Você está na página 1de 10

The Plant Journal (2002) 31(6), 777786

TECHNICAL ADVANCE

Virus-induced gene silencing in tomato


Yule Liu, Michael Schiff and S. P. Dinesh-Kumar*
Department of Molecular, Cellular and Developmental Biology, OML 451, Yale University, PO Box 208104, New Haven,
CT 065208104, USA

Received 12 March 2002; revised 14 May 2002; accepted 21 May 2002.


*
For correspondence (fax 203-432-3854; e-mail savithramma.dinesh-kumar@yale.edu).

Abstract
We have previously demonstrated that a tobacco rattle virus (TRV)-based vector can be used in virus-
induced gene silencing (VIGS) to study gene function in Nicotiana benthamiana. Here we show that
recombinant TRV infects tomato plants and induces efcient gene silencing. Using this system, we
suppressed the PDS, CTR1 and CTR2 genes in tomato. Suppression of CTR1 led to a constitutive
ethylene response phenotype and up-regulation of an ethylene response gene, CHITINASE B. This
phenotype is similar to Arabidopsis ctr1 mutant plants. We have constructed a modied TRV vector
based on the GATEWAY recombination system, allowing restriction- and ligation-free cloning. Our
results show that tomato expressed sequence tags (ESTs) can easily be cloned into this modied vector
using a single set of primers. Using this vector, we have silenced RbcS and an endogenous gene
homologous to the tomato EST cLED3L14. In the future, this modied vector system will facilitate large-
scale functional analysis of tomato ESTs.

Keywords: tobacco rattle virus vector, gene silencing, tomato, tCTR1, functional genomics, GATEWAY
cloning.

Introduction

The ability to sequence plant genomes has resulted in the Waterhouse et al., 1998). A modication of this method
identication of large numbers of novel open reading uses a single-stranded self-complementary (hairpin) RNA,
frames (ORFs). Large-scale functional genomic approaches containing an intron, to suppress gene function (Smith
are necessary for converting this sequence information et al., 2000). However, all of these approaches rely on the
into functional information. Traditionally, T-DNA (Azpiroz- generation of transgenic lines, which, for high throughput
Leehan and Feldman, 1997; Krysan et al., 1999) and analyses, strictly limits their use to Arabidopsis.
transposon-based (Martienssen, 1998; Parinov et al., Virus-induced gene silencing (VIGS) offers an attractive
1999; Speulman et al., 1999; Tissier et al., 1999) insertional and quick alternative for knocking out expression of a gene
mutant populations have provided the resources for the without the need to genetically transform the plant. Using
analysis of phenotypes. Large collections of such insertion this method, recombinant virus carrying a partial sequence
and deletion mutant populations have been generated for of a host gene is used to infect the plant. When the virus
plants like Arabidopsis thaliana due to the ease of trans- spreads systemically, the endogenous gene transcripts,
formation (Clough and Bent, 1998). However, these mutant which are homologous to the insert in the viral vector
collections have limitations such as difculty in disrupting (VIGS-vector), are degraded by post-transcriptional gene
or tagging all genes, gene target bias, lack of phenotype silencing (PTGS) (Baulcombe, 1999). PTGS in plants bears
due to high degree of gene duplication in the plant similarity to quelling in Neurospora and RNA interference
genome, and loss of insertions that cause lethality. More (RNAi) in Caenorhabditis elegans, Drosophila, and mam-
recently, dsRNA-mediated suppression of genes by vec- mals (Waterhouse et al., 2001). PTGS functions via a
tors that produce sense and antisense transcripts has been sequence-specic RNA degradation mechanism that is
successfully employed (Chuang and Meyerowitz, 2000; triggered by double stranded RNA (dsRNA). Recently,
2002 Blackwell Science Ltd 777
778 Yule Liu et al.

poor. Some of the genes in tomato have been cloned


successfully using targeted transposon tagging (Bishop
et al., 1996; Jones et al., 1994; Keddie et al., 1996; Takken
et al., 1998; Van der Biezen et al., 1996). A reverse genetics
approach was also developed using Ds element insertions
creating a collection of 2932 families of a miniature tomato
called Micro-Tom (Meissner et al., 2000). However, the rate
of tagged mutations is very low in this population and only
10 families exhibited a clear mutant phenotype. Even
though many resources are available to study tomato
genes, an efcient reverse genetic tool to assess gene
function is lacking.
In this paper, we describe a tobacco rattle virus (TRV)-
based VIGS vector for efcient silencing of genes in
tomato. In addition, we describe a GATEWAY-based
modication of this vector to efciently clone tomato
ESTs in a high throughput manner for silencing.

