№ 158 · biology

PCR: copy a needle until you can see it

Sometimes only a trace of DNA can be recovered, as at a crime scene, McMurry notes. His example starts below one picogram of DNA with chains 10,000 nucleotides long; a picogram of such DNA is about 105 molecules. PCR, invented by Kary Mullis in 1986, copies a chosen stretch of DNA over and over until there is enough to sequence and characterize. The surprise is how fast: within a few hours that trace can become several micrograms.

What PCR needs

McMurry lists the ingredients. There is the double-stranded target DNA, Taq DNA polymerase and magnesium ions. There are the four monomers, deoxynucleotide triphosphates (dNTPs). And there are two short primers, roughly 20 bases long, added in large excess. Each primer pairs with the sequence at one end of the target stretch. Taq polymerase comes from Thermus aquaticus, a bacterium McMurry places in a Yellowstone hot spring; Clark, Douglas and Choi note it withstands the high temperatures PCR uses.

Three temperatures, one cycle

McMurry describes each cycle as three steps. At 95 °C the double strand denatures, splitting into two single strands. The temperature drops to between 37 and 50 °C so the abundant primers anneal by hydrogen bonding to their complementary ends. Then it rises to 72 °C, where Taq polymerase adds nucleotides onto the primed strands. Each original strand is now copied, so one cycle gives two copies.

Predict first

After 30 cycles, how many copies would perfect doubling give, compared with what McMurry says is routinely achieved?

Check your answer

Perfect doubling gives 230, roughly 109. McMurry says each cycle falls short of 100% efficiency, so 30 cycles routinely yield about 106 to 108.

Why the count is exponential

Ideally each cycle doubles the number of target molecules, so the total grows as a power of two. McMurry gives the series: two cycles give four copies, three give eight, and so on. He notes that 30 or so cycles can run in an hour. The gap between the theoretical 230 and the observed 106 to 108 comes from imperfect efficiency per cycle.

Interactive Drag the cycle slider and the efficiency slider to see how many copies accumulate; the readout compares the count with McMurry's routine range for 30 cycles.

Copies after n cycles are 2n times the starting amount at 100% efficiency, or (1 + efficiency)n below that, plotted on a log scale. The shaded band marks McMurry's routine 106 to 108 copies at 30 cycles. The efficiency slider is illustrative; McMurry gives no per-cycle figure.

Watching the copies pile up

Standard PCR measures product only at the end, which Artika and colleagues say makes quantification imprecise. Real-time PCR instead reads a fluorescent signal during each cycle. Artika et al. describe the amplification curve's phases: an early linear stretch below the baseline, an exponential phase where product doubles each cycle, and a plateau once components become limiting. Calculations use the exponential phase; Artika et al. say the plateau is not useful for them.

The threshold cycle, Ct, is the fractional cycle at which fluorescence rises clearly above background. More starting template means fewer cycles to reach it. Artika et al. caution that Ct is a relative value: it needs a standard curve of known material to become a viral load.

Where it is used, and where it fails

Clark, Douglas and Choi list uses from paternity testing to detecting genetic diseases. Artika et al. note that as of 2022 real-time RT-PCR was considered the gold standard for confirming SARS-CoV-2 infection, with a Ct below 40 read as a positive detection. They report a false-negative rate near 9.3% in one large study, and a false-positive rate of 0.5%. Mutations that change the sequences the primers and probes target can cause false negatives. In PCR tests generally, contamination is the main cause of false positives, since even one stray copy can be amplified into a signal. The result, they stress, must be interpreted with care.

In short

PCR works because each cycle ideally doubles a chosen sequence, and a few hours of cycling can lift a trace sample into a measurable amount. Real-time versions track that doubling as it happens, which Artika et al. say makes quantification reliable.

Where this comes from

  1. Biology 2e, 17.1 Biotechnology (OpenStax) linked only, not reproduced
    Clark, Douglas, Choi · 2018
    openstax.org/books/biology-2e/pages/17-1-biotechnology
  2. Organic Chemistry, 28.8 The Polymerase Chain Reaction (OpenStax) linked only, not reproduced
    McMurry · 2023
    openstax.org/books/organic-chemistry/pages/28-8-the-polymerase-chain-reaction
  3. Real-Time Polymerase Chain Reaction: Current Techniques, Applications, and Role in COVID-19 Diagnosis (Genes 13:2387) linked only, not reproduced
    Artika, Dewi, Nainggolan, Siregar, Antonjaya · 2022
    pmc.ncbi.nlm.nih.gov/articles/PMC9778061/