under the control of a constitutive promoter to facilitate the selec-
tive propagation of transformed cells, and a ‘primary transgene’ or
‘gene of interest’ which could be under the control of any promoter
Engineering metabolic pathways in plants
by multigene transformation
UXUE ZORRILLA-LÓPEZ#,1, GEMMA MASIP#,1, GEMMA ARJÓ2, CHAO BAI1, RAVIRAJ BANAKAR1,
LUDOVIC BASSIE1, JUDIT BERMAN1, GEMMA FARRÉ1, BRUNA MIRALPEIX1, EDUARD PÉREZ-MASSOT1,
MAITE SABALZA1, GEORGINA SANAHUJA1, EVANGELIA VAMVAKA1, RICHARD M. TWYMAN3,
PAUL CHRISTOU1,4, CHANGFU ZHU1 and TERESA CAPELL*,1
1Department of Plant Production and Forestry Science, School of Agrifood and Forestry Science and Engineering
(ETSEA), University of Lleida-Agrotecnio Center, Lleida, Spain, 2Department of Medicine, Institute of Biomedical
Research (IRB), University of Lleida, Lleida, Spain, 3School of Life Sciences, University of Warwick, Coventry, UK
and 4Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain
ABSTRACT Metabolic engineering in plants can be used to increase the abundance of specific
valuable metabolites, but single-point interventions generally do not improve the yields of target
metabolites unless that product is immediately downstream of the intervention point and there is
a plentiful supply of precursors. In many cases, an intervention is necessary at an early bottleneck,
sometimes the first committed step in the pathway, but is often only successful in shifting the
bottleneck downstream, sometimes also causing the accumulation of an undesirable metabolic
intermediate. Occasionally it has been possible to induce multiple genes in a pathway by controlling
the expression of a key regulator, such as a transcription factor, but this strategy is only possible
if such master regulators exist and can be identified. A more robust approach is the simultane-
ous expression of multiple genes in the pathway, preferably representing every critical enzymatic
step, therefore removing all bottlenecks and ensuring completely unrestricted metabolic flux. This
approach requires the transfer of multiple enzyme-encoding genes to the recipient plant, which
is achieved most efficiently if all genes are transferred at the same time. Here we review the state
of the art in multigene transformation as applied to metabolic engineering in plants, highlighting
some of the most significant recent advances in the field.
KEY WORDS: direct DNA transfer, multigene transformation, metabolic pathway, genetic engineering
Most agronomic traits in plants are controlled by multiple genes,
as is also the case for the synthesis of complex organic compounds
from primary and secondary metabolisms, which often represent
the outputs of long and convoluted metabolic pathways. Therefore,
genetic engineering has seen a progressive change from single-
gene intervention to multigene transformation to tackle increasingly
ambitious objectives (Halpin, 2005).
In the early years of plant biotechnology, gene transfer experi-
ments typically involved two transgenes: one selectable marker
Int. J. Dev. Biol. UNCORRECTED PROOF
*Address correspondence to: Teresa Capell. University of Lleida-Agrotecnio Center, Avenida Alcalde Rovira Roure 191, E-25198 Lleida, Spain. Tel: +34 973 702831.
Fax: +34 973 702690. e-mail: email@example.com
#Note: Both authors contributed equally to this work.
Final, author-corrected PDF published online: xx xxxx 2013.
ISSN: Online 1696-3547, Print 0214-6282
© 2013 UBC Press
Printed in Spain
Abbreviations used in this paper: BAC, bacterial artificial chromosome; CaMV35S,
cauliflower mosaic virus 35S; ORF, open reading frame; MGT, multigene trans-
formation; PHB, polyhydroxybutyrate; PUFA, polyunsaturated fatty acid.
and was intended to alter the phenotype of the plant in a specific
manner (Peremarti et al., 2010). This principle was adopted in
the first examples of metabolic engineering, which involves the
modulation of endogenous metabolic pathways to increase flux
towards particular desirable molecules or even new molecules
(Capell and Christou, 2004). Multigene transformation (MGT) is
being gradually accepted as an approach to generate plants with
more ambitious phenotypes, including more complex examples of
metabolic engineering (Naqvi et al., 2009). To this end, methods
had to be developed for the coordinated expression of larger groups
2 U. Zorrilla-López et al.
of genes (Capell and Christou, 2004) to attain objectives, such
as: (a) enhance the activity of enzymes at multiple rate-limiting
steps in target pathways, e.g. by expression of enzymes that are
released from feedback inhibition; (b) increase the availability of
upstream precursors to increase flux through the target pathway;
(c) modulate pathway branch points to prevent the loss of flux; and
(d) promote the development of sink compartments to store target
compounds (Fig. 1) (Zhu et al., 2013).
Examples of metabolic engineering in plants include primary
metabolic pathways (carbohydrates, amino acids, and lipids) and
secondary metabolic pathways (e.g. alkaloids, terpenoids, flavo-
noids, lignins, quinones, and other benzoic acid derivatives; Gomez
Galera et al., 2007). These pathways generate a large number of
compounds that are useful to humans, including energy-rich foods,
vitamins and many different pharmaceuticals. In this review, we
focus on the metabolic engineering of vitamins, polyunsaturated
fatty acids, and secondary metabolites, because they provide il-
lustrative examples of applied MGT.
The scope of the challenge
The simultaneous transfer of multiple genes into plants (co-
transformation) can be achieved using two main approaches, one
involving linked genes (multiple genes on the same plasmid) and
the other involving unlinked genes (different genes on different
plasmids). The two methods can be used with both major strategies
for gene transfer to plants, i.e. transformation with Agrobacterium
tumefaciens and direct DNA transfer (Naqvi et al., 2009).
Multiple linked genes can be transferred by Agrobacterium-
mediated transformation using standard binary vectors contain-
ing multiple genes within a single T-DNA or multiple T-DNAs
each containing a single gene, whereas for direct DNA transfer
methods the genes can be linked on conventional vectors (Naqvi
et al., 2009). The transgenes tend to integrate at a single locus,
although the precise arrangement of multiple T-DNAs depends on
the bacterial strain (Twyman et al., 2002). This strategy is robust
for a small number of input genes, but as the number increases,
the vectors become increasingly cumbersome and unstable; the
effective upper capacity using standard vectors is approximately
50 kb due to dwindling efficiency (Naqvi et al., 2009). High-capacity
binary vectors (BIBAC, BIBAC2, and TAC) that allow the transfer
of up to 200 kb of insert DNA are discussed below.
Multiple unlinked genes can be introduced by Agrobacterium-
mediated transformation if the bacteria contain multiple compatible
plasmids carrying separate T-DNAs or if the inoculum comprises
a mixture of bacterial strains carrying different vectors; however,
the ratio of different input genes is difficult to control and multiple
T-DNAs tend to integrate inefficiently (Naqvi et al., 2009). Cur-
rently, only direct DNA transfer can introduce routinely and reliably
multiple unlinked genes into plants, allowing plants carrying up to
15 different transgenes to be produced in one generation (Naqvi
et al., 2009).
Direct DNA transfer with separate vectors usually results in
transgene integration at a single random locus in the form of a
multigene array, regardless of how many different transformation
cassettes have been used (Altpeter et al., 2005; Kohli et al., 2006).
The integrated array may contain any number of transgenes from 1
to n (where n is the maximum input gene number) with the distribu-
tion within the transgenic population tending to describe a skewed
normal curve as would be expected from random sampling with
selection against zero integration events and for larger numbers
of integrated transgenes (Kohli et al., 2003). Input transgenes
once integrated remain linked and do not segregate in subsequent
generations (Wu et al., 2002; Altpeter et al., 2005). This feature is
important when large numbers of genes are considered, because
a much larger transgenic population would be required if each
integration event were independent (Altpeter et al., 2005).
Chen et al., (1998) successfully transformed rice (Oryza sativa)
plants by particle bombardment with 13 separate plasmids con-
taining different marker genes, and regenerated plants carrying
and expressing all the input genes at one locus. Subsequently,
Wu et al., (2002) transformed rice with nine transgenes also by
particle bombardment and found that nonselected transgenes
were present along with the selectable marker in approximately
70% of the plants and that 56% carried seven or more genes. This
percentage was much higher than expected given the independent
integration frequencies, in accordance with a model suggesting
that the integration of one transgene promotes the cointegration
of more input DNA at the same locus (Kohli et al., 1998). All nine
transgenes were expressed, and the expression of each gene was
independent of the others (Wu et al., 2002) (Table 1).
