In vitro reconstruction and analysis of evolutionaryvariation of the tomato acylsucrose metabolic networkPengxiang Fana, Abigail M. Millera, Anthony L. Schilmillera, Xiaoxiao Liub, Itai Ofnerc, A. Daniel Jonesa,b, Dani Zamirc,and Robert L. Lasta,d,1

aDepartment of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI 48824; bDepartment of Chemistry, Michigan StateUniversity, East Lansing, MI 48824; cThe Institute of Plant Sciences and Genetics in Agriculture, Faculty of Agriculture, The Hebrew University of Jerusalem,Rehovot 76100, Israel; and dDepartment of Plant Biology, Michigan State University, East Lansing, MI 48824

Edited by Jerrold Meinwald, Cornell University, Ithaca, NY, and approved December 7, 2015 (received for review September 9, 2015)

Plant glandular secreting trichomes are epidermal protuberances thatproduce structurally diverse specialized metabolites, including medi-cally important compounds. Trichomes of many plants in the night-shade family (Solanaceae) produce O-acylsugars, and in cultivated andwild tomatoes these are mixtures of aliphatic esters of sucrose andglucose of varying structures and quantities documented to contributeto insect defense. We characterized the first two enzymes of acylsu-crose biosynthesis in the cultivated tomato Solanum lycopersicum.These are type I/IV trichome-expressed BAHD acyltransferases encodedby Solyc12g006330─or S. lycopersicum acylsucrose acyltransferase 1(Sl-ASAT1)─and Solyc04g012020 (Sl-ASAT2). These enzymes wereused—in concert with two previously identified BAHD acyltrans-ferases—to reconstruct the entire cultivated tomato acylsucrose bio-synthetic pathway in vitro using sucrose and acyl-CoA substrates.Comparative genomics and biochemical analysis of ASAT enzymeswere combined with in vitro mutagenesis to identify amino acids thatinfluence CoA ester substrate specificity and contribute to differencesin types of acylsucroses that accumulate in cultivated andwild tomatospecies. This work demonstrates the feasibility of the metabolic en-gineering of these insecticidal metabolites in plants and microbes.

Solanum | glandular trichomes | acylsugar | specialized metabolism |genotype to phenotype

Plants are masters of metabolism, producing hundreds ofthousands of small molecules known as specialized metabolites,which vary widely in structure, abundance, and physical and biologicalproperties. These metabolites tend to be produced by enzymes thatevolve faster than those that produce “central” metabolites such asamino acids, nucleotides, sugars, and cofactors (1–3), and the path-ways and metabolic intermediates involved in biosynthesis of manyspecialized metabolites remain mysterious. Despite the growingavailability of genomic DNA sequences, understanding the geneticand biochemical mechanisms that contribute to this phenotypic di-versity and plasticity presents enduring and major challenges in plantbiochemistry. It is of great interest to understand and manipulate thebiosynthesis of these biologically active molecules.Specialized metabolites typically are produced in a cell- or

tissue-specific manner and are generally limited in their taxonomicdistribution. Glandular secreting trichomes provide an example ofsuch a differentiated structure; these epidermal “hairs” produce avariety of metabolites of importance to humans, including aromaticflavor components (e.g., in hops for beer and Mediterranean herbsfor cooking), psychoactive cannabinoids in Cannabis, and the an-timalarial drug artemisinin in Artemisia annua (4, 5).Some trichome-produced metabolites have documented direct

and indirect antiherbivore activities (4, 6–8). For example, acylsugarsare a group of structurally related specialized metabolites producedin plants of the nightshade family—the Solanaceae (9, 10). Char-acterized examples in the tomato group of Solanum consist of eithera glucose or a sucrose backbone with three to four aliphatic acylgroups of varying carbon numbers ranging from 2 to 12 esterified tothe sugar hydroxyl groups (11–15). Nicotiana attenuata acylsucrosesare at the center of a multitrophic defense interaction where they are

metabolized to volatile fatty acids byManduca sexta larvae, and theseairborne products attract predatory ants (6). Protective propertiesagainst herbivores have made increasing total acylsugars or alteringacyl chain types a target for breeding insect-resistant cultivated to-matoes (Solanum lycopersicum) (16–18). In addition, synthetic su-crose esters mimicking natural acylsugars have been applied as safe,biodegradable insecticidal compounds (19–21) and also have com-mercial value in the food, cosmetic, and pharmaceutical industries(22, 23).Recent work on acylsucrose biosynthesis in trichomes of the to-

mato clade revealed that this relatively closely related group ofplants produces a surprisingly diverse group of acylsucroses intip cells of the long hair-like type I/IV trichomes (14, 24–26). Thegenetic and biochemical mechanisms underlying some of this phe-notypic diversity have begun to be revealed. For example, 2-meth-ylpropanoic acid (iC4) and 3-methylbutanoic acid (iC5) acyl chainvariation in Solanum pennellii accessions is influenced by varia-tion in the activity of a set of truncated isopropylmalate synthase3 (IPMS3) enzymes (26). Differences in patterns of Solanum hab-rochaites acylsucrose acetylation (24) and acyl chain length andposition variation (25) are due to genetic variation in two BAHD[BEAT, AHCT, HCBT, DAT (27, 28)] acyltransferases. Solanumlycopersicum AcylSugar AcylTransferase 3 (Sl-ASAT3) catalyzesacylation of diacylsucroses on the five-membered (furanose) ring tomake triacylsucroses (25). Variant forms of ASAT3 were describedin wild tomato accessions: these use different chain length acyl-CoAesters with diacylsucrose, or acylate the six-membered (pyranose)ring of monoacylated sucrose, demonstrating recent evolutionary

Significance

Throughout the course of human history, plant-derived naturalproducts have been used in medicines, in cooking, as pest controlagents, and in rituals of cultural importance. Plants produce rap-idly diversifying specialized metabolites as protective agents andto mediate interactions with beneficial organisms. In vitro re-construction of the cultivated tomato insect protective acylsucrosebiosynthetic network showed that four acyltransferase enzymesare sufficient to produce the full set of naturally occurring com-pounds. This system enabled identification of simple changes inenzyme structure leading to much of the acylsucrose diversityproduced in epidermal trichomes of wild tomato. These findingswill enable analysis of trichome specialized metabolites through-out the Solanaceae and demonstrate the feasibility of engineeringthese metabolites in plants and microorganisms.

Author contributions: P.F., A.L.S., and R.L.L. designed research; P.F., A.M.M., A.L.S., andX.L. performed research; I.O. and D.Z. contributed new reagents/analytic tools; P.F., A.L.S.,X.L., I.O., A.D.J., D.Z., and R.L.L. analyzed data; and P.F. and R.L.L. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1To whom correspondence should be addressed. Email: lastr@msu.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1517930113 PNAS | Published online December 29, 2015 | E239–E248

PLANTBIOLO

GY

PNASPL

US

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://crossmark.crossref.org/dialog/?doi=10.1073/pnas.1517930113&domain=pdfmailto:lastr@msu.eduhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplementalhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplementalwww.pnas.org/cgi/doi/10.1073/pnas.1517930113

changes leading to diversification of enzymatic activity. The phylo-genetically related ASAT4 enzyme (formerly AT2) is responsiblefor making tetraacylsucroses in S. lycopersicum and S. habrochaitestrichomes by acetylating triacylsucroses using acetyl-CoA (C2-CoA)(29). A variety of loss-of-function alleles of this enzyme are found inpopulations of S. habrochaites from northern Peru and Ecuador(24), reinforcing the idea that ASAT diversification plays an im-portant role in shaping the strong phenotypic diversity in trichomesof wild tomato. Although variation in these enzymes influences theacylsugar phenotypes seen in cultivated and wild tomatoes, much ofthe diversity remains to be explained at a molecular level, includingthe order of acysugar assembly. In addition, identification of theenzymes that produce the ASAT3 substrates is needed to un-derstand cultivated and wild tomato acylsucrose biosynthesis ingreater detail.In this study, two trichome-specific S. lycopersicum BAHD acyl-

transferases that catalyze consecutive reactions to produce thesediacylsucrose intermediates were identified. Sl-ASAT1 was found touse sucrose and various acyl-CoAs to make monoacylsucroses withpyranose R4 acylation. Sl-ASAT2 adds a second acyl chain to the R3position of the ASAT1 S1:5 (iC5R4) product (note that “S” refersto a sucrose backbone, “1:5” indicates the presence of a single iC5ester decoration on sucrose, and the superscript “R4” describes theacylation position) to make the R3,4 diacylsucrose Sl-ASAT3 sub-strates. With the four ASAT enzymes in hand, we reconstructed thecultivated tomato acylsucrose biosynthetic network using sucroseand acyl-CoA substrates. Comparative functional analysis of ASAT2variants from nine wild tomato relatives led to identification of tworesidues affecting ASAT2 iso-C5-CoA (iC5-CoA) and anteiso-C5-CoA (aiC5-CoA; 2-methylbutyryl-CoA) substrate preference. Usinga similar approach, a residue controlling the ability of ASAT3 to uselong chain acyl-CoAs was identified. The in vitro reconstructionsystem allowed us to test the impact of these variant enzymes andchanges in acyl-CoA substrate concentrations on the S. lycopersicumacylsucrose biosynthetic network, demonstrating the value of usingthe in vitro pathway to understand acylsugar evolution. These resultsprovide a model for understanding how small changes in enzymesequence lead to large changes in metabolic diversity.

