Biology Reference
In-Depth Information
cation.
89
A database, the Birmingham Metabolite Library,
containing 1D and 2D JRES NMR spectra of 208
metabolite standards, has been developed to
aide metabolite identi
processing techniques, such as linear prediction
in the second dimension to improve the signal
to noise ratio, may not always accurately retain
low-intensity peaks.
Most of the early NMR studies dealt with toxi-
cological effects of model hepatic and renal toxins
in rodents; hundreds of samples were analyzed
and data from 1D
1
H spectra provided satisfac-
tory information.
84,85
Although 2D spectra could
provide better information, it is prohibitively
slow (possibly hours or more per sample). On
occasion, it is inevitable that additional time
must be invested to acquire better resolved spec-
tral data to obtain additional information unat-
tainable from simple, fast methodologies. Hence
in recent years, NMR spectroscopists have begun
to develop multidimensional methodologies
speci
information to aide metabolite identi
cation.
90
J-spectroscopy
is plagued by two drawbacks: artifacts from
strong scalar coupling and the signals are
phase-modulated in both dimensions, which
require strong window functions resulting in
sensitivity loss and intensity distortions. Ludwig
and Viant recently detailed improvements in the
2D JRES experiment to address these drawbacks
and presented recommendations for optimal
spectral acquisition and data processing.
91
COSY/TOCSY
Traditionally, the 2D correlation spectroscopy
(COSY)
92
and total correlation spectroscopy
(TOCSY)
93
experiments are used for tracing out
coupled spin systems within a molecule;
thereby, neighboring protons are identi
cally for metabolomics. The 2D experi-
ments fall into two general categories: homo-
nuclear (
1
H-
1
H) and heteronuclear (
1
H-X).
Separation of components in a mixture by differ-
ences in diffusion rates in the second dimension
has also been proposed but has not been widely
applied to metabolomics.
86,87
ed.
Magnetization is transferred through scalar
coupling to generate off-diagonal peaks from
coupled spins in the COSY and the entire
coupling network in the TOCSY experiment.
Singlet peaks which have no coupling partners
will only have peaks on the diagonal in the spec-
trum. The off-diagonal peaks (also referred to as
cross-peaks) provide a way to spread out the
peaks.
In comparing 2D TOCSY to 1D
1
H spectra of
mice urine, Van et al. demonstrated that statis-
tical analysis of 1D
1
H data are biased towards
changes in the higher abundant metabolites
and showed that peaks from low abundance
metabolites were better represented in models
generated from the 2D TOCSY data.
36
Expan-
sions of the aromatic region of a 1D
1
H and 2D
TOCSY spectra of mice urine are shown in
Figure 2
. Recording high-resolution 2D TOCSY
spectra requires a large investment in time and
thus this methodology may not be suitable for
all studies. However, information on lower
abundant metabolites obtainable only from the
2D TOCSY spectra provided a more complete
Homonuclear 2D
J-Resolved Spectroscopy
The small 12 ppm chemical shift dispersion of
1
H often results in highly congested spectra for
complex mixtures that is worsened by the split-
ting of peaks into multiplet structures from
coupling to neighboring protons. In the 2D
J-resolved (JRES) experiments, chemical shifts
and homonuclear scalar couplings are separated
into different dimensions.
88
A broadband
proton-decoupled
1
H spectrum (pJRES) is
obtained from the skyline projection of the 2D
JRES experiment. Peaks that normally appear as
multiplets due to proton
e
proton scalar coupling
are now reduced to single lines. The reducedpeak
congestions in pJRES provide several advan-
tages: (1) more accurate integration of speci
c
metabolites, (2) broadmacromolecule resonances
are minimized, and (3) spin
e
spin coupling
Search WWH ::
Custom Search