The position of transgene integration also influences the level
and stability of expression in both transformation methods. For
example, the transgenes can be integrated at a silencing locus
(position-dependent silencing) or influenced by nearby regulatory
sequences, such as enhancers (Topping et al., 1991). The inte-
gration mechanism does not appear to be sequence dependent.
Contrary to the prevailing view that the repetitious use of the same
promoter may lead to the likelihood of transcriptional silencing,
Fig. 1. Strategies to modulate organic compound levels in plants. A
and B are the precursors of C; C is the target product; D is the result of
the target product conversion. (1) Modification of the activity of enzymes
implicated in rate-limiting steps in the target pathway by modulation of
one or two key enzymes, or multiple enzymes. (2) Upstream precursors
enhancement by increasing flux through the pathway by overexpressing
the enzyme(s) that catalyze(s) the first committed step of the pathway.
(3) Blocked pathway branch points by RNA interference or antisense. (4)
Enhanced accumulation of target metabolite by increasing sink compart-
ments. (Zhu et al., 2013).
1. Increase enzyme activity
A B C
2. Upstream precursors overexpression
4. Generation of a metabolic sink
3. Blocking by gene silencing
Multigene engineering of plant metabolic pathways 3
a number of transgenic plants have been generated containing
five or more transgenes controlled by the same promoter with
no untoward effects (Naqvi et al., 2009). Particle bombardment
often generates large, high-copy-number transgenic loci, which
are believed to be prone to instability and silencing, but there are
many instances where this is not the case. For example, Golden
Rice provides a clear example in which higher transgene copy
numbers correspond to higher expression levels, ultimately lead-
ing to more b-carotene production in the endosperm (Datta et al.,
2003). Bacillus thurigiensis (Bt)-resistant rice containing multiple
transgene copies also performed well against a number of insect
pests in the greenhouse (Maqbool and Christou, 1999; Maqbool
et al., 2001) and under field conditions (Tu et al., 2000; Ye et al.,
2001), indicating that the transgenes were expressed efficiently.
Development of multigene transformation methods
Stacking and retransformation
Prior to the development of simultaneous transfer methods suit-
able for many genes, multiple transgenes could be stacked in plants
through successive rounds of crosses between different transgenic
lines (Ma et al., 1995; Datta et al., 2002) or by the retransformation
of transformed plants with additional transgenes (Jobling et al.,
2002). However, both methods are time-consuming and labor-
intensive because of the need of multiple breeding generations
to complete the stacking process and the important segregation
risk, unless all the genes can be stacked close together on the
same chromosome. Both the time necessary for stacking and the
segregation risk increase with the number of transgenes. In the
case of sequential transformation, multiple selectable markers (or
marker excision and reuse) are also required.
Standard T-DNA and bombardment vectors
Both Agrobacterium-mediated transformation and direct DNA
transfer involve the use of vectors that are optimized to replicate
efficiently in Escherichia coli and to facilitate subcloning, which
benefit from the vector remaining small. Vectors become increas-
ingly unstable and prone to eject DNA when too much is inserted.
The shear forces during particle bombardment can lead to frag-
mentation as well. It also becomes increasingly difficult to find
restriction enzymes that cut at a unique site as more transgenes
are introduced into the vector. Therefore, as the number of input
genes increases, standard vectors are largely restricted to use with
unlinked transgenes. As discussed above, Agrobacterium-mediated
transformation turns out to be progressively less efficient as the
number of separate T-DNAs increases, so for the highest numbers
of transgenes only direct DNA transfer can be carried out with
standard vectors. This problem has been solved to a certain extent
by transforming plants with two bacterial strains, each carrying
T-DNAs containing two or more transgenes; however, direct DNA
transfer remains efficient with up to 15 unlinked transgenes and
no upper limit has yet been determined. In the context of metabolic
engineering, standard expression vectors have allowed the stable
expression of several transgenes in maize (Zea mays) to recreate
partial metabolic pathways (Zhu et al., 2008; Naqvi et al., 2010).
High-capacity T-DNA vectors
The limitations of MGT using A. tumefaciens have been ad-
dressed in part by the development of systems based on high-
capacity artificial chromosome vectors with the ability to integrate
large DNA fragments. These systems use the capacity of bacterial
artificial chromosome (BAC) vectors and combine them with the
components of standard binary vectors, resulting in chimeric binary
vectors, such as BIBAC and TAC. Initially, these vectors still suf-
fered from the cumbersome cloning procedure due to the lack of
unique restriction sites, but this issue has been taken care of by
combining the vectors with Gateway site-specific recombination
technology (Vega et al., 2008). Multisite and MultiRound Gateway
systems have been used to integrate up to seven genes into the
plant genome (Buntru et al., 2013).
Split reading frames
The use of linker peptides can also facilitate MGT by letting
several polypeptides to be encoded in a single open reading
frame (ORF) controlled by a single promoter. For example, the
Foot-and-mouth disease virus 2A polyprotein system allows the
coexpression of up to four polypeptides in tobacco (Nicotiana
tabacum) plants (Møldrup et al., 2011; Lee et al., 2012; Sun et
Number of input
transgenes Plant Results Reference
17% of plants contained all input transgenes.
60% of all transgenic lines carried all three transgenes.
Introduced psy1 and crtI (carotenoid pathway), Dhar (ascorbate pathway) and folE (folate pathway) using an
unlinked direct DNA transfer co-transformation strategy to increase levels of β-carotene, folate and ascorbate
in the endosperm. Achieved significant increases in all three nutrients providing the first example towards
50% of transgenic plants contained all four input transgenes.
More than 20% of the plants contained and expressed all four input transgenes (fully assembled secretory
All transgenic plants contained at least two transgenes (mostly marker genes) and 16% contained all input
Introduced psy1, crtI, lycb, bch and crtW genes using an unlinked direct DNA transfer co-transformation
strategy aiming to generate a range of genotypes and phenotypes to dissect the carotenoid pathway.
Recovered maize plants with a range of phenotypes reflecting different carotenoid profiles.
Non-selected transgenes were present along with the selectable marker: 70% of the plants; 56% carried
seven or more transgenes.
85% of the plants contained more than two, and 17% more than nine of the introduced transgenes.
Romano et al. (2003)
Sivamani et al. (1999); Maqbool et al. (2001)
Naqvi et al. (2009)
Wu et al. (2002); Altpeter et al. (2005)
Nicholson et al. (2005)
Five Rice Agrawal et al. (2005)
Up to five Maize
Zhu et al. (2008)
Wu et al. (2002)
Thirteen Rice Chen et al. (1998)
EXAMPLES OF UNLINKED GENES FOR CO-TRANSFORMATION USING MULTIGENE TRANSFORMATION IN PLANTS
Adapted from (Naqvi et al., 2010).
4 U. Zorrilla-López et al.
al., 2012). The 2A linker is less than 20 amino acids in length
and has the ability to cleave its own C-terminus, thus releasing
downstream polypeptides after synthesis (Halpin, 2005). In the
context of metabolic engineering, the Paracoccus crtW and crtZ
genes were simultaneously expressed as a polyprotein with an
intervening 2A linker in transgenic tobacco and tomato (Solanum
lycopersicum) plants to generate novel ketocarotenoids (Ralley et
al., 2004). More recently, the genes for phytoene synthase and caro-
tene desaturase have been expressed in soybean (Glycine max)
seeds with an intervening 2A linker, using either the b-conglycinin
or the cauliflower mosaic virus 35S (CaMV35S) promoter (Kim et
al., 2012). Only the b-conglycinin promoter produced seeds with
orange endosperm, indicating the accumulation of b-carotene, and
this corresponded to high mRNA levels in the transgenic seeds. In
contrast, the CaMV35S construct generated high mRNA levels in
the leaves of transgenic plants (Kim et al., 2012). Attributes and
limitations of the key MGT methods are illustrated in Fig. 2.
