Limits...
In silico screening, genotyping, molecular dynamics simulation and activity studies of SNPs in pyruvate kinase M2.

Kalaiarasan P, Kumar B, Chopra R, Gupta V, Subbarao N, Bamezai RN - PLoS ONE (2015)

Bottom Line: SNPeffect and HOPE showed three substitutions (C31F, Q310P and S437Y) in-silico as deleterious and functionally important.The 5 intronic SNPs of PKM2, associated with sporadic breast cancer in a case-control study, when subjected to different computational analyses, indicated that 3 SNPs (rs2856929, rs8192381 and rs8192431) could generate an alternative transcript by influencing splicing factor binding to PKM2.We propose that these, potentially functional and important variations, both within exons and introns, could have a bearing on cancer metabolism, since PKM2 has been implicated in cancer in the recent past.

View Article: PubMed Central - PubMed

Affiliation: School of Biotechnology, Shri Mata Vaishno Devi University, Katra, Jammu and Kashmir, India.

ABSTRACT
Role of, 29-non-synonymous, 15-intronic, 3-close to UTR, single nucleotide polymorphisms (SNPs) and 2 mutations of Human Pyruvate Kinase (PK) M2 were investigated by in-silico and in-vitro functional studies. Prediction of deleterious substitutions based on sequence homology and structure based servers, SIFT, PANTHER, SNPs&GO, PhD-SNP, SNAP and PolyPhen, depicted that 19% emerged common between all the mentioned programs. SNPeffect and HOPE showed three substitutions (C31F, Q310P and S437Y) in-silico as deleterious and functionally important. In-vitro activity assays showed C31F and S437Y variants of PKM2 with reduced activity, while Q310P variant was catalytically inactive. The allosteric activation due to binding of fructose 1-6 bisphosphate (FBP) was compromised in case of S437Y nsSNP variant protein. This was corroborated through molecular dynamics (MD) simulation study, which was also carried out in other two variant proteins. The 5 intronic SNPs of PKM2, associated with sporadic breast cancer in a case-control study, when subjected to different computational analyses, indicated that 3 SNPs (rs2856929, rs8192381 and rs8192431) could generate an alternative transcript by influencing splicing factor binding to PKM2. We propose that these, potentially functional and important variations, both within exons and introns, could have a bearing on cancer metabolism, since PKM2 has been implicated in cancer in the recent past.

No MeSH data available.


Related in: MedlinePlus

PKM2 structural properties of wild and nsSVPs.(A) Backbone RMSD of PKM2 (B) Radius of gyration of protein (C) Solvent accessible surface area (D) intra-molecular hydrogen bonds.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4359060&req=5

pone.0120469.g003: PKM2 structural properties of wild and nsSVPs.(A) Backbone RMSD of PKM2 (B) Radius of gyration of protein (C) Solvent accessible surface area (D) intra-molecular hydrogen bonds.