Results and discussion

Recombinant TRV infects tomato


We have developed a TRV-based VIGS vector to study
gene function in N. benthamiana (Liu et al., 2002). We
assessed the ability of this vector to cause systemic
infection in tomato in order to employ it in VIGS. A
mixture of Agrobacterium cultures containing TRV-RNA1
(pTRV1) and TRV-RNA2 (pTRV2) T-DNA constructs (Figure
Figure 1. Recombinant TRV infects tomato.
1a,b) was inltrated onto the lower leaves of 3-week-old
(a,b) TRV based VIGS vectors described in Liu et al. 2002). TRV cDNA
clones were placed in between duplicated CaMV 35S promoter (2 3 35S) Lycopersicon esculentum cultivar VF36 (referred to as
and nopaline synthase terminator (NOSt) in a T-DNA vector. RdRp, RNA- VF36) or L. esculentum cultivar Micro-tom (referred to as
dependent RNA polymerase; 16K, 16 kDa cysteine rich protein; MP,
Micro-tom) plants (Scott and Harbaugh, 1989). Ten days
movement protein; CP, coat protein; LB and RB, left and right borders of
T-DNA; Rz, self-cleaving ribozyme; MCS, multiple cloning sites. post-Agro-inltration, total RNA was prepared from the
(c) Accumulation of genomic TRV RNA1 (gRNA1) and RNA2 (gRNA2) and upper un-inltrated leaves. RNA blots were hybridized
subgenomic RNA1a (sgRNA1a) and RNA1b (sgRNA1b) in the systemic
with probes derived from the 3 ends of RNA1 and RNA2.
leaves of recombinant TRV-infected tomato plants. Tomato plants were
inltrated with Agrobacterium alone (lane 1) or Agrobacterium- Genomic RNA1 and 2, and sub-genomic RNA1a and 1b
containing pTRV1 and pTRV2 (lane 2). Ten days after inltration, blots were detected only in those plants inltrated with
were prepared using 5 mg of total RNA and probed with 32P-labeled
Agrobacterium-containing TRV clones (Figure 1c; lane 1)
cDNA fragments corresponding to the 3 ends of RNA1 and RNA2. The
picture of the ethidium bromide stained gel shown below the blot and were absent in the control Agro-inltrated plants
demonstrates the equal loading of RNA. (Figure 1c; lane 2). These results clearly showed that
recombinant TRV can efciently replicate and spread
systemically in tomato plants.
RNAi was used efciently to study the function of all the
genes on Chromosomes I and III in C. elegans (Fraser et al.,
Silencing of the tomato PDS gene using TRV-VIGS vector
2000; Gonczy et al., 2000).
Many VIGS vectors have been described to study gene Next, we tested whether the TRV clones could induce gene
function in plants (Kjemtrup et al., 1998; Kumagi et al., silencing in tomato plants. In order to do so, we examined
1995; Liu et al., 2002; Ratcliff et al., 2001; Ruiz et al., 1998). the ability of our TRV-VIGS vector to suppress the expres-
Most of these vectors are designed to work in Nicotiana sion of the endogenous phytoene desaturase gene (PDS)
benthamiana. Tomato is an economically important crop in Micro-tom tomato. A mixture of Agrobacterium cultures
and like N. benthamiana, it belongs to the Solanaceous containing pTRV2, carrying tomato PDS (pTRV2-tPDS), and
family. A large number of expressed sequence tags (EST) pTRV1, was inltrated onto the lower leaves of 3-week-old-
are available for tomato. While the transformation of tomato plants (Figure 2a). PDS silencing in N. benthami-
tomato is feasible, the efciency of transformation is very ana inhibits carotenoid biosynthesis, causing the plants to

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


VIGS in tomato 779

Figure 2. Method of inoculation of


Agrobacterium-containing TRV clones onto
tomato plants.
Agrobacterium cultures containing pTRV1
and pTRV2 carrying tomato PDS (pTRV2-
tPDS) were mixed in a 1 : 1 ratio and
inltrated on to 3-week-old tomato plants
using 1 ml needleless syringe (a) or sprayed
using an artist's airbrush (b). The
silencing effect was visible 10 days after
Agrobacterium inltration or spray.

Figure 3. Silencing of the tomato PDS gene.


Infection of tomato plants with recombinant TRV alone (a) or TRV carrying the tomato PDS (TRV-tPDS) (b). Infection with TRV-tPDS silences endogenous
PDS in Micro-tom tomato plants and causes inhibition of carotenoid biosynthesis resulting in photo-bleaching phenotype (b).

