An arginine decarboxylase gene PtADC from Poncirus trifoliata confers abiotic stress tolerance and

16 Pages·2011·2.28 MB·English

JournalofExperimentalBotany,Vol.62,No.8,pp.2899–2914,2011
doi:10.1093/jxb/erq463 AdvanceAccesspublication31January,2011
Thispaperisavailableonlinefreeofallaccesscharges(seehttp://jxb.oxfordjournals.org/open_access.htmlforfurtherdetails)
RESEARCH PAPER
An arginine decarboxylase gene PtADC from Poncirus
trifoliata confers abiotic stress tolerance and promotes
primary root growth in Arabidopsis
Jing Wang1,*, Pei-PeiSun1,*, Chun-LiChen2,YinWang1, Xing-Zheng Fu1and Ji-Hong Liu1,†
D
1 NationalKeyLaboratoryofCropGeneticImprovement,KeyLaboratoryofHorticulturalPlantBiologyoftheMinistryofEducation, ow
n
HuazhongAgriculturalUniversity,Wuhan430070,China lo
a
2 CollegeofLifeSciences,HuazhongAgriculturalUniversity,Wuhan430070,China d
e
d
*Theseauthorscontributedequallytothiswork. fro
yTowhomcorrespondenceshouldbeaddressed.E-mail:[email protected] m
h
ttp
s
Received8September2010;Revised21December2010;Accepted23December2010 ://a
c
a
d
e
m
Abstract ic
.o
u
p
Arginine decarboxylase (ADC) is an important enzyme responsible for polyamine synthesis under stress conditions. .c
o
In this study, the gene encoding ADC in Poncirus trifoliata (PtADC) was isolated and it existed as a single-copy m
/jx
member. Transcript levels of PtADC were up-regulated by low temperature and dehydration. Overexpression of b
/a
PtADC in an Arabidopsis thaliana ADC mutant adc1-1 promoted putrescine synthesis in the transgenic line and the rtic
stomatal density was reverted to that in the wild type. The transgenic line showed enhanced resistance to high le
-a
osmoticum, dehydration, long-term drought, and cold stress compared with the wild type and the mutant. The b
s
accumulationofreactiveoxygenspecies(ROS)inthetransgeniclinewasappreciablydecreasedunderthestresses, tra
c
but ROS scavenging capacity was compromised when the transgenic plants were treated with the ADC inhibitor t/6
2
D-arginine prior to stress treatment. In addition, the transgenic line had longer roots than the wild type and the /8
/2
mutant under both normal and stressful conditions, consistent with larger cell number and length of the root 8
9
meristematic zone. Taken together, these results demonstrated that PtADC is involved in tolerance to multiple 9/4
7
stresses,anditsfunctionmaybedue,atleastpartly,toefficientROSeliminationandtoitsinfluenceonrootgrowth 8
0
conducivetodroughttolerance. 39
b
y
g
Keywords: Abioticstress,argininedecarboxylase,polyamine,Poncirustrifoliata,reactiveoxygenspecies,rootgrowth. u
e
s
t o
n
0
6
Introduction A
p
ril 2
Drought and low temperature are two of the most rootstock and cold tolerance in scions may extend citrus 0
1
devastating environmental stresses that affect plant growth production regions and maintain a sustainable citrus in- 9
and development, productivity, and geographic distribu- dustry. Genetic engineering via transfer of stress-responsive
tion. Trifoliate orange [Poncirus trifoliata (L.) Raf.] is genesoffersanewapproachforcropimprovement,andhas
widely used as a rootstock for commercial production of been proven to be a powerful tool for enhancing stress
citrus, a significant fruit crop in the world. However, resistance (Nakashima et al., 2009). As a first step towards
susceptibility to drought limits its use in areas where applyingthisapproachtocitrus/trifoliateorangegermplasm
irrigation is not readily available. In addition, as most of enhancement, favourable genes that can confer stress
thecommercialscionvarietiesarenotcoldhardy,freezingis toleranceneedto be identified.
afrequentoccurrence,causingsignificantlossestothecitrus Both cold and drought exert common effects on plants,
industry throughout the world (Sahin-Cxevik and Moore, such as alteration of osmotic potential and accumulation of
2006). Therefore, improvement of drought tolerance in the reactiveoxygenspecies(ROS),whichmayinduceanarsenal
ª2011TheAuthor(s).
ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommonsAttributionNon-CommercialLicense(http://creativecommons.org/licenses/by-
nc/2.5),whichpermitsunrestrictednon-commercialuse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
2900 | Wang et al.