ResultsIdentification of Two Trichome-Specific BAHD Acyltransferases Involvedin Tomato Acylsugar Biosynthesis. Cultivated tomato produces pri-marily tri- and tetraacylated sucrose esters in the tip cell of the longmulticellular “type I/IV” trichomes (14, 29). Although two trichomeapical cell-expressed BAHD acyltransferases that produce tri- andtetraacylsucroses from diacylsucroses were recently described (25,29), the enzymes that convert sucrose to diacylated sucrose have notbeen previously reported. We used functional genomics approachesto identify candidate enzymes for these earlier steps in the pathway.Bioinformatic analysis revealed 92 genes predicted to encodeBAHD acyltransferase sequences in the S. lycopersicum genome(SI Appendix, Fig. S1A). Twenty-two genes were selected as targetsfor RNAi suppression in M82; these were predicted to encode full-length proteins and had evidence of trichome expression based on aS. lycopersicum trichome EST database (30) (see the proteins in-dicated with arrows in SI Appendix, Fig. S1A). Acylsugars in RNAiT0 primary transgenic plants were analyzed, and lines targetingSolyc12g006330, which we renamed Sl-ASAT1, showed reduction oftotal acylsugar levels compared with the M82 parental line (Fig. 1A)and no detectable acylsucrose intermediates, pointing to a role forSl-ASAT1 in acylsugar biosynthesis in M82.Published studies revealed major differences in trichome metab-

olite accumulation (15, 25, 26, 29) in several introgression lines (ILs),which have regions of the S. pennellii LA0716 genome substituted inplace of the S. lycopersicum M82 genome (31). Rescreening the ILsby liquid chromatography–time of flight mass spectrometry (LC–ToFMS) identified a more subtle phenotype: accumulation of two low-abundance species in IL4-1 not seen in the M82 parent (Fig. 1B).

Positive-ion mode MS fragment ion masses revealed that thesetriacylsucroses—S3:15-P (5, 5, 5) and S3:22-P (5, 5, 12)—presumably have all three acyl chains on the pyranose ring (SIAppendix, Fig. S2A). This is in contrast to the acylsucroses S3:15

35S:ASAT1-intron-1TASA (RNAi) in M82

T 0-1

0T 0

-04

T 0-1

3T 0

-26

T 0-0

6T 0

-02

T 0-3

3T 0

-01

T 0-3

4T 0

-19

T 0-2

0T 0

-07

T 0-3

2T 0

-16

T 0-1

7T 0

-29

T 0-0

3T 0

-08

T 0-1

5T 0

-11

T 0-2

2T 0

-24

T 0-3

1T 0

-05

T 0-2

1T 0

-28

T 0-2

5T 0

-23

T 0-2

7T 0

-18

T 0-0

9T 0

-30

T 0-1

2M

82

0

2

4

6

8

10

Pea

k ar

ea/In

tern

al s

tand

ard

Tota

l Acy

lsug

ars

A

2.00 4.00 6.00 8.00 10.00 12.00 14.000

1000

100 1: TOF MS ES- BPI

5.00e4

1: TOF MS ES- BPI

5.00e4

S3:

22 (5

,5,1

2)S

3:22

(5,5

,12)

S4:

24 (2

,5,5

,12)

S3:

15 (5

,5,5

)S

3:15

(5,5

,5)

S4:

17 (2

,5,5

,5)

S3:1

5-P

(5,5

,5)

S3:2

2-P

(5,5

,12)

S4:

16 (2

,4,5

,5)

Rel

ativ

e ab

unda

nce

%

B

S4:

16 (2

,4,5

,5)

S4:

17 (2

,5,5

,5)

S4:

24 (2

,5,5

,12)

Rel

ativ

e ab

unda

nce

%

IL4-1

M82

Retention time (min)

IL4-

1T 0

-19

T 0-1

7T 0

-04

T 0-0

9T 0

-23

T 0-2

0T 0

-24

T 0-2

1T 0

-26

T 0-2

2T 0

-11

T 0-1

3T 0

-25

T 0-0

6T 0

-01

T 0-2

7T 0

-02

T 0-1

0T 0

-14

T 0-1

5T 0

-12

T 0-0

7T 0

-08

M82

0.0

0.1

0.2

0.3

Sl-ASAT2:ASAT2 in IL4-1

S3:

22-P

/S3:

22P

eak

area

ratio

C

Fig. 1. The S. lycopersicum M82 trichome acylsugar profile is affected bychanges in two BAHD acyltransferases. (A) RNAi suppression of Sl-ASAT1 causesreduction of M82 tomato total acylsugar peak areas. Summed extracted ionchromatogram peak areas divided by internal standard peak areas for all de-tectable acylsugars are shown for each independent T0 primary transformant.M82 data are from five biological replicates ± SD. (B) Introgression of S. pen-nellii chromosome 4 region IL4-1 causes accumulation of acylsucroses S3:15-P(5, 5, 5) and S3:22-P (5, 5, 12), which are not seen in M82 extracts. Negative-ion-mode base-peak intensity LC/MS chromatograms are shown for IL4-1 and M82.Fragment ion masses in positive-ion-mode mass spectra (SI Appendix, Fig. S2A)indicate that these metabolites contain all three acyl chains on one ring. “-P”means that three acyl chains are presumably on the pyranose ring. (C) Sl-ASAT2transgenic expression causes reversal of the IL4-1 mutant phenotype. LC/MSpeak area ratios for the IL4-1–specific S3:22-P and S3:22 acylsugars are shownfor each independent T0 primary transgenic line. Data for IL4-1 and M82 areeach from five biological replicates ± SD.

E240 | www.pnas.org/cgi/doi/10.1073/pnas.1517930113 Fan et al.

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfwww.pnas.org/cgi/doi/10.1073/pnas.1517930113

(5, 5, 5) and S3:22 (5, 5, 12), which are typically seen in the M82parent and which have two acyl chains on the pyranose ring and

one on the furanose ring (SI Appendix, Fig. S2A). The locuscontrolling this acylsugar phenotype was narrowed down to a

Sl-ASAT2

Retention time (min)1.00 2.00 3.00 4.00 5.00

%

0

100

%

0

100

%

0

100

%

0

100

%

0

100 1: TOF MS ES- 471.1+541.2 0.3000Da

7.90e4

1: TOF MS ES- 471.1+555.2 0.3000Da

7.90e4

1: TOF MS ES- 471.1+555.2 0.3000Da

7.90e4

1: TOF MS ES- 471.1+625.3 0.3000Da

7.90e4

1: TOF MS ES- 471.1+653.3 0.3000Da

7.90e4

S-CoA

O

S-CoA

O

S-CoA

O

S-CoAO

S-CoAO

+S1:5 (iC5R4)

S1:5 (iC5R4)C

iC4-CoA

iC5-CoA

aiC5-CoA

nC10-CoA

nC12-CoA

S2:9 (iC5,iC4)

S2:10 (iC5,iC5)

S2:10 (iC5R4,aiC5R3)

S2:15 (iC5,nC10)

S2:17 (iC5,nC12)

1: TOF MS ES- 387.1+461.1 0.2000Da

4.78e4

%

100

6

4

3 2

1’

3’ 4’

6’

Sl-ASAT1

Sp-ASAT1

A B

2.00 3.00 4.00 5.00 6.00 7.00 8.000

R4= iC4, iC5, aiC5, nC10, nC12Retention time (min)

Rel

ativ

e ab

unda

nce Sucrose

S1:5

S1:5 isomer

or

[M+Cl]-

[M+formate]-

Fig. 2. Consecutive in vitro reactions with Sl-ASAT1 and Sl-ASAT2 proteins produce diacylsucroses from sucrose. (A) Result of Sl-ASAT1 enzyme activity assay using sucroseand iC5-CoA as substrates. Negative-ion-mode LC/MS extracted ion chromatograms form/z 387.1 (sucrose; [M+formate]−) and the corresponding ion for reaction productm/z 461.1 (S1:5; [M+Cl]−) are shown. “S” represents an acylsucrose backbone, “1” indicates the number of acyl chains, and “5” corresponds to the total number of carbonsin the acyl chain. The minor peak is a S1:5 (iC5R6) isomer produced by nonenzymatic rearrangement. (B) Summary of the reactions catalyzed by Sl-ASAT1 or Sp-ASAT1(Sopen12g002290) with sucrose and acyl-CoA substrates of different chain lengths. The R4 acylation of sucrose by Sl-ASAT1 was verified by NMR for the monoacylsucrosescontaining an iC5 acyl chain as shown in table S4 of Schilmiller et al. (25) or the nC12 chain (SI Appendix, Table S1). (C) In vitro production of R3-acylated diacylsucroses by Sl-ASAT2 using S1:5 (iC5R4) and different acyl-CoAs (iC4-, iC5-, aiC5-, nC10-, and nC12-CoA) as substrates. Negative-ion-mode LC/MS extracted ion chromatograms for S1:5 anddifferent diacylsucrose products are shown. The S2:10 (iC5R4, aiC5R3) structure was verified by NMR as shown in table S3 of Schilmiller et al. (25).

