REGULAR PAPER

RNAi-mediated suppression of the phenylalanine ammonia-lyasegene in Salvia miltiorrhiza causes abnormal phenotypesand a reduction in rosmarinic acid biosynthesis

Jie Song • Zhezhi Wang

Received: 17 August 2009 / Accepted: 6 April 2010 / Published online: 28 May 2010

� The Botanical Society of Japan and Springer 2010

Abstract Medicinal Salvia miltiorrhiza contains two

main groups of active pharmaceutical ingredients: lipid-

soluble tanshinones and water-soluble phenolic acids,

including rosmarinic acid and salvianolic acid B. Phenyl-

alanine ammonia-lyase (PAL) catalyzes the first step in the

phenylpropanoid pathway and is assumed to be closely

related to the accumulation of rosmarinic acid and its

derivatives. We selected a 217-bp fragment, located at the

30 end of the coding region of PAL1, to establish an RNA

interference construct that was introduced into S. mil-

tiorrhiza via Agrobacterium tumefaciens-mediated trans-

formation. PAL-suppressed plants exhibited several

unusual phenotypes such as stunted growth, delayed root

formation, altered leaves, and reduced lignin deposition.

The total phenolic content was decreased by 20–70% in

PAL-suppressed lines, and was accompanied by lower PAL

activity. Down-regulation of PAL also affected the

expression of C4H, 4CL2, and TAT, which are related

genes in the rosmarinic acid pathway. Moreover, rosmari-

nic acid and salvianolic acid B were markedly reduced in

PAL-suppressed lines, as detected by HPLC analysis. Our

results indicate that PAL is very important for the synthesis

of major water-soluble pharmaceutical ingredients within

S. miltiorrhiza.

Keywords Phenotype � Phenylalanine ammonia-lyase �RNA interference � Rosmarinic acid � Salvia miltiorrhiza

Introduction

Salvia miltiorrhiza Bunge (‘‘Danshen’’ in Chinese) is widely

grown across China and in other Asian countries. As a tra-

ditional medicine, its roots are renowned for their curative

effects on coronary heart diseases, particularly angina pec-

toris and myocardial infarction (Zhang et al. 2002; Zhou

et al. 2005). Natural-products chemists and medicinal cli-

nicians have focused on this species because of its important

biological activities, including antioxidant, antitumor, and

antimicrobial properties. Its active pharmaceutical ingredi-

ents are divided into two main groups: lipid-soluble tanshi-

nones and water-soluble phenolic acids, such as rosmarinic

acid, salvianolic acids, lithospermic acid, and 3,4-dihydr-

oxyphenyllactic acid (Liu et al. 2006a; Ma et al. 2006). These

phenolics now attract more attention because they are the

main components of water decoction, which is the most

common form of dosing administered to patients in Chinese

clinics (Hu et al. 2005; Liu et al. 2006a; Ma et al. 2006).

Among these compounds, salvianolic acid B is a major

water-soluble constituent of S. miltiorrhiza ([3% of dry

weight). When those plants are processed traditionally by

extraction with water, salvianolic acid B is responsible for

many therapeutic actions, especially antioxidation and the

scavenging of free radicals (Zhang et al. 2002).

Rosmarinic acid, which has antiviral, antimicrobial, and

anti-inflammatory activities, is thought to be the core

structure of most hydrophilic compounds in S. miltiorrhiza,

such as salvianolic and lithospermic acids (Petersen and

Simmonds 2003). A proposed biosynthetic pathway in

Coleus blumei suggests that rosmarinic acid is an ester of

Electronic supplementary material The online version of thisarticle (doi:10.1007/s10265-010-0350-5) contains supplementarymaterial, which is available to authorized users.

J. Song � Z. Wang (&)

Key Laboratory of the Ministry of Education for Medicinal

Resources and Natural Pharmaceutical Chemistry, National

Engineering Laboratory for Resource Development of

Endangered Crude Drugs in Northwest of China, Shaanxi

Normal University, Xi’an 710062, People’s Republic of China

e-mail: zzwang@snnu.edu.cn

123

J Plant Res (2011) 124:183–192

DOI 10.1007/s10265-010-0350-5

3,4-dihydroxyphenyllactic acid and caffeic acid. Those

two compounds are synthesized via the tyrosine-derived

pathway and the phenylpropanoid pathway, respectively

(Petersen et al. 1993) (Fig. 1).

Phenylalanine ammonia-lyase (PAL, EC 4.3.1.5) is the

first enzyme in the phenylpropanoid pathway, and is also

the branch point between primary and secondary metabo-

lism. It catalyzes the conversion of L-Phe to trans-cin-

namic acid, which is further transformed into many

phenylpropanoid compounds. Because of its crucial role in

that pathway, PAL has been widely studied since its dis-

covery. This enzyme participates in many physiological

processes that involve anthocyanin accumulation, lignifi-

cation, flavonoid synthesis, and pathogen defense (Lois

et al. 1989; Dixon and Paiva 1995; Shadle et al. 2003;

MacDonald and D’Cunha 2007). It also acts in the syn-

thesis of main active ingredients within many plants, e.g.,

caffeoyl quinic acids in Coffea canephora (Mahesh et al.

2006) and quercetin in Astragalus membranaceus var.

mongholicus (Liu et al. 2006b). Razzaque and Ellis (1977)

have shown that PAL is important for rosmarinic acid

biosynthesis in Coleus blumei, and Chen and Chen (2000)

have suggested that this enzyme has the same role in

S. miltiorrhiza.

PAL is encoded, in most plants, by a small multi-gene

family, and sequences for different members are highly

conserved. PAL genes have been cloned from various

species, such as Arabidopsis thaliana (Wanner et al. 1995;

Cochrane et al. 2004) and Petroselinum crispum (Lois et al.