of analogous stress responses in plants (Wang et al., 2003). transformants still remains elusive. In addition, in previous
It is well documented that plants have evolved a multitude reports, genetic engineering of ADC genes provided toler-
of physiological, biochemical, and molecular mechanisms ance to only a single stress, and the question thus arises as
enabling them to adapt to or tolerate harsh abiotic stresses. to whether or not manipulation of an ADC gene can confer
During the last decade, our understanding of the complex tolerance todifferentstresses,ashasbeenshown inthecase
regulatory networks associated with stress adaptation and of transformation with another polyamine biosynthetic
tolerance has advanced significantly, as evidenced by the gene, spermidine synthase (SPDS) (Kasukabe et al., 2004;
exploration of an array of stress-responsive genes (Urano Wen et al., 2008, 2010). Moreover, it is noteworthy that
et al., 2010). These genes are categorized into two main although ADC genes have been cloned from several fruit-
groups: functional genes and regulatory genes. Products of bearing plants, such as peach (Liu et al., 2009), grape
the former group, such as osmolyte biosynthetic enzymes, (Primikirios and Roubelakis-Angelakis, 1999), and apple
detoxification enzymes, and chaperones, are thought to (Hao et al., 2005), their functions have never been
function directly in protecting cells from external stresses elucidated. To address these issues, an ADC gene was
(Shinozaki and Yamaguchi-Shinozaki, 2007). Overexpres- cloned from trifoliate orange, and its function in tolerance
D
sion of these functional genes is expected to be an efficient to abiotic stresses, including high osmoticum, dehydration, o
w
strategy to enhance stress tolerance. In keeping with this drought, and cold, was analysed. The data presented here n
lo
rationale, attempts have been made to develop transgenic clearly show that the ADCgene of trifoliate orange encodes ad
e
plants with increased stress tolerance through manipulation a functional protein and its ectopic expression confers d
of functionalgenes (Umezawa etal.,2006). tolerancetomultiplestressesinArabidopsis,whichislargely fro
m
Polyamines, primarily putrescine (Put), spermidine (Spd), correlated with reduced accumulation of ROS. Further- h
and spermine (Spm), are low molecular weight aliphatic more, the results indicate that root growth in the transgenic ttps
polycations that are ubiquitously present in living organ- line was altered in a manner conducive to adaptation to ://a
c
a
isms. Being positively charged, they can interact with and/or survival under droughtconditions. d
e
negativelychargedmoleculessuchasproteins,nucleicacids, m
ic
and phospholipids (Martin-Tanguy, 2001), indicating that .o
u
they can stabilize cell membranes under stressful conditions Materials and methods p.c
o
andthereforealleviatecellinjury.Ithasalsobeensuggested m
that polyamines actasosmoticregulatorsandscavengersof Plantmaterials /jx
b
RThOuSs,(iDt rioslectonecteiavla.,bl1e9t8h6a;tEhvigahnesraancdcuMmuallamtiboenrgo,f1p9o89ly)-. TCreinfoteliraatetHoruaanzgheonsgeeAdsgroicbutlatuinreadlUfrnoimverNsitaytiwoneraelsCuirtfraucse-Bstreereildizinedg /article
amines through stimulated de novo biosynthesis can be aTnudckseorw,n19o6n9)m,e0d.7iu%m(wco/vm)paogsaerd, oanfdM5T%ba(wsa/vl)saslutscr(oMseur(apsHhig5e.8a)nind -abs
conducive to ameliorating stress-induced damage. order to obtain in vitro seedlings, which were grown in a growth tra
c
As cellular polyamine content is primarily modulated at chamberat2561 (cid:1)Cundera16hlight(providedbyincandescent t/6
the transcriptional level, polyamine biosynthesis may be light, 50lmol m(cid:1)2 s(cid:1)1)/8h dark regime. The seedlings were used 2/8
modified via overexpression of polyamine biosynthetic forgenecloningandanalysisofexpressionpatternsunderstresses. /2
8
Seeds of A. thaliana ecotype Wassilewskija (WS) and an adc1-1 9
genes. Polyamine biosynthesis involves a series of steps 9
mutant (kindly provided by Professor K. Shinozaki, RIKEN, /4
catalysed by specific enzymes (Bagni and Tassoni, 2001; Japan) were first vernalized for 2d at 4(cid:1)C, surface-sterilized 78
0
Alca´zar et al., 2010a), including arginine decarboxylase based on standard protocols (Weigel and Glazebrook, 2002), 3
9
(ADC, EC 4.1.1.19), which has been suggested to play an and then sown on germination medium containing half-strength b
y
important role in stress response (Urano et al., 2004; J. H. MS(MurashigeandSkoog,1962)salts,supplementedwith3%(w/v) g
u
sucrose and 0.75% (w/v) agar (pH 5.8). Petri dishes were e
Liu et al., 2006). Stress-inducible ADC genes have been s
incubated inanincubator at22(cid:1)C witha 16h light(50lmol m(cid:1)2 t o
cloned from various plants, and overexpression of the s(cid:1)1)/8h dark regime unless the seedlings were used for different n 0
cloned ADCgenes has also been documented. For example, experiments. 6
A
Roy and Wu (2001) were the first to introduce an oat p
(AvenasativaL.)ADCgeneintorice,resultingintransgenic Cloningandbioinformaticsanalysisofthefull-lengthADCgene ril 2
0
plants that showed enhanced tolerance to high salinity. fromP.trifoliata 19
Moreover, transgenic rice plants expressing the Datura
Insilicohomologysearchofacitrusexpressedsequencetag(EST)
stramonium ADC gene had greater drought tolerance than database, HarvEST (http://harvest.ucr.edu), was carried out in
the wild type (Capell et al., 2004). More recently, Alca´zar order to assemble an ADC contig. For amplification of a full-
and colleagues (2010b) demonstrated that overexpression of length cDNA encoding ADC, total RNA was isolated from
P. trifoliata seedlings treated at 4 (cid:1)C for 2 h in accordance with
the ADC2 gene led to enhanced drought tolerance in
