Ligation is the joining of two nucleotides, or two nucleic acid fragments, into a single polymeric chain through the action of an enzyme known as a ligase. The reaction involves the formation of a phosphodiester bond between the 3'-hydroxyl terminus of one nucleotide and the 5'-phosphoryl terminus of another nucleotide, which results in the two nucleotides being linked consecutively on a single strand. Ligation works in fundamentally the same way for both DNA and RNA. A cofactor is generally involved in the reaction, usually ATP or NAD+. Eukaryotic ligases belong to the ATP type, while the NAD+ type are found in bacteria (e.g. E. coli).
Ligation occurs naturally as part of numerous cellular processes, including DNA replication, transcription, splicing, and recombination, and is also an essential laboratory procedure in molecular cloning, whereby DNA fragments are joined to create recombinant DNA molecules (such as when a foreign DNA fragment is inserted into a plasmid). The discovery of DNA ligase dates back to 1967 and was an important event in the field of molecular biology. Ligation in the laboratory is normally performed using T4 DNA ligase. It is broadly used in vitro due to its capability of joining sticky-ended fragments as well as blunt-ended fragments. However, procedures for ligation without the use of standard DNA ligase are also popular. Human DNA ligase abnormalities have been linked to pathological disorders characterized by immunodeficiency, radiation sensitivity, and developmental problems.
Ligation reaction The mechanism of the ligation reaction was first elucidated in the laboratory of I. Robert Lehman. Two fragments of DNA may be joined by DNA ligase which catalyzes the formation of a phosphodiester bond between the 3'-hydroxyl group (-OH) at one end of a strand of DNA and the 5'-phosphate group (-PO4) of another. In animals and bacteriophages, ATP is used as the energy source for the ligation, while in bacteria, NAD+ is used.
The DNA ligase first reacts with ATP or NAD+, forming a ligase-AMP intermediate with the AMP linked to the ε-amino group of lysine in the active site of the ligase via a phosphoramide bond. This adenylyl group is then transferred to the phosphate group at the 5' end of a DNA chain, forming a DNA-adenylate complex. Finally, a phosphodiester bond between the two DNA ends is formed via the nucleophilic attack of the 3'-hydroxyl at the end of a DNA strand on the activated 5′-phosphoryl group of another.
A nick in the DNA (i.e. a break in one strand of a double-stranded DNA) can be repaired very efficiently by the ligase. However, a complicating feature of ligation conducted presents itself when ligating two separate DNA ends as the two ends need to come together before the ligation reaction can proceed. In the ligation reactions conducted in a laboratory, the ligation of DNA with sticky or cohesive ends, the protruding strands of DNA may be annealed together already, therefore it is a relatively efficient process as it is equivalent to repairing two nicks in the DNA. However, in the ligation of blunt-ends, which lack protruding ends for the DNA to anneal together, the process is dependent on random collision for the ends to align together before they can be ligated, and is consequently a much less efficient process. The DNA ligase from E. coli cannot ligate blunt-ended DNA except under conditions of molecular crowding, and it is therefore not normally used for ligation in the laboratory. Instead the DNA ligase from phage T4 is used as it can ligate blunt-ended DNA as well as single-stranded DNA.
Factors affecting ligation In the laboratory, factors that affect an enzyme-mediated chemical reaction would naturally affect a ligation reaction, these include the concentration of enzyme and the reactants, the temperature of reaction and the length of time of incubation. Ligation is complicated by the fact that the reaction can involve both inter- and intra-molecular reactions, but the desired ligation products in many ligation reactions (e.g. ligating a DNA fragment into a vector) needed first to be inter-molecular, i.e. between two different DNA molecules, followed by an intra-molecular reaction to seal and circularize the molecule. For efficient ligation, an additional annealing step is also necessary.
The three steps to form a new phosphodiester bond during ligation are: enzyme adenylylation, adenylyl transfer to DNA, and nick sealing. Mg(2+) is a cofactor for catalysis, therefore at high concentration of Mg(2+) the ligation efficiency is high. If the concentration of Mg(2+) is limited, the nick- sealing is the rate- limiting reaction of the process, and adenylylated DNA intermediate stays in the solution. Such adenylylation of the enzyme restrains the rebinding to the adenylylated DNA intermediate comparison of an Achilles' heel of LIG1, and represents a risk if they are not fixed.
DNA concentration The concentration of DNA can affect the rate of ligation, and whether the ligation is an inter-molecular or intra-molecular reaction. Ligation involves joining up the ends of a DNA with other ends, however, each DNA fragment has two ends, and if the ends are compatible, a DNA molecule can circularize by joining its own ends. At high DNA concentration, there is a greater chance of one end of a DNA molecule meeting the end of another DNA, thereby forming intermolecular ligation. At a lower DNA concentration, the chance that one end of a DNA molecule would meet the other end of the same molecule increases, therefore intramolecular reaction that circularizes the DNA is more likely. The transformation efficiency of linear DNA is also much lower than circular DNA, and for the DNA to circularize, the DNA concentration should not be too high. As a general rule, the total DNA concentration should be less than 10 μg/ml.
