bestatin, an inhibitor of aminopeptidases, provides a chemical genetics approach to dissect
TRANSCRIPT
Bestatin, an Inhibitor of Aminopeptidases, Providesa Chemical Genetics Approach to Dissect JasmonateSignaling in Arabidopsis1[W][OA]
Wenguang Zheng, Qingzhe Zhai, Jiaqiang Sun, Chang-Bao Li, Lei Zhang, Hongmei Li, Xiaoli Zhang,Shuyu Li, Yingxiu Xu, Hongling Jiang, Xiaoyan Wu, and Chuanyou Li*
State Key Laboratory of Plant Genomics and Center for Plant Gene Research, Institute of Geneticsand Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China (W.Z., Q.Z., J.S.,C.-B.L., L.Z., H.L., X.Z., S.L., Y.X., H.J., X.W., C.L.); Graduate School of the Chinese Academyof Sciences, Beijing 100039, China (W.Z., L.Z., H.L., X.Z., S.L., Y.X.); and Horticulture College (Q.Z.)and Key Laboratory of Crop Biology, Agronomy College (C.-B.L., X.W.), Shandong Agricultural University,Taian 271018, China
Bestatin, a potent inhibitor of some aminopeptidases,was shownpreviously to be apowerful inducer ofwound-response genes intomato (Lycopersicon esculentum). Here, we present several lines of evidence showing that bestatin specifically activates jasmonicacid (JA) signaling in plants. First, bestatin specifically activates the expression of JA-inducible genes in tomato and Arabidopsis(Arabidopsis thaliana). Second, the induction of JA-responsive genes by bestatin requires the COI1-dependent JA-signalingpathway, but does not depend strictly on JA biosynthesis. Third, microarray analysis using Arabidopsis whole-genome chipdemonstrates that the gene expression profile of bestatin-treated plants is similar to that of JA-treated plants. Fourth, bestatinpromotes a series of JA-related developmental phenotypes. Taken together, the unique action mode of bestatin in regulating JA-signaled processes leads us to the hypothesis that bestatin exerts its effects through the modulation of some key regulators in JAsignaling.We have employed bestatin as an experimental tool to dissect JA signaling through a chemical genetic screening, whichyielded a collection of Arabidopsis bestatin-resistant (ber) mutants that are insensitive to the inhibitory effects of bestatin on rootelongation. Further characterization efforts demonstrate that some ber mutants are defective in various JA-induced responses,which allowed us to classify the ber mutants into three phenotypic groups: JA-insensitive ber mutants, JA-hypersensitive bermutants, and mutants insensitive to bestatin but showing normal response to JA. Genetic and phenotypic analyses of the bermutants with altered JA responses indicate that we have identified several novel loci involved in JA signaling.
Plants defend themselves against insect attack,wounding, and some pathogen infection by activat-ing the expression of genes involved in herbivoredeterrence, wound healing, and other defense-relatedprocesses. A fascinating feature of induced defenseresponses is their occurrence both locally at the site ofwounding and systemically in undamaged partsthroughout the plant (Green and Ryan, 1972). Wound-inducible proteinase inhibitors (PIs) in tomato (Lyco-persicon esculentum) provide an attractive model
system in which to identify the proposed woundsignals for defense genes expression. Among the iden-tified candidate signals that can activate wound-induced PIs expression in tomato are systemin, an18-amino acid peptide, and jasmonic acid (JA), anoxylipin-derived phytohormone (Farmer and Ryan,1990, 1992; Pearce et al., 1991; Constabel et al., 1995;Bowles, 1998; Walling, 2000; Leon et al., 2001; Browse,2005). A wealth of evidence indicates that systeminand JA actually work together in the same signalingpathway to promote systemic expression of PIs andother defense-related genes (Constabel et al., 1995;Bowles, 1998). However, relatively little is knownabout the specific roles of systemin and JA in long-distance wound signaling. Recently, a series of graftingexperiments conducted with tomato mutants defectivein JA biosynthesis or signaling and, as a consequence,compromised in systemic defense responses led to thecurrent notion that systemin acts at or near the site ofwounding (locally) to amplify the production of JA,which in turn acts as a long-distance wound signal topromote defense genes expression (Li et al., 2002; Ryanand Moura, 2002; Turner et al., 2002; Stratmann, 2003;Howe, 2004; Schilmiller and Howe, 2005). These stud-ies challenge the previous paradigm that systemin is
1 This work was supported by the National Natural ScienceFoundation of China (grant nos. 30425033 and 30530440 to C.L.) andthe Chinese Academy of Sciences (grant no. CXTD–S2005–2 to C.L.).
* Corresponding author; e-mail [email protected]; fax 8610–6487–3428.
The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Chuanyou Li ([email protected]).
[W] The online version of this article contains Web-only data.[OA] Open Access articles can be viewed online without a sub-
scription.Article, publication date, and citation information can be found at
www.plantphysiol.org/cgi/doi/10.1104/pp.106.080390.
1400 Plant Physiology, August 2006, Vol. 141, pp. 1400–1413, www.plantphysiol.org � 2006 American Society of Plant Biologists
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
the sole long-distance wound signal and ascribe JA acentral role in regulating systemic defense responsesin virtually all species throughout the plant kingdom.In addition to defense, the jasmonate family of signal-ing molecules, including JA, methyl JA (MeJA), andother bioactive derivatives (collectively referred tohere as jasmonates), also play important roles in regu-lating developmental processes such as root growth,tuberization, senescence, and reproduction (Creelmanand Mullet, 1997; Turner et al., 2002; Wasternack andHause, 2002; Devoto and Turner, 2003; Browse, 2005).In spite of the importance of jasmonates as stressand growth regulators is well recognized, relativelylittle is known about the molecular mechanisms gov-erning JA perception and subsequent signaling eventsleading to activation of downstream JA-responsivegenes.Conventional forward genetic screens for JA-
resistant (insensitive) mutants have led to the identi-fication of a few molecular players in the JA-signalingnetwork, includingubiquitin-mediatedproteolysisma-chinery (Turner et al., 2002; Devoto and Turner, 2003;Browse, 2005). A significant advance in our under-standing of JA signaling came from the analysis of theArabidopsis (Arabidopsis thaliana) coronatine insensitive1 (coi1) mutant that is insensitive to JA. The identifi-cation of COI1 encoding an F-box protein (Xie et al.,1998) and the demonstration that COI1 interacts withSkp1 and Cullin1 to assemble a functional SCFCOI1
ubiquitin ligase complex in vivo (Devoto et al., 2002;Xu et al., 2002) suggest that a ubiquitin-mediatedprotein degradation machinery is involved in JA sig-naling. Significantly, it was demonstrated that loss offunction of a tomato homolog of COI1 (LeCOI1) resultsin a JA-insensitive phenotype both in terms of defenseresponse and development, suggesting the existenceof a COI1 function is conserved among plant speciesother than Arabidopsis (Li et al., 2004). Moreover,plants deficient in the general regulators of SCFcomplexes, such as Cullin1 (Ren et al., 2005), SGT1b(Gray et al., 2003), AXR1 (Leyser et al., 1993), and CSN(Feng et al., 2003), also show impaired JA responses.Taken together, accumulating evidence shows thatthe ubiquitin-mediated protein degradation process,which also has been demonstrated in many differenthormone-signaling pathways, is a central regulatorymechanism in JA signaling.By classic forward genetic screening, other impor-
tant components of the JA-response pathway have alsobeen identified. Compared with coi1, the JA-resistantmutants jar1 (Staswick et al., 1992) and jin1 (Bergeret al., 1996) show a relatively weak phenotype inresponse to JA. JAR1 encodes an enzyme that has JAadenylation activity to form JA-amino acid conjugates,suggesting covalent modification of JA plays a keyrole in its action (Staswick et al., 2002). JIN1 encodesa nuclear-localized bHLH-type transcription factorknown as AtMYC2 (Lorenzo et al., 2004). The tomatohomologs of AtMYC2, named JAMYC2 and JA-MYC10, specifically recognize a T/G box in the pro-
moters of JA-responsive genes and activate theirexpression (Boter et al., 2004). To identify negativeregulators of JA signaling, several groups conductedgenetic screens for constitutive JA-response mutants(Ellis and Turner, 2001; Hilpert et al., 2001; Xu et al.,2001). To date, only one of these genes (CEV1) has beenidentified. The finding that CEV1 encodes a cellulosesynthase suggests a link between cell wall biosynthesisand JA signaling (Ellis et al., 2002).