the method of Y. Z. Liu et al. (2006). After treatment with
Arabidopsis thaliana. All of these results suggest that the
amplification-grade DNase I (Takara, Dalian, China) at 37 (cid:1)C to
ADC gene is a promising candidate for enhancing stress remove contaminant genomic DNA, 3 lg of total RNA was used
toleranceandmightbeofgreatvalueinmolecularbreeding. tosynthesizefirst-strandcDNAusingtheRevertAid(cid:2)FirstStrand
Although ADC genes have been cloned from various cDNA Synthesis Kit (Fermentas, Lithuania) according to the
manufacturer’s instructions. A pair of gene-specific primers
plants and genetic transformation of ADC genes yields
(ADC1,Table 1)thatcanamplifyfull-length cDNAwasdesigned
interesting outcomes, knowledge of the physiological mech- usingPrimerPremier5.0basedontheassembledcontig.ThePCR
anismsunderlyingtheenhancedstress tolerance intheADC cocktail consisted of 100ng of cDNA, 2 mM MgCl , 0.2 mM
2
Cloning and functional characterization of PtADC | 2901
Table1. Oligonucleotideprimersusedinthisstudy
Primers Primersequences(5’–3’)
Forward Reverse
ADC1 CCCCCTCGTTTTTTCTTTTTCTT TGTTCAACTGCTTCCATCTTTTG
ADC2 GGTCGAAACCGGAGCTGTTG- GCCCCCGATGTCAATCACTT
ADC3 CTTGTGGTGCCATCCTCTTGT AATCCTCCCCGTTGCAATAGT
ADC4 CCGCTCGAGCCCCCTCGTTTTTTCTTTTTCTT(XhoIsiteisunderlined) CGGGGTACCTGTTCAACTGCTTCCATCTTTTG(KpnIsiteisunderlined)
Actin TGGTTTTTTGTGTCTCGTTTCG CGAGATATGCCTCCTTCAATGG
NPT AGACAATCGGCTGCTCTGAT TCATTTCGAACCCCAGAGTC
35S-ADC2 GATGTGATATCTCCACTGACGTAAG ADC2reverse
SDD1 TCGGCGTGCTTGATACTGG TCCCGTTGAATCTCTTGCG
YDA TTCTACTCGGGTTTCACGC CATTTTCACAATCAGACGCC
CYCB1;1 ACTGTGCAAAGCTGTTGGCGTAT TTCGTCCTTGGAGTATTTCTTTCGT D
o
Tubulin CCTCCTGCACTTCCACTTCGTCTTC CGTGGATCACAGCAATACAGAGCC w
n
lo
a
dNTP, 0.5 U of Taq polymerase (Fermentas, Lithuania), 0.4 lM relative expression level of PtADC. The PCR solution (8 ll) de
d
PofCeRacahmpprliimficeartaionndw13asrepaecrtfioornmbeduffienr,ainthaertmotoaclyvcolelurm(AeBoIf 29070l0l., Bcoionstyasinteemds)4,l5l0onfg 2o3f cRDeNalAM,aasntedrM0.i5xl(MSYoBfReaGchreepnr,imAerp.pRlieed- from
Applied Biosystems) for 30 cycles of denaturation at 94 (cid:1)C for action programs consisted of an initial incubation at 50 (cid:1)C for h
30s, annealing at 51(cid:1)C for 30s, elongation at 72(cid:1)C for 3 min, 2 min and 95(cid:1)C for 10min, followed by 45 cycles of 95 (cid:1)C for ttp
s
followed by 72(cid:1)C for 7 min. The amplified cDNA was recovered 15s and 60C for 1 min. Each sample was analysed in four ://a
and cloned into pGEM-T Easy vector (Promega) and sequenced replicates, and the amplification efficiency and relative expression c
a
(UnitedGene, Shanghai, China). Sequence analysis was done in resultweredeterminedasdescribedpreviously(Wangetal.,2010). de
accordance with a previous study (Liu et al., 2009), including m
ic
homologysearchintheNCBI(NationalCenterforBiotechnology ConstructionofPtADCplantexpressionvector,plant .ou
Information)database,multiplealignmentsusingClustalX,analy- p
sis of typical motifs, and calculation of the isoelectric point (pI) transformation,andconfirmation .co
m
andmolecularweight. TEhaseyPvteAcDtoCr ucsoidnigngprriemgeiorsn cwoanstaiPnCinRg-aemithpelirfieadnfrXohmoIpoGrEKMp-nTI /jxb
DNAextractionandSouthernblottinganalysis rthesotsreictfioorntshietef(uAllD-leCn4g,thTacbDleN1A)ucnlodnerintgh,eesxacmepetPfCoRr tchoenadnitnioenalsinags /artic
le
Genomic DNA was extracted from trifoliate orange leaves using temperature (61(cid:1)C). The PCR product was double-digested with -a
the cetyltrimethyl ammonium bromide (CTAB) method as has XhoI and KpnI, and the resultant product was purified and bs
been described by Cheng et al. (2003). For Southern blotting subcloned in the sense orientation into XhoI/KpnI-linearized tra
c
analysis, 15lg of DNA was digested overnight with three re- binary vector pMV (kindly provided by Professor Zhi-Biao Ye, t/6
strictionenzymes(KpnI,XhoI,andXbaI)withoutcleavagesitesin HuazhongAgriculturalUniversity)harbouringakanamycin(Km) 2/8
PtADC. The digested products were fractionated on a 0.8% resistance marker and a cauliflower mosaic virus (CaMV) 35S /2
agarose gel, followed by transfer to Hybond-N+ nylon membrane promoter. After fidelity confirmation by sequencing, the construct 89
9
(Amersham Pharmacia Biotech, NJ, USA) in 0.4 M NaOH. To wasintroduced into Agrobacterium tumefaciens strain EHA105by /4
7
produce a probe, a 500bp PCR product was amplified from the electroporation,whichwasthenusedtotransformtheArabidopsis 8
0
pGEM-T Easy plasmid using primer ADC2 (Table 1) with the mutant adc1-1 by floral dipping (Clough and Bent, 1998). The T 3
0 9
following PCR program, 30 cycles at 94(cid:1)C for 30s, 61(cid:1)C for seeds were harvested and sown on MS supplemented with 50mg b
y
30s, 72(cid:1)C for 45s, followed by 72(cid:1)C for 5 min. The PCR l(cid:1)1Kmasaselectionmarker,andKm-resistantplantswerePCR- g
product was labelled as a probe with [a-32P]dCTP via random confirmed using primers specific to the NPTII gene (NPT) and ue
s
priming. Membrane hybridization and washing under high strin- a pair of primers covering CaMV 35S and PtADC (35S-ADC2, t o
gency conditions were performed according to the method of Table 1), which produced 750bp and 1400bp amplicons, re- n 0
Sambrooketal.(1989),followedbyexposure ofthemembrane to spectively. The expression level of PtADC in the positive plants 6 A
X-rayfilm(Fuji,Japan)at–70(cid:1)Cfor4h. was analysed by semi-quantitative RT-PCR following the same p
procedure as for probe preparation, and the expression level of ril 2
Analysisofthegeneexpressionbyquantitativereal-timePCR Tubulin (Table 1) was used as an internal control. The transgenic 019
line withthehighest PtADC expression wasselected to harvest T
(QRT-PCR) 3
homozygouslinesandutilizedforthesubsequentexperiments.
