Polymerase Chain Reaction (PCR) Explained (Video)

Written guide: Polymerase Chain Reaction: Steps, Types, and Uses

This video explains the principles and steps of polymerase chain reaction for DNA amplification.

Watch on YouTube

Chapters

  1. 0:00 Why PCR was invented
  2. 2:11 How DNA polymerase copies DNA
  3. 4:03 What goes into the PCR reaction tube
  4. 5:38 The three steps of a PCR cycle
  5. 7:42 How the copies multiply
  6. 9:20 How to design PCR primers (Wallace rule example)
  7. 11:36 Taq polymerase and proofreading polymerases
  8. 13:17 Checking the PCR product on an agarose gel
  9. 14:25 Real-time PCR (qPCR), Ct values and RT-qPCR
  10. 16:43 Common PCR problems and how to fix them
  11. 18:47 Recap and question
Video transcript

0:00 Why PCR was invented

The polymerase chain reaction, or PCR, copies one chosen stretch of DNA into billions of copies in a few hours. We will start with why PCR was invented and how DNA polymerase copies DNA, then go through what goes into the reaction tube, the three steps of each cycle and how the copies grow exponentially. After that we will design primers, compare the heat-stable enzymes that make it work, check the product on a gel, measure DNA as it is made with real-time PCR, and finish with the most common problems and their fixes. First, we will look at why PCR was invented. Before PCR, studying a single gene meant separating it from a genome of about three billion base pairs, and one gene is a very small share of that DNA. To get enough copies of that gene, scientists cloned it in bacteria, which took days to weeks. In 1983, Kary Mullis, a chemist at the Cetus Corporation in California, worked out a way to copy a chosen stretch of DNA again and again in a tube. His method uses two primers, short single-stranded DNA molecules that mark the two ends of the stretch we want to copy. Repeated heating and cooling lets a DNA polymerase extend each primer, so the chosen stretch is copied many times over. The first version used a DNA polymerase from E. coli, and heat destroyed that enzyme in each cycle, so someone had to add fresh enzyme by hand after each heating step. The answer was a heat-stable polymerase from Thermus aquaticus, a bacterium that Thomas Brock isolated from a hot spring in Yellowstone National Park and described in 1969. Its DNA polymerase, called Taq polymerase, survives the heating steps, so one dose lasts the whole run, and a machine could then run the whole process automatically. Mullis shared the 1993 Nobel Prize in Chemistry for the invention of PCR. Today laboratories use PCR in disease diagnosis, forensic DNA profiling, cloning, sequencing and genetic testing. So PCR answered a practical need for many copies of one gene, and a heat-stable enzyme made the method practical.

2:11 How DNA polymerase copies DNA

Next, we will learn how DNA polymerase copies DNA, because PCR borrows the same enzyme that copies DNA in the cell. We covered how cells copy their DNA in our DNA replication video, and here we need only the parts that PCR relies on. DNA is two strands wound into a double helix, and the strands run in opposite directions, one from its five prime end to its three prime end and the other from three prime to five prime. Biologists call this arrangement antiparallel. Complementary base pairing holds the two strands together. Adenine pairs with thymine through two hydrogen bonds, and guanine pairs with cytosine through three. A single hydrogen bond is weak, so heat can separate the two strands without breaking the sugar-phosphate backbone of either strand. DNA polymerase builds a new strand by reading a single-stranded template, the strand it copies, and adding the nucleotide that pairs with each base. The new strand always grows from its five prime end toward its three prime end. The enzyme cannot start a strand from nothing. It adds nucleotides only to an existing three prime hydroxyl end, so it needs a short primer paired to the template. A primer is a short single-stranded DNA oligonucleotide, usually 18 to 25 bases long, and its free three prime hydroxyl end is where the polymerase begins. In the cell, an enzyme called primase makes a short RNA primer. In PCR, the scientist supplies DNA primers, and the choice of primers decides which stretch of the template gets copied. In the cell, the enzyme helicase separates the two strands, while in PCR heat does that job, and otherwise PCR copies DNA much the way a cell does during replication. With base pairing, strand direction and the need for a primer in place, we can assemble the reaction itself.