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


780 Yule Liu et al.

exhibit a photo-bleached phenotype (Kumagi et al., 1995). plants are severely dwarfed and constitutively express
Tomato plants infected with pTRV-tPDS developed a ethylene inducible genes (Kieber et al., 1993). In tomato,
photo-bleached phenotype in the upper leaves 10 days there are two CTR1-like genes, tCTR1 (Wang and Li, 1997;
post-Agro-inltration and remained white for at least Zegzouti et al., 1999) and tCTR2 (AJ005077). tCTR1 is 58%
1 month (Figure 3). The Agrobacterium inltration method identical and 65% similar to Arabidopsis CTR1 at the
of infecting pTRV-tPDS resulted in the PDS silencing amino acid level. tCTR2 is 60% identical and 76% similar to
phenotype in only ve out of 10 tomato plants (50% tCTR1 in the C-terminus kinase domain but only 38%
efciency). In contrast, when this technique was applied in identical and 55% similar in the N-terminus non-kinase
N. benthamiana, all plants infected with pTRV-NbPDS domain. The tCTR2 kinase domain bears a high degree of
exhibited PDS silencing (Liu et al., 2002). homology to Arabidopsis EDR1 (85% identical) at the
To improve the silencing efciency, we tested a spray amino acid level. The Arabidopsis EDR1 gene encodes a
technique for the delivery of TRV into tomato plants putative MAPKKK and the mutant, edr1, has an elevated
(Figure 2b). The Agrobacterium mixture was sprayed onto resistance to Pseudomonas syringae and Erysiphe cichor-
3-week-old tomato plants using an artist's airbrush (see acearum (Frye et al., 2001). The biological functions of
the Experimental procedures section). This method tCTR1 and tCTR2 in tomato have not been examined.
resulted in substantial improvement in the silencing To examine whether tCTR1 and tCTR2 are true homo-
efciency. Of the 10 plants sprayed with pTRV-tPDS, nine logs of the Arabidopsis CTR1 gene, we silenced these
(90%) exhibited the PDS suppression phenotype. These genes in tomato and in N. benthamiana using the TRV-
results suggest that spraying Agrobacterium is more VIGS assay. Suppression of tCTR1 in VF36 tomato plants
effective than inltration in the induction of silencing in resulted in a constitutive ethylene response phenotype
tomato plants. Perhaps the Agrobacterium inltration similar to that observed in Arabidopsis. They were
method is not very efcient due to the compact architec- severely dwarfed compared with the non-silenced plants
ture of the young tomato leaves. Additionally, wounding (Figure 5a). On the other hand, the suppression of tCTR2
caused by the spray technique may mobilize T-DNA had no effect on plant growth or development (data not
transfer more effectively into the tomato cells. shown). Since tomato and tobacco share very high
Semi-quantitative RT-PCR was performed to conrm PDS sequence similarity, we investigated whether CTR1 homo-
silencing. The primers that anneal to the PDS gene outside logs of tobacco can induce a similar phenotype in N.
the region targeted for silencing were used. In pTRV-tPDS benthamiana. As expected, suppression of NbCTR1 leads
infected plants, the PDS message was reduced by more to a severe dwarf phenotype similar to that observed in
than 78% compared with the TRV infected controls (Figure tomato (Figure 5b) and NbCTR2 suppression had no effect
4b). The level of EF1a RNA was similar in TRV-tPDS and TRV (data not shown).
alone infected tissue and served as an internal control for To conrm the tCTR1 suppression at the molecular level,
RNA quality and RT-PCR amplication (Figure 4a). The level we performed semiquantitative RT-PCR. In TRV-tCTR1
of suppression of PDS in tomato by the TRV-VIGS vector is infected plants, the tCTR1 message was reduced by more
comparable with PDS silencing in N. benthamiana (data not than 81% compared to the controls infected with TRV alone
shown). The fact that TRV effectively caused the VIGS of (Figure 4c). In both tissue RNA samples, EF1a expression
PDS in tomato suggests that other nuclear genes could be levels were similar (data not shown) and served as an
targeted for silencing in a similar manner. internal control. Because the region targeted for silencing
tCTR1 has 70% similarity to tCTR2 at the nucleotide level,
we tested whether TRV-tCTR1 could also suppress tCTR2.
Silencing of the CTR1 homolog in tomato and N.
Semi-quantitative RT-PCR analysis-using primers that
benthamiana leads to a constitutive ethylene response
anneal to tCTR2, showed that the tCTR2 message level is
phenotype
not affected in the TRV-tCTR1 silenced plants (Figure 4d).
The phytohormone ethylene participates in a variety of To rule out the possibility that the absence of a develop-
physiological processes in plants including germination, mental phenotype in TRV-tCTR2 suppressed plants is due
cell elongation, ower and leaf senescence, sex determin- to lack of suppression of endogenous tCTR2, we performed
ation, fruit ripening and abscission, wounding and patho- semiquantitative RT-PCR analysis. In tCTR2 silenced plants
gen infection (Abeles et al., 1992; Johnson and Ecker, there was an 85% reduction of tCTR2 mRNA compared with
1998). In Arabidopsis, the CTR1 (constitutive triple the TRV infected control plants (Figure 4e). These results
response 1) gene encodes a Raf-like mitogen-activated suggest that tCTR2 is effectively silenced by VIGS although
protein kinase kinase kinase (MAPKKK) that functions there is no visible phenotype.
downstream of an ethylene receptor and negatively regu- The mutation in CTR1 in Arabidopsis causes constitutive
lates the ethylene response (Kieber et al., 1993). The ctr1 expression of CHITINASE B (CHIB), an ethylene inducible
loss-of-function mutation confers a phenotype in which gene (Kieber et al., 1993). We examined the level of CHIB

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


VIGS in tomato 781

Figure 4. RT-PCR analysis showing the


effect of VIGS on tPDS, tCTR1, tCTR2, tRbcS
and EST cLED3L14 transcription.
Ethidium bromide-stained agarose gels
showing RT-PCR products. The rst strand
cDNA was generated from total RNA
isolated from silenced and non-silenced
plants using an oligo (dT) primer and
reverse trascriptase. This rst strand cDNA
was used in a PCR reaction using gene
specic primers.
(a) Typical PCR products for EF1a derived
from TRV alone infected (left) and TRV-VIGS
vector-infected (right) tomato plants.
(b) Typical PCR products of Micro-tom
tomato plants, either PDS silenced (right) or
non-silenced, infected with TRV alone (left).
(c,d) Typical PCR products for tCTR1 (c)and
tCTR2 (d)derived from non-silenced TRV-
infected (left) and TRV-tCTR1 infected (right)
VF36 tomato plants.
(e, f, g) Typical PCR products for tCTR2,
tRbcS and EST cLED3L14 silenced (right) or
non-silenced TRV-alone infected (left) VF36
tomato plants, respectively.
Lanes 16 correspond to products from PCR
cycle number 15, 18, 21, 24, 27, and 30.
Lane 7 represents the control, in which the
RT reaction mix without reverse
transcriptase was used as a template in the
reaction. Lane M represents marker.

