Difference between revisions of "Task 7: Structure-based mutation analysis"

From Bioinformatikpedia
(R261Q)
(R408W)
 
(11 intermediate revisions by the same user not shown)
Line 1,457: Line 1,457:
 
| 2053.34
 
| 2053.34
 
| -1228.48
 
| -1228.48
| [[File:PAH_R261Q_GROMACS_Bond.png |thumb | 300px | Figure 22: Gromacs energy of R261Q]]
+
| [[File:PAH_R261Q_GROMACS_Bond.png |thumb | 300px | Figure 24: Gromacs energy of R261Q]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,464: Line 1,464:
 
| 179.696
 
| 179.696
 
| 95.0119
 
| 95.0119
| [[File:PAH_R261Q_GROMACS_Angle.png |thumb | 300px | Figure 23: Gromacs energy of R261Q]]
+
| [[File:PAH_R261Q_GROMACS_Angle.png |thumb | 300px | Figure 25: Gromacs energy of R261Q]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,471: Line 1,471:
 
| 3020.89
 
| 3020.89
 
| -4437.22
 
| -4437.22
| [[File:PAH_R261Q_GROMACS_Potential.png |thumb | 300px | Figure 24: Gromacs energy of R261Q]]
+
| [[File:PAH_R261Q_GROMACS_Potential.png |thumb | 300px | Figure 26: Gromacs energy of R261Q]]
 
|}
 
|}
   
Line 1,488: Line 1,488:
 
| 2209.81
 
| 2209.81
 
| -1459.51
 
| -1459.51
| [[File:PAH_T266A_GROMACS_Bond.png| 300px]]
+
| [[File:PAH_T266A_GROMACS_Bond.png |thumb | 300px | Figure 27: Gromacs energy of T266A]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,495: Line 1,495:
 
| 192.08
 
| 192.08
 
| 65.6116
 
| 65.6116
| [[File:PAH_T266A_GROMACS_Angle.png| 300px]]
+
| [[File:PAH_T266A_GROMACS_Angle.png |thumb | 300px | Figure 28: Gromacs energy of T266A]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,502: Line 1,502:
 
| 3239.42
 
| 3239.42
 
| -4913.99
 
| -4913.99
| [[File:PAH_T266A_GROMACS_Potential.png| 300px]]
+
| [[File:PAH_T266A_GROMACS_Potential.png |thumb | 300px | Figure 29: Gromacs energy of T266A]]
 
|}
 
|}
   
Line 1,519: Line 1,519:
 
| -nan
 
| -nan
 
| -2575.37
 
| -2575.37
| [[File:PAH_P275S_GROMACS_Bond.png| 300px]]
+
| [[File:PAH_P275S_GROMACS_Bond.png |thumb | 300px | Figure 30: Gromacs energy of P275S]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,526: Line 1,526:
 
| -nan
 
| -nan
 
| -61.3735
 
| -61.3735
| [[File:PAH_P275S_GROMACS_Angle.png| 300px]]
+
| [[File:PAH_P275S_GROMACS_Angle.png |thumb | 300px | Figure 31: Gromacs energy of P275S]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,533: Line 1,533:
 
| -nan
 
| -nan
 
| -6645.9
 
| -6645.9
| [[File:PAH_P275S_GROMACS_Potential.png| 300px]]
+
| [[File:PAH_P275S_GROMACS_Potential.png |thumb | 300px | Figure 32: Gromacs energy of P275S]]
 
|}
 
|}
   
Line 1,550: Line 1,550:
 
| 2383
 
| 2383
 
| -1740.86
 
| -1740.86
| [[File:PAH_T278N_GROMACS_Bond.png| 300px]]
+
| [[File:PAH_T278N_GROMACS_Bond.png |thumb | 300px | Figure 33: Gromacs energy of T278N]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,557: Line 1,557:
 
| 207.028
 
| 207.028
 
| 26.8598
 
| 26.8598
| [[File:PAH_T278N_GROMACS_Angle.png| 300px]]
+
| [[File:PAH_T278N_GROMACS_Angle.png |thumb | 300px | Figure 34: Gromacs energy of T278N]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,564: Line 1,564:
 
| 3493.71
 
| 3493.71
 
| -5504.87
 
| -5504.87
| [[File:PAH_T278N_GROMACS_Potential.png| 300px]]
+
| [[File:PAH_T278N_GROMACS_Potential.png |thumb | 300px | Figure 35: Gromacs energy of T278N]]
 
|}
 
|}
   
Line 1,581: Line 1,581:
 
| 2141.57
 
| 2141.57
 
| -1346.93
 
| -1346.93
| [[File:PAH_P281L_GROMACS_Bond.png| 300px]]
+
| [[File:PAH_P281L_GROMACS_Bond.png |thumb | 300px | Figure 36: Gromacs energy of P281L]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,588: Line 1,588:
 
| 187.673
 
| 187.673
 
| 80.9581
 
| 80.9581
| [[File:PAH_P281L_GROMACS_Angle.png| 300px]]
+
| [[File:PAH_P281L_GROMACS_Angle.png |thumb | 300px | Figure 37: Gromacs energy of P281L]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,595: Line 1,595:
 
| 3172.53
 
| 3172.53
 
| -4826.89
 
| -4826.89
| [[File:PAH_P281L_GROMACS_Potential.png| 300px]]
+
| [[File:PAH_P281L_GROMACS_Potential.png |thumb | 300px | Figure 38: Gromacs energy of P281L]]
 