4:03 What goes into the PCR reaction tube

After that, we will look at what goes into the reaction tube, since each component supplies something the polymerase needs. We mix a standard reaction on ice in a thin-walled tube of 0.2 milliliters, in a total volume of 25 or 50 microliters. The template is the DNA that contains the target sequence. A few nanograms of genomic DNA are enough, and in principle the reaction can start from a single copy. The target is the stretch of DNA between the two primers, and the copied product is called the amplicon. The forward primer pairs with one strand at one end of the target, and the reverse primer pairs with the other strand at the other end, so the two primers point toward each other. We add both primers in large excess over the template. The four dNTPs, dATP, dCTP, dGTP and dTTP, supply the building blocks for the new strands. A heat-stable DNA polymerase, usually Taq, joins those nucleotides onto the primers. A buffer keeps the pH stable and supplies magnesium ions, usually as 1.5 to 2.5 millimolar magnesium chloride. The polymerase needs magnesium as a cofactor, so a reaction with too little magnesium gives little or no product. Water brings the mix to its final volume, and the tubes then go into a thermal cycler, the instrument that heats and cools them through the programmed temperatures. The tube now holds a template, two primers, four building blocks, an enzyme and its magnesium, and the thermal cycler takes over from here.

5:38 The three steps of a PCR cycle

Next, we will follow the three steps of a PCR cycle, which turn that mixture into new DNA. The run opens with an initial denaturation of 2 to 3 minutes at 94 to 98 degrees Celsius, because long genomic DNA needs extra time to separate completely. The first step of each cycle is denaturation. The thermal cycler holds the tube at 94 to 98 degrees Celsius for 15 to 30 seconds, the hydrogen bonds between the strands break, and the double-stranded DNA separates into two single strands. The second step is annealing. The cycler cools the tube to about 50 to 65 degrees Celsius and holds it there for about 30 seconds. The annealing temperature sits a few degrees below the melting temperature of the primers, or Tm, which is the temperature at which half of a primer is paired with its complementary sequence and half is single-stranded. At the annealing temperature, the primers pair with their complementary sequences on the single strands. The primers reach their sites before the long template strands can pair with each other again, because primers are short and present in large excess. The third step is extension. The cycler raises the temperature to 72 degrees Celsius, close to the optimum temperature of Taq polymerase. The polymerase binds at each primer and adds dNTPs to its three prime end, building a new strand complementary to the template. A common guide allows about 1 minute of extension per 1,000 bases of target. Denaturation, annealing and extension together make one cycle, which takes a few minutes, and a typical run repeats 25 to 35 cycles. After the last cycle, a final extension of 5 to 10 minutes at 72 degrees Celsius completes any unfinished strands, and the cycler then holds the tubes at 4 degrees Celsius. So each cycle moves the tube through three temperatures, and each temperature drives one molecular change.

7:42 How the copies multiply

Then, we will see how the copies multiply when the cycle repeats. Each cycle doubles the number of copies of the target, because each double-stranded molecule separates into two templates, and each template gains a new partner strand. In the first cycle, the new strands start at a primer but run past the far end of the target, because nothing stops the polymerase on the long template. In the second cycle, copies made from those new strands stop exactly at the primer site, which gives the first strands of the exact target length. By the end of the third cycle, the first double-stranded copies bounded by both primers appear, and from then on these exact-length copies dominate the reaction. In an ideal reaction, n cycles give 2 to the power n copies of each starting target. After 30 cycles, one starting molecule gives 2 to the power 30, which is about 1 billion copies. The longer strands from the first cycles increase only in a straight line, so they soon make up a negligible fraction of the product. Real reactions do not double forever. The reaction uses up primers and dNTPs, the polymerase gradually loses activity, and the abundant product strands pair with each other instead of with primers, so the reaction slows and reaches a plateau. For that reason, the amount of product at the end of a standard run does not show how much template was present at the start. Real-time PCR solves this problem, and we will come to it later in this lesson. The copies grow exponentially at first and then level off, and the primers decide which stretch of DNA grows.

