全 文 ：Genomics Grand for Diversified Plant Secondary Metabolites
Xin FANG1,2, Chang鄄Qing YANG2, Yu鄄Kun WEI1, Qi鄄Xia MA1,
Lei YANG1,2, Xiao鄄Ya CHEN1,2*
(1 Plant Science Research Center, Shanghai Chenshan Botanical Garden, Shanghai 201602, China;
2 Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences,
Chinese Academy of Sciences, Shanghai 200032, China)
Abstract: Plants can generate an overwhelming variety of structurally diversified organic compounds called seconda鄄
ry metabolites. These compounds usually perform interesting biological activities and important functions in influen鄄
cing interactions between plants and other organisms. They are also widely utilized as pharmaceuticals, insecticides,
dyes, flavors and fragrances. Plant genome sequencing, transcriptome and metabolome analyses have provided huge
amounts of data to explain the great diversity of secondary metabolites. This knowledge in turn will help us better un鄄
derstand their ecological role and is a creating novel tool for genetic engineering of plant secondary metabolism.
Key words: Plant secondary metabolites; Biosynthetic pathways; Genomics; Chemical diversity; Bioinformatics
CLC number: Q 946. 8摇 摇 摇 摇 Document Code: A摇 摇 摇 摇 摇 摇 Article ID: 2095-0845(2011)01-053-12
摇 Plants produce structurally diversified seconda鄄
ry metabolites which are distributed differentially
among limited taxonomic groups. Recent estimates
are that a single plant species can produce 5 000 to
25 000 different compounds and that the over
100 000 known chemical structures represent a frac鄄
tion of the total in the plant kingdom (Trethewey,
2004). They have been extensively utilized as dye,
spices, glue, perfume and drugs for centuries by hu鄄
man beings. The early investigations of these organic
compounds were mainly undertaken by organic
chemists who developed separation techniques,
spectroscopic approaches and synthetic methodolo鄄
gies to elucidate their structures, the complex bio鄄
synthetic pathways of most of these compounds, al鄄
though frequently deduced, were unveiled during
that time. There is no doubt that characterizations in
chemistry have helped us to better utilize these sec鄄
ondary metabolites. However, for a long time, most
plant biologists treated them as waste products of pri鄄
mary metabolism and paid little attention to them as
chemical treasures as they do not appear to contrib鄄
ute to plant growth and development directly. Only
recently, plant secondary metabolites were consid鄄
ered to influence ecological interactions between the
plant and its environments in many aspects, such as
inter鄄 and intra鄄plant priming, direct and indirect
defense, allelopathy, attraction, UV鄄B protection,
responses to temperature, drought or nutrient stres鄄
ses (Chen et al., 2009). These ecological functions
are of vital importance for plant survival and have at鄄
tracted ever鄄increasing attention from plant biolo鄄
gists.
The identification and classification of the com鄄
plex and diversified structures of plant secondary
metabolites and the enzymes involved in the corre鄄
sponding pathway are the basis for the development
of organic chemistry and plant ecology. However,
the natural products obtained by traditional chemistry
methods and enzymes by biochemistry methods usu鄄
ally could not comprehensively account for the bio鄄
synthetic capacity of the organism.
Genome sequencing has opened a new era of
life science research. Plant genomics provides both a
植 物 分 类 与 资 源 学 报摇 2011, 33 (1): 53 ~ 64
Plant Diversity and Resources摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 DOI: 10. 3724 / SP. J. 1143. 2011. 10233
* Author for correspondence; E鄄mail: xychen@ sibs. ac. cn
Received date: 2010-12-17, Accepted date: 2010-12-30
powerful tool and a rich source of data to dissect in
order to better understand the origin of the great di鄄
versity of secondary metabolites. A plant genome
may harbor more genes responsible for secondary
metabolites than previously estimated. For example,
of the 120 Mb Arabidopsis genome, 32 were annota鄄
ted to code for terpene synthases, although functions
of most of them remain to be experimentally charac鄄
terized (Aubourg et al., 2002). These genes were
suggested to derive from individual gene tandem du鄄
plication and large鄄scale genome segmental duplica鄄
tion events. Plants could also create chemical diver鄄
sity by using a matrix pathway, mix genes, and tran鄄
scription factors without createing new genes. Ge鄄
nome mining could uncover compounds that have not
been characterized by classical natural product isola鄄
tion methods. We will then discuss the genomics
grand for diversified plant secondary metabolites.
1摇 Major groups of plant secondary me鄄
tabolites
Plant secondary metabolites arrive from limited
precursor molecules, shared with plant primary me鄄
tabolism, and are further transformed into complex
and specific products through enzymatic catalyzed
reactions in the plant. Based on biosynthetic ori鄄
gins, these diversified compounds can be divided
into three major categories: terpenoids, alkaloids, and
phenylpropanoids / allied phenolic compounds. There
are other minor groups which cannot be included in
these three groups, such as amines, cyanogenetic,
glycosides, glucosinolates, acetlenes and psoralens.
1. 1摇 Terpenoids
Terpenoids represent the most abundant and
structurally diverse group of plant secondary metabo鄄
lites, in which more than 36 000 individual struc鄄
tures have been identified. The diversified structures
of terpenoids derive from a sequential assembly of
five鄄carbon building blocks called isoprene units (C5
H8), the so called biogenetic isoprene rule. Ac鄄
cordingly, one, two, three, four, five and six units
construct hemiterpenes, monoterpenes, sesquiterpe鄄
nes, diterpenes, sesterterpenes and triterpenes, re鄄
spectively. The isoprene units are often joined in a
“head鄄to鄄tail冶 and “ head鄄to鄄head冶 fashions and a
few are formed by head鄄to鄄middle fusions. After the
formation of the basic terpenoid skeletons, subse鄄
quent modifications, including oxidation, reduction,
isomerization, conjugation, and even degradation,
lead to various structures of terpenoids. For exam鄄
ple, steroids, a group of physiologically important
natural products including cholesterol, ergosterol,
and androgen, are degraded triterpenes.
