posted on 2012-06-14, 00:00authored byAlex R. Jones, Henry J. Russell, Gregory
M. Greetham, Michael Towrie, Sam Hay, Nigel S. Scrutton
Vitamin B12 (cyanocobalamin, CNCbl) and its
derivatives
are structurally complex and functionally diverse biomolecules. The
excited state and radical pair reaction dynamics that follow their
photoexcitation have been previously studied in detail using UV–visible
techniques. Similar time-resolved infrared (TRIR) data are limited,
however. Herein we present TRIR difference spectra in the 1300–1700
cm–1 region between 2 ps and 2 ns for adenosylcobalamin
(AdoCbl), methylcobalamin (MeCbl), CNCbl, and hydroxocobalamin (OHCbl).
The spectral profiles of all four cobalamins are complex, with broad
similarities that suggest the vibrational excited states are related,
but with a number of identifiable variations. The majority of the
signals from AdoCbl and MeCbl decay with kinetics similar to those
reported in the literature from UV–visible studies. However,
there are regions of rapid (<10 ps) vibrational relaxation (peak
shifts to higher frequencies from 1551, 1442, and 1337 cm–1) that are more pronounced in AdoCbl than in MeCbl. The AdoCbl data
also exhibit more substantial changes in the amide I region and a
number of more gradual peak shifts elsewhere (e.g., from 1549 to 1563
cm–1), which are not apparent in the MeCbl data.
We attribute these differences to interactions between the bulky adenosyl
and the corrin ring after photoexcitation and during radical pair
recombination, respectively. Although spectrally similar to the initial
excited state, the long-lived metal-to-ligand charge transfer state
of MeCbl is clearly resolved in the kinetic analysis. The excited
states of CNCbl and OHCbl relax to the ground state within 40 ps with
few significant peak shifts, suggesting little or no homolysis of
the bond between the Co and the upper axial ligand. Difference spectra
from density functional theory calculations (where spectra from simplified
cobalamins with an upper axial methyl were subtracted from those without)
show qualitative agreement with the experimental data. They imply
the excited state intermediates in the TRIR difference spectra resemble
the dissociated states vibrationally (the cobalamin with the upper
axial ligand missing) relative to the ground state with a methyl in
this position. They also indicate that most of the TRIR signals arise
from vibrations involving some degree of motion in the corrin ring.
Such coupling of motions throughout the ring makes specific peak assignments
neither trivial nor always meaningful, suggesting our data should
be regarded as IR spectral fingerprints.