1989). Although we have previously reported the cloning

of SmPAL1 from S. miltiorrhiza (Song and Wang 2009),

details of its role in the biosynthesis of rosmarinic acid and

its derivatives in that species are unclear.

In the post-genome era, analysis of loss-of-function

mutants is critical to determining gene functioning. RNA

interference (RNAi) is one of the most powerful means

for generating knockdown plants (Higuchi et al. 2009).

Therefore, to gain insight into the nature of rosmarinic acid

biosynthesis, we applied RNAi technology to suppress the

expression of SmPAL1, based on a genetics transformation

system established in our laboratory (Yan and Wang 2007).

Here, we investigated the role of PAL in the production

of hydrophilic active pharmaceutical ingredients in

S. miltiorrhiza. Our objectives were to assess any altera-

tions in phenotype and to monitor changes in the contents

of rosmarinic acid and salvianolic acid B within PAL-

suppressed lines.

Materials and methods

Plant materials

Sterile plantlets of Salvia miltiorrhiza Bunge were cultured

on an MS basal medium (Murashige and Skoog 1962), as

described by Yan and Wang (2007).

Vector construction

Primers for PAL1 were designed according to the sequence

submitted at GenBank (Accession Number EF462460).

Pairs included sense F (50-CCGCTCGAGACCCCTTGA

TGCAGAAGCTGAGG-30) and sense R (50-GGGGTA

CCGGGTACGACCTGCACTCCGCTAT-30), with under-

lined sites for XhoI and KpnI, respectively; and anti-sense F

(50-GGCGGATCCACCCCTTGATGCAGAAGCTGAGG-30)and anti-sense R (50-GTATCGATGGGTACGACCTGCA

CTCCGCTAT-30), with italicized sites for BamHI and

ClaI, respectively. These were used to amplify 217-bp

fragments of PAL1 (located from coding sequence posi-

tions 1744–1961, and sharing [70% identity with PAL

genes reported from other species). The fragments were

digested with XhoI/KpnI and BamHI/ClaI (Takara, Japan),

then inserted into vector pKANNIBAL (Wesley et al.

2001) to generate plasmid pKanPAL1. An interfering box

(Fig. 2a) with the pyruvate orthophosphate dikinase (PDK)

intron was removed with NotI (Takara, Japan) from

Fig. 1 Proposed biosynthetic pathway for rosmarinic acid. C4Hcinnamate 4-hydroxylase; 4CL 4-coumarate:coenzyme A ligase;

4-HPLA 4-hydroxyphenyllactic acid; 4-HPPA 4-hydroxyphenylpyru-

vic acid; HPPR hydroxyphenylpyruvate reductase; PAL phenylala-

nine ammonia-lyase; RAS rosmarinic acid synthase; TAT tyrosine

aminotransferase (Petersen et al. 1993; Yan et al. 2006)

184 J Plant Res (2011) 124:183–192

123

pKanPAL1 and cloned into the site for NotI in vector

pART27 (Gleave 1992), thereby generating RNAi vector

pPAL1. The vector carried a spectinomycin-resistance

gene (Spe) as a bacterial selection marker plus NPT-II as

our plant selection marker. Plasmid pPAL1 was sequenced

by Sangon Biological Engineering Technology & Services

Co., Ltd. (Shanghai, China), and then introduced into

Agrobacterium tumefaciens EHA105. Resistant colonies

were verified by PCR-amplification and used in our

transformation experiments. An empty pART27 vector

without the interfering box was separately introduced into

EHA105 to serve as our vector control.

Agrobacterium-mediated gene transfer

Gene transfer was performed as described by Yan and

Wang (2007). A single clone of A. tumefaciens EHA105

harboring RNAi vector pPAL1 was inoculated into 10 mL

of a liquid LB medium that contained 50 mg L-1 specti-

nomycin and 40 mg L-1 rifampicin, and then grown on a

shaker (180 rpm) at 28�C for 16–18 h. Cells were collected

by centrifugation when the OD600 reached 0.6, and were

re-suspended in 2–3 vol of a liquid MS medium. Sterile

leaves were cut into 0.5 9 0.5 cm discs and pre-cultured

for 1 day on an MS basal medium supplemented with

1.0 mg L-1 BAP and 0.1 mg L-1 NAA. Afterward, the

discs were submerged with shaking in a bacterial suspen-

sion for 30 min. Excess bacteria were later blotted, and the

discs were transferred to the same media type and cultured

for 3 days. They were then moved to a selection medium

(MS basal medium supplemented with 1.0 mg L-1 BAP,

0.1 mg L-1 NAA, 200 mg L-1 cefotaxime sodium, and

60 mg L-1 kanamycin). After four cycles of selection

(10 days each) with 60 mg L-1 kanamycin, the regener-

ated buds were transferred to a �-strength MS basal

medium supplemented with 30 mg L-1 kanamycin for root

formation and elongation. For our untransformed control,

we used buds regenerated from leaf discs that had not been

submerged in bacteria but were only cultured on an MS

basal medium supplemented with 1.0 mg L-1 BAP and

0.1 mg L-1 NAA. Rooted plantlets were cut from their

internodes into segments and were cultured on a �-strength

MS basal medium for propagation. One month old plantlets

were used for PCR-screening, and for evaluating gene

expression, enzyme activity, and phenolic acids. Positively

transformed plants were then grown in soil for further

observation and HPLC analysis.