Controlling the expression of multiple transgenes
As discussed above, a number of studies have shown that the
same promoter can be used to drive multiple transgenes without
negative effects, such as. the strong endosperm-specific expression
of three transgenes in maize achieved using the barley (Hordeum
vulgare) dhordein promoter (Naqvi et al., 2009). Other studies have
indicated that repetitive use of the same promoter can encourage
(although probably not directly trigger) transgene silencing (Mourrain
et al., 2007). This observation may reflect several underlying factors,
such as the presence of potential secondary structures that could
interact in trans to promote de novo methylation, or the intrinsic
activity of the promoters generating enough mRNA to saturate the
polyadenylation machinery of the cell, allowing the formation of
hairpin RNAs. These effects are also context dependent, based
on the integration site and the juxtaposition of transgene copies,
some of which may integrate ‘head-to-head’, thus encouraging the
formation of double-stranded RNA at the junction of two opposing
promoters (Kohli et al., 2006).
Therefore, although it is by no means certain that using the
same promoter for different transgenes will have a negative impact
on transgene expression, various strategies have been devised to
avoid the possibility. Examples include the use of natural diverse
promoters with the same or similar activity (for instance, five differ-
ent endosperm-specific promoters were used in maize to achieve
Gene stacking Re-transformation
x + +
+ + +
A B C D E F
Fig. 2. Multigene transformation (MGT) methods for metabolic engineering. Schematic summary of the principles of different methods for mul-
tigene transfer. Each panel shows a different method and charts the origin, fate, and activity of two different transgenes (red and blue blocks, with
promoters shown as sideways arrows). The corresponding products at the level of mRNA (undulating lines) and polypeptides (discs) are shown in
matching colors. (A) In the gene stacking approach, plants already carrying transgenes 1 and 2 are crossed to bring both genes into the same line. The
genes are integrated and expressed independently (diagonal slash) and may therefore segregate in later generations. Therefore, a backcross program
is needed to bring the two transgenes to the same locus. (B) In the retransformation approach, plants already carrying transgene 1 are transformed
with transgene 2, to bring both genes into the same line. The genes are integrated and expressed independently (diagonal slash) and may therefore
segregate in later generations. A backcross program is also needed in this case to bring the two transgenes to the same locus. (C) In the unlinked
transformation approach, transgenes 1 and 2 are introduced into wild-type plants using separate vectors. All genes tend to integrate at the same locus,
which is random, and may integrate in tandem (shown here) or in head-to-head or tail-to-tail conformations, occasionally with intervening genomic DNA
sequences. Although panels (A-C) show individual transgenes as blue and red blocks, the same principles of integration and segregation also apply to
groups of linked transgenes. (D) In the linked transformation approach, the transgenes are arranged in tandem on a single vector. The entire construct
tends to integrate so the integrated transgenes are arranged in the same order as on the vector. This approach becomes increasingly difficult with
more transgenes, unless high-capacity BIBAC/TAC vectors are employed. (E) In the split reading frame approach, two genes are expressed as a fusion
protein linked by the 2A peptide from the Food-and-mouth disease virus, resulting in the expression of polycistronic mRNA and a polyprotein, which
is self-cleaved into proteins 1 and 2, although each retains part of the 2A peptide (black circles). (F) In the operon approach, two or more genes are
expressed as an operon yielding a polycistronic mRNA, but the proteins are translated independently via internal ribosome entry sites. This approach is
only feasible for genes expressed in plastids and is therefore suitable for plants, such as tobacco and a small number of other species that are amenable
to plastid transformation (as shown), but not currently for cereal crops, such as maize (shown in the other panels).
Multigene engineering of plant metabolic pathways 5
the high-level expression of five carotenogenic genes; Zhu et al.,
2008), and the use of synthetic or modified promoters to reduce
the amount of sequence identity (Naqvi et al., 2010; Peremarti et
Most promoters used in plant biotechnology are unidirectional,
but bidirectional promoters are becoming increasingly useful for
MGT because they allow the simultaneous expression of two gene
products. For example, the human b-casein gene and a bacterial
marker gene encoding luciferase have been expressed using the
auxin-inducible, bidirectional mannopine synthase (mas) promoter
in transgenic potato (Solanum tuberosum cv. Bintje) plants to
increase their nutritional value (Chong et al., 1997).
Promoter activity depends on the availability and activity of the
transcription factors, so that the expression of such transcription
factors can activate several target genes. For example, ectopic
expression of the maize C1 and R chimeric transcription factors in
soybean upregulated a suite of endogenous isoflavonoid biosyn-
thetic genes encoding phenylalanine ammonia-lyase, cinnamic acid
4-hydroxylase, chalcone isomerase, chalcone reductase, flavanone
3-hydroxylase, dihydroflavonol reductase, and flavonol synthase,
doubling the isoflavonoid levels in the seeds (Yu et al., 2003).
The number of promoters can also be reduced by using the
split ORF method based on the 2A linker peptide discussed above,
or operon-based methods in which the genes are arranged in
tandem to yield a polycistronic mRNA, of which the ORFs are
translated independently. The latter method is only suitable for
plastid transformation, because the plastid genome is arranged
into operons reflecting its prokaryotic origin. Plastid transformation
Pyruvate + GA3P
Fig. 3. Carotenoid biosynthetic pathway in plants and equivalent steps in bacteria (Farré et al., 2010, 2011). Enzymes in the red ovals are from
bacteria. Abbreviations: CRTB, bacterial phytoene synthase; CRTI, bacterial phytoene desaturase, which catalyze all desaturation and isomerization reac-
tion from phytoene to lycopene; CRTISO, carotenoid isomerase; CRTY, bacterial lycopene b-cyclase; CRTZ, bacterial b-carotene hydroxylase; CYP97C,
heme-containing cytochrome P450 carotene e-ring hydroxylase; DMAPP , dimethylallyl diphosphate; DXP , 1-deoxy-D-xylulose 5-phosphate; DXR, DXP
reductoisomerase; DXS, DXP synthase; GA3P , glyceraldehyde 3-phosphate; GGPP , geranylgeranyl diphosphate; GGPPS, GGPP synthase; HDR, HMBPP
reductase; HMBPP , hydroxymethylbutenyl 4-diphosphate; HYDB, b-carotene hydroxylase [non-heme di-iron b-carotene hydroxylase (BCH) and heme-
containing cytochrome P450 b-ring hydroxyalses (CYP97A and CYP97B)]; IPP , isopentenyl diphosphate; IPPI, isopentenyl diphosphate isomerase; LYCB,
lycopene b-cyclase; LYCE, lycopene ε-cyclase; MEP , methylerythritol 4-phosphate; PDS, phytoene desaturase; PSY, phytoene synthase; VDE, violaxanthin
deepoxidase; ZDS, ζ-carotene desaturase; ZEP , zeaxanthin epoxidase; Z-ISO, ζ-carotene isomerase.
6 U. Zorrilla-López et al.
with operon-like multigene constructs has been used to produce
astaxanthin in tobacco by expressing b-carotene ketolase and
b-carotene hydroxylase (Hasunuma et al., 2008). Similarly, the
production of polyhydroxybutyric acid (PHB) in plastids has been
achieved by expressing the phbC-phbB-phbA genes of Ralstonia
eutropha using the T7g10 promoter (Lössl et al., 2005).
One of the key challenges in metabolic engineering is that any
targeted pathway must be understood in detail before interventions
are made, to avoid wasting resources on the development of futile
transgenic lines. In other words, for the longer and more complex
pathways, large numbers of transgenic lines must be developed and
tested independently before the most suitable intervention points are
identified. Combinatorial transformation, a concept developed by
Zhu et al., (2008), elegantly solves this challenge and simultaneously
turns the irritating random nature of transgene integration during
gene transfer to plants into an advantage. The approach is based
on the creation of metabolic libraries comprising plants transformed
with random selections of particular transgenes. For example, the
targeted analysis of five transgenes would require the generation
of five transgenic lines carrying individual transgenes, plus other
lines carrying combinations (perhaps created by stacking), each
of which would then be subject to metabolic profiling to determine
the impact on the target pathway. In combinatorial transformation,
this idea is reversed by taking advantage of the scattergun nature
of transgene integration: instead of selecting specific transgenic
lines containing particular combinations of transgenes, the aim is
to look at all the transgenic lines and with as much diversity as
possible. Combinatorial transformation with five transgenes would
therefore generate many different lines, some containing single
transgenes, others two or three or four, and some with all five. These
lines constitute a diverse library of metabolic potential, produced
in a single generation. Hence, subsequent metabolic profiling
helps to identify bottlenecks in the pathway and the best interven-
tion points, even if effective intervention can only be achieved by
multiple transgenes. The combinatorial approach is analogous to
the use of factorial designs to test different parameters rather than
focusing on the variation of one parameter at a time.
