Six useful viral qRT-PCR tips

Quantitative real-time PCR (qRT-PCR) is a sensitive and accurate method for detecting viruses in samples, or to test for transcript variation in infected cells. However, some viruses might add additional levels of trickiness to this otherwise well-established technique. Here are six tips for using viral qRT-PCR correctly (or at least for realizing if something is going wrong)!

Tip#1 Relative vs. absolute quantification

A good qRT-PCR will efficiently amplify a specific target sequence, resulting in accumulation of fluorescent signal. The Ct value (cycle threshold) of a qRT-PCR is the number of cycles required for this fluorescent signal to cross a threshold (e.g., to exceed the background level). However, Ct values alone cannot tell us how much virus is in a sample or by how much the expression of a gene has changed because of infection.
There are two principal types of quantification for qRT-PCR data, relative and absolute. Absolute quantification reports the actual number of copies of a target nucleic acid sequence in a sample based on a known quantity (e.g., using a standard curve). Relative quantification reports changes in gene expression in a sample relative to another reference sample (e.g., an uninfected control sample).
Choosing the correct method will depend on your aim. For example, if the aim is to correlate viral copy number with a disease state, absolute quantification is recommended. If the aim is to measure virus-induced changes in cellular gene expression, relative quantification is more commonly used.
Bonus tip: Although Ct values are sometimes used in the clinic to report changes in viral load, in most cases this data needs further analysis for proper quantification.

Tip #2. If using relative quantification, select a suitable reference gene

Good quantitative PCRs often depend on choosing a suitable reference gene. A reference gene is used as an internal control to normalize transcript-of-interest data before making comparisons between samples. This normalization accounts for variations between samples (e.g., because of pipetting or sampling error). However, if your reference gene is not stably expressed under your test conditions (e.g., infected vs. non-infected cells), you’d better stop there!
For example, you are testing whether a transcript-of-interest is more abundant in virus-infected vs. non-infected cells. Now imagine that the expression of your reference gene is affected by the virus. In a situation like this, you could easily come to an erroneous conclusion. So, try to find reference genes that are stably expressed under your experimental conditions. If you can’t find these from the literature, you might have to test a selection of promising reference genes to identify the most suitable. As a starting point, we recommend Radonić et al.’s 2005 Virology Journal article. Here they tested the performance of ten potential reference genes in cells infected with various viruses, including DNA and RNA viruses (coronavirus [SARS-coronavirus], flavivirus [yellow fever virus], herpesvirus [Human herpesvirus-6] cytomegalovirus and orthopoxvirus camelpox). Based on their findings they recommend TATA-Box binding protein (TBP) and peptidyl prolyl isomerase A (PPI) as the most suitable for normalizing qRT-PCR data from virus infected human cell lines.
Sometimes, however, the situation gets even more complicated, as certain viruses are known to inhibit cellular transcription and translation. In such cases, using a low multiplicity of infection or early time-points might help. Alternatively, if you are lucky enough to have a mutant that reverts the phenotype that you are investigating, then you could compare the transcript-of-interest between mutant and wild-type viruses, as they presumably display the same levels of reference gene transcript. The take home message here is to proceed with viral qRT-PCR with caution and, where possible, support your qRT-PCR findings with a complementary method.

Tip #3. Start with the same amounts

Good qRT-PCR is all about comparing similar samples, and it is best to check that this is the case at every step of the process. That means starting with the same sample amounts and using the same quantity of RNA for your cDNA synthesis. This is not always as easy is it seems. For example, HIV patients with advanced immunosuppression have fewer cells per milliliter of blood than patients with less advanced immunosuppression. In this case, therefore, it would be better to begin with equal cell numbers than equal blood volumes.
It is also important to ensure that the RNA concentrations of your samples are equal before progressing to cDNA synthesis. There may be an exception to this rule. What if, for instance, your virus of interests reduces cellular RNA synthesis but boosts the synthesis of viral RNA? While in this case we might decide just to go with equal volumes of RNA rather than equal concentrations, the data would have to be interpreted with caution. Again, in these extreme cases, qRT-PCR might not be the best technique to use.

Tip #4. Beware of false-positives and false-negatives

qRT-PCR is a powerful technique, but sometimes things go wrong. When qRT-PCR detects something that isn’t there, this is called a false-positive. When qRT-PCR fails to detect something that is there, this is called a false-negative. False-positives might result from lab contamination, while the presence of inhibitors of your PCR reaction might produce false-negatives.
For example, the active intracellular metabolite of acyclovir, an antiviral drug used to treat immunocompromised patients, can produce false negatives by inhibiting the diagnostic PCR for cytomegalovirus. If you are concerned about false negatives, consider using commercially available exogenous internal positive control reagents that can distinguish true target negatives from PCR inhibition (e.g., TaqMan® Exogenous Internal Positive Control Reagents).
Another way to gain confidence in your qRT-PCR data is to back it up with a complementary but independent approach. For example, if there is an antibody available that targets your protein-of-interest, support your qRT-PCR data with Western blotting. Reviewers will find it difficult to disbelieve your findings if you have used multiple independent approaches. For clinicians, we recommend the CDS’s guidance on the use of RT-PCR for diagnosis.
Bonus tip: Always include positive and negative controls in each run and each plate.

Tip #5. Choose your primers wisely

Primer design is critical to qRT-PCR, but good primer design can be a challenge. This is especially true for viral PCRs. Co-evolution of viruses with their hosts means that they often share significant amounts of homology. Also, you might need to discriminate between closely related virus subtypes. This can make it difficult to achieve robust amplification while also distinguishing between the target and non-target sequences.
Luckily, there are several online database resources to help speed your search for appropriate viral primers, including MRPrimerV, VirOligo, PrimerHunter, and The Influenza Primer Design Resource.
Bonus tip: Every time you use a new set of primers, remember to run a melting curve to check for non-specific binding.

Tip #6. Fully describe how your qRT-PCR was done

Finally, you can do the best qRT-PCR in the world, ever. But unless you properly describe how your experiment was done and the data analyzed, reviewers will find it difficult to assess the validity of our results and other investigators might be unable to reproduce your findings. For a useful checklist of what you should report, see the Clinical Chemistry review article by Bustin et al. (2009) entitled The MIQE Guidelines (for Minimum Information for Publication of Quantitative Real-Time PCR Experiments).