However, compared to the five classic and otherplant hormones, our knowledge of the JA-signalingpathway is still very limited. For example, extensive con-ventional forward genetic screens failed to identify theJA receptor(s) and the substrate(s) of the SCFCOI1
ubiquitin ligase complex, two of the most critical andapparent gaps in the understanding of JA signal-ing. The inability to identify the JA receptor(s), SCFCOI1
substrates, and other components in forward geneticscreens suggests functional redundancy or lethalmutation in the JA perception and other signalingapparatus. With this in mind, we are looking for analternative approach that may complement the limi-tations of conventional genetics to further dissect JAsignaling. So-called ‘‘chemical genetics’’ is a powerfulapproach to dissect biological processes that may beintractable using conventional genetics because of genefunction redundancy or lethality (Surpin et al., 2005).Chemical genetics is centered around the tenet thatsmall organic molecules can be used like mutations inclassic genetics to modulate protein functions and toassist in delineation of biological pathways (Blackwelland Zhao, 2003). Once small molecules that affect(inhibit or activate) a biological process of interest areidentified, integration of the small molecule-basedforward chemical genetics with the ever-growing ge-nomics tools in model plant system (for example,Arabidopsis) will significantly facilitate the identifica-tion of relevant gene products, including targets of theactive compounds, which are intractable by the classicgenetics approach. Indeed, recent work has demon-strated the feasibility of this approach in studies ofseveral biological pathways, including auxin signal-ing (Blackwell and Zhao, 2003; Zhao et al., 2003;Armstrong et al., 2004; Dai et al., 2005), brassinosteroidsignaling (Asami et al., 2003), vacuole sorting (Zouharet al., 2004), endomembrane trafficking, and gravitrop-ism (Surpin et al., 2005).
To identify new components in JA signaling, wehave developed a forward chemical genetic screen,which combines the advantages of chemical geneticsand the classic genetic screen, to identify novel mu-tants with altered JA responses. Bestatin, a potentinhibitor of some aminopeptidases in plants and an-imals, was shown previously to be a powerful inducerof wound-response genes in tomato (Schaller et al.,1995). Bestatin has been widely used as an experimen-tal tool in elucidating the physiological role of somemammalian exopeptidases in the regulation of theimmune system, in the growth of tumors, and in thedegradation of cellular proteins (Scornik and Botbol,
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1401
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
2001). We have conducted a series of genetic and mole-cular analyses in tomato and Arabidopsis, two ofthe most extensively studied model systems for JAresponses, showing that bestatin specifically modu-lates JA signaling. Bestatin specifically activates theexpression of JA-inducible defense-related genes, andthe induction kinetics of these genes in response tobestatin are similar to those in response to JA. Fur-thermore, the action of bestatin to activate defensegene expression requires the function of the F-boxprotein COI1/Jai1, a central regulator of JA-signaling,but does not depend strictly on JA biosynthesis.Microarray analysis using Arabidopsis whole-genomechip demonstrates that the gene transcriptional profileof bestatin-treated plants is similar to that of JA-treatedplants. Finally, bestatin promotes several JA-relateddevelopmental phenotypes. Taken together, the uniqueaction mode of bestatin in regulating JA-signaledprocesses leads us to the hypothesis that bestatinprobably exerts its effects by modulating the JA-signaling pathway. To test this hypothesis and, more-over, to identify novel components involved in JAresponses, we have employed bestatin as an experi-mental tool to dissect JA signaling using a chemicalgenetics approach, which yielded a collection of Arabi-dopsis bestatin-resistant (ber) mutants insensitive to theinhibitory effects of bestatin on primary root elon-gation. Further characterization efforts demonstratethat some ber mutants are defective in various JA- andwound-induced responses, which allowed us to clas-sify them into three phenotypic groups: mutants in-sensitive to both bestatin and JA, mutants insensitiveto bestatin but hypersensitive to JA, and mutantsinsensitive to bestatin but showing normal responseto JA. Among the three groups of mutants, the secondgroup, which is insensitive to bestatin but is hyper-sensitive to JA, is distinct from other identifiedJA-related mutants. Our ongoing effort to identifythe gene functions defined by these novel ber mutantspromises to shed new light on the molecular basis ofJA signaling and on the molecular mechanisms of thebiological function of bestatin in plants.
RESULTS
Bestatin Specifically Activates JA Signaling
in Tomato and Arabidopsis
It was previously reported that bestatin is a power-ful inducer of tomato defense genes, which also areinduced by herbivore attack, mechanical wounding,and the potent wound-response elicitors, such as sys-temin and JA (Schaller et al., 1995). Given that JAworks as a systemic wound signal and plays a centralrole in regulating defense genes expression in re-sponse to herbivore attack and mechanical wounding,we further compared the action mode of bestatin withthat of JA in tomato and Arabidopsis, two of the mostextensively studied systems on JA and wound signal-
ing. The induction kinetics of proteinase inhibitor II(PI-II) mRNA, a well-known representative marker ofa group of wound-inducible mRNAs in tomato, inresponse to exogenous bestatin and JA, was found tobe similar (Fig. 1A). Significant increase in PI-IImRNAlevels was initially observed 2 h after feeding plantswith both bestatin and JA, and mRNA abundance
Figure 1. Activation of JA-inducible genes by bestatin in tomato. A,Induction of PI-IImRNA by bestatin and JA. RNAwas isolated from wild-type (WT) tomato leaves at the times shown after treatment of the cuttingstems with bestatin (40 nmol/plant, dissolved in 15 mM phosphatebuffer), JA (20 nmol/plant, dissolved in 15 mM phosphate buffer), orbuffer control (15 mM phosphate buffer). RNA-blot hybridization wasperformed using 32P-labeled cDNAs for PI-II. A duplicate gel stainedwith ethidium bromide was used as RNA loading control. B, Effects of thespr2 and jai1-1 mutations on bestatin- and JA-induced PI-II mRNAaccumulation. Eighteen-day-old wild-type, spr2, and jai1-1 plants weretreated with bestatin, JA, and buffer control as described in A. Twelvehours after treatment, RNA was isolated from leaves of each genotypeand probed with 32P-labeled PI-II cDNA. A duplicate gel stained withethidium bromide was used as RNA loading control. C, Effects of the spr2and jai1-1 mutations on bestatin- and JA-induced PI-II protein accumu-lation. Eighteen-day-old wild-type, spr2, and jai1-1 plants were treatedwith buffer control (white bars), JA (black bars), and bestatin (gray bars),as described in A. Twenty-four hours after treatment, PI-II protein levelswere measured. Values represent mean 6 SD of 12 plants per genotype.
Zheng et al.
1402 Plant Physiol. Vol. 141, 2006
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
continued to rise during the time course, up to 24 h(Fig. 1A).To examine whether the action of bestatin to activate
the expression of JA-related defense genes is ubiqui-tous, we investigated the action of bestatin inArabidopsis. Treatment of wild-type plants with ex-ogenous bestatin readily induced enhanced expres-sion of VEGETATIVE STORAGE PROTEIN1 (AtVSP1;At5g24780), a widely used marker gene of JA re-sponses in Arabidopsis (Fig. 2B). Moreover, the induc-tion kinetics ofAtVSP1 transcript in response to bestatinis very similar to that in response to JA (Fig. 2, A andB). We further constructed a JA reporter line in whichthe promoter of the JA-inducible AtVSP1 gene wasfusedwith the b-glucuronidase (GUS) reporter. Bestatininduces the expression of AtVSP1-GUS throughoutthe plant, with highest expression in young leaves(Fig. 2E). Moreover, the GUS expression pattern of theJA reporter line in response to bestatin is similar to that
in response to JA (Fig. 2, D and E). Taken together,these data demonstrate that bestatin activates JA sig-naling both in tomato and Arabidopsis.
The Action of Bestatin to Activate JA SignalingStrictly Depends on COI1 Function
To determine whether bestatin-induced defensegene expression requires a functional JA-signalingpathway, we further compared the capacity of exoge-nous bestatin and JA to activate JA-inducible markergenes in tomato and Arabidopsis mutants that areknown to be defective in JA biosynthesis or signaling.Application of bestatin to the tomato spr2 mutant,which is deficient in JA biosynthesis (Li et al., 2003),induced the accumulation of PI-II protein to levelscomparable to those in wild-type plant (Fig. 1C).However, bestatin failed to induce PI-II accumulationin the jai1-1 plant (Fig. 1C), which is a loss-of-functionmutant of the tomato homolog of COI1, an F-boxprotein required for JA signaling (Li et al., 2004). Thecapacity of bestatin to induce PI-II protein accumula-tion in wild-type, spr2, and jai1-1 plants is very similarto that of JA, as shown in our parallel experiments (Fig.1C). We also examined the response of spr2 and jai1-1to exogenous bestatin at the level of PI-II transcriptaccumulation. For these experiments, PI-II transcriptlevels were measured 12 h after treatment with bestatinor JA, and the results obtained were in good agree-ment with the measurements of PI-II protein levels.Both bestatin and JA induced the accumulation of PI-IItranscripts in wild-type and spr2 plants, but not injai1-1 plants (Fig. 1B). These results demonstrate thatthe bestatin-induced PI-II accumulation in tomatoleaves requires the Jai1/COI1-dependent JA-signalingpathway, but does not depend on JA biosynthesis.Consistent with the results obtained from the tomatosystem, both bestatin and JA induce AtVSP1 expres-sion in wild-type Arabidopsis plants but not in coi1plants (Fig. 2, A and B), a well-known JA-signalingmutant in Arabidopsis (Feys et al., 1994; Xie et al.,1998). These results indicate that the induction of
Figure 3. Diagrams of the number of overlapping or nonoverlappinggenes up-regulated (A) or down-regulated (B) in response to bestatin(left) and MeJA (right) treatments. Expression data for all of the dif-ferentially regulated genes are listed in Supplemental Tables I to III.
