DNA ligases play an essential role in DNA replication, repair, and recombination and are classified as ATP-dependent or NAD-dependent based on their adenylation cofactor requirement. ATP-dependent ligases have been found in bacteriophages, viruses, archaea, eukaryotes and bacteria. NAD-dependent ligases exist in eubacteria and entomopoxviruses. Both types of ligases share a catalytic core which consists of conserved motifs found in nucleotidyl transferase superfamily. These motifs are essential for adenylation cofactor binding, metal ion coordination, and ligation chemistry, as validated by X-ray crystal structure determination and mutational analyses. Viral ATP ligases are generally small in size and have been used as models for elucidating the catalytic mechanism. More complex genomes, such as yeast and humans, encode multiple ATP ligases with additional non-catalytically essential domains at N-terminal and / or C-teminal regions of the catalytic core. Some of these additional domains are involved in protein-protein interactions, which target different ATP ligases to specific loci to fulfill specific cellular replication and repair functions. As a consequence of multiple protein interactions, ligase activity may be enhanced to complete a specific strand joining reaction. In humans, three ligase genes encoding four ligases (ligase I, ligase IIIα, ligase IIIβ, and ligase IV) are required for replication, recombination, repair, and nonhomologous end-joining (NHEJ). In addition to ligation activities, other enzymatic activities such as AP lyase activity have been detected in ATP ligases, implicating an expanded role of ligases in DNA repair. The discovery of ATP ligases in eubacteria indicates their ubiquity in living organisms. NAD ligases are composed of a nucleotidyl transferase catalytic core and additional C-terminal domains. The discovery of NAD ligases in entomopox virus genomes changes the notion that NAD ligases are only associated with eubacteria. While NAD and ATP ligases complement each other in rescuing growth defects, specific cellular functions such as DNA repair appear to require proteinprotein interactions specified by domains associated with NAD or ATP ligases. A ligase-DNA complex structure is needed to better understand the recognition and catalytic mechanism during DNA strand joining. The complex protein-protein interactions have to be elucidated to define the cellular functions. Newly identified putative ATP ligases in bacteria may open a new dimension in DNA ligation.