1
$\begingroup$

I'm currently working on a dataset were I'm trying to identify substrates of an kinase in silico, I have a dataset which contains the concentrations every 0h 2h,4h,6h,8h,10h,24h for a lot of proteins found in the sample when the kinase was active. However some of these aren't targeted by the kinase but other proteins that got activated or were indirect activated by the kinase. So what I did I used R to plot the concentration of the detected proteins only if they fitted the concentration line of the kinase of interest(with a little devition allowed) this gave me the following graph:
enter image description here the blue line is the kinase and the other lines are proteins found in the dataset. I know there could be postive or negative feedback which has effect on the concentation of these proteins.So my question is could the lines shown here be a possible substrate?

$\endgroup$
1
$\begingroup$

The concentration of a kinase is completely independent of the concentration of the substrate, and vice versa. In a typical kinase cascade (such as the Ras→Raf→MEK→Erk pathway), signal transduction is generally amplified with each successive step. A relative small population of one kinase (like Raf) phosphorylates orders of magnitude more of its substrates (like MEK1/2), which in turn phosphorylate orders of magnitude more p44/42 MAPK (Erk1/2) molecules. These then phosphorylate transcription factors like c-Myc which translocate to the nucleus and affect the expression of target genes. Over time, protein levels inside the cell are changed, including the levels of the various components of the pathway or their inhibitors (such as phosphatases, components of the protein degradation machinery like ubiquitin ligases, etc.), allowing for regulation of the signal.

Therefore, for any given kinase at any given time post-stimulation, such as with a growth factor, the concentration and phosphorylation state of its substrate(s) can vary tremendously. For example, in the classical NK-κB pathway, activated by TNFα, one of the main substrates of the IKKα/IKKβ kinase complex is the inhibitor IκBα, which upon phosphorylation is rapidly ubiquitinated and degraded by the proteasome. However, one of the first target genes of the activated NF-κB p65/p50 transcription factor dimer is Ikba itself, which negatively regulates activation of the pathway.

I don't know exactly what information your data set contains, so I can't comment too much on your results. However, just graphing substrate concentrations that correspond to kinase concentrations at various time points will only give you one piece of information - that some substrates have grossly similar concentrations to the kinase - which means nothing. All that is is confirmation bias - you've decided on your result, then picked the data that confirm it. I bet if you were to graph the concentrations of direct substrates along with your kinase, you would find no relationship whatsoever, because as I've described above, there is none biologically.

$\endgroup$
  • $\begingroup$ Thankyou for your explanation! I have a dataset which contains all proteins downstream of the kinase (so substrates and non substrates), but I found it quite remarkable that the pink line has almost exaclty the same concentration changes as the kinase has, maybe this could be coincidence? @MattDMo $\endgroup$ – KingBoomie Oct 21 '16 at 15:06
  • $\begingroup$ @RickBeeloo most likely it's just a coincidence. It could also be that your protein of interest is some sort of regulator of the kinase, and expression levels may correlate with it. You'll have to dig into the biology to answer that. $\endgroup$ – MattDMo Oct 21 '16 at 17:28
  • $\begingroup$ The hardest part is that there is nothing to find about the proteins of interest because they aren't identified before, but thankyou anyway! @MattDMo $\endgroup$ – KingBoomie Oct 22 '16 at 7:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.