Running head: Biosynthesis of tomato fruit triterpenoids

1

Running head: Biosynthesis of tomato fruit triterpenoids

Corresponding Author: Reinhard Jetter

Department of Botany

University of British Columbia

6270 University Boulevard

Vancouver, BC V6T 1Z4

Canada

Phone: (604) 822-2477

Fax: (604) 822-6089

Email: [email protected]

RESEARCH AREA/MONITORING EDITOR:

Biochemical Processes and Macromolecular Structures/Joe Chappell

Plant Physiology Preview. Published on November 8, 2010, as DOI:10.1104/pp.110.162883

Copyright 2010 by the American Society of Plant Biologists

www.plantphysiol.orgon January 21, 2019 - Published by Downloaded from Copyright © 2010 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

2

Two Oxidosqualene Cyclases Responsible for Biosynthesis of Tomato (Solanum

lycopersicum) Fruit Cuticular Triterpenoids 1

Zhonghua Wang, Ortwin Guhling, Ruonan Yao, Fengling Li, Trevor H. Yeats, Jocelyn K.C.

Rose and Reinhard Jetter*

Department of Botany, University of British Columbia, 6270 University Boulevard, Vancouver,

BC, V6T 1Z4, Canada (Z.W., O.G., R.Y., F.L., R.J.);

Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, BC,

Canada, V6T 1Z1, Canada (R.J.)

Department of Plant Biology, 412 Mann Library Building, Cornell University, Ithaca, NY

14853, U.S.A.,

(T.H.Y., J.K.C.R.).



3

Keywords

Amyrin, cuticular wax, fruit surface, pentacyclic triterpenoids, enzymatic cyclization

Footnotes:

1This work was supported by a grant from the Natural Sciences and Engineering Research

Council of Canada

*Corresponding author: Reinhard Jetter

Fax: (604) 822-6089

Email: [email protected]



4

ABSTRACT

The first committed step in triterpenoid biosynthesis is the cyclization of epoxysqualene into

various triterpene alcohol isomers, a reaction catalyzed by oxidosqualene cyclases (OSCs). The

different OSCs have characteristic product specificities, which are mainly due to differences in

the numbers of high-energy intermediates the enzymes can stabilize. The goal of the present

investigation was to clone and characterize OSCs from tomato (Solanum lycopersicum), a

species known to accumulate -amyrin in its fruit cuticular wax, in order to gain insights into the

enzymatic formation of this particular triterpenoid. We used a homology-based approach to

isolate two tomato OSCs, and tested their biochemical properties by heterologous expression in

yeast, as well as overexpression in tomato. One of the enzymes was found to be a product-

specific -amyrin synthase, while the other one was a multifunctional OSC synthesizing 48% -

amyrin and six other products. The product spectra of both OSCs together account for both the

range and the relative amounts of the triterpenoids found in the fruit cuticle. Both enzymes were

expressed exclusively in the epidermis of the tomato fruit, indicating that their major function is

to form the cuticular triterpenoids. The relative expression levels of both OSC genes, determined

by quantitative RT PCR, were consistent with product profiles in fruit and leaves of the tomato

cultivar MicroTom. However, the transcript ratios were only partially consistent with the

differences in amounts of product triterpenoids between the tomato cultivars MicroTom, M82

and Ailsa Craig and, thus, transcriptional control of the two OSCs alone cannot explain the fruit

triterpenoid profiles of the cultivars.



5

INTRODUCTION

Triterpenoids are a very diverse group of natural products with wide distribution and

particularly high chemical diversity in plants. They include compounds such as betulinic acid,

the avenacins and glycyrrhizin, which have important biological functions and medicinal

properties (Reichardt et al., 1984; Papadopoulou et al., 1999; Hayashi et al., 2001). The

biosynthetic pathway towards triterpenoids proceeds by joining six isoprene units together to

form the branched long-chain hydrocarbon squalene (Eschenmoser et al., 1955). In prokaryotes,

squalene is directly cyclized into hopanoid triterpenes, whereas in eukaryotes it is first activated

into 2,3-epoxysqualene and then cyclized (Abe, 2007). The cyclizations are highly regio- and

stereospecific, establishing the final carbon structure of the triterpenoid products.

The overall cyclization reaction comprises 1) an initial protonation step, 2) a polyene

addition cascade forming the up to five carbon cycles, 3) a series of 1,2-shifts of hydride and/or

methyl groups, and 3) a final deprotonation (Fig. 1) (Xu et al., 2004; Phillips et al., 2006). The

entire sequence of steps is catalyzed by single enzymes that are designated as triterpenoid

synthases after their preferred products, or as oxidosqualene cyclases (OSCs) after their common

substrate (Abe et al., 1993). The great diversity of triterpenoid structures, with more than 100

different carbon skeletons, is due to different OSCs and, in particular, to the numbers of

rearrangement steps the different enzymes can catalyze in the third stage of the reaction (Xu et

al., 2004). Approximately 50 OSCs have been cloned from various plant species and have been

characterized, typically using heterologous expression in yeast. Many of the plant OSCs were

found to form predominantly one triterpenoid product, but some were also reported to be



6

multifunctional (Morita et al., 2000; Kushiro et al., 2000; Basyuni et al., 2006). For a few plant

species, more than one OSC has been characterized, so that the suite of OSCs accounted for at

least part of the triterpenoid profile in those species (Morita et al., 1997; Morita et al., 2000;

Hayashi et al., 2000; Hayashi et al., 2001; Iturbe-Ormaetxe et al., 2006; Sawai et al., 2006a;

Sawai et al., 2006b). It is generally assumed that the OSC product specificities determined in

vivo in yeast would accurately reflect the true enzyme activities, however, in planta data to

corroborate the specificities have only rarely been reported (Han et al., 2006), and in vitro results

are missing probably due to problems with handling the lipopilic substrate and the membrane-

associatied enzymes.

A glimpse of the biochemical diversity within OSCs can be seen in the case of

Arabidopsis, where the genome was found to contain 13 OSC genes (Fazio et al., 2004). Six of

the corresponding gene products were characterized as monofunctional enzymes forming

cycloartenol, thalianol, marneral, arabidiol, lanosterol or β-amyrin (Corey et al., 1993; Fazio et

al., 2004; Xiong et al., 2006; Xiang et al., 2006; Suzuki et al., 2006; Shibuya et al., 2009). Six

other Arabidopsis OSC genes were found to encode multifunctional enzymes (Herrera et al.,

1998; Segura et al., 2000; Kushiro et al., 2000; Husselstein-Muller et al., 2001; Ebizuka et al.,

2003; Shibuya et al., 2007; Lodeiro et al., 2007; Kolesnikova et al., 2010), while the function of

one gene remains unknown. Even though the previous studies provide substantial information

about the sequence variability and biochemical specificity within this large gene family, the

information on the cyclization mechanism is still fairly limited. This is mainly due to the fact that

the large majority of OSCs characterized to date form lupeol and -amyrin, the two triterpenoids

most commonly found throughout the plant kingdom. Only rarely have OSCs forming other



7

pentacyclic triterpenoid products been described. That such OSCs with diverse product profiles

exist can be seen from a recent report, in which three Kalanchoe daigremontiana OSCs were

shown to synthesize taraxerol, glutinol and friedelin, i.e. products with carbon skeletons

requiring relatively many rearrangements in the course of the OSC reaction (Wang et al., 2010).

For some of the known triterpenoid products, including -amyrin, no OSC activities have been

described to date. However, information on a broad range of OSCs from various plant species,

and with varying product specificities, would help us understand how the different enzymes can

catalyze specific numbers of rearrangement steps, stabilize the high-energy intermediates

involved, and quench the reaction by deprotonation of a particular carbocation to form specific

end products.

A number of studies have reported that -amyrin accumulates to relatively high

concentrations in the lipid mixture coating the surface of tomato fruit (Vogg et al., 2004; Bauer

et al., 2004b; Saladie et al., 2007; Hovav et al., 2007; Isaacson et al., 2009). Besides, ten other

triterpenoids have also been identified in the fruit cuticular waxes extracted from the surfaces of

various tomato cultivars that have been studied in much detail in recent years (Baker et al., 1982;

Smith et al., 1996; Vogg et al., 2004; Bauer et al., 2004a; Bauer et al., 2004b; Saladie et al.,

2007; Leide et al., 2007; Hovav et al., 2007; Mintz-Oron et al., 2008; Isaacson et al., 2009;

Adato et al., 2009; Kosma et al., 2010). In particular, the surface wax on mature fruit of the

tomato variety MicroTom contains approximately 25% of triterpenoids, and 36% of this fraction

is -amyrin (Vogg et al., 2004). The relative portions of the triterpenoid fraction within the

waxes decreases in the course of fruit development, while the relative amounts of very long

chain aliphatics derived from fatty acid metabolism increases, most importantly unbranched and



8

methyl-branched alkanes (Leide et al., 2007; Mintz-Oron et al., 2008). Thus, the chemical data

indicate that triterpenoids are formed during early fruit expansion rather than ripening, and so

expression of OSC genes should peak relatively early. Various studies have further shown that

fruit waxes from different tomato cultivars differ in the relative portions of triterpenoids,

especially of -amyrin, with ratios of 3:3:2 between -amyrin, -amyrin and -amyrin in

MicroTom, and ratios of 3:2:2 in most other cultivars including M82 and Ailsa Craig (Vogg et

al., 2004; Mintz-Oron et al., 2008; Isaacson et al., 2009). This suggests that more than one OSC

should be involved in formation of the different triterpenoids, and that the varying product ratios

should reflect different expression levels and/or enzyme activities in the different cultivars.