To investigate how PtADC responds to stress, 2-month-old
trifoliate orange seedlings with fully expanded leaves were placed
Quantificationoffreepolyaminesbyhigh-performanceliquid
in empty conical flasks at ambient temperature for dehydration
chromatography(HPLC)
and the leaves were sampled at different time points (0, 0.5, 1, 3,
and 7 h). In addition, another set of seedlings were cultured at Free polyamines in 6-week-old seedlings of WS, adc1-1, and the
4 (cid:1)C,andtheleavesweresampledatdifferenttimepoints(0,0.25, transgenic line grown under the same conditions were extracted
1, 3, and 7 d). Total RNA isolation and cDNA synthesis were andmeasuredaccordingtothemethodofWuetal.(2009).A20ll
performed as described above. QRT-PCR with the SYBR Green aliquot of the prepared polyamine solutions was loaded into an
dye method was carried out in an ABI7500 cycler (Applied HPLC(Waters,USA)equippedwithaC reversedphasecolumn
18
Biosystems, Foster City, CA, USA) to evaluate gene expression. (4.6mm3250mm, particle size 5 lm) at 30(cid:1)C using a methanol/
Gene-specificprimers(ADC3,Table1)weredesignedbasedonthe HPLC-grade water gradient, changing from 60%:40% (v/v, meth-
non-conserved regions of the full-length sequence, while a Tubulin anol:water)to95%:5%in23minataflowrateof1 mlmin(cid:1)1.The
gene (Table 1) was used as an internal control to normalize the eluates were detected using a fluorescence spectrophotometer with
2902 | Wang et al.
excitation and emission wavelengths of 365nm and 510nm, andincubatedin5 mlofdistilledwater,andshakenonagyratory
respectively. Quantification of free polyamines was repeated three shaker (120 rpm) at room temperature for 1 h. The initial
times,withthreetriplicatesforeachline. conductivity (C1) was measured with a conductivity meter (DSS-
307,Shanghai,China).Thesampleswerethenboiledfor10minto
induce complete leakage, and cooled down at room temperature
AbioticstresstreatmentofWS,adc1-1,andthetransgenicline
before measurement of electrolyte conductivity (C2). Electrolyte
Fortheassayofsensitivitytoosmoticstress,seedsharvestedfrom leakage (C) was calculated according to the equation C
WS, adc1-1, and the transgenic plants were sown separately on (%)¼1003C1/C2.
germination medium, or on germination medium supplemented MDAcontentwasdeterminedbythethiobarbituricacid(TBA)-
with either 400mM mannitol or 125mM NaCl. The germination based colorimetric method as described by Heath and Packer
statusof the threelineson thesemediawasmonitored andscored (1968).Theleafsamplesharvestedafterdehydrationwerehomog-
at2 dintervalsduringa14dexperimentationperiod. enized in 5% (w/v) trichloroaceticic acid (TCA) and then centri-
For dehydration analysis, 10-day-old in vitro seedlings or fully fugedat10000rpmfor10min.A2 mlaliquotofthesupernatant
expandedleavesdetachedfrompottedplantswereputondryfilter washeatedwith2.0 mlof0.67%(w/v)TBAat100(cid:1)Cfor 30min.
paper for up to 50min, and the fresh weight (FW) was measured After cooling to the ambient temperature, the mixture was
every 10min to determine the rate of water loss. Electrolyte centrifuged at 10000rpm for 5 min. Absorbance of the superna-
leakage (EL) and malondialdehyde (MDA) were measured on the tantwasmeasuredat450,532,and600nminaspectrophotometer
D
samplescollectedatthelasttimepoint. (Varian Cary 50 Scan, Australia). The MDA concentration, o
w
For drought stress, 30 plants of each line grown in pots (10cm expressed as nmol ml(cid:1)1, was calculated according to the formula: n
lo
diameter) filled with vermiculite were constantly watered for 3 MDAcontent¼6.45(A –A )–0.56A . a
532 600 450 d
weeks before water was withheld for up to 18d. During drought Leaf chlorophyll content was calculated as the mean of three e
d
treatment, morphological changes in the plants were monitored readings for each leaf from a portable chlorophyll meter (SPAD- fro
and photographed when differences were apparent, and EL and 502, Minolta, Japan), and three leaves were read for each plant. m
swuerrveivcaolnrsaidteerwederedeeaxdamifinaeldl tahtetlheeaveensdtuorfntehdeyeexlploewrim. Iennt.anPolathnetsr Majorveinswereavoidedwhenmeterreadingsweretaken. http
s
edxrpoeurgihmtefnotr,1p0ladn,tsaantdtphheorteopgrroadpuhcstiwveersetatgaekewne.realsosubjectedto DetectionofO2– andH2O2 ://aca
For cold stress, 4-week-old plants (17 for each line) grown at Accumulation of O2– and H2O2 was detected by a histochemical dem
22(cid:1)Cweremovedtoagrowthchambersetat0 (cid:1)Candtreatedfor staining method using nitro blue tetrazolium (NBT) and diamino- ic
24h under continuous light. Survival rate, EL, and chlorophyll benzidine (DAB), respectively, according to Shi et al. (2010). For .o
