Integrating genomics and metabolomics for engineering
plant metabolic pathways
Kirsi-Marja Oksman-Caldentey1
and Kazuki Saito2,3,4
Plant metabolites are characterized by an enormous chemical
diversity, every plant having its own complex set of
metabolites. This variety poses analytical challenges, both
for profiling multiple metabolites in parallel and for the
quantitative analysis of selected metabolites. We are only
just starting to understand the roles of these metabolites,
many of them being involved in adaptations to specific
ecological niches and some finding beneficial use (e.g.
as pharmaceuticals). Spectacular advances in plant
metabolomics offer new possibilities, together with the aid of
systems biology, to explore the extraordinary complexity of
the plant biochemical capacity. State-of-the art genomics
tools can be combined with metabolic profiling to identify
key genes that could be engineered for the production of
improved crop plants.
Addresses
1
VTT Biotechnology, PO Box 1500, 02044 VTT (Espoo), Finland
2
Graduate School of Pharmaceutical Sciences, Chiba University,
Inage-ku, Chiba 263-8522, Japan
3
Kazusa DNA Research Institute, Kisarazu, Chiba 292-0818, Japan
4
RIKEN Plant Science Center, Tsurumi-ku, Yokohama 230-0045, Japan
Corresponding author: Oksman-Caldentey,
Kirsi-Marja (kirsi-marja.oksman@vtt.fi)
Current Opinion in Biotechnology 2005, 16:174­179
This review comes from a themed issue on
Plant biotechnology
Edited by Dirk Inze´
Available online 3rd March 2005
0958-1669/$ ­ see front matter
# 2005 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.copbio.2005.02.007
Introduction
Plants are the most excellent designers and producers of a
variety of small compounds that are beneficial to mankind
as foods, medicines and industrial raw materials [1
].
Plants produce materials independently of fossil energies
and resources and are thus regarded as `ultimate factories'.
However, to rely solely on naturally growing
plants as a production system (e.g. for pharmaceuticals)
can be extremely dangerous, as shown recently by severe
shortage problems of artemisinin [2]. Therefore, an alternative
biotechnological production system based on
genetically engineered plant cells is an attractive
approach. Engineering plant metabolic pathways is, however,
a difficult endeavor, because our fundamental
knowledge of metabolic control in plants is not yet
sufficient for rational engineering of complicated metabolic
networks. Of course the relatively simple approaches,
such as the introduction of single genes for
rate-limiting enzymes or new branch pathway enzymes
(often from heterologous organisms), have led to spectacular
successes [3,4]. However, there are also `failures'
that have not been published, and thus it is difficult to
estimate the success rate of all trials of metabolic engineering.
In most cases, difficulties arise from the fundamental
lack of precise understanding of the entire
network between genes, transcripts, proteins and metabolites
in biological systems [5].
In contrast to traditional analysis that deals with a limited
number of genes, proteins and metabolites, a genomewide
large-scale approach is now possible for the holistic
systems analysis of some model plants such as Arabidopsis
thaliana. Non-targeted analysis of cell components, integrated
with genomics, analytical chemistry and bioinformatics,
allows a better understanding of plant systems and
will increase the practical potential of metabolic engineering.
In particular, recent technological advances in
metabolomics could open new avenues for precise metabolic
engineering in plants. This article describes the
cutting edge of the integrated analysis of genomics and
metabolomics for metabolic engineering in plants. We
also discuss some limitations of the technology and future
challenges that need to be faced.
The development of plant metabolomics
Metabolomics, which deals with all cellular metabolites,
has been recently recognized as an important sector of
post-genome science. The general idea of `metabolomics'
or the `metabolome' was first defined several years ago in
the field of microbiology [6], and its importance in plant
science was pointed out soon after [7]. Today, metabolomics
is also a powerful tool in drug discovery and
development; for instance, in the identification of drug
metabolites or biomarkers for organ-specific toxicities [8].
Even in the absence of any visible change in a cell or
individual plant, metabolomics, which allows phenotyping
by exhaustive metabolic profiling, can show how cells
respond as a system. Plant metabolomics is of particular
importance because of the huge chemical diversity of
plants compared with microorganisms and animals. The
number of metabolites from the plant kingdom has been
estimated at 200 000 [9] or even more [10]. Even a
single plant species such as A. thaliana might produce
5000 metabolites [11
,12
,13]. These numbers are significantly
greater than those of microorganisms (1500)
Current Opinion in Biotechnology 2005, 16:174­179 www.sciencedirect.com
and animals (2500). Although the term `metabolite (or
metabolic) profiling' has been understood in the past to
mean comprehensive profiling of metabolic change,
metabolomics has an intrinsically distinct connotation
because it implies integration with the other `omic'
sciences (Figure 1).
The comprehensive chemical analysis of metabolites and
the computation of huge datasets are the key components
of metabolomics [11
,14
,15,16]. Metabolomics is principally
required to determine all metabolites in a plant
extract; however, no single technology for metabolomics,
such as a DNA sequencer for genomics or DNA arrays for
transcriptomics, is available, and such a method may
never be possible. This is because the analysis of metabolites
of divergent physicochemical properties needs a
wide range of chemistries: a single chemistry cannot deal
with metabolomics as it can with nucleic acids and
proteins. Metaphorically speaking, metabolomics is like
attempting whole-genome sequencing without either the
Sanger method or the Maxam­Gilbert method. At present,
combinations of different analytical methods of high
sensitivity are generally used for comprehensive nontargeted
chemical analysis [17,18].
One of the key challenges of metabolite profiling and
analysis is to find an optimal balance between accuracy
and coverage of metabolite measurements. This has
become particularly apparent when analysing plant secondary
metabolites, which have very different chemical
natures compared with primary metabolites. If one general
extraction and analytical system is used it is likely
that many metabolites will remain in the plant matrix and
will not be profiled. One possibility is to split the metabolomics
platform into subgroups of compounds sharing