|}
 
|}
   
Line 1,612: Line 1,612:
 
| 3492.54
 
| 3492.54
 
| -3394.39
 
| -3394.39
| [[File:PAH_G312D_GROMACS_Bond.png| 300px]]
+
| [[File:PAH_G312D_GROMACS_Bond.png |thumb | 300px | Figure 39: Gromacs energy of G312D]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,619: Line 1,619:
 
| 198.398
 
| 198.398
 
| -68.8
 
| -68.8
| [[File:PAH_G312D_GROMACS_Angle.png| 300px]]
+
| [[File:PAH_G312D_GROMACS_Angle.png |thumb | 300px | Figure 40: Gromacs energy of G312D]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,626: Line 1,626:
 
| 328802
 
| 328802
 
| -143251
 
| -143251
| [[File:PAH_G312D_GROMACS_Potential.png| 300px]]
+
| [[File:PAH_G312D_GROMACS_Potential.png |thumb | 300px | Figure 41: Gromacs energy of G312D]]
 
|}
 
|}
 
   
 
==== R408W ====
 
==== R408W ====
Line 1,644: Line 1,643:
 
| 4742.06
 
| 4742.06
 
| -5893.55
 
| -5893.55
| [[File:PAH_R408W_GROMACS_Bond.png| 300px]]
+
| [[File:PAH_R408W_GROMACS_Bond.png |thumb | 300px | Figure 42: Gromacs energy of R408W]]
 
|-
 
|-
 
| Angle
 
| Angle
Line 1,651: Line 1,650:
 
| 386.434
 
| 386.434
 
| -483.529
 
| -483.529
| [[File:PAH_R408W_GROMACS_Angle.png| 300px]]
+
| [[File:PAH_R408W_GROMACS_Angle.png |thumb | 300px | Figure 43: Gromacs energy of R408W]]
 
|-
 
|-
 
| Potential
 
| Potential
Line 1,658: Line 1,657:
 
| 6.93881e+08
 
| 6.93881e+08
 
| -2.71572e+08
 
| -2.71572e+08
| [[File:PAH_R408W_GROMACS_Potential.png| 300px]]
+
| [[File:PAH_R408W_GROMACS_Potential.png |thumb | 300px | Figure 44: Gromacs energy of R408W]]
 
|}
 
|}
   
 
===Timerun===
 
===Timerun===
 
The calculation of the time runs could not be done by a script, therefore we ran gromacs for just AMBER03, AMBERGS and CHARMM each with the nsteps of 125, 250, 500 and 1000.
 
The calculation of the time runs could not be done by a script, therefore we ran gromacs for just AMBER03, AMBERGS and CHARMM each with the nsteps of 125, 250, 500 and 1000.
[[Image:PAH_GROMACS_Timeplot.png]]
 
   
  +
{| border="0"
  +
|-
  +
| [[Image:PAH_GROMACS_Timeplot.png | thumb | 500px | Figure 45: shows the runtime vs nsteps in GROMACS for AMBER03, AMBERGS and CHARMM]]
  +
|}
   
 
== Final results and discussion ==
 
== Final results and discussion ==
Line 1,794: Line 1,796:
   
 
=== R261Q ===
 
=== R261Q ===
[[File:PAH_R261Q_interpretation.png|400px]]
 
   
  +
{| border="0"
The illustration shows the residue 261 (red), the residue 238 (yellow), the iron (orange) and the BH4 molecule (blue).
 
  +
|-
  +
| [[File:PAH_R261Q_interpretation.png | thumb | 500px | Figure 46: shows the residue 261 (red), the residue 238 (yellow), the iron (orange) and the BH4 molecule (blue).]]
  +
|}
  +
   
 
The mutation seems to disrupt the H-bond to Thr238. This H-bond is probably necessary for the correct
 
The mutation seems to disrupt the H-bond to Thr238. This H-bond is probably necessary for the correct
Line 1,845: Line 1,850:
   
 
=== P281L ===
 
=== P281L ===
[[File:PAH_R261Q_interpretation.png|400px]]
 
   
  +
{| border="0"
The illustration above shows residue 281 (red), residue 268 (yellow), the iron atom (orange) and the bh4 molecule (blue).
 
  +
|-
  +
| [[File:PAH_R261Q_interpretation.png | thumb | 500px | Figure 47: shows residue 281 (red), residue 268 (yellow), the iron atom (orange) and the bh4 molecule (blue) molecule (blue).]]
  +
|}
  +
   
 
This mutation removes the helix-breaker P from the end of a helix. In the prediction of SCWRL the effect of such a mutation
 
This mutation removes the helix-breaker P from the end of a helix. In the prediction of SCWRL the effect of such a mutation
Line 1,856: Line 1,864:
   
 
=== G312D ===
 
=== G312D ===
[[File:PAH_G312D_int.png|400px]]
 
   
  +
{| border="0"
The illustration above shows residue 312 (red), the iron atom (orange) and the BH4 molecule (blue).
 
  +
|-
  +
| [[File:PAH_G312D_int.png | thumb | 500px | Figure 48: shows residue 312 (red), the iron atom (orange) and the BH4 molecule (blue)]]
  +
|}
  +
   
 
The residue 312 has no interactions. The mutation does not seem to influence any parts of the catalytic site or the binding sites.
 