However, it has not been determined how many OSCs are involved in forming the tomato fruit

cuticular triterpenoids, and which of those OSCs are product-specific or multifunctional

enzymes.

Based on all the previous evidence, the goal of the present investigation was to clone and

characterize multiple OSCs from tomato fruit. In particular, one or more enzymes forming -

amyrin were targeted. The primary focus of the investigations was on the cultivar MicroTom;

however, other cultivars such as M82 and Ailsa Craig were also included in order to compare

OSC sequences and expression patterns. With this, we sought to answer the question which

OSCs are responsible for the formation of the triterpenoids accumulating in the tomato fruit

cuticle and, thus, contribute to the important ecophysiological functions of the fruit epidermis.



9

RESULTS

The goal of the present investigation was to isolate and characterize OSCs from tomato

fruit, and to test their involvement in the formation of cuticular triterpenoids accumulating in the

fruit skin. To this end, a set of four experiments was carried out centered on the tomato cultivar

MicroTom: First, a homology-based PCR approach was used to clone the enzyme(s), initially

targeting a core segment of the sequence and then extending it by RACE experiments at the 5’

and 3’ termini. Second, the biochemical characterization of the enzyme(s) was carried out by

heterologous expression in yeast and by overexpression in tomato. Third, the expression of OSC

enzyme(s) in the epidermis and/or in inner parts of the fruit was evaluated. Finally, qRT-PCR

was used to determine the relative expression levels of the OSCs, both in the leaves and fruit of

MicroTom. The expression studies were extended to the tomato cultivars M82 and Ailsa Craig,

and complemented by chemical analyses of the triterpenoids in the M82 and MictroTom

cuticular waxes.

Cloning and Sequence Analysis of Oxidosqualene Cyclases from Tomato

In order to obtain a core sequence, a PCR was performed with a set of degenerate primers

designed using conserved OSC amino acid sequences and cDNAs isolated from the epidermis of

growing fruit. The primary PCR product corresponded to the expected size of approximately

1000 bp. A second PCR was used to extend the fragment, the product was isolated and cloned

into E. coli, and sequencing inserts from five bacterial colonies revealed the presence of only two



10

different cDNAs. Both had substantial similarity with other OSC sequences, and were subjected

to 5’ and 3’ RACE resulting in two full-length cDNAs that were designated as SlTTS1 and

SlTTS2.

The open reading frames (ORFs) of SlTTS1 and SlTTS2 are predicted to encode proteins

of 761 and 763 amino acids with molecular weights of 89.7 kDa and 88.1 kDa, respectively

(Figure 2). The two protein sequences are 88% identical, and both contain the DCTAE motif

thought to be involved in substrate binding and the four QW motifs characteristic of the OSC

superfamily. Furthermore, SlTTS1 and SlTTS2 are highly similar to previously reported -

amyrin synthases and multifunctional triterpene synthases (Table I). Phylogenetic analysis using

neighbour-joining methods showed that SlTTS1 and SlTTS2 are more closely related to each

other than to any other OSC and that they, together with the Panax ginseng -amyrin synthases

(Kushiro et al., 1998a; Kushiro et al., 1998b), form a subclade within a group of OSC enzymes

that were all characterized as -amyrin synthases from various plant species (Figure 3).

In order to analyze the gene structure of SlTTS1 and SlTTS2, the tomato genome

sequence database (www.solgenomics.net) was queried using the two corresponding cDNA

sequences. One BAC clone, Hba0131G17, was found to contain both OSC genes and,

accordingly, SlTTS1 and SlTTS2 are arranged in tandem in a 23 kb region on chromosome 12.

The intron patterns and exon lengths of the two SlTTS genes are very similar to those of other

OSCs (Figure 4), with SlTTS2 gene organization most closely resembling OSC3 of Lotus

japonica (Sawai et al., 2006b). However, SlTTS2 differs from L. japonica OSC3 in the length of

the first and last exons. The two tomato genes differ from each other in that exons 16 and 17 in


http://www.solgenomics.net/


11

SlTTS2 are fused into a single exon 16 SlTTS1, which indeed distinguishes SlTTS1 from all other

OSC genes studied to date.

Biochemical Characterization of the Tomato SlTTS Proteins

In order to biochemically characterize SlTTS1 and SlTTS2, two sets of experiments

using heterologous expression in yeast and overexpression in tomato fruit were carried out. For

yeast expression, the full-length cDNAs of the two putative OSCs were cloned into the

expression vector pYES2, and this construct was transformed into the yeast mutant GIL77 (gal2

hem3-6 erg7 ura3-167). This host strain lacks lanosterol synthase and so accumulates 2,3-

oxidosqualene, which can serve as a substrate for heterologously expressed OSCs (Morita et al.,

1997). Yeast transformants harbouring one of the pYES-TTS constructs or the empty pYES

vector were grown, gene expression was induced with galactose for 24 h, and then lipophilic

compounds were extracted with hexane. After purification by TLC, the triterpenoids were

identified by comparing their GC-MS characteristics with those of authentic standards and

literature data (Figures 5 and 6).

Yeast cells expressing SlTTS1 cDNA were found to produce a single triterpenoid product

that was identified as -amyrin (Figure 7). In contrast, heterologous expression of SlTTS2 led to

the formation of a mixture of triterpenoids, comprising 48 ± 0.3% of -amyrin, 13 ± 0.1% of -

amyrin, 18 ± 0.8% of -amyrin and 3 - 7% each of multiflorenol, -taraxasterol, taraxasterol

and one unidentified triterpene alcohol isomer. The yeast expression experiments were repeated



12

three times, always with very similar results. Overall, they indicate that SlTTS1 is a

monofunctional -amyrin synthase, while SlTTS2 is a multifunctional OSC enzyme catalyzing

the formation of seven different triterpenoid isomers.

In a second experiment, the potential OSC activities of the two tomato SlTTS proteins

were further assessed by overexpression in tomato and chemical analysis of the resulting fruit

surface waxes. Both SlTTS cDNAs were cloned into the pGWB5 vector, transformed into tomato

cv. MicroTom, and expressed under transcriptional control of the 35S promoter. Two sets of

twenty independent lines constitutively expressing SlTTS1 or SlTTS2 were recovered, and the

surface materials from fruit from the F1 generation were extracted for wax analysis. The SlTTS1

overexpressor lines were found to have amounts of -amyrin and -amyrin similar to those on

wildtype tomato fruit surfaces (Figure 8). In contrast, the amounts of -amyrin were doubled

from 0.7 g/cm2 in wild type to 1.4 g/cm

2 in SlTTS1 overexpressors. All the non-amyrin

triterpenoids remained unchanged after SlTTS1 overexpression (data not shown).

The transgenic lines harboring the SlTTS2 overexpression construct showed highly

variable triterpenoid amounts in their fruit cuticular waxes, while the very long chain fatty acid

derivatives were present at constant levels identical to those in the wild type. The transgenic fruit

could be classified into three distinct categories with significantly different triterpenoid profiles

(Student’s t-test, p<0.01). One of them, found for four of the lines, had triterpenoid amounts very

similar to wildtype levels. Three of the lines showed a gain of function phenotype, characterized

by significant increases of all seven fruit wax triterpenoids. Notably, the amounts of -, - and -

amyrin were all approximately doubled while the ratios between them were unchanged in



13

comparison with wild type (Figure 8). The remaining four lines recovered in the course of the

SlTTS2 overexpression experiment had fruit surface waxes containing increased levels of -

amyrin and drastically reduced levels of -amyrin and -amyrin when compared with wild type

(Figure 8). Similarly, the quantities of the non-amyrin triterpenoids were also decreased, in these

cases to trace levels that were detectable only by single ion monitoring GC-MS (data not shown).