similar chemical extraction conditions, chromatographic
separation and subsequent instrumental analysis [5].
Advances in instrument development have been tremendous.
Traditionally, mass spectrometry (MS) combined
with a number of chromatographic methods is used for
metabolome analysis. These methods include gas chromatograpy
mass spectrometry (GC-MS) [19
,20
], highperformance
liquid chromatography mass spectrometry
(HPLC-MS) [12
,21,22] and, more recently, capillary
electrophoresis mass spectrometry (CE-MS) [23]. Analyti-
calmethodswithoutpre-separationbychromatography,for
example, Fourier-transform ion cyclotron resonance (FTICR)
mass spectrometry and time-of-flight (TOF) mass
spectrometry, are also used, although largely for fingerprinting
purposes [24,25]. Nuclear magnetic resonance
(NMR) spectroscopy analysis of the whole-cell extract
and on-line NMR analysis coupled with a liquid chromatography
stop flow method are also available [26­29].
Data analysis forms the second major component of
metabolomics. Chemometrics and multivariate analysis,
such as a principal component analysis, hierarchical cluster
analysis, and self-organization mapping, are often used
for data mining [14
,30­32,33
,34]. Integration of metabolome
data with other omics data, such as transcriptomic
data, can be performed in silico using a software tool to
depict gene-protein-metabolite networks [35]. Cell simulation
can also be carried out to predict the metabolic
phenotype of a particular gene knockout [36].
Metabolomics-based functional genomics
in model plants
An advantage of working with A. thaliana, a model plant
for modern plant science, is that genome-related
Integrating plant genomics and metabolomics Oksman-Caldentey and Saito 175
Figure 1
Molecular breeding of
stress-resistant plants
Development of pharmaceuticals
and functional foods
Production of `green' industrial
materials and bioenergy
ˇ
ˇ
ˇ
Multiple arrays of
a given plant species
* single compounds
* mixture of compounds
Biological activity arrays
Metabolome arrays
Proteome arrays
Transcriptome arrays
Genome arrays
Current Opinion in Biotechnology
Data mining and
bioinformatics
The `phytochemical array' -- a concept for using various functional genomics-based arrays in plants. Multiple arrays to analyse the genome,
transcriptome, proteome, metabolome and bioactivity of metabolites of a given plant species are the modern tools for understanding plants at
various biological levels. These tools can lead to applications for crop improvement (e.g. molecular breeding of biotic and abiotic stress-resistant
plants), to the discovery and development of plant-based pharmaceuticals, to the development of functional foods, and to the production of
plant-derived industrial materials and energy. All of these advances will be valuable in the future and can be developed through genome-based
plant biotechnology. The phytochemical array exhibits all of the links from the genome to the activity of metabolites for desired traits, and will
lead to accelerated development.
www.sciencedirect.com Current Opinion in Biotechnology 2005, 16:174­179
resources including the whole-genome sequence with
functional annotations, DNA microarrays, DNA-tagged
insertion mutants and metabolic maps (AraCyc; http://
www.arabidopsis.org/tools/aracyc) are readily accessible
[37].
GC-MS was used to compare the metabolic profiles of
four Arabidopsis genotypes (two ecotypes, Col-2 and C24,
and a mutant of each ecotype) and indicated that the
metabolic phenotypes of the two ecotypes diverged more
from each other than did those of each mutant from its
parent ecoptype, suggesting that the cell metabolome is
influenced more predominantly by the difference in
ecotype than by a single gene mutation [19
].
Pair-wise metabolite-to-metabolite [38
] and transcriptto-metabolite
[39
] correlation analysis of transgenic
potato plants expressing particular genes revealed novel
correlations that had not been suggested by classical
targeted approaches. Most of these transgenic plants
exhibited no visible phenotypes. Arabidopsis pal1 and
pal2 mutants lacking the function of two phenylalanine
ammonia lyase genes also showed no clear phenotypic
alterations. Phenotyping of the single mutants and the
double-mutant by a combination of transcript profiling
and detailed targeted metabolite profiling for sugars,
amino acids and phenylpropanoids suggested the specific
function of PAL1 and PAL2 [40
].
Integrated analyses of the transcriptome and a detailed
chemical analysis employing LC-MS and FT-ICR MS
were carried out for Arabidopsis lines overexpressing the
PAP1 gene, which codes for a Myb-like transcription
factor [41
]. The changes in the metabolic profiles
caused by PAP1 gene expression were specific to the
increased accumulation of anthocyanins. The expression
of genes known to be involved in anthocyanin production
was upregulated, and thus other upregulated genes could
be tentatively assigned a role in the formation of anthocyanins,
such as members of the glycosyltransferase,
acyltransferase and glutathione S-transferase families.
The function of some of these predicted genes was
confirmed experimentally through classical analysis of
T-DNA-inserted knockout mutant lines and in vitro
enzymatic studies with recombinant proteins [42]. These
approaches indicate the feasibility of integrating metabolomic
and transcriptomic analyses in functional genomics
studies of Arabidopsis [41
].
Nutritional and abiotic stresses modulate re-programming
of the transcriptome and metabolome. Thus, an
integrated analysis leads to the identification of gene
functions that are modulated by these stresses. A good
example of this was obtained from a study of sulfur
starvation in Arabidopsis [33
]. Glucosinolate-related
metabolites and genes involved in their metabolism were
coordinately regulated by nutritional sulfur stress.
Further detailed investigation resulted in the prediction
and identification of particular gene functions involved in
glucosinolate metabolism (MY Hirai and K Saito, unpublished).
Through the investigation of stress responses to
nitrogen [43] and low temperature [44], wide investigation
of gene expression and metabolite levels, respectively,
have revealed novel pathways responding to each
stress.
Attempts to functionally identify genes involved in saponin
biosynthesis by Medicago truncatula, a model Leguminosae
plant, have also been made using a combination
of genomics and metabolic profiling [45,46]. These studies
demonstrated the differential transcriptional re-programming
of multiple genes in the phenylpropanoid and
triterpene pathways in response to methyl jasmonate and
yeast extracts (see Update).
Gene and metabolite expression profiles
in non-model plants
For decades, most beneficial compounds identified and