Fan et al. PNAS | Published online December 29, 2015 | E241

PLANTBIOLO

GY

PNASPL

US

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdf

region containing 64 genes by screening selected backcross in-bred lines (BILs) that have recombination breakpoints on chro-mosome 4 (SI Appendix, Fig. S2B). Among the 64 genes, a strongcandidate gene—Solyc04g012020 (Sl-ASAT2)—and its putativeorthologous gene Sopen04g006140 (Sp-ASAT2) in LA0716 wereidentified based on the prediction that they encode enzymesbelonging to the BAHD acyltransferase family.The in vivo function of Sl-ASAT2 was tested by transgenic

plant experiments. F1 plants generated by crossing IL4-1 to M82showed the M82 acylsugar phenotype, consistent with the hy-pothesis that Sp-ASAT2 is recessive to Sl-ASAT2. Thus, wepredicted that transformation of IL4-1 with an Sl-ASAT2 trans-gene driven by its own promoter would restore IL4-1 acylsugarprofiles to the wild-type M82 phenotype. Indeed, 15 of 23 in-dependent IL4-1 T0 transformant plants showed varying levels ofcomplementation with the peak area ratio of S3:22-P to S3:22restored from 0.23 in IL4-1 to less than 0.05, which is close to theratio observed in M82 (Fig. 1C). Sl-ASAT2 RNAi lines weregenerated in M82, and reduced total acylsugar levels were ob-served for 24 of 29 independent transgenic lines (SI Appendix,Fig. S2C); this is consistent with the hypothesis that Sl-ASAT2plays an in vivo role in wild-type M82 acylsugar production. TheT0 transgenic RNAi (SI Appendix, Fig. S3) and complementationresults (SI Appendix, Fig. S4) were confirmed in T1 progeny linesgenerated by self-crossing.Deep RNA-seq analysis revealed that both Sl-ASAT1 and Sl-

ASAT2 mRNAs are highly enriched in M82 tomato trichomes (26),and trichome-enriched expression of Sl-ASAT1 and Sl-ASAT2 inM82 stem trichomes and Sp-ASAT2 in S. pennellii LA0716 leaftrichomes were confirmed by RT-PCR (SI Appendix, Figs. S1C andS2D). These results were further validated and refined by producingM82 transgenic lines expressing a green fluorescent protein–β-glucuronidase (GFP–GUS) reporter driven by the promoter ofSl-ASAT1 or Sl-ASAT2. Each promoter drove GFP expression inthe tip cell of type I/IV trichomes in stably transformed M82plants (SI Appendix, Figs. S1D and S2E). This pattern is identicalto that of Sl-ASAT3 (25), Sl-ASAT4 (29), and Sl-IPMS3 (26) andsupports the hypothesis that Sl-ASAT1 and Sl-ASAT2 havefunctions in acylsugar biosynthesis.

Sl-ASAT1 and Sl-ASAT2 Work Sequentially to Produce Diacylsucrosesin Vitro. The combination of lack of accumulation of partially acyl-ated acylsugar pathway intermediates in RNAi plants and type I/IVapical cell expression led to the hypothesis that Sl-ASAT1 catalyzesthe first step of acylsucrose biosynthesis. Indeed, recombinant Sl-ASAT1 protein expressed in Escherichia coli converted sucrose andiC5-CoA to make the monoacylsucrose S1:5 (Fig. 2A). Sl-ASAT1can also use other short chain (iC4-CoA, aiC5-CoA) or long-chain(nC10-CoA, nC12-CoA) acyl-CoAs as in vitro acyl donors to makethe respective monoacylsucroses (Fig. 2B). To determine theSl-ASAT1 reaction product structures, S1:5 and S1:12 were purifiedand analyzed using NMR spectroscopy. Acylation at the sucrose R4position was observed for both, with the NMR chemical shift datafor S1:12 (nC12R4) shown in SI Appendix, Table S1 and for S1:5(iC5R4) in table S4 in Schilmiller et al. (25). A chromatographicallyseparable S1:5 isomer, seen as a minor later eluting peak (Fig. 2A),was purified from the original S1:5 product and found to be acylatedat the R6 position; this was demonstrated to be an in vitro artifactcaused by acyl chain rearrangement that is promoted by highpH. The putative ortholog of ASAT1 in S. pennellii LA0716,Sopen12g002290, encodes a protein that shares 97.9% amino acididentity with Sl-ASAT1. This protein also produces monoacylsucroses(Fig. 2B), indicating that ASAT1 has a conserved function in thetomato branch of Solanum. Taken together, our results indicatethat ASAT1 catalyzes the first step of sucrose acylation andproduces an R4 monoacylated sucrose product.Sl-ASAT2 in vitro enzyme activity was tested using protein

expressed in E. coli. We found that Sl-ASAT2 uses S1:5 (iC5R4)—

the product of Sl-ASAT1—and the structurally diverse acyl-CoAdonor substrates iC4-CoA, aiC5-CoA, nC10-CoA, and nC12-CoA tomake the corresponding diacylsucroses (Fig. 2C). The enzyme pro-duced only a small amount of product with iC5-CoA compared withother substrates, which indicates that iC5-CoA is not a preferredsubstrate for Sl-ASAT2. NMR analysis was performed on the pu-rified compound S2:10 (iC5, aiC5) made by sequential reaction ofsucrose with iC5-CoA catalyzed by Sl-ASAT1 and the subsequentreaction of this product with aiC5-CoA catalyzed by Sl-ASAT2.NMR chemical shift data indicate that Sl-ASAT2 added the aiC5group to the R3 position of the sucrose backbone as shown in tableS3 of Schilmiller et al. (25). The diacylsucrose S2:17 (iC5, nC12),made by sequential reaction of sucrose with iC5-CoA (catalyzed bySl-ASAT1) and its product with nC12-CoA (catalyzed by Sl-ASAT2), has the same chromatographic retention time as S2:17(iC5R4, nC12R3) purified from IL11-3 (SI Appendix, Fig. S5),which has NMR-resolved structural information showing the iC5group at the R4 position and the nC12 group at the R3 position(25). This result suggests that Sl-ASAT2 added the long acylchain to the R3 position of the S1:5 monoacylsucrose substrate tomake R3, R4 substituted diacylsucroses. In contrast, Sl-ASAT2does not efficiently use purified S1:12 (nC12R4) as the acyl ac-ceptor using any of the acyl-CoA donors tested.Kinetic analyses were performed to obtain more detailed in-

formation regarding the in vitro properties of the acyltransferases.ASAT1 and ASAT2 used short chain acyl-CoA esters with apparentKm values in the 20- to 50-μM range, and the longer chain nC12-CoA exhibited substrate inhibition of both enzymes (SI Appendix,Table S2 and Fig. S6), as was also previously reported for ASAT3(25). An apparent Km of 2.3 mM was measured for ASAT1 and theacceptor substrate sucrose (SI Appendix, Table S2 and Fig. S6). Inaddition, the ASAT1 enzyme showed evidence of substrate pro-miscuity as was previously documented for other BAHD enzymes(32). The aromatic benzoyl-CoA was efficiently used as a donorsubstrate with sucrose, whereas the negatively charged malonyl-CoAdid not yield detectable product (SI Appendix, Table S3 and Fig. S7).Although ASAT1 had no activity with the monosaccharide glucose,it used the glucose-containing disaccharides cellobiose, lactose,maltose, and trehalose as acceptor substrates when acyl-CoA sub-strates were used, albeit much less efficiently than with sucrose (lessthan 3% of the monoacylsucrose products peak areas) (SI Appendix,Fig. S7). Finally, all four ASAT enzymes were found to be readilyreversible when incubated with their usual products and 100 μMCoA (SI Appendix, Fig. S8), as reported for other BAHD enzymes(33, 34). In addition, S1:5 (iC5R4) tended to hydrolyze to sucroseincubated with CoA in the absence of enzyme.