In the context of metabolic engineering, the carotenoid biosyn-
thesis pathway in maize has been investigated by combinatorial
transformation, allowing the identification and complementation of
rate-limiting steps that affect the accumulation of b-carotene and
other nutritionally important carotenoids, such as lutein, zeaxanthin,
and lycopene. This approach has also allowed the pathway to be
extended beyond its natural end-point to produce compounds,
such as astaxanthin, revealing competition between b-carotene
Fig. 4. Vitamin E biosynthesis in plants (Farré
et al., 2012). Tocochromanols are synthesized
on the inner chloroplast membrane from
precursors derived from the shikimate and
2-C-methyl-D-erythritol 4-phosphate (MEP)
pathways. The shikimate pathway contributes
the head-group precursor homogentisic acid
(HGA), whereas the MEP pathway gives rise
to the side-chain precursors phytyldiphosphate
(PDP) and geranylgeranyldiphosphate (GGDP).
The first committed step in the reaction is the
cytosolic conversion of q-hydroxyphenylpyruvic
acid (HPP) to HGA by q–hydroxyphenylpyruvic
acid dioxygenase (HPPD). HGA is then prenyl-
ated with either PDP or GGDP to produce the
intermediates 2-methyl-6-phytyl benzoquinone
(MPBQ) and 2-methyl-6-geranylgeranylplasto-
quinol (MGGBQ). A second methyl group is
added by MPBQ methyltransferase (MPBQ-
MT) in the tocopherol branch and MGGBQ
methyltransferase (MGGBQ-MT) in the tocotri-
enol branch, producing the intermediates 3-di-
(DMGGBQ). All four of these intermediates
are substrates for tocopherol cyclase (TC),
which produces d and g tocopherols and toco-
trienols. Finally, g-tocopherol methyltransferase
(g-TMT) catalyses a second ring methylation
to yield a and b tocopherols and tocotrienols.
Other abbreviations: GGDR, geranylgeranyl
diphosphate reductase; HGGT; homogentisate
geranylgeranyl transferase; HPT, homogentis-
ate phytyltransferase. PDS, VTE1, VTE2, VTE3
and VTE4 correspond to genes cloned from
Arabidopsis Thaliana that are homologous to
HPPD, HPT, MPBQ-MT, TC, and g-TMT genes,
Multigene engineering of plant metabolic pathways 7
hydroxylase and bacterial b-carotene ketolase for substrates (Zhu
et al., 2008).
Combinatorial transformation has also been used to combine
genes from several different metabolic pathways to identify com-
binations that allow the simultaneous accumulation of different
compounds. For example, maize plants had been generated that
coincidently accumulated high levels of vitamins A, C and B9 (folate)
(Naqvi et al., 2009).
Synthetic biology as the next step for multigene meta-
Synthetic biology describes the de novo assembly of genetic
systems using prevalidated components (Haseloff and Ajioka,
2009). In the context of metabolic engineering in plants, a syn-
thetic biology approach would utilize specific promoters, genes,
and other regulatory elements to create ideal genetic circuits that
facilitate the accumulation of particular metabolites. The concept of
synthetic biology creates engineering and mathematical modeling
to predict and test the behavior of the resulting system, which can
be considered as the next step in multigene metabolic engineer-
ing because it removes any dependence on naturally occurring
sequences and allows the design of ideal functional genetic circuits
from first principles. Thus far, most work on synthetic biology has
been accomplished with microorganisms, in spite of still some
limiting factors, such as the ability of current methods to assemble
complex DNA molecules encoding multiple genetic components in
predefined arrangements (Weber et al., 2011). Simple synthetic
biology approaches have been described in plants, mostly in the
context of signaling pathways and development, but also in the
development of phytodetectors (Zurbriggen et al., 2012) and bio-
fortified crops (Naqvi et al., 2009).
The use of synthetic biology in development as well as me-
tabolism is important because it not only controls the metabolic
capacity of a cell, but also steps one level up in terms of organiza-
tion and use of particular promoters and genes that control devel-
opmental processes to generate novel tissues, in which the cells
have specialized biosynthetic or storage functions to accumulate
target products in particular organs. This approach will facilitate
the achievement of goals that are unattainable by conventional
genetic engineering, such as the development of novel organisms
with medical functions, the production of biofuels, and the removal
of hazardous waste (Purnick and Weiss, 2009).
Applications of MGT for pathway engineering
Metabolic pathways leading to complex organic molecules, such
as vitamins (Figs 3, 4 and 5), polyunsaturated fatty acids (Fig. 6),
and secondary metabolites often comprise a large number of genes,
enzymes, and feedback mechanisms, limiting our ability to modulate
these pathways by single-gene transformation. The introduction
of multiple genes is necessary to understand the bottlenecks and
identify and complement the rate-limiting steps (Zhu et al., 2008).
The metabolic engineering of vitamin synthesis is necessary
because many staple crops lack adequate amounts of these vital
compounds. For example, vitamin A is required (as retinal) for
blindness prevention and (as retinoic acid) for development and
maintenance of a healthy immune system; vitamin E is an important
antioxidant defense compound that quenches free radicals and
protects against lipid peroxidation (Zhu et al., 2012); and folate
plays a central metabolic role, including DNA synthesis. None of
these compounds are present at high levels in cereal grains, and
more than one half of the world’s population suffers from deficiency
diseases because they rely on a cereal-based diet (Fitzpatrick et
al., 2012). Consequently, multigene metabolic engineering in plants
has focused on carotenoid biosynthesis (vitamin A), tocochromanol
synthesis (vitamin E), and folate synthesis (Naqvi et al., 2009, 2011).
Polyunsaturated fatty acids (PUFAs) are lipids that are needed
Fig. 5. The plant folate biosynthesis pathway (Naqvi et
al., 2009). Folates are tripartite molecules consisting of
pteridine, p-aminobenzoate (PABA), and glutamate moi-
eties, with pteridines synthesized in the cytosol and PABA
in the plastids. These moieties are transported to the mitochondria, where they condense to form dihydropteroate and are conjugated to glutamate.
DHN, dihydroneopterin; -P/-PP/-PPP , mono/di/triphosphate; DHM, dihydromonapterin; HMDHP , hydroxymethyldihydropterin.
8 U. Zorrilla-López et al.
not only as energy molecules, but for more specific activities, such
as maintenance of the nervous system, the immune system, and
prevention of atherosclerosis (Benatti et al., 2004). PUFAs with
dual roles as energy providers and essential nutrients include the
omega-3 group, e.g. a-linolenic acid (ALA), eicosapentaenoic
acid (EPA), and docosahexaenoic acid (DHA), and the omega-6
group, e.g. linoleic acid (LA) and arachidonic acid (AA) (Fig. 6).
Humans cannot synthetize PUFAs because they lack methyl-end
desaturases, and such modules must be accessed through the
diet, particularly in fish and other seafood (Benatti et al., 2004).
Because there is little access to seafood across large parts of
the world, metabolic engineering has been used to increase the
abundance of essential PUFAs in transgenic plants.
Finally, secondary metabolites are complex molecules that are
not required for housekeeping functions, but, nevertheless, provide
advantages to plants, e.g. by attracting pollinators and repelling
pests and pathogens. The three major types of secondary metabolite
are the alkaloids, terpenoids/isoprenoids, and phenolics. Because
plants have evolved to produce such molecules to control the be-
havior of animals and microbes, many secondary metabolites have
pharmacological properties in humans or can be used as flavors,
fragrances, and crop protection products (Miralpeix et al., 2013).
The first use of multigene engineering to modulate the vitamin
content of plants was the development of Golden Rice (Ye et al.,
2000). This is a transgenic rice line engineered to produce high
levels of b-carotene through the expression of Pantoea ananatis
phytoene desaturase (PaCrtI), daffodil (Narcissus spp.) phytoene
synthase (psy1), and daffodil lycopene b-cyclase (lycb) (Fig. 3).
The original Golden Rice line produced 1.6 mg/g dry weight (DW)
of b-carotene, but the replacement of daffodil psy1 with the more
active maize enzyme in Golden Rice 2 boosted the b-carotene
content to 31 mg/g DW (Paine et al., 2005).