Screening transgenic plants by PCR

DNA was isolated from the leaves of putative transfor-

mants by the CTAB method (Doyle and Doyle 1987).

Primers PDK F (50-GTGATGTGTAAGACGAAGAAG-30)and PDK R (50-GATAGATCTTGCGCTTTG-30) were

used to amplify a 427-bp fragment from the intron

sequence of the interfering box. To test for the presence of

contaminating Agrobacterium cells in plant tissues, we

included Spe-specific primers F (50-TTGATTTGCTGGT

TACGG-30) and R (50-ATGACGGGCTGATACTGG-30).PCR was conducted in a PTC-200 thermocycler (Bio-Rad,

USA) in 20 lL of solution containing 50 ng of DNA, 2 lL

of 109 PCR reaction buffer, 2 lL of dNTP (2.5 mM),

2 lL of Mg2? (2.5 mM), 0.5 lM of each primer, and 0.5 U

of rTaq DNA polymerase (Takara, Japan). These mixtures

were treated at 94�C for 5 min, then subjected to 35 cycles

of amplification (94�C for 30 s, 52�C for 1 min, 72�C for

40 s), followed by a final elongation at 72�C for 10 min.

RNAi vector pPAL1 was used as the positive control, while

genomic DNA from untransformed plants was our negative

control. Amplified products were electrophoresed on a

Fig. 2 Vector construction and

molecular analysis of transgenic

plants. a Sketch map of

interfering box. Restriction sites

and target gene are marked.

b PCR-screening of PAL-

suppressed Salvia miltiorrhiza.

Pyruvate orthophosphate

dikinase (PDK) intron was

amplified to screen positive

transgenic lines.

Spectinomycin-resistance gene

(Spe) was amplified to test for

presence of contaminating

Agrobacterium cells in plant

tissues. M marker; ? positive

control; – no template control;

C untransformed control

J Plant Res (2011) 124:183–192 185

123

1.0% agarose gel containing 0.5 lg mL-1 ethidium bro-

mide, and visualized under ultraviolet (UV) light.

Gene expression analysis by quantitative real-time PCR

Total RNAs were extracted from the 1 month old plantlets

(including roots, stems and leaves) with Trizol reagent

(BioFlux, China). cDNA was obtained with an oligo(dT)18

primer and a RevertAIDTM

First Strand cDNA Synthesis Kit

(MBI, USA), per the manufacturer’s protocol. A fragment

of Ubiquitin was amplified as an internal control for cali-

brating relative levels of gene expression. Table 1 lists our

gene-specific primers for quantitative PCR and the lengths

of the target fragments. PCR-amplification was done in a

20-lL total volume containing appropriate diluted cDNA,

0.2 lM for each primer, and 10 lL of 29 Q-PCR Mix

(with SYBR Green I; Takara, Japan). Negative controls

that lacked cDNA template were included in this experi-

ment. Real-time quantitative PCR (qRT-PCR) was per-

formed in an iQ5 thermocycler (Bio-Rad) as follows:

1 min of pre-denaturation at 94�C; then 35 cycles of

denaturation at 94�C for 10 s, annealing at 60�C for 20 s,

and fluorescence collection at 82�C for 15 s. Products were

then run on a 1.5% agarose gel containing 0.5 lg mL-1

ethidium bromide to obtain a band of the predicted size.

Gene expression was quantified by the comparative CT

method (Schmittgen and Livak 2008).

Assay of PAL activity

Enzymes were extracted from 1 month old plantlets

(including roots, stems and leaves) by using 100 mM

phosphate buffer (pH 6.0) that contained 2 mM EDTA,

4 mM dithiothreitol, and 2% (w/w) polyvinylpyrrolidone.

Fresh samples were ground on ice for 5 min in

0.25 g mL-1 of the extraction buffer, then centrifuged for

25 min at 17,0009g and 4�C to obtain a solid-free extract.

PAL activity of that extract was determined by the method

of Yan et al. (2006), with slightly modifications. Briefly,

the enzyme extract (0.2 mL) was incubated at 30�C for

60 min with 2 mL of 0.01 M borate buffer (pH 8.7) and

1 mL of 0.02 M L-phenylalanine (pre-dissolved in 0.01 M

borate buffer, pH 8.7). This reaction was stopped by the

addition of 1 ml of 6 M HCl. The reaction was then cen-

trifuged for 10 min at 12,0009g to pellet the denatured

protein. Absorbance was measured at 290 nm before and

after incubation. One unit of activity (katal) was defined as

the amount of PAL that produced 1 mol of cinnamic acid

in 1 s, and was expressed as nkatal mg-1 protein. A

reaction without the substrate was our blank control.

Triplicate assays were performed for each extract. Protein

contents were determined by the method of Bradford

(1976), using bovine serum albumin as a standard.

Histochemical determination of lignin

The third to fourth internodes from 1 month old plantlets

were used for histochemical analysis of lignin. Free-hand

sections were immediately immersed for 2 min in 10%

phloroglucinol (dissolved in 100% ethanol), then incubated

in concentrated HCl. Specimens were then photographed

within 30 min at high magnification (1009) under a bright

field with a Leica microscope (DMLB2, Germany) (Bate

et al. 1994).

