PCR Primer Design

PCR can be easy to do, if someone designs the experiment and gives you all the ingredients. From that point, it's all pipetting. The greater challenge is to understand the process well enough that you could design your own experiment and use the technique to amplify a new region of DNA.

When designing a new PCR experiment, one  of the first considerations will be designing the primers. The binding of primers to the template determines where the polymerase will start copying DNA; if this step goes right, then the rest of the PCR steps should be easy. If this step goes wrong, there's no hope for your PCR.

A simplified example

On the polymerase chain reaction page, I showed a simple diagram of how PCR works. In that diagram, the nucleotides of the DNA are represented by a zippered line. Suppose we add some nucleotide sequences to the ends of these DNA strands:

Each strand of the starting template is the reverse complement of the other strand (the base pairs are complementary, and the 5'—3' direction is opposite). In order for the primer to anneal to the template, the primer must be the reverse complement of the template. Therefore, each primer sequence is exactly the same as the beginning (5') end of one of the two strands! This also means that each primer is the reverse complement of the 3' end of the other strand.

How to specify the primer sequences

If the DNA sequence shown above was real, and you wanted to copy it with PCR, you'd start by looking at the target sequence (the part you want to copy) and designing the primers. In the diagram above, I showed the nucleotide sequence of both the upper and lower strands, but when you look up a sequence in a database (which you'll be doing soon) or get a result from DNA sequencing, you only see the sequence of one strand. It's assumed that you are smart enough to know that the DNA is actually double-stranded, and that if you're given the sequence of one strand, you can determine the sequence of the other. So if we look up the imaginary target sequence shown above, it would look like this:

5' CGGGG..........TTTTA 3'

Sequences are always given from 5' to 3' by convention, so the 5' and 3' aren't usually shown. The dotted lines in the middle indicate that I'm leaving out part of the sequence; I'm only showing the regions where the primers bind. If you want to do this PCR, you would go online and order the appropriate custom-made primers. The primers are called oligonucleotides (oligos), because they are short stretches of nucleic acid. In our lab, we usually order from Thermo Fisher; you can see the order form here. To order your oligos, you go to the order form and type in your sequences:

First (left) primer: CGGGG

Second (right) primer: TAAAA

Notice that the first primer is exactly the same as the target sequence, and the second primer is the reverse complement of the target sequence. This will always be true. You enter the rest of the information, complete the order, and they'll synthesize it and ship the same day! And they're cheap!

This example is much too simple, for several reasons:

• Real PCR primers are longer, often from 20 to 40 nucleotides long.
• Real DNA sequences are more complex; in fact, a repetitive sequence like GGGGG would be difficult to copy correctly.
• Real PCR products are longer, often ranging from a hundred to thousands of nucleotides long.

So let's look at a more realistic example.

Primer sequences for a taste receptor gene

Here is an example based on the human TAS2R38 gene, which encodes a taste receptor protein. In Bio 6B (assuming that you're not taking the class online), you'll use PCR to amplify a segment of this gene from your own genome. The goal of that experiment will be to examine sequence differences from one person to another (see the experiment here).

Here is the complete sequence of the protein-coding region of the TAS2R38 gene from GenBank:

  1 ATGTTGACTC TAACTCGCAT CCGCACTGTG TCCTATGAAG TCAGGAGTAC ATTTCTGTTC
 61 ATTTCAGTCC TGGAGTTTGC AGTGGGGTTT CTGACCAATG CCTTCGTTTT CTTGGTGAAT
121 TTTTGGGATG TAGTGAAGAG GCAGGCACTG AGCAACAGTG ATTGTGTGCT GCTGTGTCTC
181 AGCATCAGCC GGCTTTTCCT GCATGGACTG CTGTTCCTGA GTGCTATCCA GCTTACCCAC
241 TTCCAGAAGT TGAGTGAACC ACTGAACCAC AGCTACCAAG CCATCATCAT GCTATGGATG
301 ATTGCAAACC AAGCCAACCT CTGGCTTGCT GCCTGCCTCA GCCTGCTTTA CTGCTCCAAG
361 CTCATCCGTT TCTCTCACAC CTTCCTGATC TGCTTGGCAA GCTGGGTCTC CAGGAAGATC
421 TCCCAGATGC TCCTGGGTAT TATTCTTTGC TCCTGCATCT GCACTGTCCT CTGTGTTTGG
481 TGCTTTTTTA GCAGACCTCA CTTCACAGTC ACAACTGTGC TATTCATGAA TAACAATACA
541 AGGCTCAACT GGCAGAATAA AGATCTCAAT TTATTTTATT CCTTTCTCTT CTGCTATCTG
601 TGGTCTGTGC CTCCTTTCCT ATTGTTTCTG GTTTCTTCTG GGATGCTGAC TGTCTCCCTG
661 GGAAGGCACA TGAGGACAAT GAAGGTCTAT ACCAGAAACT CTCGTGACCC CAGCCTGGAG
721 GCCCACATTA AAGCCCTCAA GTCTCTTGTC TCCTTTTTCT GCTTCTTTGT GATATCATCC
781 TGTGTTGCCT TCATCTCTGT GCCCCTACTG ATTCTGTGGC GCGACAAAAT AGGGGTGATG
841 GTTTGTGTTG GGATAATGGC AGCTTGTCCC TCTGGGCATG CAGCCATCCT GATCTCAGGC
901 AATGCCAAGT TGAGGAGAGC TGTGATGACC ATTCTGCTCT GGGCTCAGAG CAGCCTGAAG
961 GTAAGAGCCG ACCACAAGGC AGATTCCCGG ACACTGTGCT GA

The target sequence (the region you will try to amplify using PCR) is highlighted in yellow.

Here is the sequence of the entire 303-bp (base pairs, or nucleotide pairs) PCR product shown above:

      AACT GGCAGAATAA AGATCTCAAT TTATTTTATT CCTTTCTCTT CTGCTATCTG TGGTCTGTGC CTCCTTTCCT ATTGTTTCTG GTTTCTTCTG GGATGCTGAC TGTCTCCCTG GGAAGGCACA TGAGGACAAT GAAGGTCTAT ACCAGAAACT CTCGTGACCC CAGCCTGGAG GCCCACATTA AAGCCCTCAA GTCTCTTGTC TCCTTTTTCT GCTTCTTTGT GATATCATCC TGTGTTGCCT TCATCTCTGT GCCCCTACTG ATTCTGTGGC GCGACAAAAT AGGGGTGATG GTTTGTGTT

The primer binding regions are highlighted in orange. To show how the primers fit this template sequence, here is a version of the PCR product showing only the primer binding regions, with the rest of the nucleotides represented by dots:

5' AACTGGCAGAATAAAGATCTCAATTTAT..........AAAATAGGGGTGATGGTTTGTGTT 3'

As usual, this nucleotide sequence shows only one of the two strands of the DNA. It's easier to visualize the primers if you see the sequence as double-stranded, like this:

5' AACTGGCAGAATAAAGATCTCAATTTAT..........AAAATAGGGGTGATGGTTTGTGTT 3'
3' TTGACCGTCTTATTTCTAGAGTTAAATA..........TTTTATCCCCACTACCAAACACAA 5'

The primer sequences used in this experiment, shown in red, are: 

Forward Primer 5′ AACTGGCAGAATAAAGATCTCAATTTAT 3′

Reverse Primer 5′ AACACAAACCATCACCCCTATTTT 3′.

Take a moment to study how the primers relate to the template sequence. Each primer is the reverse complement of one of the strands of DNA and identical to the other strand. This example is like the simplified one at the top of the page, but with more realistic sequences.

Keep in mind that primer sequences, like other nucleotide sequences, are normally given from 5' to 3'.  The template DNA is double-stranded, but the primers are single-stranded.

Ready for a quiz?

Soon I will give you a quiz on this, and one of the questions will ask you to determine the correct primer sequences for a given target sequence. The way to answer that quiz question is simple:

• First (left) primer: Copy the start of the target sequence. For the quiz, I'd have to tell you how many nucleotides long the primer should be. (This is sometimes referred to as the forward primer.)
• Second (right) primer: take the reverse complement of the 3' end of the target sequence. (This is sometimes referred to as the reverse primer.)

How to figure out the reverse complement: I hope you can do this yourself, following the example shown above. It's easy to make a mistake, though; if I was doing it, I'd probably use the Reverse Complement tool at Bioinformatics.org.

Keep in mind, I wouldn't be giving you such explicit instructions if I hadn't seen a lot of students get it wrong!

Designing better primers

The example above turns out to be too simple.

What determines the target sequence?