General strategies to increase carotenoid levels in plants include
increasing the availability of carotenoid precursors, expressing en-
zymes in the common (linear) part of the pathway, and shifting the
flux from the a- to the b-branch (Fig. 3). In canola (Brassica napus),
Agrobacterium-mediated MGT was used to introduce seven different
transgenes in order to reconstruct the entire carotenoid pathway,
including an extension which allowed the production of ketocarot-
enoids (Fujisawa et al., 2009). The input genes were isopentenyl
pyrophosphate isomerase (idi), geranylgeranyldiphosphate (GGPP)
synthase (CrtE), bacterial phytoene synthase (CrtB), CrtI, lycopene
b-cyclase (CrtY), and the genes for two additional enzymes (CrtZ
and CrtW) that catalyze downstream steps converting b-carotene
into ketocarotenoids. This strategy achieved a 30-fold increase in
total carotenoid content (657 mg/g fresh weight [FW]) and a 1070-
fold increase in b-carotene (214 mg/g FW). In maize, MGT with
maize psy1, PaCrtI, lycb of Gentiana lutea (great yellow gentian)
(Gllycb), and Paracoccus sp. CrtW produced 35.64 mg/g DW of b-
carotene (Zhu et al., 2008). Carotenoid multigene engineering has
also been applied in tomato, potato, and wheat (Triticum aestivum)
(Dharmapuri et al., 2002; Diretto et al., 2007; Cong et al., 2009).
The folate biosynthesis pathway (Fig. 4) involves the integra-
tion of two independent branches (pterin and p-aminobenzoate).
The total folate content can be increased by modulating individual
enzymes in either branch, but the best results are achieved by
the simultaneous modulation of both branches by multigene
engineering. In the most successful report, a 100-fold increase
of total folate (38.3 nmol/g FW) was achieved in rice by expres-
sion of the Arabidopsis Thaliana GTP cyclohydrolase 1 (GCH1)
and aminodeoxychorismate synthase (ADCS) that enhances the
the cytosolic (pterin) branch and p-aminobenzoate branch of the
pathway, respectively (Storozhenko et al., 2007).
The synthesis of tocochromanols (vitamin E) involves a complex
pathway (Fig. 5). Vitamin E levels can be elevated by increasing
the total tocopherol content or enhancing the production of spe-
cific tocochromanols with the most potent vitamin E activity (a-
tocopherol). The constitutive expression of two Arabidopsis cDNA
Linoleic acid (LA) C18:2n-6
γ-Linoleic acid (GLA) C18:3n-6
Dihomo- γ Linoleic
acid (DGLA) C20:3n-6
Adrenic acid (ADA)
acid (DPA) C22:5n-6
α-linoleic acid (ALA) C18:2n-3
Stearidonic acid (SDA) C18:4n-3
Eicosatetraenoic acid C20:4n-3
ω3/ Δ17 desaturase
Fig. 6. The biosynthetic pathway of poly-
unsaturated fatty acids (PUFAs) in plants.
The conventional Δ6-desaturase/Δ6-elongase
and the alternative Δ9-elongase pathways start
from linolenic (LA; 18:2n-6) and a-linolenic acid
(ALA; 18:3n-3), respectively. PUFAs synthesis
takes place in plastids. Since mammals lack the
desaturases which are responsible to produce LA
and ALA, both of them are considered essential
fatty acids and must be obtained from the diet.
Adapted from Vrinten et al., (2007).
Multigene engineering of plant metabolic pathways 9
clones encoding p-hydroxyphenylpyruvate dioxygenase (HPPD)
and 2-methyl-6-phytylplastoquinol methyltransferase (MPBQ MT)
increased the tocopherol content threefold in transgenic maize
(Naqvi et al., 2011). In soybean, the expression of homogentisate
phytyltransferase (HPT1), HPPD, TyrA (responsible for the syn-
thesis of HPP from prephenate), and geranylgeranyldiphosphate
reductase (GGDR) increased the tocochromanol content by 15-fold
(4806 mg/g DW) (Karunanandaa et al., 2005).
The current state-of-the-art in vitamin engineering is the si-
multaneous modulation of multiple vitamin pathways in the same
plant, as reported by Naqvi et al., (2009) through the expression of
maize psy1 and PaCrtI, representing the carotenoid biosynthesis
pathway, rice dehydroascorbate reductase (DHAR) to increase
vitamin C (ascorbate) levels, and E. coli FolE to enhance folate
accumulation (Figs 3, 4 and 5). The transgenic kernels contained
169-fold the normal amount of b-carotene, 6-fold the normal amount
of ascorbate, and 2-fold the normal amount of folate.
Long-chain polyunsaturated fatty acids
Several different oil-seed crops have been transformed with
multiple genes representing the PUFA biosynthesis pathway
(Sayanova and Napier, 2004). For example, to obtain very long
chain PUFAs from ALA and LA, it is necessary to introduce at least
three transgenes encoding the desaturases and elongases required
for sequential enzymatic reactions (Beaudoin et al., 2000; Hong,
2002) (Fig. 6). Soybean seeds with higher levels of EPA have
been produced by expressing Mortierella alpina Δ6-desaturase,
Δ5-desaturase and Δ6-elongase transgenes plus omega-3Δ17-
desaturase from Saprolegnia diclina, and omega-3Δ15-desaturase
from Arabidopsis (Kinney et al., 2011) (Fig. 6). This strategy was
chosen to maximize the accumulation of omega-3 very long
chain PUFAs by converting omega-6 PUFAs into their omega-3
Although metabolic engineering can be used to enhance the
production of secondary metabolites, it is challenging because
of the complexity of the pathways and the shuffling of precursors
and intermediates between compartments (Miralpeix et al., 2013).
The availability of precursors can be augmented by modulating
the accessibility of basic nitrogen, carbon, and sulfur compounds,
including the synthesis of amino acids, such as phenylalanine,
tryptophan and tyrosine, and enhance both primary and secondary
metabolism simultaneously (Pichersky and Gang, 2000).
Artemisinin is used as a drug against malaria caused by Plas-
modium falciparum and has been produced in transgenic tobacco
by multigene engineering (Farhi et al., 2011). A mega-vector was
constructed, containing the Artemisia annua (sweet wormwood)
genes for cytochrome P450 reductase (CPR), amorpha-4,11-diene
synthase (ADS), amorpha-4,11-diene monooxygenase (CY-
P71AV1), and artemisinic aldehyde Δ-11(13) reductase (DBR2),
and the yeast 3-hydroxy-3-methylglutaryl coenzyme A reductase
(tHMG), each under the control of a different promoter. In a separate
vector, the ADS sequence was fused to a COX4 signal peptide for
import into the mitochondria, to boost the production of terpenoids.
The vectors were introduced into tobacco plants by Agrobacteri-
um-mediated transformation and the resulting transgenic plants
produced amorpha-4,11-diene at levels of 26–72 ng/g FW (normal
ADS) and 137–827 ng/g fresh weight (mitochondrial ADS).
MGT has also been used to boost the production of the natural
polyester PHB in sugarcane (Saccharum spp.), but required the
introduction of the enzymes b-ketothiolase (PHAA), acetoacetyl-
reductase (PHAB), and PHB synthase (PHAC) from Ralstonia
eutropha, and was achieved by particle bombardment with separate
vectors (Petrasovits et al., 2007). The resulting plants accumulated
PHB to 1.88% DW in their leaves.
Opium poppy (Papaver somniferum) is one of the most impor-
tant medicinal plants because it is the source of the cancer drug
noscapine, the muscle relaxant papaverine, and analgesic and
narcotic drugs, such as morphine and codeine. Morphine-type
alkaloids have been produced in plant cell cultures since the
1970s (Rischer et al., 2013). MGT has been used in this case to
inhibit several genes in a pathway by virus-induced gene silenc-
ing to gain insight into the final six steps of morphine biosynthesis
(Wijekoon and Facchini, 2012). The inhibition of SalSyn, SalR,
T6ODM, and CODM protein levels correlated with lower morphine
yields and a substantial increase in the accumulation of reticuline,
salutaridine, thebaine, and codeine, respectively. In contrast, the
inhibition of SalAT and COR resulted in higher levels of salutaridine
Looking to the future
MGT is becoming essential as a strategy for metabolic engineer-
ing because it is clear that single-point interventions are inadequate
to achieve ambitious metabolic goals even when dealing with a
single pathway, and are unsuitable for the simultaneous engineering
of different pathways, as illustrated bymultivitamin corn.
