qPCR You Can Trust: Samples, Controls and Shipping

Quantitative PCR follows the amount of amplified DNA in real time through a fluorescent signal collected at every cycle. The primary readout is the Cq value (also written Ct): the cycle number at which the signal crosses a defined threshold. The more target molecules there are at the start, the earlier the threshold is crossed and the lower the Cq. A Cq is not a copy number by itself; it allows comparison between samples that were measured under the same conditions.

There are two ways to turn that value into a result. In relative quantification, the amount of the target is compared between samples against a reference gene and a calibrator sample, and the result is reported as a fold change. In absolute quantification, a standard curve is built from serial dilutions of a standard of known copy number and the sample Cq is read off that curve. The standard curve is also a source of the reaction efficiency, so neither approach is meaningful without a trustworthy curve or a validated efficiency. Deciding which approach is needed before the analysis directly shapes the sample and control plan.

If the RNA is degraded, no qPCR protocol can repair it. Integrity is reported by capillary-electrophoresis systems as a numerical score such as RIN or RQN; the score is an indicator of degradation, but there is no single correct cut-off for every sample type. Short amplicons tolerate partly degraded RNA better than long ones. What matters is to set the acceptance level in advance and to make sure that all samples to be compared have similar integrity. If the degree of degradation differs between groups, a real biological difference becomes indistinguishable from a technical one.

The most common purity hints are the spectrophotometric A260/280 and A260/230 ratios. A low A260/280 can point to residual protein or phenol, and a low A260/230 to organic carry-over such as guanidine salts, phenol or carbohydrate. They are only hints, however: pH and buffer composition shift the values, the ratios become unreliable when the nucleic acid concentration is very low, and an inhibitor can be present even when the ratios look good. Concentration has a similar limitation. A spectrophotometer cannot tell RNA from DNA, free nucleotides or degradation products, whereas fluorometric dye-based methods (Qubit-type assays) are specific to the target molecule and usually give a more accurate concentration for dilute or impure samples. Reporting both measurements together is the most informative approach.

When RNA targets are measured, a reverse transcription (RT) step must precede qPCR. This is an enzymatic reaction that converts RNA into cDNA and it carries its own variability: its yield depends on the sample, on the priming strategy (random hexamers, oligo-dT or gene-specific primers) and on the RNA quality. For that reason, samples to be compared should be processed with the same RT protocol, ideally in the same batch and with a similar RNA input. MIQE asks for the RT conditions to be reported in detail because this step is one of the least visible but most influential sources of variation in the final result.

The second major risk is residual genomic DNA (gDNA) in the sample, which can amplify the target and inflate the signal artificially. We manage it with three measures. The first is a DNase treatment during RNA preparation. The second is, where possible, to design primers that span an intron or sit across an exon-exon junction, so that gDNA gives no product or a product of a different size. The third, and the most important, is to run a no-RT control, a reaction that contains no reverse transcriptase. If the no-RT control shows a clear signal, that signal comes from gDNA and the true RNA signal of the sample has to be interpreted accordingly. It is therefore good practice to send enough RNA to cover a no-RT control for every sample.

A few controls are indispensable in every qPCR study. The no-template control (NTC) contains water in place of template and reveals reagent or environmental contamination and primer dimers. The no-RT control, as described above, catches gDNA carry-over. A positive control proves that the reaction and the primers work, and in particular lets you tell a failed assay from a true negative when the result is negative. In studies spread over more than one plate, an inter-run calibrator, the same sample placed on every plate, allows plate-to-plate differences to be corrected.

In relative quantification the reference gene is itself an assumption. There is no single universal housekeeping gene; expression can vary with tissue, treatment and experimental condition. MIQE recommends that reference genes be validated for each experimental system and, preferably, that more than one be used. Replicates also call for two concepts that should not be confused. Technical replicates (usually triplicates) measure pipetting and instrument error, whereas biological replicates represent the real variability between individuals or independent samples and are the basis for any statistical inference. Three technical replicates do not stand in for three biological replicates.

A primer pair or probe must be validated for efficiency before it is used. Efficiency is calculated from the slope of a standard curve made from a serial dilution of sample cDNA or of a known standard. A commonly used acceptance range is 90 to 110 percent; efficiency outside that range can point to poorly designed primers, secondary structures or inhibition. When efficiency is not close to one hundred percent, the assumption of a perfect doubling at every cycle used in comparative calculations no longer holds and results are shifted systematically. For intercalating dyes such as SYBR Green, melt-curve analysis is also needed: a single sharp peak indicates the expected product, whereas several peaks, or a peak at a low temperature, indicate non-specific products or primer dimers. Hydrolysis-probe chemistries have no melt curve; specificity there comes from the probe design.

Inhibitors can come from matrices such as blood, plant tissue, soil or cell lysate, or from carry-over of the extraction reagents, and they delay Cq values or sometimes stop the reaction altogether. For suspect samples a dilution test is used: the sample is diluted over several steps and the Cq is checked to shift by the expected amount. If the Cq shifts much less than expected, or becomes earlier in the diluted sample, inhibition is present. As an alternative, a known amount of an external spike-in control can be added, which shows directly whether inhibition occurs. Noticing such problems before the run begins saves both time and sample.

Nucleic acid samples are damaged mainly in two ways: nuclease contamination and repeated freeze-thaw cycles. RNA is especially sensitive. Keep the sample in nuclease-free tubes and in nuclease-free water or a suitable buffer, split it into small aliquots to avoid repeated freezing and thawing, work on ice and ship without breaking the cold chain. Label the tubes clearly with a durable label carrying the sample ID, type and date; making the names on the shipping list identical to those on the tubes prevents mix-ups. The shipping method (dry ice, gel packs, stabilised sample) is best agreed with the laboratory before the analysis.

In the request form, state the target gene or sequences and, if you have them, the details of the primers or probe you want to use; the preferred chemistry (SYBR dye or hydrolysis probe); the sample type and extraction method; the number of samples and how much material is available; the number of technical and biological replicates wanted; the controls and reference genes to be used; and the expected output (raw Cq values, relative expression, absolute copy number). NAKTEM offers qPCR molecular analysis, and our team agrees the sample requirements, the controls and the shipping method with you before the run. That conversation beforehand is very often the single most valuable step for the reliability of the result.