Science Wiki

DNA ligase (EC 6.5.1.1 for ATP-dependent ligases; EC 6.5.1.2 for NAD⁺-dependent ligases) is an essential class of nucleotidyltransferase enzymes that functions as the universal catalytic "molecular glue" of the cell, permanently joining adjacent DNA strands by catalyzing the formation of a covalent 3′→5′ phosphodiester bond between a 3′-hydroxyl (3′-OH) terminus and a juxtaposed 5′-phosphate (5′-PO₄²⁻) terminus. First discovered and purified independently in 1967 by the laboratories of Martin Gellert, Charles C. Richardson, Robert Lehman, and Jerard Hurwitz, DNA ligase is indispensable across all domains of life for the maturation of Okazaki fragments during DNA replication, the resolution of genetic recombination, and the terminal sealing step of DNA repair pathways.

The Three-Step Nucleotidyl Transfer Cascade

Because the direct condensation of a 3′-hydroxyl group with a 5′-monophosphate group is thermodynamically unfavorable in aqueous solution (ΔG°′ ≈ +20 to +22 kJ/mol), all DNA ligases couple phosphodiester bond synthesis to the cleavage of a high-energy phosphoanhydride cofactor (ATP or NAD⁺) in the presence of divalent magnesium ions (Mg²⁺). The reaction proceeds via three conserved, sequential covalent intermediates:

1. Step 1 — Enzyme Adenylylation (Lys-AMP Formation): The ε-amino group (-NH₂) of a strictly conserved catalytic lysine residue within the conserved KxDG motif of the ligase nucleotidyltransferase (NTase) active site launches a nucleophilic attack on the α-phosphorus of ATP or NAD⁺. This forms a covalent ligase–(lysyl-Nζ)–AMP phosphoramidate intermediate and releases inorganic pyrophosphate (PP_i, from ATP) or nicotinamide mononucleotide (NMN, from NAD⁺):

E-Lys + ATP (or NAD⁺) → E-Lys-AMP + PP_i (or NMN)

1. Step 2 — Adenylyl Transfer to the 5′-Phosphate (DNA-Adenylate Formation): Upon binding a nicked DNA duplex and encircling the helix via its OB-fold and DNA-binding domains, the enzyme transfers the AMP moiety from the catalytic lysine onto the exposed 5′-phosphate oxygen at the nick site. This generates an activated DNA-adenylate intermediate bearing an inverted 5′–5′ pyrophosphate linkage (AppDNA):

E-Lys-AMP + 5′-P-DNA → E-Lys + AppDNA

1. Step 3 — Nick Sealing and AMP Release (Phosphodiester Bond Formation): Coordinated by catalytic Mg²⁺ ions, the adjacent 3′-hydroxyl (-OH) group attacks the 5′-phosphorus of the AppDNA intermediate via inline S_N2 displacement, forming a native 3′–5′ phosphodiester bond and releasing free AMP as the leaving group:

DNA-3′-OH + AppDNA → DNA-3′-O-P-O-5′-DNA + AMP

Cofactor Divergence Across Domains of Life (ATP vs. NAD⁺)

While the three-step adenylylation chemistry and the core NTase–OB fold are universally conserved (sharing structural ancestry with eukaryotic mRNA capping enzymes), the identity of the adenylate donor strictly divides the domains of life:

• Eukaryotes, Archaea, and Bacteriophages/Viruses (EC 6.5.1.1): Utilize ATP (adenosine 5′-triphosphate) as their sole high-energy adenylate cofactor.
• Prokaryotes / Replicative Bacterial Ligases (EC 6.5.1.2): All eubacteria encode an essential replicative DNA ligase (LigA) that exclusively utilizes NAD⁺ (nicotinamide adenine dinucleotide) as its adenylate cofactor, recognized via a unique bacterial BRCA1 C-terminus (BRCT) and Ia sub-domain. Because human ligases cannot use NAD⁺ and show minimal sequence identity near the NMN-binding pocket, bacterial LigA represents a validated target for narrow- and broad-spectrum antibacterial drug discovery.

Okazaki Fragment Maturation with PCNA and FEN1

During cellular division, the antiparallel geometry of the double helix and the strict 5′→3′ directionality of DNA polymerase mandate discontinuous synthesis of the lagging strand as a series of short Okazaki fragments (~1,000–2,000 nt in bacteria; ~100–200 nt in eukaryotes, totaling >20 million fragments per human cell division).

Once RNase H, DNA polymerase I (in bacteria), or DNA polymerase δ paired with Flap Endonuclease 1 (FEN1) (in eukaryotes) excises the initiating RNA primers and fills the resulting gaps with deoxyribonucleotides, a single-stranded phosphodiester nick remains between the 3′-OH of the upstream fragment and the 5′-phosphate of the downstream fragment.

In eukaryotes, DNA ligase I is recruited directly to the replication factory via an N-terminal PCNA-interacting peptide (PIP box) that binds the homotrimeric PCNA sliding clamp encircling the DNA. DNA ligase I undergoes a large conformational switch from an extended state to a closed three-domain ring (DNA-binding domain, NTase domain, and OB-fold domain) that completely encircles the nicked B-DNA duplex, verifies Watson-Crick base-pair geometry on both sides of the nick, and seals the backbone into a continuous chromosome.