Although it is now straightforward to introduce and express
5-10 transgenes in the same transgenic line, this is not the ceiling
of the technique but rather the current status quo. Theoretically,
there is no maximum number of transgenes that can be introduced
at once, as demonstrated in microbes in which large, low-copy
number vectors, such as BACs, P1-derived artificial chromosomes
in bacteria, and yeast artificial chromosomes, are suitable for the
introduction of hundreds of genes. This trend is emerging in plants,
with large-capacity T-DNA-based vectors, but unlinked genes and
direct DNA transfer allow the use of smaller vectors and achieve
the same goals, because integration occurs at a single locus,
therefore providing a suitable platform for strategies based on
synthetic biology. In order to improve the potential of multigene
transfer, a combination of linked and unlinked strategies could be
developed to engineer more complex novel high-flux pathways and
even combine these with the strategies shown in Fig. 1.
Multigene transfer must still overcome certain practical barriers
that occur after gene integration, e.g. silencing, transgene rear-
rangement, and interactions between transgenes. As discussed
above, repeated use of the same promoter does not necessarily
encourage silencing, but may be a factor when another trigger
is present; therefore, strategies have been developed based on
promoter diversity or use of artificial or chimeric promoters to re-
duce the risk of unproductive interactions (Peremarti et al., 2010).
Novel strategies to assess the risk of transgene rearrangement and
interactions with surrounding loci include site-specific recombina-
tion, targeted integration, and the use of engineered restriction
enzymes, especially those based on zinc fingers and transcription
activator-like effectors (Li et al., 2012).
Metabolic pathways display a high degree of connectivity in
10 U. Zorrilla-López et al.
larger networks, especially when metabolites are involved in two
or more pathways; hence, the introduction of a large number of
input genes has the potential to generate unintended and unpre-
dicted effects. However, an interesting study showed that transfer
of the entire pathway for dhurrin biosynthesis (a tyrosine-derived
cyanogenic glucoside) into Arabidopsis had no significant im-
pact on the wider transcriptome and metabolome, whereas the
transfer of an incomplete pathway induced significant changes in
morphology, transcriptome, and metabolome, probably through
metabolic crosstalk or detoxification reactions (Kristensen et al.,
2005). Monitoring changes at the gene, transcript, protein, and
metabolite levels is a challenge. In the future, it will be necessary
to integrate these data in the context of systems biology, in which
modeling is becoming a standard analytical tool for understanding
whole biological systems and predicting gene behavior (Purnick
and Weiss, 2009). Systems biology is also a necessary component
of synthetic biology, because it is critical to foresee the behavior
of synthetic genetic circuits in the context of the wider organism.
Advances in systems biology and synthetic biology offer enormous
potential in terms of development of novel materials and energy
sources, improvement of agronomic traits, human health applica-
tions, and a better understanding of natural gene regulation (Naqvi
et al., 2009; Zurbriggen et al., 2012). For example, the expression
of three genes required for the conversion of acetyl-CoA to PHB
in plastids allows the production of bioplastics in plants (Bohmert-
Tatarev et al., 2011), and the introduction of five genes of the E.
coli glycolate catabolic pathway into Arabidopsis Thaliana plastids
reduces the loss of fixed carbon and nitrogen during photorespira-
tion, increasing plant biomass (Kebeish et al., 2007).
Research at the Universitat de Lleida is supported by the Ministerio
de Ciencia e Innovación (grants no. BFU2007-61413, BIO2011-23324,
BIO02011-22525, PIM2010PKB-0074, Acciones complementarias
BIO2007-30738-E and BIO2011-22525, and the Centre CONSOLIDER
on Agrigenomics), the European Union Framework 7 Program (SmartCell
Integrated Project 222716), the European Research Council IDEAS Ad-
vanced Grant Program (BIOFORCE) (to PC), the European Cooperation
in Science and Technology (COST Action FA0804), and RecerCaixa.
AGRAWAL P K, KOHLI A, TWYMAN R M, CHRISTOU P (2005). Transformation of
plants with multiple cassettes generates simple transgene integration patterns
and high expression levels. Mol Breed 16: 247-260.
ALTPETER F, BAISAKH N, BEACHY R, BOCK R, CAPELL T, CHRISTOU P, DANIELL
H, DATTA K, DATTA S, DIX P J, FAUQUET C, HUANG N, KOHLI A, MOOIBROEK
H, NICHOLSON L, NGUYEN T H, NUGENT G, RAEMAKERS K, ROMANO A,
SOMERS D A, STOGER E, TAYLOR N, VISSER R (2005). Particle bombardment
and the genetic enhancement of crops: myths and realities. Mol Breed 15: 305-327.
BEAUDOIN F, MICHAELSON L V, HEY S J, LEWIS M J, SHEWRY P R, SAYANOVA
O, NAPIER J A (2000). Heterologous reconstitution in yeast of the polyunsaturated
fatty acid biosynthetic pathway. Proc Natl Acad Sci USA 97: 6421-6426.
BENATTI P, PELUSO G, NICOLAI R, CALVANI M (2004). Polyunsaturated fatty acids:
Biochemical, nutritional and epigenetic properties. J Am Coll Nutr 23: 281-302.
BOHMERT-TATAREV K, MCAVOY S, DAUGHTRY S, PEOPLES O P, SNELL K D
(2011). High levels of bioplastic are produced in fertile transplastomic tobacco
plants engineered with a synthetic operon for the production of polyhydroxybutyr-
ate. Plant Physiol 155: 1690-1708.
BUNTRU M, GÄRTNER S, STAIB L, KREUZALER F, SCHLAICH N (2013). Delivery of
multiple transgenes to plant cells by an improved version of MultiRound Gateway
technology. Transgenic Res 22: 153-167.
CAPELL T, CHRISTOU P (2004). Progress in plant metabolic engineering. Curr Opin
Biotechnol 15: 148-154.
CHEN L, MARMEY P, TAYLOR N J, BRIZARD J-P, ESPINOZA C, D’CRUZ P, HUET
H, ZHANG S, DE KOCHKO A, BEACHY R N, FAUQUET C M (1998). Expression
and inheritance of multiple transgenes in rice plants. Nat Biotechnol 16: 1060-1064.
CHONG D K X, ROBERTS W, ARAKAWA T, ILLES K, BAGI G, SLATTERY C W,
LANGRIDGE W H R (1997). Expression of the human milk protein b-casein in
transgenic potato plants. Transgenic Res 6: 289-296.
CONG L, WANG C, CHEN L, LIU H, YANG G, HE G (2009). Expression of phytoene
synthase1 and carotene desaturase crtI genes result in an increase in the total
carotenoids content in transgenic elite wheat (Triticum aestivum L.). J Agric Food
Chem 57: 8652-8660.
DATTA K, BAISAKH N, MAUNG THET K, TU J, DATTA S K (2002). Pyramiding
transgenes for multiple resistance in rice against bacterial blight, yellow stem
borer and sheath blight. Theor Appl Genet 106: 1-8.
DATTA K, BAISAKH N, OLIVA N, TORRIZO L, ABRIGO E, TAN J, RAI M, REHANA
S, AL-BABILI S, BEYER P, POTRYKUS I, DATTA S K (2003). Bioengineered
‘golden’ indica rice cultivars with b-carotene metabolism in the endosperm with
hygromycin and mannose selection systems. Plant Biotechnol J 1: 81-90.
DHARMAPURI S, ROSATI C, PALLARA P, AQUILANI R, BOUVIER F, CAMARA
B, GIULIANO G (2002). Metabolic engineering of xanthophyll content in tomato
fruits. FEBS Lett 519: 30-34.
DIRETTO G, AL-BABILI S, TAVAZZA R, PAPACCHIOLI V, BEYER P, GIULIANO G
(2007). Metabolic engineering of potato carotenoid content through tuber-specific
overexpression of a bacterial mini-pathway. PLoS One 2:e350.
FARHI M, MARHEVKA E, BEN-ARI J, ALGAMAS-DIMANTOV A, LIANG Z, ZEEVI
V, EDELBAUM O, SPITZER-RIMON B, ABELIOVICH H, SCHWARTZ B, TZFIRA
T, VAINSTAIN A (2011). Generation of the potent anti-malarial drug artemisinin in
tobacco. Nat Biotechnol 29: 1072-1074.
