How can 'opposite' genetic mutations have the same effect?

December 23, 2005 10:05 AM

MolecularBiologyFilter: How can knocking out one copy of a gene have the same effect as a mutation that leads to increased activity?

I hope this isn't too specific or obscure. But anyway -- there's this gene, a particular receptor-associated kinase. A particular point mutation results in increased activity of the translated protein. A nonsense mutation that should effectively knock out one allele results in the same phenotype as the activating mutation. Both mutations are dominant. How would you explain this?

posted by greatgefilte to science & nature (17 comments total)

Cell signaling is complicated: lots of crosstalk and feedback.
How many targets get phosphorylated by the kinase (rhetorical question)? I assume the increased activity is being measured in vitro against a particular target? Also, how sure are you that the nonsense mutation is doing what you think it's doing?
I can imagine a scenario wherein the cell/organism has some sort of compensation mechanism explaining the phenotype.
posted by exogenous at 10:17 AM on December 23, 2005

I can imagine a scenario wherein the cell/organism has some sort of compensation mechanism explaining the phenotype.

Yup, that's what I'm trying to figure out. :)

I should probably also mention that the activating mutation is in humans, and the nonense one in a mouse.
posted by greatgefilte at 10:30 AM on December 23, 2005

Do you know if the activity of corresponding phosphatases has changed?
posted by gaspode at 10:32 AM on December 23, 2005

I should probably also mention that the activating mutation is in humans, and the nonense one in a mouse.

I would be surprised if this alone is responsible for the phenomenon.

Further to gaspode's comment, I suggest learning what you can about the downstream signals from your kinase and how they relate to the phenotype you see, and developing some possible hypotheses. Depending on how interested you are, you can then run some experiments to test them. Of course, science being what it is, you might not find the answer you're looking for, but you can at least collect some data :-)
posted by exogenous at 10:45 AM on December 23, 2005

Here's a strict biochemical hypothesis:

Perhaps the kinase is able to phosphorylate itself (or other members of its family), and the phosphorylated kinase is less active, reducing its own activity in a form of negative feedback. The activating mutation renders the protein insensible to the effects of phosphorylation; knockout of one allele causes a reduced gene dosage, rendering any given kinase less likely to be phosphorylated at any given time.

Could this be possible? Is there a handy serine or tyrosine nearby (primary or tertiary structure-wise) the activating mutation site?
posted by ikkyu2 at 10:57 AM on December 23, 2005

ikkyu2 -- Good hypothesis, but it's been shown that the kinase needs to be phosphorylated in order to be active. I think some kind of interaction with the corresponding phosphatase just might be it...
posted by greatgefilte at 11:07 AM on December 23, 2005

Is the phenotype defined as increased activity of the enzyme on a molecule-by-molecule basis, as just overall enzymatic activity, or as something else?
posted by shoos at 12:43 PM on December 23, 2005

shoos -- the phenotype is an increased level of a particular circulating cell type.
posted by greatgefilte at 12:57 PM on December 23, 2005

So can the situation be rephrased this way?:

A kinase gene mutation in humans that results in increased activity of the kinase causes the number of a particular cell type to go up...

and

...introduction of a nonsense mutation in one allele of the murine homolog of that gene also causes the number of the same cell type to go up.
posted by shoos at 1:23 PM on December 23, 2005

Yup.
posted by greatgefilte at 7:16 PM on December 23, 2005

Why are you not naming names here, gg? What kinase are we talking about, what cell type?
posted by sennoma at 8:08 PM on December 23, 2005

sennoma -- I'm just a mere lab tech for a genotyping service, and it's quite possible that I'm not supposed to reveal other researchers' findings before they do so themselves. I was just curious about this particular situation, and I'm trying to play it safe by not naming names.
posted by greatgefilte at 8:52 PM on December 23, 2005

Gg -- fair enough. (I was, I admit, wondering if you were one of those "no one must ever have the chance to STEAL my WONDERFUL ideas" types, of whom the bioresearch community seems to have an abundance.) But playing it safe with someone else's work is the ethical thing to do.

(As regards your question, I got nothing I'm afraid. Lots of the guesses upthread seem plausible to me.)
posted by sennoma at 9:26 PM on December 23, 2005

Yeah, there are a very large number of possible explanations.

Here's one simple (and probably wrong) one: the nonsense mutation produces a truncated kinase that is constitutively active. Maybe the sequence c-terminal to the introduced stop codon contains the regulatory region, and without it, the kinase is always active in the mice.
posted by shoos at 12:19 AM on December 24, 2005

shoos -- Yeah, if only it were that easy! I'm working from the assumption that the nonsense mRNA gets degraded, especially since they did do RT-PCR and only found one transcript.
posted by greatgefilte at 10:22 AM on December 24, 2005

Maybe their primers were in the part of the transcript encoding that regulatory region, so they couldn't detect even if it were there! Where was the nonsense mutation placed?
posted by shoos at 10:44 AM on December 24, 2005

Ah, never mind that, I just got up.
posted by shoos at 10:45 AM on December 24, 2005