The residue 312 has no interactions. The mutation does not seem to influence any parts of the catalytic site or the binding sites.
Line 1,864: Line 1,875:
   
 
=== R408W ===
 
=== R408W ===
[[File:PAH R408W int.png]]
 
   
  +
{| border="0"
The illustration above shows the residue 408 (red), the residues 311, 309, 308 (yellow), the oligomerization domain (blue) and the H-bonds between the oligomerization domain and the catalytic domain (yellow lines).
 
  +
|-
  +
| [[File:PAH R408W int.png | thumb | 500px | Figure 49: shows the residue 408 (red), the residues 311, 309, 308 (yellow), the oligomerization domain (blue) and the H-bonds between the oligomerization domain and the catalytic domain (yellow lines).]]
  +
|}
  +
   
 
The mutation R408W is in the flexible loop identified in the B-factor analysis during the discussion of the disorder prediction (task 3). This is a "hinge" loop between the oligomerization and the catalytic domain of the protein. As one can see in the illustration above, there are only five H-bonds between these two domains and three of them are defined by R408. The mutation causes several clashes in the prediction of SCWRL and probably these three h-bonds are broken. Therefore the flexibility between the two domains is somehow changed.
 
The mutation R408W is in the flexible loop identified in the B-factor analysis during the discussion of the disorder prediction (task 3). This is a "hinge" loop between the oligomerization and the catalytic domain of the protein. As one can see in the illustration above, there are only five H-bonds between these two domains and three of them are defined by R408. The mutation causes several clashes in the prediction of SCWRL and probably these three h-bonds are broken. Therefore the flexibility between the two domains is somehow changed.

Latest revision as of 19:55, 31 August 2011

Task description

A detailed task description can be found here.

Selection of protein structure

We had the following choice of reference structures for PAH:

Entry Method Resolution (A) Chain Positions
1DMW X-Ray 2.00 A 118-424
1J8T X-Ray 1.70 A 103-427
1J8U X-Ray 1.50 A 103-427
1KW0 X-Ray 2.50 A 103-427
1LRM X-Ray 2.10 A 103-427
1MMK X-Ray 2.00 A 103-427
1MMT X-Ray 2.00 A 103-427
1PAH X-Ray 2.00 A 117-424
1TDW X-Ray 2.10 A 117-424
1TG2 X-Ray 2.20 A 117-424
2PAH X-Ray 3.10 A/B 118-452
3PAH X-Ray 2.00 A 117-424
4PAH X-Ray 2.00 A 117-424
5PAH X-Ray 2.10 A 117-424
6PAH X-Ray 2.15 A 117-424

All these structures have in common that they did not solve the structure of the whole PAH protein. They only solve the catalytic domain of PAH, the missing parts are the tetramerisation domain and the regulatory domain which are located at the N- and C- terminal ends. In addition, there is no complete true apo structure available either. All structures have at least a Fe2+ atom bound. Because of this we thought it might be better if we select a structure which has all reaction components or at least most of them bound in the catalytic site in order to get a good picture of the binding site configuration. Though, only 1KW0 and 1MMK fulfilled the constrains that all reaction components are bound.

In the end we did not select 1KW0 or 1MMK, we decided us for the structure 1J8U which is complexed with Fe2+ and BH4 (5,6,7,8-TETRAHYDROBIOPTERIN). Only. This has simple reasons. First of all it has the lowest resolution (1.5 Angstrom) and secondly we already used this structure in previous task as our reference structure for PAH. So we think to keep our experiments more consistent we should stay with this structure. Furthermore, we identified this structure to have no gaps and it solves the complete catalytic domain (as all available structures). Also, the R-Value looked good to us which is 0.157.

To sum it up our selected structure 1J8U has the following experimental metrics (taken from PDBe):

Figure 1: Summary of experimental details for structure 1J8U

Mapping mutations to the structure

We identifier the following functional residues and catalytic sites with the help of UniProt entry P00439 and Catalytic Site Atlas. We looked for catalytic sites in the structure of 1J8U.

We identified the following functional residues and catalytic sites:

  • HIS 285, functional part: side chain (from CSA)
  • HIS 290 (from UniProt)
  • GLU 330 (from UniProt)
  • SER 349, functional part: side chain (from CSA)

In the following picture we can see the Fe+2 atom as a brown sphere, BH4 as a cloud of green blue and red spheres, the location of the mutated residues in orange (mutation I65T and R71H are not included) and the four identified catalytic sites in yellow:

Figure 2: visualization of catalytic site residues and position of selected mutations



I65T

This mutation is not part of our structure but we would say probably no effect on catalytic site because it is too far away.

R71H

This mutation is not part of our structure but we would say probably no effect on catalytic site because it is too far away.

R158Q

Probably no effect on catalytic site because it is too far away.

R261Q

Probably no effect on catalytic site because it is too far away.

T266A

No direct influence on catalytic site residue. However, this residue is located what we would define as the catalytic center.

P275S

Probably no effect on catalytic site because it is too far away.

T278N

No direct influence on catalytic site residue. However, this residue is located what we would define as the catalytic center. ´

P281L

Probably direct influence on catalytic site residue HIS 285. In addition, this residue is located what we would define as the catalytic center.

G312D

Probably no effect on catalytic site because it is too far away.

R408W

Probably no effect on catalytic site because it is too far away.

Binding Site

Figure 3: Binding site of the compounds visualized in orange

To get the binding sites of the two compounds (red) we selected the residues (orange) within 6 Angstrom range of the compounds.

