Metabolomics: Approaches to assessing
oocyte and embryo quality
R. Singh, K.D. Sinclair *
School of Biosciences, University of Nottingham, Sutton Bonington Campus, Leicestershire LE12 5RD, UK
Abstract
Morphological evaluation remains the primary method of embryo assessment during IVF cycles, but its modest predictive power
and inherent inter- and intra-observer variability limits its value. Low-molecular weight metabolites represent the end products of
cell regulatory processes and therefore reveal the response of biological systems to a variety of genetic, nutrient or environmental
influences. It follows that the non-invasive quantification of oocyte and embryo metabolism, from the analyses of follicular fluid or
culture media, may be a useful predictor of pregnancy outcome following embryo transfer, a potential supported by recent clinical
studies working with specific classes of metabolites such as glycolytic intermediates and amino acids. Such selective approaches,
however, whilst adhering closely to known cellular processes, may fail to harness the full potential of contemporary metabolomic
methodologies, which can measure a wider spectrum of metabolites. However, an important technical drawback with many existing
methodologies is the limited number of metabolites that can be determined by a single analytical platform. Vibrational spectroscopy
methodologies such as Fourier transform infrared and near infrared spectroscopy may overcome these limitations by generating
unique spectral signatures of functional groups and bonds, but their application in embryo quality assessment remains to be fully
validated. Ultimately, a combination of evaluation criteria that include morphometry with metabolomics may provide the best
predictive assessment of embryo viability.
# 2007 Elsevier Inc. All rights reserved.
Keywords: Metabolomics; Oocyte quality; Embryo quality; Metabolites; Mass spectroscopy
1. Introduction
At present morphological assessment is the primary
method used to determine embryo viability during IVF
cycles. It entails observations of the developmental
pattern of embryos during culture such as the timing and
rate of cleavage, fragmentation, inclusion bodies, cell
number and allocation to the inner cell mass [1] and has
the advantage of being a quick, convenient and
inexpensive means of assessment but with modest
predictive value [2,3]. Whereas the majority of human
embryos are transferred during the early cleavage
stages, the majority of embryos from domestic animal
species are transferred at the morula or blastocyst stage.
It follows that greater emphasis is placed on pronuclear
morphology, cleavage timing and cell number in human
embryos which tend to be handled in small numbers [2].
Embryo stage and overall morphology are the primary
determinants in animal embryos where an internationally
recognized system has been in practice for cattle
since 1998 [3]. In general, however, morphological
assessments are hampered by a lack of suitable
standards and the inherent inter- and intra-observer
variability associated with a subjective grading system.
Furthermore, it does not necessarily equate to functional
status, and morphologically normal looking embryos
can still harbor genetic or epigenetic defects [4].
www.theriojournal.com
Theriogenology 68S (2007) S56­S62
* Corresponding author. Tel.: +44 115 951 6053;
fax: +44 115 951 6060.
E-mail address: kevin.sinclair@nottingham.ac.uk (K.D. Sinclair).
0093-691X/$ ­ see front matter # 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.theriogenology.2007.04.007
Quantitative techniques have been developed for the
non-invasive assessment of embryo metabolism and
their value as predictors of embryo viability and
pregnancy outcome is the subject of ongoing investigations
[5­7]. The field is presently moving away from
metabolic target analysis (analysis restricted to single or
selectively defined metabolites) for diagnostic purposes
towards metabolite profiling (which looks for selective
groups of target compounds and their metabolic
intermediates using a single analytical technique) and
a more comprehensive analysis of the metabolome,
which is the dynamic quantitative complement of all
low molecular weight molecules (<1000 Da) present in
cells at a particular physiological or developmental
stage [8]. The hope is that the latter may enhance the
value of embryo metabolism as a predictor of pregnancy
outcome following transfer.
Low-molecular weight metabolites represent the end
products of cell regulatory processes and therefore
reveal the response of biological systems to a variety of
genetic, nutrient or environmental influences. Since
alterations to a cell's physiology, resulting from gene
overexpression or deletion/silencing, are amplified
through the hierarchy of the transcriptome and
proteome, these changes can be more readily measured
through the metabolome. In addition, the metabolome is
downstream of gene function and therefore is a superior
measure of cellular activities at the physiological level
compared to trancriptomic and proteomic approaches.