isolated from plants were discovered through classical,
complicated extraction systems combined with targeted
chemical analysis. Secondary metabolites have been
identified as a particularly rich source of useful compounds
and have a crucial role in this regard. However,
only a fragment of all higher plants has been investigated
[47]. There is no doubt that the chemical diversity of
plants is much greater than any chemical library made by
man, and thus the plant kingdom represents an enormous
reservoir of novel molecules waiting to be discovered.
Efficient metabolome analysis is likely to be a key element
for future success in this field.
Elicitation is a process used to mimic the natural reactions
of plants towards various environmental stresses and has
been successfully applied to plants and plant cell cultures
to induce secondary metabolite production [48]. It has
been shown that upon elicitation huge numbers of genes
are activated, many of them involved in metabolite biosynthesis.
Goossens and coworkers [49
] treated tobacco
BY-2 cells with methyljasmonate and discovered by using
cDNA-AFLP (amplified fragment length polymorphism)
transcript profiling that close to 600 genes were differentially
regulated by this elicitor. By linking this data to
metabolite analysis in a time-course experiment, it was
possible to build an ample inventory of genes that were
already known or novel genes. Potentially, these genes
could be involved -- in either a structural or regulatory
capacity -- with tobacco secondary metabolism, and
possibly with plant secondary metabolism in general.
The great advantage of this technology is that it allows
gene identification without prior sequence knowledge.
This is crucial when working with non-model plants or
rare (e.g. medicinal) plant species for which the genome is
unknown. Furthermore, this approach provides quantitative
information on gene expression. Applying transcript
176 Plant biotechnology
Current Opinion in Biotechnology 2005, 16:174­179 www.sciencedirect.com
profiling in Catharanthus roseus, for example, the number
of jasmonate-modulated genes was found to be of similar
magnitude to that in tobacco cells. However, according to
our preliminary metabolome experiments, more than
2000 metabolites were seen to respond to methyljasmonate
treatment in C. roseus (H Rischer, M Oresic and KM
Oksman-Caldentey, unpublished). Another huge challenge
is to identify the function of the candidate genes
obtained from this kind of transcript profiling [49
] and to
find out their possible role in secondary metabolism.
Again, high-throughput metabolite analysis will play a
crucial role.
Some genes involved in the formation of volatile compounds
in strawberry [50], rose [51
] and spider mint [52
]
have been identified using DNA microarrays in combination
with targeted analysis of volatile metabolites. The
chemical analysis of fragrance-related metabolites by GCMS
is possible and leads to a sensitive chemical analysis
which, together with gene expression profiles, is sufficient
for gene identification. Activation-tagged lines, in which a
gene is over-expressed by random insertion of an enhancer
sequence in the genome, are good resources for gene
hunting. By screening activation-tagged lines in tomato, a
Myb transcription factor gene and co-regulated structural
genes involved in anthocyanin formation were successfully
identified [53].
The classical biochemical approach has resulted in considerable
knowledge of the genes involved in the synthesis
of flavonoids [54] and terpenoid indole alkaloids [55].
However, genetic maps of biosynthetic pathways, in
general, are still far from complete and the regulation
of these pathways is not fully understood. This point is
well illustrated by a recent investigation of morphine
biosynthesis in the opium poppy. Silencing codeinone
reductase using a chimeric hairpin RNA construct led,
through a feed-back mechanism, to the accumulation of
(S)-reticuline, an intermediate of morphine biosynthesis;
however, between (S)-reticuline and codeinone there are
eight enzymatic steps and, surprisingly, all were inhibited
by this RNA construct [56].
Conclusions
Tremendous advances in metabolomics and its integration
into other omics (genomics, transcriptomics and
proteomics) have brought us closer to understanding
the links between different levels of biological systems,
leading to the realization of systems biology [57]. This has
been made possible by combining expertise from the
areas of biology, chemistry, instrumentation and bioinformatics.
In particular, in silico metabolic experiments could
facilitate the network analysis of complicated metabolic
pathways [58
,59]. Considering the huge chemical diversity
of plants compared with those of animals and microorganisms,
a major future challenge will be to explore the
molecular genetic origins of chemical diversity in nonmodel
exotic plants. A further prospect beyond conventional
omics would be the comprehensive analysis of the
function and activity of an array of plant metabolites,
leading to the `phytochemical array' concept (Figure 1).
The phytochemical array consists of genomics, transcriptomics,
proteomics, metabolomics and activity arrays of a
given plant species. Such an array would allow visualization
of all the connections between genes, transcripts,
proteins, metabolites and their activities. Drawing links
from the genome to the activity of metabolites will be
necessary for the high-throughput discovery of plantbased
pharmaceuticals and for the development of functional
foods and stress-resistant plants. Although there are
still several limitations in metabolomics, such as the
urgent need for precise spacio-resolution (e.g. single cell
and subcellular analysis) and temporal-resolution and the
issue of metabolic channeling by protein complexes, we
believe that many exciting developments are to be
expected in the coming years.
Update
Recent work has demonstrated the differential transcriptional
re-programming of multiple genes in the phenylpropanoid
and triterpene pathways in response to methyl
jasmonate and yeast extracts in M. truncatula [60].
Acknowledgements
Tekes, The National Technology Agency, Finland is kindly
acknowledged for financial support from the NeoBio program to KMOC.
KS thanks the Grants-in-Aid for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology, Japan and
CREST of Japan Science and Technology Agency (JST) for financial
support. Finally, the authors express their warm thanks to John
Londesborough (VTT Biotechnology) for kindly revising the manuscript.
References and recommended reading
Papers of particular interest, published within the annual period of
review, have been highlighted as:
 of special interest
 of outstanding interest
1.