Binding residues:

245 247-251 254-255 263-266 281 285-286 290 307 321-322 325-326 330 345-346
Figure 4: Binding site of the compounds visualized in orange, mutated residues in range of the binding site are in white and residues out of range of the binding site are colored in blue


In this case the residues in range of the compounds are orange, the mutated residues out of range are blue and the mutated residues in range are white. The residues 266 and 281 are affected by our mutations and in range of the compounds.

Introducing mutations to 1J8U

Introducing mutations to 1J8U with SCWRL

We had to employ several steps to introduce our mutated residues to our structure with SCWRL:

1. extract amino acid sequence from PDB file

/apps/scripts/repairPDB 1J8U.pdb -seq > 1J8U_seq.txt

2. convert all upper case residues to lower case

vim 1J8U_seq.txt
:%s/.*/\L&/g

3. create one sequence file for each mutation and put the residue to mutate as an uppercase letter

4. execute SCWRL for each mutation

/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_R158Q.txt -o 1J8U_R158Q.pdb | tee scwrl_r158q.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_R261Q.txt -o 1J8U_R261Q.pdb | tee scwrl_r261q.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_T266A.txt -o 1J8U_T266A.pdb | tee scwrl_t266a.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_P275S.txt -o 1J8U_P275S.pdb | tee scwrl_p275s.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_T278N.txt -o 1J8U_T278N.pdb | tee scwrl_t278n.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_P281L.txt -o 1J8U_P281L.pdb | tee scwrl_p281l.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_G312D.txt -o 1J8U_G312D.pdb | tee scwrl_g312d.out
/apps/scwrl4/Scwrl4 -i 1J8U.pdb -s 1J8U_seq_R408W.txt -o 1J8U_R408W.pdb | tee scwrl_r408w.out

SCWRL versus PyMol: Comparison of the rotation of the side chains

I65T

Could not compare the side chain of this mutation since this position is not included in 1J8U.

R71H

Could not compare the side chain of this mutation since this position is not included in 1J8U.

R158Q

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 5: sidechain of R158Q


As seen in the picture the mutated side chain of SCWRL (in pink) points now in the same direction as the lower part of the WT side chain (seen in orange). In contrast to that the calculated side chain rotation of pymol points towards the adjacent alpha-helix.

R261Q

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 6: sidechain of R261Q


All three side chains point into the same direction. However, we could observe that the side chain calculated by SCWRL is rotated around the Y-axis.

T266A

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 7: sidechain of T266A


All three side chains point into the same direction. the rotation of pymol and SCWRL is the same. However, this is not surprising since alanine has no side chain.

P275S

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 8: sidechain of P275S


All three side chains point into the same direction. Also the rotation of the mutatant side chain of pymol and SCWRL is the same.

T278N

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 9: sidechain of T278N


The distinction between the mutated side chain position of pymol and SCWRL is that the mutated side chain of SCWRL is flipped to the empty C branch of the WT and pymols side chain is flipped to the CO branch of the WT. Hence, we may assume that they form different polar interactions.

P281L

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 10: sidechain of P281L


The two mutated side chains of pymol and SCWRL point into the same direction. However, they are differently rotated. The side chain produced by poymol points to the FE atom and the side chain produced by SCWRL points to the BH4 molecule.

G312D

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 11: sidechain of G312D


The mutated side chains of pymol and SCWRL are rotated into different directions by approximately 190° around the Y-axis. This leads to a different orientation of the COO ends in pymol and SCWRL.

R408W

Orange: side chain of the WT Yellow: side chain of mutated residue (PyMol) Pink: side chain of mutated residue (SCWRL)

Figure 12: sidechain of R408W


Also in our last mutation the mutated residue is rotated differently in pymol and SCWRL. The mutated residue of SCWRL points into the same direction as the WT whereas the the side chain produced by pymol is somehow horizontal to that.

Hydrogen bond network of the WT and mutants

The given H-bond distances in the tables below are calculated from N-O. However, normally an H-bond is defined to be from NH-O. Hence, this explains the great values for H-bonds we obtained during our measurements. So we assume that the distance from N-H is 1 Angstrom. In order to get the "real" H-bond length now we have to substract this 1 Angstrom from our N-O distance. For example if our measured distance was 3 Angstrom then our corrected H-bond distance would be 3-1 = 2 Angstrom.

Furthermore, it seems like SCWRL confused the atom indices during its calculation. This has the effect that they may differ from the original structure. This leads to the problem that the atom index can not be taken directly to compare whether an H-bond in the original structure and in the mutated structure go from and to the same atoms. In order to solve this problem we set an remark whether this is the same h-bond in the WT or mutant structure.

I65T

We could not analyze the hydrogen bonding network of this position since it is not included in the J18U PDB structure.

R71H

We could not analyze the hydrogen bonding network of this position since it is not included in the J18U PDB structure.

R158Q

R158 Q158
Figure 13a: Hydrogen network of R158
Figure 13b: Hydrogen network of Q158

WT (R158) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
R158 SC E280 SC N 344 O 1360 2.9 -
R158 SC E280 SC N 343 O 1365 2.8 -
R158 SC E141 SC N 344 O 201 3.6 -
R158 SC Y154 SC N 343 O 306 3 -
R158 BB Y154 BB N 334 O 298 2.9 -
R158 BB F161 BB O 337 N 363 3.2 -
R158 BB A162 BB O 337 N 374 2.9 -


Mutant (Q158) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
Q158 SC Y154 SC N 341 O 306 3.1 -
Q158 SC I269 SC N 341 O 1264 3.5 -
Q158 BB Y154 BB N 334 O 298 2.9 -
Q158 BB F161 BB O 337 N 363 3.2 -
Q158 BB A162 BB O 337 N 374 2.9 -