Increases in mRNA do not always correlate with
increased protein expression [9] and a protein may or
may not be enzymatically active [10]. It can be argued
that changes in the transcriptome or proteome do
not always correspond to altered cellular phenotype.
This provides a further rationale for measuring the
metabolome which could be an important complement
to assessing gene function. The central dogma of
molecular biology describes a linear unidirectional flow
of information from gene to transcript to protein.
Enzymes then regulate metabolic pathways and lead to
phenotypic changes. This `linear' flow of information is
challenged by the fact that cellular phenomena are
intimately networked with various feedback-loops and
therefore should be represented as dynamic gene and
protein complexes interacting with neighboring metabolites
[11].
Metabolomics has, in recent years, proven to be a
robust, high throughput and rapid post-genomic
technology for pattern recognition analyses of biological
samples [8]. Alterations to steady-state concentrations
and transient changes in intracellular
metabolites resulting from processes such as cell
signaling can be readily investigated using metabolomic
techniques such as gas chromatography­mass spectrometry
(GC­MS), liquid chromatography­MS (LCMS),
capillary electrophoresis­MS (CE­MS) and
nuclear magnetic resonance spectroscopy (NMRS),
which are capable of detecting hundreds of individual
chemical structures. Integrative analyses using genomic,
proteomic and metabolomic techniques will help
enhance the understanding of cellular physiology and
lead to the identification of multiple physiological
biomarkers that are not accessible with targeted studies.
In this review we discuss current state-of-the-art
metabolomic techniques and their applicability to the
assessment of embryo quality and in developmental
research.
2. Metabolomic analysis
The ultimate goal of a metabolomic experiment is to
quantify all of the metabolites in a cellular system under
a given set of physiological conditions at a particular
time point. However, measuring the metabolome is a
considerable analytical challenge. This is due to the
labile nature, and wide dynamic and diverse range of
many metabolites, their chemical complexity and
heterogeneity, lack of simple automated analytical
techniques which can quantitatively measure a large
number of metabolites of unknown structure, the
throughput of measurements and the elaborate nature
of extraction protocols [12]. All these challenges need
to be addressed by the analytical platform employed.
Unlike transcriptome analysis, methods to amplify
metabolites and therefore increase sensitivity do not
exist. Furthermore, in contrast to genomics and
proteomics, where only one class of compound is
analyzed, metabolomic-based analyses have to deal
with a diverse class of molecules with different
properties.
Cellular metabolism has to be stopped as quickly as
possible in order to terminate the activities of enzymes.
Tissue samples should be snap-frozen to abruptly
terminate metabolism. The metabolites then can be
isolated by cold methanol or boiling ethanol, cold
methanol:chloroform buffer, organic solvents, perchloric
acid, or alkali. The advantages and disadvantages of
these procedures along with their applications have
been reviewed elsewhere [13].
The platforms for metabolome measurement should
be unbiased, sensitive and have a capacity for highthroughput
analysis to screen a large number of
metabolites. Despite the enormous success in development
of metabolomic technologies, it is not yet possible
R. Singh, K.D. Sinclair / Theriogenology 68S (2007) S56­S62 S57
R.Singh,K.D.Sinclair/Theriogenology68S(2007)S56­S62S58
Table 1
Contemporary metabolomic platforms and their applications
Metabolomic
platform
Application and selectivity Sensitivity
(M)
Examples of metabolite analysis Comments
NMRS Little chemical bias and negligible sample preparation. Magic
angle spinning NMRS can analyse small pieces of intact tissues.
Useful for tissue extracts and biofluids
106
Carbohydrates, oligosaccharides, amino acids,
glycolytic intermediates, carboxylic acids
Provides detailed structural information.
Low sensitivity reduces selectivity. Matrix
effects can limit identification
LC­UV Suitable for identification and quantification of a broad range of
metabolites. No derivatization required. Suitable for thermo-labile
compounds
109
Glycolytic intermediates, carboxylic acids,
amino acids, nucleotides, isoprenoids
Less suitable for identification unless
coupled to MS
GC­MS Broad selectivity, suitable for volatile and non-volatile metabolite
separation, identification and quantification. Non-volatile compounds
require derivatization, so can be difficult to identify. Not suitable for
thermo-labile compounds
1012
Fatty acids, lipids in general, glycolytic
metabolites, carboxylic acids, amino acids
Good separation but moderate throughput
and solvent bias. GC on its own is less
useful for identification unless coupled
to MS
LC­MS Suitable for identification and quantification of a broad range of
metabolites. Derivatization prior to analysis not normally needed.