Investigation of SlTTS Expression Patterns in Tomato Fruit

RT-PCR analyses were employed to test the tissue specificity of OSC transcription within

the green tomato fruit. To this end, two pairs of primers were designed based on the SlTTS1 and

SlTTS2 sequences, and proved to be gene-specific when tested with plasmid templates

harbouring either of the two OSCs cDNAs (data not shown). To determine whether the SlTTS

genes were expressed differentially between inner parts of the tomato fruit and the epidermis

layer, total RNA was isolated from both tissues. RT-PCR analysis using subsequently derived

cDNA templates showed that both SlTTS transcripts are expressed in the epidermal cells, but not

in the inner tissues of the fruit (Figure 9). Parallel metabolite analyses revealed that the major

part of the triterpenoids in the tomato fruit reside in the cuticle, with 74%, 79% and 74% of the

total -amyrin, -amyrin and -amyrin aglycone amounts present in the surface wax,

respectively, and the remaining 21 – 26% in the underlying epidermal cell layer and the internal

tissues of fruit.



14

Relative Expression Levels of SlTTS1 and SlTTS2 in Fruit and Leaves of Different Tomato

Cultivars

Finally, quantitative RT-PCR analyses were carried out to determine the relative

expression level of SlTTS1 and SlTTS2 within leaves and fruit. Since the ratios of different

triterpenoid products vary not only between these two organs, but also between fruit of different

tomato cultivars (Vogg et al., 2004; Mintz-Oron et al., 2008; Isaacson et al., 2009), this

experiment was to also compare expression patterns of both SlTTS genes between MicroTom and

the cultivars M82 and Ailsa Craig. The same primers originally designed for the cloning of the

full-length sequence from MicroTom were used to amplify corresponding cDNAs from M82 and

Ailsa Craig. One allele each of SlTTS1 and SlTTS2 was cloned and sequenced from both

cultivars, and the ORFs of the orthologous genes were found to be very similar or identical

between M82 and Ailsa Craig (Suppl. Figure 1). However, both SlTTS1 and SlTTS2 differed

between these cultivars and MicroTom, with up to eight nucleotide differences between cultivars

resulting in up to four amino acid changes, and >99.6% DNA and >99.4% protein sequence

identity (Table I). The similarities between both genes were approximately 89% on the

nucleotide level across cultivars, and 87 - 88% on the amino acid level.

The expression levels of SlTTS1 and SlTTS2 were found to be similar to each other in the

leaves and fruit of the cultivar MicroTom (Figure 10). Both genes were also expressed at similar

levels to each other in leaves of M82, whereas in the fruit of this cultivar SlTTS2 was expressed

at a much higher level than SlTTS1. The cultivar Ailsa Craig differed further, with higher

expression levels of SlTTS2 in fruit and leaves.



15

Terpenoid Profiles in Fruit and Leaves of Different Tomato Cultivars

The terpenoid profiles of the cuticular wax mixtures of leaves and fruits from the

MicroTom, M82 and Ailsa Craig cultivars were assessed, with particular focus on the relative

compositions of amyrin isomers. The compositions of fruit waxes from the three cultivars (Vogg

et al., 2004; Mintz-Oron et al., 2008; Isaacson et al., 2009) and also the leaf wax composition of

MicroTom had been reported before (Vogg et al., 2004). However, exact data on the leaf

triterpenoid profiles of all three cultivars were missing. Therefore, a detailed study of the leaf

wax from M82 and MicroTom was carried out. It was found that the M82 leaf was covered by

4.8 ± 0.5 g/cm2 of cuticular wax, and that the wax mixture contained 3.0 ± 1.0% of

triterpenoids, 64.3% ± 12.5% of very long chain fatty acid derivatives and 32.7 ± 13.5% of

unidentified compounds (Figure 11). n-Alkanes (C27 to C33) were the dominating compound

class with 45.6% ± 8.9%, accompanied by 16.6% ± 3.1% of branched alkanes (C29 – C33) and

20.5% ± 7.4% of alcohols (C28 and C32). The triterpenoid fraction was dominated by amyrin

isomers, comprising 29%, 33.4% and 14.4% of -amyrin, -amyrin and -amyrin, respectively.

Other triterpenoid alcohols were detected in the form of lupeol (23.3%) and trace amounts of

multiflorenol, -taraxasterol and taraxasterol. MicroTom leaf wax had a similar composition,

with a conspicuous ratio of 2:3:1 between of -amyrin, -amyrin and -amyrin, respectively.



16

DISCUSSION

In the present study, a homology-based approach was used to clone two tomato OSCs,

SlTTS1 and SlTTS2, which were found to encode highly similar amino acid sequences. Our

principal goal was to characterize the product spectrum of the enzymes, with the expectation that

a -amyrin synthase might be uncovered. Furthermore, we aimed to elucidate the role of the

enzymes in the formation of the triterpenoids accumulating in the fruit cuticular wax.

A phylogenetic analysis showed that SlTTS1 and SlTTS2 fall into a clade consisting

entirely of -amyrin synthases. Thus, the primary sequence of the new OSCs is in accordance

with the prediction that the tomato enzymes catalyze the formation of amyrins. However, based

on genetic information alone, it was not possible to predict the product specificities of the tomato

enzymes. It should also be noted that, in contrast to the similarity of exon sequences with amyrin

synthases, the intron pattern of SlTTS2 was identical to that of OSC3 from Lotus japonica, a

lupeol synthase that had fairly dissimilar sequence (Sawai et al., 2006b).

The product profiles of SlTTS1 and SlTTS2 were determined by heterologous expression

in yeast. It was found that SlTTS1 forms -amyrin as its sole product, while SlTTS2 catalyzes

the formation of seven different triterpenoids, with -amyrin as the major product. The in planta

overexpression of SlTTS1 and SlTTS2 led to increased accumulation of cuticular triterpenoids, in

both cases of the same products that had also been found in yeast. This finding is very important,

since most previous reports on biochemical characterizations of OSCs had relied entirely on

yeast expression systems, and it had only rarely been tested whether the heterologous



17

environment might alter the enzyme specificity (Han et al., 2006). The match between yeast and

in planta expression results in the present study confirms the validity of the yeast expression

system, at least for the two tomato OSCs studied. Overall, we conclude that the two enzymes are

a single product -amyrin synthase and a multifunctional OSC, respectively. SlTTS2 is the first

enzyme reported to synthesize predominantly -amyrin.

Interestingly, some of the tomato lines harboring the SlTTS2 construct exhibited a wax

triterpenoid phenotype opposite to that of the SlTTS2 overexpressors, suggesting that in these

lines the native gene was down-regulated by co-suppression. The observed increase in -amyrin

may be the result of additional 2,3-oxidosqualene substrate availability for TTS1 upon silencing

of SlTTS2. The SlTTS2 loss-of-function lines thus not only confirm the biochemical function of

the enzyme, but they also indicate that this enzyme must play a central role in the formation of

the tomato fruit wax triterpenoids. We conclude that SlTTS2 is crucial for the biosynthesis of six

of the seven triterpenoid isomers found in the wax, and partially also contributes to the formation

of the seventh component, -amyrin. Another part of the latter compound is formed by the

closely related enzyme SlTTS1. Both enzyme product profiles taken together completely match

the wax triterpenoid composition and, therefore, SlTTS1 and SlTTS2 are likely the two major

enzymes forming the triterpenoids found in the tomato fruit cuticular wax. The involvement of

additional OSCs cannot be excluded at this point, but it seems unlikely that any OSC other than

SlTTS1 and SlTTS2 would contribute substantially to the triterpenoid amounts or spectrum.

Two apparently orthologous OSCs were identified from each of the tomato cultivars M82

and Ailsa Craig, with a high degree of sequence identity to those from MicroTom. The



18

differences between the two genes within one cultivar were much greater than the differences

between the alleles in the cultivars. As differences between orthologous proteins were restricted

to changes in three or four amino acids, it is very likely that both SlTTS1 and SlTTS2 were

functional enzymes in all three cultivars, and that their biochemical functions should be very

similar in MicroTom, M82 and Ailsa Craig. We conclude that all three cultivars have at least one

-amyrin synthase (SlTTS1) and one multifunctional OSC producing -amyrin and other

isomers (SlTTS2).

The relative expression levels of the SlTTS1 and SlTTS2 genes varied between fruit of the

three tomato cultivars investigated, with transcripts of both genes showing similar abundance in

MicroTom, but SlTTS2 accumulating to higher levels than SlTTS1 in M82 and Ailsa Craig. This

expression pattern is in accordance with the amyrin profiles in the corresponding fruit waxes,

where the latter two cultivars had lower levels of the SlTTS1 product -amyrin than MicroTom

(Vogg et al., 2004; Mintz-Oron et al., 2008; Isaacson et al., 2009). However, it should be noted

that the differences in triterpenoid profiles between cultivars may further be due to factors other

than differential gene expression, possibly including differences in enzyme activities and/or

additional OSCs being present.