Oksman-Caldentey KM, Inze´ D: Plant cell factories in the postgenomic
era: new ways to produce designer secondary
metabolites. Trends Plant Sci 2004, 9:433-440.
A concise and well-written up-to-date review on strategies and future
trends for engineering plant secondary products, including a few attractive
examples. New possibilities to use plant-cell cultures as a production
host are presented.
2. Cyranoski D: Campaign to fight malaria hit by surge in demand
for medicine. Nature 2004, 432:259.
3. Ye X, Al-Babili S, Kloti A, Zhang J, Lucca P, Beyer P, Potrykus I:
Engineering the provitamin A (b-carotene) biosynthetic
pathway into (carotenoid-free) rice endosperm. Science 2000,
287:303-305.
4. Niggeweg R, Michael AJ, Martin C: Engineering plants with
increased levels of the antioxidant chlorogenic acid. Nat
Biotechnol 2004, 22:746-754.
5. Oksman-Caldentey K-M, Inze´ D, Oresˇicˇ M: Connecting genes to
metabolites by a systems biology approach. Proc Natl Acad Sci
USA 2004, 101:9949-9950.
6. Tweeddale H, Notley-McRobb L, Ferenci T: Effect of slow growth
on metabolism of Escherichia coli, as revealed by global
metabolite pool (`Metabolome') analysis. J Bacteriol 1998,
180:5109-5116.
Integrating plant genomics and metabolomics Oksman-Caldentey and Saito 177
www.sciencedirect.com Current Opinion in Biotechnology 2005, 16:174­179
7. Trethewey RN, Krotzky AJ, Willmitzer L: Metabolic profiling: a
Rosetta stone for genomics? Curr Opin Plant Biol 1999, 2:83-85.
8. Harrigan GG, Goodacre R (eds): Metabolic Profiling: its Role in
Biomarker Discovery and Gene Function Analysis. Kluwer
Academic Publishers: Boston; 2003.
9. Dixon RA, Strack D: Phytochemistry meets genome analysis,
and beyond. . .. Phytochemistry 2003, 62:815-816.
10. Trethewey R: Metabolite profiling as an aid to metabolic
engineering in plants. Curr Opin Plant Biol 2004, 7:196-201.
11.

