I'm performing reverse transfection on SH-SY5Y using Viromer Green, and I get wonderful transfection efficiency (90+%). See picture (RFP labeled siRNA fluorescence overlapped on phase contrast image).
However, when I qPCR for KO I see no knockdown for any of my siRNAs. It's odd, because I had previously validated this protocol for two of my siRNAs and they worked fine (>50% KO with low standard errors, replicated across multiple biological replicates, at multiple time points, with a clear and significant dose response curve). So I know that at-least two are targeting my transcript fairly well.
For qPCR, I'm using TaqMan chemistry with best coverage primer probe sets. I'm extracting total RNA with Directzol (column-based RNA separation from Trizol reagent), and cDNA synthesis with MultiScribe.
Almost everything in my protocol is identical to what I performed in the validation. The only thing that is different is that I forgot a step in the RNA extraction process. In particular, Directzol indicates to homogenize cells in Trizol, add an equal volume of ethanol to the homogenate, run through column, DNase treatment on column, several ethanol washes, and elute in water. This time around, I forgot to add ethanol to the homogenate. Stupid move on my part, I know. Obviously, it precipitates the nucleic acids out of solution, so that they more readily interact with the column. However, I still recovered a good deal of pure RNA, which was lower though roughly consistent with my yields from the validation. So I reran my extraction protocol with the ethanol addition and my yields were somewhere between 1.5-2x greater.
My question is would this necessarily be a causative issue with the lack of KO in the TaqMan experiment? I'm testing this hypothesis in the next couple days, but I would like your input to see if this reasoning is solid. If it's not then maybe I can save some reagents. I just find it odd that I got a decent quantity of high quality RNA from the extractions without the ethanol precipitation. Plus, when I run the cDNA from these batches I still get good amplification curves (all thresholding between 15-25 cycles) with little variability between technical and biological replicates. So I don't think it's an issue with cDNA synthesis or the assay itself. Could the lack of ethanol precipitation bias my pool of total RNA? Wouldn't any bias that results from RNA loss in the extraction process be relatively random? In turn, shouldn't I still be able to see my KO effects at the mRNA level? Or could the lack of ethanol precipitation cause the total RNA pool to be of lower quality or integrity? I've ran samples on both nanodrop and qubit, and both my 260/280 & 260/230 are consistently between 2-2.1.
Any ideas or questions to help me work through this would be greatly appreciated?
Marty
How many days after transfection do you collect the ran samples? You may need to harvest at a latter time point.
Two days. I was under the impression that mRNA knockdown was quite rapid, within 24-48 hr. Whereas longer collection times were more to see changes in protein level.
Transfection occurs within 6hrs, and I'm not entirely sure if that would be an indicator of how long it would take for KO to occur in SH-SY5Y. There is one paper in particular that has done westerns successfully at 48 hours with this cell line, so if the proteins not there at that time point I'm assuming the same for the mRNA. If I might ask, what do you think about the potential issue with the ethanol precipitation?
Thanks,
Marty
"Typically, efficient siRNA-meidated transcript down-regulation is considered to be > 10-fold, ideally up to 100-fold." I'm talking about a positive control. So I don't think that siRNA design is the issue, especially since we've validated it previously and have gotten up to 16 fold change.
I don't think it's the transfection reagent. We've tested messengerMAX, RNAiMAX, lipofectamine 3000, and electroporation. As you can see from my picture, viromer is giving 90+% transfection efficiency 6 hours post exposure. Also, viromer gave the strongest knockdown at equimolar concentrations of siRNA (~ 90% KO at high concentration). Until very recently it was reproducible, and I'm trying to figure out what's gone wrong. Maybe it has to do with siRNA concentration, though I'm doubtful.
Also, RNAiMAX kills all of my cells, and gives a statistically insignificant knockdown. Not entirely sure what makes it best-in-class for my cell line.
Well I followed the manufactorers protocol. More than happy to do another transfection with RFP labeled siRNA to confirm again. Also Im not the only person to report difficulties with RNAiMAX or even Lipo3000 for these cells. Particularly because these are difficult to transfect cells. Actually Lipo3000 worked better at
So the issue was with the RNA extraction mistake afterall. I nanodroped new samples today, and had close to 3x the yield. Moreover, I ran the assay, and the got the expected KO I was looking for.
sometimes siRNA is delivered into the cells, but is stuck in some compartments where it can not display biological activity- and gradually gets degraded or expelled (depends on the delivery reagent etc). Also siRNA can have low or no activity (poor design or synthesis). Or isolation and downstream analysis were suboptimal
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Medical Pharmacology
- Clinical Pharmacology