Recently, I want to knockdown the expression of one gene with siRNA. There are four siRNA sequence targeting the gene. They can silence target gene expression by at least 75% at the mRNA level, but there is no change at the protein level. I wonder the reason for these discrepancy. Can someone help me ?
Perhaps you could tell us how you've measured your mRNA levels and your protein levels?
There are a few possibilities I can come up off the top of my head:
Perhaps you could tell us how you've measured your mRNA levels and your protein levels?
There are a few possibilities I can come up off the top of my head:
Hi Shu,
I would agree with Petra almost totally. Except for point 3, which she also doubts, and 4, where if it were true you would actually see an increase in your target mRNA or no change, if the whole transcriptome were affected, the other observations are fundamental ones: still, GOOD that you checked the protein level of your targeted gene!! That's the most important point, and people tend to forget that proteins have a katabolism of their own which needs to be investigated when knocking down the mRNA.
And WB is a tricky tool: don't forget the opposite situation not mentioned by Petra, if your protein is very highly expressed and you load too much protein extract on gel, you can saturate the membrane (particularly if you are using PVDF), so that what you transfer to the membrane is not anymore proportional to what is in the extract. It's typical with the classical proteins used for calibration, actin, alpha-tubulin, GAPDH: if you load 30-40 ug of extract, of course the calibration looks all so nice and even, you saturate the membrane capacity for these proteins with far less extract.... So even with a digital acquisition system that has a much longer linear range than a film you won't be quantifying correctly your extracts.
Pietro makes good points (and he's right about 4 - it should have been increase).
Western blot is often miss-used for quantitative protein determination when you really should be using it only for qualitative statements with >25% difference. There are several ways of making it more accurate (loading dilutions of your samples, measuring and making sure there's a linear relation between the diluted samples; or the newer fluorescent antibodies with a proper digital system and the dilutions...) but really, why put yourself through the pain when ELISA is much better in that way?
If there's no off-the-shelf ELISA assay or antibody available then I'd go for something like this:
https://www.aptagen.com/aptamer-technologies#apta-beacons-and-apta-sensors
Dear Pietro and Petra，thank you very much! I blasted the 4 siRNA sequence in NCBI . I found only one siRNA sequence could target all isoform of the protein ( isoform alfa and beta), the other three sequence only target one isoform(isoform alfa). In addition, the protein I detect is mainly isoform beta which is abundant in the tissue .I would repeat my experiment .
I have another question. I found the target protein band is different in differernt protein sample. In tissue protein, the protein band is among 40-55 KD, but in cell protein, the protein band is about 55kD. As to the primary antibody is polyclonal, I have diffcult to be sure the position of target protein.
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science
- Reasoning