Figure 2. Activation of JA-inducible genes by bestatin in Arabidopsis. Aand B, Parallel experiments to show the induction of the ArabidopsisAtVSP1 expression in wild type and coi1 mutant by JA (A) and bestatin(B). Two-week-old Arabidopsis seedlings (Col-0 wild type and coi1mutant) were treated with 50 mM MeJA, 50 mM bestatin, or not treated(0), and samples were harvested at indicated time points. Total RNAwas isolated and hybridized to 32P-labeled AtVSP1 cDNA, as describedin ‘‘Materials and Methods.’’ A duplicate gel stained with ethidiumbromide was used as RNA loading control. C to E, Activation of the JAreporter gene AtVSP::GUS by bestatin and JA. Seedlings of theArabidopsis AtVSP1::GUS transgenic line grown on MS medium (C),or MS medium containing 25 mM bestatin (D) or 25 mM MeJA (E) for 9 dwere visualized by GUS staining assay.
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1403
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
AtVSP1 expression by bestatin, just like by JA, alsooccurs through the COI1-dependent JA-signaling path-way in Arabidopsis. Taken together, our results indi-cate that bestatin promotes plant defense responsesby activating the JA-signaling pathway, and furthersuggest that bestatin might exert its action down-stream of JA biosynthesis but upstream of the F-boxprotein COI1.
Gene Transcriptional Profile of Bestatin-TreatedArabidopsis Plants Is Similar to That ofJA-Treated Plants
To gain a deeper understanding on the action mech-anism of bestatin, we performed a microarray analysisusing the Arabidopsis whole-genome chip (Affyme-trix) to compare the global gene transcriptional pro-files regulated by bestatin and MeJA in wild-typeArabidopsis seedlings. The arrays of bestatin- orMeJA-treated samples at the four time points tested(15, 30, 60, and 360 min) were compared with that ofuntreated sample by GeneChip Operating Software.
Using the statistical criteria described in ‘‘Materialsand Methods,’’ we selected the differentially ex-pressed genes (expression levels up or down for atleast 2.3-fold; Log2 values of 1.2 and 21.2, respec-tively) in response to bestatin andMeJA treatments forfurther analysis. Microarray data analysis revealedthat MeJA treatment caused 1,320 genes showing up-regulation and 1,069 genes showing down-regulation(Fig. 3). Bestatin treatment, on the other hand, caused1,007 genes showingup-regulation and 844 genes show-ing down-regulation (Fig. 3). The microarray data alsorevealed a high percentage of overlap between bestatin-regulated genes andMeJA-regulated genes; 83% of thegenes induced by bestatin were also induced byMeJA,and 66% of the genes repressed by bestatin were alsorepressed by MeJA (Fig. 3).
Further analyses indicated that genes regulated bybestatin and MeJA treatments at all the time pointsshowed significant correlation. Figure 4 shows theresults of 15- and 60-min time points, showing goodcorrelation between the changes of gene transcrip-tional level in response to bestatin and MeJA treat-ments. These analyses not only verified the reliabilityof our microarray experiment but also substantiatedthe notion that the gene expression profile of bestatin-treated plants is similar to that of JA-treated plants.
Functional classification of genes regulated by bothbestatin and JA treatments was carried out using thetools forGeneOntology (GO)categories (http://plexdb.org/index.php andhttps://www.affymetrix.com) andrevised manually, which demonstrated that bestatincanmimic the biological function of JA in the regulationof gene expression (Table I). For example, most of thestress-related genes regulated by MeJA treatment arealso regulated by bestatin treatment. We also noticedthat many genes related to senescence or cell death are
Figure 4. Correlation analysis for genes regulated by bestatin and MeJA treatments at 15-min (A) and 60-min (B) time points,respectively.
Table I. Functional classification of genes regulated by both bestatinand JA treatments using GO categories
Biological ProcessesGene No.
Up-Regulated Down-Regulated
Metabolism 312 197Transcription 71 34Transport 55 20Response to biotic stimulus 50 8Response to abiotic stimulus 20 12Signal transduction 42 14Senescence and cell death 26 1Photosynthesis and light
response8 16
Response to auxin 5 5Other 399 326
Zheng et al.
1404 Plant Physiol. Vol. 141, 2006
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
Table II. The list of selected genes regulated by both bestatin and MeJA
Signal Log2 ratio refers to the change in expression level for a transcript between bestatin or MeJA treatments and untreatment. P value representsthe Change P value.
Gene Function AGI
Bestatin Treatment MeJA Treatment
15 min 30 min 60 min 360 min 15 min 30 min 60 min 360 min
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Log2
Ratio
P
Value
Metabolism
b-Glucosidase At1g75940 3.3 0.004925 22 0.532344 4.5 0.00002 4 0.000068 4.1 0.00002 3.2 0.000023 4.4 0.00002 23.1 0.945978
Lipoxygenase At1g17420 3.7 0.00002 3.7 0.00002 4.4 0.00002 0.6 0.342437 4.8 0.00002 6.2 0.00002 6.7 0.00002 2.5 0.000052
Putative polygalacturonase At2g43870 23.7 0.998349 20.5 0.961585 21.7 0.998168 20.6 0.921063 20.7 0.999226 21.4 0.994067 24 0.996959 22.1 0.995927
Flavonol synthase At5g08640 21.7 0.99997 21.3 0.999948 21.9 0.99998 22.3 0.999977 21.8 0.99998 21.9 0.99998 21.8 0.999954 20.9 0.99996
Putative lipase At2g42690 20.7 0.999308 21 0.99998 21.3 0.99998 22.4 0.99998 20.4 0.999226 20.6 0.999034 22.1 0.99998 21.9 0.99998
Transcription
DREB2A At5g05410 1.9 0.00003 1.9 0.00002 2.5 0.00002 3.4 0.00002 1.9 0.000046 1.6 0.00002 3 0.00002 2.4 0.00002
R2R3-MYB
transcription factor
At3g50060 4.6 0.00002 3.1 0.00002 2.3 0.00004 1.3 0.065566 4.4 0.00002 3.3 0.00002 1 0.00003 1.3 0.004481
WRKY transcription factor At1g80840 6.1 0.00002 4.6 0.00002 4.5 0.00002 5.2 0.00002 5.8 0.00002 5.7 0.00002 6.5 0.00002 5 0.00002
MYB-related
transcription factor
At3g23250 5.8 0.000088 4.7 0.002753 5 0.00002 4.9 0.000241 6.4 0.00002 6.7 0.00003 8.1 0.00002 4.1 0.00712
C2H2-type zinc-finger
protein
At1g68130 20.8 0.994067 21.3 0.998349 20.7 0.995927 21.8 0.99994 20.6 0.997968 21.7 0.999899 21.6 0.99998 20.6 0.99997
Transport
Sugar transporter-like
protein
At4g36670 1 0.330589 2.1 0.000068 3.1 0.00002 1.4 0.000241 1.3 0.000078 3.2 0.00002 3.6 0.00002 1.4 0.000241
Peptide transporter At1g22570 0 0.441923 2.4 0.00002 2.8 0.00002 0 0.493524 1.8 0.00002 2.3 0.00002 1.4 0.00003 0.6 0.366593
Ammonium transporter At3g24290 22.6 0.999759 20.1 0.881991 20.7 0.999922 23.9 0.991489 0 0.921063 21.9 0.99997 22.5 0.998799 23.4 0.999786
Nitrate transporter At3g21670 20.2 0.998923 22 0.99998 22.3 0.99998 23.3 0.99998 20.6 0.999948 20.5 0.999833 23.1 0.99998 21.9 0.99998
Abiotic stress response
Glutathione S-transferase At2g29470 0.8 0.506476 2.8 0.018128 2.7 0.000167 7.6 0.00002 20.3 0.814019 1.7 0.044149 4.4 0.00002 7.3 0.00002
Drought-induced-19-like 1 At4g02200 1.9 0.000023 1.6 0.00002 0.4 0.004481 20.4 0.846768 1.6 0.00002 1.8 0.00002 0.9 0.000492 0.1 0.5
Heat-shock protein At5g52640 0.8 0.002753 0.4 0.161038 2 0.00002 3.8 0.00002 0.2 0.5 0.2 0.5 2.5 0.00002 2.6 0.00002
DRE CRT-binding protein
DREB1C
At4g25470 3.7 0.000023 2.3 0.000101 0.8 0.366593 20.4 0.657563 3.7 0.000023 2.7 0.000046 20.6 0.747149 20.5 0.5
Biotic stress response