Bino RJ, Hall RD, Fiehn O, Kopka J, Saito K, Draper J, Nikolau BJ,
Mendes P, Roessner-Tunali U, Beale MH et al.: Potential of
metabolomics as a functional genomics tool. Trends Plant Sci
2004, 9:418-425.
This article proposes a framework for the study of plant metabolomics
presented by the international plant metabolomics platform. Several
community-based efforts are proposed for the development of plant
metabolomics.
12.

Von Roepenack-Lahaye E, Degenkolb T, Zerjeski M, Franz M,
Roth U, Wessjohann L, Schmidt J, Scheel D, Clemens S: Profiling
of Arabidopsis secondary metabolites by capillary liquid
chromatography coupled to electrospray ionization
quadrupole time-of-flight mass spectrometry. Plant Physiol
2004, 134:548-559.
This report describes the chemical analysis of Arabidopsis metabolites
using HPLC coupled with a quadrupole time-of-flight mass spectrometer.
Using this method, the authors detected 2000 metabolites in leaves and
roots of A. thaliana.
13. Facchini PJ, Bird DA, St-Pierre B: Can Arabidopsis make
complex alkaloids? Trends Plant Sci 2004, 9:116-122.
14.

Fiehn O: Metabolomics ­ the link between genotypes and
phenotypes. Plant Mol Biol 2002, 48:155-171.
The first comprehensive review covering all aspects of plant metabolomics,
definition, analytical methodology and data analysis.
15. Sumner LW, Mendes P, Dixon RA: Plant metabolomics:
large-scale phytochemistry in the functional genomics era.
Phytochemistry 2003, 62:817-836.
16. Weckwerth W: Metabolomics in systems biology. Annu Rev
Plant Biol 2003, 54:669-689.
17. Kopka J, Fernie A, Weckwerth W, Gibon Y, Stitt M: Metabolite
profiling in plant biology: platforms and destinations.
Genome Biol 2004, 5:109.
18. Stitt M, Fernie AR: From measurements of metabolites to
metabolomics: an `on the fly' perspective illustrated by recent
studies of carbon­nitrogen interactions. Curr Opin Biotechnol
2003, 14:136-144.
19.

Fiehn O, Kopka J, Doermann P, Altman T, Trethewey RN,
Willmitzer L: Metabolite profiling for plant functional genomics.
Nat Biotechnol 2000, 18:1157-1161.
Together with Rossener et al. [20
], this is one of the first reports to show
how comprehensive metabolic profiling using high-throughput GC-MS
technology could help us to understand the systematic response of plants
to genetic and environmental alterations.
20.

