QSAR Studies of Conformational Parameters for Alpha-helix/Beta - sheet, and Reverse Turn Residues and Residue Frequencies in Loops

Hanqing Wu,   hanqing@csd.uwm.edu

Department of Chemistry, University of Wisconsin-Milwaukee

Milwaukee, WI 53201, USA


ABSTRACT

INTRODUCTION

RESULTS

DISCUSSION

CONCLUSION

REFERENCES


                                             ABSTRACT

         More than fifty parameters with different properties of amino acid residues are collected including DeltaG, DeltaH, DeltaS values, inorganic, organic property values and residue electronegativity value calculated by empirical methods. The correlation among them is also studied by QSAR method. The conformational parameters (CP) for alpha-helix / beta - sheet, and reverse turn residues are highly correlated with LFlex, PoPoty, and Hphox / Charge. Residue frequencies (RF) in loops are highly correlated with V, Hypat and FreeEng.


                                           INTRODUCTION

        The secondary structure of proteins falls into three classes: alpha-helix, beta-sheet, and reverse turns. Another category of nonregular secondary structure is the "loop" described by George D. Rose(Science, 234, 849(1986)). More than fifty parameters with different properties of amino acid residues are collected including DeltaG, DeltaH, DeltaS values, inorganic(I), organic(O) property values and residue electronegativity values calculated by empirical methods. All of these parameters can be used for protein sequence analyses which have been done for many years, and many programs have been developed (PROFILEGRAPH, SEQAID can be freely distributed). The correlation among them can be studied by QSAR method. If two or more parameters have high correlation with each other, only one of them is enough used to make profile of protein sequence by PROFILEGRAPH program with desired window width. In this poster, author interested in the correlation of conformational parameters for Alpha-helix / Beta - sheet, and Reverse Turn Residues and Residue Frequencies in Loops with other parameters.


                                                     RESULTS

      The correlation among all the collected  parameters can be studied by QSAR method (some parameters and their definition can be see from another attached file). The following table only list the correlation among 14 parameters.



          The conformational parameters (CP) for alpha-helix/beta-sheet, and reverse turn residues  (from Peter Y. Chou and Gerald D. Fasman, Ann. Rev. Biochem., 47, 251 - 276 (1978)) are highly correlated with LFlex, PoPoty, and Hphox/Charge. Residue frequencies (RF) in loops are highly correlated with V, Hypat and FreeEng, the followings are the equations:


Where: LFlex is local flexibility (Ragone et al. 1989), PoPoty is polar polarity (Ponnuswamy et al., Biochim. Biophys. Acta 623, 301 (1980) ), Hphox is hydrophobic index (Ponnuswamy et al., Biochim. Biophys. Acta 623, 301 (1980) ), Charge is charge of amino acids (example scale), V is volume (Chothia, Nature 254, 304 (1975) ), Hypat is hydropathy (Kyte and Doolittle, J. Mol. Biol. 157, 105 (1982) ), and FreeEng is the free energy of transfer to surface (Bull & Breese, Arch. Biophys. Biochem. 161, 665 (1974) ).Their values can be seen in another file.

      There are many normalized frequencies calculated using differently weighted protein data (Michael Levitt, Biochemistry,17,4277-4284(1978)) and other scales of conformational parameters. By using above series of parameters, the similar QSAR equations can also be obtained (not provided here).


     Using (0.002196*LFlex, 0.05237*Popoty, 2.321*Hphox) as (x,y,z), the 3-D plot can be obtained by most computer modelling software (see the figures below). Two groups of amino acids can be devided, the small one consists of 3 dark grey("C"), 3 light grey("H"), and 1 yellow("S") amino acids, the large one consists of two small clusters of amino acids and separated sites for amino acid. It can be known that the amino acids in the small group are mostly found in alpha helix/beta sheet, while the amino acids in two small clusters of the large group are mainly found in turns or breaks.Some amino acids in separated sites are also highly frequently found in secondary structure in the proteins. The following figures also tell us the complication of classification the preferences of formation secondary structure in proteins even by three parameters of amino acids. Specific type of amino acid residues can be selected, the symble or color have the following definition:

The dark grey color<--->"C"---represents aliphatic amino acids: they are Glycine[Gly,G]; Alanine[Ala,A]; Valine[Val.V]; Leucine[Leu,L]; Isoleucine[Ile,I]; Proline[Pro,P].

The light grey color<--->"H"---represents aromatic amino acids: they are Phenylanine[Phe,F]; Tyrosine[Y]; Tryptophan[Trp,W].

 The red color<--->"O"---represents basic and acidic amino acids: they are Histidine[His,H]; Lysine[Lys,K]; Arginine[Arg,R]; Aspartate[Asp,D]; Glutamate[Glu,E].

The blue color<--->"N'---represents alcohol and amide amino acids: they are Serine[Ser,S]; Threonine[Thr,T]; Asparagine[Asn,N]; Glutamine[Gln,Q].

The yellow color<--->"S"---represents sulfur-containing amino acids: they are Methionine[Met,M]; Cysteine[Cys,C].



                                                  DISCUSSION

        James U. Bowie et al (Science, 253, 164(1991)) divided six side-chain environment categories, most of eighteen 3D-1D scores are highly correlated with two or three amino acid residue parameters (r >0.9). Author find that the conformational parameters for alpha-helical residues have less correlation with any two or three parameters, the same results for the relative preference values for position-specific amino acid preferences in helices calculated by Jane S. Richardson and David C. Richardson (Science, 240, 1648(1988)). The possible reason is that the ith residue in an alpha-helix is also dependent on i+4 and i-4 residues on the sequence besides dependent on its neighboring i+1 and i-1 residues. The tendency of a polypeptide chain to form alpha-helix or beta-sheet secondary structure dependents upon local and nonlocal effects. Local effects reflect the intrinsic properties of the amino acid residues for particular secondary structure, while nonlocal effects reflect the positioning of the individual residues in the context of the entire amino acid sequence. If we consider local and non local effects of above mentioned parameters (for instance average of seven residues' parameters in a consecutive protein sequence), we possible can precisely predict secondary structure of proteins.


                                              CONCLUSION

1.  The residue parameters of LFlex  (local flexibility), PoPoty (polar polarity), Hphox (hydrophobic index) are important in the formation of secondary structure (alpha helix or beta sheet) as well as for the reverse turn formation in proteins.

2.  The charge in the amino acid residue is also important for formation beta sheet. V (volume), Hypat (hydropathy), and FreeEng (the free energy of transfer to surface) are important in the formation of "loops" in proteins.

3.  Some parameters are highly correlated with each other: (V, Maximal_A); (Flex, FreeEng); (Hypat, AccSA, Bur, Exp); (DeltaH, DeltaG) etc. Computer graphics can be used for classification high or low preference of amino acids to form secondary structure.
                                          REFERENCES

1.  Peter Y. Chou and Gerald D. Fasman, Empirical Predictions of Protein Conformation, Ann. Rev. Biochem., 47, 251-276(1978).

2.  Michael Levitt, Conformational Preferences of Amino Acids in Globular Proteins, Biochemsitry, 17, 42774285(1978).

3.  Jacquelyn F. Leszczynski and George D. Rose, Loops in Globular Proteins: A Novel Category of Secondary Structure, Science,234, 849-855(1986).

4. James U. Bowie et al, A Method to Identify Protein Sequences That Fold into a Known Three-Dimensional Structure, Science, 253,164-170(1991).

Thanks to Mr. Mark Bosley for setting up inter-page linkage and some helpfull correction.


Return to main WATOC Poster page