Suitable for thermo-labile compounds
1015
Glycolytic intermediates, carboxylic acids,
amino acids, nucleotides, isoprenoids
More suitable for polar compounds
(solvent bias). But is also used for lipids.
Matrix effects can limit identification.
Moderate throughput
MS Direct injection MS (DIMS) suitable for identification of a broad
range of metabolites. Organic solvent extraction leads to loss of
some compounds
1015
Carbohydrates, amino acids Rapid throughput, but poor quantification.
Metabolite identification of unknown
compounds requires MS­MS. Matrix
effects can be problematical
LC-LIF Suitable for small metabolites since it has a mass detection limit
between 1018
to 1020
1019
Peptides, amino acids, catecholamines Good for applications where the
compounds have native
fluorescence or can be derivatized
CE-LIF Suitable for small (1018
to 1020
) polar metabolites using very
small volumes. Provides high resolution but has low selectivity
1023
Amino acids, catecholamines, saccharides As above or where the concentration
and/or sample volumes are very small
(volume = nL, concentration = 108
M)
FT-IR Used to obtain metabolic fingerprints for a variety of biofluids
including follicular fluid. Requires no reagents or sample
preparation
106
Vibrations associated with bonds and functional
groups (e.g. C O, N­H, C­N), amino acids,
fatty acids, saccharides
Useful for identification of functional
groups. Very high throughput. Little
chemical bias. Direct use but requires
sample dehydration
to analyze the diverse range of complex metabolites
with a single analytical platform. The techniques most
commonly employed for metabolomic analysis are
NMRS and MS which can be used alone or preceded by
a chromatographic step (Table 1). The data acquired
through these platforms generate a large number of
complex spectral signals, which reflect the metabolic
status of the cell or organism. These large data sets
require powerful statistical and bioinformatic tools for
adequate data analysis and interpretation.
2.1. Nuclear magnetic resonance (NMRS)
techniques
Various NMRS techniques have been used successfully
to analyze the metabolome in biological samples.
The use of this technique has been pioneered by
Nicholson and colleagues [14,15] who used 1
H NMRS
to analyze bio-fluids in order to diagnose disease.
Although not as sensitive as MS, it is potentially fully
quantitative using a single internal standard, requires
little or no sample preparation, is highly reproducible
and is capable of analyzing liquid and solid samples
directly. NMRS can also detect isotopes of many
elements such as 13
C, 31
P, 15
N and can therefore be used
for examining metabolic pathways following exposure
of biological systems to labeled low molecular
substrates [12]. NMRS can be used also to investigate
metabolic changes in intact tissues using magic angle
spinning techniques [16]. The major disadvantage of
NMRS is its relatively low sensitivity, which necessitates
larger volumes. The issue of sensitivity has been
addressed partly by improved detectors and development
of magnets with increased field strength [17].
Whilst limiting the utility of this technique for direct,
non-invasive measurements in the egg or pre-implantation
embryo, the issue of sensitivity may be of less
concern for analysis of metabolites in associated
follicular fluid, granulosa cells or culture media.
2.2. Mass spectrometry techniques
Mass spectrometry (MS) is an extremely powerful
analytical metabolomic tool since it provides a unique
blend of sensitive, rapid, qualitative and potentially
quantitative analysis of a large population of metabolites
[13]. MS can be used alone or in combination with
other chromatographic techniques. Without chromatographic
separation, direct injection mass spectrometry
(DIMS) involves the injection or continuous infusion of
a crude sample extract straight into an electrospray mass
spectrometer, an approach suitable for quick screening
but not particularly quantitative [13]. Mass spectrometric
analysis requires extensive sample preparation
and extraction of samples usually with organic solvents
which may lead to loss of certain compounds. Variable
ionization efficiencies of metabolites hamper absolute
quantification although concentrations of metabolites
can be determined precisely by MS using stable-isotope
internal standards [12].
MS can be combined with chromatographic techniques
such as liquid chromatography (analytes separated
by their chemical properties such as hydrophobicity),
gas chromatography (analytes separated by their boiling
point and interaction with the liquid layer covering the
capillary in the gas phase), capillary electrophoresis
(analytes separated by their mobility in a capillary filled
with electrolyte under the influence of an electric field)
or ultra performance liquid chromatography (UPLC).
