The computational study of the interaction of POT1 with SSDNA and TPP1

Abstract

Linear DNA sequences in eukaryotes have telomere sequences at the chromosome ends. In humans, telomeres consists of repeating TTAGGG sequences. Shelterin complex mainly interacts with telomere sequences. It interacts with both telomeric dsDNA and ssDNA. It protects it from DNA repair mechanism. POT1 is one of the members of shelterin complex. It interacts with telomeric ssDNA. It controls the length of the telomere sequence by forming complex with TPP1. The N-terminal domain of POT1, POT1N, binds to ssDNA while C terminal domain of POT1, POT1C, is known to interact with TPP1. The interaction between POT1 and ssDNA depends on many factors. In the shorter sequences, POT1 prefers to bind 3' end of the sequence by 10 fold. The preference changes as TPP1 binds to POT1. POT1 slides on ssDNA after forming complex with TPP1. There is no information about the mechanism of action in the literature. TPP1 interacts with POT1 and increases its affinity to ssDNA. The affinity of POT1 on ssDNA increases after interacting with POT1 and POT1-TPP1 interaction is regulated by cell cycle as well. It is likely that the POT1-TPP1 complex has at least 2 different conformations and these conformations have different interactions, conformations and geometries. There is no information about these structures in the literature. The preference of POT1 on ssDNA, potential initial binding and sliding mechanisms, POT1-TPP1 complex and the allosteric effect of TPP1 were investigated by standard Molecular Dynamics (MD) simulations and Replica Exchange Molecular Dynamics (REMD) simulations. Free energy of the system was calculated by Molecular Mechanic/Poisson-Boltzmann Surface Area (MM/PBSA) method. Another candidate for the mechanism was investigated by Potemtial of Mean Force (PMF) method. According to our results, the binding preference of POT1 depends on the length of ssDNA. In a longer DNA sequence (22 nucleotides), POT1 has the highest abundance in the middle of sequence, ~27%, followed by 5'end, ~ 17% and 3' end ~ 16%. In a shorter DNA sequence (16 nucleotides), POT1 prefers binding to the 3' end over 5' end (~32% and ~22%). This results shows that the preference of POT1 is influenced by ssDNA length. Our results indicate that extra nucleotide in the longer ssDNA simulations form secondary structure with each other and this might explain 5' end preference over 3' end. In the shorter ssDNA simulations, 3' end is loosely bound to POT1 ~23% of the time and ~8% in the solvent, while 5' end has ~8% protein bound and ~14% in the solvent and this explains 3' preference in the shorter ssDNA simulations. Our results show that there are multiple potential initial binding and sliding mechanisms. Most of the time, POT1 slides only 1 OB fold at a time while other OB fold conserves its position. The sliding happens per nucleotide in this mechanism. The relative abundance of these OB shifted structures were ~ 21%. Here, OB2 fold shifted structures had higher abundance than OB1 fold shifted ones which confirms the literature. Another potential mechanism involves the repetetive nature of telomere sequence. POT1 might prefer to skip a telomere repeat and OB fold can bind to the next telomere sequence. When OB1 fold binds to telomere sequence, OB2 fold moves 6 nucleotides and then OB1 folds shall slide on ssDNA. Furthermore, when ssDNA was interacted with only OB1 fold or OB2 fold, it clearly showed that OB1 fold prefers to bind ssDNA. The relative abundance of these structures were ~4.5% and ~1.5%, respectively. When whole sequence was shifted by 1 nucleotide, which is considered a potential mechanism, the structure has the lowest abundance. In order to test this for 2, 3, 4 and 5 nucleotides, 5 new initial structure was created. In these structures, ssDNA was 11 nucleotide long. The crystal structure and 6 new structures were simulated. G results show that all of these structures are not a candidate for a potential intermediate structures during the sliding. There are 2 different complexes for POT1-TPP1. In the first complex, it should slide easily on ssDNA and it should have less specific interactions. In second complex, it should bind it tightly and have specific interactions. Both structures were constructed based on O. nova. In the simulations, they survived more than 1 µs.