< 2nd term

The interior of proteins and macromolecular complexes

Last update on the 20th of February, 2017

This work is about describing hydrophobic clusters in proteins, DNA structural parts and modeling interaction of DNA and protein. All needed visual material is presented in the applet below, disscussion goes onwards.

Scripts

Table 1. Desription of scripts and links to them
Button name link PDB ID Load speed
Hydrophobe script 1ZK7 Slow
Residue script 1ZK7 Medium
DNA1 script 5KFF, modified Fast
DNA2 script 5KFF, modified Fast

Hydrophobic cores

Script: Hydrophobe

As it is known, some of aminoacid residues exhibit hydrophobic properties, whereas others - hydrophilic. The solvent - water - forms a network of hydrogen bonds, which are unlikely to be cleaved because of thermodynamic effects. So, protein molecule exposes on its surface hydrophilic residues and hides into depth hydrophobic ones. In such a way hydrophobic cores are formed[1].

The presented subunit of mercuric reductase consists of 3451 atoms, its two hydrophobic cores of 661 and 78 atoms, the sare of 19.15% and 2.26%, respectively. These clusters were identified with CluD algorithm. Clearly seen, the cores fill almost all inner space of protein.

Interestingly, visibly huge share of surface is hydrophobic due to bigger (green in viewing) core. There are several possible explanations of that. Firstly, the protein may be localized in any membrane, but the presented protein is cytoplasmatic[2]. Secondly, it may bind otherhydrophobic molecules. This statement is more likely to be true, because 1) the protein is homodimer[3]; 2) binds FAD[4]. The applet contains a representation of two divided monomers with ligands. The observation makes the suggestion even more plausible. Specific functions of smaller core has yet to be explained.

Packing density of atoms

Script: Residue

The important feature of any atom heap is its density. That value becomes more intriguing when it comes to hydrophobic core. Indeed, how tightly are atoms packaged? In order to determine the answer, I did following steps.

First, I've chosen one of residues in very depth of bigger core - Phe195. Surrounded atoms are defined as those which are not far then 7 Å from the Phe195. That is 29 residues with 217 atoms. Then, I visualized step-by-step atoms not far then 1, 2, ..., 7 Å. Finally, I did some math and came up with the results. Main points are:

  1. Ammount of surrounded atoms increases rapidly since the distance value is not less than 4 Å;
  2. The Phe195 is visibly hidden by other atoms placed not far then 5 - 6 Å;
  3. Characteristic distance between two covalently non-binded atoms is counted by the weighted arithmetic mean[5] function based on the data from Table 2.
Table 2. Pocket table of atoms surrounding the Phe195 in the hydrophobic core of MerA protein.
Pocket, Å Number of atoms
0-1 0
1-2 2
2-3 7
3-4 23
4-5 30
5-6 33
6-7 41
Total 136
Weighted arithmetic mean, 0-6* 4.89
*As the Phe195 is invisible from the distance of 6 Å, there is no need in taking into account pocket 6 - 7 Å in WAM.

The counts were made in J(S)Mol script and Google sheets. The link to script source code was fetched above; the table with formula in .xlsx format is available for download here.

So, the characteristic distance between two non-binded atoms (centers of atoms) in hydrophobic core is about 4.89 Å. Is it possible to place any other atom? Or molecule of water, which at approximation is an oxygen atom. To count that, I used Van der Waals radius of oxygen - 1.4 Å, the smallest one among all "life" elements except for hydrogen[6]. Imagine, that we have two atoms of oxygen, placed in 4.89 Å distance counting from their centers. So, the free space is 4.89 - 2 * 1.4 = 2.09 Å. The oxygen atom diameter is 1.4 * 2 = 2.8 Å, that is larger, then the remaining space. So, it is impossible to insert any additional atom into current hydrophobic core, that may give evidence for rather tight atom packing in the core.

DNA-protein interaction

Script: DNA1

The presented model is a restricted part of human DNA polymerase eta-DNA[7] ternary complex with Mn2+. The DNA is fully provided on this applet, whereas protein remains is a continious chain of four beta strands and two alpha helices. It looks like the protein somehow "cuddles" DNA in area of major groove, interacting by two beta strands with DNA backbone. It is clearly seen, how sidechains of beta strand residues are aimed to the backbone (this point will be disscussed later). For what reason may this occur? Possibly, this portion of protein "support" DNA chains whereas another portion provides the connection of dATP to one of the DNA chains or simply grab such a huge molecule in order to immobilise it - on the whole, carry out auxiliary function to polymerization.

Hydrogen bonds between DNA and protein, or once more about the previous object

Script: DNA2

This script presents some structural properties of DNA molecule: minor and major groove, hydrogen donors and acceptors in terms of making hydrogen bonds with protein in both grooves and issues connected to them. First, with which part of DNA molecule - minor, major grooves or backbone - does the protein interact? The solution is to count the number of atoms of those protein residues which contain atoms that are placed not far then 5 Å to relative structures with priority to bakcbone (for instance, protein does not hover minor grove, so it wolud be unfair to rank "backboned" atoms for minor groove). The results are that 212 atoms interact with backbone and only 34 with major groove, minor groove is the loneliest one. So it is likely to say that the current portion of protein interacts with the DNA backbone.

The assertion is more or less confirmed by the fact that the protein forms 9 hidrogen bonds with DNA backbone by beta strand atoms.

Mysterious reaction

The reaction presented in this video is a dephosphorylation of paraoxon[8] (diethyl p-nitrophenyl phosphate). According to the scheme of reaction the mechanism is bimolecular nucleophilic substitution (SN2). Electrophile is a phosphorus atom, which lacks electrons because of four oxygen atoms covalently binded to it, whereas nucleophile is an oxygen atom of tyrosine hydroxyl. The steps are following:

  1. Nucleophile activation: proton transition from tyrosine hydroxyl to "free" oxygen atom in phosphate, covalent bond is replaced with hydrogen one;
  2. Transition state: oxygen of tyrosine attacks phosphorous atom, a bond between phosphorus and oxygen of p-nitrophenyl is being cleaved, a bond between phosphorus and oxygen of tyrosine is being created - quinquivalent state of phosphorus atom;
  3. Product release: p-nitrophenolate gets protonated with hydrogen derived from "free" oxygen atom of phospate, stereochemical inversion of diethyl phosphate bonded to tyrosine.

Paraoxon usually acts as an acetylcholinesterase inhibitor and was used as highly effective insecticide, but later depricated because of its high toxicity to vertebrates[8].

References

  1. Bruce Alberts et al. Molecular biology of the cell 6th ed. Garland science, 2014;
  2. Wikipedia article on mercuric reductase;
  3. MerA structure paragraph on Wikipedia;
  4. MerA mechanism paragraph on Wikipedia;
  5. Wikipedia article on Weighted arithmetic mean;
  6. Jhon Emsley. The Elements. Clarendon Press, 1991;
  7. DNA polymerase article on Wikipedia;
  8. Wikipedia article on paraoxon.