Peroxidase At5g19890 1.4 0.013078 1.7 0.00002 1.2 0.000438 0.9 0.003699 0.9 0.001651 2.5 0.00002 2.2 0.00002 1.9 0.000241
Oxidoreductase-like protein At5g38000 1.5 0.000241 20.6 0.868657 2.9 0.00002 2.1 0.001486 2.6 0.000027 2.4 0.00002 2.8 0.00002 1.2 0.001201
Lectin protein At1g65400 5.8 0.00004 5.8 0.00002 6.9 0.00002 4.8 0.000966 4.5 0.00002 6.4 0.00002 7.6 0.00002 4.5 0.00002
Disease resistance protein At2g32140 4.9 0.000241 5 0.001651 3.8 0.000492 5 0.000078 5.7 0.000101 6 0.00004 5.7 0.000046 4 0.000046
Peroxidase At5g64110 20.5 0.99775 22.8 0.995519 22.1 0.999977 21.3 0.998514 21.3 0.999226 21.1 0.999654 21 0.999562 21.9 0.999654
Pathogenesis-related
protein (PR-1)
At2g19990 21 0.996959 20.3 0.532344 20.8 0.999226 24.1 0.99997 21.7 0.99775 20.7 0.975245 23.2 0.999977 20.7 0.988955
Signal transduction
Putative receptor
protein kinase
At1g74360 2 0.001336 1.8 0.011045 2.8 0.000241 5.2 0.00002 1.5 0.094279 2.4 0.010135 4.1 0.00002 3.9 0.00002
Calcineurin B-like protein 1 At4g17615 1 0.105663 1.1 0.000167 1.5 0.00002 2.8 0.00004 1.5 0.000147 2.1 0.00002 2.5 0.000088 1.9 0.00003
Phosphoinositide-specific
phospholipase C
At5g58670 1 0.296089 1 0.035785 2.7 0.00002 1.4 0.001832 0.9 0.000389 2.5 0.00002 4.5 0.00002 1.3 0.000214
Protein phosphatase 2C At1g07160 4.3 0.111714 1.2 0.318909 5.8 0.000307 6.7 0.000966 0.6 0.583567 1 0.185981 6.7 0.000346 6.7 0.000346
Putative receptor
protein kinase
At5g65600 1.7 0.263341 2.9 0.078937 4.3 0.000692 6.6 0.000167 2.8 0.069813 5.5 0.009292 5.8 0.000552 5.8 0.000346
Photosynthesis and
light response
Lhcb3 chlorophyll
a/b-binding protein
At5g54270 2.1 0.00002 2.4 0.00002 0.4 0.000865 0 0.506476 2.5 0.00002 2.7 0.00002 20.4 0.998799 20.7 0.99998
PSII type I chlorophyll
a/b-binding protein
At2g34430 1.4 0.000027 1.4 0.000027 20.2 0.99998 0.2 0.5 1.3 0.00002 1.5 0.00002 21.7 0.99998 20.5 0.999977
Nonphototropic
hypocotyl protein
At3g15570 0 0.938478 21.5 0.999911 24.6 0.99998 25.2 0.99997 20.6 0.997968 21.1 0.99998 25.1 0.99998 22.9 0.99998
Putative chlorophyll
a/b-binding protein
At2g40100 20.7 0.99998 22.1 0.99998 21 0.99998 22 0.99998 21 0.99998 21.6 0.99998 21 0.999977 22 0.99998
Putative nonphototropic
hypocotyl
At1g52770 20.8 0.981872 22.2 0.999759 23.6 0.999922 20.8 0.998923 20.2 0.838962 22.4 0.999693 22.1 0.999727 21 0.997247
Senescence and cell death
Disease resistance protein At1g56510 1.4 0.001201 3.3 0.00002 2.3 0.00002 0.9 0.354442 2.1 0.000023 3.1 0.00002 2.1 0.000101 0.3 0.454766
Disease resistance-like
protein
At5g04720 1.7 0.000027 1.2 0.00002 1.3 0.00002 0.9 0.074268 1.7 0.00002 1.9 0.00002 1.3 0.00003 0.9 0.004073
Senescence-associated
protein sen1
At4g35770 4.4 0.00002 1.2 0.000052 0.8 0.000552 3.5 0.00002 2.2 0.00002 4.9 0.00002 0.6 0.078937 4.1 0.00002
Disease resistance
protein-like
At5g41740 4.5 0.00002 4.9 0.000692 4.8 0.000068 5 0.00002 4.9 0.000088 4.6 0.000068 5.1 0.000114 4 0.006503
Auxin response
Auxin-induced proteins At3g09870 7.6 0.000046 7 0.000241 4.2 0.000774 0.6 0.366593 7.6 0.00002 6.2 0.00002 3.8 0.000774 4.3 0.000438
Putative GH3-like protein At4g03400 0.8 0.026698 2.2 0.00003 2.6 0.00002 20.2 0.938478 0.9 0.000114 3 0.00002 3 0.00003 22 0.999886
auxin-induced protein At1g29510 20.6 0.99751 22.7 0.999308 20.8 0.999448 21.7 0.990708 21.3 0.99987 21.3 0.998349 20.4 0.971234 22.5 0.999786
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1405
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
induced by bestatin and MeJA treatments, and manygenes related to photosynthesis or light response arerepressed by bestatin andMeJA treatments. The repre-sentative genes regulated by bestatin and MeJA treat-ments are listed in Table II. Expressiondata for all of thedifferentially regulated genes are listed in the sup-plemental data (Supplemental Tables I–III). The highsimilarity of gene transcriptional profiles betweenbestatin- and JA-treated Arabidopsis plants providedfurther support to the hypothesis that bestatin exerts itsbiological function through specific modulation of theJA-signaling pathway.
Isolation of ber Mutants
The above-mentioned unique biological function ofbestatin to specifically modulate the JA-responsepathway prompted us to employ this chemical as anexperimental tool to further dissect JA signaling inArabidopsis using the so-called chemical geneticsapproach, which potentially could circumvent theproblems of gene functional redundancy or lethal mu-tation (Surpin et al., 2005). To this end, we attemptedto screen for ber Arabidopsis mutants as a first step to
elucidate the molecular mechanisms of bestatin actionin plants. In addition to mimicking JA function interms of gene expression, bestatin also stimulatesseveral JA-related developmental phenotypes, includinginhibition of primary root elongation, promotion oflateral root formation, and anthocyanin accumulation(Fig. 5, A–C), which potentially could provide us witha convenient assay to screen for ber mutants. We nextinvestigated the responses of coi1 and other knownJA-related mutants to bestatin treatment in terms ofinhibition of root elongation. Interestingly, the coi1mutant is as responsive as wild type to bestatin interms of inhibition of primary root elongation(Fig. 5D). This result indicates that even though thebestatin-induced defense gene expression strictly de-pends on the COI1 protein function, the bestatin-induced root growth inhibition does not. However,our parallel root growth assay indicated that theJA-insensitive mutant jin1 (Berger et al., 1996; Lorenzoet al., 2004) was significantly less sensitive to bestatinthan the wild type (Fig. 5D). Together, these resultsimply that it is possible to identify genetic loci in-volved in JA signaling through screening for bestatin-response mutants. Based on these observations, we set
Figure 5. The chemical structures of bestatinand JA and their effects on root growthof Arabidopsis seedlings. A, The chemicalstructure of JA and 7-d-old wild-typeArabidopsis (Col-0) seedlings grown on in-dicated concentrations of JA. B, The chem-ical structure of bestatin and 7-d-old Col-0seedlings grown on indicated concentra-tions of bestatin. C, Quantitative analysisof the inhibitory effects on primary rootelongation of bestatin and JA. Root lengthsof seedlings as given in Figure 5, A and B,were measured. Data represent mean 6 SD
of 10 plants per genotype. Similar resultswere obtained from at least three repeatsof this experiment. D, Comparison of rootgrowth phenotypes of Col-0, coi1, and jin1in response to bestatin. Seedlings werephotographed 6 d after germination onMS medium (top) or MS medium contain-ing 25 mM bestatin (bottom).
Zheng et al.
1406 Plant Physiol. Vol. 141, 2006
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
up a large-scale genetic screen for Arabidopsis mu-tants that are resistant to the inhibitory effects ofbestatin on primary root elongation. Approximately150,000 Arabidopsis seeds derived from an ethyl-methane sulfonate (EMS)-mutagenized M2 populationwere examined on Murashige and Skoog (MS) me-dium containing 25mMbestatin, and 300 seedlingswithelongated roots were selected as putative bestatin-resistantmutants. Seeds harvested from theprimarypu-tative mutants were rescreened on bestatin-containingmedium, and 19 ber mutants were retained for furtheranalysis.
