20000131 Construction of the scale model of the spatial structure of the protein on its nucleotide sequence

A.Yu. Kushelev, S.A. Pisarzhevsky
The Nanoworld Laboratory
Scientific-industrial association "Politechnology"
107005, Russia, Moscow, Gospitalniy per., 10,
The Bauman Moscow State Technical University
http://ftp.decsy.ru/nanoworld/

The strong correlation dependence of spatial structure of the protein from its nucleotide sequence was theoretically predicted by physical modelling, experimentally discovered and statistically confirmed.

In the process of biosynthesis the third nucleotide of the codon controls the orientation of the amino acid forming the concrete spatial isomer that is the conformation of the protein molecule cutting off competition ways of the forming of 2D and 3D structures.

Keywords: nanoworld, circular model of electron, composition genetical code, prediction of protein structure.

The prediction of the spatial structure of the proteins is the important problem of modern molecular biology [1, 2, 3].

The circular model of the electron was created by us what allows to apply the laws of classical physics to "nonclassical" objects, in particular to create the scale geometrical models of the electron covers of the atoms and the molecules. In our classical models of the microobjects the electrons are presented by the circular magnets. We marked for the simplification of further reasonings that the consideration of the inner structure of the electron, the dependence of the radius of the electron ring from the tension on of the nuclear field and other physical nuances of the circular model of the electron go out from the limits of this article.

The main research method was the physical modeling. It consisted in carrying out model experiments with circular magnets representing the electrons. The number of the circular magnets corresponded to the number of the valent electrons of the molecule.

The supplementary surfaces imitating compensation of electrostatic influence of the nuclei of the atoms and the other electrons of the molecule were used in the experiments.

Some models were made by the method of the symmetrical placing of the set number of the rings with due regard for all kinds of the properties of the real electrons.

The methods of looking through all the versions, analogy, optimization, systematization and others were used.

The first experimental result was obtained in the model experiment with the circular magnets which was carried out with the aim of verification of the hypothesis: "The rings-electrons forms the electron cover the polyhedron". Really 8 circular magnets supports the form of the octahedron from the rings packed up on the sphere of the electrostatic equilibrium of the electrons. This model of electron cover corresponds to the experimentally discovered fermi-surface (fig. 1).

The substitution of the four rings-electrons by the unfinished octahedrons (without one ring) turns the form of the octahedron atom to the familiar to us form the tetrahedron molecule. Electron surfaces of the molecules of carbon tetrachloride and phosphoric acid have such a form.

We constructed more than 1500 circular models of the various chemical compounds.

The special class of chemical compounds in conjugated systems. The model experiment showed that mutual influence of the similar to carbon atoms appeared to be the cause of the reconstruction of the electron structure of the molecule from polyhedron to polylayer what leads to formation of the "flat" conjugated systems, for example, molecule of benzol.

The composition of polyhedron and polylayer atoms and molecules allowed us to compose ringsided structure of DNA nuleotides (fig. 2).The structure of our model of DNA doesn’t contradict generally accepted model.

We constructed the model of the group CCON. The different versions of the mutual arrangement of the groups CCON in the protein structure are given on the figure 3.

T-RNA holds amino acid by the ACC-end. The feature of the structure of t-RNA found by us consists in the loop structure of the ACC-end.The last nucleotide of the triplet ACC is turned on the hinge-molecule of the phosphoric acid thus that the process of the formation of the additional diether bond by its free group PO3 with the group PO3 of the complementary chain of RNA has appeared possible.

The last from the two nucleotides forming 3’-end of t-RNA has inversion magnetic properties,which distinguish all nucleotides of the reverse chains of DNA and RNA from the nucleotides of the straight chains.

The northern magnetic poles of the two electrons (one from oxygen and the second from nitrogen),by which the cytosine -N ends ( in which the northern poles of the electrons forming hydrogen bonds with the complementary base are inverted outside) and the southern magnetic poles of the two electrons of the cytosine-S (in which the southern poles of the electrons forming hydrogen bonds with the complementary base are inverted outside) are disposed complementary to the four electrons of the peptide group,which oriented by the group of nitrogen along the rotation axis of t-RNA.The amino acid may be placed by the manipulator-t-RNA in the growing protein chain with the different angle of the turn around the rotation axis.The angle of the turn is regulated by the varible loop of t-RNA and the distance between the third mucleitide of the codon of m-RNA and the first nucleotide of the anticodon of t-RNA.The distance between the first nucleotide of the anticodon (inosine) and the third nucleotide of the codon depends of the nature of the base complementary for inosine (A, C, G, T).