The relative concentration of the DNA fragments, their length, as well as buffer conditions are also factors that can affect whether intermolecular or intramolecular reactions are favored.
The concentration of DNA can be artificially increased by adding condensing agents such as cobalt hexamine and biogenic polyamines such as spermidine, or by using crowding agents such as polyethylene glycol (PEG) which also increase the effective concentration of enzymes. Note however that additives such as cobalt hexamine can produce exclusively intermolecular reaction, resulting in linear concatemers rather than the circular DNA more suitable for transformation of plasmid DNA, and is therefore undesirable for plasmid ligation. If it is necessary to use additives in plasmid ligation, the use of PEG is preferable as it can promote intramolecular as well as intermolecular ligation.
Ligase concentration As is usual for an enzyme, the higher the ligase concentration, the faster is the rate of ligation. Blunt-end ligation is much less efficient than sticky end ligation, so a higher concentration of ligase is used in blunt-end ligations. High DNA ligase concentration may be used in conjunction with PEG for a faster ligation, and they are the components often found in commercial kits designed for rapid ligation.
Temperature Two issues are involved when considering the temperature of a ligation reaction. First, the optimum temperature for DNA ligase activity which is 37°C, and second, the melting temperature (Tm) of the DNA ends to be ligated. The melting temperature is dependent on length and base composition of the DNA overhang—the greater the number of G and C, the higher the Tm since there are three hydrogen bonds formed between G-C base pair compared to two for A-T base pair—with some contribution from the stacking of the bases between fragments. For the ligation reaction to proceed efficiently, the ends should be stably annealed, and in ligation experiments, the Tm of the DNA ends is generally much lower than 37°C. The optimal temperature for ligating cohesive ends is therefore a compromise between the best temperature for DNA ligase activity and the Tm where the ends can associate. However, different restriction enzymes generates different ends, and the base composition of the ends produced by these enzymes may also differ, the melting temperature and therefore the optimal temperature can vary widely depending on the restriction enzymes used, and the optimum temperature for ligation may be between 4-15°C depending on the ends. Ligations also often involve ligating ends generated from different restriction enzymes in the same reaction mixture, therefore it may not be practical to select optimal temperature for a particular ligation reaction and most protocols simply choose 12-16°C, room temperature, or 4°C. When conducting a ligation at 4°C, it is necessary to increase the time of ligation reaction, for example by leaving the ligation mixture overnight or longer in the fridge.
Buffer composition The ionic strength of the buffer used can affect the ligation. The kinds of cations presence can also influence the ligation reaction, for example, excess amount of Na+ can cause the DNA to become more rigid and increase the likelihood of intermolecular ligation. At high concentration of monovalent cation (>200 mM) ligation can also be almost completely inhibited. The standard buffer used for ligation is designed to minimize ionic effects.
Sticky-end ligation Restriction enzymes can generate a wide variety of ends in the DNA they digest, but in cloning experiments most commonly used restriction enzymes generate a 4-base single-stranded overhang called the sticky or cohesive end (exceptions include NdeI which generates a 2-base overhang, and those that generate blunt ends). These sticky ends can anneal to other compatible ends and become ligated in a sticky-end (or cohesive end) ligation. EcoRI for example generates an AATT end, and since A and T have lower melting temperature than C and G, its melting temperature Tm is low at around 6°C. For most restriction enzymes, the overhangs generated have a Tm that is around 15°C. For practical purposes, sticky end ligations are performed at 12-16°C, or at room temperature, or alternatively at 4°C for a longer period.
For the insertion of a DNA fragment into a plasmid vector, it is preferable to use two different restriction enzymes to digest the DNA so that different ends are generated. The two different ends can prevent the religation of the vector without any insert, and it also allows the fragment to be inserted in a directional manner.
When it is not possible to use two different sites, then the vector DNA may need to be dephosphorylated to avoid a high background of recircularized vector DNA with no insert. Without a phosphate group at the ends the vector cannot ligate to itself, but can be ligated to an insert with a phosphate group. Dephosphorylation is commonly done using calf-intestinal alkaline phosphatase (CIAP) which removes the phosphate group from the 5′ end of digested DNA, but note that CIAP is not easy to inactivate and can interfere with ligation without an additional step to remove the CIAP, thereby resulting in failure of ligation. CIAP should not be used in excessive amount and should only be used when necessary. Shrimp alkaline phosphatase (SAP) or Antarctic phosphatase (AP) are suitable alternative as they can be easily inactivated.
Blunt-end ligation Blunt end ligation does not involve base-pairing of the protruding ends, so any blunt end may be ligated to another blunt end. Blunt ends may be generated by restriction enzymes such as SmaI and EcoRV. A major advantage of blunt-end cloning is that the desired insert does not require any restriction sites in its sequence as blunt-ends are usually generated in a PCR, and the PCR generated blunt-ended DNA fragment may then be ligated into a blunt-ended vector generated from restriction digest.