Response of ber Mutants to JA
The unique biological function of bestatin to mod-ulate JA-signaled processes led us to the hypothesisthat our collection of bestatin-response mutants couldrecover novel genetic loci involved in JA signaling. Totest this possibility, we examined the responses of the
ber mutants to exogenous JA. For these experiments,seeds of each of the 19 ber mutants were further testedseparately on medium containing exogenous JA. Asexpected, we identified ber mutants with altered JAresponse. Based on the root elongation phenotype ofthese ber mutants in response to bestatin and JA, the19 ber mutants were classified into three phenotypicgroups. The first group of ber mutants, which includesber6, ber7, ber188, ber229, ber328, and ber759, is insen-sitive to both bestatin and JA and therefore wasdesignated as JA-insensitive ber mutants (Fig. 6). TheJA-insensitive phenotype of this group of ber mutantssuggests they could define positive regulators in theJA-signaling pathway. The second group containedthree independently isolated mutants: ber157, ber426,and ber544. Because these mutants are insensitive tobestatin but are hypersensitive to JA, they were des-ignated as JA-hypersensitive ber mutants (Fig. 7). Thisgroup of ber mutants could define negative regulatorsin JA signaling. The third and largest group of ber
Figure 6. Root elongation phenotype of the groupof JA-insensitive ber mutants. A, Root phenotypesof Col-0, ber6, ber7, ber188, ber229, ber328, andber759 in response to bestatin and JA. Seedlingswere photographed 7 d after germination on MSmedium and MS medium containing 25 mM
bestatin or 20 mM MeJA. B, The root length wasmeasured for seedlings as given in Figure 6A. Thevalues represent mean 6 SD of 10 plants pergenotype.
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1407
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
mutants, which includes ber38, ber227, ber281, ber336,ber338, ber369, ber374, ber399, ber581, and ber670, isinsensitive to bestatin but shows normal response toJA (Fig. 8). This group of mutants might define genesinvolved in the uptake, transport, metabolism, and/orperception of bestatin by plants. Taken together, theseresults indicate that we have recovered some JA-related mutants from the ber mutants identified bythe bestatin-based chemical genetic screening.
Genetic and Phenotypic Analyses of ber Mutantswith Altered JA Responses
Further characterization efforts were primarily fo-cused on the two phenotypic groups of ber mutantswith altered JA responses, which may define novel
components in JA signaling. We conducted a set ofcrosses to determine the genetic basis of these bermutants. Each of these nine mutants was backcrossedtowild type (Columbia-0 [Col-0]), and the phenotype ofthe resulting F1 plantswas scored on bestatin-containingmedium and comparedwith those of the parental lines(i.e. mutant and Col-0). F1 plants derived from crossesfor all of the nine mutants (including six JA-insensitiveber mutants and three JA-hypersensitive ber mutants)displayed wild-type phenotype (i.e. normal responseto bestatin in terms of inhibition of root elongation;data not shown). F1 plants were self-pollinated and thephenotype of F2 progeny was scored. The segregationratio of bestatin-sensitive plants:bestatin-resistant plantswas close to 3:1 in all F2 populations (data not shown).These results indicate that the bestatin-resistant phe-notype of each of the nine ber mutants is caused by asingle recessive mutation in a nuclear gene.
Complementation tests among the six lines of JA-insensitive ber mutants revealed that they are notallelic to each other. Test crosses were also conductedbetween each of the six mutants and the reportedJA-insensitivemutants, including jai1/jin1 (Bergeret al.,1996; Lorenzo et al., 2004) and coi1 (Feys et al., 1994;Xie et al., 1998), the results indicated that ber328 isallelic to jin1, which was recently shown to encode abHLH protein called AtMYC2 (Lorenzo et al., 2004),and none of the mutants is allelic to coi1.
Allelism analysis among the three JA-hypersensitiveber mutants demonstrated that each of them representsan independent locus. Each of the three JA-hypersen-sitive ber mutants also displays specific morphologicalphenotypes. For example, ber157 seedlings exhibitstunted growth and short roots without JA treatment(Fig. 7A). Another striking phenotype of ber157 is vires-cence, i.e. young leaves or newly expanded tissues werepale, whereas more mature tissues were as green aswild type (Fig. 9, A and B). Similar to the Arabidopsistransparent testa mutants that are deficient in pigmentaccumulation in the testa (seed coat; Shirley et al., 1995),seeds of the ber426 mutant are yellow or pale brown incolor, easily distinguishable from the brown seeds ofwild-type plants (Fig. 9, C and D). In the absence ofexogenous JA, the ber544 seedlings mimic phenotypesof JA-treated, wild-type plants, i.e. stunted growth andrelatively shorter roots (Fig. 7A). The typical morphol-ogy of ber544 adult plants includes narrow leaf blade,semidwarfism, and small stature (Fig. 9E). To test if thethree JA-hypersensitive mutants define novel loci in JAsignaling or represent new alleles of the reported cev1(Ellis and Turner, 2001) and cex1 (Xu et al., 2001)mutants, which show constitutive expression of JA-related genes, we determined the chromosome locationsof the three genes through standard genetic mapping.BER157 was located on chromosome 4 and linked tomarkers ciw7 and nga1107. BER426 was mapped to aregion between markers nga249 and ciw8 on the shortarm of chromosome 5. BER544 was initially mappedbetweenmarkers CTR1 and nga249 on the short arm ofchromosome 5 and later was delimited to a region
Figure 7. Root elongation phenotype of the group of JA-hypersensitiveber mutants. A, Root phenotypes of Col-0, ber157, ber426, and ber544in response to bestatin and JA. Seedlings were photographed 7 d aftergermination on MS medium and MS medium containing 25 mM bestatinor 20 mM MeJA. B, The root length was measured for seedlings as givenin Figure 7A. The values represent mean 6 SD of 10 plants per genotype.
Zheng et al.
1408 Plant Physiol. Vol. 141, 2006
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
containing five BAC clones. To our knowledge, none ofthese loci represents a known genetically identifiedlocus with JA-hypersensitive phenotypes, implyingthat these three JA-hypersensitive ber mutants definenovel genes in JA signaling.
JA-Induced Defense Gene Expression in ber Mutants
To examine the effects of ber mutations on the JA-or wound-response pathway, we further compareddefense-related marker gene expression betweenwild type and representative ber mutants. Seedlingswere either mechanically wounded or treated withJA, and the AtVSP1 transcripts were measured atdifferent time after treatment. Wild-type plants accu-mulated high levels of AtVSP1 transcripts in response
to JA or mechanical wounding (Fig. 10). However, theJA-insensitive ber759 was significantly impaired inJA-induced AtVSP1 transcript accumulation (Fig.10A). In contrast, a JA-hypersensitive ber mutant,ber544, constitutively showed elevated AtVSP1 ex-pression without treatment (Fig. 10, B and C). Whentreated with JA and mechanical wounding, AtVSP1mRNA accumulation levels in this mutant were sig-nificantly higher than those in wild type (Fig. 10, Band C).
DISCUSSION
More and more laboratories are exploring the useof synthetic molecules as tools to solve biological
Figure 8. Root elongation phenotype of the group of ber mutants with normal JA responses. A, Root phenotypes of Col-0, ber38,ber227, ber281, ber336, ber338, ber369, ber374, ber399, ber581, and ber670 in response to bestatin and JA. Seedlings werephotographed 7 d after germination on MS medium and MS medium containing 25 mM bestatin or 20 mM MeJA. B, The root lengthwas measured for seedlings as given in Figure 8A. Data represent mean 6 SD of 10 plants per genotype.