The most typical hemiterpene is isoprene, a vol鄄
atile product, which is released from photosyntheti鄄
cally active tissues. The monoterpenes and sesquiter鄄
penes are components of volatiles and are widely used
as flavors and perfumes. The diterpenes are mainly
found in resin and triterpenes mainly occur as sapo鄄
nins. There are more than 30 basic monoterpene skel鄄
etons which could be divided into the acyclic (such
as myrcene and citronellol), monocyclic ( including
the camphene, isocamphane and ionone types), bi鄄
cyclic (including the thujane, carane, pinane, cam鄄
phene, isocamphane, and fenchane types), and the
iridoids (Grayson, 2000). The reported ecological
functions of monoterpenes include attraction, allelop鄄
athy and defense, and some monoterpenes play differ鄄
ent functions in different plants. The basic skeletons
of sesquiterpenes are most abundant in the terpe鄄
noids, of which there were 60 known types in 1995
(Fraga, 1997). According to the structures, ses鄄
quiterpenes also could be grouped into the acyclic
( such as farmesane), monocyclic ( including the
bisabolane, germacrane, humulane, and elemane
types), bicyclic (including the caryophyllane, cadi鄄
nane, eremophilane, valeriane, guaiane, eudes鄄
mane, illudane, acorane, himachanlane, laurane,
chamigrane, and neopinguisene types ), tricyclic
( including the aristolane, maliane, thujopsane,
proto鄄illudane, hirsutane, cedrane, marasmane,
cubebane, longifolane, clovane, and junicedrane
types). Similary, diterpenes also have acyclic and
cyclic skeletons. The acylic diterpenes include such
45摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 33 卷
compounds as geranylgeranilo, the monocyclic diter鄄
penes include cembrene, the bicyclic structures are
represented by the labdane and clerodane types, the
tricyclic by the pimarane and abietane types, and
the tetracyclic by the kauranes, gibberellin,
grayanane and phorbol types. Other important diter鄄
penoids include taxanes and ginkgolides (Hanson,
2004). The ecological functions of sesquiterpenes,
diterpenes and triterpenes are believed to participate
in direct or indirect defense of the plant against her鄄
bivores and microbial pathogens. Triterpenoid are
synthesized from the cyclization of squalene and most
of them have tetracyclic and pentacyclic structures.
The lanostane, dammarane, protostane, cucurbi鄄
tane, sipholane, limonoid and quassinane types are
the most common tetracyclic triterpenoids, whilst the
oleanane, ursane, lupane, hopane, serratabe, mal鄄
abaricane and ferane types are the most abundant
pentacyclic triterpenoids (Connolly and Hill, 2010).
Many terpenoids are of great medicinal vaule, for
example, the sesquiterpene lacone artemisinin and
the diterpene alkaloid taxol are drugs used for the
treatments of malaria and cancer, respectively.
1. 2摇 Alkaloids
Alkaloids are a group of alkaline, have a low
molecular weight, and are nitrogen鄄containing com鄄
pounds with at least one ring with the nitrogen atoms
usually present in rings. They are the most widely
distributed secondary metabolites and are found not
only in plants, but also in microorganism animals,
and play an important role in plant defense systems.
Alkaloid鄄containing plants were, for mankind, the
original “material medica冶 and many are still in use
today as prescription drugs, such as vinblastine,
quinine, atropine, and camptothecin. There are
more than 12 000 alkaloids reported from 100 fami鄄
lies of plants, and they are especially abundant in
families of Leguminosae, Solanaceae, Manisper鄄
maceae, Papaveraceae, Ranurculaceae, and Berbe鄄
ridaceae ( Buchanan et al., 2000). Alkaloids sel鄄
dom co鄄exist with terpenoids and volatile oils in the
same plant and most of them exist in the form of salts
in the plant, except for some with weak basis.
Aalkaloids can be classified on the basis of the
plants from which they were isolated, the chemical
structures, and the biosynthetic origins, and the last
has an obvious advantage of reflecting the relation鄄
ship between biosynthetic pathways and chemical
structures. Alkaloids could thus be further classified
into three groups according to their biosynthesis ori鄄
gin: true alkaloid, protoalkaloid, and pseudoalka鄄
loid ( Dewick, 2002 ). True alkaloids are formed
from L鄄amino acids, such as tryptophan ( Trp ),
tyrosine (Tyr), lysine (Lys), and arginine (Arg)
(Dewick, 2002). For example, pyrrolizidine, pyr鄄
rolizidine and tropand alkaloids are formed from or鄄
nithine; piperidine, indolizidine and quiolizidine al鄄
kaloids (mainly found in Leguminosae) from lysine;
quinoline, quinazoline and acridine alkaloids from
anthranilic acid. Simple isoquinoline, benzyliso鄄
qunoline, bisbenzylisoquinoline, aporphine, berber鄄
ine and protoberberine, protopine, emetine, 琢鄄
naphtha鄄phenanthridine and morphine alkaloids are
generated from Tyr; simple 茁鄄carboline, monoterpe鄄
noid indole and ergot alkaloids are originated from
tryptophan. Protoalkaloids, such as ephedrine,
pseudoephedrine and capsaicinoid, are a group of
aromatic amine originating from phenylalanine
(Phe) but with the nitrogen atom located inside the
chain instead of the ring (Dewick, 2002). The pre鄄
cursors of pseudoalkaloids are not L鄄amino acids,
and the nitrogen atoms are introduced into the struc鄄
ture in a later stage in the biosynthetic pathway, de鄄
spite the location of nitrogen atom in ring (Dewick,
2002). These types of products include terpenoid
alkaloids and steroidal alkaloids.
1. 3摇 Phenylpropanoids
Phenylpropanoids / allied phenolic compounds
contain at least one aromatic ring with one or more
hydroxyl groups, and the majority of which are
formed through shikimate / arogenate pathways alone
or in combination with acetate / melonate pathways.
More than 10 000 plant phenolic structures have
been reported. Phenolics are of great importance as
551 期摇 摇 摇 摇 摇 摇 摇 摇 Xin FANG et al. : Genomics Grand for Diversified Plant Secondary Metabolites摇 摇 摇 摇 摇 摇 摇 摇 摇
cellular support materials, such as lignins present in
the cell wall, for mechanical support and as barriers
against microbial invasion. The most magnificent
function of the phenolic flavonoids, especially the
anthocyanins, together with flavones and flavonols as
co鄄pigments, is their contribution to flower and fruit
colors (Dey and Harborne, 1997).
On the basis of their chemical structures and
biosynthetic pathway, phenylpropanoids / allied phe鄄
nolic compounds could be divided into following
groups: lignins, lignans, coumarins, flavonoids,
tannins, stibenes, styrylpyrones and arylpyrones.
Lignins and lignans, two biosynthetically related
secondary metabolites, are polymer, oligomer or
dimmer of phenylpropanoids, respectively. Most lig鄄
nans are optically active, but all isolated lignins are
not. Cinnamyl alcohol, cinnamic acid, propenyl
phenol and allyl phenol are the phenylpropanoid u鄄
nits consisting of lignans and their coupling modes
are relatively few comparing to the several thousand
known lignans. There are still no methods available
for isolating lignins in their native state that do not
markedly alter the original structure of the biopoly鄄
mers during dissolution, and no methods of com鄄
pletely and effectively breaking down the biopoly鄄
mers for structure analysis. The real structure of lig鄄
nins remains a question, but we do know that gym鄄
nosperm lignins are primarily derived from coniferyl
alcohol, and to a lesser extent, p鄄coumaryl alcohol,
whereas angiosperms contain coniferyl and sinapyl
alcohols in roughly equal proportions.
