Transgenic Res (2014) 23:643–655 DOI 10.1007/s11248-014-9803-2

ORIGINAL PAPER

OsLOX2, a rice type I lipoxygenase, confers opposite effects on seed germination and longevity Jiexue Huang • Maohong Cai • Qizhang Long Linglong Liu • Qiuyun Lin • Ling Jiang • Saihua Chen • Jianmin Wan

•

Received: 7 September 2013 / Accepted: 18 April 2014 / Published online: 4 May 2014  Springer International Publishing Switzerland 2014

Abstract Rice production and seed storage are confronted with grain deterioration and loss of seed viability. Some members of the lipoxygenase (LOX) family function in degradation of storage lipids during the seed germination, but little is known about their influence on seed longevity during storage. We characterized the role of rice OsLOX2 gene in seed germination and longevity via over-expression and knock-down approaches. Abundant expression of OsLOX2 was detected in panicles, roots, and stems, but not in leaves. Moreover, OsLOX2 was highly

Electronic supplementary material The online version of this article (doi:10.1007/s11248-014-9803-2) contains supplementary material, which is available to authorized users. J. Huang  M. Cai  Q. Long  L. Liu  Q. Lin  L. Jiang  S. Chen (&)  J. Wan (&) State Key Laboratory for Crop Genetics and Germplasm Enhancement, Jiangsu Provincial Center of Plant Gene Engineering, Nanjing Agricultural University, Weigang 1, Nanjing 210095, China e-mail: [email protected] J. Wan e-mail: [email protected] J. Wan Institute of Crop Science, The National Key Facility for Crop Gene Resources and Genetic Improvement, Chinese Academy of Agricultural Sciences, Beijing 10081, China

induced during germination. OsLOX2 protein, located in the cytoplasm, showed a wide range of temperature adaptation (20–50 C) and a substrate preference to linoleic acid. Lines over-expressing OsLOX2 showed accelerated seed germination under normal condition and lower seed viability after accelerated aging. RNA interference (RNAi) of OsLOX2 caused delayed germination and enhanced seed longevity. RNAi lines with strongly repressed OsLOX2 activity completely lost the capability of germination after accelerated aging. More lipid hydroperoxide were found in OE15 than the control, but less in RNAi lines than in the WT Nipponbare. Therefore, OsLOX2 acts in opposite directions during seed germination and longevity during storage. Appropriate repression of the OsLOX2 gene may delay the aging process during the storage without compromising germination under normal conditions. Keywords Oryza sativa L.  OsLOX2  Seed germination  Seed longevity  Grain storage Abbreviations LOXs Lipoxygenase LA a-Linoleic acid LeA Linolenic acid GFP Green fluorescent protein OE Over-expression RNAi RNA interference HPOD Hydroperoxide LHP Lipid hydroperoxide

123

644

Introduction Lipoxygenases (LOXs; linoleate:oxygen oxidoreductase, EC 1.13.11.12) belong to a family of non-heme iron-containing dioxygenases that catalyze oxidation of polyunsaturated fatty acids (PUFAs) to produce unsaturated fatty acid hydroperoxides (Feussner and Wasternack 2002). They are widely distributed throughout the plant, animal, and microorganism kingdoms (Kuhn and Thiele 1999; Oliw 2002; Rosahl 1996). Metabolism of PUFAs via a LOX-catalyzed step and subsequent reactions are collectively named LOX pathways (Blee 2002). Plant LOXs are classified with respect to their positional specificities for linoleic acid (LA) oxygenation. LA is oxygenated at either carbon atom 9 (9-LOX) or 13 (13-LOX) of the hydrocarbon backbone of the fatty acid, generating two groups of compounds, viz. (9S)- hydroperoxy- or the (13S)-hydroperoxy derivatives of LA (Brash 1999). Another classification of plant LOXs was proposed according to their primary structures (Shibata et al. 1994). Type I-LOXs lack a putative chloroplast transit peptide and show high sequence similarities ([ 75 %); Type II-LOXs carry a plastidic transit peptide on the N-terminal and have about 35 % sequence similarity to each other. In plants, LOXs are an enzyme group with various physiological functions, such as regulation of plant development, synthesis of signal substances like phytodienoic acid (OPDA), jasmonic acid (JA) and abscisic acid (ABA), and hypersensitive response to Pseudomonas syringae infection (Beckman and Ingram 1994; Feussner et al. 2001; Siedow 1991). The LOX pathway is usually in a quiescent state, and is activated only when an organism develops to a certain stage, or is subjected to environmental stress (Gardner 1995). The intracellular localizations of LOX proteins provide a clue for their differential physiological functions in plants (Feussner and Wasternack 2002). LOX-dependent degradation of storage lipids occurs during seed germination (Feussner et al. 2001). At the phospholipid monolayer of lipid bodies in various oilseed and cucumber seedlings, a specific linoleate, 13-LOX, is capable of oxygenating esterified linoleate residues without the preceding action of a lipid hydrolysing