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1409
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
problems (Schreiber, 2000; Blackwell and Zhao, 2003;Zhaoet al., 2003;Armstronget al., 2004;Vermaetal., 2004;Zouhar et al., 2004; Dai et al., 2005). Among the majorchallenges of using a synthetic compound in a biolog-ical system are to determine the specificity of thecompound in the biological process of interest and todefine the targets and action mechanism of the usedcompound. In general, the specificity of a particularcompound can be determined if (1) the compoundregulates a set of genes known to be involved in theprocess of interest (e.g. this can often be assessed by aDNAmicroarray experiment), (2) the compound causesany known developmental phenotypes related to theprocess of interest, and/or (3) knownmutants involvedin the process respond to the compound as predicted(Blackwell andZhao, 2003).Wehaveprovideda series ofgenetic and molecular evidence showing that bestatinactivates the expression of JA-inducible, defense-related genes in tomato and Arabidopsis (Figs. 1Aand 2, A–C). Furthermore, the action of bestatin toactivate defense genes expression requires the functionof the F-box proteinCOI1/Jai1, a central regulator of JAsignaling (Figs. 1, B and C, and 2, A and B). Microarrayanalysis demonstrates that the gene transcriptionalprofile of bestatin-treated Arabidopsis plants is similarto that of JA-treated plants (Figs. 3 and 4; SupplementalTables I–III). Finally, bestatin promotes several JA-related developmental phenotypes, including inhibi-tion of root growth (Fig. 5). Our data clearly support theconclusion that the action of bestatin to activate defense
genes expression strictly depends on COI1-mediatedJA signaling. However, the fact that the JA-insensitivecoi1 mutant is fully responsive to bestatin treatment interms of root growth inhibition (Fig. 5D) leads to theconclusion that certain JA-related phenotypes trig-gered by bestatin are COI1 independent. One possibleexplanation for this phenomenon is that bestatin mighttarget different proteins to trigger defense gene expres-sion or inhibit root elongation. In contrast to the coi1mutant, our parallel root growth assay indicated thatthe JA-insensitive mutant jin1 (Berger et al., 1996;Lorenzo et al., 2004) was significantly less sensitive tobestatin than the wild type (Fig. 5D). Taken together,these results ledus to thehypothesis that it is possible toidentify genetic loci involved in JA signaling throughscreening for bestatin-response mutants. Using bestatinas a chemical tool, we have applied a chemical geneticapproach to dissect JA signaling, with the hope to
Figure 9. Developmental phenotypes of JA-hypersensitive ber mutants.A and B, Phenotypic comparison between wild type (Col-0) andber157. Shown is 2-week-old wild type (A) and ber157 (B) on MSmedium. The young leaves of ber157 are pale. C and D, Comparison ofseed coat color between wild type and the ber426 mutant. ber426 (C)displays lighter color in seed coat compare to wild type (D). E, Five-week-old wild type (left) and ber544 mutant (right).
Figure 10. JA- and wound-inducedAtVSP1 expression in representativeber mutants showing altered JA response. A, JA-induced AtVSP1 ex-pression is attenuated in the JA-insensitive ber759. Two-week-old Col-0and ber759 mutant seedlings were treated with 50 mM MeJA or nottreated (0). Plant tissues were harvested at indicated times for total RNAextraction. RNA-blot hybridization was performed using 32P-labeledAtVSP1 as probe. A duplicate gel stained with ethidium bromide wasused as RNA loading control. B, JA-induced AtVSP1 expression iselevated in the JA-hypersensitive ber544. Two-week-old Col-0 andber544 seedlings were treated with 50 mM MeJA or not treated (0). Planttissues were harvested 6 h after treatment for total RNA extraction. RNA-blot hybridization was performed using 32P-labeled AtVSP1 as probe. Aduplicate gel stained with ethidium bromide was used as RNA loadingcontrol. C, Wound-induced AtVSP1 expression is elevated in theJA-hypersensitive ber544. Twenty-day-old Col-0 and ber544 seedlingswere wounded with a hemostat or not wounded (0). Plant tissues wereharvested 6 h after treatment for total RNA extraction. RNA-blot hybrid-ization was performed using 32P-labeled AtVSP1 as probe. A duplicategel stained with ethidium bromide was used as RNA loading control.
Zheng et al.
1410 Plant Physiol. Vol. 141, 2006
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
circumvent the gene redundancy and mutation le-thality problems, which are intractable by the conven-tional genetic approach. Indeed, our genetic screen forbestatin-resistantArabidopsis plants did lead to identi-fication of two phenotypic groups of mutants withaltered JA responses, i.e. JA-insensitive bermutants andJA-hypersensitive ber mutants. Among the group ofJA-insensitive ber mutants are additional alleles ofknown JA-response mutants (for example, jin1/atmyc2;Lorenzo et al., 2004) andnovel ones.Wenoted thatmostof the JA-insensitive ber mutants show strong insen-sitivity to bestatin but display relatively weak in-sensitivity to JA (Fig. 6), which probably is the reasonwhy these mutants did not come out from previousdirect JA-resistant mutant screens. The strong bestatin-insensitive phenotype of these mutants also providesuswithafacileassaytoidentifythecorrespondinggenesusing a standard map-based cloning method, againshowing the power of the bestatin-based approach toovercome the difficulty caused by gene redundancy inprevious JA studies. Apparently, the identification ofnew components involved in the JA-signaling pathwaydefined by these ber mutants will extend our under-standing of the molecular mechanisms that underlieJA-dependent responses. These results also demon-strated that bestatin-based chemical genetic screeningis a very effective tool to dissect JA signaling inArabidopsis.The other concern of using the chemical genetic
approach to solve biological problems is to identify thesmall molecule targets and to elucidate their actionmechanisms. Recent elegant workwith the small mole-cule Sirtinol in auxin studies has demonstrated thepower of combining genetic studies and chemicalanalyses in determining the action mechanisms of asynthetic compound in the model plant Arabidopsis(Zhao et al., 2003; Dai et al., 2005). Because bestatin andJA are not structurally related (Fig. 5, A and B) butinduce similar responses with respect to defense-related gene expression, we hypothesized that bestatinand JA most likely target different components in JAsignaling and that bestatin targets should be importantcomponents in JA signaling. It is expected that ourgenetic screen for mutants insensitive to the effects ofbestatin should identify genes involved in bestatinuptake, transport, and metabolism, as well as targetsand downstream signaling components. Among these,targets for bestatin are of most interest and signifi-cance. Given our systematic analyses have demon-strated that bestatin, an inhibitor of metalloprotease(Scornik and Botbol, 2001), causes JA responses in to-mato andArabidopsis in a JA biosynthesis-independentbut COI1-dependent manner, this action mode leadsus to the hypothesis that bestatin likely inhibits aprotease that serves as a negative regulator in JA sig-naling. Furthermore, the action mode of bestatin alsosupports the prediction that a mutant plant harboringmutation of bestatin target will be insensitive tobestatin but hypersensitive to JA; this phenotype ex-actly fits the phenotype of our second group of ber
mutants (i.e. JA-hypersensitive ber mutants) that in-cludes ber157, ber544, and ber426. Our ongoing effortsto characterize these mutants, identify the correspond-ing genes, and perform functional analyses will be ofgreat importance for the elucidation of the molecularmechanisms of bestatin biological function and fur-ther understanding of the molecular basis of JAsignaling.
MATERIALS AND METHODS
Plant Materials and Bioassay
Tomato (Lycopersicon esculentum) cv Castlemart was used as the wild-type
cultivar for all experiments involving bioassay of bestatin and JA. The tomato
wound-response mutants spr2 (Li et al., 2003) and jai1-1 (Li et al., 2004) were in
the genetic background of cv Castlemart. Tomato plants were grown at standard
conditions as described previously (Schaller et al., 1995; Li et al., 2003). Eighteen
to 20 d after planting, plants were excised at the base of the stem and placed in
0.5-mL microfuge tubes containing 300 mL of the inducing compound for 2 h.
Plants were then transferred to glass vials containing 20 mL of water and
incubated in a Lucite box for 24 h under continuous light. PI-II levels in leaves
were measured by radial immunodiffusion assay (Ryan, 1967). Briefly, leaves of
the treated seedlings were collected and ground together using a glass pestle on
ceramic to get leave juice. Five microliters of leaf juice was loaded on PIN-II
plates containing tomato PI-II antiserum. After 24 h of incubation on bench top,
the PIN-II plates were developed with 7% acetic acid for 5 min. Diameters (mm)
of the PIN-II radial immunodiffusion rings were measured with a caliper. The
PI-II concentrations (mg/mL leaf juice) were determined using the following
formula: (diameter 3 diameter 2 625) 3 0.016.
JA, MeJA, and bestatin were purchased from Sigma (catalog nos. J2500,
392707, and B8385 respectively), and the inducers were diluted from stock
solutions into sodium phosphate buffer (15 mM sodium phosphate, pH 6.5)
prior to use.
The Col-0 ecotype of Arabidopsis (Arabidopsis thaliana) was used as wild
type. The JA-response mutant coi1 was kindly provided by Dr. Daoxin Xie
(Tsinghua University, Beijing). For JA- and bestatin-response analysis, 2-week-
old seedlings grown on 0.53 MS medium were sprayed with solutions
containing the chemical compounds, and then incubated in growth chamber
under continuous light. Tissues were then harvested at the time intervals as
indicated for RNA extraction. Twenty-day-old Arabidopsis seedlings grown
on 0.53 MS medium were wounded with a hemostat across the midvein of
two expanded leaves. Before wounding or 6 h after wounding, leaves were
harvested for total RNA isolation as described below.