In Vitro Reconstruction of M82 Acylsucrose Biosynthesis. In vitroreconstruction serves as an excellent approach to validate bio-synthetic pathways and provides the opportunity to explore thefeasibility of metabolic engineering approaches. We asked whetherthe four recombinant enzymes—ASAT1 through ASAT4—coulduse acyl-CoA substrates to produce the acylsucroses extracted fromcultivated tomato. Reconstruction of the M82 acylsucrose bio-synthetic network starting with sucrose was performed by sequen-tially adding the four enzymes and appropriate acyl-CoA substratesin a single tube (Fig. 3A). Because M82 tomato trichomes produceacylsucroses with long or short chains at the R3 position (14),parallel Sl-ASAT2 reactions were performed using either aiC5-CoA or nC12-CoA as the acyl donor substrate and S1:5 (iC5R4),produced by Sl-ASAT1, as the acyl acceptor substrate. The diac-ylsucrose products from the second step were then used as sub-strates for Sl-ASAT3, followed by reaction of these triacylsucroseproducts with C2-CoA and Sl-ASAT4 (chromatograms “S” and“L,” respectively, in Fig. 3B). The sequential assays producedmono, di-, tri-, and tetraacylsucroses, and the resultant S3:15 (5, 5,5), S4:17 (2, 5, 5, 5), S3:22 (5, 5, 12), and S4:24 (2, 5, 5, 12) hadchromatographic retention times identical to the M82 trichome

E242 | www.pnas.org/cgi/doi/10.1073/pnas.1517930113 Fan et al.

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfwww.pnas.org/cgi/doi/10.1073/pnas.1517930113

acylsucroses (Fig. 3B). The high-collision-energy negative-ion modemass spectra of these four compounds revealed fragment ion spectraindistinguishable from the tri- and tetraacylsucroses extracted fromM82 that were previously documented by Schilmiller et al. (29).Although the sequential reconstruction experiments provided

strong support that these four enzymes are sufficient to produce themajor products in leaf-surface extracts from sucrose and CoA es-ters, simultaneous presence of enzymes and substrates presumablymore closely reflects the in vivo reaction conditions. The four en-zymes, sucrose, and acyl-CoA substrates were added simultaneouslyfor these mixed assays: iC5-CoA, which is the acyl donor for Sl-ASAT1 and Sl-ASAT3, and C2-CoA, the acyl donor for Sl-ASAT4,were added to the mixed reactions, together with varied types ofacyl-CoA substrates for Sl-ASAT2, either with short (iC4-CoA,aiC5-CoA) or with long acyl chains (nC10-CoA, nC12-CoA). Col-lectively, the reactions produced the full set of tri- and tetraa-cylsucroses that accumulate in vivo, including the iC4- and nC10-containing compounds that are relatively minor components intomato plants (SI Appendix, Fig. S9 A and B). All in vitro-producedtri- and tetraacylsucroses shared the samem/z and chromatographicretention times with the corresponding acylsucroses extracted fromM82 trichomes (SI Appendix, Fig. S9 A and B). The in vitro pro-duction of the full set of M82 acylsucroses provides strong evidencethat Sl-ASAT1, Sl-ASAT2, Sl-ASAT3, and Sl-ASAT4 are the majorenzymes in the acylsugar metabolic network in the apical cell ofcultivated tomato type I/IV trichomes.

In Vitro System Responds to Changes in Acyl-CoA Precursor Availability.Acylsugar phenotypic diversity was observed for various wild tomatospecies and for accessions within S. pennellii and S. habrochaites(14, 24–26). For example, we recently demonstrated that intro-gression of a region of S. pennellii LA0716 from the top of chro-mosome 8 (IL8-1-1) causes increased accumulation of iC4-containing acylsucroses due to introduction of a gene encoding atruncated Leu feedback-insensitive isopropylmalate synthase(IPMS)-like enzyme (Sp-IPMS3) (15, 26). Unlike the enzymaticallyactive Sl-IPMS3, this S. pennellii isoform has a defect in in vitroIPMS activity, and we hypothesized that this defect blocks pro-duction of iC5-CoA and diverts its precursor to iC4-CoA, leading toaccumulation of iC4 acyl chain-containing acylsugars (26). We usedthe in vitro reconstructed S. lycopersicum acylsucrose pathway totest the hypothesis that the IL8-1-1 high-iC4 acylsucrose phenotypeis due to an increase in the ratio of iC4-CoA to iC5-CoA substrates.We varied the relative amounts of iC4-CoA and iC5-CoA added toin vitro reactions that contained all substrates and enzymes addedsimultaneously (SI Appendix, Fig. S9C). As shown in Fig. 4, both theabsolute and relative amounts of the four C4-containing metabolitesincreased as the percentage of total iC4-CoA [100 × iC4-CoA/(iC4-CoA + iC5-CoA)] went from 12.5% to 50%. Increases in S3:20(4, 4, 12) and S4:22 (2, 4, 4, 12) were especially sensitive to iC4-CoAsubstrate availability, presumably because their synthesis is com-pletely dependent on this substrate. These in vitro results are con-sistent with the hypothesis that provision of acyl-CoA estersinfluences the overall composition of acylsucroses in trichomes of

A

B

Fig. 3. In vitro reconstruction of production of the fourmajorM82 acylsugars by sequential addition of ASAT1, ASAT2, ASAT3, and ASAT4 using sucrose and acyl-CoAsubstrates. (A) Schematic representation of sequential enzyme assays. The first reaction used sucrose and iC5-CoA to make the monoacylsucrose product S1:5 (iC5R4)catalyzed by Sl-ASAT1. After enzyme heat inactivation, the acyl-CoA short-chain aiC5-CoA (S) or long-chain nC12-CoA (L) was added with Sl-ASAT2, and the corre-sponding diacylsucroses were produced. Next, Sl-ASAT3 and iC5-CoA were added followed by Sl-ASAT4 and C2-CoA to produce tri- and tetraacylsucroses, respectively.(B) LC/MS-extracted ion chromatogram analysis of the products of the sequential reactions. “S” and “L” represent the sequential assay for which the short-chain aiC5-CoA or long-chain nC12-CoA, respectively, were used with Sl-ASAT2. Color coding of the product peaks and names corresponds to enzyme names and acyl chains in A.Relative abundance for each chromatogram is based on setting the major peak to 100%; M82: S4:17 (2, 5, 5, 5); L: S3:22 (5, 5, 12); S: S4:17 (2, 5, 5, 5).

Fan et al. PNAS | Published online December 29, 2015 | E243

PLANTBIOLO

GY

PNASPL

US

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdf

S. lycopersicum. They validate the idea that differences in iC4-con-taining acylsugars in the IL8-1-1 introgression line and S. pennelliiaccessions are due to changes in iC5- and iC4-CoA availabilitycaused by differences in IPMS3 enzyme structure and function (26).

Comparative Analysis of Natural Variant Enzymes Reveals Amino AcidResidues Affecting ASAT2 Acyl-CoA Substrate Specificity. Differ-ences in ASAT3 and ASAT4 enzyme activities were previouslydemonstrated to contribute to metabolite diversity in S. hab-rochaites and S. pennellii (24, 25). Sp-ASAT2, which shares95.1% protein identity with Sl-ASAT2, showed barely detectableactivity with the Sl-ASAT2 acyl acceptor substrates S1:5 (iC5R4)and different acyl-CoAs. This observation suggests that di-versification of ASAT2 substrate selectivity could also influencethe acylsugar diversity observed in various wild tomato species(25). To test this prediction, nine putative ASAT2 orthologswere cloned from wild tomato relatives that are phylogeneticallypositioned between cultivated tomato and S. pennellii, usingprimers based on published genomic resequencing data (35).Enzyme activities were then tested using S1:5 (iC5R4) and dif-ferent acyl-CoAs as substrates (Fig. 5A). The ASAT2 isoforms ofthe closest relatives of cultivated tomato showed similar acyl-CoA substrate preference to that of Sl-ASAT2 when using S1:5(iC5R4) as the acyl acceptor substrate: Solanum pimpinellifoliumLA1578 and Solanum galapagense LA1401 had barely detectableactivities with iC5-CoA as a donor substrate, but used a variety ofother acyl-CoAs (Fig. 5A). This in vitro activity is consistent withthe lack of iC5 acylation at the R3 position of cultivated tomatoacylsucroses (14). In contrast, ASAT2 isoforms from theremaining species used iC5-CoA more efficiently (Fig. 5A). Aninteresting exception is that aiC5-CoA was a poor substrate forthe S. habrochaites LA1718-ASAT2 isoform (Fig. 5A). The rel-atively high protein sequence identity of ASAT2 variants allowedus to recognize a small number of amino acid residues thatcorrelate with differences in ASAT2 variant substrate specificity(Fig. 5A; SI Appendix, Fig. S10). This analysis led to identifica-tion of four candidate residues associated with the preferencefor iC5-CoA and six candidate residues that correlate withaiC5-CoA utilization.We reasoned that variant amino acids near to the predicted acyl-