Analysis of total phenolics

Fresh samples (0.5 g, 1 month old plantlets including

roots, stems and leaves) were ground into powder in liquid

nitrogen, and extracted with 7 mL of phosphate buffer

Table 1 List of primers used

for gene expression analysisGenes Product (bp) Sequence

Ubiquitin 207 F: 50-ACCCTCACGGGGAAGACCATC-30

R: 50-ACCACGGAGACGGAGGACAAG-30

PAL1 203 F: 50-GATAGCGGAGTGCAGGTCGTAC-30

R: 50-CGAACTAGCAGATTGGCAGAGG-30

PAL2 218 F: 50-GGCGGCGATTGAGAGCAGGA-30

R: 50-ATCAGCAGATAGGAAGAGGAGCACC-30

C4H 183 F: 50-CCAGGAGTCCAAATAA CAGAGCCG-30

R: 50-GCCACCAAGCGTTCACCAAG AT-30

4CL2 134 F: 50-TCGCCAAATACGACCTTTCC-30

R: 50-TGCTTCAGTCATCCCATACCC-30

TAT 143 F: 50-CAACTGCTGGTCTTCCACAAAC-30

R: 50-GCGAGCCAAAACGGACA-30

HPPR 139 F: 50-TGACTCCAGAAACAACCCACATT-30

R: 50-CCCAGACGACCCTCCACAAG-30

186 J Plant Res (2011) 124:183–192

123

(75 mM, pH 7.0) by vortexing for 2–3 min. Each mixture

was centrifuged at 13,0009g for 25 min, and the super-

natant was collected for assaying total phenolics after being

diluted to a suitable concentration with extraction buffer

(Yan et al. 2006). The extract solution (200 lL) was mixed

in a test tube with 1 mL of Folin-Ciocalteu reagent and

0.8 mL of Na2CO3 (7.5%). This mixture was incubated at

40�C for 30 min, then measured for absorbance at 765 nm

(Zheng and Wang 2001). Gallic acid was used as a stan-

dard. The total phenolic content was depicted as milligrams

of gallic-acid equivalent per gram of fresh weight. Each

extract was analyzed in triplicate.

HPLC analysis of hydrophilic active pharmaceutical

ingredients

Three-month-old plantlets were dried at 30�C to a constant

weight. Afterward, 300-mg samples of roots, stems, and

leaves were ground into powder and extracted with 15 mL

of 70% methanol under sonication for 30 min. Following

centrifugation at 1,7009g for 5 min, the extracts were

separated on a reverse-phase HPLC column (LC-2010A;

SHIMADZU, Kyoto, Japan) equipped with an auto-injec-

tor, UV detector, and Empower software. Spectral data

were recorded at 280 nm during the entire run. A C18

column (150 9 4.6 mm, 5-lm particle size) (SHIMADZU)

and a 20.0 9 4.6 mm guard column were used for phenolic

separations at 30�C. The mobile phase comprised 0.5%

acetic acid: water solution (Solvent A) and methanol con-

taining 0.5% acetic acid (Solvent B). The gradient was

0–30 min, 10% B to 25% B; 30–45 min, 25% B to 35% B;

and 45–60 min, 35% B to 50% B. A flow rate of

1.0 mL min-1 was used and 20 lL of each sample was

injected. All samples and mobile phases were first passed

through a 0.22-lm filter, type GV (Jinteng, China).

Statistical analysis

Statistical analysis was performed with SPSS 13.0 soft-

ware. Analysis of variance (ANOVA) was followed by

Tukey’s pairwise comparison tests, at a level of P \ 0.05,

to determine significant differences between means.

Results

Generation of PAL-suppressed S. miltiorrhiza

A 217-bp fragment of PAL1, located at the 30 end of its

coding region, shares high identity with PAL genes repor-

ted from Agastache rugosa (AF326116), Solenostemon

scutellarioides (EU019242), and P. crispum (X17462). It

was used here to generate an intron-spliced hpRNA vector

(pPAL1). When T-DNA was transferred into S. miltiorrh-

iza, adventitious buds appeared on transformed explants

after 20 days. In contrast, untransformed explants on

selection media blackened and died. As soon as buds were

observed, they were transferred to fresh media. Following

four cycles of selection, green regenerated buds (1-cm-

long) were moved to a basal medium supplemented with

kanamycin for root formation and elongation. Leaves from

1 month old transgenic plantlets were used for DNA

extraction and PCR-screening. An expected 427-bp frag-

ment of the PDK intron was amplified in the positive

control and (except for escaped Lines 5, 9, and 23) in our

RNAi lines. Negative results were obtained from PCR

assays performed on genomic DNA samples of RNAi lines

when primers specific for spectinomycin-resistance genes

were used. This abolished any suspicion of false positives

due to contamination from Agrobacterium cells in plant

tissues (Fig. 2b). Verified transgenics were screened by

amplifying the sequence of the PDK intron. We then

determined that PAL1 expression was suppressed to vary-

ing degrees in different RNAi lines (Fig. 3a, Supplemen-

tary Fig. S1).

Morphological characterization of PAL-suppressed

S. miltiorrhiza

Segments from verified transgenics were cultured for

morphologic observations. Untransformed plantlets as well

as our vector-control line and some PAL-suppressed lines

began to root within 4 days. In contrast, most suppressed

lines exhibited delayed root formation and started rooting

only at 10–30 days after sub-culturing began. A few indi-

viduals (28%) died during this period. In addition, lignin

deposition was reduced in PAL-suppressed plants, as

indicated by the poor staining of stem sections with

phloroglucinol, compared with untransformed plantlets and

the vector-control line (Fig. 4a).