content were measured after the treatment was stopped. Mean- O– detection, the samples were immersed in 1 mg ml(cid:1)1 NBT up
2 .c
while, the treated plants were shifted to normal growth condition solutionpreparedin10mMphosphatebuffer(pH7.8)atambient o
m
forIn15ads.eparate experiment, 15-day-old in vitro seedlings of the toefmHperOatu,rethuendsaermlpiglehstuwnetrileblpuleacsepdotsinap1pemagredm.lF(cid:1)o1rflorecsahlizDatAioBn /jxb
2 2 /a
twrahnicshg,enailcolningewwiethreWincSubaantdedaidnc13-1m,MweDr-earegiitnhienred(Deh-ayrdgr)aftoedr3fodr, suonltuiltiobnro(wpnHs3p.o8)tspwreepraerevdisiibnle1.0TmhMe sptahionsepdhastaembpulefsferw(epreHt7h.e8n) rticle
50minortreatedat4 (cid:1)Cfor12h.Leavesfromtheseedlingswere bleachedinconcentratedethanolandkeptin70%ethanolpriorto -a
b
sampledforROSstainingasmentionedbelow. takingpictureswithadigitalcamera. s
tra
c
t/6
Rootgrowthassayandrelativewatercontent(RWC)measurement Analysisofstomataldensityandexpressionofgenesinvolvedin 2
/8
Seeds from the three lines were sown in Petri dishes containing stomataldevelopment /2
8
germination medium as mentioned above, and were placed Stomatal density of 6-week-old in vitro seedlings or potted plants 99
vertically in a growth chamber. The primary root length of 30 grown under the same conditions was examined by light micros- /4
7
plantswasmeasuredevery2 dovera14dperiodstartingfrom3 d copy and scanning electron microscopy (SEM). For light micros- 8
0
aftersowing.Inanotherexperiment,35-day-oldpottedplantsthat copy,theepidermiswasstrippedandexaminedunderanOlympus 39
had been watered regularly were either continuously watered or BH-2 microscope equipped with an Olympus DP70 CCD camera. b
y
subjected to drought treatment. Ten days later, the root length of For SEM, ultrathin leaf sections were prepared as previously g
u
the plants was measured, and the leaves were simultaneously described (Dong et al., 2009), and observed with a scanning es
sampled to perform an RWC assay using a Moisture Balance electronic microscope (H-7650, Hitachi, Ltd, Tokyo, Japan). The t o
n
(Mettler, HG63, Switzerland) according to the manufacturer’s pictureswerecollectedasJPEGdigitalfiles.Atotalof150images 0
instruction. were obtained for each line, and the number of stomata in each 6 A
For microscopic observation, root tips of 3-day-old in vitro imagewascountedandconvertedintofinalstomataldensityusing p
seedlings were sampled and cleared in a mixture of chloral Image Pro Plus 5.0. Using the same software, epidermal cell size ril 2
hydrate:distilled water:glycerol (8:3:1, v/v/v) and visualized using wasalsomeasured,basedon15cellsfromthreerandomlyselected 01
9
Nomarski optics on a Nikon 80i (Tokyo, Japan) microscope photos of each line. In addition, expression levels of three genes,
equipped with a Nikon DXM1200 digital camera. Root meristem SDD1, YDA, and CYCB1;1, which are related to stomatal
size was measured and the number of cortex cells in a file development (Nadeau, 2009), were assessed using leaves from the
extending from the quiescent centre to the first elongated cortex potted plants that were sampled together with those for micros-
cell (Ioio et al., 2008) was counted using ImageJ software (http:// copy. After RNA extraction and cDNA synthesis, RT-PCR was
rsb.info.nih.gov/ij/). For analysis of the expression level of performed following the procedure mentioned above, except for
CYCB1;1, a metaphase gene, roots of 3-day-old in vitro seedlings theuseofprimersspecifictothesegenes(Table1).
were collected. RNA extraction, cDNA synthesis, and RT-PCR
using specific primers were done as mentioned for probe
Statisticalanalysis
generation.
Thetreatmentswererepeatedatleasttwicewithconsistentresults.
The results from one representative experiment are shown herein,
MeasurementofEL,MDA,andchlorophyllcontent
expressedasmeansandstandarderrorscalculatedusingMicrosoft
EL was measured based on the method of J. H. Liu et al. (2006) Excel.Thedatawereprocessedusingone-wayanalysisofvariance
with slight modification. The collected leaves were cut into strips (ANOVA), and statistical differences were compared based on
Cloning and functional characterization of PtADC | 2903
Student’s t-test, taking P <0.05 (*), P <0.01 (**), and P <0.001 levelofPtADCwasinducedafterinitiationofbothstresses,
(***)assignificant. which progressed until a maximum level was reached at 3 d
(cold, Fig. 3A) or 3 h (dehydration, Fig. 3B). Although
there was a decline at the last time point, the expression
Results levels of PtADC were still higher than those at onset of the
stresses. These results suggest that PtADC is a stress-
Cloning and analysis of PtADC from trifoliate orange
responsivegene.