Recently developed UPLC systems use columns with
very small particle size (<2 mm), which leads to almost
a doubling of peak capacity and offers the highest
efficiency attainable at relatively high linear velocities.
These result in a 10-fold increase in speed and a threeto
five-fold increase in sensitivity compared to
conventional reversed phase high performance liquid
chromatography (HPLC) [18]. LC­MS and UPLC­MS
can be used for separation, identification and quantification
of a broad array of metabolites and these
techniques are highly sensitive, require no derivatization
and enable analyses of thermolabile compounds
(Table 1). The chromatographic step reduces the
number of competing analytes entering the mass
spectrometer and alleviates ion suppression and matrix
effects in addition to facilitating separation of complex
mixtures of metabolites, leading to improved MS data
quality [19]. Capillary electrophoresis MS is a
promising technology particularly suitable for the
analysis of polar and thermolabile compounds, using
very small volumes. GC­MS and LC­MS have
moderate sample throughput but provide incontrovertible
identification and quantification of individual
compounds in a complex mixture. GC­MS is an
extremely popular metabolomic platform due to its
comprehensive nature and exceptionally high sensitivity.
However, sample preparation is extensive, nonvolatile
metabolites require derivatization prior to
analysis and thermolabile metabolites are essentially
missed. Mass spectral deconvolution is needed to
quantify metabolites that are not resolved by GC.
Modern GC­TOF­MS applications are capable of
faster spectral acquisition, and integrated deconvolution
algorithms successfully accomplish these requirements
[20]. In addition, Fourier transform­ion cyclotron
R. Singh, K.D. Sinclair / Theriogenology 68S (2007) S56­S62 S59
resonance (FT­ICR)­MS analysis is becoming increasingly
popular since it is highly sensitive, and with its
high mass resolution coupled to software that can use
the information in isotope patterns, it can generate
empirical formulae of metabolites directly [21]. The
ability of FT­ICR­MS to generate spectral data relating
to the elemental composition of metabolites suggests
that this technique is set to play a major role in future
metabolomic research.
2.3. Vibrational spectroscopy
Fourier transform infrared spectroscopy (FT-IR) is a
high-throughput, non-invasive technique, which
requires no reagents or sample preparation. It measures
the vibrations of bonds within functional groups [22].
When a sample is subjected to light or electromagnetic
radiation, chemical bonds absorb this light and exhibit
stretching or bending vibrations, which can be,
correlated to single bonds or functional groups of
molecules for identification of unknown compounds. A
large number of investigations have used FT-IR to
analyze cells, tissues and biofluids for rapid detection of
disease or malfunction [22,23]. Although most of the
metabolomic investigations, which use vibrational
spectroscopy have employed FT-IR, the use of Raman
which, unlike FT-IR mainly measures non-polar bonds,
is an emerging technology which has been used to
assess embryo quality (see later). Near Infrared (NIR)
spectroscopy, which primarily measures overtones and
combination vibrations, also has significant analytical
potential and has been used in various metabolomic
investigations including measurement of lactate in
human blood [22].
3. Metabolomic assessment of oocyte and
embryo quality
Most of the relevant investigations into the assessment
of oocyte and embryo viability have, to date, used
targeted metabolic approaches [24] or have undertaken
profiling for a selective class of metabolites such as
amino acids or fatty acids using conventional chromatographic
techniques [25,26]. The emphasis has been on
hypothesis driven research, which has given rise to
much of our current understanding of the metabolic
requirements of mammalian gametes and the preimplantation
embryo [27]. This has since given way, at
least to a certain extent, to metabolomic investigations
that attempt to predict embryo quality. The profiling of
one particular class of metabolite (i.e. amino acids) by
HPLC has attracted greatest attention, and has been
proposed as a means of non-invasive selection of
developmentally competent embryos for transfer [5,6].
A recent retrospective clinical study employed this
non-invasive method of embryo selection, which is
based on the depletion/appearance of amino acids in
culture medium [28]. The study involved 53 cycles of
IVF treatment using ICSI. HPLC analysis of the spent
medium showed that the turnover of three amino acids,
asparagine, glycine and leucine, was significantly
correlated with clinical pregnancy and live birth.