9:20 How to design PCR primers (Wallace rule example)

Next, we will learn how to design good primers, since the primers decide whether the reaction copies the right target. The forward primer has the same sequence as the start of the top strand, so it pairs with the bottom strand. The reverse primer pairs with the top strand at the other end, so we write it as the reverse complement of that region. The two primers point toward each other, and the product runs from the five prime end of one primer to the five prime end of the other. Good primers are usually 18 to 25 bases long, which is long enough to be unique in a genome, with a GC content of 40 to 60 percent. The two primers of a pair should have melting temperatures within about 5 degrees Celsius of each other, usually between 52 and 65 degrees Celsius, so that a single annealing temperature suits both primers. A GC clamp, a G or C base in the last two positions at the three prime end, helps that end bind firmly. More than three G or C bases in the last five positions encourage mispriming, where a primer binds to a sequence other than its target. A primer must not pair with itself, which forms a hairpin, and the two primers must not pair with each other, especially at their three prime ends. Such pairing forms primer-dimers, short products that the polymerase extends. To see how melting temperature sets the annealing step, take a 20-base primer with 10 G or C bases and 10 A or T bases. The Wallace rule multiplies the number of A and T bases by 2 and the number of G and C bases by 4. Two times 10 plus four times 10 gives a melting temperature of 60 degrees Celsius. We then set the annealing temperature about 3 to 5 degrees below the lower melting temperature of the pair, so about 55 to 57 degrees Celsius. The Wallace rule is only a quick estimate for short primers, and primer design software uses more exact nearest-neighbor calculations. Primer-BLAST at NCBI designs primers and checks that they match only the intended target in the genome. A well-designed pair binds one site on each strand at a known temperature, and the next question is which enzyme extends the primers.

11:36 Taq polymerase and proofreading polymerases

After that, we will compare Taq with other DNA polymerases, because the enzyme decides how well the reaction survives the heat and how accurate the copies are. Taq polymerase keeps working through dozens of heating steps. Its half-life at 95 degrees Celsius is about 40 minutes, and it works fastest at about 72 to 75 degrees Celsius, the temperature of the extension step. Taq has no three prime to five prime exonuclease activity, so it cannot proofread. Proofreading is the activity that removes a wrongly added nucleotide before the new strand grows further. Without proofreading, Taq leaves roughly one wrong base per 10,000 to 100,000 bases copied. Such an error rate is acceptable for detecting a target, but cloning and sequencing need more accurate copies. Proofreading polymerases such as Pfu, from the archaeon Pyrococcus furiosus, remove wrongly added bases and make far fewer errors, so laboratories use them for cloning and other high-fidelity work. Taq usually adds one extra adenine to the three prime end of each product, without a template. TA cloning makes use of this single A overhang, while proofreading enzymes leave blunt ends. Hot-start polymerases stay inactive, held by an antibody or a chemical block, until the first heating step. The block stops the enzyme from extending loosely bound primers while the reaction sits at room temperature during setup, so fewer nonspecific products and primer-dimers form. The choice of enzyme therefore sets the accuracy of the copies and the shape of their ends.

13:17 Checking the PCR product on an agarose gel

Next, we will learn how to check the product on a gel once the run ends, because the tube itself shows nothing about what the reaction made. Agarose gel electrophoresis separates DNA fragments by size. DNA carries a negative charge on its phosphate groups, so in an electric field it moves through the gel toward the positive electrode, and shorter fragments move faster than longer ones. A DNA ladder with fragments of known sizes runs in one lane, and a successful reaction gives a single sharp band at the expected size of the amplicon. Each run also includes a no-template control, which contains water in place of template DNA. A band in that lane means that a reagent or the bench carries contaminating DNA, and then we cannot trust any result from that run. A positive control contains a template known to give the product. It shows that the reagents and the program work, so a blank sample lane can count as a true negative. A gel therefore tells us whether the target is present and whether the controls behaved, though a gel cannot measure how much target the sample held at the start.