We have constructed the circularside models of the proteins with the known structure in particular myoglobin, insulin, troponin, oxytocin. The sorting out helped to choose version of the connection of the models of group CCON, which gave the forms of these proteins. It turned out that ,for example, the amino acid glycin is encoded by the code GGA if it was a member of beta-layer, by the code GGG if it was a member of 3/10 alpha-helix where hydrogen bond wasn’t with the forth but with the third amino acid residue and finally by the code GGT if it was the second amino acid residue of the beta-turning. Thus the composition genetical code was composed showing how the amino acid is disposed relatively previous one depending on its genetical code(table 1).

The repetition of the version 2 leads to the formation of beta-layer. The turn of the chain in beta-layer demands the sequence of the versions 1-4-1.

The repetition of the versions 3 leads to the formation of xi-helix deviding alpha and beta parts of the proteins.

The repetition of the version 4 leads to the formation of the diminished alph-helix.Its structure differs from the structure of the ordinary alpha-helix :the hydrogen dond between the group CO and the nitrogen group is closed not on the fourth but on the third amino acid residue.

The special case is the helix of collagen. The versions 1, 2, 3, 4 is alternate in it. Collagen has the unusual structure.Its peculiarity is the presence of the soft axis of the symmetry rolled up into helix. This helix has elastic properties.

The geometry of the amino acids and the table of the composition genetical code we layed to the foundation of the program "Pikotechnology", which assemble the structure of the protein molecule on the screen of the PC. Had used accessible for us codes the program reckoned 30 varieties of keratins to the alpha-structural proteins and 3 types of fibroins to beta-structural proteins.

We compared the data about 2D structure of the proteins from the Brookhaven Protein Data Bank with the data obtained with the help of our program by the codes of the same proteins from Oxford Bank of Nucleotide Sequences.

Eight corresponsing sequences from the Oxford Bank were found for 314 structures of the Brookhavon Bank.Per cent of the content of alpha-codons corresponding to the alpha-helix parts (according to X-ray analysis) is indicated below.

The data obtained by us demands the revision of the existing views on the structures of the proteins, in particular on interpretation of the data of the X-ray structural analysis. The fact is that many globular proteins according to our model experiments consist in sewed together alpha-helix strips, which is interpreted as beta-strips now.

The growing polypeptide chain is synthesized deeply inside the ribosome. However every aminoacid may get the conformation of the extended beta structure (in this case it is encoded by beta codon) and alpha structure (in this case it is encoded by alpha codon). The diameter of the alpha structure practically does not differ from the diameter of the extended beta structure. Moreover, the diameter of the structure depends on the length of the radicals which are included in the composition of the alpha and beta structures more strongly. For example, the replacement of the radicals of Gly by the radicals of Trp enlarges the diameter of beta and alpha structures approximately in three times, while the diameter of the alpha helix exceeds the diameter of beta structure less than in 1,5 times.

Except the extended beta structure (in our model it corresponds to the version 2, for example, for Gly it is the triplet gga) in the channel of the ribosome the alpha helix (version 1, for example, for Gly it is the triplet ggc), 3/10 alpha helix (the version 4, for example, for Gly it is the triplet ggt) and at last the so called xi structure (the version 3, for example, for Gly it is the triplet ggg) may be placed.

More than 20 residues forming one of these four structures really should be incorporated before the chain quits from the channel in the ribosome. However the nature of the codon of any residue may influence its conformation in the protein structure forming either the site of extended beta structure or the site of alpha helix or the site of 3/10 alpha helix or the site of xi structure, which a little differ on diameter in comparison with the length of the largest radicals.

Comparing structures of the proteins received with the help of this program with the data of X-ray structural analysis the specialists engaged in study of the protein structure discover divergences consisting in displacing of the ends of alpha-helix at time even for five aminoacid residues.