Importantly, this association was found to be largely
independent of other known determinants of pregnancy
such as maternal age, ovarian reserve, embryo cell
number and morphological grading, so that a combination
of assessments (e.g. cell number, morphological
grade and amino acid metabolism) could potentially
increase the chances of selecting developmentally
competent embryos. Allowing for differences in the
composition of embryo culture media between studies,
there appears to be great similarity in the metabolism of
amino acids between pre-implantation embryos of
contrasting mammalian species [26,28,29], indicating
that non-invasive metabolic foot-printing of amino acid
turnover has considerable generic potential to improve
significantly the prospective selection of viable
embryos for transfer.
Another study investigated fatty acid metabolism
and uptake in donated human pre-implantation embryos
by gas chromatography and showed that, compared to
embryos which did not develop beyond the 4-cell stage,
those that did develop beyond this point had significantly
greater concentrations of the unsaturated fatty
acids (particularly linoleic and oleic acids) and a lower
concentration of saturated fatty acids [30]. Consequently,
knowledge of the fatty acid requirements of the
pre-implantation embryo could be used to enhance the
success of IVF treatment. Non-invasive foot-printing of
glycolytic activity has also been used for assessing
embryo viability in mouse embryos [7]. Blastocysts
with low glycolytic activity (i.e. low lactate production),
in a range closer to that of in vivo derived
embryos, were more viable than those with an
abnormally high glycolytic rate. The use of glycolysis
as an index for selecting viable embryos led to a fourfold
increase in pregnancy rate compared to the transfer
of embryos selected at random [7]. Metabolite profiling
for glycolytic intermediates, therefore, offers another
possible criterion for the assessment of embryo
viability.
Less specific approaches than those listed above have
employed techniques such as FT-IR to obtain `biochemical
fingerprints' for the composition of follicular
R. Singh, K.D. Sinclair / Theriogenology 68S (2007) S56­S62S60
fluids from large and small antral luteinized follicles
[23]. Given the importance of oocyte quality in
determining post-fertilization development [31], the
analysis of follicular fluid at the point of egg recovery
could yield valuable predictive information concerning
subsequent embryo viability. The results of Thomas and
colleagues [23] showed that the FT-IR spectra of fluids
collected from large antral follicles were tightly
clustered, indicating a similar biochemical profile,
whereas fluids collected from small follicles were
heterogenous reflecting differences in their maturational
stage. Further investigations are required to relate
the FT-IR spectra of follicular fluid to the developmental
competence of oocytes.
Finally, a recent prospective multi-centre study has
shown that non-invasive metabolomic profiling of
human embryo culture media correlated with pregnancy
outcome [32]. This investigation used both Raman and
Near Infrared Spectroscopy (NIR) to analyze culture
media collected after embryo transfer on day 3
following fertilization. Results from this study showed
that spectral profiles describing differences in ­CH, NH
and ­OH concentrations exhibit discrete differences
between the culture media of embryos that resulted in
pregnancy and those that failed to establish a pregnancy.
The ratio of ­CH to R­OH content in the media is
reflective of oxidative stress. Using wavelength
selective genetic algorithms (Molecular Biometrics,
LLC, Chester, NJ) with Raman, the authors found four
spectral regions associated with these molecular species
to be predictive of pregnancy establishment. Raman
analysis resulted in a specificity of 80% and a sensitivity
of 95% whereas NIR provided a specificity of 83% and
sensitivity of 73%. A similar investigation using NMR,
Raman and NIR, analyzed 228 embryo media, 72
follicular fluid (FF) and 133 seminal plasma samples
and identified specific oxidative stress biomarkers of ROH,
CH, OH and NH groups using wavelength selective
genetic algorithms [33]. Results from this study showed
that unique metabolomic oxidative stress profiles were
consistently produced from discarded culture media,
follicular fluid and seminal plasma and these profiles
correlated well with pregnancy versus non-pregnancy
outcomes.
4. Conclusions and perspectives
Metabolic profiling of follicular fluid or culture
media can usefully serve as a high throughput and noninvasive
means for the assessment of gamete and
embryo viability, which could lower the number of
embryos transferred during human ART procedures.