The positive plants were transplanted into soil for

further observation. Compared with untransformed

plantlets and those from the vector-control line, RNAi

lines showed abnormal phenotypes, including dwarfism

and an underdeveloped root system (Fig. 4b). Leaves

from most PAL-suppressed lines were curled or perished

(Fig. 4c). Based on these characteristics, we divided the

phenotypes of PAL-suppressed S. miltiorrhiza into three

classes: Class I (3 of 25 independent transformants), with

only dwarfism and an underdeveloped root system (i.e.,

RNAi-1, -8, and -10); Class II (15 of 25 independent

transformants), with dwarfism, delayed rooting, and

altered leaf formation (RNAi-3, -4, -6, -7, -15, -18, -21,

and -24); and Class III (7 of 25 independent transfor-

mants), with the same Class-II phenotype plus mortality

during the culture period.

J Plant Res (2011) 124:183–192 187

123

Analyses of PAL activity and metabolites

Three transgenic lines (RNAi-4, RNAi-8, and RNAi-21)

that presented obvious phenotypic changes and had sur-

vived during culturing were analyzed for PAL activity and

total phenolics. In all lines, PAL activity was reduced,

especially in RNAi-4 and RNAi-21, both within Class II

(Fig. 5). Compared with the untransformed plantlets and

vector-control line, total phenolics contents were 20–70%

lower in PAL-suppressed plants (Fig. 6a). HPLC results

showed a large decrease in levels of rosmarinic acid and

salvianolic acid B in RNAi-4 and RNAi-21, two lines with

serious abnormalities (Fig. 6b). However, those changes

were more moderate in RNAi-8, a line with only mild

symptoms. The contents of other water-soluble ingredients,

such as 3,4-dihydroxyphenyllactic acid and protocatechu-

aldehyde, were too low to be quantified.

Gene expression in PAL-suppressed S. miltiorrhiza

Expression of PAL2 was decreased in all RNAi lines,

especially RNAi-8 in Class I (Fig. 3b). Transcript levels of

C4H and 4CL2, which encode enzymes that act in the

phenylpropanoid pathway and downstream of PAL, were

increased significantly in Class-II RNAi-4 and RNAi-21

(Fig. 3c, d). We also evaluated the first two genes within

the tyrosine-derived pathway, and found that expression of

TAT was obviously elevated in PAL-suppressed lines

(Fig. 3e), whereas that of HPPR showed no clear change

(i.e., \two fold) (Fig. 3f).

Discussion

PAL-suppressed S. miltiorrhiza exhibits abnormal

vegetative development

The PAL enzyme in plants has been extensively studied for

its crucial functioning in the biosynthesis of various sec-

ondary metabolites (Hahlbrock and Scheel 1989). Intro-

duction of the bean PAL2 gene into transgenic tobacco

paradoxically leads to homology-dependent silencing of

the endogenous tobacco PAL gene, resulting in a series of

unusual and unexpected visible phenotypes, such as stunted

growth, localized lesions, altered leaf shape, and less lig-

nification in the xylem (Elkind et al. 1990; Bate et al. 1994;

Sewalt et al. 1997). In Medicago sativa, down-regulation of

PAL can cause a reduction in growth rate and lignin content

(Chen et al. 2006). However, investigations with T-DNA

insertional mutants for PAL genes in A. thaliana have

revealed no visible phenotypic alterations in PAL1 and

PAL2 single and double mutants, except for a slight

decrease in lignin accumulation in the double mutant

(Rohde et al. 2004). In contrast, we demonstrated obvious

Fig. 3 Expression analysis of

PAL1 (a), PAL2 (b), C4H (c),

4CL2 (d), TAT (e), and HPPR(f) in PAL-suppressed

S. miltiorrhiza. Data represent

average of 3 experiments; errorbars show standard deviations.

Asterisk indicates that

difference is significant at

P \ 0.05 compared with

untransformed control.

VC vector control; UCuntransformed control; R-4, -8,-21 PAL-suppressed lines

188 J Plant Res (2011) 124:183–192

123

phenotypic changes in S. miltiorrhiza caused by RNAi-

associated suppression of PAL. Almost all PAL-suppressed

plants exhibited some degree of abnormality, including

dwarfism, altered leaves, and less lignin, as well as delayed

root formation and an underdeveloped root system.

Reduced PAL activity in our transgenic plants might

suggest that all of these symptoms resulted from a pertur-

bation of phenylpropanoid metabolism. PAL catalyzes the

first step in the synthesis of diverse natural products based

on the phenylpropane skeleton, which fulfill many essential

functions in higher plants (MacDonald and D’Cunha

2007). For example, hydroxycinnamic acids and flavonoids

can modulate auxin metabolism and its polar transport,

respectively (Elkind et al. 1990). Unusual auxin metabo-

lism might lead to dwarfism, delayed root formation, and

Fig. 4 Suppression of PAL via

RNAi in S. miltiorrhiza caused

abnormal phenotypes, including

reduced lignin content (a),

dwarfism and underdeveloped

root system (b), and altered

leaves (c). Scale bars 50 lm.

P pith; X xylem; Co cortex;

VC vector control; UCuntransformed control; R-4, -8,-21 PAL-suppressed lines

Fig. 5 PAL activity in RNAi lines. Data represent average of 3

experiments, error bars show standard deviations. Asterisk indicates

that difference is significant at P \ 0.05 compared with untrans-

formed control. VC vector control; UC untransformed control; R-4,-8, -21 PAL-suppressed lines

J Plant Res (2011) 124:183–192 189

123

an underdeveloped root system. Here, an abnormal phe-

notype was also associated with a reduced accumulation of

total phenolic acids, confirming that PAL is a key regula-

tory step in the phenylpropanoid pathway.