Our qRT-PCR results showed that the SlTTS1 and SlTTS2 genes are expressed not only in

tomato fruit, but also in the leaves of MicroTom, M82 and Ailsa Craig. Therefore, it seems likely

that both genes also contribute to the formation of leaf cuticular triterpenoids. However, it must

be noted that the leaf waxes have triterpenoid profiles differing from those in the fruit in

quantitative and in qualitative terms, as shown by our detailed analyses of M82 leaf wax in



19

comparison with literature data. The most prominent quantitative difference is the shift in amyrin

ratios from 3:2:2 for -amyrin, -amyrin and -amyrin in the fruit to 2:2:1 in the leaf wax. This

trend correlates with the differences in expression levels of SlTTS1 and SlTTS2 between both

organs of M82, and can thus, at least in part, be explained by transcriptional control. It should be

noted that Ailsa Craig leaves showed a very conspicuous expression pattern with particularly low

levels of SlTTS2 transcript. Based on this finding, it may be expected that the leaf waxes of this

cultivar should contain relatively little -amyrin, however, the amyrin profiles of the leaf wax of

Ailsa Craig have not been determined to date.

One outstanding qualitative difference between the triterpenoid compositions of fruit and

leaves, at least of the cultivars MicroTom and M82, is the presence of lupeol in the latter organ, a

product that is not formed by either SlTTS1 or SlTTS2. Therefore, one or more other OSCs must

be involved in formation of leaf cuticular triterpenoids. More detailed investigations are needed

to determine the contribution of SlTTS1 and SlTTS2 to the biosynthesis of tomato leaf cuticles.

Our results on the tissue-specific expression of SlTTS1 and SlTTS2 showed that these

enzymes are localized exclusively in the epidermis of tomato fruit. We further found that the

triterpenoid products are also restricted to the fruit skin, as they accumulate to high

concentrations in the cuticular waxes coating the epidermis, but not in the internal parts of the

fruit. The close match in localization of transcripts and metabolites makes it very likely that

SlTTS1 and SlTTS2 are dedicated entirely to making the triterpenoids destined for the cuticular

wax of the fruit surface. With this, a major biological function can be assigned to these two

OSCs, as the cuticular triterpenoids contribute significantly to the chemical composition and to



20

the ecophysiological properties of the fruit cuticle (Vogg et al., 2004; Isaacson et al., 2009). It

should be noted that similar biological functions had previously been attributed to a few other

OSCs, for example a glutinol synthase and a friedelin synthase from Kalanchoe daigremontiana

(Wang et al., 2010).

The sequences of SlTTS1 and SlTTS2 are relatively similar to each other, containing

only 92 amino acid changes. On the other hand, the two enzymes were found to have fairly

distinct product profiles, with SlTTS1 yielding exclusively -amyrin whereas SlTTS2 is

catalyzing the formation of -amyrin (among other products). It has been predicted that the

formation of -amyrin requires only one rearrangement step less than -amyrin in the course of

the OSC-catalyzed cyclization (Fig. 1) (Xu et al., 2004; Phillips et al., 2006). This relatively

subtle difference in the mechanism of formation of both triterpenoids must be due to small

differences in protein architecture, possibly involving specific changes in amino acids lining the

active site cavity. The sequence differences between SlTTS1 and SlTTS2 give first information

on the candidate residues that may be involved in determining the amyrin isomer specificity.

The list of candidate residues for defining -amyrin synthases can be somewhat narrowed

down based on comparisons between the two OSCs characterized here in three different tomato

cultivars, and with -amyrin synthases from other species. A second observation might further

help select candidate residues: many of the amino acid changes between both enzymes are

clustered together, and many involve aromatic residues. Since it is well established that such

aromatic amino acids play important roles in stabilizing the positive charge of high-energy

intermediates of the cyclization reaction, it has been speculated that the number and exact



21

position of aromatic residues lining the active site might limit the number of rearrangement steps

and, therefore, determine the product outcome of OSC-catalyzed reactions (Phillips et al., 2006;

Christianson, 2006). Consequently, the changes between SlTTS1 and SlTTS2 involving aromatic

amino acids at positions 48/49, 103, 318, 388, 399, 412, 421, 478, 496, 560, 675, 742 and 760

make these residues especially interesting candidates for defining the OSC specificity of -

amyrin and -amyrin synthases. It appears very promising to use site-directed mutagenesis

experiments to test this hypothesis, in order to further our detailed understanding of the enzyme

mechanisms involved in triterpenoid biosynthesis.



22

MATERIALS AND METHODS

Plant Material and Surface Wax Analysis

Tomato (Solanum lycopersicum cv. MicroTom) plants were grown in standard soil under

ambient conditions in a greenhouse at the University of British Columbia. Fruits and leaves were

harvested and immediately immersed in CHCl3 for 1 min at room temperature to extract the

cuticular waxes. For direct GC-MS analyses, the wax mixtures were derivatized using bis-N,O-

(trimethylsilyl)trifluoroacetamide (BSTFA) (in pyridine at 70 °C for 60 min), dried under N2 and

dissolved in CHCl3. The qualitative composition was studied using a 6890N Network GC

(Agilent) equipped with a mass spectrometric detector (Agilent 5973N) and an HP-1 capillary

column (Agilent, length 30 m, inner diameter 320 µm, 1 µm film thickness). 1 µl of each sample

was injected on-column into a flow of helium gas held constant at 1.4 ml/min. The oven

temperature was programmed for 2 min at 50 ˚C, followed by a 40 ˚C/min ramp to 200 ˚C, held

at 200 ˚C for 2 min, increased by 3 ˚C/min to 320 ˚C, and held at 320 ˚C for 30 min.

Triterpenoids were identified by comparison with authentic compounds (-amyrin, -amyrin, -

amyrin, lupeol) and with literature data. The quantitative composition was studied using a similar

GC system equipped with flame ionization detector (FID) under the same GC conditions as

above, but H2 carrier gas inlet pressure programmed for 2 ml/min. A known amount of n-

tetracosane was added to the solvent prior to extraction, and used as internal standard for

quantifying compound amounts. The extracted surface areas of leaves were determined using

digital images and ImageJ software.



23

For analysis of triterpenoids contained in internal tissues, fruit were surface-extracted

with CHCl3 to remove cuticular waxes, frozen in liquid nitrogen and then ground to powder. The

powder was extracted with CHCl3 and the resulting lipid mixture fractionated by TLC

(20 × 20 cm, silica gel, 0.5 mm; Merck) using CHCl3 as the mobile phase. After the plate was

stained with primuline and viewed under UV-light, distinct bands were scratched off, extracted

with CHCl3 and analyzed by GC-MS and GC-FID for identification and quantification of the

triterpenoids, respectively. GC conditions were used as described above for wax analyses.

Cloning of OSC Genes from Tomato

Green tomato fruits were harvested and immediately plunged into liquid nitrogen and

ground thoroughly with a mortar and pestle. Total RNA was extracted from the powder using the

Trizol Reagent (Invitrogen). The RNA was used for cDNA synthesis by SuperScript II Reverse

Transcriptase (Invitrogen) following standard protocols. The resulting cDNA mixture served

directly as a template for the following PCRs.

Sequence alignments of previously characterized plant OSCs revealed conserved regions,

which were used to design degenerate oligonucleotide primers for the specific amplification of

the core fragments of OSCs from the cDNA mixture. The antisense primer AMYT was derived

from the EST clone TC160438 (TIGR tomato EST database) suspected to encode part of an OSC

sequence. To obtain the triterpenoid synthase core sequence, a PCR was performed with the

degenerate sense primer OGA1S (5’TTYGGHAGYCAARMRTGGGAT3’) combined with



24

antisense primer AMYT (5’CGGTATTCAGCCAAACCCCA3’), with recombinant Taq DNA

Polymerase (Invitrogen) under the following cycling conditions: 2 min 94 °C, (20 s 94 °C, 40 s

54 °C, 70 s 72 °C) × 30 cycles, 10 min final extension at 72 °C. The resulting PCR products

were separated by gel electrophoresis (1% agarose) and extracted using the QIAquick Gel

Extraction Kit (Qiagen). A DNA band of 1 kb was recovered and cloned into the pGEM-T vector

(Promega, Madison, WI, USA) and transformed into Top10 competent cells. Plasmid DNA was

purified from transformed cells using QIAprep Spin Miniprep Kit (Qiagen) and sequenced. All

further PCR products mentioned below were subcloned and sequenced by the same procedure.