This would lower the incidence of multiple births, and
lead to a general improvement in pregnancy outcomes
across species. Larger prospective studies, however,
will be required to further refine and validate these
methodologies in order to fully determine their value as
predictors of egg or embryo quality. An important
shortcoming with many of these techniques is the
limited number of metabolites that can be determined
by a single analytical platform under a given set of
conditions. That is, the predictive value of metabolomic
analysis may only be fully realized when the broadest
range of metabolites can be determined. Vibrational
spectroscopy methodologies such as FT-IR and NIR
may overcome these limitations but their application
remains to be fully validated. Ultimately, a combination
of assessment criteria that includes morphometry with
metabolomic analysis may provide the best predictive
assessments of embryo viability.
Finally, two further considerations. First, the focus of
current research endeavors has been directed primarily
towards determining metabolites in culture media with
little attention given to metabolites in follicular fluid or
granulosa cells which are also retrieved at egg recovery.
Could the metabolomic analysis of these superfluous
materials provide additional predictive information on
egg and embryo quality? Secondly, it remains to be
determined if metabolomic analysis could indicate
epigenetic defects, which may not affect embryo
development and pregnancy establishment, but which
could influence fetal development and the health and
wellbeing of offspring.
Acknowledgements
Research in the authors' laboratory is supported by a
cooperative agreement from the National Institutes of
Health, NICHD (U01-HD044638) and the Biotechnology
and Biological Sciences Research Council (BBS/B/
06164). The authors thank Dr. D.A. Barrett and Dr. J.T.
Posillico for useful discussions.
References
[1] Alikani M, Sadowy S, Cohen J. Human embryo morphology and
developmental capacity. In: Van Soom A, Boerjan M, editors.
Assessment of Mammalian Embryo Quality, Invasive and Noninvasive
techniques. Kluwer Academic Publishers; 2002. p. 1­31.
[2] Sakkas D, Gardner DK. Noninvasive methods to assess embryo
quality. Curr Opin Obstet Gynecol 2005;17:283­8.
[3] Merton S. Morphological evaluation of embryos in domestic
species. In: Van Soom A, Boerjan M, editors. Assessment of
Mammalian Embryo Quality, Invasive and Non-invasive Techniques.
Kluwer academic Publishers; 2002. p. 33­55.
R. Singh, K.D. Sinclair / Theriogenology 68S (2007) S56­S62 S61
[4] Sinclair KD, Lea RG, Rees WD, Young LE. The developmental
origins of health and disease: current theories and epigenetic
mechanisms. Reproduction 2007;64(Suppl):425­43.
[5] Houghton FD, Hawkhead JA, Humpherson PG, Hogg JE, Balen
AH, Rutherford AJ, et al. Non-invasive amino acid turnover
predicts human embryo developmental capacity. Hum Reprod
2002;17:999­1005.
[6] Houghton FD, Leese HJ. Metabolism and developmental competence
of the preimplantation embryo. Eur J Obstet Gynecol
Reprod Biol 2004;115(Suppl 1):S92­6.
[7] Lane M, Gardner DK. Selection of viable mouse blastocysts
prior to transfer using a metabolic criterion. Hum Reprod
1996;11:1975­8.
[8] Goodacre R, Vaidyanathan S, Dunn WB, Harrigan GG, Kell DB.
Metabolomics by numbers: acquiring and understanding global
metabolite data. Trends Biotechnol 2004;22:245­52.
[9] Gygi SP, Rochon Y, Franza BR, Aebersold R. Correlation
between protein and mRNA abundance in yeast. Mol Cell Biol
1999;19:1720­30.
[10] Sumner LW, Mendes P, Dixon RA. Plant metabolomics: largescale
phytochemistry in the functional genomics era. Phytochemistry
2003;62:817­36.
[11] Barabasi AL, Oltvai ZN. Network biology: understanding the
cell's functional organization. Nat Rev Genet 2004;5:101­13.
[12] Whitfield PD, German AJ, Noble PJ. Metabolomics: an emerging
post-genomic tool for nutrition. Br J Nutr 2004;92:549­55.
[13] Villas-Boas SG, Mas S, Akesson M, Smedsgaard J, Nielsen J.
Mass spectrometry in metabolome analysis. Mass Spectrom Rev
2005;24:613­46.
[14] Nicholson JK, Wilson ID. Opinion: understanding `global'
systems biology: metabonomics and the continuum of metabolism.
Nat Rev Drug Discov 2003;2:668­76.
[15] Nicholson JK, Lindon JC, Holmes E. `Metabonomics': understanding
the metabolic responses of living systems to pathophysiological
stimuli via multivariate statistical analysis of
biological NMR spectroscopic data. Xenobiotica 1999;29:
1181­9.