CoA–binding pocket were the strongest candidates for affecting acyl

donor specificity. Protein structure homology modeling was per-formed for Sl-ASAT2 using a trichothecene 3-O-acetyltransferase−acyl CoA complex (3B2S) (36). This analysis revealed that—of thecandidate residues possibly affecting iC5-CoA and aiC5-CoA sub-strate specificity identified in the comparative sequence analysis (Fig.5A)—Phe408 and Ile44 were closest to the putative acyl-CoA–bindingpocket (Fig. 5B).Site-directed mutagenesis and in vitro enzyme assays were used to

test the hypothesis that these candidate amino acids influence acyl-CoA substrate preference. Consistent with expectation, mutagenesisof Sl-ASAT2 from Phe408 to Val408 increased the ability of the en-zyme to use iC5-CoA to produce S2:10 (5, 5) (Fig. 6): the variantenzyme has an apparent Km value for iC5-CoA of 26.3 ± 3.2 μM.This is similar to the apparent Km value 27.1 ± 5.2 μM for iC5-CoAfor Solanum arcanum LA2172 ASAT2, an isoform that efficientlyuses iC5-CoA as a substrate (SI Appendix, Fig. S11). The reciprocalmutagenesis change—with S. arcanum LA2172 ASAT2 mutatedfrom Val408 to Phe408—led to an enzyme with undetectable activityfor iC5-CoA (Fig. 6) while retaining the ability to use aiC5-CoA andnC12-CoA as substrates. In contrast, mutating position 43 of theS. habrochaites LA1718 ASAT2 from Leu to Ile conferred the abilityto use aiC5-CoA as a substrate (Fig. 6) with the apparent Km valueobserved of 159 ± 23 μM (SI Appendix, Fig. S11). The conversion ofIle to Leu at amino acid 44 of Sl-ASAT2 abolishes its ability to useaiC5-CoA as a substrate (Fig. 6; SI Appendix, Fig. S11) without af-fecting the ability to accept the long-chain nC12-CoA as a substrate.Taken together, this comparative biochemical analysis identified tworesidues affecting ASAT2 acyl-CoA substrate specificity.

Identification of an ASAT3 Residue Associated with Long-Chain Acylationof the Acylsucrose Furanose Ring. In contrast to acylsucroses extractedfrom S. lycopersicum M82, which have an iC5 acyl chain at the R3′position of the furanose ring, some S. habrochaites accessions pro-duce acylsucroses with long chains (C10–C12) at this position (14,25). These differences correlate with variation in in vitro ASAT3activities from those accessions (25). We used the comparativebiochemical approach to identify amino acids responsible for thesedifferences in ASAT3 acyl-CoA substrate specificities. As shown inSI Appendix, Fig. S12, 19 residues correlated with activity differencesbetween Sl-ASAT3 and the long-chain acyl-CoA−using ASAT3variants from S. habrochaites LA1777 and LA1731. Homologymodeling of Sl-ASAT3 with the trichothecene 3-O-acetyltransferase−acyl CoA complex 3B2S structure suggested that amino acids 35 and41 are close to the acyl-CoA–binding pocket (Fig. 7A). Indeed,mutation of Tyr41—found in short acyl chain-using enzymes—to theCys41 found in the S. habrochaites Sh-ASAT3-F enzymes convertedSl-ASAT3 into an enzyme that adds nC12 to the furanose ring ofS2:10 (5, 5) to produce S3:22 (5, 5, 12), a product not seen in M82trichome extracts (Fig. 7 B and C). Sl-ASAT3_Y41C had an ap-parent Km value of 1.1 ± 0.5 μM for nC12-CoA and also exhibitedsubstrate inhibition by nC12-CoA with an apparent Ki value of 4.5 ±2.0 μM (SI Appendix, Fig. S11). Again, these results are similar tovalues observed for LA1777 Sh-ASAT3-F (25).The ability of a single amino acid substitution to change Sl-

ASAT3 into a Sh-ASAT3–type activity suggested that we couldtransform the S. lycopersicummetabolic network into that seen inS. habrochaites LA1777 by substitution of this variant mutantenzyme. Indeed, use of this Sl-ASAT3_Y41C variant enzyme inthe sequential in vitro metabolic network reconstruction systemled to accumulation of four different acylsucroses containing C10acylations on the furanose ring (SI Appendix, Fig. S13). Thesefour in vitro-synthesized acylsucroses shared the same chro-matographic retention time with the corresponding acylsugarsextracted from LA1777 leaf trichomes (SI Appendix, Fig. S13A),and positive-mode high-collision-energy mass spectra revealedlong-chain acylation on the furanose ring (SI Appendix, Fig.S13B). These results are consistent with the hypothesis that asingle amino acid change explains some of the major differences

01234

10

20

30

40

S4:23 (2,4,5,12)S3:21 (4,5,12)S4:22 (2,4,4,12)S3:20 (4,4,12)

Pea

k ar

ea/In

tern

al s

tand

ard

Fo

ld c

hang

e

12.5% 25% 37.5% 50%[100 × iC4-CoA/(iC4-CoA + iC5-CoA)]

Fig. 4. Varying the ratio of iC4-CoA and iC5-CoA causes changes in accumu-lation of iC4-containing acylsucroses in vitro. Quantification of LC/MS-extractedion chromatogram peak areas divided by internal standard peak areas for thein vitro-produced acylsucroses S3:20 (4, 4, 12), S4:22 (2, 4, 4, 12), S3:21 (4, 5, 12),and S4:23 (2, 4, 5, 12) showed that the accumulation of iC4-containing acyl-sucroses increased as the percentage of total iC4-CoA [100 × iC4-CoA/(iC4-CoA +iC5-CoA)] went from 12.5% to 50%. The fold change for each acylsucrose isshown with replicates ± SE.

E244 | www.pnas.org/cgi/doi/10.1073/pnas.1517930113 Fan et al.

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfwww.pnas.org/cgi/doi/10.1073/pnas.1517930113

in acylsucrose structures observed among accessions of S. hab-rochaites and between S. lycopersicum and S. habrochaites.

DiscussionDuring the past 10 y the glandular-secreting trichomes of cultivatedand wild tomatoes have emerged as a model for studying the evo-lution of previously uncharacterized specialized metabolic networks,including terpenes in type VI glands (37–41), methylated flavonoids(42), and acylsugars in type I/IV trichomes (24–26, 29). These studiesrevealed strong metabolic diversity both within tomato species andacross the tomato clade of the genus Solanum, leading us to studythe genetic and biochemical mechanisms associated with this evo-lutionarily rapid diversification. In this work we describe the char-acterization of ASAT1 and ASAT2 BAHD acyltransferases thatcatalyze the first two steps of acylsucrose biosynthesis from sucrose inS. lycopersicum. These enzymes were used in combination with thepreviously described Sl-ASAT3 and Sl-ASAT4 to reconstitute thesynthesis of the major S. lycopersicum tri- and tetraacylsucroses invitro. When combined with phenotypic analyses and demonstrationthat these enzymes are expressed in the acylsucrose-producing typeI/IV trichome apical cells, these in vitro reconstruction experimentsconfirm the roles of these enzymes in acylsugar synthesis.ASAT1, which catalyzes the first acylation step in tomato acyl-

sucrose biosynthesis, has several features that distinguish it fromother characterized ASAT enzymes. First, its RNAi lines do notaccumulate detectable acylsucrose intermediates, consistent with

its role in catalyzing the first step of the biosynthetic pathway.Second, in contrast to ASAT2-ASAT4, we found no in vivo phe-notypic or in vitro enzyme activity evidence for ASAT1 geneticvariation leading to altered enzyme activities across the tomatoclade, which shares a last common ancestor several million years ago(43). As acylsugar metabolic networks are characterized in Sol-anaceae species outside of the tomato group, it will be interesting tolearn whether this enzyme was also conserved as the committing stepover tens of millions of years of evolution.As seen for other BAHD acyltransferases (for example, see ref.

