Conceptual question about PCR vs. qPCR

[newmanreb]

Hi all, 
I'm doing some research at my company about implementing qPCR, because we are still in the dark ages with conventional PCR (cPCR). I came across this information in a publication, and I haven't been able to find more help online to clarify what the authors are saying. I was hoping someone here could illuminate things for me.

This is a quote from a 2002 paper, citation below. It says, "A major reason why classical PCR has not been adopted by most plant disease regulatory and diagnostic laboratories is the time and labor required to confirm the identification of the PCR amplification product. The simple presence of a particular molecular weight DNA fragment in an agarose gel does not prove the identity of the resulting band, and verification of the amplified product must be done by Southern blot hybridization. Another major factor is that the technique is not much more sensitive than ELISA, and it is much less sensitive than isolation of the organism on semiselective agar media (Wang et al. 1999). However, these concerns do not apply to qPCR." Later in the article, the authors include this statement: "Real-time PCR has many important advantages over classical PCR: (i) it eliminates the need to do a Southern blot to confirm identification of PCR product...." 

I have a few questions about these statements: 
1. Has cPCR advanced since 2002 to the point where confirmation of the resulting sequence is not always needed? I'm fairly certain cPCR is more sensitive than ELISA, potentially by a lot. We do some plant diagnostic tests where a strong band is considered a positive, but usually we will use Sanger sequencing for identification. (That could just be lazy assay design on our part - we're working on that.) Back in 2002 Sanger sequencing was definitely around, but maybe it was significantly more expensive than Southern blotting?
2. My main question is this - why is it that qPCR products do not need sequence identification? My understanding of qPCR is that it's almost the same chemistry as a cPCR reaction, it is simply monitored with fluorescence. Is it because of the increased sensitivity? How would that be explained?

Thank you so much for any thoughts & discussion. 

 

Rebekah


Schaad, Norman W., and Reid D. Frederick. "Real-time PCR and its application for rapid plant disease diagnostics." Canadian journal of plant pathology 24.3 (2002): 250-258.

[CharonY]

 Fundamentally PCR has not changed (let's just call it PCR, it is what it has been called forever). So not much has changed in diagnosis of the correct amplicon. But I think the confusion might be due to the fact that not all qPCRs are created equal. Some use intercalating dyes to detect double strand DNA and in this case, there is no real additional information of the amplicon over PCR (aside from a melting curve, which can be generated after the run and which is kind of helpful in that regard).

However, there are qPCRs that use a probe that binds to the target region, similar to a Southern blot. That one is what the author refers to as being more specific.

[newmanreb]

Thank you CharonY! This helps me out a lot. As for the probe that binds to the target region, how long in bp would that tend to be? I suppose I assumed that the probes would be about the same length as primers, and although primers specifically bind to the target, they are clearly not enough to determine the identity sequence. However, they may be thinking of universal primers used to amplify the 16S region, for example, which are not designed to be species-specific and therefore not intended to identify the sequence. 

[CharonY]

As a whole qPCR tend to use fairly short target regions (usually <250 bp), this helps to keep amplification cycles really short and probes usually are only between 18-30 bp. Part of the limitation s that the labelled probe needs to be quenched and with longer probes it can cause issues. But they are ways around that (e.g. using free quenchers), so in theory one could design longer probes. But often that is not ideal for the performance of the assay.

30 bp or shorter is typically enough to be highly specific for a target gene within an organism, if run under sufficiently stringent conditions (your signal has to come from successful binding of the primers as well as a probe between them).

However things get tricky when we are e.g. looking for SNPs in mixed samples. There are approaches for community analyses where probes can be designed to fit certain taxa, but obviously the require quite a bit of validation work.

There is now a move toward doing more sequencing for validation, but despite cost reductions it may still be a bad hit to the budget.