R261Q

R261 Q261
Figure 14a: Hydrogen network of R261
Figure 14b: Hydrogen network of Q261

WT (R261) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom To Atom Distance (Ansgstrom)
R261 SC THR238 BB NH2 O 2.3
R261 BB LEU258 BB N O 2.9
R261 BB ARG241 BB O N 2.9


Mutant (Q261) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom To Atom Distance (Ansgstrom)
Q261 BB LEU258 BB N O 2.9
Q261 BB ARG241 BB O N 2.9

T266A

T266 A266
Figure 15a: Hydrogen network of T266
Figure 15b: Hydrogen network of A266

WT (T266) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
T266 SC E286 SC O 1247 O 1413 3.1 -
T266 BB E286 SC N 1241 O 1413 2.8 Same H-Bond as A266 to E286


Mutant (A266) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
A266 BB E286 SC N 1507 O 1710 2.8 Same H-Bond as T266 to E286


P275S

P275 S275
Figure 16a: Hydrogen network of P275
Figure 16b: Hydrogen network of S275


WT (P275) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
P275 BB R270 SC O 1320 N 1286 2.9 Same H-Bond as S275 to R270


Mutant (S275) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
S275 BB R270 SC O 1609 N 1281 3.1 Same H-Bond as P275 to R270


T278N

T278 N278
Figure 17a: Hydrogen network of T278
Figure 17b: Hydrogen network of N278

WT (T278) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
T278 SC E280 BB O 1350 O 1361 2.7 -
T278 SC E280 BB O 1350 N 1358 3.3 Same H-Bond as N278 to E280


Mutant (N278) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom From Atom Index To Atom To Atom Index Distance (Ansgstrom) Remark
N278 SC E280 BB O 1345 N 1353 3.4 Same H-Bond as T278 to E280


P281L

P281 L281
Figure 18a: Hydrogen network of P281
Figure 18b: Hydrogen network of L281

WT (P281) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom To Atom Distance (Ansgstrom)
P281 BB TYR268 SC O OH 2.3


Mutant (Q261) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom To Atom Distance (Ansgstrom)
L281 BB TYR268 SC O OH 2.3

G312D

G312 D312
Figure 19a: Hydrogen network of G312
Figure 19b: Hydrogen network of D312

WT (G312) H-Bonds

There are no hydrogen bonds to residue 312.


Mutant (Q261) H-Bonds

There are no hydrogen bonds to residue 312.

R408W

R408 W408
Figure 20a: Hydrogen network of R408
Figure 20b: Hydrogen network of W408

WT (R408) H-Bonds

From AA From SC/BB? To AA To SC/BB? From Atom To Atom Distance (Ansgstrom)
R408 SC LEU311 BB NH2 O 2.8
R408 SC ALA309 BB NH2 O 3.0
R408 SC LEU308 BB NH2 O 2.9


Mutant (W408) H-Bonds

Clashes with PHE410 and VAL412.

FoldX

Execution

We run foldX in the runfile mode and used the following runfile:

<TITLE>FOLDX_runscript;
<JOBSTART>#;
<PDBS>#;
<BATCH>list_stability.txt;
<COMMANDS>FOLDX_commandfile;
<Stability>Stability.txt;
<END>#;
<OPTIONS>FOLDX_optionfile;
<Temperature>298;
<R>#;
<pH>7;
<IonStrength>0.050;
<water>-CRYSTAL;
<metal>-CRYSTAL;
<VdWDesign>2;
<OutPDB>false;
<pdb_hydrogens>false;
<END>#;
<JOBEND>#;
<ENDFILE>#; 

This says we are calculating the stability (energy) of the PDB structures defined in the list_stability.txt with the following content:

1J8U.pdb
_1J8U_G312D.pdb
_1J8U_P275S.pdb
_1J8U_P281L.pdb
_1J8U_R158Q.pdb
_1J8U_R261Q.pdb
_1J8U_R408W.pdb
_1J8U_T266A.pdb
_1J8U_T278N.pdb

The other lines define some options like that the calculation runs at 298K and with a physiological PH of 7. Then we executed this runfile with the following command:

/apps/FoldX_30b5/foldx -runfile runfile_stability.txt | tee stability_stdout.txt


Results

total energy Backbone Hbond Sidechain Hbond Van der Waals Electrostatics Solvation Polar Solvation Hydrophobic Van der Waals clashes entropy sidechain entropy mainchain sloop_entropy mloop_entropy cis_bond torsional clash backbone clash helix dipole water bridge disulfide electrostatic kon partial covalent bonds energy Ionisation Entropy Complex Number of Residues
1J8U.pdb 13.58 -196.05 -55.77 -379.28 -19.47 492.46 -495.68 34.69 194.54 454.27 0 0 0 11.69 227.66 -15.13 -14.15 0 0 0 1.45 0 307
_1J8U_G312D.pdb 93.54 -193.28 -49.89 -380.58 -22.09 497.08 -497.31 90.61 191.44 452.73 0 0 0 18.93 227.32 -15.85 0 0 0 0 1.77 0 307
_1J8U_P275S.pdb 78.76 -193.22 -49.86 -378.73 -19.74 493.76 -494.26 72.12 190.73 453.21 0 0 0 18.92 228.52 -15.95 0 0 0 0 1.77 0 307
_1J8U_P281L.pdb 77.32 -193.26 -49.88 -379.66 -20.05 494.58 -496.29 72.83 191.19 452.95 0 0 0 18.99 227.31 -15.95 0 0 0 0 1.87 0 307
_1J8U_R158Q.pdb 77.89 -193.3 -47.69 -377.81 -17.08 490.87 -494.58 71.81 189.62 451.69 0 0 0 18.72 227.12 -16.12 0 0 0 0 1.76 0 307
_1J8U_R261Q.pdb 75.18 -193.14 -48.65 -378.39 -19.89 492.32 -494.84 71.79 189.89 451.43 0 0 0 18.85 227.08 -15.95 0 0 0 0 1.77 0 307
_1J8U_R408W.pdb 139.7 -191.5 -48.31 -379.92 -19.75 491.96 -498.23 137.59 189.86 452.27 0 0 0 18.81 226.94 -14.86 0 0 0 0 1.77 0 307
_1J8U_T266A.pdb 74.01 -193.48 -49.47 -378.06 -19.73 491.38 -494.45 71.98 190.15 450.98 0 0 0 18.87 227.18 -15.95 0 0 0 0 1.77 0 307
_1J8U_T278N.pdb 79.9 -192.73 -49.32 -379.04 -19.77 493.36 -495.18 74.69 190.87 451.81 0 0 0 19.38 227.34 -15.95 0 0 0 0 1.79 0 307

As expected our WT structure has the lowest total energy with a value of 13.58 kcal/mol. Interestingly, the mutations R408W and G312D cause the most increase of energy to the protein with values of 139.7 kcal/mol and 93.54 kcal/mol. The other mutations are also harmful in terms of stability to the protein. However, these mutations only introduce a total energy of 75-79 kcal/mol.

Minimise

Mutation Energy
WT -7383.985291
I65T -
R71H -
R158Q -7400.825142
R261Q -7456.793410
T266A -7392.572699
P275S -7418.432874
T278N -7379.215571
P281L -7401.621858
G312D -5643.645312
R408W -5438.301688

Gromacs

Mutations

Mutation Steps Potential Energy Maximum Force Norm of Force
WT 36 -3.7828219e+04 2.0795129e+02 2.1377096e+01
R158Q 328 -3.7326676e+04 5.1247034e+02 1.7521566e+01
R261Q 363 -3.7302664e+04 1.4321185e+02 1.3723365e+01
T266A 315 -3.7422707e+04 3.7855130e+02 1.6683250e+01
P275S 256 -3.7569789e+04 3.6591925e+02 1.9855389e+01
T278N 272 -3.7567461e+04 6.8363385e+02 2.0818382e+01
P281L 336 -3.7656289e+04 1.6796155e+02 1.4324168e+01
G312D 334 -3.7583254e+04 8.6771161e+02 2.3178673e+01
R408W 320 -3.6683172e+04 6.0292523e+02 2.2048130e+01


I65T

Could not compare the side chain of this mutation since this position is not included in 1J8U.

R71H

Could not compare the side chain of this mutation since this position is not included in 1J8U.

R158Q

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 726.495 230 2165.47 -1392.61
Figure 21: Gromacs energy of R158Q
Angle 2351.59 22 189.806 80.2662
Figure 22: Gromacs energy of R158Q
Potential -36202.9 740 3182.59 -4873.47
Figure 23: Gromacs energy of R158Q

R261Q

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 705.135 210 2053.34 -1228.48
Figure 24: Gromacs energy of R261Q
Angle 2356.01 21 179.696 95.0119
Figure 25: Gromacs energy of R261Q
Potential -36270.8 690 3020.89 -4437.22
Figure 26: Gromacs energy of R261Q

T266A

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 740.335 240 2209.81 -1459.51
Figure 27: Gromacs energy of T266A
Angle 2352.51 22 192.08 65.6116
Figure 28: Gromacs energy of T266A
Potential -36283.8 750 3239.42 -4913.99
Figure 29: Gromacs energy of T266A

P275S

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 911.556 410 -nan -2575.37
Figure 30: Gromacs energy of P275S
Angle 2338.07 33 -nan -61.3735
Figure 31: Gromacs energy of P275S
Potential -36106.6 1000 -nan -6645.9
Figure 32: Gromacs energy of P275S

T278N

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 778.485 280 2383 -1740.86
Figure 33: Gromacs energy of T278N
Angle 2344.25 25 207.028 26.8598
Figure 34: Gromacs energy of T278N
Potential -36291.6 820 3493.71 -5504.87
Figure 35: Gromacs energy of T278N

P281L

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 728.439 220 2141.57 -1346.93
Figure 36: Gromacs energy of P281L
Angle 2352.62 21 187.673 80.9581
Figure 37: Gromacs energy of P281L
Potential -36549.7 740 3172.53 -4826.89
Figure 38: Gromacs energy of P281L

G312D

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 1074.93 550 3492.54 -3394.39
Figure 39: Gromacs energy of G312D
Angle 2396.13 28 198.398 -68.8
Figure 40: Gromacs energy of G312D
Potential -13546.2 23000 328802 -143251
Figure 41: Gromacs energy of G312D

R408W

Category Average Err.Est. RMSD Tot-Drift Graph
Bond 1508.76 980 4742.06 -5893.55
Figure 42: Gromacs energy of R408W
Angle 2482.65 99 386.434 -483.529
Figure 43: Gromacs energy of R408W
Potential 4.48468e+07 4.3e+07 6.93881e+08 -2.71572e+08
Figure 44: Gromacs energy of R408W

Timerun

The calculation of the time runs could not be done by a script, therefore we ran gromacs for just AMBER03, AMBERGS and CHARMM each with the nsteps of 125, 250, 500 and 1000.