14:25 Real-time PCR (qPCR), Ct values and RT-qPCR

Next, we will learn about real-time PCR, which answers the question a gel cannot answer. Real-time PCR, also called quantitative PCR or qPCR, measures the product during each cycle with a fluorescent signal. One method uses SYBR Green, a dye that fluoresces strongly when it binds double-stranded DNA, so the fluorescence rises as the product doubles. SYBR Green binds any double-stranded DNA, including primer-dimers, so a melt curve at the end of the run checks that only one product formed. Another method uses a hydrolysis probe, often called a TaqMan probe. The probe is a short oligonucleotide that pairs inside the target, with a fluorescent reporter dye at its five prime end and a quencher at its three prime end. While the probe is intact, the quencher absorbs the reporter's signal. During extension, the five prime to three prime exonuclease activity of Taq cuts the probe, the reporter separates from the quencher, and the sample fluoresces. Only the intended target releases this signal, so probes are more specific than SYBR Green. The instrument plots fluorescence against cycle number for each sample. The Ct, or cycle threshold, is the cycle at which a sample's fluorescence crosses a threshold set above background, and the same value is also called Cq. A sample with more starting target crosses the threshold earlier, so a lower Ct means more template. With perfect doubling, a difference of 1 cycle means a two-fold difference in starting amount, and a difference of about 3.3 cycles means a ten-fold difference. A standard curve made from known amounts of template converts Ct values into copy numbers. To measure RNA, such as the messenger RNA of a gene or the genome of an RNA virus, we first copy the RNA into complementary DNA, or cDNA, with the enzyme reverse transcriptase. This combination is reverse transcription PCR, or RT-PCR. With real-time detection added, it becomes RT-qPCR, the method used for SARS-CoV-2 testing and for measuring gene expression. The letters RT in these names stand for reverse transcription, not for real time. Real-time PCR therefore turns the timing of the signal into a measure of the starting amount.

16:43 Common PCR problems and how to fix them

In the final section, we will learn how to read a failed PCR on a gel and how to fix each result, using the controls and the cycle we have already covered. When a gel shows no band, the annealing temperature may be too high, the magnesium too low, the extension too short, or the template degraded or present in too small an amount. Inhibitors carried over from the sample also block the polymerase. Heme from blood, heparin, phenol and ethanol all interfere, and EDTA binds the magnesium that the enzyme needs. To fix an empty lane, lower the annealing temperature in small steps, check the template, and dilute or purify the sample again if inhibitors are likely. Extra bands of the wrong size mean that the primers bind to other sites. Raising the annealing temperature usually helps, and so does lowering the magnesium, primer or template amount, using fewer cycles, or switching to a hot-start enzyme. Gradient PCR finds the best annealing temperature quickly. It runs one sample at a range of annealing temperatures across the block of the thermal cycler in a single run. Touchdown PCR starts with an annealing temperature above the Tm of the primers and lowers it by about 1 degree each cycle. The first copies made are the specific ones, and they then outcompete nonspecific products. A faint band near the bottom of the gel, usually under 100 base pairs, is a primer-dimer. Redesigning primers that pair at their three prime ends, lowering the primer amount and using a hot-start enzyme all reduce primer-dimers. A smear usually comes from too much template, too many cycles or degraded DNA. A band in the no-template control means contamination, most often from earlier PCR products in the lab. To prevent it, set up reactions in a separate pre-PCR area, never open product tubes there, use filter tips, and split reagents into small aliquots. Each of these gel patterns points to a cause, and each cause has a direct change to the reaction or the program.

18:47 Recap and question

Here is what we covered in this lesson. PCR was invented to make many copies of one gene, and it became practical once Taq polymerase from a Yellowstone bacterium survived the heating steps. The reaction relies on DNA polymerase, which extends a primer from its three prime end, so a tube with template, two primers, dNTPs, Taq and magnesium can copy a chosen target. Each cycle separates the strands at about 95 degrees, anneals the primers a few degrees below their melting temperature, and extends them at 72 degrees, so the target doubles with each cycle until the reaction reaches a plateau. Well-designed primers make the product specific, the choice of polymerase sets its accuracy, and a gel with its controls shows whether the right product formed. Real-time PCR follows the product as it grows, so the Ct value shows the starting amount, and the common failures on a gel each point to a fix. Here is a question to test what you learned. In a real-time PCR run, sample A crosses the threshold at cycle 22 and sample B crosses it at cycle 25. Which sample started with more target DNA, and roughly how many times more? Please write your answer and your reasoning in the comments. Thank you for following this lesson on the polymerase chain reaction.

Read more

All videos