br Results and discussion In the present study we used
Results and discussion In the present study, we used a set of compounds (Fig. 2) consisting of ATP (a), dATP (b), an ATP analog containing β,γ-nonhydrolyzable hypophosphate P-P bond (c), and three pairs of P-diastereoisomers of α-thio-ATP analogs (d,e,f) containing a sulfur tgf beta receptor inhibitor instead of one of the non-bridging oxygen atoms at the α-phosphate group. Notably, the oxygen→sulfur replacement creates a new stereogenic center at the P-atom. Thus, both (RP and SP) diastereomers of ATPαS (d), dATPαS (e) and β,γ-hypo-ATPαS (f) were used. Since P-diastereomers may interact with the enzyme in a stereodependent manner the use of d-f ATP analogs offered a possibility to correlate the stereochemistry of the PS-stereogenic center as well as the presence of β,γ-hypo-phosphate modification on the T4 DNA ligase activity. Because β,γ-hypo-ATP (oxo) (c) and β,γ-hypo-ATPαS (f) are not commercially available we synthesized them. The β,γ-hypo-ATPαS (f) was obtained based on an oxathiaphospholane approach developed in our laboratory , ,  where hypophosphate salt reacts with 1,3,2-oxathiaphospholane (Scheme 1). Two diastereomers of the β,γ-hypo-ATPαS were separated using RP-HPLC and assigned as “fast” and “slow” in respect to their chromatographic mobility. The β,γ-hypo-ATP (oxo) (c) was obtained by selective oxidation of thio derivative (f) with iodoxybenzene (PhIO2). The β,γ-hypo-ATP can be also prepared by Michelson’s anion exchange approach , .
Experimental model In order to facilitate ligation product analysis, a set of three oligonucleotides was used (Fig. 3). A 30-nucleotide long oligomer served as a template for the ligation of two shorter single stranded DNA fragments (a 20 and 15nt strands) giving rise to a 35nt product with a 5-nt overhang. Due to the size difference, the post-reaction components could be separated by electrophoretic analysis (20% polyacrylamide/7M urea gel electrophoresis, and Stains-All detection).
Conclusions The aim of this research was to determine whether the configuration at the α-phosphate of ATPαS is important for T4 DNA ligase activity. Through the use of β,γ-hypo-ATP derivatives, we assessed the impact of the β,γ-pyrophosphate moiety in ATP and ATPαS analogs on enzyme activity. For this purpose a series of ATP analogs; β,γ-hypo-ATP and pure P-epimers of β,γ-hypo-ATPαS were synthesized and used as potential cofactors in the model DNA ligation reaction. Thus, the influence of separate diastereomers of ATPαS on the enzymatic activity of T4 DNA ligase was tested for the first time. The correlation between the β,γ-phosphorus linkage (hypophosphate) is also reported here. Based on the PAGE studies and structural analysis we have determined that only β,γ-hypo-ATP (oxo) and the SP-epimer of ATPαS are efficient cofactors for T4 DNA ligase, while the RP-epimer and both epimers of β,γ-hypo-ATPαS are not. None of the tested ATP analogs exerted inhibition activity towards this enzyme. In summary, we have demonstrated for the first time that commonly used T4 DNA ligase exhibit stereoselective properties towards modified ATP analogs.
Acknowledgments This research was supported by statutory funds of CMMS PAS in Lodz. Authors want to thank Ms. Barbara Mikolajczyk for all the help in HPLC analysis and Prof. Piotr Guga for fruitful discussion.
Introduction DNA ligases play an essential role in maintaining genomic integrity by joining breaks in the phosphodiester backbone of DNA that occur during replication and recombination, and as a consequence of DNA damage and its repair. Three human genes, LIG1, LIG3 and LIG4 encode ATP-dependent DNA ligases. These enzymes have related catalytic regions that catalyze the same three-step ligation reaction but different flanking domains that mediate protein:protein interactions with different partners (Ellenberger and Tomkinson, 2008). While almost all eukaryotes have homologs of the LIG1 and LIG4 genes, the LIG3 gene is less widely distributed. Initially, it was thought that the LIG3 gene was restricted to vertebrates but, with the sequencing of more genomes, it has now been found in about 30% of eukaryotes, including members of 4 of the 6 ancestral eukaryotic groups (Simsek and Jasin, 2011). This distribution suggests that the LIG3 gene arose relatively early during the evolution of eukaryotes but was not always retained. As eukaryotes became more complex, it was presumably advantageous to have multiple LIG genes that encoded a broader repertoire of DNA ligases to participate in the increasing number of specialized DNA transactions, including immunoglobulin gene rearrangements in immune cells, meiosis and germ cell development, the use of poly(ADP-ribose) to signal DNA damage and the different DNA repair pathways in proliferating and terminally differentiated cells. Notably, the DNA ligases encoded by vertebrate LIG3 genes have acquired several conserved accessory domains that flank the core catalytic region during evolution (Simsek and Jasin, 2011). As discussed below, these domains play critical roles in dictating the multiple cellular functions of the DNA ligases encoded by the LIG3 gene in vertebrate DNA metabolism.