RNA expression in tCTR1 suppressed tomato plants using 29 000 unique sequence clones (Uni ESTs). Many of these
RNA blots hybridized with the CHIB gene. CHIB was not ESTs show homology to genes in Arabidopsis. The TRV-
detected in wild-type plants (Figure 5c, lane 1) and very VIGS approach described in this manuscript offers great
low expression was observed in non-silenced TRV infected promise for studying tomato Uni-ESTs function. However,
plants (Figure 5c, lane 3). However, in the tCTR1 silenced insertion of tomato ESTs into pTRV2 using a traditional
plants, the CHIB gene was highly expressed (Figure 5c, cloning method is labor-intensive and time-consuming.
lane 2). These results suggest that suppression of tCTR1 in Therefore, we modied the pTRV2 clone using the
tomato leads to constitutive expression of ethylene-regu- GATEWAY system (Invitrogen, CA, USA). The GATEWAY
lated genes. Taken together, our results show that the TRV- technology allows fast and easy cloning that is restriction
based VIGS vector can efciently phenocopy the effects of enzyme- and ligation-free. Consequently, the Uni ESTs can
mutations in different nuclear genes in tomato. be cloned en masse into this pTRV2 vector.
The construct, pTRV2-attP1-attP2, was generated to
facilitate en masse cloning of tomato ESTs (Figure 6a).
Modication of the TRV2 vector for high throughput
The PCR products anked by attB1 and attB2 sequences
cloning
directionally recombine in vitro at attP1 and attP2 sites
In the TIGR (The Institute of Genomic Research) database contained in the plasmid when incubated with the BP
there are over 100 000 tomato ESTs corresponding to CLONASE enzyme (Figure 6b). When this reaction mixture

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


782 Yule Liu et al.

is transformed into an E. coli strain such as DH10B, only


recombinants can grow because the ccdB gene, contained
in the plasmid, is lethal. The resulting recombinant
plasmid will contain anking attL1 and attL2 sequences
(Figure 6b). We cloned the N. benthamiana PDS gene into
the pTRV2-attP1-attP2 vector to test the efciency of
cloning and silencing. Efciency of cloning a PCR product
anked by attB1 and attB2 into this vector is about 95%.
However, when we tested this vector for the efciency of
silencing PDS gene in N. benthamiana, a very low level of
patchy suppression was observed (data not shown). RT-
PCR conrmed that the clone was infectious, but the PCR
fragment obtained from amplication of RNA extracted
from infected plants was smaller than expected (data not
shown). Since, both attL1 and attL2 sequences are direct
repeats of 100 bp, we reasoned that PDS anked by these
repeat sequences may have been deleted upon infection of
plants with the vector, and failure to suppress PDS would
result.
In order to overcome this deletion problem, we gener-
ated the vector pTRV2-attR2-attR1 (Figure 7a). In the rst
step, the PCR products anked by attB1 and attB2
sequences directionally recombine in vitro at attP1 and
attP2 sites on the pDONR-mod plasmid when incubated
with the BP CLONASE enzyme (Figure 7b). In the second
step, the intermediate attL1-attL2-containing pDONR-mod
vector directionally recombines with attR2 and attR1 sites
in the pTRV2 vector, when incubated with the LR
CLONASE enzyme (Figure 7b). Therefore, the resulting
plasmid has only a 23-bp anking sequence of attB1 and
attB2. Moreover, these sequences are not direct repeats.

Figure 5. Silencing of the tomato CTR1 and N. benthamiana CTR1 genes.


(a )tCTR1 silenced (right) and non-silenced (left) phenotype in VF36
tomato plants. Figure 6. Map of pTRV2 GATEWAY vector and schematic representation
(b) NbCTR1 silenced (right) and non-silenced (left) phenotype in N. of cloning PCR products.
benthamiana plants. (a) Modied pTRV2 vector based on the GATEWAY cloning technology
(c) Expression of CHIB gene in wild-type (lane 1), tCTR1-silenced (lane 2) containing attP1 and attP2 recombination sites.
and non-silenced (lane 3) VF36 tomato plants. Ten mg of total RNA (b) PCR products anked by attB1 and attB2 sites can be directly
isolated from these plants was blotted and probed with 32P-labeled CHIB recombined into attP1 and attP2 containing pTRV2 vector using BP
cDNA fragment. The picture of the ethidium bromide-stained gel shown CLONASE enzyme. The resulting attL1- and attL2-containing recombinant
below the blot demonstrates equal loading of RNA. clone is shown.