Mentions: We performed molecular dynamics simulations for wild type and other three substitutions (C31F, Q30P and S437Y). The average value of RMSD for backbone atoms was 0.363 nm in wild type PKM2. Whereas, the average values of RMSD for backbone atoms in C31F, Q310P and S437Y were 0.282, 0.314 and 0.312, respectively (Fig. 3A). RMSD of native and nsSVPs as a function of time were within 0.5Å +/- during the simulation, which suggested that the simulation is stable for further structural comparison. The radius of gyration (Rg) value of protein for wild varied between 2.336 nm to 2.484 nm. For C31F, Q310P and S437Y, the Rg value varied between 2.322 nm to 2.463 nm, 2.365 nm to 2.527 nm and 2.343 nm to 2.545 nm, respectively (Fig. 3B). The average values of radius of gyration (Rg-protein) for wild and 3 variant PKM2 proteins were 2.390 nm, 2.364 nm, 2.445 nm and 2.422 nm, respectively. The lower values of radius of gyration in wild protein structure suggested it to be stable. These data revealed that the stability of the PKM2 decreased upon two (Q310P and S437Y) substitutions; and in case of C31F variant protein, it did not show a significant decrease in stability when compared to wild PKM2 (Fig. 3B). However, when stability analysis in tetrameric PKM2 structure using FOLDX (based on total energy) was carried out, it showed a significant difference in stability between the wild and variant C31F PKM2. Further, glycerol gradient results of the C31F-PKM2 protein showed more of the relatively inactive dimeric form as compared to functionally active tetramer, apparently supporting that C31F nsSVP is less stable (S2 Fig). Solvent accessible surface area (SASA) was calculated for wild type and variant proteins trajectories (Fig. 3C) for each residue [37] and values averaged. The average SASA of wild type and nsSVPs were: 129.768 nm2, 128.339 nm2, 132.527 nm2 and 132.161 nm2, respectively. The SASA of wild type PKM2 varied from 118.497 nm2 to 148.950 nm2; and that of variant proteins C31F, Q310P and S437Y, differed from 120.376 nm2 to 147.990 nm2, 123.296 nm2 to 147.333 nm2 and 123.634 nm2 to 1487.204 nm2, respectively. The number of intra-molecular hydrogen bonds were calculated for wild and variant proteins to assess the fluctuation of the rigidity of the proteins. The wild type and three nsSVPs, C31F, Q310P and S437Y, showed 389, 397, 386 and 399, intra-molecular hydrogen bonds, respectively (Fig. 3D). The RMSF differences indicated that the dynamics of the wild and the 3 variants had four loop regions that displayed larger flexibility. One of the loops (the loop-1) was located at the PEP binding site and the rest of the 3 were near FBP binding site. This fluctuation in the structure allowed us to calculate the distance between each domain of PKM2 (Fig. 4). The distance between each domain was calculated to identify the effect of these variations in the domains of PKM2; where the distance between the domains A and B decreased significantly in the case of nsSVPs when compared to wild type PKM2 (Fig. 5A). Incidentally, the binding pocket of the substrate (PEP) was located in-between domains A and B (Fig. 2). The nsSVPs of PKM2 showed a slightly higher distance between the domains A and C, when compared to wild type PKM2 (Fig. 5B). The wild type and variant C31F followed the same pattern in the domain distance between A and N, showing larger distance comparable to other two, Q310P and S437Y, variant proteins which followed the same pattern (Fig. 5C). The S437Y variant protein showed a sudden decrease in the distance between domain B and C between 10ns to 40ns; and after 40ns the distance was stable (Fig. 5D). The distance between the domains A and N was reflected in-between domains, B and N, as well (Fig. 5E). Whereas, a larger distance between domains C and N was observed in wild type and C31F variant, when compared to Q310P and S437Y nsSVPs (Fig. 5F).


In silico screening, genotyping, molecular dynamics simulation and activity studies of SNPs in pyruvate kinase M2.

Kalaiarasan P, Kumar B, Chopra R, Gupta V, Subbarao N, Bamezai RN - PLoS ONE (2015)

PKM2 structural properties of wild and nsSVPs.(A) Backbone RMSD of PKM2 (B) Radius of gyration of protein (C) Solvent accessible surface area (D) intra-molecular hydrogen bonds.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4359060&req=5