Lignin, an important product of phenylpropanoid

metabolism, is involved in many physiological processes.

It is a main element of the secondary cell wall and is

essential for mechanical support and water transport in

vascular plants. The resistance by xylem to stresses is

critical to the growth and development of higher plants

(Campbell and Sederoff 1996; Goujon et al. 2003). Here, a

lower lignin accumulation could have been another key

contributor to the traits found in our PAL-suppressed Salvia

plants.

Silencing PAL affects the expression of other

rosmarinic acid synthesis-related genes

in S. miltiorrhiza

PAL seems to be encoded by a multi-gene family in higher

plants, with the number of isoforms varying among species.

The genome of Arabidopsis thaliana harbors four PAL

genes that are expressed differently over the course of

development (Cochrane et al. 2004; Rohde et al. 2004;

Olsen et al. 2008). Significantly increased transcript levels

of PAL4 have been discovered in both the PAL1 and PAL2

single and double mutants of A. thaliana, indicating that

disruption of PAL1 and PAL2 can be compensated for

by other PAL isoforms (Rohde et al. 2004). PAL in

S. miltiorrhiza also is probably encoded by a multi-gene

family of perhaps three members (Hu et al. 2009). We

utilized the 30 end of PAL2 (1,784 bp; GenBank Accession

Number GQ249111) for transcription analysis in PAL1-

suppressed S. miltiorrhiza. In contrast to its functioning in

A. thaliana, expression of PAL2 was decreased here to

various degrees in RNAi lines. This might have been

because the target fragment we used for RNAi vector

construction shared a high identity (74%) with PAL2

(Supplementary Fig. S2). Hence, the similarity between

those two PAL genes led to the silencing of both isoforms.

Moreover, morphological changes, PAL activity, and

phenolic acid content in our RNAi lines were consistent

with the expression of PAL1 but not PAL2. Finally, the

transcription level for PAL2 in normal plants was about

eight-fold less than for PAL1 (Supplementary Fig. S3).

Thus, we can assume that PAL1 plays a major role during

the development and metabolic course of S. miltiorrhiza.

According to the pathway proposed by Petersen et al.

(1993) (Fig. 1), several enzymes participate in rosmarinic

acid biosynthesis; these are PAL, cinnamate 4-hydroxylase

(C4H, EC 1.14.13.11) and 4-coumarate:coenzyme A ligase

(4CL, EC 6.2.1.12), which belong to the phenylpropanoid

pathway; and tyrosine amino-transferase (TAT, EC 2.6.1.5)

and hydroxyphenylpyruvate reductase (HPPR, EC

1.1.1.237), from the tyrosine-derived pathway. C4H, the

second enzyme within the general phenylpropanoid path-

way, is often studied together with PAL (Blount et al.

2000; Achnine et al. 2004). In PAL-suppressed tobacco

plants, C4H activity remains unchanged (Blount et al.

2000); a similar phenomenon is found in A. thaliana pal1

mutants (Rohde et al. 2004). However, an NAPDH P450

reductase gene, ATR3, that supports C4H activity is up-

regulated in pal1 mutants (Rohde et al. 2004). 4CL plays a

pivotal role in the general pathway of phenylpropanoid

metabolism by providing activated coenzyme A esters of

hydroxycinnamic acids. Its transcription is induced in

A. thaliana pal1 mutants (Rohde et al. 2004). In S. mil-

tiorrhiza, 4CL exists in two isogenes and 4CL2 is assumed

to play a more critical role than does 4CL1 in the biosyn-

thesis of water-soluble phenolic compounds (Zhao et al.

2006). Here, C4H and 4CL2 expression was increased in

PAL-suppressed plants, demonstrating that the reduced flux

through PAL could enhance transcription of those con-

secutive genes. TAT, which catalyzes transamination from

L-tyrosine to 4-hydroxyphenylpyruvate, is the first enzyme

in the tyrosine-derived branch pathway of rosmarinic acid

Fig. 6 Suppression of PAL resulted in reduction of phenolic-acid

biosynthesis. a Content of total phenolics. b Content of rosmarinic

acid and salvianolic acid B. Data represent average of 3 experiments;

error bars show standard deviations. Asterisk indicates that difference

is significant at P \ 0.05 compared with untransformed control.

VC vector control; UC untransformed control; R-4, -8, -21 PAL-

suppressed lines

190 J Plant Res (2011) 124:183–192

123

biosynthesis. It degrades both Phe and Tyr and is trans-

criptionally down-regulated in A. thaliana pal1 mutants

(Rohde et al. 2004). In contrast, suppression of PAL in

S. miltiorrhiza results in the up-regulation of TAT expres-

sion while that of HPPR, which encodes the enzyme that

acts downstream of TAT, is not obviously changed.

PAL plays an important role in the synthesis of major

hydrophilic active pharmaceutical ingredients

from S. miltiorrhiza

Thorough study of the biochemical constituents of S. mil-

tiorrhiza has revealed 25 caffeic acid derivatives that have

been isolated and identified from aqueous extracts,

including caffeic, rosmarinic, and lithospermic acids, and

salvianolic acid B (Lu and Foo 2002; Petersen and Sim-

monds 2003; Hu et al. 2005). Although the water-soluble

fraction of S. miltiorrhiza has been formulated and used

clinically for many years in China for disease treatments,

the biosynthesis of salvianolic acid B and other rosmarinic

acid derivatives is still unclear. PAL is the first enzyme in

the phenylalanine-derived branch of the biosynthetic

pathway for rosmarinic acid (Petersen et al. 1993). Chen

and Chen (2000) have treated cell cultures of S. mil-

tiorrhiza with a PAL inhibitor and found that formation of

rosmarinic acid is blocked, suggesting that PAL plays an

important role in such biosynthesis. However, Yan et al.