The structural feature of coumarins is a benzo鄄
pyranone core, and they could be divided into sim鄄
ple coumarins, linear furanocoumarins, angular
furanocoumarins, pyranocoumarins, and pyromesub鄄
stituted coumarins. There are more than 1 500 cou鄄
marins found in more than 800 species, occurring in
the seed coats, fruits, flowers, roots, leaves and
stems, although in general the greatest concentra鄄
tions are found in fruits and flowers.
The structure of flavonoids contains a C6 鄄C3 鄄C6
core which is two benzene ( A ring and B ring)
linked through a three carbons bridge or a pyran ring
(C ring). They are biosynthesized through the con鄄
densation of three molecules of acetate鄄derived malo鄄
nyl鄄CoA and one molecule of p鄄coumaryl鄄CoA. Ac鄄
cording to the oxidation of the C3 chain and the link鄄
age of the B ring, flavonoids could be classified into
such subgroups as chalcones, aurones, flavonones,
isoflavonoids, flavones, flavonols, leucoanthocyani鄄
dins, xanthones, aurones, furanochromones, ho鄄
moisoflavones, phenylchromones, catechins, and
anthocyanins. They could occur as monomers, dim鄄
mers and higher oligomers in most plant tissues, of鄄
ten in vacuoles. They are also found as mixtures of
colored oligomeric / polymeric components in various
heartwoods and barks. The coupling of flavoids
could produce condensed tannins which add a dis鄄
tinct bitterness or astringency to the taste of certain
plant tissues and function as antifeedants. In certain
plant species cinnamoyl鄄CoA and malonyl鄄CoA path鄄
ways could also undergo condensation reactions to
yield the corresponding stilbenes, styrylpyrones, and
arylpyrones. The stilbene combretastatin has impor鄄
tant antineoplastic activities, and resveratrol, pres鄄
ent in red wine, helps suppress tumor formation.
2 摇 Biosynthetic origins of secondary me鄄
tabolism pathways
Secondary metabolites are derived from primary
metabolism pathways. In many cases, primary and
secondary metabolites cannot readily be distinguish鄄
ed on the basis of precursor molecules, chemical
structures, or biosynthetic origins. As we have men鄄
tioned above, the investigation of the biosynthesis of
secondary metabolism was mainly carried out by or鄄
ganic chemists in the early twenty century. The “ i鄄
soprene rule冶 proposed by Otto Wallach, “biogenet鄄
ic isoprene rule冶 by Leopold Ruzicka, and the alka鄄
loid biosynthesis pathway suggested by Sir Robert
Robinson, Clemens Sch觟pf, Ernst Winterstein and
Georg Trier were a few of the most famous examples.
From 1950忆s to 1970忆s, the establishment of precur鄄
sor鄄feeding experiments and suspension cultures of
65摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 33 卷
plant cells allowed scientists to investigate the bio鄄
synthesis of secondary metabolites at the enzyme lev鄄
el, although this was typically limited to a small
group of proteins that were abundant, stable, and
soluble and for which substrates were commercially
available. Since the 1980s, the ability to isolate mu鄄
tant plants defective in multiple steps of biochemical
pathways, along with the expression of plant proteins
in microbes to assay enzyme activities in vitro or to
complement existing mutations in the microorgan鄄
ism, led to hundreds of plant enzymes being identi鄄
fied and cloned. Once full鄄length cDNAs became a鄄
vailable, the problem of obtaining enough protein for
detailed enzymatic analysis could be circumvented
by expression and purification of target proteins pro鄄
duced in microbes and even in animal cells. By u鄄
sing these methods, the biosynthesis pathways of
some terpenoids, lignans, lignins and flavonoids have
been characterized. As for alkaloids, at least eight
pathways have been elucidated for the enzyme and
gene level, including ajmaline, vindoline, berberine,
corydaline, macarpine, morphine, berbamunine, and
scopolamine (Ziegler and Facchini, 2008). The sec鄄
ondary metabolites are often stored and even formed
in special structures of plants. The hydrophilic me鄄
tabolites are often stored in vacuole, laticifers, and
apoplast or cell walls, whilst the lipophilic compunds
in cuticles, trichomes, resin ducts, laticifers, oil
cells, and plastid membranes.
2. 1摇 Biosynthesis of terpenoids
The investigation of the biosynthesis of terpe鄄
noids revealed that the real precursors of terpenoids
are not isoprene, but isopentenyl diphosphate (IPP)
and its allylic isomer dimethylallyl diphosphate
(DMAPP), and the geranyl diphosphate ( GPP),
farmesyl diphosphate ( FDP ) and geranylgeranyl
diphosphate ( GGPP) produced by repetitive addi鄄
tions of IPP. The elaboration of these allylic prenyl
diphosphates by specific terpenoid syntheses yield
terpenoid skeletons, and then the secondary enzymat鄄
ic modifications to the skeletons give rise to the func鄄
tional properties and great chemical diversity of this
family of natural products. The processes from IPP to
terpenoid skeletons have been extensively studied but
knowledge about modification steps are relatively
poor. The processes of biosynthesis of terpenoids are
regulated by two broad factors: the spatial and the
temporal.
The formation of IPP in the plant involves two
independent pathways located in separate subcellular
compartments. In cytosol, IPP is derived from the
long鄄known mevalonic acid ( MVA) pathway that
starts with the condensation of acetyl鄄CoA (Newman
and Chappell, 1999). In plastids, IPP is formed
from pyruvate and glyceraldehydes 3鄄phosphate,
which is called MEP pathway, named after the key
intermediate methylerythritol phosphate (Lichtentha鄄
ler, 1999). The cytosolic IPP may serve as a pre鄄
cursor of FPP for sesquiterpenes and triterpenes,
whilst the plastidial IPP provides the precursors for
GPP and GGPP for mono鄄, di鄄, and tetra鄄terpenes.
However, cross鄄talk between these two IPP genera鄄
tion pathways is prevalent (Dudareva et al., 2005),
particularly in going from direction from plastids to
cytosol. Then, the prenyltransferase enzymes gener鄄
ate GPP, FPP and GGPP from IPP in head鄄to鄄tail
condensation reactions. Squalene, the direct precur鄄
sor of triterpene, is formed by a head鄄to鄄head con鄄
densation reaction of two molecules of FPP, cata鄄
lyzed by squalene synthase. And phytoene, the pre鄄
cursor of tetraterpenes, is formed by two molecules
of GGPP in a manner analogous to that of squalene,
which is catalyzed by phytoene synthase. These en鄄
zymes function at the branch point of terpenoid me鄄
tabolism, thus playing a regulatory role in controlling
IPP flux into different families of terpenoids.
