Articles

Cloning PCR Products: TA, Blunt-End and Phosphorylation

Taq 3' A-tailing, proofreading blunt ends, 5' phosphorylation with T4 PNK, and orientation screening

Molecular Cloning

Cloning PCR Products: TA, Blunt-End and Phosphorylation

Sometimes you only need a PCR product in a stable plasmid: for sequencing, for storage, or as a starting point for subcloning later. Designing restriction sites or assembly overlaps for a clone like this costs more effort than it returns.

Cloning the product directly into a vector is the short route. There are two ways to do it, they use different polymerases, and both depend on a phosphorylation step.

This guide covers TA cloning, blunt-end cloning, how to choose between them, and the phosphate problem that is easy to miss.

At a glance

  • TA cloning: Fast and efficient, but the kits are expensive and it needs a Taq product.
  • Blunt-end cloning: Cheap and works with high-fidelity products, but much less efficient.
  • Watch for: Both are non-directional, so design your screening primers before you start.

TA cloning

TA cloning uses a quirk of standard Taq polymerase. Taq has no proofreading activity, and it tends to add a single adenine to the 3' ends of its products, leaving a one-base A overhang.

You mix the product with a linearized vector carrying matching single thymine overhangs. The ends pair up, and T4 DNA ligase closes the circle.

The advantages are speed and efficiency. A single-base overhang is not much, but it is enough to hold the fragments together while the ligase works, and TA ligations are far more efficient than blunt ones.

The disadvantage is cost. Commercial kits such as pGEM-T are expensive per reaction, and using them for routine work uses up a budget quickly. Thermo Fisher also sells a topoisomerase-based version, TOPO TA, covered in our TOPO Cloning guide.

Because Taq's error rate is high, keep TA cloning for archiving fragments and preparing sequencing templates. For anything you intend to express, amplify with a high-fidelity polymerase instead and use one of the other routes.

Blunt-end cloning

If you are cloning a gene for expression, use a proofreading polymerase such as Q5, Phusion or PrimeSTAR. Their 3'-to-5' exonuclease activity removes overhanging bases, so these enzymes leave blunt ends. You can clone blunt fragments into any vector cut with a blunt-cutting enzyme such as EcoRV or SmaI.

This is the cheap option, and the main reason to choose it. You can prepare your own blunt vectors from standard plasmids with inexpensive enzymes, so there is no kit to buy and nothing that runs out.

The trade-off is efficiency. Blunt ends have no overhangs to hold the fragments together while the ligase works, so the ligation is far less efficient than with sticky ends and needs more attention to the molar ratio, the PEG in the buffer and the incubation. Expect to screen more colonies.

Phosphorylation: the step people miss

T4 DNA ligase needs a 5' phosphate to form a phosphodiester bond. Standard commercial primers do not carry one, so a normal PCR product is unphosphorylated at both ends.

This is fine on its own. It stops being fine the moment you dephosphorylate the vector to prevent self-ligation, because then neither partner has a phosphate and nothing can be joined at all. You get a bare plate and nothing on it to tell you why.

Two ways to supply the missing phosphate:

  • Order phosphorylated primers: Suppliers will add a 5' phosphate during synthesis, at a premium. Worth it if you will use the primers repeatedly.
  • Treat the product with T4 polynucleotide kinase: PNK and ATP, typically for 30 minutes. Check your protocol: PNK works best on purified product, and some ligase buffers already supply the ATP it needs, which saves a separate addition.

The rule to remember is that one of the two partners has to carry the phosphates. If you dephosphorylate the vector, phosphorylating the insert is not optional.

Plan it first

Even an archiving clone deserves a few minutes of planning, because both of these methods are non-directional. Your insert has an equal chance of going in forwards or backwards, and half your correct clones will be the wrong way round.

For storage the orientation makes no difference. If you later want to express from the clone, or use it as a template with a direction-dependent primer, an inverted insert will not work.

In Photo51, you can build both orientations before you start (see also: TA Cloning in Photo51). The point is not to discover which one you will get, since you will get both. It is to design the colony PCR primers that will tell them apart, so you find the right clone during screening rather than after you have paid for sequencing.

Troubleshooting

No colonies from a TA cloning reaction

Symptom: You set up a standard TA reaction and the plates are bare.

Cause: You amplified with a proofreading polymerase. High-fidelity enzymes leave blunt ends, and a blunt end cannot pair with the vector's T overhang.

Fix: You can rescue the product by A-tailing it. Purify the blunt product to remove the proofreading enzyme, then incubate it with standard Taq, Taq buffer and dATP at 72 °C for 15 to 30 minutes. Taq adds the A overhangs, and the product is then compatible with your kit. Check the composition of commercial mixes too, since some long-range mixes contain Taq and behave unpredictably here.

Plenty of colonies from a blunt ligation, and all of them empty

Symptom: You get hundreds of colonies, and every colony PCR shows an empty vector.

Cause: A blunt vector closes back on itself far more readily than it accepts a blunt insert. Both ends are identical, and there is no overhang to favor the insert.

Fix: Treat the vector with an alkaline phosphatase (such as CIP or rSAP) to remove its 5' phosphates so it cannot self-ligate. Then phosphorylate your insert with T4 PNK. Doing only the first half of this leaves nothing phosphorylated and you will get no colonies at all, which is the failure one step along from this one.

The clone is correct and the insert is backwards

Symptom: Your insert is present and the sequence is right, but the gene faces the wrong way relative to the promoter.

Cause: Non-directional cloning, working exactly as expected. About half your correct clones will be inverted.

Fix: Screen more colonies, and screen them better. Instead of two universal plasmid primers, use one primer in the vector backbone and one inside your insert. This pair only gives a product when the insert is the right way round, so a single gel tells you which colonies to grow up.