DNA Ligase III and Single-Strand Break / Mitochondrial Repair

The cellular genome sustains tens of thousands of lesions daily from reactive oxygen species, spontaneous hydrolysis, and ultraviolet or ionizing radiation. Vertebrates partition repair ligation across three specialized ATP-dependent isoforms:

• Single-Strand Breaks and Excision Repair (DNA Ligase III & I): Following removal of damaged bases in Base Excision Repair (BER) or bulky helix-distorting photoproducts in Nucleotide Excision Repair (NER), DNA ligase IIIα—stabilized via a C-terminal BRCT domain heterodimerization with the scaffold protein XRCC1 and recruited by PARP1—seals single-nucleotide (short-patch) nicks, while DNA ligase I seals PCNA-dependent long-patch repair tracts. Crucially, an alternative translation initiation site generates a mitochondrial targeting sequence on DNA ligase III, making *LIG3* the sole DNA ligase inside mammalian mitochondria and indispensable for embryonic survival.

DNA Ligase IV and Non-Homologous End Joining (NHEJ)

• Double-Strand Breaks and V(D)J Recombination (DNA Ligase IV): When ionizing radiation or replication fork collapse severs both strands of the DNA helix—or when RAG1/RAG2 endonucleases cleave antigen receptor loci during V(D)J recombination in developing B and T lymphocytes—the cell deploys Non-Homologous End Joining (NHEJ). DNA ligase IV, bound constitutively via tandem BRCT domains to the homodimeric coiled-coil scaffold XRCC4 and stimulated by XLF, is uniquely capable of ligating incompatible, single-stranded, or damaged ends across a synaptic double-strand break. Hypomorphic mutations in human *LIG4* cause LIG4 syndrome, characterized by radiosensitivity, microcephaly, growth retardation, and severe combined immunodeficiency (SCID).

Sticky-End vs. Blunt-End Ligation in Recombinant DNA

The purification of DNA ligase in 1967 and its pairing with restriction endonucleases by Paul Berg, Stanley Cohen, and Herbert Boyer in 1972–1973 launched recombinant DNA technology, biotechnology, and synthetic biology.

• Molecular Cloning and Adapter Ligation: In the laboratory, researchers cleave a vector plasmid and a target gene insert using type II restriction enzymes (or Golden Gate Type IIS assembly) to generate complementary single-stranded 5′ or 3′ overhangs ("cohesive" or "sticky" ends) bearing 5′-phosphate and 3′-hydroxyl groups.
• T4 DNA Ligase: The predominant enzyme used in modern genetic engineering and next-generation sequencing (NGS) library preparation is bacteriophage T4 DNA ligase. Unlike *E. coli* LigA, T4 DNA ligase is ATP-dependent and uniquely capable of efficiently catalyzing both cohesive-end ligation and blunt-end ligation (joining flush duplex ends lacking overhanging base pairs, typically enhanced in vitro by macromolecular crowding agents such as polyethylene glycol, PEG, at 16–25 °C). Conversely, thermostable NAD⁺-dependent ligases (such as *Taq* DNA ligase from *Thermus aquaticus*) that have zero blunt-end activity and high mismatch discrimination at 45–65 °C are used in the Ligase Chain Reaction (LCR) and Gibson isothermal assembly.

Why can't DNA ligase join two DNA fragments if the 5′ end has been treated with calf intestinal alkaline phosphatase (CIP/rSAP)?

DNA ligase strictly requires an intact 5′-monophosphate group (-PO₄²⁻) at the nick site so that it can transfer the AMP moiety from its catalytic lysine to form the reactive 5′–5′ DNA-adenylate (AppDNA) intermediate in Step 2 of catalysis. Treating a restriction-digested plasmid vector with alkaline phosphatase removes its terminal 5′-phosphate groups (leaving two 5′-OH ends), which prevents the vector from self-ligating without an insert while still allowing ligation to a phosphorylated foreign gene insert.

Why is bacterial NAD⁺-dependent DNA ligase (LigA) considered an ideal target for novel antibiotics?

All known bacteria rely on an essential replicative DNA ligase (LigA) that uses NAD⁺ (nicotinamide adenine dinucleotide) as its adenylate donor cofactor, whereas all human nuclear and mitochondrial DNA ligases (Ligase I, III, and IV) exclusively utilize ATP. Because the NMN-binding pocket of bacterial LigA is structurally absent from human ligases, small-molecule inhibitors targeting the NAD⁺ site of LigA selectively block bacterial DNA replication without inhibiting human cellular enzymes.

Why are blunt-end ligations with T4 DNA ligase typically performed at 16 °C or room temperature even though the enzyme's optimal catalytic temperature is 37 °C?

Cohesive ("sticky") overhangs of 2 to 4 base pairs have a very low thermal melting temperature (*T*_m ≈ 5–15 °C), and blunt ends have zero hydrogen-bonding attraction at all. At 37 °C, thermal kinetic motion rapidly dissociates transiently aligned DNA ends before T4 DNA ligase can bind and adenylate the junction; incubating at 16–22 °C represents a kinetic compromise that slows Brownian end-dissociation while maintaining sufficient enzymatic turnover.