Northern-Blot Analysis
Total RNA was prepared by a guanidine thiocyanate extraction method,
and RNA gel-blot analysis was performed as described previously (Li et al.,
2005). Ten micrograms of total RNAwas separated in an agarose gel contain-
ing 10% formaldehyde, blotted onto a Hybond N1 membrane (Amersham),
and probed with the PCR-amplified DNA fragments using the following primers:
PI-II F (5#-TCAGAAGGAAGTCCGCTAAATC-3#) and PI-II R (5#-TTGCCTTG-GGTTCATCACTCT-3#) for PI-II from tomato, and AtVSP1F (5#-ATGAAAAT-
CCTCTCACTTTCA-3#) and AtVSP1R (5#-TATCCATATTTAGCGTAGTAGG-3#)for AtVSP1 from Arabidopsis (Col-0).
Plasmid Construction and Plant Transformation
A 740-bp fragment upstream of theAtVSP1 coding sequence was isolated by
PCR from genomic DNA (Col-0) using the primers 5#-AAGAAAATCAAGCT-
TTAACCTAAAATCAAC-3# and 5#-GTCGGATCCAGTTTATGGTGTTTATTT-
GTG-3#. The PCR product was digested with HindIII and BamHI and inserted
into the same sites of pBI121 (CLONTECH) to yield the construct AtVSP1::GUS.
The construct was then transformed into Agrobacterium tumefaciens strain
GV3101 (pMP90), whichwas used for transformation of Arabidopsis by vacuum
infiltration (Bechtold et al., 1993). A homozygous transgenic line of AtVSP1::
GUS was identified and used for GUS staining experiment.
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1411
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
Analysis of GUS Activity
Histochemical staining for GUS activity in transgenic plants was per-
formed as described previously (Jefferson et al., 1987).
Microarray Experiments and Data Analysis
Two-week-old, light-grownArabidopsis seedlingswere sprayedwith 50mM
bestatin or 50mMMeJA and then incubated in growth chamber. Seedlings were
then harvested at different time points (15, 30, 60, and 360 min) after treatment
for total RNA preparation. The untreated seedlings were harvested for total
RNApreparation used as control. Total RNAwas used for preparing probes for
the microarray experiments, which were carried out according to the protocols
provided by the gene chip manufacturer Affymetrix.
The signal intensity data files of scanned probe arrays were analyzed by
GeneChip Operating Software. The one-sidedWilcoxon’s signed rank test was
the statistical method employed to generate the Detection P value, which is
evaluated against defined cutoffs to determine the Detection call. Signal
values were calculated using the one-step Tukey’s biweight estimate, which
assigns a relative measure of abundance to the transcript. Before comparing
two arrays between bestatin- or MeJA-treated samples and untreated sample,
we set the average signal intensity of the array to a target signal of 500 to
scale/normalize the data. The Wilcoxon’s signed rank test was used to com-
pute each Change P value. The signal Log2 ratio was calculated by comparing
each probe pair on the experiment array to the corresponding probe pair on
the baseline array.
The criteria for selecting the significantly differential expressed genes in
response to bestatin or MeJA treatment are as follows. For up-regulated genes,
the criteria are as follows. (1) The signal Detection calls of at least two of the
four time points are Present. (2) The Change call of at least two of the four time
points is Increase. (3) The signal Log2 ratios of at least two of the four time
points are 1.2 (2.3-fold change) or greater. For down-regulated genes, the
criteria are as follows. (1) The signal Detection call of untreated sample is
Present. (2) The Change call of at least two of the four time points is Decrease.
(3) The signal Log2 ratios of at least two of the four time points are 21.2 (2.3-
fold change) or lower. Signal Detection call, Change call, and signal Log2 ratio
are calculated separately using three independent metrics. Change call and
signal Log2 ratio represent the change in expression level for a transcript
between bestatin or MeJA treatments and nontreatment. Correlation analyses
for the changes of gene transcriptional level in response to bestatin and MeJA
treatments were performed in Microsoft Excel. Functional classification of
genes regulated by both bestatin and MeJA treatments was carried out using
the tools for GO categories (http://plexdb.org/index.Php and https://
www.affymetrix.com) and revised manually.
Bestatin-Resistant Mutant Screen and Genetic Analysis
EMS-mutagenized Arabidopsis Col-0 M2 seeds were kindly provided by
Dr. Jianru Zuo (Institute of Genetics and Developmental Biology, Chinese
Academy of Sciences, Beijing). The M2 seeds were germinated and grown on
0.53 MSmedium containing 25mM bestatin underwhite light (16-h-light/8-h-
dark cycle) at 22�C for 8 to 10 d. Seedlingswith elongated rootswere selected as
putative bermutants and directly transplanted to soil. ResultingM3 seeds from
the putative mutants were retested on 25 mM bestatin by measuring root
elongation. The confirmed ber mutants were then tested for their responses to
applied JA. The nine ber mutants showing altered JA responses were back-
crossed at least twice towild type before physiological andmolecular analyses.
For mapping, the ber mutants in Col-0 background were crossed to the
polymorphic ecotype Landsberg erecta, and the resulting F1 plants were self-
pollinated to yield F2 populations segregating for the ber mutant phenotype.
Simple sequence length polymorphism markers were used for linkage anal-
ysis using the standard procedures as described (Lukowitz et al., 2000).
ACKNOWLEDGMENTS
We gratefully acknowledge Dr. Clarence Ryan (Washington State Univer-
sity, Pullman, WA) for providing tomato PI-II antiserum, Dr. Gregg Howe
(Michigan State University, East Lansing, MI) for providing the tomato spr2
and jai1-1 seeds, Dr. Jianru Zuo (Institute of Genetics and Developmental
Biology, Chinese Academy of Sciences, Beijing) for providing EMS-mutagenized
Arabidopsis (Col-0) M2 seeds, Dr. Daoxin Xie (Tsinghua University, Beijing)
for providing coi1 seeds, and Dr. Salome Prat (Institut de Biologia Molecular
de Barcelona) for providing homozygous atmyc2-1 (SALK_040500) and
atmyc2-2 (SALK_083483) seeds. We also thank Wenying Xu and Wenkun
Zhou for their help in microarray data analyses.
Received March 14, 2006; revised June 10, 2006; accepted June 11, 2006;
published June 23, 2006.
LITERATURE CITED
Armstrong JI, Yuan S, Dale JM, Tanner VN, Theologis A (2004) Identification
of inhibitors of auxin transcriptional activation by means of chemical
genetics in Arabidopsis. Proc Natl Acad Sci USA 101: 14978–14983
Asami T, Nakano T, Nakashita H, Sekimata K, Shimada Y, Yoshida S
(2003) The influence of chemical genetics on plant science: shedding
light on functions and mechanism of action of brassinosteroids using
biosynthesis inhibitors. J Plant Growth Regul 22: 336–349
Bechtold N, Ellis J, Pelletier G (1993) In planta Agrobacterium-mediated
gene transfer by infiltration of adult Arabidopsis thaliana plants. C R
Acad Sci III Sci Vie 316: 1194–1199
Berger S, Bell E, Mullet JE (1996) Two methyl jasmonate-insensitive
mutants show altered expression of AtVsp in response to methyl
jasmonate and wounding. Plant Physiol 111: 525–531
Blackwell HE, Zhao Y (2003) Chemical genetic approaches to plant biology.
Plant Physiol 133: 448–455
Boter M, Ruiz-Rivero O, Abdeen A, Prat S (2004) Conserved MYC
transcription factors play a key role in jasmonate signaling both in
tomato and Arabidopsis. Genes Dev 18: 1577–1591
Bowles D (1998) Signal transduction in the wound response of tomato
plants. Philos Trans R Soc Lond B Biol Sci 353: 1495–1510
Browse J (2005) Jasmonate: an oxylipin signal with many roles in plants.
Vitam Horm 72: 431–456
Constabel CP, Bergey DR, Ryan CA (1995) Systemin activates synthesis of
wound-inducible tomato leaf polyphenol oxidase via the octadecanoid
defense signaling pathway. Proc Natl Acad Sci USA 92: 407–411
Creelman RA, Mullet JE (1997) Biosynthesis and action of jasmonates in
plants. Annu Rev Plant Physiol Plant Mol Biol 48: 355–381
Dai X, Hayashi K, Nozaki H, Cheng Y, Zhao Y (2005) Genetic and chemical
analyses of the action mechanisms of sirtinol in Arabidopsis. Proc Natl
Acad Sci USA 102: 3129–3134
Devoto A, Nieto-Rostro M, Xie D, Ellis C, Harmston R, Patrick E, Davis J,
Sherratt L, Coleman M, Turner JG (2002) COI1 links jasmonate signal-
ling and fertility to the SCF ubiquitin-ligase complex in Arabidopsis.
Plant J 32: 457–466
Devoto A, Turner JG (2003) Regulation of jasmonate-mediated plant
responses in arabidopsis. Ann Bot (Lond) 92: 329–337
Ellis C, Karafyllidis I, Wasternack C, Turner JG (2002) The Arabidopsis
mutant cev1 links cell wall signaling to jasmonate and ethylene re-
sponses. Plant Cell 14: 1557–1566
Ellis C, Turner JG (2001) The Arabidopsis mutant cev1 has constitutively
active jasmonate and ethylene signal pathways and enhanced resistance
to pathogens. Plant Cell 13: 1025–1033
Farmer EE, Ryan CA (1990) Interplant communication: Airborne methyl
jasmonate induces synthesis of proteinase inhibitors in plant leaves.