The allylic prenyldiphosphaes of GPP, FPP and
GGPP are used by terpene synthases (TPSs) to form
mono鄄, sequi鄄, and diterpenes, respectively. After
the construction of the basic parent skeletons pro鄄
duced by the TPSs, subsequent modifications inclu鄄
ding oxidation, reduction, isomerization, and conju鄄
gation reactions impact functional properties to the
terpenoid molecules. Among these modifications,
751 期摇 摇 摇 摇 摇 摇 摇 摇 Xin FANG et al. : Genomics Grand for Diversified Plant Secondary Metabolites摇 摇 摇 摇 摇 摇 摇 摇 摇
the hydroxylations or epoxidations involved in intro鄄
ducing oxygen atoms into the terpenoid skeletons
have been extensively investigated. These reactions
are performed by cytochrome P450 enzyme systems,
in which the P450 monooxygenase plays a key role.
Several important terpenoids biosynthesis pathways
have been thoroughly studied. For example, Artemis鄄
ia annus contains abundant terpenes, of which arte鄄
misinin is a sesquiterpene lactone with excellent anti鄄
malarial activity. Synthesis of amorpha鄄4, 11鄄diene
from IPP is the first step in artemisinin biosynthesis.
The amorpha鄄4, 11鄄diene synthase was cloned, ex鄄
pressed from Artemisia annua (Bouwmeester et al.,
1999; Wallaart et al., 2001; Picaud et al., 2005).
Eight cDNAs encoding terpenes synthase have been i鄄
solated from Artemisia annus, in which four enzymes
are characterized, including ( -)鄄茁鄄pinene synthase
and b鄄caryophyllene synthase (Jia et al., 1999; Cai
et al., 2002; Lu et al., 2002).
Gossypol, a Malvaceae specific sesquiterpene
aldehyde, is synthesized by the condensation of two
moleculars of hemigossypol, which is the product of
the modification of sesquiterpene ( +)鄄啄鄄cadinene.
One P450 monooxygenase, CYP706B1, was identi鄄
fied to be ( +)鄄啄鄄cadinene鄄8鄄hydroxylase, and the
product 8鄄hydroxyl鄄( +)鄄啄鄄cadinene is then conver鄄
ted to gossypol derivates (Luo et al., 2001).
2. 2摇 Biosynthesis of Phenylpropanoids
The biosynthesis pathways of lignans, lignins
and flavonoids are perhaps the most studied of plant
secondary metabolisms. The direct precursors of lig鄄
nans and lignins are monolignols derived out of Phe
and Tyr; whilst the precursor molecules of flavonoids
are Phe, Tyr and malonly CoA. Phenylalanine am鄄
monia鄄lyase (PAL) catalyzes the conversion of Phe
to cinnamic acid (Koukol and Conn, 1961) and Tyr
to p鄄coumaric acid (Neish, 1961), the first step in
the phenylpropanoid pathway. In dicot Phe is the
highly preferred substrate, but in monocot both Phe
and Tyr could be utilized. In some plants, PAL ap鄄
pears to be encoded by a single gene, whereas in
others it is the product of a multigene family. The
ammonium ion liberated by the PAL reaction is recy鄄
cled by way of glutamine synthetase and glutamate
synthetase. Cinnamate鄄4鄄hydroxylase (C4H), func鄄
tioning in aromatic ring hydroxylation, is an oxygen鄄
requiring, NADPH鄄dependent, cytochrome P450
enzyme that catalyzes the regiospecific hydroxylation
at the para鄄positon of cinnamic acid to give p鄄cou鄄
maric acid (Russell and Conn, 1967). O鄄Methyl鄄
transferases, catalyzing the transformation of a meth鄄
yl group into the meta鄄position, uses S鄄adenosylmethi鄄
onine (SAM) as a donor (Finkle and Nelson, 1963),
whereas CoA ligation requires ATP and CoASH.
This two鄄step ligation first generates the AMP deriva鄄
tive, and then converts it into the corresponding CoA
ester, and two sequential NADPH鄄dependent reduc鄄
tions produce the monolignols. The monolignols are
primarily converted into lignans and lignins, the first
of which requires a dirigent protein to orient the pu鄄
tative free radical substrates in such a way that ran鄄
dom coupling cannot occur (Davin et al., 1997).
However, confusion remains on whether the biosyn鄄
thesis of lignins requires enzymes. We also do not
yet fully understand the metabolic flux and compart鄄
mentalization of the phenylpropanoid pathway.
The flavonoid pathway is branched off the phe鄄
nylpropanoid pathway from p鄄coumaryl鄄CoA to con鄄
dense three molecules of acetate鄄derived malonly鄄
CoA to generate a 6鄄deoxychalcone, which is cata鄄
lyzed by chalcone synthase (CHS). CHS is a di鄄
meric polyketide synthase with each subunit at about
42 kDa. Then, chalcone isomerase (CHI) catalyzes
a stereospecific ring closure isomerization step to
form some flavanones, which is shared by most of
the flavonoid biosynthesis pathways. The isomeriza鄄
tion of the flavanones leads to the isoflavonoid
branch point catalyzed by two enzymes. Isoflavone
synthase, an NADPH鄄dependent cytochrome P450
enzyme, catalyzes the first step of a 1, 2 aryl migra鄄
tion and hydroxylation to give the 2鄄hydroxyisofla鄄
vanones. Dehydration of the 2鄄hydroxyisoflavanone de鄄
hydratase (IFD), forms the isoflavonoids. In general
flavonoid metabolism, the second branching point in鄄
85摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 33 卷
volves dehydration of naringenin at the C鄄2 / C鄄3 posi鄄
tions to give abundant flavones. This conversion is cat鄄
alyzed by flavone synthase (FNS), which varies in en鄄
zymatic type depending on the plant species. The third
branch point is stereospecific 3鄄hydroxylation of narin鄄
genin to give dihydroflavonols, which is catalyzed by
flavanone 3鄄hydroxylase, a Fe2+鄄requiring, 琢鄄ketoglut鄄
arate鄄dependent dioxygenase. The subsequent species鄄
and tissue鄄specific enzymatic conversions could create
vast arrays of structurally diverse groups of flavonoids.
Condensed tannins are formed from flavonoids, how鄄
ever, the enzymology associated with those coupling
processes, chain extension mechanisms, and oxida鄄
tive modifications is not yet established.
3摇 Plant genomes for secondary metabolism
Early efforts at natural product isolation and en鄄
zyme discovery have their limits. The compounds
that can be obtained by traditional isolation method,
even those of trace components, are consequently
not a comprehensive accounting of the organism忆s bi鄄
osynthetic capacity, but a reflection of the state of
the tissues upon harvest. As natural product biosyn鄄
thesis is heavily influenced by external stimuli such
as microbial infection or herbivory, the extracts of
un鄄induced plants consequently contain only a subset
of the products that these organisms can biosynthe鄄
size. The enzymes obtained by the precursor鄄feeding,
mutant screening and cDNA expression experiments
are also not thoroughly reflecting the organism忆s bio鄄
synthetic capacity, since the enzymes involved in
secondary metabolim are often tissue鄄specific and reg鄄
ulated under environmental influences. A complete
understanding of plant secondary metabolism and
their biosynthesis pathway thus requires new and
comprehensive methods to reveal their molecular basis
and to overcome the limits mentioned above.