Rossener U, Luedemann A, Brust D, Fiehn O, Linke T, Willmitzer L,
Fernie AR: Metabolic profiling allows comprehensive
phenotyping of genetically or environmentally modified plant
systems. Plant Cell 2001, 13:11-29.
See the annotation for [19
].
21. Yamazaki M, Nakajima J, Yamanashi M, Sugiyama M, Makita Y,
Springob K, Awazuhara M, Saito K: Metabolomics and
differential gene expression in anthocyanin chemo-varietal
froms of Perilla frutescens. Phytochemistry 2003, 62:987-995.
22. Yamazaki Y, Urano A, Sudo H, Kitajima M, Takayama H,
Yamazaki M, Aimi N, Saito K: Metabolite profiling of alkaloids
and strictosidine synthase activity in camptothecin producing
plants. Phytochemistry 2003, 62:461-470.
23. Sato S, Soga T, Nishioka T, Tomita M: Simultaneous
determination of the main metabolites in rice leaves using
capillary electrophoresis mass spectrometry and capillary
electrophoresis diode array detection. Plant J 2004,
40:151-163.
24. Aharoni A, de Vos CH, Verhoeven HA, Mailiepaard CA, Kruppa G,
Bino RJ, Goodenowe DB: Nontargeted metabolome analysis by
use of Fourier transform ion cyclotron mass spectrometry.
OMICS 2002, 6:217-234.
25. Brown SC, Kruppa G, Dasseux JL: Metabolomics applications
of FT-ICR mass spectrometry. Mass Spectrom Rev 2005,
24:223-231.
26. 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, 7:648-654.
27. Wolfender JL, Ndjoko K, Hostettmann K: Liquid chromatography
with ultraviolet absorbance-mass spectrometric detection
and with nuclear magnetic resonance spectroscopy: a
powerful combination for the on-line structural investigation
of plant metabolites. J Chromatogr A 2003, 1000:437-455.
28. Ward JL, Harris C, Lewis J, Beale MH: Assessment of 1
H NMR
spectroscopy and multivariate analysis as a technique for
metabolite fingerprinting of Arabidopsis thaliana.
Phytochemistry 2003, 62:949-957.
29. Kikuchi J, Shinozaki K, Hirayama T: Stable isotope labeling of
Arabidopsis thaliana for an NMR-based metabolomics
approach. Plant Cell Physiol 2004, 45:1099-1104.
30. Kose F, Weckwerth W, Linke T, Fiehn O: Visualizing plant
metabolomics correlation networks using clique-metabolite
matrices. Bioinformatics 2001, 17:1198-1208.
31. Duran AL, Yang J, Wang L, Sumner LW: Metabolomics spectral
formatting, alignment and conversion tools (MSFACTs).
Bioinformatics 2003, 19:2283-2293.
32. Jonsson P, Gullberg J, Nordstro¨ m A, Kusano M, Kowalczyk M,
Sjo¨ stro¨ m M, Moritz T: A strategy for identifying differences in
large series of metabolomics samples analyzed GC/MS. Anal
Chem 2004, 76:1738-1745.
33.

Hirai MY, Yano M, Goodenowe D, Kanaya S, Kimura T,
Awazuhara M, Arita M, Fujiwara T, Saito K: Integration of
transcriptomics and metabolomics for understanding of
global responses to nutritional stresses in Arabidopsis
thaliana. Proc Natl Acad Sci USA 2004, 101:10205-10210.
In this excellent article, the authors combined DNA array analysis with
targeted and non-targeted chemical analysis for holistic investigation
under conditions of nutritional stress in Arabidopsis. They found new
networks of pathways, genes and metabolites in primary and secondary
metabolism, enabled by the use of batch-learning self-organizing map
analysis.
34. Jenkins H, Hardy N, Beckmann M, Draper J, Smith AR, Taylor J,
Fiehn O, Goodacre R, Bino R, Hall R et al.: A proposed framework
for the description of plant metabolomics experiments and
their results. Nat Biotechnol 2004, 2:1601-1606.
35. Thimm O, Bla¨ sing O, Gibon Y, Nagel A, Meyer S, Kru¨ ger P,
Selbig J, Mu¨ ller LA, Rhee SY, Stitt M: MAPMAN: a user-driven
tool to display genomics data sets onto diagrams of metabolic
pathways and other biological processes. Plant J 2004,
37:914-939.
36. Tomita M, Hashimoto K, Takahashi K, Shimizu TS, Matsuzaki Y,
Miyoshi F, Saito K, Tanida S, Yugi K, Venter JC: E-CELL: Software
environment for whole-cell simulation. Bioinformatics 1999,
15:72-84.
37. Mueller LA, Zhang P, Rhee SY: AraCyc. A biochemical pathway
database for Arabidopsis. Plant Physiol 2003, 132:453-460.
38.

