Biology Reference
In-Depth Information
and with the system of internal cavities/tunnels found in other globins
(
Nardini et al., 2007; Salter et al., 2008
).
Tunnel 1 is a straight apolar protein matrix tunnel (
7
˚
in diameter and
16
˚
in length), lined by residues Ile(56)B5, Thr(59)B8, Trp(60)B9,
Phe(63)B12, Phe(93)E11, Ile(97)E15, Phe(145)G7, Pro(148)G10, Ile(149)
G11, Thr(152)G14 and Met(153)G15. Tunnel 2 (
5
˚
in diameter and
10
˚
in length) is a straight and short opening to the haem distal cavity
nestled among residues Tyr(61)B10, Leu(64)B13, Gly(65)B14, Leu(71)
C5, Phe(74)CD1 and Leu(86)E4. Both tunnels host one water molecule
at their solvent side aperture. In addition, a core cavity of
75
˚
3
is located
between the distal and proximal haem sides; the cavity hosts four mutually
hydrogen-bonded water molecules. All the residues lining the protein tun-
nels and the inner cavity are conserved in known Pgbs, which would be
consistent with their implication in diatomic ligand diffusion to and from
the haem, multi-ligand storage and/or (pseudo-)enzymatic reactivity. Such
functional roles would rely on structural principles that are entirely different
from those shown for Xenon cavities in Mb and in truncated 2/2Hbs
(
Brunori & Gibson, 2001; Milani et al., 2005
). Indeed, the structures of fer-
ric
Ma
Pgb
*
in complex with Xenon (in the presence of cyanide or azide as
haem Fe-ligand) clearly show a Xenon atom trapped inside tunnel 1, in a
hydrophobic cavity efficiently sealed by Trp(60)B9 side chain, which moves
into the distal site upon cyanide/azide binding (see below). On the contrary,
no Xenon atoms bind at tunnel 2, due to its short length and more hydro-
philic nature (
Pesce, Tilleman, et al., 2013
).
Molecular dynamics simulations showed that while tunnel 2 is always
accessible to diatomic ligands in both deoxygenated and oxygenated forms
of the protein, the accessibility of tunnel 1 is controlled through the syner-
gistic effect of both the ligation and the oligomerization states of the protein.
In particular, steric hindrance between Phe(93)E11 and the haem-bound
ligand would alter the structural and dynamical behaviour of the B- and
E-helices, thus facilitating the opening of tunnel 1, while dimerisation
would affect the spatial organisation of the G-helix, which, in turn, would
modify the structure of tunnel 1 (
Forti et al., 2011
). More controversial is the
role of Phe(145)G8. Molecular dynamics simulations and electron paramag-
netic resonance experiments have suggested that the accessibility of ligands
through tunnel 1 is also regulated by the side chain of Phe(145)G8, which
can adopt open and closed conformations (
Forti et al., 2011; Van Doorslaer
et al., 2012
). However, there are no crystallographic evidences of alternate
conformations
structures of
Ma
Pgb
*
for
this
residue in all crystal