Publication of the first two bacterial genomes in
1995 marked the beginning of the genomic era (Del鄄
laPenna and Last, 2008). It has been proved that
the genomics and bioinformatics could serve as pre鄄
dictors of new molecules and enzymes. The se鄄
quenced streptomycetes of S. avermitilis (魶mura et
al., 2001) and S. griseus ( Ohishi et al., 2008)
could make 2-3 natural products but harbor 25-30
predicted biosynthetic gene clusters. To date, we
are missing 90% of the natural product biosynthetic
capacity of even the workhorse producers. If even
20% of the 20-25 cryptic molecules were novel, the
current knowledge base from streptomyctes would
double (Walsh and Fischabch, 2010). Thus bacte鄄
rial genomics and bioinformatics have become as im鄄
portant as chemistry in categorizing known natural
products and identifying likely unknown variants to
be discovered.
Since the genes for plant natural product path鄄
ways are rarely physically clustered as bacterial
genes do, efforts to decipher plant natural product
pathways have lagged behind those to bacterial path鄄
ways. However, bioinformatics is a useful tool to i鄄
dentify the metabolic functions of unknown plant
genes once the completed genomes of the plant are
obtained. The development of such tools based on
plant genomics and bioinformatics as protein family鄄
based analysis, contextual genomics approaches,
cell鄄specific comparisons, and co鄄expression analy鄄
sis allows genome鄄based research in plant metabo鄄
lism to be more feasible ( DellaPenna and Last,
2008). Thalianol is a novel triterpene first uncov鄄
ered by heterologous expression of enzymes exploited
from Arabidopsis thaliana genomic information and
then detected in A. thaliana at a low level, which is
a good indicator that genome mining can uncover sec鄄
ondary metabolites that eluded classical methodologies
(Fazio et al., 2004). It is estimated that in Arabidopsis
thaliana, about 5 000 genes (about a quarter of all)
are involved in secondary metabolism (Gierl and Frey,
2001). Therefore, the current public availability of
draft or completed genomes for rapidly increasing num鄄
bers of organisms of different taxonomic groups creates
unprecedented opportunities to study individual plant
enzymes, pathways, and metabolic networks.
Investigations using genomics and bioinformat鄄
ics methods have produced extensive knowledge and
951 期摇 摇 摇 摇 摇 摇 摇 摇 Xin FANG et al. : Genomics Grand for Diversified Plant Secondary Metabolites摇 摇 摇 摇 摇 摇 摇 摇 摇
interpretation of the chemical diversity in plants. As
discussed above, plants use limited building blocks
to construct structurally diversified and complex sec鄄
ondary metabolites. Not surprisingly, the biosynthet鄄
ic pathways of these secondary metabolites are ac鄄
cordingly complex ( Fischbach et al., 2008). The
set of proteins that comprise a complete biosynthetic
pathway can be twice the size of the ribosome, even
though the ribosome translates thousands of different
proteins, whereas the biosynthetic pathway produces
a few small molecules. Furthermore, some enzymes
involved in natural product biosynthesis have a broad
substrate tolerance, which is now firmly supported
by experimental evidence gained for all major natural
product pathways (Firn and Jones, 2003). A good
example of this tolerance is the multifunctional en鄄
zyme of Arabidopis LUP1 ( At1G78970) that con鄄
verts oxidosqualene to mixtures of at least 6 -7 dis鄄
tinct triterpene alcohols. Such enzymes capable of
acting on more than one substrate would be expected
to facilitate branching pathways, and at the extreme,
to participate in a matrix grid, which creates more
chemical diversity with limited enzymes. In petunia
flowers, for example, three enzymes (F3H, F3忆5忆H
and F3忆H) can produce five different products (eri鄄
odictyol, pentahydroxyflanone, dihydromyricetin,
dihydroquercetic anddihydrokaempferol) from narin鄄
genin (Holton and Cornish, 1989).
Also, genes functioning in secondary metabo鄄
lism are generally more divergent than those coding
for proteins involved in primary metabolism. For ex鄄
ample, in Arabidopsis thaliana, a family of 40 terpe鄄
noid synthase genes (AtTPS) was discovered by ge鄄
nome sequence analysis. Among them, thirty鄄two
AtTPS genes are attributed as putative monoterpene
synthases, sesquiterpene synthases or diterpene syn鄄
thases of secondary metabolism. In contrast, only
two AtTPS genes have known functions in hormone
metabolism, namely gibberellin biosynthesis ( Au鄄
bourg et al., 2002). This striking difference in rates
of gene diversification in primary ( hormone) and
secondary metabolisms is relevant for an understand鄄
ing of the evolution of natural product diversity. A鄄
lignment of the amino acid sequences of plant ter鄄
pene syntheses divided the TPS family into seven
subfamilies, designated TPSa to TPSg. Each sub鄄
family has a minimum of 40% sequence identity a鄄
mong members and has, in general, similar func鄄
tions. However, specific product profiles of members
of the same subfamily can be quite diverse and can鄄
not be predicted based on sequence alone. There are
two strategies used in terpenoids synthesis to create
chemical diversity. One is that different terpene syn鄄
thases use the same substrate to produce different
products, and the other is one terpene synthase pro鄄
duces multiple products.
Plants could also mix genes to create chemical
diversity by juxtaposing distinct but chemically com鄄
patible biosynthetic systems. Tailored natural prod鄄
ucts are the most important results of this strategy
(Walsh and Fischabch, 2010), which are produced
by functionalizing the core scaffold, often occurring
later in the pathway. Tailoring enzyme chemistries
can be grouped into two broad categories: group
transfer reactions and oxidative transformations.
Nearly all group transfers involve coupling an elec鄄
trophilic fragment of a cosubstrate or primary metab鄄
olite to a nucleophilic N, O, or S in the natural
product skeleton. These cosubstrates include NDP鄄
sugars (such as UDP鄄Glucose) as glycosyl donors,
S鄄adenosylmethionine as methyl donors, acyl鄄CoA as
an acyl donor and the correponding enzymes are gly鄄
cosyltransferases, methyltransferases, and acyltrans鄄
ferases, respectively. The so鄄called BAHD super鄄
family enzymes are a large group of plant鄄specific
acyl鄄CoA dependent O鄄or N鄄acyltransferases identi鄄
fied recently. Most of them fall into two functional
families: alcohol acetyltransferases responsible for
forming aroma / flavor volatile acetate esters, such as
geranyl acetate, phenylethyl acetate, and benzyl ac鄄
etate (Dudareva and Pichersky, 2000) and, antho鄄
cyanin / flavonoid acyltransferases, primarily malo鄄
nyltranserases and hydroxycinnamoyltransferases, re鄄
sponsible for modifying polyphenolics ( Yu et al.,
06摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 33 卷
2008 ). Several BAHD members synthesize and
modify a variety of other metabolites, such as shiki鄄
mate鄄phenylpropanoid鄄derived phytoalexins, alka鄄
loids, terpenoids, and polyamines ( Luo et al.,
2009). The sequences of BAHD genes are highly
divergent, showing only 10-30% similarity at the a鄄
mino鄄acid level, consistent with their functional di鄄
versity (D忆Auria, 2006). Using the conserved se鄄
quence motifs HXXXD and DFGWG of BAHD mem鄄
bers, 94 and 61 putative BAHD genes were identi鄄
fied from the genome sequences of Populus and Ara鄄
bidopsis, respectively (Yu et al., 2009). The ratio
of the number of BAHD genes in Populus to that of
Arabidopsis is consistent with the estimation of the
entire genome. Gene chromosomal distribution dem鄄
onstrates that both individual gene tandem duplica鄄
tion and large鄄scale genome segmental duplication e鄄
vents appear to have exclusively contributed to the
current complexity of the BAHD gene superfamilies
in both Populus and Arabidopsis. Having a greater
number of diverse BAHD enzymes meets the needs
of the biosynthesis and modification of a repertoire of
the secondary metabolites in the plants.