Weckwerth W, Loureiro ME, Wenzel K, Fiehn O: Differential
metabolic networks unravel the effects of silent plant
phenotypes. Proc Natl Acad Sci USA 2004, 101:7809-7814.
Non-targeted metabolite analysis by GC-TOF MS was applied to the
phenotyping of silent (i.e. no visible changes) transgenic potato plants
suppressing a sucrose synthase gene. Metabolic networks were constructed
to reveal alterations even if no apparent difference was seen in
metabolite levels.
39.

Urbanczyk-Wochniak E, Luedemann A, Kopka J, Selbig J,
Roessner-Tunali U, Willmitzer L, Fernie AR: Parallel analysis of
transcript and metabolic profiles: a new approach in systems
biology. EMBO Rep 2003, 4:989-993.
178 Plant biotechnology
Current Opinion in Biotechnology 2005, 16:174­179 www.sciencedirect.com
Correlations of gene expression and metabolite accumulation were made
by the pair-wise analysis of transcript and metabolite levels. From 27 000
possible pairs, 571 showed significant correlation.
40.

Rohde A, Morreel K, Ralph J, Goeminne G, Hostyn V, De Rycke R,
Kushnir S, Van Doorsselaere J, Joseleau JP, Vuylsteke M et al.:
Molecular phenotyping of the pal1 and pal2 mutants of
Arabidopsis thaliana reveals far-reaching consequences on
phenylpropanoid, amino acid, and carbohydrate metabolism.
Plant Cell 2004, 16:2749-2771.
Extensive analysis of the metabolites and transcripts from pal1 and pal2
single- and double-mutants of Arabidopsis enabled the distinct functions
of the PAL1 and PAL2 genes to be elucidated.
41.

Tohge T, Nishiyama Y, Hirai MY, Yano M, Nakajima J,
Awazuhara M, Inoue E, Takahashi H, Goodenowe DB, Kitayama M
et al: Functional genomics by integrated analysis of
metabolome and transcriptome of Arabidopsis plants
overexpressing a MYB transcription factor. Plant J 2005,
in press.
Detailed analysis of flavonoids together with non-targeted chemical
analysis were used to characterize the metabolome of several Arabidopsis
lines engineered to overexpress the transcription factor PAP1. Some
of the genes upregulated in the PAP1 overexpression lines were found to
encode proteins involved in the biosynthesis of anthocyanins. This is a
good example of how the combined efforts of transcriptome and metabolome
analyses can lead to the identification of gene function.
42. Jones P, Messner B, Nakajima J, Scha¨ ffner AR, Saito K: UGT73C6
and UGT78D1, glycosyltransferases involved in flavonol
glycoside biosynthesis in Arabidopsis thaliana. J Biol Chem
2003, 278:43910-43918.
43. Scheible WR, Morcuende R, Czechowski T, Fritz C, Osuna D,
Palacios-Rojas N, Schindelasch D, Thimm O, Udvardi MK, Stitt M:
Genome-wide reprogramming of primary and secondary
metabolism, protein synthesis, cellular growth processes, and
the regulatory infrastructure of Arabidopsis in response to
nitrogen. Plant Physiol 2004, 136:2483-2499.
44. Cook D, Fowler S, Fiehn O, Thomashow MF: A prominent role for
the CBF cold response pathway in configuring the lowtemperature
metabolome of Arabidopsis. Proc Natl Acad Sci
USA 2004, 101:15243-15248.
45. Suzuki H, Achnine L, Xu R, Matsuda SPT, Dixon RA: A genomics
approach to the early stages of triterpene saponin
biosynthesis in Medicago truncatula. Plant J 2002,
32:1033-1048.
46. Huhman DV, Sumner LW: Metabolic profiling of saponins in
Medicago sativa and Medicago truncatula using HPLC
coupled to an electrospray ion-trap mass spectrometer.
Phytochemistry 2002, 59:347-360.
47. Hostettmann K, Terreaux C: Search for new lead compounds
from higher plants. Chimia (Aarau) 2000, 54:652-657.
48. Dixon RA: Natural products and plant disease resistance.
Nature 2001, 411:843-847.
49.