Figure 45: shows the runtime vs nsteps in GROMACS for AMBER03, AMBERGS and CHARMM

Final results and discussion

Comparison of the delta energies of the different methods and mutations

Mutation FoldX – Total WT Energy FoldX – Total Mutant Energy FoldX – DELTA E Minimise – Total WT Energy Minimise– Total Mutant Energy Minimise – DELTA E Gromacs – Total WT Energy Gromacs – Total Mutant Energy Gromacs – DELTA E
R158Q 13.58 77.89 64.31 -7383.985291 -7400.825142 -16.839851 -37828.219 -37326.676 501.543
R261Q 13.58 75.18 61.6 -7383.985291 -7456.793410 -72.808119 -37828.219 -37302.664 525.555
T266A 13.58 74.01 60.43 -7383.985291 -7392.572699 -8.587408 -37828.219 -37422.707 405.512
P275S 13.58 78.76 65.18 -7383.985291 -7418.432874 -34.447583 -37828.219 -37569.789 258.43
T278N 13.58 79.9 66.32 -7383.985291 -7379.215571 4.76972 -37828.219 -37567.461 260.758
P281L 13.58 77.32 63.74 -7383.985291 -7401.621858 -17.636567 -37828.219 -37656.289 171.93
G312D 13.58 93.54 79.96 -7383.985291 -5643.645312 1740.339979 -37828.219 -37583.254 244.965
R408W 13.58 139.7 126.12 -7383.985291 -5438.301688 1945.683603 -37828.219 -36683.172 1145.047

I65T

We could not perform a structure based mutation analysis for this mutation since it was not solved in our structure.

R71H

We could not perform a structure based mutation analysis for this mutation since it was not solved in our structure.

R158Q

The mutation from amino acid arginine to glutamine on position 158 is not really closely located to any functional or catalytic residue. However, it is located in the catalytic domain of PAH (as all our mutations, though with exception of I65T and R71H which are located on the regulatory domain). Hence, we do not assume that this mutation has any direct influence on these sites.

When we compared the rotation of the mutant generated by SCWRL with the rotation of the mutant generated by pymol with the rotation of the WT side chain we found out that the mutated side chain of SCWRL now in the same direction as the lower part of the WT side chain. In contrast to that the calculated side chain rotation of pymol points towards the adjacent alpha-helix.

During our H-bond network analysis we found out that the WT, R158 is strongly interconnected with other residues by h-bonds. To be more precise we found 7 h-bonds, 4 h-bonds are side chain to side chain h-bonds and the other three are backbone to backbone h-bonds. When mutations are incorporated side chain to side chain and side chain to backbone interactions run in danger to get lost, hence the protein stability might be affected negatively. Whereas h-bonds between backbone to backbone and backbone to side chain are not affected by mutations since they start from a backbone which always stays as it is.

So we found that three side chain to side chain interactions got lost. In our mutant Q158 we now find the three backbone to backbone h-bonds as in our WT and two side chain to side chain h-bonds. One of these h-bonds is the old one Q158 -> Y154 which got preserved and we also found a new h-bond which does not exists in our WT, which is Q158 -> I269. However, the distance of this H-bond is 3.5 angstrom (real h-bond length is 2.5) which is kind of far apart for an h-bond. We assume that the threshold for an H-bond is 2.5 angstrom so we guess this h-bond interaction here is very weak.

In our energy calculations with the methods foldX, minimise and gromacs we found indeed an increase of the energy which is visible from a positive delta value. However, this positive delta value occurred only for the calculations of gromacs and foldx. Minimise calculated a slighly negative delta value which means that our mutant is more stable than the WT. This seems to be a little awkward since we found out that several H-bonds get lost due to the mutation.

R261Q

Figure 46: shows the residue 261 (red), the residue 238 (yellow), the iron (orange) and the BH4 molecule (blue).


The mutation seems to disrupt the H-bond to Thr238. This H-bond is probably necessary for the correct positioning of the helix next to R261 in the catalytic site. This helix contains the BH4 binding site. A shift of the helix would result in a change in the distance of the iron atom and the BH4 molecule. These two molecules need to interact in the reaction. Therefore it is likely, that the catalytic activity of this mutant is at least lowered. The illustration above shows the key players.

FoldX predicts an increase in energy, mostly caused by Van-der-Waals-clashes. Minimise predict the mutant to be more stable than the WT. Gromacs indicates an increase in potential energy, whereas the maximum Force decreased and the norm of force also decreased. Probably this is caused by a loss of interactions, one of them is probably the H-bond between residue 238 and 261.

T266A

The mutation from amino acid threonine to alanine on position 266 is not really closely located to any functional or catalytic residue. However, it is located in the catalytic domain of PAH (as all our mutations, though with exception of I65T and R71H which are located on the regulatory domain). Hence, we do not assume that this mutation has any direct influence on these sites.

When we compared the rotation of the mutant generated by SCWRL with the rotation of the mutant generated by pymol with the rotation of the WT side chain we found out that all three side chains point into the same direction. The rotation of pymol and SCWRL is the same. However, this is not surprising since alanine has no side chain.

During our H-bond network analysis we found out that the WT, T266 is interconnected with other residues by h-bonds. To be more precise we found 2 h-bonds, 1 h-bonds is a side chain to side chain h-bond and the other one is a backbone to side chain h-bond. When mutations are incorporated side chain to side chain and side chain to backbone interactions run in danger to get lost, hence the protein stability might be affected negatively. Whereas h-bonds between backbone to backbone and backbone to side chain are not affected by mutations since they start from a backbone which always stays as it is.