Oxidative transformations may plan an important
role in creating chemical diversity in plants. Terpe鄄
noid pathways generate product with dramatic struc鄄
tural diversities from simple building blocks by first
generating a reduced, unreactive polycyclic interme鄄
diate and then tailoring it with oxygen鄄based func鄄
tionality ( Walsh and Fischabch, 2010 ). Unlike
group transfer reactions that simply introduce a
structural fragment in the core skeleton, oxidative
transformations can also generate a novel skeleton
through structural rearrangement as the oxygen鄄based
functionality serves as a new reaction centre. In the
flavonoid, lignan, and lignin pathway, oxidative
transformations also catalyze aromatic and aliphatic
hydroxylations, and skeleton formation (Ayabe and
Akashi, 2006). In plants, heme iron鄄containing en鄄
zymes of the P450 monooxygenase superfamily are
widespread oxidative tailoring enzymes of natural
products, adding thousands of genes falling into 126
families and 464 subfamilies. Plant genome encodes
more P450 enzymes than other organisms. For exam鄄
ple, there are 246 genes in Arabidopsis thaliana,
356 in rice, 312 in poplar, and 457 in grape, and
the number of P450 genes is estimated at up to 1%
of total gene annotations of each plant species (Mi鄄
zutani and Ohta, 2010); whereas there are 57 P450
genes in Homo sapiens, 105 in Mus musculus (Nel鄄
son et al., 2004), 86 in Drosophila melanogaster
(Chung et al., 2008), and 80 in Caenorhabditis el鄄
egans (Menzel et al., 2001). This observation is in
accordance with the diversified reactions P450 en鄄
zymes catalyzed in plant secondary metabolism path鄄
way such as hydroxylation, epoxidation, dealkyla鄄
tion, isomerization, dehydration, carbon鄄carbon
cleavage, decarboxylation, nitrogen and sulfur oxi鄄
dation, dehalogenation, and deamination. P450 di鄄
versification during evolution was one of the primary
driving forces of phytochemical diversity.
Interestingly, P450 enzymes can form gene
cluster for biosynthesis of secondary metabolites. For
instance, a triterpene gene cluster ( Field and Os鄄
bourn, 2008), three diterpene gene clusters (Shimu鄄
ra et al., 2007; Swaminathan et al., 2009), and a
benzoxazinoid gene cluster (Gierl and Frey, 2001)
were found in Arabidopsis, rice and maize, respec鄄
tively (Table 1). Clustering in plants facilitates the
inheritance of beneficial combinations of genes, and
avoids the accumulation of toxic intermediates.
The production of secondary metabolites by
plants is influenced by developmental, environmen鄄
tal, pathogen and symbiont signals, which leave the
investigation of secondary metabolites pathway even
more complicated. It is now generally believed that,
in plants, transcription factors play a major role in
the regulation of secondary metabolism pathways, as
many other aspects of plant growth and develop鄄
ment. The phenylpropanoid pathway has been the
leading model for studies on plant gene regulation
but little is known about any of the other major meta鄄
bolic pathways (Davies and Schwinn, 2003). Some
transcription factors of flavonoid pathways have been
161 期摇 摇 摇 摇 摇 摇 摇 摇 Xin FANG et al. : Genomics Grand for Diversified Plant Secondary Metabolites摇 摇 摇 摇 摇 摇 摇 摇 摇
cloned, such as C1, PAR1, ANTHOCYANIN2
(MYB); R, ANTHOCYANIN1, TT8 (Bhlh), etc.
Interestingly, some of the anthocyanin regulatory
genes also impact other processes. Rosea1 of Antir鄄
rhinum and An1, An2 and An11 of Petunia also reg鄄
ulate vacuolar pH, and additionally, An1 influences
seed coat epidermal cell development ( Davies and
Schwinn, 2003). The best model for studies of such
overlapping pathways is Arabidopsis, in which the
regulatory pathways for anthocyanin and proanthocy鄄
anidin production share components with those con鄄
trolling developmental processes such as trichome
formation, root epidermal cell development and seed
mucilage production (Davies and Schwinn, 2003).
Transcription factors can be the main target of the
secondary metabolic engineering. As the activity of
the biosynthetic genes of secondary metabolism ap鄄
pears to be determined primarily by the expression
patterns of the regulatory genes, altering the patterns
of regulatory gene expression may allow the temporal
and spatial modification of secondary metabolite pro鄄
duction. For example, in apple MYB10 is an antho鄄
cyanin鄄regulating transcription factor; an allelic re鄄
arrangement in the gene promoter of MYB10 has gene鄄
Table 1摇 Example of plant P450 clusters identified functionally
P450s
name
Other genes
name Products Species
At5g48000
At5g47990
At5g48010
At5g47980
Desaturated thalina鄄diol
( triterpenoid) Arabidopsis
CYP99A2
CYP99A3
AK103462
OsCycl
Momilactone A
(diterpenoid) Rice
OsKS4
Cyp71Z6
Cyp71Z7
Os鄄CPS2
Os鄄KSL7
Phytocassanes A鄄E
(diterpenoid) Rice
Cyp76M5 Os鄄KSL5
Cyp76M6 Os鄄KSL6
Cyp76M7
Cyp76M8
CYP99A2
CYP99A3
Os鄄CPS4
Os鄄KSL4
Momilactone A&B
(diterpenoid) Rice
Os鄄MAS
BX2
BX3
DIBOA
(benzoxazinone) Maize
BX4
BX5
rated an autoregulatory locus, which is sufficient to
account for the increase in MYB10 transcript levels
and subsequent ectopic accumulation of anthocyanins
in the plant, leading to a striking phenotype that in鄄
cludes red foliage and red fruit flesh (Espley et al.,
2009 ). Another example is the GaWRKY1, a
WRKY transcription factor in cotton that regulates
gossypol content and expression of biosynthesis path鄄
way gene ( +)鄄啄鄄cadinene synthase, by binding to
the promoter and activates its spacial and temporal
expression. (Xu, 2004) Therefore, the identifica鄄
tion of defined transcription factor genes provides
tools for modulating both the amount and distribution
of secondary metabolites in plants.
