VMD-L Mailing List

From: Vermaas, Joshua (Joshua.Vermaas_at_nrel.gov)
Date: Thu Mar 15 2018 - 12:21:21 CDT

Hi Anna,

I'd be surprised if there was a ready-made patch for your specific
situation (at least in CHARMM). When I was looking at a bacterial
reaction center, which features a non-heme iron coordinated somewhat
similarly to what you describe (4 HSD + a glutamate), the solution was a
custom patch (given below) that enforced all the interactions as though
they were bonds and kept the charge on the iron reasonable. Since bonded
atoms are excluded from direct electrostatic interaction, this was good
enough to catch the general electrostatics of the center for distant
parts of the protein that we were actually studying, and the bonds kept
the interaction geometry intact so the iron didn't float off. A more
rigorous approach would be to explicitly parameterize your clusters,
somewhat like what was done recently for cobalamin
(10.1021/acs.jctc.7b01236).

-Josh

PRES RCFE 1.00 ! patch for FE2+ in PRC
                       ! 2 bonds to COO- terminal of GLU
                       ! 4 bonds to unprotonated N of the 4 HIS (HSD)
! usage in psfgen (the order must be precisely the following):
! patch RCFE HSD230 HSD266 HSD219 HSD190 GLU234 FE
GROUP !
! HSD 1 - atom types !
! containing letter W ! 1HE1 2HE1
ATOM 1CE1 CPW2 0.27 ! | |
ATOM 1HE1 HR1 0.13 ! --1CE1 2CE1--
ATOM 1NE2 NRW -0.65 ! || ||
ATOM 1CD2 CPW1 0.25 ! || ||
ATOM 1HD2 HR3 0.10 ! \\ || || //
! HSD 2 - atom types ! 1HD2-1CD2--1NE2 2NE2--2CD2-2HD2
! containing letter X ! \ /
ATOM 2CE1 CPX2 0.27 ! --3CE1 \ / 4CE2--
ATOM 2HE1 HR1 0.13 ! / \\ \ / // \
ATOM 2NE2 NRX -0.65 ! 3HE1 \\ \ + / // 4HE1
ATOM 2CD2 CPX1 0.25 ! 3NE2----6FE----4NE2
ATOM 2HD2 HR3 0.10 ! 4HD2 / / + \ \ 4HD2
! HSD 3 - atom types ! \ / / \ \ /
! containing letter Y ! ==3CD2 / \ 4CD2==
ATOM 3CE1 CPY2 0.27 ! | |
ATOM 3HE1 HR1 0.13 ! 5OE1 5OE2(-)
ATOM 3NE2 NRY -0.65 ! \\ /
ATOM 3CD2 CPY1 0.25 ! \\ /
ATOM 3HD2 HR3 0.10 ! 5CD
! HSD 4 - atom types ! |
! containing letter Z ! |
ATOM 4CE1 CPZ2 0.27 ! 5HG1-5CG-5HG2
ATOM 4HE1 HR1 0.13 ! |
ATOM 4NE2 NRZ -0.65 ! |
ATOM 4CD2 CPZ1 0.25 !
ATOM 4HD2 HR3 0.10 !
! GLU !
ATOM 5CG CT2 -0.23 !
ATOM 5HG1 HA 0.09 !
ATOM 5HG2 HA 0.09 !
ATOM 5CD CC 0.75 !
ATOM 5OE1 OC1 -0.55 !
ATOM 5OE2 OC2 -0.55 !
! !
GROUP !
! FE !
ATOM 6FE FE 1.00 !

! patching bonds
BOND 6FE 1NE2 6FE 2NE2 6FE 3NE2 6FE 4NE2
BOND 6FE 5OE1 6FE 5OE2

! patching angles
ANGLE 6FE 1NE2 1CE1 6FE 1NE2 1CD2
ANGLE 6FE 2NE2 2CE1 6FE 2NE2 2CD2
ANGLE 6FE 3NE2 3CE1 6FE 3NE2 3CD2
ANGLE 6FE 4NE2 4CE1 6FE 4NE2 4CD2
ANGLE 6FE 5OE1 5CD 6FE 5OE2 5CD
ANGLE 1NE2 6FE 2NE2 1NE2 6FE 3NE2 1NE2 6FE 5OE1 1NE2 6FE 5OE2
ANGLE 2NE2 6FE 3NE2 2NE2 6FE 4NE2 2NE2 6FE 5OE1 2NE2 6FE 5OE2
ANGLE 3NE2 6FE 4NE2 3NE2 6FE 5OE1 3NE2 6FE 5OE2
ANGLE 4NE2 6FE 1NE2 4NE2 6FE 5OE1 4NE2 6FE 5OE2
ANGLE 5OE1 6FE 5OE2

On 03/15/2018 10:28 AM, Anna Petroff wrote:
> Hello,
>
> I'm seeking recommendations on how to identify non-heme coordinations
> between iron and His in VMD/NAMD. From what I've read, there are some
> nice patch options for heme coordination, but I haven't found one
> specific to non-heme coordination. Is there a best-practice way to
> handle this situation?
>
> My protein of interest has a diiron active site. One of the irons is
> coordinated with five histidine residues (all HSE). The other iron is
> coordinated with four histidines (also all HSE), as well as an
> interaction with a glutamine residue that is mediated by a water
> molecule.
>
> I would like to set these coordinations in NAMD2, possibly through a
> combination of patches. If you have experience related to this type of
> problem, I'd really appreciate your thoughts.
>
> Thanks for your time,
> Anna
>