Goossens A, Ha¨ kkinen ST, Laakso I, Seppa¨ nen-Laakso T,
Biondi S, De Sutter V, Lammertyn F, Nuutila AM, Soderlund H,
Zabeau M et al.: A functional genomics approach toward
the understanding of secondary metabolism in plant cells.
Proc Natl Acad Sci USA 2003, 100:8595-8600.
A cDNA-AFLP technology was applied to the high-throughput transcript
profiling of methyljasmonate-treated tobacco BY-2 cells. The correlation
of transcripts with the accumulation patterns of secondary metabolites
allowed the identification of putative genes responsible for the inducible
accumulation of secondary products. The method is generally applicable
for any non-model or exotic plants for which genome information is
unavailable. Furthermore, this approach provides quantitative data on
gene expression.
50. Aharoni A, Keizer LCP, Bouwmeester HJ, Sun Z, Alvarez-Huerta
M, Verhoven HA, Blaas J, van Houwelingen AMML, De Vos RCH,
van der Voet H et al.: Identification of the SAAT gene involved in
strawberry flavor biogenesis by use of DNA microarrays. Plant
Cell 2000, 12:647-661.
51.

Guterman I, Shalit M, Menda N, Piestun D, Dafy-Yelin M, Shalev G,
Bar E, Davydov O, Ovadis M, Emanuel M et al.: Rose scent:
genomics approach to discovering novel floral fragrancerelated
genes. Plant Cell 2002, 14:2325-2338.
DNA chips were prepared from the expressed sequence tag information
of rose and used for the expression profiling of transcripts in rose petal
from different cultivars. Combined with a detailed chemical analysis of
volatile compounds this work led to the discovery of novel flower scentrelated
genes.
52.

Mercke P, Kappers IF, Verstappen FWA, Vorst O, Dicke M,
Bouwmeester HJ: Combined transcript and metabolite analysis
reveals genes involved in spider mint induced volatile
formation in cucumber plants. Plant Physiol 2004,
135:2012-2024.
Using a similar approach to that described by Guterman et al. [51
], DNA
microarray analysis was combined with the chemical analysis of volatile
metabolites in cucumber. Insight was gained into the involvement of the
studied genes in the production of volatile compounds.
53. Mathews H, Clendennen SK, Caldwell CG, Liu XL, Connors K,
Matheis N, Schuster DK, Menasco DJ, Wagoner W, Lightner J,
Wagner DR: Activation tagging in tomato identifies a
transcriptional regulator of anthocyanin biosynthesis,
modification, and transport. Plant Cell 2003, 15:1689-1703.
54. Springob K, Nakajima J, Yamazaki M, Saito K: Recent advances
in the biosynthesis of and accumulation of anthocyanins.
Nat Prod Rep 2003, 20:288-303.
55. Verpoorte R, Memelink J: Engineering secondary metabolite
production in plants. Curr Opin Biotechnol 2002, 13:181-187.
56. Allen RS, Millgate AG, Chitty JA, Thisleton J, Miller JAC, Fist AJ,
Gerlach WL, Larkin PJ: RNAi-mediated replacement of
morphine with the nonnarcotic alkaloid reticuline in opium
poppy. Nat Biotechnol 2004, 22:1559-1566.
57. Fernie AR, Trethewey RN, Krotzky AJ, Willmitzer L: Metabolite
profiling: from diagnostics to systems biology. Nat Rev Mol Cell
Biol 2004, 5:763-769.
58.

Arita M: In silico atomic tracing by substrate-product
relationships in Escherichia coli intermediary metabolism.
Genome Res 2003, 13:2455-2466.
A report on a specific software that allows in silico tracer experiments to
be carried out at the atomic level for all possible pathways.
59. Yamazaki Y, Kitajima M, Arita M, Takayama H, Sudo H,
Yamazaki M, Aimi N, Saito K: Biosynthesis of camptothecin.
In silico and in vivo tracer study from [1-13
C]glucose.
Plant Physiol 2004, 134:161-170.
60. Suzuki H, Srinivasa Reddy MS, Naoumkina M, Aziz N, May GD,
Huhman DV, Sumner LW, Blount JW, Mendes P, Dixon RA: Methyl
jasmonate and yeast elicitor induce differential transcriptional
and metabolic re-programming in cell suspension cultures of
the model legume Medicago truncatula. Planta 2005. DOI:
10.1007/s00425-004-1387-2.
Integrating plant genomics and metabolomics Oksman-Caldentey and Saito 179
www.sciencedirect.com Current Opinion in Biotechnology 2005, 16:174­179