought about kinases was that they can phosphorylate a substrate by "passing" their phosphate group to the substrate ( in this case the kinase gets dephosphorylated); but a pseudo phosphosylated form can not pass on the phosphate group to phosphorylate a substrate. I dont know if this is not a correct assumption.

In # 2)
I am just trying to analyze the possibility of a heat shock protein (that is not a kinase, but can autophosphorylate) to phosphorylate another protein. There is some evidence that over expression of this hsp activates the MAPK pathway. I have to add that this protein does not have a kinase domain.

thanks for the reply

So in order to activate another protein, the pseudophosphorylated kinase does not need to transfer the phosphate group, but then, how does it phosphorylate its substrate?
unless Im mixing togather active site with kinase domain of the protein.

some kinases like PKA take phosphate from ATP

I'm not really a kinase expert but this is what I've picked up while reading around the subject a while back.

The general layout of a kinase is that you have a Glycine-rich loop which is involved in the binding of the ATP for the phospho-transfer process. Then there is a catalytic loop which usually contains a important lysine residue involved in kinases by anchoring ATP and contributes to the correct orientation of the triphosphate and the γ-phosphate for catalysis. Mutation of this lysine residue nearly always results in a 'dead' mutant which has no phosphorylating activity.

Most kinases can autophosphorylate themselves at different residues, and can either be activated upon phosphorylation or inactivated - though the former is more common. Mutation of these activation sites to give the impression of it being phosphorylated generally leads to constituitive activation of the kinase.

Normally a kinase works by taking a phosphate group, either from ATP or another protein and transfering that onto its target substrate, not sure if they will take their own phosphate group for transfer as this will result in activation of itself??

I've not come across a protein that does not have a kinase domain/catalytic loop and can autophosphorylate itself...

With regards to your pseudophosphorylated form not being able to pass on to the substrate - that will depend on whether the residue changed results in a constituitively active or inactive protein...

I've also not come across HSPs phosphorylating other proteins. It could be the fact that your overexpressing the hsp and that it somehow has a binding site or recognition site that results in the activation of the MAPK pathway?? How did you identify that it has autophosphorylating activity? Did you run an in vitro assay with purified HSP?

Hope this info helps!

On another note...

Ceri
You talked about a kinase substrate-trapping mutant, is there such a thing? Do you know if it is the result of mutations to certain residues within the kinase domain that are conserved throughout kinases? I'd be interested in your answer!

1) Can a phosphorylation mutant (i.e. pseudophosphorylation mutant, exchanging Serine for glutamate) of a protein kinase able to phosphorylate other proteins??
Then, if it can not phosphorylate other proteins, how a constitutively active kinase able to activate a signal cascade (i.e. MAPK pathway) ?

2) Does anybody knows of any protein that can phosphorylate other proteins without having a kinase domain, but just the ability to autophosphorylate in vitro??

Please, anybody in the kinase field help me!!!

Thank you very much

Thanks a lot for all that info, it was really helpfull. Actually, I know that the hsp has autokinase activity bc it was published already, that was the thing. I also agree with you, it should be binding to some protein to induce activation of MAPK.

thanks for your comments

I'm not really a kinase expert but this is what I've picked up while reading around the subject a while back.

The general layout of a kinase is that you have a Glycine-rich loop which is involved in the binding of the ATP for the phospho-transfer process. Then there is a catalytic loop which usually contains a important lysine residue involved in kinases by anchoring ATP and contributes to the correct orientation of the triphosphate and the γ-phosphate for catalysis. Mutation of this lysine residue nearly always results in a 'dead' mutant which has no phosphorylating activity.

Most kinases can autophosphorylate themselves at different residues, and can either be activated upon phosphorylation or inactivated - though the former is more common. Mutation of these activation sites to give the impression of it being phosphorylated generally leads to constituitive activation of the kinase.

Normally a kinase works by taking a phosphate group, either from ATP or another protein and transfering that onto its target substrate, not sure if they will take their own phosphate group for transfer as this will result in activation of itself??

I've not come across a protein that does not have a kinase domain/catalytic loop and can autophosphorylate itself...

With regards to your pseudophosphorylated form not being able to pass on to the substrate - that will depend on whether the residue changed results in a constituitively active or inactive protein...

I've also not come across HSPs phosphorylating other proteins. It could be the fact that your overexpressing the hsp and that it somehow has a binding site or recognition site that results in the activation of the MAPK pathway?? How did you identify that it has autophosphorylating activity? Did you run an in vitro assay with purified HSP?

Hope this info helps!

On another note...

Ceri
You talked about a kinase substrate-trapping mutant, is there such a thing? Do you know if it is the result of mutations to certain residues within the kinase domain that are conserved throughout kinases? I'd be interested in your answer!

Kinases transfer the terminal phosphate from ATP to their substrates. If a kinase is active when phosphorylated or pseudophosphorylated, it still uses ATP. This is true for all kinases. They do not transfer phosphates from their own phosphorylated residues - if they did they would thereby become inactive, and such things as constitutively active kinases would not exist. The process of phosphorylation will result in conformational changes within the kinase, allowing it to bind ATP and/or its substrates, in the case of a kinase that becomes active when phosphorylated. Kinases like GSK3B and Src are inactive when phosphorylated.

HSP is most likely acting as a cofactor, as someone suggested. There are examples of kinases that are stabilized by HSPs and as a result have increased activity. And with regards to autophosphoylation, there are two things which need to be distinguished. Most kinases can autophosphorylate, and by this I mean that if we have two identical kinase molecules, A and B, A can phosphorylate B and vice versa. But A cannot phosphorylate itself. So a kinase cannot take the phosphate from ATP and attach it directly to itself. Do you know what your HSP is doing? If one HSP molecule is phosphorylating another, then by definition it is a kinase, whether or not it has a kinase domain and whether or not it has any other substrates. However, I suppose a plausible alternative is that one HSP molecule attaches an ATP-derived phosphate directly to itself as opposed to another molecule. This would not make it a kinase.

If you do a kinase assay with HSP and myelin basic protein as a substrate (which all kinases can phosphorylate) you could distinguish between these possibilities.

One last point. Just because it does not have a kinase domain in its sequence, does not mean it does not have a kinase domain. There are many examples of kinases that showed no homology to other kinases at the level of sequence, but when their structures were determined they clearly shared structural homology to other kinases. And structural homology is the deciding point.