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


VIGS in tomato 783

The efciency of cloning a target sequence for silencing alone infected control plants (Figure 8). This phenotype is
into this vector is about 90%. similar to that reported by Ratcliff et al. (2001) in N.
To determine whether the pTRV2-attR2-attR1 vector benthamiana. Furthermore, semiquantitative RT-PCR
efciently mediates gene silencing similar to original analysis suggests that the endogenous tRbcS mRNA is
pTRV2 vector, we cloned PDS and the small sub-unit of reduced by 76% compared with the TRV infected control
the ribulose bisphosphate carboxylase (tRbcS). Compared plants (Figure 4F).
with PDS, the RbcS is encoded by a multigene family and In order to demonstrate that tomato ESTs can be
is expressed in abundance. The efciency of silencing PDS easily cloned into the pTRV2-attR2-attR1 destination
was similar to that of the original pTRV2 vector (data not vector, we cloned 10 tomato ESTs that bear homology
shown). The tRbcS silenced plants developed pale yellow to a serine/threonine kinase (The Institute of Genomic
leaves 12 days post-Agro-inltration compared to the TRV- Research). Primer sequences were designed that

Figure 7. Schematic representation of


cloning tomato ESTs into the pTRV2-attR2-
attR1 vector.
(a) Modied pTRV2 vector based on the
GATEWAY cloning technology containing
attR1 and attR2 recombination sites.
(b) PCR products anked by attB1 and attB2
sites can be directly recombined into attP1-
and attP2-containing pDONR-mod using BP
CLONASE enzyme. The resulting attL1- and
attL2-containing pDONR-mod vector can be
recombined with pTRV2 containing attR1
and attR2 using LR CLONASE enzyme. The
resulting attB1- and attB2-containing
recombinant clone is shown.
(c) Tomato EST products generated by PCR
using a common set primer with attB1 and
attB2 sites shown in (b). M, marker; 110,
different tomato EST PCR products.
(d) Restriction digestion of plasmids
prepared from two independent colonies
recovered from E. coli transformed with
each recombination reaction mixture
containing pTRV2-attR2-attR1 and tomato
EST PCR products. Plasmids were cut with
XbaI-SacI to release the inserts. Some
inserts do not match the PCR product size
shown in (c) because of internal restriction
sites.

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


784 Yule Liu et al.

Figure 8. Silencing of the tomato RbcS using the pTRV2 GATEWAY vector.
Infection of VF36 tomato plants with recombinant TRV GATEWAY alone (a) or TRV GATEWAY carrying the tomato RbcS (TRV-tRbcS) (b). Infection with
TRV-tRbcS silences endogenous RbcS and causes development of pale yellow leaves (b).

annealed to vector sequences anking the tomato ESTs. Conclusions


The attB1 sequence was included in the forward primer
Gene silencing methods that do not rely on transformation
and the attB2 sequence was included in reverse primer
offer a tremendous advantage for gene function analysis. In
(Figure 7b). These primer sets were then used to
this report we show that recombinant TRV infects tomato
amplify ESTs by PCR (Figure 7c). The PCR products
plants and can be used to silence genes efciently. Our
with terminal attB1 and attB2 sequences were incubated
results suggest that a spray technique to deliver
with the pDONR-mod vector containing the attP1 and
Agrobacterium to tomato plant cells is better for obtaining
attP2 recombination site and the BP CLONASE enzyme.
high efciency silencing compared with the routinely used
The pTRV2-attR2-attR1 destination vector containing the
Agrobacterium inltration method. Using this TRV-based
attR1 and attR2 recombination site and the LR
VIGS system, we demonstrate that tCTR1 is a true homolog
CLONASE enzyme were added. This mixture was trans-
of the Arabidopsis CTR1 gene. Suppression of tCTR1 in
formed into DH10B chemical competent cells and
tomato and N. benthamiana induces a constitutive ethylene
selected on kanamycin-containing LB plates. In our response phenotype, while suppression of tCTR2 does not.
hands, this single-tube protocol for cloning attB-PCR Even though tCTR1 and tCTR2 share up to 60% sequence
products directly into pTRV2-attR2-attR1 resulted in a similarity at the nucleotide level, suppression of tCTR1 using
95100% success rate. Inserts containing clones were the VIGS vector had no effect on tCTR2 gene expression.
veried by restriction enzyme digestion (Figure 7d) and Finally, we have modied the TRV2 vector using the
by sequencing the vector-insert junctions. These results GATEWAY recombination system in order to clone tomato
suggest that tomato ESTs can be cloned into the ESTs en masse with one set of primers. This provides a rapid
pTRV2-attR2-attR1 vector for silencing en masse. way to test tomato EST function. Using this vector, we show
In order to demonstrate that the above described TRV-EST that tRbcS and an endogenous gene corresponding to
clones can be used for silencing corresponding endogenous tomato EST cLED3L14 can be successfully silenced.
genes, we silenced EST cLED3L14 (TIGR) (corresponding Therefore, the modied TRV2 vector will facilitate produc-
PCR product and the clone is shown in Figures 7c,d; lane 2). tion of normalized cDNA libraries or the cloning of large sets
This EST shows highest homology to potato protein kinase of genes for large-scale functional genomics in the future.
StCPK1 (Lakatos et al., 1998). Suppression of the endogen-
ous gene corresponding to this EST showed no visible
phenotype (data not shown) even though the mRNA is Experimental procedures
reduced by 82% compared with the TRV-alone control
(Figure 4g). These results indicate that the tomato EST Plasmid construction
clones in TRV-attR2-attR1 vector can be used successfully to pTRV1 and pTRV2 VIGS vectors have been described in (Liu et al.,
silence corresponding endogenous genes in tomato. 2002).