(2006) have examined rosmarinic acid accumulation and

PAL activity in S. miltiorrhiza hairy roots treated with

yeast extract and Ag?, and have concluded that the elicitor-

induced accumulation of rosmarinic acid is not correlated

with PAL activity.

To further examine that relationship, we used HPLC to

analyze the soluble phenolic extracts of PAL-suppressed

lines in which PAL activity was diminished. Accumula-

tions of both rosmarinic acid and salvianolic acid B were

reduced, indicating that PAL is of great importance to the

synthesis of these hydrophilic active pharmaceutical

ingredients in S. miltiorrhiza.

Acknowledgments This work was supported by the 10th–11th

‘‘five-year technique project’’ of the Ministry of Science and Tech-

nology of the People’s Republic of China (2004BA701A35,

2006BAI06A12-04).

References

Achnine L, Blancaflor EB, Rasmussen S, Dixon RA (2004) Colocal-

ization of L-phenylalanine ammonia-lyase and cinnamate

4-hydroxylase for metabolic channeling in phenylpropanoid

biosynthesis. Plant Cell 16:3098–3109

Bate NJ, Orr J, Ni W, Meromi A, Nadler-Hassar T, Doerner PW,

Dixon RA, Lamb CJ, Elkind Y (1994) Quantitative relationship

between phenylalanine ammonia-lyase levels and phenylpropa-

noid accumulation in transgenic tobacco identifies a rate-

determining step in natural product synthesis. Proc Natl Acad

Sci USA 91:7608–7612

Blount JW, Korth KL, Masoud SA, Rasmussen S, Lamb C, Dixon RA

(2000) Altering expression of cinnamic acid 4-hydroxylase in

transgenic plants provides evidence for a feedback loop at the

entry point into the phenylpropanoid pathway. Plant Physiol

122:107–116

Bradford MM (1976) A rapid and sensitive method for the

quantitation of microgram quantities of protein utilizing the

principle of protein-dye binding. Anal Biochem 72:248–254

Campbell MM, Sederoff RR (1996) Variation in lignin content and

composition, mechanisms of control and implications for the

genetic improvement of plants. Plant Physiol 110:3–13

Chen H, Chen F (2000) Effect of yeast elicitor on the secondary

metabolism of Ti-transformed Salvia miltiorrhiza cell suspen-

sion cultures. Plant Cell Rep 19:710–717

Chen F, Reddy MS, Temple S, Jackson L, Shadle G, Dixon RA

(2006) Multi-site genetic modulation of monolignol biosynthesis

suggests new routes for formation of syringyl lignin and wall-

bound ferulic acid in alfalfa (Medicago sativa L.). Plant J

48:113–124

Cochrane FC, Davin LB, Lewis NG (2004) The Arabidopsisphenylalanine ammonia lyase gene family: kinetic character-

ization of the four PAL isoforms. Phytochemistry 65:1557–

1564

Dixon RA, Paiva NL (1995) Stress-induced phenylpropanoid metab-

olism. Plant Cell 7:1085–1097

Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure from

small quantities of fresh leaf tissue. Phytochem Bull 19:11–15

Elkind Y, Edwards R, Mavandad M, Hedrick SA, Ribak O, Dixon

RA, Lamb CJ (1990) Abnormal plant development and down-

regulation of phenylpropanoid biosynthesis in transgenic tobacco

containing a heterologous phenylalanine ammonia-lyase gene.

Proc Natl Acad Sci USA 81:9057–9061

Gleave AP (1992) A versatile binary vector system with a T-DNA

organisational structure conducive to efficient integration of

cloned DNA into the plant genome. Plant Mol Biol 20:1203–

1207

Goujon T, Sibout R, Eudes A, Mackay J, Jouanin L (2003) Genes

involved in the biosynthesis of lignin precursors in Arabidopsisthaliana. Plant Physiol Biochem 41:677–687

Hahlbrock K, Scheel D (1989) Physiology and molecular biology of

phenylpropanoid metabolism. Annu Rev Plant Physiol Plant Mol

Biol 40:347–369

Higuchi M, Yoshizumi T, Kuriyama T, Hara H, Akagi C, Shimada H,

Matsui M (2009) Simple construction of plant RNAi vectors

using long oligonucleotides. J Plant Res 122:477–482

Hu P, Luo G, Zhao Z, Jiang Z (2005) Quality assessment of Radix

Salviae Miltiorrhizae. Chem Pharm Bull (Tokyo) 53:481–486

Hu YS, Zhang L, Di P, Chen WS (2009) Cloning and induction of

phenylalanine ammonia-lyase gene from Salvia miltiorrhiza and

its effect on hydrophilic phenolic acids levels. Chin J Nat Med

7:449–457

Liu AH, Li L, Xu M, Lin YH, Guo HZ, Guo DA (2006a)

Simultaneous quantification of six major phenolic acids in the

roots of Salvia miltiorrhiza and four related traditional Chinese

medicinal preparations by HPLC-DAD method. J Pharm Biomed

Anal 41:48–56

Liu R, Xu S, Li J, Hu Y, Lin Z (2006b) Expression profile of a PAL

gene from Astragalus membranaceus var mongholicus and its

crucial role in flux into flavonoid biosynthesis. Plant Cell Rep

25:705–710

Lois R, Dietrich A, Hahlbrock K, Schulz W (1989) A phenylalanine

ammonia-lyase gene from parsley: structure, regulation and

J Plant Res (2011) 124:183–192 191

123

identification of elicitor and light responsive cis-acting elements.