32), Sl-ASAT1 showed evidence of promiscuity in vitro, acylatingsucrose at the R4 position using acyl-CoAs with different acyl chainlengths (iC4, iC5, nC10, iC12) or branching patterns (subterminallybranched aiC5 and terminally branched iC5). Despite the abilityof the enzyme to use longer chain CoAs, only iC4 and iC5 chainsthus far have been reported at the R4 position of acylsucroses ofS. lycopersicumM82 and three different accessions of S. habrochaites(14). This seeming conflict likely arises because the next enzyme—Sl-ASAT2—is unable to use the S1:12 (nC12R4) ASAT1 products assubstrates to produce diacylsucroses. This presumably causes pro-duction of dead-end monoacylated products that may be degradedby an acylsucrose acylhydrolase that cleaves the R4 position ofacylsucroses—or converted to compounds of sufficiently differentstructure that they are not detected by our analytical methods.In contrast to the uniform composition of the R4 acyl groups

of acylsucroses extracted from S. lycopersicum, the R3 position

30 106 408 44 185 257 295 298 320

P E F I P T L H A

P E F I P T L H A

P E F I P T L H A

H K V I P T L H A

H K V I P T L H A

H K V M P T L H A

H K V M P T L H A

H K V M P T L H A

H K V I P T L H A

H K V L A I V N G

iC5-CoA

- + +

- + +

- + +

+ + +

+ + +

+ + +

+ + +

+ + +

+ + +

+ - +

aiC5-CoA

nC12-CoA

S.lycopersicum M82

S.pimpinellifolium LA1578

S.galapagense LA1401

S.neorickii LA2133

S.arcanum LA2172

S.chilense LA1969

S.peruvianum LA1278

S.corneliomulleri LA0107

S.habrochaites LA1777

S.habrochaites LA1718

Sotub04g011000

98

57

71

91 97

99

8798

0.02

S1:5 (iC5R4) as acyl acceptorResidue positionsEnzyme activityA

BP-185

P-30

H-298

L-295

I-44

E-106F-408

T-257

A-320

Acyl-CoA

ASAT2

Acyl-CoA

V

V

V

I

I

I

I

I

I

I

104

V-104

Fig. 5. ASAT2 amino acid polymorphisms and variation in acyl-CoA substrate specificities. (A) Alignment of amino acid polymorphisms of ASAT2 putativeorthologs with acyl-CoA substrate specificities. ASAT2 phylogenetic tree obtained using protein sequences. “+”means good enzyme activities with detectableproduct peaks; “−” means no or barely detectable enzyme activity. Polymorphisms at amino acid residues 30, 104, 106, and 408 correspond with differentASAT2 activity using iC5-CoA as the substrate. LA1718 ASAT2, which is the only enzyme that does not efficiently use aiC5-CoA, has residues at positions 44,185, 257, 295, 298, and 320 that are unique compared with other ASAT2. (B, Left) The alignment of homology-modeled cartoon structure of Sl-ASAT2(orange) superimposed upon the crystallographic cartoon structure of a trichothecene 3-O-acetyltransferase−acyl CoA complex (3B2S) (gray) (36), which hasacyl-CoA (red) and protein crystallized in a complex. (Right) Highlights of Sl-ASAT2 residues that correlate with differences in acyl-CoA substrate specificityand model near the 3B2S acyl-CoA.

Fan et al. PNAS | Published online December 29, 2015 | E245

PLANTBIOLO

GY

PNASPL

US

Dow

nloa

ded

by g

uest

on

June

14,

202

1

acyl chain length and branching pattern is quite variable, withiC4, aiC5, iC10, nC10, and nC12 acyl chains observed (14, 25).This in vivo diversity correlates well with the ability of Sl-ASAT2to acylate S1:5 (iC5R4) at the R3 position using diverse acyl-CoAs. The next two enzymes—Sl-ASAT3 and Sl-ASAT4—canuse acyl acceptor substrates with varied R3 substitutions in vitroto produce the major acylsucroses found in M82 (Fig. 3 and SIAppendix, Fig. S9).Although we have reconstituted a four-enzyme pathway in culti-

vated tomato, open questions remain regarding the pathway in dif-ferent tomato species. For example, what enzyme(s) are responsiblefor production of the acylsugars containing all acylchains on thepyranose ring in IL4-1 (SI Appendix, Fig. S2A)? How does IL4-1produce most of the same major acylsugars as M82 despite the in-trogression of the putative SpASAT2 ortholog, which does not useS1:5 (iC5R4) as substrate. These suggest that unknown enzymaticactivities remain to be discovered to add to the current acylsugarmetabolic network.Understanding the genetic mechanisms leading to changes in the

types of specialized metabolites that accumulate over evolutionarytime and the biochemical basis for substrate specificity and enzy-matic promiscuity is central to plant improvement efforts. Enzymestructure and function analysis depends upon identification of aminoacids that influence activity, but even using directed evolution and“semi-rational” approaches to protein engineering by in vitro mu-tagenesis can be quite time-consuming, requiring construction andtesting of large numbers of single or multiple amino acid variants(44–47) and the availability of suitable substrates.We took advantage of existing acylsucrose variation within

accessions of S. habrochaites and other tomato species to seekASAT2 and ASAT3 amino acids that are responsible for thesephenotypic differences. Comparisons of primary sequence vari-ation with in vitro assays performed using a variety of acyl-CoAsubstrates revealed relatively small numbers of amino acids ascandidates for influencing substrate specificity. These candidateresidues were screened further for those that might be positionedto interact with the CoA substrates using homology modeling. Inall three cases the approach led to identification of amino acid

positions that impacted substrate utilization. ASAT2 Val/Phe408

strongly influenced utilization of terminally branched iC5-CoAas a donor substrate. The second ASAT2 residue identifiedin this analysis, Leu43 found in S. habrochaites LA1718 (Ile44 inSl-ASAT2), influences utilization of aiC5-CoA as donor sub-strate. The ability to facilely identify a single amino acid residuethat influences discrimination between such structurally similarsubstrates (terminally and subterminally branched C5-CoAs)speaks to the power of this approach. Finally, comparison of Sh-ASAT3 variants that can add a longer chain to the R3′ position to

44

ASAT2

0.0 0.5 1.0

LA1718-ASAT2_L43I

LA1718-ASAT2

Sl-ASAT2_I44L

Sl-ASAT2

Peak area/ Internal standard0.0 0.1 0.2 0.3

LA2172-ASAT2_V408F

LA2172-ASAT2

Sl-ASAT2_F408V

Sl-ASAT2

Peak area/ Internal standard

Ile

Ile

Leu

Leu

Phe

Val

Phe

Val

408

S2:10 (iC5, aiC5) S2:10 (iC5, iC5)

Fig. 6. Single-residue substitutions affect ASAT2 substrate preference foriC5-CoA or aiC5-CoA. (Left) The amount of S2:10 (iC5,aiC5) produced byASAT2 of varying structures using S1:5 (iC5R4) and aiC5-CoA as substrates.The ASAT2 variants with Ile44 have higher activity using aiC5-CoA as sub-strate than those with Leu44. (Right) The amount of S2:10 (iC5,iC5) producedby different ASAT2s using S1:5 (iC5R4) and iC5-CoA as substrates. The Val408

ASAT2 variants have higher activity using iC5-CoA as substrate than thosewith Phe408. Average LC/MS peak areas divided by internal standard peakareas of the S2:10 products with ±SD were calculated from three replicates.

Retention time (min)1.00 2.00 3.00 4.00 5.00 6.00

%

0

100 1: TOF MS ES- 555.2+737.4 0.3000Da

1.84e5

S2:10 (iC5R4, aiC5R3)

S3:22 (5, 5, 12)

m/z0 100 200 300 400 500 600 700 800

%

0

100 715.45

345.26

+

[M+Na]+

ESI+ 40V

A

B

C

SlASAT3_Y41C + S2:10 + nC12-CoA

Fig. 7. ASAT3 position 41 polymorphisms control acyl chain length preference.(A) The predicted homology-modeled structure of Sl-ASAT3 superimposed onthe 3B2S structure. The Sl-ASAT3 structure (green ribbon) and acyl-CoA (red) areshown. The amino acids highlighted as blue or orange are the residues inSl-ASAT3 that are not found in two S. habrochaites ASAT3 isoforms that canuse long-chain acyl-CoAs as substrates. The residue Y41 is shown in orange.(B) Sl-ASAT3_Y41C uses purified S2:10 (iC5R4, aiC5R3) and nC12-CoA as substratesto produce a triacylsucrose S3:22 (5, 5, 12). LC/MS-extracted ion chromatogramsare shown form/z 555.2 (S2:10) andm/z 737.4 (S3:22). (C) Positive-ion-modemassspectrum with a collision potential of 40 V is shown for S3:22 (5, 5, 12). Presenceof the fragment ion withm/z 345 indicates that the long acyl chain was added tothe furanose ring of sucrose.

E246 | www.pnas.org/cgi/doi/10.1073/pnas.1517930113 Fan et al.