FARRÉ G, SANAHUJA G, NAQVI S, BAI C, CAPELL T, ZHU C, CHRISTOU P (2010).
Travel advice on the road to carotenoids in plants. Plant Sci 179: 28-48.
FARRÉ G, BAI C, TWYMAN R M, CAPELL T, CHRISTOU P, ZHU C (2011). Nutritious
crops producing multiple carotenoids – a metabolic balancing act.
Trends Plant Sci 16: 532–540.
FARRÉ G, ZORRILLA U, BERMAN J, ZHU C, CHRISTOU P, CAPELL T (2012).
Increasing the vitamin E content of food by in-plant production. CAB Rev. 7: 030.
FITZPATRICK T B, BASSET G J C, BOREL P, CARRARI F, DELLAPENNA D, FRASER
P D, HELLMANN H, OSORIO S, ROTHAN C, VALPUESTA V, CARIS-VEYRAR
C, FERNIE A R (2012). Vitamin deficiencies in humans: can plant science help?
Plant Cell 24: 395-414.
FUJISAWA M, TAKITA E, HARADA H, SAKURAI N, SUZUKI H, OHYAMA K, SHIBATA
D, MISAWA N (2009). Pathway engineering of Brassica napus seeds using multiple
key enzyme genes involved in ketocarotenoid formation. J Exp Bot 60: 1319-1332.
GOMEZ GALERA S, PELACHO A M, GENE A, CAPELL T, CHRISTOU P (2007).
The genetic manipulation of medicinal and aromatic plants. Plant Cell Rep 26:
HALPIN C (2005). Gene stacking in transgenic plants – the challenge for 21st century
plant biotechnology. Plant Biotechnol J 3: 141-155.
HASELOFF J, AJIOKA J (2009). Synthetic biology: history, challenges and prospects.
J R Soc Interface 6: S389-S391.
HASUNUMA T, MIYAZAWA S-I, YOSHIMURA S, SHINZAKI Y, TOMIZAWA K-I,
SHINDO K, CHOI S-K, MISAWA N, MIYAKE C (2008). Biosynthesis of astaxanthin
in tobacco leaves by transplastomic engineering. Plant J 55: 857-868.
HONG H, DATTA N, REED D W, COVELLO P S, MACKENZIE S L, QIU X (2002).
High-level production of g-linolenic acid in Brassica juncea using a Δ6 desaturase
from Pythium irregulare. Plant Physiol 129: 354-362.
JOBLING S A, WESTCOTT R J, TAYAL A, JEFFCOAT R, SCHWALL G P (2002).
Production of a freeze–thaw-stable potato starch by antisense inhibition of three
starch synthase genes. Nat Biotechnol 20: 295-299.
KARUNANANDAA B, QI Q, HAO M, BASZIS S R, JENSEN P K, WONG Y-H H, JIANG
J, VENKATRAMESH M, GRUYS K J, MOSHIRI F, POST-BEITTENMILLER D,
WEISS J D, VALENTIN H E (2005). Metabolically engineered oilseed crops with
enhanced seed tocopherol. Metab Eng 7: 384–400.
KEBEISH R, NIESSEN M, THIRUVEEDHI K, BARI R, HIRSCH H-J, ROSENKRANZ
R, STÄBLER N, SCHÖNFELD B, KREUZALER F, PETERHÄNSEL C (2007).
Multigene engineering of plant metabolic pathways 11
Chloroplastic photorespiratory bypass increases photosynthesis and biomass
production in Arabidopsis Thaliana. Nat Biotechnol 25: 593-599.
KIM M-J, KIM J K, KIM H J, PAK J H, LEE J-H, KIM D-H, CHOI H K, JUNG H W, LEE
J-D, CHUNG Y-S, HA S-H (2012). Genetic modification of the soybean to enhance
the b-carotene content through seed-specific expression. PloS One 7: e48287.
KINNEY A J, CAHOON E B, DAMUDE H G, LIU Z-B (2011). Production of very long
chain polyunsaturated fatty acids in oilseeds plants. US Patent Application US
2011/0269983 A1, 66 United States: E. I. Du Pont de Nemours and Company.
KOHLI A, LEECH M, VAIN P, LAURIE D A, CHRISTOU P (1998). Transgene orga-
nization in rice engineered through direct DNA transfer supports a two-phase
integration mechanism mediated by the establishment of integration hot spots.
Proc Natl Acad Sci USA 95: 7203-7208.
KOHLI A, TWYMAN R M, ABRANCHES R, WEGEL E, STOGER E, CHRISTOU P
(2003). Transgene integration, organization and interaction in plants. Plant Mol
Biol 52: 247-258.
KOHLI A, GONZÁLEZ-MELENDI P, ABRANCHES R, CAPELL T, STOGER E, CHRIS-
TOU P (2006). The quest to understand the basis and mechanisms that control
expression of introduced transgenes in crop plants. Plant Signal Behav 1: 185-195.
KRISTENSEN C, MORANT M, OLSEN C E, EKSTRØM C T, GALBRAITH D W,
MØLLER B L, BAK S (2005). Metabolic engineering of dhurrin in transgenic
Arabidopsis plants with marginal inadvertent effects on the metabolome and
transcriptome. Proc Natl Acad Sci USA 102: 1779-1784.
LEE D-S, LEE K-H, JUNG S, JO E-J, HAN K-H, BAE H-J (2012). Synergistic effects
of 2A-mediated polyproteins on the production of lignocellulose degradation
enzymes in tobacco plants. J Exp Bot 63: 4797-4810.
LI L, PIATEK M J, ATEF A, PIATEK A, WIBOWO A, FANG X, SABIR J S M, ZHU J-K,
MAHFOUZ M M (2012). Rapid and highly efficient construction of TALE-based
transcriptional regulators and nucleases for genome modification. Plant Mol Biol
LÖSSL A, BOHMERT K, HARLOFF H, EIBL C, MÜHLBAUER S, KOOP H-U (2005).
Inducible trans-activation of plastid transgenes: expression of the R. eutropha phb
operon in transplastomic tobacco. Plant Cell Physiol 46: 1462-1471.
MA J K-C, HIATT A, HEIN M, VINE N D, WANG F, STABILA P, VAN DOLLEWEERD C,
MOSTOV K, LEHNER T (1995). Generation and assembly of secretory antibodies
in plants. Science 268: 716-719.
MAQBOOL S B, RIAZUDDIN S, NGUYEN T L, GATEHOUSE A M R, GATEHOUSE J
A, CHRISTOU P (2001). Expression of multiple insecticidal genes confers broad
resistance against a range of different rice pests. Mol Breed 7: 85-93.
MAQBOOL S B, CHRISTOU P (1999). Multiple traits of agronomic importance in
transgenic indica rice plants: analysis of transgene integration patterns, expres-
sion levels and stability. Mol Breed 5: 471-480.
MIRALPEIX B, RISCHER H, HÄKKINEN S T, RITALA A, SEPPÄNEN-LAAKSO T,
OKSMAN-CALDENTEY K-M, CAPELL T, CHRISTOU P (2013). Metabolic engineer-
ing of plant secondary products: which way forward? Curr Pharm Des (in press).
MOURRAIN P, VAN BLOKLAND R, KOOTER J M, VAUCHERET H (2007). A single
transgene locus triggers both transcriptional and post-transcriptional silencing
through double-stranded RNA production. Planta 225: 365-379.
MØLDRUP M E, GEU-FLORES F, OLSEN CE, HALKIER B A (2011). Modulation of sulfur
metabolism enables efficient glucosinolate engineering. BMC Biotechnol 11: 12.
NAQVI S, ZHU C, FARRE G, RAMESSAR K, BASSIE L, BREITENBACH J, PEREZ
CONESA D, ROS G, SANDMANN G, CAPELL T, CHRISTOU P (2009). Transgenic
multivitamin corn through biofortification of endosperm with three vitamins repre-
senting three distinct metabolic pathways. Proc Natl Acad Sci USA 106: 7762-7767.
NAQVI S, FARRÉ G, SANAHUJA G, CAPELL T, ZHU C, CHRISTOU P (2010). When
more is better: multigene engineering in plants. Trends Plant Sci 15: 48-56.
NAQVI S, FARRÉ G, ZHU C, SANDMANN G, CAPELL T, CHRISTOU P (2011).
