The spore photoproduct lyase is a radical SAM enzyme, which repairs 5-(a-thyminyl)-5,6-dihydrothymidine. Here we show that the enzyme establishes a complex radical transfer cascade and creates a cysteine and a tyrosyl radical dyade to establish repair. This allows the enzyme to solve topological and energetic problems associated with the radical based repair reaction.UV irradiation causes the formation of a variety of dinucleotide lesions, which are typically formed between two pyrimidines.
1,2In cellular DNA, UV irradiation causes the formation of the well studied cyclobutane pyrimidine dimers (CPDs), pyrimidine (6-4)-pyrimidone photoproducts (6-4PPs), and their Dewar-valence isomers.1 In bacterial endospores, in contrast, these bipyrimidine photoproducts are only formed in low amounts. Due to the unusual packing of the DNA in spores, UV irradiation was found to create selectively the (5R)-5-(a-thyminyl)-5,6-dihydrothymidine lesion (spore photoproduct or SP-lesion). [3][4][5] This dimeric thymidine-photoproduct is repaired during germination by the spore photoproduct lyase (SPL), which is a radical SAM enzyme.
6The SPL catalyzes the direct repair of the SP-lesion, utilizing a radical mediated mechanism (Fig. 1A) ( Fig. 1B) shows the topological problem associated with closing the catalytic cycle. While the allyl radical is situated and reduced at the 3 0 -side, transfer of the radical center back to the 5 0 -dAdoH requires moving the radical back to the 5 0 -part in the active side over a distance of roughly 10 Å. This creates next to a topological problem also an energetic obstacle, because regeneration of the adenosyl radical by the thiyl radical would be endothermic by 62 kJ mol
À1.
16Here we provide first evidence that the enzyme uses a further tyrosyl radical intermediate to solve the energetic and topological problem. In the crystal structure the tyrosine bridges the conserved cysteine and the 5 0 -dAdoH. The structure shows that Tyr98 is located 3.6 Å and 5.1 Å away from both centers, respectively. In order to clarify the role of Cys140 and Tyr98 we prepared two mutants C140A and Y98F ( Fig. 2A) and determined their catalytic activity. To this end we quantified product formation after 3 h and after reaction overnight. The measured K m and V max values are listed in Table 1. The data show that both mutants have a strongly reduced catalytic activity. The Y98F mutant shows in addition a reduced substrate binding affinity, which indicates that the phenolic hydroxyl group is important to organize the substrate in the active site. Direct interactions with the lesion are, however, not observed in the crystal structure. For the catalytically competent wild type (wt) enzyme we measured a ratio of product/5 0 -dAdoH of greater than two indicating that the cleavage of one SAM molecule repairs more than one lesion (turnover). The mutants in contrast use more than one SAM per product showing that here the turnover is smaller than one product molecule (0.7 and 0.25, Fig. 2B) formed per SAM. This result shows that Cys1...