Proc Natl Acad Sci USA 87: 7713–7716
Farmer EE, Ryan CA (1992) Octadecanoid precursors of jasmonic acid
activate the synthesis of wound-inducible proteinase inhibitors. Plant
Cell 4: 129–134
Feng S, Ma L, Wang X, Xie D, Dinesh-Kumar SP, Wei N, Deng XW (2003)
The COP9 signalosome interacts physically with SCF COI1 and mod-
ulates jasmonate responses. Plant Cell 15: 1083–1094
Feys B, Benedetti CE, Penfold CN, Turner JG (1994) Arabidopsis mutants
selected for resistance to the phytotoxin coronatine are male sterile,
insensitive to methyl jasmonate, and resistant to a bacterial pathogen.
Plant Cell 6: 751–759
Gray WM, Muskett PR, Chuang HW, Parker JE (2003) Arabidopsis SGT1b
is required for SCF(TIR1)-mediated auxin response. Plant Cell 15:
1310–1319
Green T, Ryan C (1972) Wound inducible proteinase inhibitors in plant
leaves: a possible defense mechanism against insects. Science 175: 776–777
Hilpert B, Bohlmann H, op den Camp RO, Przybyla D, Miersch O,
Buchala A, Apel K (2001) Isolation and characterization of signal
Zheng et al.
1412 Plant Physiol. Vol. 141, 2006
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021
transduction mutants of Arabidopsis thaliana that constitutively
activate the octadecanoid pathway and form necrotic microlesions.
Plant J 26: 435–446
Howe GA (2004) Jasmonates as signals in the wound response. J Plant
Growth Regul 23: 223–237
Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: beta-
glucuronidase as a sensitive and versatile gene fusion marker in higher
plants. EMBO J 6: 3901–3907
Leon J, Rojo E, Sanchez-Serrano JJ (2001) Wound signalling in plants. J Exp
Bot 52: 1–9
Leyser HM, Lincoln CA, Timpte C, Lammer D, Turner J, Estelle M (1993)
Arabidopsis auxin-resistance gene AXR1 encodes a protein related to
ubiquitin-activating enzyme E1. Nature 364: 161–164
Li C, Liu G, Xu C, Lee GI, Bauer P, Ling HQ, Ganal MW, Howe GA (2003)
The tomato suppressor of prosystemin-mediated responses2 gene en-
codes a fatty acid desaturase required for the biosynthesis of jasmonic
acid and the production of a systemic wound signal for defense gene
expression. Plant Cell 15: 1646–1661
Li C, Schilmiller AL, Liu G, Lee GI, Jayanty S, Sageman C, Vrebalov J,
Giovannoni JJ, Yagi K, Kobayashi Y, et al (2005) Role of beta-oxidation
in jasmonate biosynthesis and systemic wound signaling in tomato.
Plant Cell 17: 971–986
Li L, Li C, Lee GI, Howe GA (2002) Distinct roles for jasmonate synthesis
and action in the systemic wound response of tomato. Proc Natl Acad
Sci USA 99: 6416–6421
Li L, Zhao Y, McCaig BC, Wingerd BA, Wang J, Whalon ME, Pichersky E,
HoweGA (2004) The tomato homolog of CORONATINE-INSENSITIVE1
is required for the maternal control of seed maturation, jasmonate-signaled
defense responses, and glandular trichomedevelopment. Plant Cell 16: 126–143
Lukowitz W, Gillmor S, Scheible WR (2000) Positional cloning in Arabi-
dopsis. Why it feels good to have a genome initiative working for you.
Plant Physiol 123: 795–805
Lorenzo O, Chico JM, Sanchez-Serrano JJ, Solano R (2004) JASMONATE-
INSENSITIVE1 encodes a MYC transcription factor essential to dis-
criminate between different jasmonate-regulated defense responses in
Arabidopsis. Plant Cell 16: 1938–1950
Pearce G, Strydom D, Johnson S, Ryan CA (1991) A polypeptide from
tomato leaves induces wound-inducible proteinase inhibitor proteins.
Science 253: 895–898
Ren C, Pan J, Peng W, Genschik P, Hobbie L, Hellmann H, Estelle M, Gao
B, Peng J, Sun C, et al (2005) Point mutations in Arabidopsis Cullin1
reveal its essential role in jasmonate response. Plant J 42: 514–524
Ryan CA (1967) Quantitative determination of soluble cellular proteins by radial
diffusion in agar gels containing antibodies. Anal Biochem 19: 434–440
Ryan CA, Moura DS (2002) Systemic wound signaling in plants: a new
perception. Proc Natl Acad Sci USA 99: 6519–6520
Schaller A, Bergey DR, Ryan CA (1995) Induction of wound response
genes in tomato leaves by bestatin, an inhibitor of aminopeptidases.
Plant Cell 7: 1893–1898
Schilmiller AL, Howe GA (2005) Systemic signaling in the wound re-
sponse. Curr Opin Plant Biol 8: 369–377
Schreiber SL (2000) Target-oriented and diversity-oriented organic syn-
thesis in drug discovery. Science 287: 1964–1969
Scornik OA, Botbol V (2001) Bestatin as an experimental tool in mammals.
Curr Drug Metab 2: 67–85
Shirley BW, Kubasek WL, Storz G, Bruggemann E, Koornneef M,
Ausubel FM, Goodman HM (1995) Analysis of Arabidopsis mutants
deficient in flavonoid biosynthesis. Plant J 8: 659–671
Staswick PE, Su W, Howell SH (1992) Methyl jasmonate inhibition of root
growth and induction of a leaf protein are decreased in an Arabidopsis
thaliana mutant. Proc Natl Acad Sci USA 89: 6837–6840
Staswick PE, Tiryaki I, Rowe ML (2002) Jasmonate response locus JAR1
and several related Arabidopsis genes encode enzymes of the firefly
luciferase superfamily that show activity on jasmonic, salicylic, and
indole-3-acetic acids in an assay for adenylation. Plant Cell 14:
1405–1415
Stratmann JW (2003) Long distance run in the wound response: Jasmonic
acid is pulling ahead. Trends Plant Sci 8: 247–250
Surpin M, Rojas-Pierce M, Carter C, Hicks GR, Vasquez J, Raikhel NV
(2005) The power of chemical genomics to study the link between
endomembrane system components and the gravitropic response. Proc
Natl Acad Sci USA 102: 4902–4907
Turner JG, Ellis C, Devoto A (2002) The jasmonate signal pathway. Plant
Cell (Suppl) 14: S153–S164
Verma R, Peters NR, D’Onofrio M, Tochtrop GP, Sakamoto KM, Varadan
R, Zhang M, Coffino P, Fushman D, Deshaies RJ, et al (2004) Ubistatins
inhibit proteasome-dependent degradation by binding the ubiquitin
chain. Science 306: 117–120
Walling M (2000) Development of contraceptive implants. Br J Fam Plann
26: 12–13
Wasternack C, Hause B (2002) Jasmonates and octadecanoids: signals in
plant stress responses and development. Prog Nucleic Acid Res Mol Biol
72: 165–221
Xie DX, Feys BF, James S, Nieto-Rostro M, Turner JG (1998) COI1: an
Arabidopsis gene required for jasmonate-regulated defense and fertil-
ity. Science 280: 1091–1094
Xu L, Liu F, Lechner E, Genschik P, Crosby WL, Ma H, Peng W, Huang D,
Xie D (2002) The SCF(COI1) ubiquitin-ligase complexes are required for
jasmonate response in Arabidopsis. Plant Cell 14: 1919–1935
Xu L, Liu F, Wang Z, Peng W, Huang R, Huang D, Xie D (2001) An
Arabidopsis mutant cex1 exhibits constant accumulation of jasmonate-
regulated AtVSP, Thi2.1 and PDF1.2. FEBS Lett 494: 161–164
Zhao Y, Dai X, Blackwell HE, Schreiber SL, Chory J (2003) SIR1, an
upstream component in auxin signaling identified by chemical genetics.
Science 301: 1107–1110
Zouhar J, Hicks GR, Raikhel NV (2004) Sorting inhibitors (Sortins):
chemical compounds to study vacuolar sorting in Arabidopsis. Proc
Natl Acad Sci USA 101: 9497–9501
The Power of Chemical Genetics to Dissect Jasmonate Signaling
Plant Physiol. Vol. 141, 2006 1413
Dow
nloaded from https://academ
ic.oup.com/plphys/article/141/4/1400/6103502 by guest on 20 Septem
ber 2021