pone.0120469.g003: PKM2 structural properties of wild and nsSVPs.(A) Backbone RMSD of PKM2 (B) Radius of gyration of protein (C) Solvent accessible surface area (D) intra-molecular hydrogen bonds.
Mentions: We performed molecular dynamics simulations for wild type and other three substitutions (C31F, Q30P and S437Y). The average value of RMSD for backbone atoms was 0.363 nm in wild type PKM2. Whereas, the average values of RMSD for backbone atoms in C31F, Q310P and S437Y were 0.282, 0.314 and 0.312, respectively (Fig. 3A). RMSD of native and nsSVPs as a function of time were within 0.5Å +/- during the simulation, which suggested that the simulation is stable for further structural comparison. The radius of gyration (Rg) value of protein for wild varied between 2.336 nm to 2.484 nm. For C31F, Q310P and S437Y, the Rg value varied between 2.322 nm to 2.463 nm, 2.365 nm to 2.527 nm and 2.343 nm to 2.545 nm, respectively (Fig. 3B). The average values of radius of gyration (Rg-protein) for wild and 3 variant PKM2 proteins were 2.390 nm, 2.364 nm, 2.445 nm and 2.422 nm, respectively. The lower values of radius of gyration in wild protein structure suggested it to be stable. These data revealed that the stability of the PKM2 decreased upon two (Q310P and S437Y) substitutions; and in case of C31F variant protein, it did not show a significant decrease in stability when compared to wild PKM2 (Fig. 3B). However, when stability analysis in tetrameric PKM2 structure using FOLDX (based on total energy) was carried out, it showed a significant difference in stability between the wild and variant C31F PKM2. Further, glycerol gradient results of the C31F-PKM2 protein showed more of the relatively inactive dimeric form as compared to functionally active tetramer, apparently supporting that C31F nsSVP is less stable (S2 Fig). Solvent accessible surface area (SASA) was calculated for wild type and variant proteins trajectories (Fig. 3C) for each residue [37] and values averaged. The average SASA of wild type and nsSVPs were: 129.768 nm2, 128.339 nm2, 132.527 nm2 and 132.161 nm2, respectively. The SASA of wild type PKM2 varied from 118.497 nm2 to 148.950 nm2; and that of variant proteins C31F, Q310P and S437Y, differed from 120.376 nm2 to 147.990 nm2, 123.296 nm2 to 147.333 nm2 and 123.634 nm2 to 1487.204 nm2, respectively. The number of intra-molecular hydrogen bonds were calculated for wild and variant proteins to assess the fluctuation of the rigidity of the proteins. The wild type and three nsSVPs, C31F, Q310P and S437Y, showed 389, 397, 386 and 399, intra-molecular hydrogen bonds, respectively (Fig. 3D). The RMSF differences indicated that the dynamics of the wild and the 3 variants had four loop regions that displayed larger flexibility. One of the loops (the loop-1) was located at the PEP binding site and the rest of the 3 were near FBP binding site. This fluctuation in the structure allowed us to calculate the distance between each domain of PKM2 (Fig. 4). The distance between each domain was calculated to identify the effect of these variations in the domains of PKM2; where the distance between the domains A and B decreased significantly in the case of nsSVPs when compared to wild type PKM2 (Fig. 5A). Incidentally, the binding pocket of the substrate (PEP) was located in-between domains A and B (Fig. 2). The nsSVPs of PKM2 showed a slightly higher distance between the domains A and C, when compared to wild type PKM2 (Fig. 5B). The wild type and variant C31F followed the same pattern in the domain distance between A and N, showing larger distance comparable to other two, Q310P and S437Y, variant proteins which followed the same pattern (Fig. 5C). The S437Y variant protein showed a sudden decrease in the distance between domain B and C between 10ns to 40ns; and after 40ns the distance was stable (Fig. 5D). The distance between the domains A and N was reflected in-between domains, B and N, as well (Fig. 5E). Whereas, a larger distance between domains C and N was observed in wild type and C31F variant, when compared to Q310P and S437Y nsSVPs (Fig. 5F).

Bottom Line: SNPeffect and HOPE showed three substitutions (C31F, Q310P and S437Y) in-silico as deleterious and functionally important.The 5 intronic SNPs of PKM2, associated with sporadic breast cancer in a case-control study, when subjected to different computational analyses, indicated that 3 SNPs (rs2856929, rs8192381 and rs8192431) could generate an alternative transcript by influencing splicing factor binding to PKM2.We propose that these, potentially functional and important variations, both within exons and introns, could have a bearing on cancer metabolism, since PKM2 has been implicated in cancer in the recent past.

View Article: PubMed Central - PubMed

Affiliation: School of Biotechnology, Shri Mata Vaishno Devi University, Katra, Jammu and Kashmir, India.

ABSTRACT
Role of, 29-non-synonymous, 15-intronic, 3-close to UTR, single nucleotide polymorphisms (SNPs) and 2 mutations of Human Pyruvate Kinase (PK) M2 were investigated by in-silico and in-vitro functional studies. Prediction of deleterious substitutions based on sequence homology and structure based servers, SIFT, PANTHER, SNPs&GO, PhD-SNP, SNAP and PolyPhen, depicted that 19% emerged common between all the mentioned programs. SNPeffect and HOPE showed three substitutions (C31F, Q310P and S437Y) in-silico as deleterious and functionally important. In-vitro activity assays showed C31F and S437Y variants of PKM2 with reduced activity, while Q310P variant was catalytically inactive. The allosteric activation due to binding of fructose 1-6 bisphosphate (FBP) was compromised in case of S437Y nsSNP variant protein. This was corroborated through molecular dynamics (MD) simulation study, which was also carried out in other two variant proteins. The 5 intronic SNPs of PKM2, associated with sporadic breast cancer in a case-control study, when subjected to different computational analyses, indicated that 3 SNPs (rs2856929, rs8192381 and rs8192431) could generate an alternative transcript by influencing splicing factor binding to PKM2. We propose that these, potentially functional and important variations, both within exons and introns, could have a bearing on cancer metabolism, since PKM2 has been implicated in cancer in the recent past.

No MeSH data available.


Related in: MedlinePlus