Structural and Energetic Origin of Different Product Specificities and Activities for SETD3 and Its Mutants on the Methylation of the ß-Actin H73K Peptide: Insights from a QM/MM Study.

The methylation of the lysine residue can affect some fundamental biological processes, and specific biological effects of the methylations are often related to product specificity of methyltransferases. The question remains concerning how active-site structural features and dynamics control the activity as well as the number (1, 2, or 3) of methyl groups on methyl lysine products. SET domain containing protein 3 (SETD3) has been identified recently as the ß-actin histidine73-N3 methyltransferase, and also, it has a weak methylation activity on the H73K ß-actin peptide for which the target H73 residue is mutated into K73. Interestingly, the K73 methylation activity of SETD3 increases significantly as a result of the N255 → A or N255 → F/W273 → A mutation, and the N255A product specificity also differs from that of wild-type. Here, we performed QM/MM molecular dynamics and potential of mean force (PMF) simulations for SETD3 and its mutants (N255A and N255F/W273A) to study how SETD3 and its mutants could have different product specificities and activities for the K73 methylation. The PMF simulations show that the barrier for the first methylation of K73 is higher compared to the barrier of the H73 methylation in SETD3. Moreover, the second methylation of K73 has been found to have a barrier from the free energy simulation that is higher by 2.2 kcal/mol compared to the barrier of the first methyl transfer to K73, agreeing with the suggestion that SETD3 is a monomethylase. For the first, second, and third methylations of K73 in the N255A mutant, the barriers obtained from the PMF simulations for transferring the second and third methyl groups are found to be lower relative to the barrier for the first methyl transfer. Thus, N255A can be considered as a trimethyl lysine methyltransferase. In addition, for the first K73 methylation, the activities from the PMF simulations follow the order of N255F/W273A > N255A > WT, in agreement with experiments. The examination of the structural and dynamic results at the active sites provides better understanding of different product specificities and activities for the K73 methylations in SETD3 and its mutants. It is demonstrated that the existence of well-balanced interactions at the active site leading to the near attack conformation is of crucial importance for the efficient methyl transfers. Moreover, the presence of potential interactions (e.g., the C-H···O and cation-π interactions) that are strengthening at the transition state can also be important. Furthermore, the activity as well as product specificity of the K73 methylation also seems to be controlled by certain active-site water molecules which may be released to provide extra space for the addition of more methyl groups on K73.

Computational Study of Methionine Methylation Process Catalyzed by SETD3.

The SETD3 enzyme has been identified as the methyltransferase for the His73 methylation in ß-actin, and such methylation plays an important role in regulating the actin's biochemical properties and fine-tuning the protein's cellular roles. Further studies have demonstrated that SETD3 may be able to methylase some other residues, including lysine and methionine, that substitute His73 in the ß-actin peptide. The activity of SETD3 on the Met73 peptide is low without turnover. Interestingly, it has been shown that the N255V and N255A mutations of SETD3 can increase the activity by about 3-fold for the methionine methylation, while such mutations lead to a significant reduction of kcat for the His73 methylation. The detailed mechanism that leads to such increase of the activity for the Met73 methylation as a result of the mutations has not been understood. In this work, QM/MM molecular dynamics (MD) and potential of mean force (PMF) free energy simulations are undertaken for investigating structural, dynamic, and energetic properties involving the complex of SETD3 and Met73 peptide and to study the SETD3-catalyzed methionine methylation and the effects of the N255V mutation. It is demonstrated that the free energy barrier in the case of the methionine methylation in SETD3 is about 10 kcal/mol higher than that for the histidine methylation. Moreover, the free energy barrier for the methionine methylation in the N255V mutant is about 1 kcal/mol lower than that in the wild-type enzyme. These results agree with previous experimental observation. The origin of the free-energy barrier changes as a result of the H to M substitution on the ß-actin peptide and the N255V mutation of SETD3 is discussed based on the data obtained from the simulations.

Unraveling the Origins of Changing Product Specificity Properties of Arginine Methyltransferase PRMT7 by the E181D and E181D/Q329A Mutations through QM/MM MD and Free-Energy Simulations.

Arginine methylations can regulate important biological processes and affect many cellular activities, and the enzymes that catalyze the methylations are protein arginine methyltransferases (PRMTs). The biological consequences of arginine methylations depend on the methylation states of arginine that are determined by the PRMT's product specificity. Nevertheless, it is still unclear how different PRMTs may generate different methylation states for the target proteins. PRMT7 is the only known member of type III PRMT that produces monomethyl arginine (MMA) product. Interestingly, its E181D and E181D/Q329A mutants can catalyze, respectively, the formation of asymmetrically dimethylated arginine (ADMA) and symmetrically dimethylated arginine (SDMA). The reasons as to why the mutants have the abilities to add the second methyl group and E181D (E181D/Q329A) has the unique product specificity in generating ADMA (SDMA) have not been understood. Here, quantum mechanics/molecular mechanics (QM/MM) molecular dynamics (MD) and potential of mean force (PMF) free-energy simulations are performed for the E181D and E181D/Q329A mutants to understand the origin for their ability to generate, respectively, ADMA and SDMA. The simulations show that the free-energy barrier for adding the second methyl group to MMA in E181D (E181D/Q329A) to produce ADMA (SDMA) is considerably lower than the corresponding barriers in wild type and E181D/Q329A (wild type and E181D), consistent with experimental observations. Some important factors that contribute to the change of the activity and product specificity due to the E181D and E181D/Q329A mutations are identified based on the data from the simulations and analysis. It is shown that the transferable methyl group (from SAM) and Nη2 (the nitrogen atom that is methylated in the substrate MMA) can only form good near-attack conformations in the E181D reaction state for the methyl transfer (not in wild type and E181D/Q329A), while the transferable methyl group and Nη1 (the nitrogen atom that is not methylated in the substrate MMA) can only form good near-attack conformations in E181D/Q329A (not in wild type and E181D). The results suggest that the steric repulsions in the reaction state between the methyl group on MMA and active-site residues (e.g., Q329) and the release of such repulsions (e.g., from the Q329A mutation) may play an important role in generating specific near-attack conformations for the methyl transfer and controlling the product specificity for the mutants. The general principle identified in this work for PRMT7 is expected to be useful for understanding the activity and product specificity of other PRMTs as well.