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


VIGS in tomato 785

pTRV2-tPDS: a 409-bp fragment of PDS cDNA fragment corres- TGC TTC CTC TGT CAT TTC TTC AGC-3 and 5- GGG GAC CAC
ponding to bases 8581266 of tomato PDS gene was PCR TTT GTA CAA GAA AGC TGG GTC CAC TTG ACG CAC ATT GTC
amplied from tomato VF36 cDNA using Taq DNA polymerase GAA TCC-3. This PCR product was recombined into pTRV2-attR2-
and the primers 5-CGG TCT AGA GGC ACT CAA CTT TAT AAA attR1 vector as described above for cloning PDS.
CC-3 and 5-CGG GGA TCC CTT CAG TTT TCT GTC AAA CC-3. pTRV2-attB2-tomato ESTs-attB1: 10 tomato ESTs that bear
The resulting PCR product was cloned into XbaI-BamHI-cut homology to serine/threonine kinases were amplied by PCR
pTRV2. using a forward primer containing the attB1 sequence and a
pTRV2-tCTR1: a 690-bp fragment of tCTR1 cDNA fragment reverse primer containing the attB2 sequence, which anneals to
corresponding to bases 19062595 of tomato CTR1 (Wang and Li, the vector pBluescript SK () containing tomato ESTs. The
1997) was PCR amplied from tomato VF36 cDNA using Taq DNA forward primer is 5-G GGG ACA AGT TTG TAC AAA AAA GCA
polymerase and the primers 5-CGG GAA TTC GTT GCA ATT ATG GGC TCC CCC GGG CTG CAG GAA TTC-3 and the reverse primer
AAG CGG TTG CG-3 and 5- CGG CTC GAG TCA TGA GAG CAA is G GGG ACC ACT TTG TAC AAG AAA GCT GGG TGG TAC CGG
CTG CAT GTC TG T-3. The resulting PCR product was cloned into GCC CCC CCT CGA G-3. The resulting PCR products with terminal
EcoRI-XhoI-cut pTRV2. attB1 and attB2 sequences were precipitated and incubated with
pTRV2-tCTR2: a 537-bp fragment of CTR2 cDNA fragment pDONR-mod vector containing the attP1 and attP2 recombination
corresponding to bases 25063042 of tomato CTR2 (GenBank sites and the BP CLONASE enzyme. To this, the pTRV2-attR2-
#AJ005077) was PCR amplied from tomato VF36 cDNA using attR1 destination vector containing the attR1 and attR2 recombi-
Taq DNA polymerase and the primers 5-CGG GAA TTC GCC CTT nation sites and the LR CLONASE enzyme was added. This
GAT GTG GCA AAG GGC AT 3 and 5-CGG CTC GAG GTA GAA mixture was transformed into DH10B chemical competent cells
TTT ACT GAG ATT TCC TG-3. The resulting PCR product was and selected on kanamycin-containing LB plates. Clones were
cloned into EcoRI-XhoI-cut pTRV2. veried by restriction enzyme digestion and by sequencing the
pTRV2-NbCTR1: a 690-bp of CTR1 cDNA fragment was PCR vector-insert junctions.
amplied from N. benthamiana cDNA using Taq DNA polymerase
and the primers used to amplify tCTR1. The resulting PCR product
was cloned into EcoRI-XhoI-cut pTRV2. Agro-inltration and spray
pTRV2-NbCTR2: a 537-bp of CTR2 cDNA fragment was PCR
N. benthamiana and tomato plants were grown in pots at 25C in
amplied from N. benthamiana cDNA using Taq DNA polymerase
a growth chamber under 16 h light/8 h dark cycle with 60%
and the primers used to amplify tCTR2. The resulting PCR product
humidity. For the VIGS assay, pTRV1 or pTRV2 and its derivatives
was cloned into EcoRI-XhoI-cut pTRV2.
were introduced into Agrobacterium strain GV3101 by electro-
pTRV2-attP1-attP2: The DNA fragment containing attP1-ccdB-
poration (BIO-RAD, Hercules, CA, USA). A 5-ml culture was grown
CmR.-attP2 amplied from pDONR201 (Invitrogen, Carlsbad, CA,
overnight at 28C in the appropriate antibiotic selection medium.
USA) using primers 5-CGG GAA TTC TAG AGG CGC GCC AAA
The next day, the culture was inoculated into a 50-ml LB medium
