Ligation-based cloning
Vector and insert are cut so that each carries a short single-stranded overhang, usually 2–4 bases, and the complementary overhangs base-pair before the ligase seals them. The most efficient and most controllable form of ligation.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
Sticky-end
Vector and insert are cut so that each carries a short single-stranded overhang, usually 2–4 bases, and the complementary overhangs base-pair before the ligase seals them. The most efficient and most controllable form of ligation.
How it differs from the standard method
- Two different enzymes give two different overhangs, so the insert can only go in one way round and the vector cannot close on itself.
- Needs very little ligase: a fraction of a Weiss unit in 10–20 µL for a few hours at 16 °C.
- Compatible overhangs from different enzymes (for example BamHI and BglII) can be joined, at the cost of losing both sites.
- A single-enzyme cut leaves identical ends: dephosphorylate the vector or expect mostly empty clones.
1One enzyme, one chemical step
T4 DNA ligase does one thing: it joins a 5' phosphate to an adjacent 3' hydroxyl in a double-stranded molecule, using ATP to pay for the new phosphodiester bond. It repairs a nick.
2Why overhangs beat flat ends
A four-base 5' overhang base-pairs with its partner and holds the two molecules together for milliseconds to seconds at 16 °C. That is an eternity to an enzyme.
Sticky-end
Two different overhangs fix the orientation and make self-closing rare. The most efficient and the most controllable option.
Blunt-end
Joins anything to anything, including the vector to itself. Low efficiency, high background, random orientation.
TA / TOPO
Uses the untemplated A that Taq leaves on a PCR product. No design work, no sites, but not directional.
3Molar ratio, not mass ratio
The reaction cares about numbers of molecules. A 1.2 kb insert weighs about a third of what a 3.8 kb vector weighs per molecule, so equal masses are already a threefold molar excess of insert.
4The four plates that tell you what happened
A colony count on its own is uninterpretable. Three control ligations turn it into data: vector plus ligase with no insert measures how much vector closes on itself; vector plus insert with no ligase measures undigested parental plasmid carried through from the digest, which transforms far better than anything you ligated; and a known supercoiled plasmid measures whether the cells were competent at all. Read the controls before you pick a colony.
- Many colonies, many on the vector-only plate
- Self-closing background. Dephosphorylate the vector, cut with two different enzymes, or raise the insert excess.
- Many colonies, many on the no-ligase plate
- Undigested vector carried through. Go back to the digest and purify the cut vector.
- No colonies anywhere, including the competence plate
- The cells, the antibiotic or the recovery step, not the ligation.
- No colonies except the competence plate
- The ligation. Check that a phosphate is present at each junction, the ends are compatible and the temperature suited them.
Key terms
Practise these as flashcards →The two groups T4 ligase joins. One partner at each junction must supply the phosphate.
The historical unit of T4 ligase activity, defined by exchange of phosphate into ATP; suppliers also quote cohesive-end units, and 1 Weiss unit is roughly 60–70 of them.
Ratio of numbers of molecules, not masses: the mass ratio that gives it depends on the two lengths.
A cut vector closing on itself. Its two ends start in the same molecule, so it is the ligation's main competitor.
The single adenine Taq-type polymerases add to most product ends; the basis of TA cloning and absent from proofread products.
A vector supplied with topoisomerase I covalently attached to each end, so the energy for the bond is pre-stored and no ligase or ATP is needed.
PCR straight from a colony with primers flanking the insertion site: sizes the junction, but cannot see point mutations.
When to use it
- You already have restriction sites that suit the construct and want a cheap, well-understood join.
- You want a PCR product in a vector today, with no design work: TA or TOPO cloning takes minutes.
- You are making a library where a scar at the junction does not matter and the vector is cut with two enzymes.
- You need a blunt fragment (a filled-in end, a shotgun fragment) in a vector and cannot redesign the ends.
Compared with alternatives
Makes the ends you ligate. Most ligation failures are digest failures: a partial digest leaves singly cut vector that closes on itself.
Prefer it when you need several fragments, a defined junction with no scar, or there are no usable sites: overlaps replace the ends entirely.
What converts the circles you made into colonies; its efficiency sets how many colonies a given ligation can give.
Prefer it for changing a few bases in a plasmid you already have, instead of cutting and re-ligating.
Where blunt and TA inserts come from, and the source of the polymerase errors you have to sequence for.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
Vector and insert are cut so that each carries a short single-stranded overhang, usually 2–4 bases, and the complementary overhangs base-pair before the ligase seals them. The most efficient and most controllable form of ligation.
- Two different enzymes give two different overhangs, so the insert can only go in one way round and the vector cannot close on itself.
- Needs very little ligase: a fraction of a Weiss unit in 10–20 µL for a few hours at 16 °C.
- Compatible overhangs from different enzymes (for example BamHI and BglII) can be joined, at the cost of losing both sites.
- A single-enzyme cut leaves identical ends: dephosphorylate the vector or expect mostly empty clones.