The core fragment was extended in a second PCR (conditions as above) using the primers

OGT4S (5’CAYCAGAAYGAAGATGGW3’) and gene-specific primer OGAS1A

(5’CATCATTCATCTCACTGGC3’) synthesized according to the obtained core sequences. Two

different core sequences were obtained, one named SlTTS1 and found to be identical to the

database sequence TC160438, and another one named SlTTS2. For 3’-end amplification of the

SlTTS2 cDNA, first strand synthesis was carried out for 1 h at 42 °C using 5 μg total RNA, AP-

primer (5’ GGCCACGCGTCGACTAGTACTTTTTTTTTTTTTTTTT 3’), and reverse

transcriptase in 20 μl. The product served as template in a PCR (conditions as above except

annealing temperature 55 °C) with the adapter primer AUAP

(5’GGCCACGCGTCGACTAGTAC 3’) and the specific primer LEASB1S

(5’AGGGTTGTGGTAGTCAATC3’). The 5’-ends of both cDNAs were amplified by the 5’

RACE system of Invitrogen (Version 2.0) with two nested gene-specific anti-sense primers

OGAS6A (5’AGAATCCATTTCCTTGCTCTA3’) and OGAS7A

(5’CACGCATTATTTACACCGCC3’).



25

The corresponding full-length cDNAs were amplified using tomato cDNA as a template

and the specific N- and C-terminal primers, respectively:

LeTTS1F: 5’TTGGAGCTCAGGATGTGGAAATTGAAAATTGCTG3’, SacI site underlined

LeTTS1R:5’CCCGAATTCTTAGTTGTTTTCTAATGGTAATAC3’, EcoRI site underlined

LeTTS2F: 5’TTGGAGCTCAAGATGTGGAAGTTGAAGATTGCAA3’, SacI site underlined

LeTTS2R: 5’CCCGAATTCCTATATGTAGTTGTGTTTTAATGGT3’, EcoRI site underlined.

The PCRs were conducted with Phusion High fidelity DNA polymerase (NEB) in a final volume

of 50 μl (1 μM of each primer and 1 μl of cDNA) under the following conditions: 30 s 98 °C

(10 s 98 °C, 30 s 55 °C, 75 s 72 °C) × 28 cycles, 10 min 72 °C. The resulting 2.3 kb PCR

products were purified by gel electrophoresis and cloned for sequencing. Sequence information

for the putative triterpenoid synthases SlTTS1 and SlTTS2 has been deposited in the GenBank

Accession Nos. HQ266579 and HQ266580, respectively.

Functional Expression of OSC cDNAs in Yeast and Product Analysis

The full-length cDNAs of putative triterpenoid synthases were double-digested with SacI

and EcoRI enzymes and ligated into the yeast expression vector pYES2 (Invitrogen) under the

control of the GAL1 promoter. The constructs were transformed into Top10 Competent Cells.

Plasmid DNAs were prepared and used to transform the mutant yeast strain GIL77 by the

LiAc/SS-DNA/PEG method (Gietz and Woods, 2002). After galactose induction and 24 h

incubation in 0.1 M potassium phosphate-containing glucose and hemin, cells were collected,



26

refluxed for 5 min in 20% KOH/50% EtOH, and extracted twice with hexane. Both hexane

solutions were combined, the solvent was removed under a gentle stream of N2 and the residue

was re-dissolved in 0.2 ml of CHCl3. The extracts were either directly derivatized using BSTFA

at 70 °C for 60 min and then analyzed by GC–FID/MS as described above or further purified by

thin layer chromatography (TLC) plate (20 × 20 cm, silica gel 60 F254, 0.25 mm; Merck,

Darmstadt, Germany) for GC-FID/MS analysis. Plates were developed using a sandwich

technique and chloroform as the mobile phase, then stained with primuline and viewed under

UV-light. The bands potentially containing cyclization products were scratched from the plates,

extracted with CHCl3, filtered, and prepared for GC analysis.

Tomato overexpression

SlTTS1 and SlTTS2 cDNAs were cloned into the pGWB5 vector under the control of 35S

promoter by using the Gateway Cloning system (Invitrogen). The vectors were transferred to

Agrobacterium tumefaciens GV3301. Plant transformation was carried out as described by Dan

et al. (2006). Transgenic plants were growing in a green house until the fruits turned to mature

red, then harvested to do wax extraction and GC/FID/MS analysis as described before.

RT-PCR analysis.



27

The epidermal cell layers were peeled from the green fruits surface of MicroTom, the

remaining inner tissue as well as the epidermis preparations were immediately frozen in liquid

nitrogen, ground into powders using a mortar and pestle, and used to extract RNA with Trizol

Reagent (Invitrogen). The RNA samples were used for cDNA synthesis by SuperScript II

Reverse Transcriptase (Invitrogen) following standard protocols. Gene-specific primers

(SlTTS1: 5’CGTCGAATGCACTGCCTCAT3’ and 5’GGACCAAATTGCACTCTATATC3’;

SlTTS2: 5’TGTTGAGTGCACTAGCTCGG3’ and 5’ACGGACAACTCGATTCACTAAGC3’)

were designed to amplify fragments of the two OSCs. Additionally, the actin gene fragment was

amplified as a positive control using the primers actinF: 5’CAAGTCATCATCCGTTTG3’ and

actinR: 5’ATACCAGTGGTACGACC3’. PCR cycle numbers and template amounts were

optimized to yield products in the linear range of the reaction. PCR conditions were: denatured at

94 °C for 2 min, followed by 28 cycles of: 94 °C for 20 s, 55 °C for 30 s, and 72 °C for 60 s.

Reactions were maintained at 72 °C for 2 min before separation of PCR products by

electrophoresis in a 1% agarose gel.

Quantitative RT-PCR analysis.

The RNA samples from tomato leaf or green fruits were used for cDNA synthesis by

SuperScript II Reverse Transcriptase (Invitrogen) following standard protocols. Gene-specific

primers were designed to amplify a fragment of SlTTS1 by TTS1F:

CGTCGAATGCACTGCCTCAT and T1R2: 5’TACCATGAACCATCAGGCATT3’, and

SlTTS2 by TTS2F: 5’TGTTGAGTGCACTAGCTCGG3’ and T2R2:

5’TACCATGAACCGTCAGGCTCC3’. Quantitative PCR was carried out using SYBR

GreenER qPCR SuperMix Universal kit (Invitrogen). The q-PCRs were programmed at 95 °C



28

for 9 min, followed by 40 cycles of: 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s on a MJ

mini opticon real-time PCR system.

Phylogenetic Analyses

Sequence alignments and phylogenetic analyses based on a neighbour-joining method

were carried out with the ClustalX program Version 1.83 (Thompson et al., 1997), using the

amino acid sequences of cloned and characterized plant OSCs. A phylogenetic tree was created

with the MEGA3.1 program. The number of bootstrap replications was 1000. The GenBank

database accession numbers of the sequences used in the analysis are summarized in table II:



29

SUPPLEMENTAL DATA

The following materials are available in the online version of this article.

Supplementary Figure 1 Amino acid sequences of the SlTTS1 and SlTTS2 alleles of tomato

cultivars MicroTom and M82.

ACKNOWLEDGEMENTS

The authors would like to thank Bangjun Wang for technical help and Dr. Y. Ebizuka

(Tokyo) for providing the yeast strain GIL77. This work has been supported by NSERC Special

Research Opportunity and Strategic Grants, by the Canada Foundation for Innovation, the British

Columbia Knowledge Development Fund and the Canada Research Chairs Program. Funding

was provided to J.K.C.R. by NSF Plant Genome Research Program grant (DBI-0606595) and

T.H.Y. was supported in part by an NIH chemistry/biology interface training grant (grant number

T32 GM008500).



30

REFERENCES

Abe I (2007) Enzymatic synthesis of cyclic triterpenes. Nat Prod Rep 24: 1311-1331

Abe I, Rohmer M, Prestwich GD (1993) Enzymatic cyclization of squalene and oxidosqualene

to sterols and triterpenes. Chem Rev 93: 2189-2206

Adato A, Mandel T, Mintz-Oron S, Venger I, Levy D, Yativ M, Dominguez E, Wang Z,

DeVos RCH, Jetter R, Schreiber L, Heredia A, Rogachev I, Aharoni A (2009) Fruit-

surface flavonoid accumulation in tomato is controlled by a SlMYB12-regulated

transcriptional network. PLoS Genetics 5:

Baker EA, Bukovac MJ, Hunt GM (1982) Composition of tomato fruit cuticle as related to

fruit growth and development. In: The plant cuticle, Linnean Society Symposium Series,

vol. 10. Cutler,D.F.; Alvin,K.L.; Price,C.E. Eds., Academic Press, London, p. 33-44

Basyuni M, Ebizuka Y, Oku H, Inafuku M, Baba S, Iwasaki H, Oshiro K, Okabe T,

Shibuya M (2006) Molecular cloning and functional expression of a multifunctional

triterpene cDNA from a mangrove species Kandelia candel (L.) Druce. Phytochemistry

67: 2517-2524

Bauer S, Schulte E, Thier H-P (2004a) Composition of the surface wax from tomatoes. 1.