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfwww.pnas.org/cgi/doi/10.1073/pnas.1517930113

those with a preference for short-chain acyl-CoAs revealed thatsubstitution of Cys in place of Tyr at position 41 is sufficientto transform substrate specificity. The comparative genomic/biochemical approach coupled with in vitro analysis has alsobeen applied to identify key residues of enzymes involved interpene (48) and artemisinin biosynthesis (49). This approachhas the potential to be generally applicable to structure–functionstudies of any enzymes that make products that are variableacross related populations or related species of plants.The results from this study have implications for engineering of

acylsugars and related compounds. The ability to test the impact ofcombinations of BAHD acyltransferase isoforms on product typesshould inform breeding and genome-editing approaches to modifybiotic stress tolerance of tomato and other Solanaceae plants. Itwill also permit regiospecific synthesis of compounds for activityscreening or large-scale production by synthetic biology ap-proaches—for example, novel pesticides, antimicrobials, phar-maceutical excipients, and emulsifiers.

Materials and MethodsTomato Transformation. Transformation of tomatoM82and IL4-1was performedusing Agrobacterium tumefaciens strain AGL0 (50). BAHD acyltransferase genesuppression was performed by cloning a fragment of each gene from M82 ge-nomic DNA into the pHELLSGATE12 binary vector (51) and transforming M82plants. Sl-ASAT2 under the control of its native promoter was cloned into pK7WG(52) and transformed into IL4-1. For in planta reporter gene analysis, the pro-moter regions of Sl-ASAT1 and Sl-ASAT2 were cloned into pKGWFS7 (52) andtransformed into M82. Detailed information is in SI Appendix, Materials andMethods.

Plant Trichome Acylsugar Extraction. Trichome acylsugars were extracted fromthe youngest expanded leaves of 3- to 4-week-old plants (15). A single leafletwas dipped in 1 mL of extraction solvent, which contained acetonitrile/iso-propanol/water (3:3:2) with 0.1% formic acid and 10 μM propyl 4-hydrox-ybenzoate as internal standard, and the mixture was gently agitated for2 min.

Mapping the IL4-1 Acylsugar Locus Using Backcross Inbred Lines. The BILpopulation was constructed and genotyped as previously described (26). BILs thathave chromosome introgression regions covering the IL4-1 and IL4-2 overlap re-gion were selected for acylsugar profile screening using LC/MS. A set of BILswith recombination breakpoints in the overlap region were tested for theiracylsugar phenotypes. Two SNP markers and one self-designed insertion/deletionmarker (forward primer 5′-TAAAACCTTAGAATCGTTCTCGT-3′ and reverse primer5′-AAATGATCACTGAAGAATTTCCA-3′) were used for further mapping analysis.

Amplification of ASAT2 Putative Orthologs from Wild Tomatoes. Accessionsof S. pimpinellifolium (LA1578), S. galapagense (LA1401), Solanum neorickii(LA2133), S. arcanum (LA2172), Solanum chilense (LA1969), Solanum peruvia-num (LA1278), Solanum corneliomulleri (LA0107), and S. habrochaites (LA1718,

LA1777) were obtained from the C. M. Rick Tomato Genetic Resource Center(tgrc.ucdavis.edu). ASAT2-coding sequences were amplified using cDNA tran-scribed from RNA extracted from 5-wk-old plant leaf tissues as the templatesand using primers that were designed based on the wild tomato whole-ge-nome resequencing data (35). Details of ASAT2 sequencing and determinationof orthology and GenBank accession numbers for different ASAT2 genes aredescribed in SI Appendix, Materials and Methods.

Protein Expression and Enzyme Assay. Recombinant proteins were generatedusing E. coli as the host for enzyme assays. The full-length ORF sequence wascloned into pET28b. ASAT1 and ASAT2 enzyme assays were performed byincubating purified recombinant proteins in 30 μL of 50 mM ammoniumacetate (pH 6.0) buffer with 100 μM acyl-CoA and an acylsucrose acceptor.Methods used to determine the apparent Km value for different acyl-CoAsubstrates were performed as previously described (25). The detailed stepsfor protein expression and enzyme activity assays are in SI Appendix, Ma-terials and Methods.

LC/MS Analysis of Acylsugars. The trichome acylsugars extracted from leaveswere analyzed using a Shimadzu LC-20ADHPLC system connected to aWatersLCT Premier ToF MS. Enzyme assay samples were analyzed using a WatersAcquity UPLC system connected to Waters Xevo G2-S QToF LC/MS. DetailedLC/MS methods are in SI Appendix, Materials and Methods.

Homology Structural Modeling of ASAT2. The Phyre web-based protein homol-ogy/analogy recognition engine (53) was used to predict the tertiary structures ofSl-ASAT2 and Sl-ASAT3. The trichothecene 3-O-acetyltransferase structure (ProteinData Bank ID: 3B2S) was used as a template to overlay with Sl-ASAT2 and Sl-ASAT3modeled structure and displayed using PyMOL (Version 1.7.4 Schrödinger).

Site-Directed Mutagenesis. Site-directed mutations were created by PCR-based plasmid amplification using the Q5 Site-Directed Mutagenesis Kit(NEB). The primers used to introduce mutations were designed based on theweb-based software NEBaseChanger (Version 1.2.2, NEB) and are listed in SIAppendix, Materials and Methods. The presence of the mutations wasconfirmed by DNA sequencing.

ACKNOWLEDGMENTS. We thank members of the Solanum Trichome Projectfor their contributions to this work, especially Kathleen Imre for help withtomato transformation; Jing Ning, Gaurav Moghe, and Bryan Leong for helpfulcomments on the manuscript; and Banibrata Ghosh for developing the LCelution methods for the large-scale purification of acylsucroses. We acknowl-edge the Michigan State University Center for Advanced Microscopy and theResearch Technology Support Facility (Mass Spectrometry and MetabolomicsCore). Work in the A.D.J. and R.L.L. laboratories was funded by National ScienceFoundation Grant IOS-1025636; A.D.J. acknowledges support from MichiganAgBioResearch Project MICL02143; research in the D.Z. laboratory was sup-ported by the European Research Council advanced grant entitled YIELD.A.M.M. was supported by Plant Genomics at Michigan State University SummerResearch Experience for Undergraduates Program and by an American Societyof Plant Biologists Summer Undergraduate Research Award. Wild tomato spe-cies seeds used in this study were obtained from the C. M. Rick Tomato GeneticsResource Center (University of California, Davis).

1. Chae L, Kim T, Nilo-Poyanco R, Rhee SY (2014) Genomic signatures of specialized

metabolism in plants. Science 344(6183):510–513.2. Weng J-K, Philippe RN, Noel JP (2012) The rise of chemodiversity in plants. Science

336(6089):1667–1670.3. Milo R, Last RL (2012) Achieving diversity in the face of constraints: Lessons from

metabolism. Science 336(6089):1663–1667.4. Werker E (2000) Trichome diversity and development. Adv Bot Res 31:1–35.5. Schilmiller AL, Last RL, Pichersky E (2008) Harnessing plant trichome biochemistry for

the production of useful compounds. Plant J 54(4):702–711.6. Weinhold A, Baldwin IT (2011) Trichome-derived O-acyl sugars are a first meal for

caterpillars that tags them for predation. Proc Natl Acad Sci USA 108(19):7855–7859.7. Shepherd RW, Bass WT, Houtz RL, Wagner GJ (2005) Phylloplanins of tobacco are defensive

proteins deployed on aerial surfaces by short glandular trichomes. Plant Cell 17(6):1851–1861.8. Simmons AT, Gurr GM (2005) Trichomes of Lycopersicon species and their hybrids:

Effects on pests and natural enemies. Agric For Entomol 7(4):265–276.9. Rodriguez AE, Tingey WM, Mutschler MA (1993) Acylsugars of Lycopersicon pennellii

deter settling and feeding of the green peach aphid. J Econ Ent 86(1):34–39.10. Juvik JA, Shapiro JA, Young TE, Mutschler MA (1994) Acylglucoses from wild toma-

toes alter behavior and reduce growth and survival of Helicoverpa zea and Spo-

doptera exigua. J Econ Ent 87(2):482–492.11. King RR, Calhoun LA, Singh RP (1988) 3,4-Di-O- and 2,3,4-tri-O-acylated glucose esters

from the glandular trichomes of nontuberous Solanum species. Phytochemistry

27(12):3765–3768.

12. King RR, Pelletier Y, Singh RP, Calhoun LA (1986) 3,4-Di-O-isobutyryl-6-O-caprylsucrose: Themajor component of a novel sucrose ester complex from the type B glandular trichomes ofSolanum berthaultii hawkes (Pl 473340). J Chem Soc Chem Commun 14:1078–1079.

13. Shapiro JA, Steffens JC, Mutschler MA (1994) Acylsugars of the wild tomato Lyco-persicon pennellii in relation to geographic distribution of the species. Biochem SystEcol 22(6):545–561.