Let's reason. What is alpha-helix? This is the helix each coil of which includes approximately four aminoacid residues.

Consequently residues with numbers 1 and 4 are located side by side and are binded by hydrogen bond.

Adjacent residues to the residue 4 (3 and 5) are also located very closely from the residue number 1 (to say not on 1.5 A but on 2 A).

It is turning out that making a mistake on 1 – 1.5 residues we may actually make a mistake on 5 residues.

So if we see that our program shows all turning of protein alpha-helix with "the mistakes up to 5 aminoacid residues" then the conclusion can be made that it works faultlessly, but X-ray making a mistake on one and a half residues may make a mistake on 5 residues.

We would like to supplement the article by the demonstration of the serviceability of our method. For demonsrating we have selected the fragment of lysozyme, which structure is investigated by different metods. Indisputable in its structure is S-S link between amino acid residues Cys96 and Cys117.

Our program "Pikotechnology" generates 3D structure of the protein autamatically, utilizing the table of the forms of the amino acid residues, the table of composition genetical code and genetical code from the EMBL-file.

Having obtained the coordinates of the atoms with the help of our program we see, that the site of the protein Cys96…Cys117 really forms the cycle througth S-S link.

[1] D.J. Wales, H.A.Scheraga. Global optimization of clusters, crystals, and biomolecules. Science, 285(5432) (1999), 1368-1372

[2] D. Shortle. Structure prediction: The state of the art. Curr Biol, 9(6) (1999), R205-209

[3] S.Takada. Go-ing for the prediction of protein folding mechanisms. Proc. Natl. Acad. Sc. USA, 96(21) (1999), 11698-11700.

N	Code	Name	Ver	N	Code	Name	Ver
00	AAA	Lys	3	32	GAA	Glu	3
01	AAC	Asn	1	33	GAC	Asp	1
02	AAG	Lys	1	34	GAG	Glu	1
03	AAT	Asn	3	35	GAT	Asp	3
04	ACA	Thr	2	36	GCA	Ala	2
05	ACC	Thr	1	37	GCC	Ala	1
06	ACG	Thr	4	38	GCG	Ala	4
07	ACT	Thr	3	39	GCT	Ala	3
08	AGA	Arg	3	40	GGA	Gly	2
09	AGC	Ser	1	41	GGC	Gly	1
10	AGG	Arg	1	42	GGG	Gly	4
11	AGT	Ser	3	43	GGT	Gly	3
12	ATA	Ile	2	44	GTA	Val	2
13	ATC	Ile	1	45	GTC	Val	1
14	ATG	Met	1	46	GTG	Val	4
15	ATT	Ile	3	47	GTT	Val	3
16	CAA	Gln	3	48	TAA	TKD	1
17	CAC	His	1	49	TAC	Tyr	1
18	CAG	Gln	1	50	TAG	TKD	1
19	CAT	His	3	51	TAT	Tyr	3
20	CCA	Pro	2	52	TCA	Ser	2
21	CCC	Pro	1	53	TCC	Ser	1
22	CCG	Pro	4	54	TCG	Ser	4
23	CCT	Pro	3	55	TCT	Ser	3
24	CGA	Arg	2	56	TGA	TKD	1
25	CGC	Arg	1	57	TGC	Cys	1
26	CGG	Arg	4	58	TGG	Trp	1
27	CGT	Arg	3	59	TGT	Cys	3
28	CTA	Leu	2	60	TTA	Leu	3
29	CTC	Leu	1	61	TTC	Phe	1
30	CTG	Leu	4	62	TTG	Leu	1
31	CTT	Leu	3	63	TTT	Phe	3

The version 2 corresponds to the entry of the residue to the classical beta-layer.

The structure of the 8-electron cover for which the majority of the atoms of the mendeleyev table are completed.

The model of the superposition of the group CCON with the previous group CCON in accordance with composition genetical code.

Version 1: the point F coincides with the point B (oved counter clock-wise for some degrees).

The version 4: the point F coincides with the point B (moved clock-wise for some degrees).

The structure of the fragment of lysozyme created by the program "Pikotechnology".