As a researcher, you can choose whatever target sequence you want. In the taste receptor experiment shown above, the primers were chosen to amplify a segment of the gene containing a polymorphism, or a region that varies from person to person. In another experiment, the goal might be to copy an entire gene or a region of noncoding DNA that might reveal something about evolutionary history. The research question determines the target.

Potential problems

Even if you design primers that look like they should copy the correct target sequence, there are still things that could go wrong.

Primers might anneal to the wrong target

If you use the approach shown on this page, you can design primers that are a perfect match to your target sequence (assuming you know your target sequence). However, those primers might also turn out to be a near-perfect match to some other region of the genome, and that similarity could allow the primers to bind in the wrong place and let the wrong target be copied.

Primers might anneal to each other (or themselves)

There are several ways for primers to bind to each other instead of binding to the template, and it's important to analyze the sequences to avoid these pitfalls.

Annealing temperature

PCR is based on repeated cycles of three temperatures for three steps: denaturation, primer annealing, and extension. The denaturation temperature is determined by DNA in general, and it's usually in the range 92°–98° C.  The annealing temperature depends on the primer, and must be determined for each PCR. The extension temperature depends on the enzyme used; in our lab that's always Taq polymerase, and we use a 72° extension temperature.

Longer primers anneal more tightly than short ones, so a longer primer will have higher melting temperature (the temperature required to separate the primer from the template). Likewise, primers with more G-C base pairs (rather than A-T) have higher melting temperature. Thus, the melting temperature of a primer is determined by its length and GC content. The annealing temperature used in PCR must be lower than the melting temperature by 5° or so.

Each PCR has one annealing temperature (you can't have different temperatures for the two different primers), so it's important that the two primers have similar melting temperatures.

Primer design software

Given the complexity of primer design, most researchers use software to help design their primers. See the references section at the bottom of this page for some examples.

Regardless of the software, what matters is how the primers work in PCR. Ultimately, every primer design is an educated guess as to what will work; you'll need to get into the lab, perform the PCR, and analyze the results to know if it's working.

Review

Terms & concepts

• 5' and 3' ends of DNA
• Annealing temperature: temperature programmed into PCR machine. Slightly cooler than melting temperature. Each PCR cycle has one annealing temperature; can't be different for the two primers.
• Melting temperature: temperature required to separate primer from template. Unique for each primer.
• Oligonucleotide
• Reverse complement
• Target sequence

Review questions

1. Given a target sequence, you should be able to determine the appropriate primers to use. An actual target sequence might be hundreds or thousands of nucleotides long, so for a quiz I would probably abbreviate some of it by using dots. Make sure you use the specified number of nucleotides (commonly 20-40 nucleotides per primer, but I'd have to tell you how many). See below for more examples.

Another primer sequence example

DNA sequences can be presented in a few different ways, but they are always given from 5' to 3', unless stated otherwise. For example, suppose you want to copy the beginning of the TAS2R38 gene shown above. Here is a target sequence:

  1 ATGTTGACTC TAACTCGCAT CCGCACTGTG TCCTATGAAG TCAGGAGTAC ATTTCTGTTC
 61 ATTTCAGTCC TGGAGTTTGC AGTGGGGTTT CTGACCAATG CCTTCGTTTT CTTGGTGAAT
121 TTTTGGGATG TAGTGAAGAG GCAGGCACTG AGCAACAGTG ATTGTGTGCT GCTGTGTCTC
181 AGCATCAGCC GGCTTTTCCT GCATGGACTG CTGTTCCTGA GTGCTATCCA GCTTACCCAC
241 TTCCAGAAGT TGAGTGAACC ACTGAACCAC AGCTACCAAG CCATCATCAT GCTATGGATG
301 ATTGCAAACC AAGCCAACCT CTGGCTTGCT GCCTGCCTCA GCCTGCTTTA CTGCTCCAAG
361 CTCATCCGTT TCTCTCACAC CTTCCTGATC TGCTTGGCAA GCTGGGTCTC CAGGAAGATC
421 TCCCAGATGC TCCTGGGTAT TATTCTTTGC TCCTGCATCT GCACTGTCCT CTGTGTTTGG
481 TGCTTTTTTA GCAGACCTCA CTTCACAGTC ACAACTGTGC TATTCATGAA TAACAATACA

What primers should you use? Let's say they should be 25 nucleotides long in this case. The primers should look like this:

 5' ATGTTGACTC TAACTCGCAT CCGCA 3'
  1 ATGTTGACTC TAACTCGCAT CCGCACTGTG TCCTATGAAG TCAGGAGTAC ATTTCTGTTC
 61 ATTTCAGTCC TGGAGTTTGC AGTGGGGTTT CTGACCAATG CCTTCGTTTT CTTGGTGAAT
121 TTTTGGGATG TAGTGAAGAG GCAGGCACTG AGCAACAGTG ATTGTGTGCT GCTGTGTCTC
181 AGCATCAGCC GGCTTTTCCT GCATGGACTG CTGTTCCTGA GTGCTATCCA GCTTACCCAC
241 TTCCAGAAGT TGAGTGAACC ACTGAACCAC AGCTACCAAG CCATCATCAT GCTATGGATG
301 ATTGCAAACC AAGCCAACCT CTGGCTTGCT GCCTGCCTCA GCCTGCTTTA CTGCTCCAAG
361 CTCATCCGTT TCTCTCACAC CTTCCTGATC TGCTTGGCAA GCTGGGTCTC CAGGAAGATC
421 TCCCAGATGC TCCTGGGTAT TATTCTTTGC TCCTGCATCT GCACTGTCCT CTGTGTTTGG
481 TGCTTTTTTA GCAGACCTCA CTTCACAGTC ACAACTGTGC TATTCATGAA TAACAATACA
                                       3' ACACG ATAAGTACTT ATTGTTATGT 5'

First (left, forward or upstream) primer: 

5' ATGTTGACTCTAACTCGCATCCGCA 3'

You simply copy the first 25 nucleotides of the target sequence. This primer is identical to the target strand that's shown; it will be the reverse complement of the other strand.

Second (right, reverse, or downstream) primer:

This one will be the reverse complement of the target strand that's shown. The primer sequence is shown above, aligned with the target. You need to flip that sequence around so it's given 5' to 3':

5' TGTATTGTTATTCATGAATAGCACA 3'

One more primer sequence example

Here's one more, with the goal of pointing out that sequences can be presented in different ways. If you see a sequence like this, the dots mean that part of the sequence isn't shown.

AGCATCAGCCGGCTTTTCCTGCATGGACTGCTGTT..........
..........GTGCTATCCAGCTTACCCACTTCCAGAAGTTGAGTG

It's not labeled with 5' and 3', but you should read it like this:

5' AGCATCAGCCGGCTTTTCCTGCATGGACTGCTGTT..........
..........GTGCTATCCAGCTTACCCACTTCCAGAAGTTGAGTG 3'

So the primers (assuming that we need them to be 25 nucleotides long) would be:

5' AGCATCAGCCGGCTTTTCCTGCATG 3'
5' AGCATCAGCCGGCTTTTCCTGCATGGACTGCTGTT..........
..........GTGCTATCCAGCTTACCCACTTCCAGAAGTTGAGTG 3'
                  3' GAATGGGTGAAGGTCTTCAACTCAC 5'

So the left and right primers would be:

<p">Left: 5' AGCATCAGCCGGCTTTTCCTGCATG 3' <p">Right: 5' CACTCAACTTCTGGAAGTGGGTAAG 3'

References & further reading

Videos

How to Design Primers for PCR from Addgene. Primer design is a subtle art. Even if you're not going to design and use your own primers, this video will help you understand what makes PCR work (or not).

Reading

PCR Primer Design Guidelines from Premier Biosoft. Describes the process of primer design, including potential problems.

Sequence tools

Reverse Complement. This simple web page lets you flip nucleotide sequences. The the default is to give you the reverse complement of the sequence you enter, but you don't want that for primers. Instead, enter your sequence (from 5' to 3'), select "reverse" and it will give you the same sequence backward (3' to 5').  This saves a little tedium and reduces the likelihood that you'll make a mistake.

Benchling. A suite of browser-based sequence analysis tools. If you sign up for a free account, you can use this package to analyze primer sequences, translate from DNA to protein sequences, perform virtual restriction digests and cloning, and many other functions. I use this all the time. Highly recommended.

PrimerBLAST from NCBI. Helps you determine whether your primers will anneal to the wrong target. Part of the powerful (but not always user-friendly) collection of resources from the National Center for Biotechnology Information, NCBI.

13 Free PCR Primer Design Programs from Tip Top Bio.