In addition to resolving the puzzle of how struc鄄
turally diversified metabolites are synthesized in
plants, the genomics and bioinformatics methods can
explain how and why these compounds occurred and
evolved and how new compounds are produced. It
seems most reasonable to assume that the precursors
and pathways for the generation of natural products
arose from mutations of enzymes involved in the syn鄄
thesis of primary metabolites. The large鄄scale ge鄄
nome segmental duplication and individual gene tan鄄
dem duplication of primary pathway genes allowed
the original enzymatic function retained in the plant,
while new functions evolve in the enzyme encoded by
the duplicate genes under the pressure of natural se鄄
lection, which can generate secondary metabolites.
This may explain why primary and secondary metab鄄
olites cannot readily be distinguished on the basis of
precursor molecules, chemical structures, or biosyn鄄
thetic origins. Similarly, duplication of secondary
pathway genes followed by divergence can form new
metabolites once a secondary pathway has been
branched from the primary pathway and this appears
to have been the most common means for diversified
secondary metabolites evolution in plants ( Gang,
2005). New metabolites may also arise due to loss
of enzymatic activity, which can occur due to loss of
gene expression. The intermediate molecules in the
affected pathway may build up to levels that were not
26摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 33 卷
previously present, enabling other enzymes to act on
these metabolites. Alternatively, changes in regula鄄
tory gene expression may not necessarily lead to
complete loss of enzyme activity from the plant, but
may cause the enzyme to be produced in a different
cell, tissue, or organ type (Gang, 2005). As a re鄄
sult, the enzyme can work on a different substrate to
produce new metabolites. If these products are fa鄄
vored by natural selection, then new chemical diver鄄
sity is generated.
4摇 Perspectives
There are more than 400 families, more than
10 000 genera and nearly 300 000 species of angio鄄
sperms on earth. Most of the secondary metabolic
pathways are, although to a different extent, taxa鄄
specific. The great richness of plant species is a huge
treasure for plant secondary metabolites, leaving the e鄄
normous biosynthetic potential of plant cells to be ex鄄
ploited. Recent advances in metabolomics, and other
areas of study, have made it possible to reveal the dy鄄
namic changes of many types of secondary metabolites
in a plant sample. The updated knowledge in metabo鄄
lomics has helped genetic engineering in introducing
whole pathways to produce medicinally valuable prod鄄
ucts in organisms that lack them. The genetic engi鄄
neering artemisin biosynthesis pathway in yeast and
resulting in the accumulation of artemisinic acid is a
pioneering success of synthetic biology. Although the
development of plant genomics and functional ge鄄
nomics has gained much progress, the genetic maps
of biosynthetic pathways are still far from complete,
and the networks regulating different pathways glob鄄
ally are still poorly understood. As plant metabolites
are of great interests and importance to human
health, we expect to see more efforts into research
and engineering of plant secondary metabolism.
References:
Aubourg S, Lecharny A, Bohlmann J, 2002. Genomic analysis of the
terpenoid synthase (AtTPS) gene family of Arabidopsis thaliana
[J] . Molecular Genetics & Genomics, 267: 730—745
Ayabe SI, Akashi T, 2006. Cytochrome P450s in flavonoid metabo鄄
lism [J] . Phytochemistry Reviews , 5: 271—282
Bouwmeester HJ, Wallaart TE, Janssen MHA et al., 1999. Amorpha鄄
4, 11鄄diene synthase catalyses the first probable step in artemis鄄
inin biosynthesis [J] . Phytochemistry, 52: 843—854
Buchanan BB, Gruissem W, Jones RL, 2000. Biochemistry and Mo鄄
lecular Biology of Plants [M]. Rockville: American Society of
Plant Physiologists
Cai Y, Jia JW, Crock J et al., 2002. A cDNA clone for 茁鄄caryo鄄
phyllene synthase from Artemisia annua [ J] . Phytochemistry,
61: 523—529
Chen F, Liu CJ, Tschaplinske TJ et al., 2009. Genomics of secondary
metabolism in Populus: interactions with biotic and abiotic envi鄄
ronments [J]. Critical Reviews in Plant Sciences, 28: 375—392
Chung H, Sztal T, Pasricha S et al., 2008. Characterization of Drosophila
melanogaster cytochrome P450 genes [J]. Proceedings of the Na鄄
tional Academy of Sciences of the United States of America, 106:
5731—5736
Connolly JD, Hill RA, 2010. Triterpenoids [ J] . Natural Product
Reports, 27: 79—132
D忆Auria JC, 2006. Acyltransferases in plants: a good time to be
BAHD [J] . Current Opinion in Plant Biology, 9: 331—340
Davies KM, Schwinn KE, 2003. Transcriptional regulation of second鄄
ary metabolism [J] . Functional Plant Biology, 30: 913—925.
