MEF2C_analysis

Analysis of variants of MEF2C for Dr Nicky Wiffin (Wellcome centre for Human Genetics, University of Oxford).

The model used was a pre-made Swissmodel homodimer based upon PDB:6BYY. (89.47% identity).

Clean-up

PyMol code to add DNA and cleanup:

load coords_2.pdb, swissmodel
fetch 6byy
select hydata, resn HOH and (byres (chain L+K and polymer) expand 3) and  (byres (chain A+B and polymer) expand 3)
create combo, hydata or (chain L+K and polymer) or swissmodel

Do note, I therefore kept some trapped waters at the interface to make the analysis more accurate.

Energy minimisation

The model was energy minimised against the 6BYY ED map with a decreasing constraint (5 cycles FastRelax at 30x, 5 at 20x, 5 at 10x). A final unconstrained minimisation 5-cycle was not done until after a change (below).

Due to the fact R15 appears to be DNA binding yet the DNA in the model is too short, the DNA was extended in PyMol:

fnab AAAAAA, name=myDNA2, mode=DNA, form=B, dbl_helix=1
align myDNA & chain A & resi 1, /MEF2C.r_1//K/DA`15/
alter myDNA & chain A, resv+=14
alter myDNA & chain A, chain='K'
alter myDNA & chain B, resv+=3
alter myDNA & chain B, chain='L'
sort
remove myDNA & chain L & resi 2
remove myDNA & chain K & resi 15

fnab TAAAAA, name=myDNA2, mode=DNA, form=B, dbl_helix=1
align myDNA2 & chain A & resi 1, /MEF2C.r_1//L/DT`15/
alter myDNA2 & chain A, resv+=14
alter myDNA2 & chain A, chain='L'
alter myDNA2 & chain B, resv+=3
alter myDNA2 & chain B, chain='K'
sort
remove myDNA2 & chain K & resi 2
remove myDNA2 & chain L & resi 15

alter all, segi=''
sort

The new parts ± 2 of the DNA were model were energy minimised with 15 FastRelax cycles. Then the whole model was energy minimised for 5 cycles.

Model additions

The model covers only 2-92. The protein beyond 136 is completely unstructured —the transactivation region. The potential helix 112-136 may go anywhere in the model as there is a large disordered linker before it.

To ascertain whether to add anything to the model other structures of MEF2A/B were checked.

• 1LEW shows a 12-mer in the 269 region bound to MAP kinase p38 —beyond the structured region
• 3P57 binds p300 TAZ2 domain in opposite side to the DNA, but no mutations studied fall there
• 1N6J likewise (CABIN1)

NB. MEF2 family binds to the mef2 motif. Thus, there is no MEF1 family.