123

Transgenic Res (2014) 23:643–655

enzyme (Feussner et al. 1996, 1997a). Particulate LOXs were also found in microsomal membranes (Feussner and Kindl 1994), plasma membranes (Nellen et al. 1995), the cytosol and vacuole (Grayburn et al. 1991; Stephenson et al. 1998) and of leaf chloroplast envelopes (Blee and Joyard 1996; Feussner et al. 1995a; Heitz et al. 1997; Royo et al. 1996). Specific location may contribute to the specific function of the different LOXs in various physiological processes. Rice (Oryza sativa L.) is both a model plant and an important world staple food crop. During the storage, rice suffers the degradation of nutritional components, production of stale flavors and loss of viability, thereby reduces commercial quality and shelf life of grain. How to keep seed longevity during long term storage is a universal problem. To avoid and/or reduce the seed deterioration during storage is a vital breeding goal for rice breeders and physiologists. Many factors affect seed longevity, including environmental conditions, physiological status of the seeds, and sub-species differences (Vertucci and Roos 1990; Zhang et al. 2007). Lipid degradation was proposed as one cause of deterioration during storage (Takano 1993). To date, four LOX genes in rice seed have been isolated and characterized, including LOX-1 (DQ389164), L-2 (X64396), LOX-3 (Os03g0700400), and r9-LOX1 (AB099850) (Mizuno et al. 2003; Shirasawa et al. 2008; Wang et al. 2008). Among them, the first three are considered to correspond to the three isozymes in seed embryos (Ida et al. 1983). In order to elucidate their functions, our lab isolated these genes one by one and uncovered their role by investigating over-expression (OE) and knock-down lines. The LOX-1 product is involved in tolerance of the rice plant to wounding and brown planthopper attack (Wang et al. 2008), whereas LOX3 has a negative effect on seed longevity (Long et al. 2013), which agrees with the generation of stale flavors during storage (Shirasawa et al. 2008). Although L-2 was previously cloned from 3-day-old seedlings (Ohta et al. 1992), its function remains elusive. In this study we isolated intact OsLOX2 (corresponding to L-2) cDNA from 3-day-old seedlings of rice cultivar Nipponbare and attempted to investigate its function both in seed germination and longevity by over expression and RNA interference (RNAi) approaches.

Transgenic Res (2014) 23:643–655

Materials and methods Plant materials Japonica cultivars Wuyujing 7 and Nipponbare were used as receptors for Agrobacterium tumefaciensmediated transformation. Transgenic T1 plantlets were screened by PCR amplification of the hygromycin phosphotransferase gene (hpt) via primer pair hpt900 (Supplemental Table 1) or analysis of hygromycin resistance of the seeds. T2 seeds from positive T1 homozygotes were obtained and used for further analysis. Rice plants were grown at the experimental station of Nanjing Agriculture University under regular cultivation conditions. Both the OE/RNAi lines and their controls were grown at the same time and adjacent to each other in the same field. Seeds used for germination and aging test were harvested 35 days after flowering and then dried in an oven for 5 days at 50 C. All seeds had the same growth conditions, storage condition and were treated simultaneously. Real-time RT-PCR Real-time RT-PCR was employed to examine expression profiles of OsLOX2 in both wild type and transgenic lines. Different tissues, including roots, stems, expanded leaves and panicles of cv. Nipponbare were collected separately at 7 days after flowering for total RNA extraction. In order to identify OsLOX2 expression during seed development, immature embryos were isolated at 6, 9, 12, 15, 18 and 21 days after flowering. Following 2-day period of water soaking, embryos, roots and leaves of 4, 6, 8, 10 and 12 day-old seedlings were obtained for total RNA isolation. In OE and RNA-interference studies, embryos of seeds immersed in water for 2 days were collected and tested for OsLOX2 gene expression. After digestion with RNase-free DNase I, 1 lg of total RNA was used for reverse transcription via AMV Primescript (Promega). Real-time RT-PCR was performed in a total volume of 20 ll containing 10 ll iTaqTM Universal SYBR Green Supermix (BioRad), 1 ll first-strand cDNA reaction products, 0.6 lL gene primer sets and 0.2 ll 250 9 ROX. Genespecific primers LOX2RT-F/R for OsLOX2 (Os03g0738600) were designed in a unique-region, compared with the OsLOX paralogs, OsLOX1,