EMBO J 8:1641–1648

Lu Y, Foo LY (2002) Polyphenolics of Salvia—a review. Photo-

chemistry 59:117–140

Ma L, Zhang X, Guo H, Gan Y (2006) Determination of four

water-soluble compounds in Salvia miltiorrhiza Bunge by high-

performance liquid chromatography with a coulometric elec-

trode array system. J Chromatogr B Analyt Technol Biomed Life

Sci 833:260–263

MacDonald MJ, D’Cunha GB (2007) A modern view of phenylal-

anine ammonia lyase. Biochem Cell Biol 85:273–282

Mahesh V, Rakotomalala JJ, Le Gal L, Vigne H, de Kochko A,

Hamon S, Noirot M, Campa C (2006) Isolation and genetic

mapping of a Coffea canephora phenylalanine ammonia-lyase

gene (CcPAL1) and its involvement in the accumulation of

caffeoyl quinic acids. Plant Cell Rep 25:986–992

Murashige T, Skoog F (1962) A revised medium for rapid growth and

bioassays with tobacco tissue culture. Physiol Plant 15:473–497

Olsen KM, Lea US, Slimestad R, Verheul M, Lillo C (2008)

Differential expression of four Arabidopsis PAL genes; PAL1

and PAL2 have functional specialization in abiotic environmen-

tal-triggered flavonoid synthesis. J Plant Physiol 165:1491–1499

Petersen M, Simmonds MS (2003) Rosmarinic acid. Phytochemistry

62:121–125

Petersen M, Hausler E, Karwatzki B, Meinhard J (1993) Proposed

biosynthetic pathway for rosmarinic acid in cell cultures of

Coleus blumei Benth. Planta 189:10–14

Razzaque A, Ellis BE (1977) Rosmarinic acid production in Coleuscell cultures. Planta 137:287–291

Rohde A, Morreel K, Ralph J, Goeminne G, Hostyn V, De Rycke R,

Kushnir S, Van Doorsselaere J, Joseleau JP, Vuylsteke M, Van

Driessche G, Van Beeumen J, Messens E, Boerjan W (2004)

Molecular phenotyping of the pal1 and pal2 mutants of A. thalianareveals far-reaching consequences on phenylpropanoid, amino

acid, and carbohydrate metabolism. Plant Cell 16:2749–2771

Schmittgen TD, Livak KJ (2008) Analyzing real-time PCR data by

the comparative CT method. Nat Protoc 3:1101–1108

Sewalt V, Ni W, Blount JW, Jung HG, Masoud SA, Howles PA,

Lamb C, Dixon RA (1997) Reduced lignin content and altered

lignin composition in transgenic tobacco down-regulated in

expression of L-phenylalanine ammonia-lyase or cinnamate

4-hydroxylase. Plant Physiol 115:41–50

Shadle GL, Wesley SV, Korth KL, Chen F, Lamb C, Dixon RA

(2003) Phenylpropanoid compounds and disease resistance in

transgenic tobacco with altered expression of L-phenylalanine

ammonia-lyase. Phytochemistry 64:153–161

Song J, Wang Z (2009) Molecular cloning, expression and charac-

terization of a phenylalanine ammonia-lyase gene (SmPAL1)

from Salvia miltiorrhiza. Mol Biol Rep 36:939–952

Wanner LA, Li G, Ware D, Somssich IE, Davis KR (1995) The

phenylalanine ammonialyase gene family in Arabidopsis thali-ana. Plant Mol Biol 27:327–338

Wesley SV, Helliwell CA, Smith NA, Wang MB, Rouse DT, Liu Q,

Gooding PS, Singh SP, Abbott D, Stoutjesdijk PA, Robinson SP,

Gleave AP, Green AG, Waterhouse PM (2001) Construct design

for efficient, effective and high throughput gene silencing in

plants. Plant J 27:581–590

Yan Y, Wang Z (2007) Genetic transformation of the medicinal plant

Salvia miltiorrhiza by Agrobacterium tumefaciens-mediated

method. Plant Cell Tissue Organ Cult 88:175–184

Yan Q, Shi M, Ng J, Wu JY (2006) Elicitor-induced rosmarinic acid

accumulation and secondary metabolism enzyme activities in

Salvia miltiorrhiza hairy roots. Plant Sci 170:853–858

Zhang H, Yu C, Jia JY, Leung SW, Siow YL, Man RY, Zhu DY

(2002) Contents of four active components in different com-

mercial crude drugs and preparations of Danshen (Salviamiltiorrhiza). Acta Pharmacol Sin 23:1163–1168

Zhao SJ, Hu ZB, Liu D, Leung FC (2006) Two divergent members of

4-coumarate:coenzyme A ligase from Salvia miltiorrhiza Bunge:

cDNA cloning and functional study. J Integr Plant Biol 48:1355–

1364

Zheng W, Wang SY (2001) Antioxidant activity and phenolic

compounds in selected herbs. J Agric Food Chem 49:5165–5170

Zhou L, Zuo Z, Chow MS (2005) Danshen: an overview of its

chemistry, pharmacology, pharmacokinetics, and clinical use.

J Clin Pharmacol 45:1345–1359

192 J Plant Res (2011) 124:183–192

123