Davin LB, Wang HB, Crowell AL et al., 1997. Stereoselective bimo鄄
lecular phenoxy tadical coupling by an auxiliary ( dirigent) pro鄄
tein without an active center [J] . Science, 275: 362—366
DellaPenna D, Last RL, 2008. Genome鄄enabled approaches shed
new light on plant metabolism [J] . Science, 320: 479—481
Dewick PM, 2002. Medicinal Natural Products [M]. Baffins lane:
John Wiley & Sons Ltd
Dey PM, Harborne JB, 1997. Plant Biochemistry [M]. San Diego:
Academic Press Ltd
Dudareva N, Andersson S, Orlova I et al., 2005. The nonmevalonate
pathway supports both monterpene and sesquiterpene formation in
snapdragon flowers [J] . Proceedings of the National Academy of
Sciences of the United States of America, 102: 933—938
Dudareva N, Pichersky E, 2000. Biochemical and molecular genetic
aspects of floral scents [J] . Plant Physiolosy, 122: 627—633
Espley RV, Brendolise C, Chagn伢 D et al., 2009. Multiple repeats of
a promoter segment causes transcription factor autoregulation in
red apples [J] . The Plant Cell, 21: 168—183
Fazio GC, Xu R, Matsuda SPT, 2004. Genome mining to identify
new plant triterpenoids [ J] . Journal of the American Chemical
Society, 126: 5678—5679
Field B, Osbourn AE, 2008. Metabolic diversification鄄independent
assembly of operon鄄like gene clusters in different plants [J] . Sci鄄
ence, 320: 543—547
Fischbach MA, Walsh CT, Clardy J, 2008. The evolution of gene
collectives: how natural selection drives chemical innovation
361 期摇 摇 摇 摇 摇 摇 摇 摇 Xin FANG et al. : Genomics Grand for Diversified Plant Secondary Metabolites摇 摇 摇 摇 摇 摇 摇 摇 摇
[J] . Proceedings of the National Academy of Sciences of the Unit鄄
ed States of America, 105: 4601—4608
Firn RD, Jones CG, 2003. Natural products鄄a simple model to ex鄄
plain chemical diversity [ J ] . Natural Product Reports, 20:
382—391
Finkle BJ, Nelson RF, 1963. Enzyme reactions with phenolic com鄄
pounds: a meta鄄O鄄methyltransferase in plants [J] . Biochimica et
Biophysica Acta, 78: 747—749
Fraga BM, 1997. Natural sesquiterpenoids [J] . Natural Product Re鄄
ports, 14: 145—162
Gang DR, 2005. Evolution of flavors and scents [J] . Annual Review
of Plant Biology, 56: 301—325
GierlA, Frey M, 2001. Evolution of benzoxazinone biosynthesis and
indole production in maize [J] . Planta, 213: 493—498
Grayson DH, 2000. Monoterpenoids [ J] . Natural Product Reports,
17: 385—419
Hanson JR, 2004. Diterpenoids [ J] . Natural Product Reports, 21:
785—793
Holton TA, Cornish EC, 1989. Genetics and biochemistry of anthocy鄄
anin biosynthesis [J] . The Plant Cell, 7: 1071—1083
Jia JW, Crock J, Lu S et al., 1999. (3R)鄄Linalool synthase from Artemi鄄
sia annua L. : cDNA isolation, characterization, and wound induction
[J]. Archives of Biochemistry and Biophysics, 372: 143—149
Koukol J, Conn EE, 1961. The metabolism of aromatic compounds in
higher plant: purification and properties of the phenylalanine de鄄
aminase of Hordeum vulgare [ J] . The Journal of Biological
Chemistry, 236: 2692—2698
Lichtenthaler HK, 1999. The 1鄄deoxy鄄D鄄xylulose鄄5鄄phosphate path鄄
way of isoprenoid biosynthesis in plants [ J] . Annual Review of
Plant Physiology and Plant Molecular Biology, 50: 47—65
Lu S , Xu R, Jia JW et al., 2002. Cloning and functional character鄄
ization of a 茁鄄pinene synthase from Artemisia annua that shows a
circadian pattern of expression [ J] . Plant Physiolosy, 130:
477—486
Luo J, Fuell C, Parr A et al., 2009. A novel polyamine acyltrans鄄
ferase responsible for the accumulation of spermidine conjugates
in Arabidopsis seed [J] . The Plant Cell, 21: 318—333
Luo P, Wang YH, Wang GD et al., 2001. Molecular cloning and
functional identification of (+)鄄d-cadinene鄄8鄄hydroxylase, a cy鄄
tochrome P450 monooxygenase (CYP706B1) of cotton sesquiter鄄
pene biosynthesis [J] . The Plant Journal, 28: 95—104
Menzel R, Bogaert T, Achazi R, 2001. Asystematic gene expression
screen of Caenorhabditis elegans cytochrome P450 genes reveals
CYP35 as strongly Xenobiotic inducible [ J] . Archives of Bio鄄
chemistry and Biophysics, 395: 158—168
Mizutani M, Ohta D, 2010. Diversification of P450 genes during land
plant evolution [J]. Annual Review of Plant Biology, 61: 1—25
Neish AC, 1961. Formation of m鄄 and p鄄coumaric acids by enzymatic
deamination of the corresponding isomers of tyrosine [J] . Phyto鄄
chemistry, 1: 1—24
Nelson DR, Zeldin DC, Hoffman SMG et al., 2004. Comparison of
cytochrome P450 (CYP) genes from mouse and human genomes,
including nomenclature recommendations for genes, pseudogenes
and alternativesplice variants [J] . Pharmacogenetics and Genom鄄
ics, 14: 1—18
Newman JD, Chappell J, 1999. Isoprenoid biosynthesis in plants: car鄄
bon partitioning within the cytoplasmic pathway [ J]. Critical
Reviews in Biochemistry and Molecular Biology, 34: 95—106
Ohishi Y, Ishikawa J, Hara H et al., 2008. Genome sequence of the
streptomycin鄄producing microorganism Streptomyces griseus IFO
13350 [J] . The Journal of Bacteriology, 190: 4050—4060
魶mura S, Ikeda H, Ishikawa J et al., 2001. Genome sequence of an
industrial microoganism Streptomyces avermitilis deducing the a鄄
bility of producing secondary metabolites [ J] . Proceedings of
the National Academy of Sciences of the United States of America,
98: 12215—12220
Picaud S, Olofsson L, Brodelius M et al., 2005. Expression, purifi鄄
cation, and characterization of recombinant amorpha鄄4, 11鄄diene
synthase from Artemisia annua L. [ J] . Archives of Biochemistry
and Biophysics, 436: 215—226
Russell DW, Conn EE, 1967. The cinnamic acid 4鄄hydraxylase of
pea seedlings [ J] . Archives of Biochemistry and Biophysics,
122: 256—258
Shimura K, Okada A, Okada K et al., 2007. Identification of a bio鄄
synthetic gene cluster in rice for momilactones [J] . The Journal
of Biological Chemistry, 282: 34013—34018
Swaminathan S, Morrone D, Wang Q et al., 2009. CYP76M7 is an
ent鄄Cassadiene C11琢鄄Hydroxylase defining a second multifunc鄄
tional diterpenoid biosynthetic gene cluster in rice [ J] . The
Plant Cell, 21: 3315—3325
Trethewey RN, 2004. Metabolite profiling as an aid to metabolic engi鄄
neering in plants [ J] . Current Opinion in Plant Biology, 7:
196—201
Wallaart TE, Bouwmeester HJ, Hille J et al., 2001. Amorpha鄄4,
11鄄diene synthase: cloning and functional expression of a key en鄄
zyme in the biosynthetic pathway of the novel antimalarial drug ar鄄
temisinin [J] . Planta, 212: 460—465
Walsh CT, Fischabch MA, 2010. Natural products version 2. 0: con鄄
necting genes to molecules [J] . Journal of the American Chemi鄄
cal Society, 132: 2469—2493
Yu XH, Gou JY, Liu CJ, 2009. BAHD superfamily of acyl鄄CoA de鄄
pendent acyltransferases in Populus and Arabidopsis: bioinforma鄄
tics and gene expression [ J] . Plant Molecular Biology, 70:
421—442
Yu XH, Chen MH, Liu CJ, 2008. Nucleocytoplasmic鄄localized acyl鄄
transferases catalyze the malonylation of 7鄄O鄄glycosidic ( iso)
flavones in Medicago truncatula [ J] . The Plant Journal, 55:
382—396
Ziegler J, Facchini PJ, 2008. Alkaloid biosynthesis: metabolism and
trafficking [J] . Annual Review of Plant Biology, 59: 735—769
46摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 33 卷