TAA TGA TTT TAT TTT GAC TGA TAG TGA C-3 and 5-C GGC
containing antibiotics, 10 mM MES and 20 mM acetosyringone.
TCG AGA GCT CAA ATA ATG ATT TTA TTT TGA CTG ATA GTG
The culture was grown overnight in a 28C shaker. Agrobacterium
AC-3. This PCR product was cloned into EcoRI-XhoI-cut pTRV2
cells were harvested and resuspended in inltration media
(Liu et al., 2002).
(10 mM MgCl2, 10 mM MES, 200 mM acetosyringone), adjusted
pDONR-mod: This vector was generated by deleting the frag-
to an O.D. of 2.0 and left at room temperature for 3 h.
ment containing the kanamycin gene between PvuI and NruI of
Agrobacterium was inltrated using a needle less 1 ml syringe
pDONR201 (Invitrogen) and religating the vector.
or sprayed using an artist's airbrush (Paasche, Harwood Heights,
pTRV2-attR2-attR1: HindIII-DraIII(T4 DNA polymerase treated)
IL, USA, model VL80) connected to a portable air compressor
fragment of pYL156 (Liu et al., 2002) containing 2xCaMV promoter (Campbell Havsfeld, Harrison, OH, USA) set at 75 psi. Plants were
and TRV-RNA2 cDNA with the NOS terminator was cloned into left covered overnight.
pBin19 to obtain pYL276. pTRV2-attR1-attR2 was obtained by
inserting the GATEWAY conversion cassette B (Invitrogen) into
pYL276 at the StuI site.
pTRV2-attL1-NbPDS-attL2 and pYL157: N. benthamiana PDS RNA isolation, Northern blot and RT-PCR analysis
cDNA containing the attB1 and attB2 sequences was obtained by Total RNA was extracted from silenced and non-silenced tomato
PCR amplication using the TRV-PDS clone (Liu et al., 2002) as plants using the RNAwiz solution (Ambion, Austin, TX, USA) and
template and primers: 5-G GGG ACA AGT TTG TAC AAA AAA treated with RNase-free DNase (Gene Hunter, Nashville, TN,
GCA GGC TCT GAC GAG CTT TCG ATG CAG-3 and 5-GGG GAC USA). First strand cDNA was synthesized using 1 mg of total RNA,
CAC TTT GTA CAA GAA AGC TGG GTA TAT ATG GAC ATT TAT oligo d(T)primer and superscript reverse transcriptase
CAC A 3. This PDS PCR product was recombined into pTRV2- (Invitrogen). Semi-quantitative RT-PCR was performed as de-
attP1-attP2 using the BP CLONASE enzyme reaction (Invitrogen). scribed in (Burton et al., 2000; Liu et al., 2002). For RT-PCR,
pTRV2-attB2-NbPDS-attB1: N. benthamiana PDS cDNA contain- primers that anneal outside the region targeted for silencing were
ing the attB1 and attB2 sequences was obtained by PCR, as used to ensure that only the endogenous gene was being tested.
described above. This PDS PCR product was recombined into The intensities of PCR-generated fragments were analyzed and
pDONR-mod vector containing the attP1 and attP2 recombination quantied using Gel Doc 2000 and Quantity One Version 4.3 (BIO-
sites using the BP CLONASE enzyme. To this, the pTRV2-attR1- RAD, CA).
attR2 destination vector and the LR CLONASE enzyme was added. RNA blots were prepared using 5 or 10 mg of total RNA. To
This mixture was transformed into DH10B chemical competent conrm TRV infection, RNA blots were hybridized with a probe
cells and selected on kanamycin-containing LB plates. derived from the 3-end of TRV RNA1 (bases 53516791) and
pTRV2-attB2-tRbcS-attB1: A 500-bp tRbcS cDNA containing the RNA2 (bases 12452103). To determine the CHIB message level, a
attB1 and attB2 sequences was obtained by PCR using VF36 cDNA fragment of CHIB was PCR amplied from tomato cDNA using
and primers 5-GGG GAC AAG TTT GTA CAA AAA AGC AGG CTC primers 5-ACT GTT TCC TTA GAG AGC AAG GTA G-3 and