A total of 26 ESTs that are highly homologous to ADC
geneswereidentifiedinthecitrusESTdatabase(HarvEST),
which were successfully assembled into one contig. Sub- Overexpression of PtADC in an A. thaliana mutant and
sequently, RT-PCR amplification of the cDNA using the analysis of transgenic plants
primers designed from the contig yielded a single fragment,
which was verified as 2493 bp by sequencing. The sequence Induction of PtADC mRNA by abiotic stress prompted
showed a high degree of homology to ADC genes in the investigation of whether this gene is involved in stress
D
database, indicating that it was an ADC gene homologue. tolerance. For this purpose, PtADC was ectopically ow
The putative gene was thus designated as PtADC (Poncirus expressed in the Arabidopsis adc1-1 mutant. A total of 15 nlo
trifoliata arginine decarboxylase) and deposited in GenBank independent transgenic lines were generated and confirmed ad
e
wcointhtatinheedacace2s2si5o6nbnpumopbeenr HreQad0i0n8g23fr7a.mTehe(OPRtAFD) CenccDodNinAg bcoypPyCtrRan(dsgaetnaenwotersehsoewlenc)t.edFotourevoafltuhaetseetlhineeesxwpriethssiaosninlegvlee-l d from
7rae5gc1iaoalncmui(lnaUoteTdaRcim)daornle.sdiwdaut.e1so5,f3al8bo0pn.5g33w’-kUitDhTaaRna.n8Pd4tAbapDp5Cr’e-dupinrcottretaedninsplaIhtaeoddf ot5h-f2eP(dtFAififDge.rC4e.nATt).hlieLneimns,eRw5N-i2tAhwlteahvseedlheoisgfihgthneseattteerdaxnpasrsgeOsesnEieonavnalrdeivesedulbainjmecoltinendge https://ac
a
5.14. to further analysis. HPLC analysis showed that adc1-1 and de
To compare sequence homology between PtADC and WS had similar levels of free polyamines, which is mic
other known ADCs, multiple sequence alignment of the consistent with a previous observation (Urano et al., 2005). .ou
deduced amino acids of PtADC was carried out with nine OE contained significantly higher free Put content p.c
ADCs from other plants. PtADC was found to share high (436.9 nmol g(cid:1)1 FW) than did either adc1-1 (216.5 nmol om
sequence identity with the known ADCs, particularly in the g(cid:1)1 FW) or WS (224.6 nmol g(cid:1)1 FW), whereas Spd was /jxb
mteridmdilnealporergtiioonnsoefxhthibeitseedquaenlocwe,erwdheegreraees othfecoNn-searvnadtioCn- o(Fnilgy.sl4igBh),tlyinidniccraetainsegdtahnadtSopvmereexxphribesitseiodnneogfligPibtAleDcChanpgrees- /article
(Fig. 1). Sequence analysis showed that PtADC contains dominantly influenced endogenousPut content inthe trans- -ab
s
two highly conserved decarboxylase motifs, pyridoxal genic line. tra
c
phosphate attachment site (FPVKCNQDRFVVEDIVKFG, t/6
2
position 166–184) and signature 2 (VRLGANMQVI- /8
/2
DIGGGLG, position 344–360). In addition, the putative Germination of Arabidopsis seeds on osmotic medium 8
9
substrate-binding site (DIGGGL; Moore and Boyle, 1990) 9/4
and the cleavage site of ADC processing (GASD; Bell and In order to compare germination efficiency, seeds of WS, 78
0
Malmberg,1990)werealsoidentified. adc1-1,andOEweresownongerminationmediumaloneor 39
on germination medium supplemented with either 400 mM b
y
mannitol or 125 mM NaCl. Without addition of mannitol g
Genomic distribution of PtADC in trifoliate orange ue
or NaCl, germination of OE was comparable with that of s
t o
Southern blotting was carried out to examine PtADC gene WS and adc1-1, although it germinated better within the n
0
copy number in the trifoliate orange genome. As shown in first 2 d after germination (Fig. 5A, C). On the medium 6
A
Fig. 2, hybridization of the genomic DNA digested with containing mannitol or NaCl, germination of WS, adc1-1, p
three restriction enzymes using a 500 bp partial fragment and OE was repressed, but clear differences were noted. ril 2
0
of PtADC as probe produced only one visible signal Whensownonthesemedia,OEseedsbegantogerminateat 19
under high stringency conditions. As the probe sequence 2 d with a dramatically lower germination rate than on
did not contain any recognition sites of all the three normal medium, but the rate continued to increase until it
enzymes, the hybridization pattern suggested that the peaked at 82.2% (mannitol) or 84.3% (NaCl). In contrast,
PtADC gene exists as a single-copy gene in the trifoliate germination of WS and adc1-1was observed at6 d (WS) or
orange genome. 8 d (adc1-1) on the mannitol-containing medium and at 4 d
(WS) or 6 d (adc1-1) on the NaCl-containing medium. The
highestgerminationratesofWSandadc1-1were14.2%and
ExpressionprofilesofPtADCinresponsetodehydration
16.3% on the mannitol-containing medium and 29.4% and
and low temperature treatment
16.8% on the NaCl-containing medium (Fig. 5B, D, E),
QRT-PCR was performed to examine the expression whichweresignificantlylower thanthoseofOEatthesame
profiles of PtADC in response to dehydration and low time point, suggesting that they were more susceptible than
temperature (4 (cid:1)C). The results indicated that the transcript OE toosmoticstresses.
2904 | Wang et al.
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/jx
b
/a
rtic
le
-a
b
s
tra
c
t/6
2
/8
/2
8
9
9
/4
7
8
0
3
9
b
y
g
u
e
s
t o
n
0
6
A
p
ril 2
0
1
9
Fig.1. MultiplealignmentsofthededucedaminoacidsequenceofPtADCandotherplantADCs,includingMdADC(AB181854),
PpADC(AB379849),VvADC(CAO47255),AtADC1(NP_179243.1),AtADC2(NM_119637),NtADC1(AF127240),GmADC(U35367),
DsADC(AJ251898),PsADC(Z37540),andOsADC1(EU220429).Blackandgreyshadedbackgroundsindicatethattheaminoacids
wereidenticalorsimilartothoseofPtADC,respectively.Orn/DAP/Argdecarboxylasesfamily2pyridoxalphosphateattachmentsite
(motifA)andsignature2(motifB)areunderlined.Thelysineresidueinvolvedinpyridoxalphosphatebindingisindicatedby‘d’,andthe
putativesubstratebindingsiteandcleavagesiteofADCprocessingareshownbyadoublelineanddashedline,respectively.