14. Ghosh B, Westbrook TC, Jones AD (2014) Comparative structural profiling of trichomespecialized metabolites in tomato (Solanum lycopersicum) and S. habrochaites: acyl-sugar profiles revealed by UHPLC/MS and NMR. Metabolomics 10(3):496–507.

15. Schilmiller A, et al. (2010) Mass spectrometry screening reveals widespread diversity in trichomespecialized metabolites of tomato chromosomal substitution lines. Plant J 62(3):391–403.

16. Maluf WR, et al. (2010) Broad-spectrum arthropod resistance in hybrids betweenhigh-and low-acylsugar tomato lines. Crop Sci 50(2):439–450.

17. Leckie BM, De Jong DM, Mutschler MA (2012) Quantitative trait loci increasingacylsugars in tomato breeding lines and their impacts on silverleaf whiteflies. MolBreed 30(4):1621–1634.

18. Leckie B, et al. (2014) Quantitative trait loci regulating the fatty acid profile ofacylsugars in tomato. Mol Breed 34:1201–1213.

19. Xia Y, Johnson AW, Chortyk OT (1998) Toxicity of synthetic sucrose esters against thetobacco aphid (Homoptera: Aphididae). J Entomol Sci 33(3):292–299.

20. Puterka GJ, Farone W, Palmer T, Barrington A (2003) Structure-function relationshipsaffecting the insecticidal and miticidal activity of sugar esters. J Econ Entomol 96(3):636–644.

Fan et al. PNAS | Published online December 29, 2015 | E247

PLANTBIOLO

GY

PNASPL

US

Dow

nloa

ded

by g

uest

on

June

14,

202

1

http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://tgrc.ucdavis.edu/http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdfhttp://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1517930113/-/DCSupplemental/pnas.1517930113.sapp.pdf

21. Chortyk OT, Kays SJ, Teng Q (1997) Characterization of insecticidal sugar esters ofPetunia. J Agric Food Chem 45(1):270–275.

22. Hill K, Rhode O (1999) Sugar-based surfactants for consumer products and technicalapplications. Fett Lipid 101(1):25–33.

23. Lerk PC, Sucker HH, Eicke HF (1996) Micellization and solubilization behavior of su-crose laurate, a new pharmaceutical excipient. Pharm Dev Technol 1(1):27–36.

24. Kim J, et al. (2012) Striking natural diversity in glandular trichome acylsugar com-position is shaped by variation at the Acyltransferase2 locus in the wild tomato So-lanum habrochaites. Plant Physiol 160(4):1854–1870.

25. Schilmiller AL, et al. (2015) Functionally divergent alleles and duplicated loci encodingan acyltransferase contribute to acylsugar metabolite diversity in Solanum trichomes.Plant Cell 27(4):1002–1017.

26. Ning J, et al. (2015) A feedback insensitive isopropylmalate synthase affects acylsugarcomposition in cultivated and wild tomato. Plant Physiol 169(3):1821–1835.

27. St. Pierre B, De Luca V (2000) Evolution of acyltransferase genes: Origin and di-versification of the BAHD superfamily of acyltransferases involved in secondary me-tabolism. Recent Advances in Phytochemistry Evolution of Metabolic Pathways, edsRomeo JT, Ibrahim R, Varin L, De Luca V (Elsevier, Oxford, UK), pp 285–315.

28. D’Auria JC (2006) Acyltransferases in plants: A good time to be BAHD. Curr Opin PlantBiol 9(3):331–340.

29. Schilmiller AL, Charbonneau AL, Last RL (2012) Identification of a BAHD acetyl-transferase that produces protective acyl sugars in tomato trichomes. Proc Natl AcadSci USA 109(40):16377–16382.

30. Schilmiller AL, et al. (2010) Studies of a biochemical factory: Tomato trichome deepexpressed sequence tag sequencing and proteomics. Plant Physiol 153(3):1212–1223.

31. Eshed Y, Zamir D (1994) A genomic library of Lycopersicon pennellii in L. esculentum :A tool for fine mapping of genes. Euphytica 79(3):175–179.

32. Landmann C, et al. (2011) Substrate promiscuity of a rosmarinic acid synthase fromlavender (Lavandula angustifolia L.). Planta 234(2):305–320.

33. Luo J, et al. (2009) A novel polyamine acyltransferase responsible for the accumula-tion of spermidine conjugates in Arabidopsis seed. Plant Cell 21(1):318–333.

34. Hoffmann L, Maury S, Martz F, Geoffroy P, Legrand M (2003) Purification, cloning,and properties of an acyltransferase controlling shikimate and quinate ester inter-mediates in phenylpropanoid metabolism. J Biol Chem 278(1):95–103.

35. 100 Tomato Genome Sequencing Consortium, et al. (2014) Exploring genetic varia-tion in the tomato (Solanum section Lycopersicon) clade by whole-genome se-quencing. Plant J 80(1):136–148.

36. Garvey GS, McCormick SP, Rayment I (2008) Structural and functional characterizationof the TRI101 trichothecene 3-O-acetyltransferase from Fusarium sporotrichioides andFusarium graminearum: Kinetic insights to combating Fusarium head blight. J BiolChem 283(3):1660–1669.

37. Sallaud C, et al. (2009) A novel pathway for sesquiterpene biosynthesis from Z,Z-farnesyl pyrophosphate in the wild tomato Solanum habrochaites. Plant Cell 21(1):301–317.

38. Schilmiller AL, et al. (2009) Monoterpenes in the glandular trichomes of tomato are

synthesized from a neryl diphosphate precursor rather than geranyl diphosphate.

Proc Natl Acad Sci USA 106(26):10865–10870.39. Gonzales-Vigil E, Hufnagel DE, Kim J, Last RL, Barry CS (2012) Evolution of TPS20-

related terpene synthases influences chemical diversity in the glandular trichomes of

the wild tomato relative Solanum habrochaites. Plant J 71(6):921–935.40. Matsuba Y, et al. (2013) Evolution of a complex locus for terpene biosynthesis in

solanum. Plant Cell 25(6):2022–2036.41. Kang J-H, et al. (2014) The flavonoid biosynthetic enzyme chalcone isomerase mod-

ulates terpenoid production in glandular trichomes of tomato. Plant Physiol 164(3):

1161–1174.42. Kim J, et al. (2014) Analysis of natural and induced variation in tomato glandular

trichome flavonoids identifies a gene not present in the reference genome. Plant Cell

26(8):3272–3285.43. Rodriguez F, Wu F, Ané C, Tanksley S, Spooner DM (2009) Do potatoes and tomatoes

have a single evolutionary history, and what proportion of the genome supports this

history? BMC Evol Biol 9:191.44. Lutz S (2010) Beyond directed evolution: Semi-rational protein engineering and de-

sign. Curr Opin Biotechnol 21(6):734–743.45. Bloom JD, et al. (2005) Evolving strategies for enzyme engineering. Curr Opin Struct

Biol 15(4):447–452.46. Dalby PA (2003) Optimising enzyme function by directed evolution. Curr Opin Struct

Biol 13(4):500–505.47. Chen R (2001) Enzyme engineering: Rational redesign versus directed evolution.

Trends Biotechnol 19(1):13–14.48. Kang J-H, Gonzales-Vigil E, Matsuba Y, Pichersky E, Barry CS (2014) Determination of

residues responsible for substrate and product specificity of Solanum habrochaites

short-chain cis-prenyltransferases. Plant Physiol 164(1):80–91.49. Komori A, et al. (2013) Comparative functional analysis of CYP71AV1 natural variants

reveals an important residue for the successive oxidation of amorpha-4,11-diene.

FEBS Lett 587(3):278–284.50. McCormick S (1991) Transformation of tomato with Agrobacterium tumefaciens.

Plant Tissue Culture Manual, ed Lindsey K (Kluwer Academic Publishers, Dordrecht,

The Netherlands), pp 1–9.51. Eamens AL, Waterhouse PM (2011) Vectors and methods for hairpin RNA and arti-

ficial microRNA-mediated gene silencing in plants. Plant Chromosome Engineering:

Methods and Protocols, ed Birchler JA (Humana Press, New York), pp 179–197.52. Karimi M, Inzé D, Depicker A (2002) GATEWAY vectors for Agrobacterium-mediated

plant transformation. Trends Plant Sci 7(5):193–195.53. Kelley LA, Sternberg MJE (2009) Protein structure prediction on the Web: A case study

using the Phyre server. Nat Protoc 4(3):363–371.

E248 | www.pnas.org/cgi/doi/10.1073/pnas.1517930113 Fan et al.

Dow

nloa

ded

by g

uest

on

June

14,

202

1

www.pnas.org/cgi/doi/10.1073/pnas.1517930113