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786


786 Yule Liu et al.

5-CAA CTA ATA GTC CGT TTC CAA AAG ACC-3 and this Gonczy, P., Echeverri, C., Oegema, K. et al. (2000) Functional
fragment was used as a probe. genomic analysis of cell division in C. elegans using RNAi of
genes of chromosome III. Nature, 408, 331336.
Johnson, P.R. and Ecker, J.R. (1998) The ethylene gas signal
Acknowledgements transduction pathway: a molecular perspective. Annu. Rev.
Genet. 32, 227254.
We thank Janet Stewart for editing the manuscript. We thank the
Jones, D.A., Thomas, C.M., Hammond-Kosack, K.E., Balint-Kurti,
members of the S.P.D-K lab for thoughtful comments and critical
P.J. and Jones, J.D.G. (1994) Isolation of the tomato Cf-9 gene
reading of the manuscript. The National Science Foundation Plant
for resistance to Cladosporium fulvum by transposon tagging.
Genome Grant DBI-0077510 to S.P.D-K supported this work.
Science, 266, 789793.
Keddie, J.S., Carroll, B., Jones, J.D.G. and Gruissem, W. (1996)
References The DCL gene of tomato is required for chloroplast
development and palisade cell morphogenesis in leaves.
Tissier, A.F., Marillonnet, S., Klimyuk, V., Patel, K., Torres, M.A., EMBO J. 15, 42084217.
Murphy, G. and Jones, J.D.J. (1999) Multiple independent Kieber, J.J., Rothenberg, M., Roman, G., Feldmann, K.A. and
defective suppressor-mutator transposon insertions in Ecker, J.R. (1993) CTR1, a negative regulator of the ethylene
Arabidopsis: a tool for functional genomics. Plant Cell, 11, response pathway in Arabidopsis, encodes a member of the Raf
18411852. family of protein kinases. Cell, 72, 427441.
Van der Biezen, E., Brandwagt, B., van Leeuwen, W., Nijkamp, H. Kjemtrup, S., Sampson, K.S., Peele, C.G., Nguyen, L.V. and
and Hille, J. (1996) Identication and isolation of the FEEBLY Conkling, M.A. (1998) Gene silencing from plant DNA carried by
gene from tomato by transposon tagging. Mol. General Genet. a geminivirus. Plant J. 14, 91100.
12, 267280. Krysan, P.J., Young, J.C. and Sussman, M.R. (1999) T-DNA as an
Wang, Y. and Li, N. (1997) A cDNA sequence isolated from the insertional mutagen in Arabidopsis. Plant Cell, 11, 22832290.
ripening tomato fruit encodes a putative protein kinase. Plant Kumagi, M.H., Donson, J., Della-Cioppa, G., Harvey, D., Hanley, K.
Physiol. 114, 1135. and Grill, L.K. (1995) Cytoplamic inhibition of carotenoid
Waterhouse, P.M., Graham, M., W. and Wang, M.-B. (1998) Virus biosynthesis with viru-derived RNA. Proc. Natl Acad. Sci. 92,
resistance and gene silencing in plants is induced by double 16791683.
stranded RNA. Proc. Natl Acad. Sci. 95, 1395913964. Lakatos, L., Hutvagner, G. and Banfalvi, Z. (1998) Potato protein
Waterhouse, P.M., Wang, M.-B. and Lough, T. (2001) Gene kinase StCPK1: a putative evolutionary link between CDPKs and
silencing as an adaptive defense against viruses. Nature, 411, CRKs. Biochim. Biophys. Acta, 1442, 101108.
834842. Liu, Y., Schiff, M., Marathe, R. and Dinesh-Kumar, S.P. (2002)
Zegzouti, H., Jones, B., Frasse, P., Marty, C., Maitre, B., Latche, A., Tobacco Rar1, EDS1 and NPR1/NIM1 like genes are required for
Pech, J.-C. and Bouzayen, M. (1999) Ethylene-regulated gene
N-mediated resistance to tobacco mosaic virus. Plant J. 30,
expression in tomato fruit: characterization of novel ethylene-
415429.
responsive ripening related genes isolated by differential
Martienssen, R.A. (1998) Functional genomics: Probing plant
display. Plant J. 18, 589600.
gene function and expression with transposons. Proc. Natl
Abeles, F.B., Morgan, P.W. and Salveit, M.E. (1992) Ethylene in
Acad. Sci. 95, 20212026.
Plant Biology 2nd edn. New York, USA: Academic Press.
Meissner, R., Chague, V., Zhu, Q., Emmanuel, E., Elkind, Y. and
Azpiroz-Leehan, R. and Feldman, K.A. (1997) T-DNA insertion
Levy, A.A. (2000) A high throughput system for transposon
mutagenesis in Arabidopsis: Going back and forth. Trends
tagging and promoter trapping in tomato. Plant J. 22, 265274.
Genet. 13, 152156.
Parinov, S., Sevugan, M.YeD., Yang, W.-C., Kumaran, M. and
Baulcombe, D.C. (1999) Fast forward genetics based on virus-
Sundaresan, V. (1999) Analysis of anking sequences from
induced gene silencing. Curr. Op Plant Biol. 2, 109113.
Bishop, G.J., Harrison, K. and Jones, J.D.G. (1996) The tomato Dissociation insertion lines: a database for reverse genetics in
Dwarf gene isolated by heterologous transposon tagging Arabidopsis. Plant Cell, 11, 22632270.
encodes the rst member of a new cytochrome P450 family. Ratcliff, F., Martin-Hernandez, A.M. and Baulcombe, D.C. (2001)
Plant Cell, 8, 959969. Tobacco rattle virus as a vector for analysis of gene function by
Burton, R.A., Gibeaut, D.M., Bacic, A., Findlay, K., Roberts, K., silencing. Plant J. 25, 237245.
Hamilton, A., Baulcombe, D.C. and Fincher, G.B. (2000) Virus- Ruiz, M.T., Voinnet, O. and Baulcombe, D.C. (1998) Initiation and
induced silencing of a plant cellulose synthase gene. Plant Cell, maintenance of virus-induced gene silencing. Plant Cell, 10,
12, 691705. 937946.
Chuang, C.F. and Meyerowitz, E.M. (2000) Specic and heritable Scott, J.W. and Harbaugh, B.K. (1989) Micro-Tom a miniature
genetic interference by double-stransded RNA in Arabidopsis dwarf tomato. Florida Agr. Expt. Sta. Circ. 370, 16.
thaliana. Proceedings of the Natl Acad. Sci. 97, 49854990. Smith, N.A., Singh, S.P., Wang, M.-B., Stoutjesdijk, P.A., Green,
Clough, S.J. and Bent, A.F. (1998) Floral dip: a simplied method A.G. and Waterhouse, P.M. (2000) Total silencing by intron-
for Agrobacterium-mediated transformation of Arabidopsis spliced hairpin RNAs. Nature, 407, 319320.
thaliana. Plant J. 16, 735743. Speulman, E., Metz, P.L.J., van Arkel, G., Hekkert, B.L., Stiekema,
Fraser, A.G., Kamath, R.S., Zipperlen, P., Martinez-Campos, M., W.J. and Pereira, A. (1999) A two-component enhancer-
Sohrmann, M. and Ahringer, J. (2000) Functional genomic inhibitor transposon mutagenesis system for functional
analysis of C. elegans chormosome I by systemic RNA analysis of the Arabidopsis genome. Plant Cell, 11, 18531866.
interference. Nature, 408, 325330. Takken, F., Schipper, D., Nijkamp, H. and Hille, J. (1998)
Frye, C.A., Tang, D. and Innes, R.W. (2001) Negative regulation of Identication and Ds-tagged isolation of a new gene at the Cf-
defense responses in plants by a conserved MAPKK kinase. 4 locus of tomato involved in disease resistance to
Proc. Natl Acad. Sci. 98, 373378. Cladosporium fulvum race 5. Plant J. 14, 401411.

Blackwell Science Ltd, The Plant Journal, (2002), 31, 777786