Identification of the components by GC/MS. Eur Food Res Technol 219: 223-228

Bauer S, Schulte E, Thier H-P (2004b) Composition of the surface wax from tomatoes. 2.

Quantification of the components at the ripe red stage and during ripening. Eur Food Res

Technol 219: 487-491

Christianson DW (2006) Structural biology and chemistry of the terpenoid cyclases. Chem Rev

106: 3412-3442

Corey EJ, Matsuda SPT, Bartel B (1993) Isolation of an Arabidopsis thaliana gene encoding

cycloartenol synthase by functional expression in a yeast mutant lacking lanosterol

synthase by the use of a chromatographic screen. Proc Natl Acad Sci USA 90: 11628-

11632

Dan Y, Yan H, Munyikwa T, Dong J, Zhang Y, Armstrong CL (2006) MIcro-Tom - a high-

throughput model transformation system for functional genomics. Plant Cell Rep 25:

432-441



31

Ebizuka Y, Katsube Y, Tsutsumi T, Kushiro T, Shibuya M (2003) Functional genomics

approach to the study of triterpene biosynthesis. Pure Appl Chem 75: 369-374

Eschenmoser A, Ruzicka L, Jeger O, Arigoni D (1955) Zur Kenntnis der Triterpene. 190. Eine

stereochemische Interpretation der biogenetischen Isoprenregel bei den Triterpenen. Helv

Chim Acta 38: 1890-1904

Fazio GC, Xu R, Matsuda SPT (2004) Genome mining to identify new plant triterpenoids. J

Am Chem Soc 126: 5678-5679

Gietz RD, Woods RA (2002) Transformation of yeast by the LiAc/SS carrier DNA/PEG

method. Methods in Enzymology 350: 87-96

Han JY, Kwon YS, Yang DC, Jung YR, Choi YE (2006) Expression and RNA interference-

induced silencing of the dammarenediol synthase gene in Panax ginseng. Plant Cell

Physiol 47: 1653-1662

Hayashi H, Hiraoka N, Ikeshiro Y, Kushiro T, Morita M, Shibuya M, Ebizuka Y (2000)

Molecular cloning and characterization of a cDNA for Glycyrrhiza glabra cycloartenol

synthase. Biol Pharm Bull 23: 231-234

Hayashi H, Huang P, Kirakosyan A, Inoue K, Hiraoka N, Ikeshiro Y, Kushiro T, Shibuya

M, Ebizuka Y (2001) Cloning and characterization of a cDNA encoding -amyrin

synthase involved in glycyrrhizin and soyasaponin biosyntheses in licorice. Biol Pharm

Bull 24: 912-916

Herrera JBR, Bartel B, Wilson WK, Matsuda SPT (1998) Cloning and characterization of the

Arabidopsis thaliana lupeol synthase gene. Phytochemistry 49: 1905-1911

Hovav R, Chehanovsky N, Moy M, Jetter R, Schaffer AA (2007) The identification of a gene

(Cwp1), silenced during Solanum evolution, which causes cuticle microfissuring and

dehydration when expressed in tomato fruit. Plant J 52: 627-639

Husselstein-Muller T, Schaller H, Benveniste P (2001) Molecular cloning and expression in

yeast of 2,3-oxidosqualene-triterpenoid cyclases from Arabidopsis thaliana. Plant Mol

Biol 45: 75-92

Isaacson T, Kosma DK, Matas AJ, Buda GJ, He Y, Yu B, Pravitasari A, Batteas JD, Stark

RE, Jenks MA, Rose JKC (2009) Cutin deficiency in the tomato fruit cuticle

consistently affects resistance to microbial infection and biomechanical properties, but

not transpirational water loss. Plant J 60: 363-377



32

Iturbe-Ormaetxe I, Haralampidis K, Papadopoulou K, Osbourn AE (2006) Molecular

cloning and characterization of triterpene synthases from Medicago truncatula and Lotus

japonicus. Plant Mol Biol 51: 731-743

Kolesnikova MD, Wilson WK, Lynch DA, Obermeyer AC, Matsuda SPT (2010)

Arabidopsis camelliol C synthase evolved from enzymes that make pentacycles. Org Lett

9: 5223-5226

Kosma DK, Parsons EP, Isaacson T, Lu S, Rose JKC, Jenks MA (2010) Fruit cuticle lipid

composition during development in tomato ripening mutants. Physiol Plant 139: 107-117

Kushiro T, Shibuya M, Ebizuka Y (1998a) Molecular cloning of oxidosqualene cyclase cDNA

from Panax ginseng: the isogene that encodes -amyrin synthase. International

Symposium on Natural Medicines, Kyoto, Japan

Kushiro T, Shibuya M, Ebizuka Y (1998b) -Amyrin synthase. Cloning of oxidosqualene

cyclase that catalyzes the formation of the most popular triterpene among higher plants.

Eur J Biochem 256: 238-244

Kushiro T, Shibuya M, Masuda K, Ebizuka Y (2000) A novel multifunctional triterpene

synthase from Arabidopsis thaliana. Tetrahedr Lett 4: 7705-7710

Leide J, Hildebrandt U, Reussing K, Riederer M, Vogg G (2007) The developmental pattern

of tomato fruit wax accumulation and its impact on cuticular transpiration barrier

properties: effects of a deficiency in a -ketoacyl-CoA synthase (LeCER6). Plant Physiol

144: 1667-1679

Lodeiro S, Xiong Q, Wilson WK, Kolesnikova MD, Onak CS, Matsuda SPT (2007) An

oxidosqualene cyclase makes numerous products by diverse mechanisms: a challenge to

prevailing concepts of triterpene biosynthesis. J Am Chem Soc 129: 11213-11222

Mintz-Oron S, Mandel T, Rogachev I, Feldberg L, Lotan O, Yativ M, Wang Z, Jetter R,

Venger I, Adato A, Aharoni A (2008) Gene expression and metabolism in tomato fruit

surface tissues. Plant Physiol 147: 823-851

Morita M, Shibuya M, Kushiro T, Masuda K, Ebizuka Y (2000) Molecular cloning and

functional expression of triterpene synthases from pea (Pisum sativum). New -amyrin-

producing enzyme is a multifunctional triterpene synthase. Eur J Biochem 267: 3453-

3460

Morita M, Shibuya M, Lee M-S, Sankawa U, Ebizuka Y (1997) Molecular cloning of pea

cDNA encoding cycloartenol synthase and its functional expression in yeast. Biol Pharm

Bull 20: 770-775



33

Papadopoulou K, Melton RE, Leggett M, Daniels MJ, Osbourn AE (1999) Compromised

disease resistance in saponin-deficient plants. Proc Natl Acad Sci USA 96: 12923-12928

Phillips DR, Rasbery JM, Bartel B, Matsuda SPT (2006) Biosynthetic diversity in plant

triterpene cyclization. Current Opinion in Plant Biology 9: 1-10

Reichardt PB, Bryant JP, Clausen TP, Wieland GD (1984) Defense of winter-dormant

Alaska paper birch against snowshoe hares. Oecologia (Berlin) 65: 58-69

Saladie M, Matas AJ, Isaacson T, Jenks MA, Goodwin SM, Niklas KJ, Xiaolin R,

Labavitch JM, Shackel KA, Fernie AR, Lytovchenko A, O'Neill MA, Watkins CB,

Rose JKC (2007) A re-evaluation of the key factors that influence tomato fruit softening

and integrity. Plant Physiol 144: 1012-1028

Sawai S, Akashi T, Sakurai N, Suzuki H, Shibata D, Ayabe S-I, Aoki T (2006a) Plant

lanosterol synthase: divergence of the sterol and triterpene biosynthetic pathways in

Eukaryotes. Plant Cell Physiol 47: 673-677

Sawai S, Shindo T, Sato S, Kaneko T, Tabata S, Ayabe S-I, Aoki T (2006b) Functional and

structural analysis of genes encoding oxidosqualene cyclases of Lotus japonicus. Plant

Sci 170: 247-257

Segura MJR, Meyer MM, Matsuda SPT (2000) Arabidopsis thaliana LUP1 converts

oxidosqualene to multiple triterpene alcohols and a triterpene diol. Org Lett 2: 2257-2259

Shibuya M, Katsube Y, Otsuka M, Zhang H, Tansakul P, Xiang T, Ebizuka Y (2009)

Identification of a product specific -amyrin synthase from Arabidopsis thaliana. Plant