Cloning and functional characterization of PtADC | 2905
Fig.3. Quantitativereal-timePCRanalysisofPtADCexpressionin
trifoliateorange.TimecoursechangeofPtADCmRNAlevelsinthe D
o
w
seedlingsthatunderwentlowtemperature(A)ordehydration(B) n
lo
treatment.Datarepresentmeans6SEoffourreplicates. a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/jx
b
Fig.2. SouthernblottinganalysisofPtADCinthetrifoliateorange /a
genome.Intotal15 lgofgenomicDNAdigestedwithKpnI,XhoI, rticle
orXbaIwasfractionatedona0.8%agarosegel,blottedtoanylon -a
b
s
membrane,andhybridizedasdescribedintheMaterialsand tra
methods.Themolecularsizeofthehybridizedsignalsisindicated Fig.4. OverexpressionofPtADCinadc1-1enhancedputrescine ct/6
ontheleft(kb). synthesis.(A)RT-PCRanalysisofPtADCmRNAexpressionin 2/8/2
seedlingsoftheadc1-1mutantandfourindependenttransgenic 89
9
AbioticstresstoleranceanalysisofWS,adc1-1,andOE lines(21-3,5-2,19-4,and14-6)expressingthePtADCtransgene /4
7
plants (forconvenienceofdescription,line5-2wasdesignatedasOEin 80
3
thefigureshereafter).Tubulinwasusedasaninternalcontrolto 9
To evaluate the stress tolerance capacity further, WS, adc1- b
1,andOEplantsweresubjectedtodehydration,drought,or showthatanequalamountofRNAwasloadedintheanalysis.(B) y g
u
Endogenousfreepolyaminelevelsin6-week-oldwild-type(WS), e
cold treatments. For dehydration, 10-day-old in vitro seed- s
adc1-1,andOEseedlings,asrevealedbyHPLCquantification. t o
lings or expanded leaves detached from potted plants of the n
Dataonfreepolyamineswerethemean6SEofthreereplicates. 0
three lines were kept in the ambient environment for up to 6
AsteriskswithorwithoutparenthesesindicatethatthevalueofOE A
50 min, and phenotype, water loss, and EL were then p
compared. The in vitro samples showed a progressive in- issignificantlydifferentfromthatofWSandadc1-1,respectively ril 2
(**P<0.01). 0
crease of water loss over the 50 min duration. The largest 19
loss was observed in adc1-1, followed by WS, whereas OE
exhibited the least water loss (Fig. 6A). Although no significantly lower in OE than in WS and adc1-1 (Fig. 6C).
significant difference was detected between OE and WS, Overall, OE was more tolerant to dehydration stress than
fresh water loss of OE was significantly lower than that of either WS or adc1-1.
adc1-1 from 30 min onwards. OE exhibited better leaf As the dehydration mentioned above is a short-term
turgor than WS and adc1-1 after 50 min dehydration (data stress, efforts were made to treat the plants in a long-term
not shown). In addition, EL of adc1-1 was significantly manner by withholding water from the potted plants for as
higher than that of WS and OE after 50 min of dehydra- longas18 d.Attheonsetofdroughttreatment,noobvious
tion. OE had lower EL than WS, but the difference was difference was observed among the tested lines. With the
statistically insignificant (Fig. 6B). Water loss of the leaves extension of drought stress, OE showed better growth than
sampledfrompottedplantsexhibitedasimilartrendtothat WS and adc1-1 (Fig. 7A). EL was significantly lower in
of the in vitro seedlings (data not shown), while EL was OE (57.4%) than in WS (75.5%) or adc1-1 (80.9%)
2906 | Wang et al.
D
o
w
n
lo
a
d
e
d
fro
m
h
Fig.5. EffectofosmoticstresscausedbymannitolandNaClongerminationofthewildtype(WS),adc1-1,andOE.(AandB) ttp
s
Germinationrateofseedssownongerminationmediumnotsupplemented(A)orsupplemnetedwith400 mMmannitolor125mMNaCl ://a
(B)duringa14dcycle;germinationwasscoredattheindicateddaysaftergermination.Asterisksshowthatthevaluesaresignificantly c
a
d
differentcomparedwithadc1-1,andasterisksinparenthesesindicateasignificantdifferencebetweenOEandWS(*P<0.05,**P<0.01, e
m
and***P<0.001).(C–E)Representativephotographstaken14daftertheseedsweresownongerminationmediumnotsupplemented ic
.o
(C)orsupplementedwith400 mMmannitol(D)or125 mMNaCl(E). u
p
.c
o
m
after 18 d drought treatment (Fig. 7B). In addition, 52.6% /jxb
of the transgenic plants survived after drought treatment, /artic
higherthanthe31.6%ofWSand26.3%ofadc1-1(Fig.7C). le
In addition, growth of OE plants under drought at -ab
s
the reproductive stages was remarkably superior to that tra
c
of WS and adc1-1 (Fig. 7D), indicating that OE was t/6
more resistant to long-term drought stress than the other 2/8
two types of plants at both vegetative and reproductive /28
9
stages. 9
/4
Afterexposureto0 (cid:1)Cfor1 d,OEplantsexhibitedbetter 7
8
0
leaf morphology than the WS and adc1-1 plants (Fig. 7E). 3
9
When the cold-treated plants were transferred to normal b
y
growth conditions for 15 d, OE plants resumed growth gu
e
and grew vigorously, whereas WS and adc1-1 plants died s
t o
(Fig. 7F). OE plants had lower EL (Fig. 7G) and higher n
0
chlorophyllcontent(Fig.7H)comparedwithWSandadc1-1, 6
A
indicating that the transgenic line was less susceptible to p
cold stress. ril 2
0
Fig.6. Dehydrationresponseofwild-type(WS),adc1-1,andOE 19
invitroseedlingsorleavesfrompottedplants.(A)Time-courseof
Production of ROS and MDA under abiotic stresses
waterlossofWS,adc1-1,andOEduringa50 minexperimenta-
tionperiod.Ten-day-oldinvitroseedlingsdetachedfromgermina- Histochemical stainingwith DAB and NBT was carriedout
tionmediumweredehydratedintheambientenvironment,andthe to check the levels of H O and O–, respectively. After
2 2 2
freshweightwasmeasuredattheindicatedtimeintervals.Water dehydration treatment the in vitro seedlings (Fig. 8A, B) or
losswascalculatedfromthedecreaseoffreshweightrelativeto leaves from potted plants (Fig. 8C, D) of adc1-1 and WS
thatbeforedehydration.(BandC)Electrolyteleakageofinvitro showed similar staining patterns with both DAB and NBT;
seedlings(B)orleavesfrompottedplants(C)ofWS,adc1-1,and however, they exhibited deeper staining than OE. Staining
OEthatweresampledatthelasttimepointofdehydration. patternsofthethreelinesunderlowtemperature(Fig.8E,F)
Asterisksshowthatthevaluesaresignificantlydifferentcompared were similar to those under dehydration, indicating that the
withadc1-1,andasterisksinparenthesesindicateasignificant transgenic line accumulated less H O and O– under these
2 2 2
differencebetweenOEandWS(*P<0.05,**P<0.01). stresses.
