How polymerase chain reaction works

Polymerase chain reaction, usually shortened to PCR, is a way to copy a chosen stretch of DNA many times so that a laboratory can see it, cut it, clone it or sequence it. The idea is older than any one instrument brand. A thermal-stable DNA polymerase, two primers and a controlled temperature cycle do the copying. This page explains the cycle, the reagent classes and the controls. It is a research explainer, not a clinical protocol and not a kit insert.

If you are comparing endpoint PCR with real-time measurement, read this page first and then the cluster on quantification. If you are ordering enzymes, nucleotides or plasticware, start from the molecular biology catalogue and send the scientific requirement with the quote request. A family name on a catalogue page is not evidence that a particular lot is on a shelf.

What the reaction is copying

PCR does not copy an entire genome in one tube unless someone has designed a very unusual experiment. It copies the region that sits between two primers. Each primer is a short synthetic DNA oligonucleotide. One primer binds one strand. The other binds the opposite strand, facing inward. The polymerase extends each primer in the 5-prime to 3-prime direction, using the annealed strand as the template and the four deoxynucleotide triphosphates as building blocks.

After the first cycles, the new strands themselves become templates. That is why the product can increase sharply. The increase is not infinite. Primers, nucleotides and enzyme are finite, and late cycles often plateau. A bright band is therefore not a quantitative count of the starting molecules. Endpoint PCR answers a presence and size question much more honestly than a "how many copies were there" question.

The three temperatures

A typical endpoint programme repeats three steps. Names vary, and the exact set-points belong to the enzyme and the primers, not to a universal recipe.

Denaturation heats the tube so the two strands of the template separate. Many Taq-type programmes use a short hold in the mid-90s Celsius. Hot-start enzymes may also ask for a longer initial activation. If the template is GC-rich or the fragment is long, incomplete denaturation shows up later as a weak or absent product.

Annealing cools the tube so primers can bind. This is the step people argue about, because it is the step that decides specificity. Too high, and the primers do not stay bound. Too low, and they bind in extra places. Calculated melting temperatures are a starting estimate. The salt and additives in the buffer move the real window. A gradient cycler, or three tubes at neighbouring temperatures, is a more honest optimisation than a single number copied from a paper that used a different buffer.

Extension warms the tube into the range where that polymerase synthesises DNA. For a classical Taq-like enzyme this is often around 72 Celsius, and a common planning figure is on the order of a minute per kilobase, adjusted for the enzyme class. High-fidelity polymerases publish their own speeds. Copying a vendor's table into a notebook without checking the enzyme you actually have is how extension times go wrong.

What is actually in the tube

A master mix is a convenience, not a mystery. The functional parts are the polymerase, a buffer with the right magnesium environment, the four dNTPs, the two primers, and the template. Some mixes add a dye so you can load a gel directly. Some use a hot-start antibody or a chemical block so the enzyme does little at room temperature while you are still pipetting.

Magnesium matters because it affects both polymerase activity and primer binding. Raising it can rescue a weak reaction and can also create extra bands. Treat magnesium titration as an experiment with a control, not as a knob to turn until something appears.

Primers are the specificity. Design them against a stated reference sequence, check them for dimers and hairpins, and blast them against the intended organism so you know what else they might bind. NCBI Primer-BLAST is a public place to start that check. A primer pair copied from a paper still needs a local test, because the template you have may not be the template they had.

Plasticware is part of the chemistry. Thin-walled tubes and plates exist so the block can change the liquid temperature quickly. A loose lid or a poor seal lets water leave the reaction. The volume drops, the salt rises, and the late cycles misbehave. Match the vessel to the block, and do not assume a plate from one instrument family seals on another.

Controls decide whether the band means anything

A no-template control contains every reagent except the DNA you claim to be testing. If it produces the same band as the samples, you have contamination or a primer artefact, and the sample lanes cannot be interpreted. A positive control, when you have a trusted template, shows that this mix and this programme can succeed. A failed positive control points at the reaction, not at every sample.

Inhibition is the quiet failure. A sample can be full of target and still look negative because heme, humic acids, ethanol carried over from a cleanup, or too much DNA suppressed the enzyme. Diluting the template, or spiking a known target into the sample matrix, separates "no target" from "the tube could not amplify". The companion note on PCR controls and contamination control goes further on how a shared bench gets into trouble.

How laboratories check the product

Most endpoint workflows still look at the DNA on an agarose gel next to a marker of known fragment sizes. The method is described in agarose gel electrophoresis for DNA. Load a defined volume, run far enough to separate the expected size from primer-dimer, and photograph the gel with the marker visible. Primer-dimer is usually a small, fuzzy product and is not your amplicon.

If the band will be cloned or sequenced, clean it. Leftover primers and nucleotides confuse a Sanger reaction. Cleanup is its own small workflow, not a rinse under the tap.

Where endpoint PCR is the wrong tool

Use endpoint PCR when you need a yes, a size, a clone or a sequencing template. Use real-time PCR, with a defined quantification strategy, when you need a careful comparison of starting amount. Use a digital method when you need partition-based precision and you have the instrument and the analysis to match. Isothermal methods such as LAMP answer some field-shaped questions with different primer rules and different false-positive modes. None of these replace a sequenced answer when identity is the claim.

Rooms, climate and shared cyclers

PCR is sensitive to DNA that arrived on a pipette barrel last week. Many laboratories physically separate preparation of the master mix from the bench where products are handled. That separation fails if the same gloves, the same tip box and the same lab coat move between the two rooms.

In a hot building, "room temperature" setup is not 22 Celsius just because a protocol printed in a temperate city said so. Enzymes and master mixes follow the storage on their own label. A cycler that rebooted after a power cut should be checked with a known positive control before a precious sample set is trusted to it. Write the programme down. Two cyclers with the same displayed temperature can still heat a plate differently if the block, the plastic and the calibration history differ.

What to send with a reagent enquiry

EVRINTH can take a sourcing question. The useful enquiry names the enzyme class or the documented performance you need, the amplicon length, the template matrix, endpoint versus real-time readout, and any constraint such as a hot-start requirement or a proofreading polymerase. It does not need a claim that a particular manufacturer relationship exists. Link the nucleic acid analysis pathway if the PCR sits inside a larger sample-to-result workflow, and use the multiplex PCR service reference only as a prompt for questions if you need a multi-target assay designed. That page is an independent method reference. Ask whether a quotation is possible. Do not read it as a statement that the assay is already running.

Safety

The main everyday hazards are hot blocks, ultraviolet light if you view a gel, and the stains used to see DNA. Follow the safety note that comes with the stain you actually use, and do not treat a research amplicon from an unknown infectious sample as harmless just because it is DNA. Your institutional biosafety rules decide the containment, not this article.