What purpose do PCR controls serve?

Too many!! Well, that's the glib answer. Controls are ideally required at every step of the PCR or in fact any experimental process - and there are even more when we create qPCR protocols. A specific type of control is the 'standard' which are required to give the user some point of reference on which to base a quantitative PCR (qPCR). The simplest are the positive and negative (also called no template, reagent and RT controls) which indicate whether a PCR has worked or is contaminated, respectively.The term 'ideally' is often used describe the implementation of controls in PCR. This is because 'realistically' controls are infrequently applied or poorly utilised generally reflecting a lack of understanding.

Some controls and related terminology include:

Purification control
Ideally, a known amount of template which differs in sequence from the desired target is added to the specimen prior to nucleic acid purification. This permits monitoring of the performance of the purification procedure. An unknown amount can be used but this will only provide a qualitative (yes/no) insight into the efficiency of the purification procedure. Alternatively you can monitor the threshold cycle values if you use real-time PCR with an unknown amount of control, however this approach is subject to external variables affecting amplification efficiency and incorrectly changing the C_T. This control can also been taken through the entire process and used as a loading control or an internal control.

When used as a control for purification, its probably best to evaluate your particular method over a period of time or for a certain number of specimens depending on your laboratory through-put as your method should not vary significantly. For quality purposes, it is important to stick to this period or number for future evaluations.

Something else to keep in mind is that different specimen types may behave differently in the same purification method so you may need to evaluate the purification efficiency for each specimen type you handle e.g. for whole blood, nasal washings, solid organ samples or paraffin embedded tissues, or at least for each different purification protocol you use. You may also wish to reintroduce your control for each new batch of purification reagents or each new kit lot.

These templates can be endogenous gene sequences (usually human, animal or plant but proviral retrovirus sequences are also useful), exogenous genetic material (addition of genomes from an entirely different preparation of another organism) or a synthetic fragment (a different purified amplicon or a cloned amplicon or gene fragment).

An additional set of primers is required to amplify this template as well as your target template of interest. This can introduce competition for PCR reagents and ultimately decrease the amplification efficiency of one or both targets so considerable optimisation is required prior to release of the protocol for general use on the bench. Nonetheless, the range of input template concentrations giving a proportional amplicon signal (dynamic range) is limited to about 3 log₁₀s. For PCR-ELISA and real-time PCR, a portion of cloned target can be modified to bind a different oligonucleotide probe (oligoprobe) whilst requiring only one set of primers.

Ideally a mutated, cloned / synthetic control requiring a separate oligoprobe (for use in PCR-ELISA or rtPCR methods) which hybridises the same primers (homologous control) is the best template as this will permit a defined amount to be added which minimises deleterious competitive effects and variable amplification efficiencies. Additionally you end up with a renewable resource of control that can be made in large batches, aliquotted and frozen. The essential caveat for using these controls is that they are accurately quantified. The use of limiting dilution PCR (LDP) provides the best, statistically verifiable method of determining the amount of amplifiable template present in a preparation. Recent improvements to spectrometry (see below) have increased its sensitivity but the lab spectro is still a source of considerable variation and therefore not the best approach to characterising controls.

Considerations when using synthetic controls

Should you chose to use a synthetic control you may wish to linearise the template and store that away rather than use circularised plasmid. However we have not seen any conclusive data that indicate this is necessary except in certain template-specific instances.

Also, when making and storing a synthetic control, include innocuous salmon-sperm or poly(A) DNA or MS2 RNA (10ng/µL). This reduces the probability that your standard molecules will degrade over time when being stored, and it helps mimic the natural environment of the sample matrix - well it helps a bit anyway!

It helps the experiment (but not the budget) to use siliconised tubes for storage as DNA non-specifically adheres to normal plasticware over time. We're not sure if it does that below -20°C.

Template loading control
This control provides some indication of the amount of NA added into the reaction tube so that results from different samples can be accurately compared. For example, twice as much amplicon resulting from amplification of sample A compared with sample B may be due the presence of twice as much total NA in sample A, rather than twice as much specific template. This could result from sample A having more cells than sample B or sample B may have experienced problems during the NA purification process which reduced the NA yield or retained more PCR inhibitors.