[16] Lindon JC, Holmes E, Nicholson JK. Metabonomics techniques
and applications to pharmaceutical research & development.
Pharm Res 2006;23:1075­88.
[17] Griffin JL. Metabonomics: NMR spectroscopy and pattern
recognition analysis of body fluids and tissues for characterisation
of xenobiotic toxicity and disease diagnosis. Curr Opin
Chem Biol 2003;7:648­54.
[18] Wilson ID, Nicholson JK, Castro-Perez J, Granger JH, Johnson
KA, Smith BW, et al. High resolution ``ultra performance''
liquid chromatography coupled to oa-TOF mass spectrometry
as a tool for differential metabolic pathway profiling in functional
genomic studies. J Proteome Res 2005;4:591­8.
[19] Pham-Tuan H, Kaskavelis L, Daykin CA, Janssen HG. Method
development in high-performance liquid chromatography for
high-throughput profiling and metabonomic studies of biofluid
samples. J Chromatogr B Analyt Technol Biomed Life Sci
2003;789:283­301.
[20] Weckwerth W, Morgenthal K. Metabolomics: from pattern
recognition to biological interpretation. Drug Discov Today
2005;10:1551­8.
[21] Aharoni A, Ric de Vos CH, Verhoeven HA, Maliepaard CA,
Kruppa G, Bino R, et al. Nontargeted metabolome analysis by
use of Fourier transform ion cyclotron mass spectrometry. Omics
2002;6:217­34.
[22] Dunn WB, Ellis DI. Metabolomics: current analytical platforms
and methodologies. Trends Anal Chem 2005;24:285­94.
[23] Thomas N, Goodacre R, Timmins EM, Gaudoin M, Fleming R.
Fourier transform infrared spectroscopy of follicular fluids from
large and small antral follicles. Hum Reprod 2000;15:1667­71.
[24] Thompson JG. The impact of nutrition of the cumulus oocyte
complex and embryo on subsequent development in ruminants. J
Reprod Dev 2006;52:169­75.
[25] Zeron Y, Ocheretny A, Kedar O, Borochov A, Sklan D, Arav A.
Seasonal changes in bovine fertility: relation to developmental
competence of oocytes, membrane properties and fatty acid
composition of follicles. Reproduction 2001;121:447­54.
[26] Booth PJ, Humpherson PG, Watson TJ, Leese HJ. Amino acid
depletion and appearance during porcine preimplantation
embryo development in vitro. Reproduction 2005;130:655­68.
[27] Sinclair KD, Rooke JA, McEvoy TG. Regulation of nutrient
uptake and metabolism in pre-elongation ruminant embryos.
Reprod 2003;61(Suppl):371­85.
[28] Brison DR, Houghton FD, Falconer D, Roberts SA, Hawkhead J,
Humpherson PG, et al. Identification of viable embryos in IVF
by non-invasive measurement of amino acid turnover. Hum
Reprod 2004;19:2319­24.
[29] Partridge RJ, Leese HJ. Consumption of amino acids by bovine
preimplantation embryos. Reprod Fertil Dev 1996;8:945­50.
[30] Haggarty P, Wood M, Ferguson E, Hoad G, Srikantharajah A,
Milne E, et al. Fatty acid metabolism in human preimplantation
embryos. Hum Reprod 2006;21:766­73.
[31] Merton JS, de Roos AP, Mullaart E, de Ruigh L, Kaal L, Vos PL,
et al. Factors affecting oocyte quality and quantity in commercial
application of embryo technologies in the cattle breeding industry.
Theriogenology 2003;59:651­74.
[32] Seli-E, Sakkas D, Behr B, Nagy P, Kwok JS, Burns DH. Noninvasive
metabolomic profiling of human embryo culture media
correlates with pregnancy outcome. Initial results of the metabolomics
study group for ART. In: Proceedings of the 62nd
annual meeting of American Society for Reproductive Medicine;
2006.
[33] Agrawal A, Behr B, Burns DH, Nagy P, Sakkas D, Scott R.
Metabolomic biomarkers of oxidative stress in ART: A rapid,
non-invasive methodology to assess the selective viable gametes
and embryos from their non-viable cohorts. In: Proceedings of
the Second international conference on cryopreservation of
human oocyte; 2006.
R. Singh, K.D. Sinclair / Theriogenology 68S (2007) S56­S62S62