Addition of DNase after RNA extraction - (Dec/27/2015 )

So I usually work with RNA viruses and after RNA extraction with Trizol I never add DNase because I don't need to. I remember when I was student I extracted RNA from cell lines and according to Qiagen kit I had to add DNase to remove genomic DNA.

Now I am working with a new team, but they never add DNase prior to cDNA synthesis. Wouldn't this create non-specific band or false band in PCR? Because the primers would bind to the genomic DNA contaminating their RNA!? Does this depend on the primer that they design, specifically for mRNA?

If they design their primers to amplify across a spliced intron boundary, then only the spliced version of the mRNA will amplify, eliminating the amplification of the gDNA (since it will contain the intron). But this relies on careful primer positioning. In general, it is safer to use DNAse (and essential if working with prokaryotes without splicing).

I need to practice this. The cells are human T47D cells. They are vimentin negative. I need to make sure the mRNA is not expressed. I am new to designing primers this way. I should have learned this long time ago.   

There's a lot of splicing data available for human cells. Try designing primers across exon-exon junctions using Primer-BLAST (one of the NCBI tools, just do a web search for "Primer-BLAST").

For your experiment, make sure you have a positive control, e.g. a cell line that should express your mRNA. Otherwise, a blank lane could just mean that your RNA extraction or cDNA synthesis didn't work.

For qPCR it's better to add a DNAse, for Trizol you need it definitely, Qiagen columns shouldn't be binding DNA so well, but you still need to get rid of DNA traces. Q have a protocol adding DNase directly to column, but my colleague still experienced DNA contamination after that.

We used to treat with TURBO DNA-free kit (originaly from Ambion, now LifeTech or whatever company bought it), since adding DNAse into a purified RNA is not such a good thing if you can't get rid of it as well (Trizol is not compatible with in-isolation digestion). This worked well, but the little sandy stuff, that binds of the DNase was sometimes hard not to aspire. We did increase the centrifugation time.

If you design intron-spanning (ideally primer that lies exactly on the exon-exon boundary, that would not bing DNA at all, but that is tricky to design.. usually spanning a large >1kb intron is fine enough) you don't need to worry much about amplifying gDNA, but in any case, without DNase treatment, you never get an acurate concentration measurement and purity. On the other hand, spectrophotometric measurement is probably pretty off anyway.

For qPCR, testing if your cDNA has gDNA contamination or if your primers bind gDNA at all, RT- reaction can be used, but we only did that on the beginning, when we get nothing we concluded,that DNA will by removed by this or just used one of the samples as a test of DNase digestion. RT- reaction uses the same amount of cDNA as your sample, you usually don't have that much of it and of reagents to spare. But of course it if was a crucial experiment, you would want to know  you don't amplify gDNA. If you have RT- for every sample, not just one to check the procedure, you can use the RT- Cts to substract the "real Ct" from your results.

Even when we sometimes did see amplification in RT- (as with Q DNase) it usually went in very high Cts, like 35. So the traces will get amplified eventually, but it truly are just traces.

It's a good practice to do the treatment, but there are so many things that can affect the precision of the assay, from the efficiency calculation (of it's lack of), to having not-so-good reference genes, to.. whatever happening during isolation, so when you get very tiny differences, it doesn't matter if you did the treatment or not, the difference probably doesn't mean anything. And if you get a huge difference there is not a way how gDNA traces could influence that. So, not a complete disaster if you don't do it and still read the results in context.

But Pfaffl and other qPCR guys.. they always say, if you want to have an assay with ability to distinguish significantly very small diferences, you need to take care for all the details, including RNA quality and integrity and so. All the MIQE guidelines. But most people don't fully follow them.