Cloning and functional characterization of PtADC | 2907
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/jx
b
/a
rtic
le
-a
b
s
tra
c
t/6
2
/8
/2
8
9
9
/4
7
8
0
3
9
b
Fig.7. Droughtandcoldstressresponseofwild-type(WS),adc1-1,andOEpottedplants.(A)Representativephotographsshowingthe y
g
phenotypeofplants0(beforedrought),15,and18dafterdroughttreatment.(BandC)Electrolyteleakage(B)andsurvivalrate(C)of ue
s
WS,adc1-1,andOEplantsthatweremeasuredattheendofdroughttreatment.Asteriskswithorwithoutparenthesesshowthatthe t o
n
valuesofOEaresignificantlydifferentcomparedwiththoseofWSandadc1-1,respectively(*P<0.05,**P<0.01).(D)Arepresentative 0
6
photographshowingthephenotypeofWS,adc1-1,andOEplantsexposedtodroughtatthereproductivestage.(EandF) A
p
Representativephotographsshowingthephenotypeofplantsthathavebeentreatedat0 (cid:1)Cfor24h(E)and15 daftertransfertothe ril 2
normalconditionforrecovery(F).(GandH)Electrolyteleakage(G)andchlorophyllcontent(H)ofWS,adc1-1,andOEplantsmeasured 0
1
attheendoflowtemperaturetreatment. 9
AccumulationofROSmaycauselipidperoxidation,which accumulation of ROS is related to the increased level of
can be estimated by the amount of MDA, a product of endogenous Put in this line. To answer this question, D-arg,
unsaturated fatty acid peroxidation. As OE differed in ROS an inhibitor of ADC, was used to treat OE seedlings before
accumulationrelativetotheothertwolines,MDAcontentin exposing them to dehydration or low temperature along
the leaves was measured after dehydration. The MDA with adc1-1 and WS. DAB and NBT staining showed that
content of WS and adc1-1 was 28.2 nmol mg(cid:1)1 protein and the D-arg-treated OE seedlings showed a staining pattern
24.8 nmol mg(cid:1)1 FW, which was significantly higher than similar to or even deeper than that of the mutant and WS
that of OE (16.4 nmol mg(cid:1)1 FW), suggesting that OE afterdehydration for50 min (Fig.8G, H).Thisobservation
underwentlesslipidperoxidation. is very different from the staining profile obtained for OE
As the Put content was significantly higher in OE than in without inhibitor treatment. As for low temperature treat-
adc1-1 and WS, it is logical to question whether the lower ment, D-arg pre-treatment of OE also compromised the
2908 | Wang et al.
D
o
w
n
lo
a
d
e
d
fro
m
h
ttp
s
://a
c
a
d
e
m
ic
.o
u
p
.c
o
m
/jx
b
/a
rtic
le
-a
b
s
tra
c
t/6
2
/8
/2
8
9
9
Fig.8. HistochemicalstainingassayofROSaccumulationwithnitrobluetetrazolium(NBT)anddiaminobenzidine(DAB)inthewildtype /4
7
8
(WS),adc1-1,andOEunderabioticstresses.(A–D)DAB(A,C)andNBT(B,D)stainingofinvitroseedlings(A,B)andleavesfrompotted 0
3
plants(C,D)thathavebeensubjectedtodehydrationfor50 min.(EandF)DAB(E)andNBT(F)stainingofleavesfrompottedplantsthat 9 b
weretreatedat0(cid:1)Cfor24h.(GandH)DAB(G)andNBT(H)stainingofinvitroplantsofwild-type(WS),adc1-1,andD-arg-treatedOE y g
u
plantsthathavebeensubjectedtodehydrationfor50min. e
s
t o
n
0
staining, although the effect was not comparable with that temperature stress. As water loss, especially within a short 6
A
observed under dehydration (data not shown). Taken period (50 min herein), is primarily related to evaporation p
together,thesedataindicatedthatROSscavengingcapacity through stomata, it was felt to be necessary to determine if ril 2
0
was impaired when OE plants were pre-treated with D-arg, stomatal density differed among the three lines. Examina- 19
suggesting that a higher level of cellular Put may be tion of potted plants (Fig. 9A–C) demonstrated that adc1-1
correlatedwithloweraccumulation ofROS under stress. had the highest stomatal density (347 mm(cid:1)2), followed by
WS (260 mm(cid:1)2) and OE (233 mm(cid:1)2). Stomatal density was
greater in the leaves of in vitro seedlings than in those of
Stomatal density and expression of related genes in
potted plants, but the three lines showed a similar trend
WS, adc1-1, and OE plants
(data not shown), which was largely consistent with their
The above data showed that adc1-1 lost fresh water more water loss under dehydration. The size of epidermal cells in
quickly than WS and OE, while the latter two were not the three lines was also compared, revealing an opposite
significantly different. Since OE contained higher free Put trend to that found for the stomatal density. OE had the
content than either WS or adc1-1, this result is inconsistent largest cell size, followed by WS and adc1-1 (Fig. 9D). As
when compared with the pronounced difference between stomatal development is partly regulated at the transcrip-
OE and the other two lines under drought and low tional level, expression levels of the three genes (SDD1,