Simultaneous expression of Arabidopsis ρ-hydroxyphenylpyruvate dioxygenase
and MPBQ methyltransferase in transgenic corn kernels triples the tocopherol
content. Transgenic Res 20: 177-181.
NICHOLSON L, GONZALEZ-MELENDI P, VAN DOLLEWEERD C, TUCK H, PERRIN
Y, MA J K-C, FISCHER R, CHRISTOU P, STOGER E (2005) A recombinant multi-
meric immunoglobulin expressed in rice shows assembly-dependent subcellular
localization in endosperm cells. Plant Biotechnol J 3: 115-127.
PAINE J A, SHIPTON C A, CHAGGAR S, HOWELLS R M, KENNEDY M J, VERNON
G, WRIGHT S Y, HINCHLIFFE E, ADAMS J L, SILVERSTONE A L, DRAKE R
(2005). Improving the nutritional value of Golden Rice through increased pro-
vitamin A content. Nat Biotechnol 23: 482-487.
PEREMARTI A, TWYMAN R M, GÓMEZ-GALERA S, NAQVI S, FARRÉ G, SABALZA
M, MIRALPEIX B, DASHEVSKAYA S, YUAN D, RAMESSAR K, CHRISTOU P,
ZHU C, BASSIE L, CAPELL T (2010). Promoter diversity in multigene transforma-
tion. Plant Mol Biol 73: 363-378.
PETRASOVITS L A, PURNELL M P, NIELSEN L K, BRUMBLEY S M (2007). Produc-
tion of polyhydroxybutyrate in sugarcane. Plant Biotechnol J 5: 162-172.
PICHERSKY E, GANG D R (2000). Genetics and biochemistry of secondary me-
tabolites in plants: An evolutionary perspective. Trends Plant Sci 5: 439-445.
PURNICK P E M, WEISS R (2009). The second wave of synthetic biology: from
modules to systems. Nat Rev Mol Cell Biol 10: 410-422.
RALLEY L, ENFISSI E M A, MISAWA N, SCHUCH W, BRAMLEY P M, FRASER
P D (2004). Metabolic engineering of ketocarotenoid formation in higher plants.
Plant J 39: 477-486.
RISCHER H, HÄKKINEN S T, RITALA A, SEPPÄNEN-LAAKSO T, MIRALPEIX B,
CAPELL T, CHRISTOU P, OKSMAN-CALDENTEY K-M (2013). Plants cells as
pharmaceutical factories. Curr Pharm Des (in press).
ROMANO A, RAEMAKERS K, BERNARDI J, VISSER R, MOOIBROEK H (2003).
Transgene organisation in potato after particle bombardment-mediated (co)
transformation using plasmids and gene cassettes. Transgenic Res 12: 461-473.
SAYANOVA O V, NAPIER J A (2004). Eicosapentaenoic acid: Biosynthetic routes
and the potential for synthesis in transgenic plants. Phytochemistry 65: 147-158.
SIVAMANI E, HUET H, SHEN P, ONG C A, DE KOCHKO A, FAUQUET C, BEACHY R
N (1999). Rice plant (Oryza sativa L.) containing Rice tungro spherical virus (RTSV)
coat protein transgenes are resistant to virus infection. Mol Breed 5: 177-185.
STOROZHENKO S, DE BROUWER V, VOLCKAERT M, NAVARRETE O, BLANC-
QUAERT D, ZHANG G F, LAMBERT W, VAN DER STRAETEN D (2007). Folate
fortification of rice by metabolic engineering. Nat Biotechnol 25: 1277-1279.
SUN H, LANG Z, ZHU L, HUANG D (2012). Acquiring transgenic tobacco plants with
insect resistance and glyphosate tolerance by fusion gene transformation. Plant
Cell Rep 31: 1877-1887.
TOPPING J F, WEI W, LINDSEY K (1991). Functional tagging of regulatory elements
in the plant genome. Development 112: 1009-1019.
TU J, ZHANG G, DATTA K, XU C, HE Y, ZHANG Q, KHUSH G S, DATTA S K (2000).
Field performance of transgenic elite commercial hybrid rice expressing Bacillus
thuringiensis δ-endotoxin. Nat Biotechnol 18: 1101-1104.
TWYMAN R M, KOHLI A, STOGER E, CHRISTOU P (2002). Foreign DNA: integra-
tion and expression in transgenic plants. Genetic Engineering: Principles and
Methods (Jane K. Setlow Ed.) (N Y) 24: 107-136.
VEGA J M, YU W, HAN F, KATO A, PETERS E M, ZHANG Z J, BIRCHLER J A (2008).
Agrobacterium-mediated transformation of maize (Zea mays) with Cre-lox site
specific recombination cassettes in BIBAC vectors. Plant Mol Biol 66: 587-598.
VRINTEN P, WU G, TRUKSA M, QIU X (2007). Production of polyunsaturated fatty
acids in transgenic plants. Biotechnol Genet Eng Rev 24: 263-280.
WEBER E, ENGLER C, GRUETZNER R, WERNER S, MARILLONNET S (2011).
A modular cloning system for standardized assembly of multigene constructs.
PLoS One 6: e16765.
WIJEKOON C P, FACCHINI P J (2012). Systematic knockdown of morphine pathway
enzymes in opium poppy using virus‐induced gene silencing. Plant J 69:1052-1063.
WU L, NANDI S, CHEN L, RODRIGUEZ R L, HUANG N (2002). Expression and
inheritance of nine transgenes in rice. Transgenic Res 11: 533-541.
YE X, AL-BABILI S, KLÖTI A, ZHANG J, LUCCA P, BEYER P, POTRYKUS I (2000).
Engineering the provitamin A (b-carotene) biosynthetic pathway into (carotenoid-
free) rice endosperm. Science 287: 303-305.
YE G, TU J, HU C, DATTA K, DATTA S K (2001). Transgenic IR72 with fused Bt gene
cry1Ab/cry1Ac from Bacillus thuringiensis is resistant against four lepidopteran
species under field conditions. Plant Biotechnol 18: 125-133.
YU O, SHI J, HESSION A O, MAXWELL C A, MCGONIGLE B, ODELL J T (2003).
Metabolic engineering to increase isoflavone biosynthesis in soybean seed.
Phytochemistry 63: 753-763.
ZHU C, NAQVI S, BREITENBACH J, SANDMANN G, CHRISTOU P, CAPELL T (2008).
Combinatorial genetic transformation generates a library of metabolic phenotypes
for the carotenoid pathway in maize. Proc Natl Acad Sci USA 105: 18232-18237.
ZHU C, SANAHUJA G, YUAN D, FARRÉ G, ARJÓ G, BERMAN J, ZORRILLA-LÓPEZ
U, BANAKAR R, BAI C, PÉREZ-MASSOT E, BASSIE L, CAPELL T, CHRISTOU P
(2013). Biofortification of plants with altered antioxidant content and composition: Download full-text
Genetic engineering strategies. Plant Biotechnol J 11: 129-141.
ZURBRIGGEN M D, MOOR A, WEBER W (2012). Plant and bacterial systems
biology as platform for plant synthetic bio(techno)logy. J Biotechnol 160: 80-90.
Further Related Reading, published previously in the Int. J. Dev. Biol.
Testicular germ cell tumors and related research from a historical point of view
Ivan Damjanov and Nicolai Wewer-Albrechtsen
Int. J. Dev. Biol. (2013) 57: 197-200
5 yr ISI Impact Factor (2011) = 2.959
Involvement of adipokines, AMPK, PI3K and the PPAR signaling pathways in ovarian
follicle development and cancer
Joëlle Dupont, Maxime Reverchon, Lucie Cloix, Pascal Froment and Christelle Ramé
Int. J. Dev. Biol. (2012) 56: 959-967
Metabolism throughout follicle and oocyte development in mammals
Esther Collado-Fernandez, Helen M. Picton and Rémi Dumollard
Int. J. Dev. Biol. (2012) 56: 799-808
Study an egg today to make an embryo tomorrow
Maurizio Zuccotti, Carlo Alberto Redi and Silvia Garagna
Int. J. Dev. Biol. (2012) 56: 761-764
Modeling of angioadaptation: insights for vascular development
Axel R. Pries, Bettina Reglin and Timothy W. Secomb
Int. J. Dev. Biol. (2011) 55: 399-405
Developmental cell biology of human villous trophoblast: current research problems
John D. Aplin
Int. J. Dev. Biol. (2010) 54: 323-329