Methods to identify macromolecule binding and aggregation prone regions in proteins and uses therefor

Abstract

The present invention provides methods and computational tools based, at least in part, on computer simulations that identify macromolecule binding regions and aggregation prone regions of a protein. Substitutions may then be made in these aggregation prone regions to engineer proteins with enhanced stability and/or a reduced propensity for aggregation. Similarly, substitutions may then be made in these macromolecule binding regions to engineer proteins with altered binding affinity for the macromolecule.

1 - 89 . (canceled) 
     
     
         90 . A method for calculating the Spatial-Aggregation-Propensity (SAP) for a particular atom in a protein, comprising
 (i) identifying one or more atoms or amino acid residues in a structural model representing the protein, wherein the one or more atoms are within a defined spatial region centered on or near the particular atom or the one or more amino acid residues have at least one atom within a defined spatial region centered on or near the particular atom;   (ii) calculating, for the one or more atoms in the defined spatial region, a ratio of the solvent accessible area (SAA) of the atoms to the SAA of atoms in an identical residue which is fully exposed;   (iii) multiplying each ratio by the atom hydrophobicity of the one or more atoms; and   (iv) summing the products of step (c); whereby the sum is the SAP for the particular atom; and   
       wherein, optionally, the SAP for the particular atom is calculated by conducting a molecular dynamics simulation prior to step (i) and repeating steps (i)-(iv), each time conducting a further molecular dynamics simulation at a plurality of time steps, thereby producing multiple sums as in step (iv), and calculating the average of the sums; 
       whereby the calculated average is the SAP for the particular atom 
     
     
         91 . The method of  claim 90  wherein the defined spatial region is a sphere having a radius of between 1-30 Å. 
     
     
         92 . The method of  claim 91  wherein the radius is 5 Å. 
     
     
         93 . The method of  claim 90  wherein the molecular dynamics simulation is performed using a simulation package chosen from the group comprising ABINIT, AMBER, Ascalaph, CASTEP, CPMD, CHARMM, DL_POLY, FIREBALL, GROMACS, GROMOS, LAMMPS, MDynaMix, MOLDY, MOSCITO, NAMD, Newton-X, ProtoMol, PWscf, SIESTA, VASP, TINKER, YASARA, ORAC, and XMD. 
     
     
         94 . A method to identify an aggregation prone region on a protein, comprising
 (a) mapping, onto a structural model of the protein the SAP as calculated according to  claim 90 ; and   (b) identifying a region within in the protein having a plurality of atoms having a SAP>0;   
       wherein the aggregation prone region comprises the amino acids comprising said plurality of atoms. 
     
     
         95 . The method of  claim 94  where the identifying comprises plotting the SAP values; calculating, for peaks in the plot, the area under the curve (AUC); and identifying one or more protein regions with a positive AUC, wherein the aggregation prone region comprises the identified protein regions. 
     
     
         96 . A method of making a protein variant which exhibits a reduced propensity for aggregation, comprising
 replacing or deleting at least one amino acid residue within an aggregation prone region in the protein,   
       wherein the aggregation prone region is identified using SAP scores calculated according to  claim 90 ; and wherein, if the amino acid residue is replaced, it is replaced with an amino acid residue which is more hydrophilic, such that the propensity for aggregation of the variant is reduced. 
     
     
         97 . A method of making a protein variant which exhibits a reduced propensity for aggregation, comprising
 (a) generating a plurality of protein variants by replacing in each variant at least one residue within an aggregation prone region in the protein,   
       wherein the aggregation prone region is identified using SAP scores calculated according to  claim 90 , wherein one or different residues, or different combinations of residues are replaced in each variant; wherein the at least one residue is replaced with a residue which is more hydrophilic; and
 (b) selecting a protein variant prepared as in (a) which exhibits a reduced propensity for aggregation. 
 
     
     
         98 . The method of  claim 96  wherein at least two amino acid residues within the aggregation prone region are replaced. 
     
     
         99 . The method of  claim 96  wherein at least one residue is replaced within more than one aggregation prone regions within the protein. 
     
     
         100 . The method of  claim 96  wherein the protein is selected from the group consisting of an antibody, a Fab fragment, a Fab′ fragment, an Fd fragment, an Fv fragment, an F(ab′) 2  fragment, and an Fc fragment. 
     
     
         101 . A method for making a pharmaceutical composition comprising a protein variant which exhibits a reduced propensity for aggregation, comprising
 formulating a protein variant obtained according to the method of  claim 96  together with a pharmaceutically acceptable carrier, adjuvant and/or excipient.   
     
     
         102 . A method to identify a macromolecule binding region on a protein, comprising
 (a) mapping, onto a structural model of the protein the SAP as calculated according to  claim 90 ; and   (b) identifying a region within in the protein having a plurality of atoms having a SAP>0;   
       wherein the macromolecule binding region comprises the amino acids comprising said plurality of atoms. 
     
     
         103 . The method of  claim 102  where the identifying comprises plotting the SAP values; calculating, for peaks in the plot, the area under the curve (AUC); and identifying one or more protein regions with a positive AUC, wherein the macromolecule binding region comprises the identified protein regions. 
     
     
         104 . A method of making a protein variant which exhibits a reduced binding affinity for a macromolecule, comprising
 replacing or deleting at least one amino acid residue within a macromolecule binding region for the macromolecule in the protein,   
       wherein the macromolecule binding region is identified using SAP scores calculated according to  claim 90 , and wherein, if the amino acid residue is replaced, it is replaced with an amino acid residue which is more hydrophilic, such that the binding affinity for the macromolecule of the variant is reduced. 
     
     
         105 . A method of making a protein variant which exhibits an altered binding affinity for a macromolecule, comprising
 (a) generating a plurality of protein variants by replacing in each variant at least one residue within a macromolecule binding region for the macromolecule in the protein, wherein the macromolecule binding region is identified using SAP scores calculated according to  claim 90 ; and wherein one or different residues, or different combinations of residues are replaced in each variant; and   (b) selecting a protein variant prepared as in (a) which exhibits an altered binding affinity for the macromolecule.   
     
     
         106 . The method of  claim 104  wherein at least two amino acid residues within the macromolecule binding region are replaced. 
     
     
         107 . The method of  claim 104  wherein at least one residue is replaced within more than one macromolecule binding regions within the protein. 
     
     
         108 . A method for making a pharmaceutical composition comprising a protein variant which exhibits an altered propensity for interaction with a binding partner, comprising formulating a protein variant obtained according to a process of  claim 104  together with a pharmaceutically acceptable carrier, adjuvant and/or excipient.