Impact of genetic variation on three dimensional structure and function of proteins

Abstract

The Protein Data Bank (PDB; http://wwpdb.org) was established in 1971 as the first open access digital data resource in biology with seven protein structures as its initial holdings. The global PDB archive now contains more than 126,000 experimentally determined atomic level three-dimensional (3D) structures of biological macromolecules (proteins, DNA, RNA), all of which are freely accessible via the Internet. Knowledge of the 3D structure of the gene product can help in understanding its function and role in disease. Of particular interest in the PDB archive are proteins for which 3D structures of genetic variant proteins have been determined, thus revealing atomic-level structural differences caused by the variation at the DNA level. Herein, we present a systematic and qualitative analysis of such cases. We observe a wide range of structural and functional changes caused by single amino acid differences, including changes in enzyme activity, aggregation propensity, structural stability, binding, and dissociation, some in the context of large assemblies. Structural comparison of wild type and mutated proteins, when both are available, provide insights into atomic-level structural differences caused by the genetic variation.

Fig 1. Distribution of SNVs based on structural position.

A. Distribution of the SNVs in our benchmark dataset based on structural position (Piechart). There are two broad categories, Surface (residue) or Buried (residue). B. Distribution of the SNVs in the dataset based on secondary structural location (Bar graph). The two broad categories, Surface (residue) or Buried (residue) are further categorized into Loop, Alpha_helix and Beta_strand based on the secondary structural element to which the SNV maps.

Fig 2. SNV consequences map to various locations within protein structures.

A) PDB: 1AZV, SNV: rs121912431 (G37R) is present on the surface of the protein in the highlighted Loop segment, where it causes the neurological disease Lou Gehrig’s disease. B) PDB: 1J04, SNV: rs121908529 (G170R) is present on the surface of the protein in the highlighted Alpha_helix, where it causes hereditary kidney stone disease primary hyperoxaluria type 1. C) PDB: 3S5E, SNV: rs138471431 (W155R) is present on the surface of the protein in the highlighted Beta_sheet, where it causes the neurodegenerative disease Friedreich’s ataxia. D) PDB: 2V7A, SNV rs121913459 (T315I) is present in the ATP-binding domain and causes resistance to the drug imatinib in patients with chronic myelogenous leukemia.

Fig 3. SNVs that affect both protein structure and function.

A) The P428L mutant form of Arylsulfatase A adopts an atypical dimeric configuration (instead of the normal octamer), which reduces protein half-life. B) The F12L mutant form of Delta-aminolevulinic acid dehydratase assembles as a hexamer (instead of the normal octamer), which shifts the pH optimum of the enzyme from pH 7→pH 9.

Fig 4. SNV related change that affects enzymatic activity.

Semi-transparent, solvent-accessible surface representation of the AMP binding site. For the wild-type structure (PDB: 2ZT5) AMP is bound in the active site (atom type color coded stick figure), while in the mutant structure (PDB: 2PMF) AMP binding is blocked by projection of the arginine sidechain (red) into the active site, thereby blocking substrate ATP binding and inactivating the enzyme.

Fig 5. SNV that affects protein structure stability.

Disease causing mutation site in protein DJ-1. The wild-type structure (PDB: 1P5F) is depicted in green and the variant (PDB: 2RK4) in red. M26 is a conserved residue in Alpha_helix A located within the hydrophobic core of the protein. The steric clash between I26 and the sidechain of I31 results in a ~0.7 Å displacement of I31 away from I26, resulting in loss of favorable packing contacts involving M26.

Fig 6. Close-up view of the nucleotide-binding region of Lys117Arg.

The mutated residue R117 is stabilized by interactions with the P-loop (Gly13, main-chain CO) and additional interaction with Asn85. Thus the mutated residue causes destabilization of nucleotide binding owing to loss of a direct contact with the ligand. Mutated PDB: 2QUZ(blue) and wild-type PDB: 2CE2 (pink).

Fig 7. In manganese superoxide dismutase, a SNV can affect protein assembly.

The wild type assembly state is tetrameric (left, but due to the mutation mapping to the dimer-dimer interface (in red), the tetrameric structure is not observed in solution (right).

Fig 8. von Willebrand factor (wild-type: green PDBID 1OAK; I546V mutant PDB: 1IJK) with the location of I546V mutation highlighted.

Substitution of Ile with Val at position 546 creates a cavity in the hydrophobic core of the I546V mutant structure, which is occupied by a water molecule (denoted by +). The resulting structure perturbation is transmitted through the interior of the protein affecting the locations of the sidechains of Y565, His563, and D560. Collectively, these changes affect Gplb binding, giving rise to von Willebrand’s disease.

Fig 9. Examples of special cases.

A) PDB: 1OPH. The highlighted residue in red represents the mutation (M358R) site. Due to this mutation, alpha1-antitrypsin loses its function as an elastase inhibitor, retains its function as a trypsin inhibitor, and gains a function as a thrombin inhibitor. B) PDB: 1J04. The two highlighted regions represent the two polymorphisms that act synergistically. The highlighted region in green represents P11L polymorphism in AGT whereas the highlighted region in red represents the disease-specific G170R mutation. C) PDB: 1J47. The highlighted red residue represents the M64I in the full-length hSRY sequence, which corresponds to M9I in the given construct and affects the extent of DNA bending.

Fig 10. Frequency distribution of the SNVs.

Bar graph indicates distribution of SNVs as Other/No effect (either neutral or does not cause a disease), Disease causing and associated with the Risk of developing a disease within each frequency category.