Physiol Biochem 47: 26-30

Shibuya M, Xiang T, Katsube Y, Otsuka M, Zhang H, Ebizuka Y (2007) Origin of structural

diversity in natural triterpenes: direct synthesis of seco-triterpene skeletons by

oxidosqualene cyclase. J Am Chem Soc 129: 1450-1455

Smith RM, Marshall JA, Davey MR, Lowe KC, Power JB (1996) Comparison of volatiles

and waxes in leaves of genetically engineered tomatoes. Phytochemistry 43: 753-758

Suzuki M, Xiang T, Ohyama K, Seki H, Saito K, Muranaka T, Hayashi H, Katsube Y,

Kushiro T, Shibuya M, Ebizuka Y (2006) Lanosterol synthase in dicotyledonous

plants. Plant Cell Physiol 47: 565-571

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The

CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment

aided by quality analysis tools. Nucleic Acids Res 25: 4876-4882



34

Vogg G, Fischer S, Leide J, Emmanuel E, Jetter R, Levy AA, Riederer M (2004) Tomato

fruit cuticular waxes and their effects on transpiration barrier properties: functional

characterization of a mutant deficient in a very-long-chain fatty acid -ketoacyl-CoA

synthase. J Exp Bot 55: 1401-1410

Wang Z, Yeats T, Han H, Jetter R (2010) Cloning and characterization of oxidosqualene

cyclases from Kalanchoe daigremontiana: enzymes catalyzing up to ten rearrangement

steps yielding friedelin and other triterpenoids. J Biol Chem, in press

Xiang T, Shibuya M, Katsube Y, Tsutsumi T, Otsuka M, Zhang H, Masuda K, Ebizuka Y

(2006) A new triterpene synthase from Arabidopsis thaliana produces a tricyclic

triterpene with two hydroxyl groups. Org Lett 8: 2835-2838

Xiong Q, Wilson WK, Matsuda SPT (2006) An Arabidopsis oxidosqualene cyclase catalyzes

iridal skeleton formation by grob fragmentation. Angew Chem Int Ed 45: 1285-1288

Xu R, Fazio GC, Matsuda SPT (2004) On the origins of triterpenoid skeletal diversity.

Phytochemistry 65: 261-291



35

FIGURE LEGENDS

Figure 1 Mechanism for the cyclization of epoxysqualene into pentacyclic triterpenoids.

The reaction starts with the protonation of oxidosqualene (1), then involves a series of

carbocationic intermediates that first undergo cyclization (2) and then various rearrangements

(3), before deprotonation (4) yields the various natural products.

Figure 2 Amino acid sequences of the two OSCs, SlTTS1 and SlTTS2, isolated from the

tomato cultivar MicroTom. The conserved QW motives thought to stabilize OSC structure are

highlighted by single underlines, while the highly conserved DCTAE motif involved in substrate

binding and protonation is marked by a double underline.

Figure 3 Phylogenetic analysis comparing the two new OSC cDNAs cloned from tomato cv.

MicroTom to previously known OSCs from other plant species. The gene names and

sequences as well as the full name of species are given in the Materials and Methods section.

Figure 4 Gene structure of the tomato cv. MicroTom OSCs in comparison with those of

other plant species. The ORFs are represented by lines, with introns as boxes. Intron lengths in

nucleotides are shown above the boxes, introns with common length between most OSCs are

shaded in grey, and introns with a characteristic length in only a few species are highlighted in

black.



36

Figure 5 Mass spectra of pentacyclic triterpenoids. Compounds 1, 2 and 4, formed by yeast

strains heterologously expressing the two tomato cv. MicroTom OSCs (left), have fragmentation

patterns identical with those of authentic standards of -amyrin, -amyrin and -amyrin (right).

Spectra are shown for the trimethylsilyl derivatives of the triterpenoid alcohols.

Figure 6 Mass spectra of pentacyclic triterpenoids. Compounds 3 and 5 - 7, formed by yeast

strains heterologously expressing the two tomato cv. MicroTom OSCs (left), have fragmentation

patterns identical with those of triterpenoids found in the cuticular wax of tomato fruit (right).

Wax constituents 5 – 7 were identified as multiflorenol, -taraxasterol and taraxasterol in

accordance with the literature, while compound 3 remained unidentified. Spectra are shown for

the trimethylsilyl derivatives of the triterpenoid alcohols.

Figure 7 Gas chromatographic analysis of triterpenoids in transgenic yeast. Yeast were

transformed with the vectors indicated in the various panels, grown and extracted. Triterpenoid

alcohols were separated from other neutral lipids by TLC, and converted into trimethylsilyl

ethers prior to GC analysis. In the empty vector control no triterpenoids were detected. In

contrast, the yeast strains harboring SlTTS1 and SlTTS2 constructs were found to contain the

triterpenoid compound 2 and a series of compounds 1 – 7, respectively. All seven compounds 1 –

7 were also detected in the MicroTom fruit cuticular wax, together with very long chain fatty

acid derivatives a – e (n-nonacosane, b: n-triacontane, c: n-hentriacontane, d: n-dotriacontane, e:

n-tritriacontane).



37

Figure 8 Relative amounts of amyrin isomers in the fruit cuticular waxes of various lines of

MicroTom. The percentages of -amyrin, -amyrin and -amyrin are given for the wild type

(WT), transgenic fruit overexpressing the OSCs (SlTTS1+ and SlTTS2+), and overexpressors

with co-suppressed SlTTS2 (SlTTS2-). Amounts of each triterpenoid were quantified as %

relative to the level of wildtype -amyrin and given as averages of three independent lines and

analyses with standard deviations.

Figure 9 RT-PCR analysis of the expression patterns of the two OSC genes in MicroTom

fruit. The SlTTS1 and SlTTS2 enzymes generating pentacyclic triterpenoids were found

expressed only in the epidermis layers, but not in the internal tissues of the fruit.

Figure 10 Quantitative RT-PCR analysis of the two OSC genes in fruit and leaves of the

three tomato cultivars MicroTom (MT), M82 and Ailsa Craig (AC). The relative expression

levels were determined in green immature fruit and normalized for each mRNA sample (n=3,

SD).

Figure 11 Chemical composition of cuticular wax on leaves of tomato cvs. M82 and

MicroTom. The absolute amounts of all identified compounds are given as averages of three

independent parallel experiments with standard deviations.



38

Table I. Identity levels in pairwise comparisons between the SlTTS1 and SlTTS2 cDNAs and alleles in the tomato cultivars MicroTom, M82 and

Ailsa Craig. Nucleotide identities are given as percentages in normal font, while percentages of amino acid identity are shown in italics.

TTS1 TTS2

MicroTom M82 Ailsa Craig MicroTom M82 Ailsa Craig

TTS1 MicroTom - 99.7% 99.6% 89.2% 89.2% 89.2%

M82 99.6% - 100% 89.1% 89.1% 89.1%

Ailsa Craig 99.5% 100% - 89.1% 89.1% 89.1%

TTS2 MicroTom 87.8% 87.7% 87.5% - 99.8% 99.8%

M82 87.7% 87.5% 87.4% 99.4% - 100%

Ailsa Craig 87.7% 87.5% 87.4% 99.4% 100% -



39

Table II: Accession data for the OSCs used in the phylogenetic analyses.

Accession No. Species Function

At4g15340 Arabidopsis thaliana Arabidiol synthase PEN1

At4g15370 Arabidopsis thaliana Baruol synthase BARS1/PEN2

At5g48010 Arabidopsis thaliana Thalianol synthase PEN4

At5g42600 Arabidopsis thaliana Marneral synthase PEN5

At1g78500 Arabidopsis thaliana Multifunctional triterpene synthase PEN6

At3g45130 Arabidopsis thaliana Lanosterol synthase LAS1/PEN7

At1g78970 Arabidopsis thaliana Multifunctional triterpene synthase LUP1


At1g78950 Arabidopsis thaliana β-amyrin synthase AtBAS LUP4


At2g07050 Arabidopsis thaliana Cycloartenol synthase CAS1

AB263204 Rhizophora stylosa Multifunctional triterpene synthase RsM2

AB257507 Kandelia candel Multifunctional triterpene synthase KcMS

AB037203 Glycyrrhiza glabra β-amyrin synthase GgbAS1

AB181244 Lotus japonicus β-amyrin synthase OSC1

AB034802 Pisum sativum β-amyrin synthase PSY

AJ430607 Medicago truncatula β-amyrin synthase AMY1

AF478455 Lotus japonicus Multifunctional triterpene synthase LjAMY2

AB034803 Pisum sativum Multifunctional triterpene synthase PSM

AB289585 Bruguiera gymnorhiza β-amyrin synthase BgbAS

AB014057 Panax ginseng β-amyrin synthase PgbAS/PNY2

AB009030 Panax ginseng β-amyrin synthase PNY

AB263203 Rhizophora stylosa Multifunctional triterpene synthase RsM1

AB206469 Euphorbia tirucalli β-amyrin synthase EtbAS

AB181245 Lotus japonicus Lupeol synthase OSC3

AB116228 Glycyrrhiza glabra Lupeol synthase GgLUS1

AB181246 Lotus japonicus Cycloartenol synthase QSC5

AB244671 Lotus japonicus Lanosterol synthase OSC7


http://www.sciencedirect.com/science?_ob=RedirectURL&_method=externObjLink&_locator=genbank&_cdi=6701&_plusSign=%2B&_targetURL=http%253A%252F%252Fwww.ncbi.nlm.nih.gov%252Fentrez%252Fquery.fcgi%253Fcmd%253Dsearch%2526db%253Dnucleotide%2526doptcmdl%253Dgenbank%2526term%253DAB037203%5Baccn%5D