So we found that one side chain to side chain interaction got lost. As expected our mutant A266 preserved the backbone to side chain h-bond from A266 to E286.

In our energy calculations with the methods foldX, minimise and gromacs we found indeed an increase of the energy which is visible from a positive delta value. However, this positive delta value occurred only for the calculations of gromacs and foldx. Minimise calculated a slighly negative delta value which means that our mutant is more stable than the WT. This seems to be a little awkward since we found out that several H-bonds get lost due to the mutation.

P275S

The mutation from amino acid proline to serine on position 275 is not really closely located to any functional or catalytic residue. However, it is located in the catalytic domain of PAH (as all our mutations, though with exception of I65T and R71H which are located on the regulatory domain). Hence, we do not assume that this mutation has any direct influence on these sites.

When we compared the rotation of the mutant generated by SCWRL with the rotation of the mutant generated by pymol with the rotation of the WT side chain we found out that all three side chains point into the same direction. Also the rotation of the mutatant side chain of pymol and SCWRL is the same.

During our H-bond network analysis we found out that the WT, P158 is interconnected with one other residue by an h-bond. This is an backbone to side chain h-bond. When mutations are incorporated side chain to side chain and side chain to backbone interactions run in danger to get lost, hence the protein stability might be affected negatively. Whereas h-bonds between backbone to backbone and backbone to side chain are not affected by mutations since they start from a backbone which always stays as it is.

As expected this mutation is preserved in our mutant S275. However the distance of this h-bond increased from 2.9 angstrom in the WT to 3.1 angstrom in the mutant. So we guess that this might have a minor influence on the protein stability.

In our energy calculations with the methods foldX, minimise and gromacs we found indeed an increase of the energy which is visible from a positive delta value. However, this positive delta value occurred only for the calculations of gromacs and foldx. Minimise calculated a slighly negative delta value which means that our mutant is more stable than the WT.

T278N

The mutation from amino acid threonine to asparagine on position 278 is not really closely located to any functional or catalytic residue. However, it is located in the catalytic domain of PAH (as all our mutations, though with exception of I65T and R71H which are located on the regulatory domain). Hence, we do not assume that this mutation has any direct influence on these sites.

When we compared the rotation of the mutant generated by SCWRL with the rotation of the mutant generated by pymol with the rotation of the WT side chain we found out that the distinction between the mutated side chain position of pymol and SCWRL is that the mutated side chain of SCWRL is flipped to the empty C branch of the WT and pymols side chain is flipped to the CO branch of the WT. Hence, we may assume that they form different polar interactions.

During our H-bond network analysis we found out that the WT,T278, is interconnected with other residues by h-bonds. To be more precise we found 2 h-bonds, which are side chain to back bone h-bonds. When mutations are incorporated side chain to side chain and side chain to backbone interactions run in danger to get lost, hence the protein stability might be affected negatively. Whereas h-bonds between backbone to backbone and backbone to side chain are not affected by mutations since they start from a backbone which always stays as it is.

So we found that one side chain to backbone interaction got lost. In our mutant N278 we now find only one H-bond preserved which is the N278 -> E280 H -bond. So we may assume that we have a slightly decreased protein stability. Furthermore the hydrophobic threonine got interchanged with a non hydrophobic residue which also might have an influence.

In our energy calculations with the methods foldX, minimise and gromacs we found indeed an increase of the energy which is visible from a positive delta value.

P281L

Figure 47: shows residue 281 (red), residue 268 (yellow), the iron atom (orange) and the bh4 molecule (blue) molecule (blue).


This mutation removes the helix-breaker P from the end of a helix. In the prediction of SCWRL the effect of such a mutation will not be seen. It is possible, that the broken helix extends and causes changes in the backbone conformation. Residue 281 is in the near of the bound iron atom. The extension of the helix could cause major changes in the surface of the binding site of the iron, which is essential for the reaction. Therefore the mutation is probably limiting the catalytic activity of PAH. The illustration above shows the key players.

G312D

Figure 48: shows residue 312 (red), the iron atom (orange) and the BH4 molecule (blue)


The residue 312 has no interactions. The mutation does not seem to influence any parts of the catalytic site or the binding sites. This mutation is probably not influencing the enzyme activity.

R408W

Figure 49: shows the residue 408 (red), the residues 311, 309, 308 (yellow), the oligomerization domain (blue) and the H-bonds between the oligomerization domain and the catalytic domain (yellow lines).


The mutation R408W is in the flexible loop identified in the B-factor analysis during the discussion of the disorder prediction (task 3). This is a "hinge" loop between the oligomerization and the catalytic domain of the protein. As one can see in the illustration above, there are only five H-bonds between these two domains and three of them are defined by R408. The mutation causes several clashes in the prediction of SCWRL and probably these three h-bonds are broken. Therefore the flexibility between the two domains is somehow changed. PAH exists in the cell in an equilibrium between its homodimer and its homotetramer form. The dimer is less active than the tetramer form. It is supposed there is a mechanism, which changes the positions of the two domains relative to each other in order to realize this equilibrium. Through the dramatic change of the h-bonds between these two domains, this mechanism is probably broken, perhaps oligomerization itself is at least to some part broken. Therefore R408W is probably a dramatic change, which results in a dramatic decrease of PAH activity.