siRNA efficiency

8 posts / 0 new
Last post
Guy Sovak
Guy Sovak's picture
siRNA efficiency

Dear All,
For quite a long time I am doing siRNA with different genes.
Lately I found out an interestin phenomena.
I am looking for the effect of kd of gene A on protein B. It seems that depending upon the amount of kd that I am getting the effect on protein B is changing. Reported to this effect is when you kd A B is increaing. I found that if the kd is ~40-50% I am getting an increase in B but if the kd is lower or higher 10-30% or 80-90% the effect is the opposite.
It is quite surprising.
Did any one got something similar?
Looking forward for a disscution in this topic.
Guy

Jon Moulton
Jon Moulton's picture
Do you expect any direct

Do you expect any direct regulatory interactions between A and B (e.g. A represses B)?

It is known that knockdown of a given sequence by siRNA will modulate the expression of up to hundreds of off-target genes (Scacheri et al, below).

It is likely that in some cases off-target gene modulation is due to direct off-target expression suppression by the siRNA, while in other cases gene expression modulation would be expected as regulatory consequences of the direct knockdowns; that is, you target gene A, which represses expression of gene B through off-target siRNA interaction with its mRNA, and as a consequence gene C might be upregulated in response to the suppression of B (in this case a regulatory effect rather than a direct siRNA effect). To restate this, even though A and C have no direct regulatory link, C can still vary as a consequence of regulatory control in response to off-target knockdown of B.

Here are some citations of papers addressing the off-target gene modulation of siRNA:

Comparison of siRNA-induced off-target RNA and protein effects. Aleman LM, Doench J, Sharp PA. RNA. 2007 Jan 19; [Epub ahead of print]

3' UTR seed matches, but not overall identity, are associated with RNAi off-targets. Birmingham A, Anderson EM, Reynolds A, Ilsley-Tyree D, Leake D, Fedorov Y, Baskerville S, Maksimova E, Robinson K, Karpilow J, Marshall WS, Khvorova A. Nat Methods. 2006 Mar;3(3):199-204.

Nonspecific, concentration-dependant stimulation and repression of mammalian gene expression by small interfering RNAs (siRNAs). Persengiev SP, Zhu X and Green M. RNA 2004; 10:12-18.

Short interfering RNAs can induce unexpected and divergent changes in the levels of untargeted proteins in mammalian cells. Scacheri PC, Rozenblatt-Rosen O, Caplen NJ, Wolfsberg TG, Umayam L, Lee JC, Hughes CM, Shanmugam KS, Bhattacharjee A, Meyerson M, Collins FS. Proc Natl Acad Sci U S A. 2004 Feb 17;101(7):1892-7. Epub 2004 Feb 09.

Guy Sovak
Guy Sovak's picture
Thank you for your response,

Thank you for your response,
To be more clear, the siRNA is of Aha1 which is an Hsp90 cochaperone. It was reported by Wang, X. et al. it Cell paper (127) that siRNA of Aha1 increases complex glycosilated CFTR.
The fact is known. Hsp90 is in complex with CFTR at the ER and Aha1 regulates this interaction.
I wanted to look at this phenomena with other cell types and found out what I described already.
The interesting thing is that under different KD eficiency the result was deferent and more surprising the opposite.
This is what I would like to discuss.
Did you ever had the same phenomena with other siRNA genes?
I got many ideas why but as a scientific tool we always try to get more then 75% KD, now from what I have found it is maybe not the right way. Or maybe we find out just half of the whole story.
Guy

Jon Moulton
Jon Moulton's picture
Hi Guy,

Hi Guy,

That's an interesting case. Obviously you have a complex system, and I will try to restate some of what you have written; please let me know if I am off-track. Perhaps I am making this more complex than it really is.

Aha1 can be knocked down with siRNA.

Aha1 mediates interaction of CFTR with Hsp90.

When Aha1 is knocked down to either 10-30% or 80-90% of the untreated concentration, there is less CFTR associated with Hsp90 than in untreated cells.

When Aha1 is 40-50% knocked down there is more CFTR associated with Hsp90 than in untreated cells.

Did I understand correctly that it is the complex of CFTR and Hsp90 that you referred to a "B" in your original post?

Guy Sovak
Guy Sovak's picture
Most of the information you

Most of the information you got correct.
In the window of 40-50% KD The amount of Aha1 are optimum for the stabilisation of the interaction between CFTR and Hsp90, Thus enables Hsp90 to interact with CFTR and maintain its folding condition and therefore CFTR it getting to the Plasma Membrane. This was reported in the Cell paper I mentioned.
In the other cases ouside of the 40-50% KD window the effect is opposite. So either to much Aha1 or to Less Aha1 is in contact with Hsp90 and thus destabilising its complex with CFTR.
Hope now thing are more clear.Did you encounter something like that.

Some how I am afraid that many scientist are not reporting evrything they get just what they falls into there theory. I am not impling anything just upon my point expireance which is not so long in siRNA.
Guy

Jon Moulton
Jon Moulton's picture
Hi Guy,

Hi Guy,

I want to see this as a purely biochem-level effect rather than an siRNA artifact, but I am having some trouble coming up with a convincing story. First came my hypothesis that at low expression there isn't enough Aha1 to stabilize the interaction, at mid-level expression the stabilization is just right and at high-level Aha1 expression the other two components (CFTR and Hsa90) are saturated with Aha1 and so rather than a single Aha1 bridging the other two to form the complex, the overabundant Aha1 saturates each of the otehr components (CFTR gets its own Aha1, Hsa90 gets a different Aha1).

It seems a nice package but it doesn't account for the untreated case. Given three levels of membrane complex formation (1 = low, 2 = mid, 3 = high) then is this true?

untreated = 2
slight knockdown = 1
moderate knockdown = 3
strong knockdown = 1

If so, my initial hypothesis can account for the pattern of the knockdowns but not for the complex formation in untreated cells, which I would expect the be 1 or less given a complex formation optimum at the level of a moderate knockdown.

So far I am stumped.

Guy Sovak
Guy Sovak's picture
So am I,

So am I,
I can not understand it, my only explanation is that the amount of Aha1 is very important to the Hsp90 Client stabilization more or less is destabilizing.
But the important Q is wether someone else saw something lie this in other siRNA experiments?
Guy

Jon Moulton
Jon Moulton's picture
I've no information to offer

I've no information to offer about siRNA effects, I use Morpholinos. Anybody else?