40

C H3

CH3

CH3

C+

C H3

C H3

C H3

OH

H

C H3

H

C H3

H

C H3

CH3

CH3

C+

CH3

C H3

OH

H

C H3

H

C H3

H

C H3

H

C H 3

H

CH3

C+

C H3

C H3

OH

H

C H3

H

C H3

H

C H3C H

3

CH3

HCH3

CH3

C+

CH3

CH3

CH3

OH

H

CH3

H

CH3

CH3

HCH3

CH3

C+

CH3

CH3

CH3

CH3

OH

H

CH3

H

CH3

H

H

CH3

CH3

C+

CH3

CH3

CH3

CH3

OH

H

CH3

C H3

H

H

CH3

CH3

C+

C H3

C H3

CH3

C H3

C H3

OH

H

H

H CH3

CH3

CH3

C+

CH3

CH3 CH3

CH3

CH3

OH

H

H

H C H3

CH3

CH3

CH+

C H3

C H3

C H3

C H3

C H3

OH

H

H

H

H CH3

CH3

CH3

C+

C H3

C H3

CH3

C H3

C H3

OH

H

H

H CH3

CH3

CH3CH3

CH3 CH3

CH3

CH3

OH

H

CH3

H

H

CH3

CH3CH3

CH3 CH3

CH3

CH3

OH

H

CH3

CH3

CH3CH3

CH3

CH3

OH

H

CH3

H

CH3

H

Germanicol -Amyrin -Amyrin Taraxerol Multiflorenol Glutinol FriedelinIsomultiflorenol WalsurenolPicenol

CH3

CH3CH3

CH3

H

CH3

H

CH3

H

CH3O

CH3

CH3

CH3

CH3CH3

CH3 CH3

CH3

CH3

O

CH3

CH3

CH3

C+

CH3

CH3 CH3

CH3

CH3

OH

H

H CH3

CH3

CH3CH3

CH3 CH3

CH3

CH3

OH

Epoxysqualene

1

2

3 3 3 3 3 3 3 3 3

444444444

H

HC H

3

CH3

CH3

C H3

C H3

CH3

C H3

C H3

OH

H

Figure 1 Mechanism for the cyclization of epoxysqualene into pentacyclic triterpenoids. The reaction starts with the protonation of

oxidosqualene (1), then involves a series of carbocationic intermediates that first undergo cyclization (2) and then various rearrangements (3), before

deprotonation (4) yields the various natural products.



41

Figure 2 Amino acid sequences of the two OSCs, SlTTS1 and SlTTS2, isolated from the tomato cultivar MicroTom. The conserved QW

motives thought to stabilize OSC structure are highlighted by single underlines, while the highly conserved DCTAE motif involved in substrate

binding and protonation is marked by a double underline.



42

Figure 3 Phylogenetic analysis comparing the two new OSC cDNAs cloned from tomato cv. MicroTom to previously known OSCs from

other plant species. The gene names and sequences as well as the full name of species are given in the Materials and Methods section.



43

Figure 4 Gene structure of the tomato cv. MicroTom OSCs in comparison with those of other plant species. The ORFs are represented by lines,

with introns as boxes. Intron lengths in nucleotides are shown above the boxes, introns with common length between most OSCs are shaded in grey,

and introns with a characteristic length in only a few species are highlighted in black.

201 186 90 198 85 167 189 114 120 123 99 57 47 144 178 165 123

SlTTS2 201 186 90 198 85 167 189 114 120 123 99 57 47 144 178 81 84 129

198 186 90 195 85 167 189 114 123 123 99 57 47 144 178 81 84 120

SlTTS1

At CAS1

201 186 90 198 85 167 192 114 120 123 99 57 369 123

218 186 90 198 85 167 189 114 120 123 99 57 47 144 178 81 84

Lj OSC1

Lj OSC3

81 84

204 186 90 198 85 167 192 114 120 123 99 57 191 114 81 84 At Lup1

178



44

Figure 5 Mass spectra of pentacyclic triterpenoids. Compounds 1, 2 and 4, formed by yeast strains heterologously expressing the two tomato cv.

MicroTom OSCs (left), have fragmentation patterns identical with those of authentic standards of -amyrin, -amyrin and -amyrin (right). Spectra

are shown for the trimethylsilyl derivatives of the triterpenoid alcohols.



45

Figure 6 Mass spectra of pentacyclic triterpenoids. Compounds 3 and 5 - 7, formed by yeast strains heterologously expressing the two tomato cv.

MicroTom OSCs (left), have fragmentation patterns identical with those of triterpenoids found in the cuticular wax of tomato fruit (right). Wax

constituents 5 – 7 were identified as multiflorenol, -taraxasterol and taraxasterol in accordance with the literature, while compound 3 remained

unidentified. Spectra are shown for the trimethylsilyl derivatives of the triterpenoid alcohols.



46

Figure 7 Gas chromatographic analysis of triterpenoids in transgenic yeast. Yeast were transformed with the vectors indicated in the various

panels, grown and extracted. Triterpenoid alcohols were separated from other neutral lipids by TLC, and converted into trimethylsilyl ethers prior to

GC analysis. In the empty vector control no triterpenoids were detected. In contrast, the yeast strains harboring SlTTS1 and SlTTS2 constructs were

found to contain the triterpenoid compound 2 and a series of compounds 1 – 7, respectively. All seven compounds 1 – 7 were also detected in the

MicroTom fruit cuticular wax, together with very long chain fatty acid derivatives (a: n-nonacosane, b: n-triacontane, c: n-hentriacontane, d: n-

dotriacontane, e: n-tritriacontane).



47

WT

SlT

TS

1(+

)

SlT

TS

2(+

)

SlT

TS

2(-

)

Rela

tive

am

yri

n a

mo

un

ts

[% o

f w

ild

typ

e

-am

yri

n]

0

50

100

150

200

250-Amyrin

-Amyrin

-Amyrin**

**

**

**

**

**

**

Figure 8 Relative amounts of amyrin isomers in the fruit cuticular waxes of various lines of MicroTom. The percentages of -amyrin, -amyrin

and -amyrin are given for the wild type (WT), transgenic fruit overexpressing the OSCs (SlTTS1+ and SlTTS2+), and overexpressors with co-

suppressed SlTTS2 (SlTTS2-). Amounts of each triterpenoid were quantified as % relative to the level of wildtype -amyrin and given as averages of

three independent lines and analyses with standard deviations.



48

Figure 9 RT-PCR analysis of the expression patterns of the two OSC genes in MicroTom fruit. The SlTTS1 and SlTTS2 enzymes generating

pentacyclic triterpenoids were found expressed only in the epidermis layers, but not in the internal tissues of the fruit.

SlTTS1

SlTTS2

Actin

Epidermal

peel

Internal

tissue



49

Fruit Leaf

MT M82 AC MT M82 AC

Rela

tive E

xp

ressio

n[%

of

tota

l T

TS

tra

nscri

pts

in

org

an

]

0

20

40

60

80

100

TTS1

TTS2

Figure 10 Quantitative RT-PCR analysis of the two OSC genes in fruit and leaves of the three tomato cultivars MicroTom (MT), M82 and

Ailsa Craig (AC). The relative expression levels were determined in green immature fruit and normalized for each mRNA sample (n=3, SD).



50

C27

C28

C29

C30

C31

C32

C33

C29

C31

C32

C33

C26

C28

C30

C32

d-A

myri

nb

-Am

yri

na

-Am

yri

nL

up

eo

l

Wax lo

ad

[

g/c

m2]

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5 M82

MicroTom

n-AlkanesBranched

alkanesAlcohols

Figure 11 Chemical composition of cuticular wax on leaves of tomato cvs. M82 and MicroTom. The absolute amounts of all identified

compounds are given as averages of three independent parallel experiments with standard deviations.



Documents

Running head: Biosynthesis of tomato fruit triterpenoids