Ideally a housekeeping gene target would be employed to analyse the amount of target NA present in a preparation. the assumption here is that the total NA present represents the amount of target NA purified. While this is expected to be the case for human, animal and plant gene targets where co-purified housekeeping targets are abundantly available, it may not be so for microbial targets since these organisms exist both intra- and extracellularly and may be present in far greater or smaller numbers than the number of cells present in the collected specimen. An external standard curve of housekeeper dilutions could be used to interpret the results of amplifying the housekeeper target present in the sample.

Because of these problems, this data is often obtained using a spectrometer to measure the total NA content at 260 or 280nm prior to the assay. The sensitivity of this approach can be enhanced using a NA-binding dye such as PicoGreen™ or RiboGreen™ together with a fluorimeter. Although this approach gives no indication of the amount of specific template present, it does provide a guide to the performance of NA purification among similar sample types (matrices). Its difficult to compare purification yields from different matrices in the same batch, e.g. blood vs. sputum vs. urine, because one matrix may inherently be more or less cellular than another and it is from the cells that the majority of the NA is released.

Reverse transcription control
The reverse transcription (RT) step is acknowledged as the most variable part of an RT-PCR. Yields of cDNA from an RNA template can be as low as 40% and the efficiency of the RT enzyme is affected by many things. RNA is the template of choice for this control, and it can be made in vitro and quantified and stored in similar ways to a synthetic DNA control (see above). One hassle with DIY RNA controls is the removal of template DNA - and it is essential that you check for this. Just to confound matters, Taq DNA polymerase can also use high concentrations of RNA as a template - so you may be trying round after round of DNase treatment only to find the RNA preparation still amplifies in by PCR in the absence of an RT step. Try diluting out the preparation before amplifying - if you stick to arithmetic dilutions you should see it drop off fairly soon, depending on how highly concentrated your RNA preparation is.

Internal control (IC)
The IC is most often used to monitor PCR inhibition but can also be used as a purification control, RT control (if in RNA form), reference and a standard. The IC can be added to the matrix or purified NA (an exogenous IC) or it can be a target present within the sample matrix which is co-purified with the target of interest (an endogenous IC). Because the IC is usually co-amplified it should be chosen/designed so that its amplification does not out-compete the target of interest and it must be able to be discriminated from the target of interest. This usually occurs via a change in length or in oligoprobe hybridisation sequence. Competition can be minimised or manipulated in the following ways:

• Add a known, small number of copies, especially if it hybridises the same primers
• If the IC uses separate primers, limit the amount included
• To identify the amount of inhibition present, add small quantities of IC (for low level inhibition) or different amounts in several co-amplifications to crudely quantify the level of inhibition
• Early qPCR was based on competition between an IC and the target of interest

Some experts advise that qPCR assays include the assumption that your templates are inhibitor-free. If you want the best results from your qPCR - check that assumption sing an IC.

Reference
Another broad term often used interchangeably with IC. A reference molecule produces a signal which may be used to normalise all experimental data. This type of control can be amplified during the PCR (an active reference) or may not be amplified and thus used to monitor non-PCR related fluctuations such as pipetting errors and non-specific fluorescence quenching (a passive reference). The active reference is analogous to an IC whereas the passive reference is a unique term developed and most cleverly applied by the Applied Biosystems team.

Some real-time PCR instrument makers claim that the function of a passive reference is not essential if the instrument is able to reproducibly amplify replicates of the same target. Some claim the use of the second fluorophore in some oligoprobe systems (e.g. HybProbe) can perform the function of a passive reference. More data is required to convince us of these claims.

The take home message is that "IC" and "reference" are very general terms and their use should be accompanied by a description to ensure the reader/user knows exactly what the intention of your IC or reference is.

Standards
These are well characterised control templates required to perform qPCR. Characterisation usually implies some mix of the following:

• Quantified;
• Generally made in bulk;
• Easily renewed;
• Ideally amplified with equivalent efficiency to the specific target.

The most simple use is for relative qPCR where a single amount can be amplified to determine an experimental threshold against which sample results are compared.