645

OsLOX3, r9-LOX1, and OsLOX5 (Os03g0179900). The house-keeping gene EF-1a was used as a control. The gene-specific primers used for amplification are described in Supplemental Table 1. Real-time RTPCR was carried out with 40 cycles of 5 s at 95 C and 34 s at 60 C in an automatic thermal cycler (Applied Biosystems: 7500 Real-Time PCR System). Each cDNA sample was assayed in triplicate. The relative amounts of OsLOX2 transcript were represented as mean 2-DCT ± SD calculated statistically (Pfaffl 2001). Cloning of intact OsLOX2 cDNA Three-day-old seedlings of Nipponbare were used for OsLOX2 cloning via RT-PCR. Based on the OsLOX2 mRNA sequence NM_001057747 in the NCBI database (http://www.ncbi.nlm.nih.gov/), we designed primer pairs LOX2-1F/MR and LOX2-MF/R, which amplified a 1,390 bp fragment covering the 50 -end of OsLOX2 cDNA and a 1,708 bp fragment covering the 30 -end. In order to produce the cDNA (2,814 bp) covering the entire coding region, a restriction site Xho I located in the 284 bp overlapping region between 50 1,390 and 30 -1,708 bp fragment was used. The fragments were firstly cloned into pMD18-T (2,692 bp, TaKaRa) and sequenced, and positive clones were designated as T-LOX2-50 and T-LOX2-30 , both of which were double digested with restriction enzymes Xho I and Hind III, respectively. A *4.0 kb fragment retrieved from T-LOX2-50 and an approximate 1.7 kb segment from T-LOX2-30 were joined via a T4 ligase (TaKaRa) reaction and transformed DH5a. Finally, an intact cDNA of OsLOX2 flanked with additional restriction sites BamH I and Hind III was cloned into pMD18-T and designated T-LOX2. Subcellular localization of OsLOX2 protein A GFP fusion construct was produced to investigate the subcellular localization of OsLOX2. Using the T-LOX2 plasmid as a template, OsLOX2 cDNA was re-amplified by primer GFP-LOX2-F/R, which has additional ends containing the restriction sites Xho I and Spe I. Due to the identical Xho I- and Spe I-termini in linear pA7-GFP double-digested by Xho I and Spe I, the PCR product was easily fused into pA7-GFP using an In-FusionTM PCR Cloning Kit (Clontech).

123

646

After transformation and sequencing, the recombinant plasmid was designated as pA7-OsLOX2-GFP. This GFP-fusion plasmid was then transfected into rice protoplasts isolated from 7 to 10-day-old rice seedlings by PEG-mediated transfection as described previously (Zhang et al. 2011). Both green fluorescent signals and spontaneous red fluorescent signals of chloroplasts were detected by confocal laser scanning microscopy (Leica TCS SP5). All fluorescence experiments were repeated at least three times. In vitro expression and enzymatic activity assays Using the T-LOX2 plasmid as template DNA, the PCR product amplified by primer LOX2-EXP-F/LOX2-R corresponded to the complete OsLOX2 cDNA. The cDNA segment without mismatch nucleotides was subsequently cloned into the expression vector pET30a(?) (Novagen) by double-digestion with BamH I and HindIII, resulting in the construct pET30a-OsLOX2. Both pET30a and pET30a-OsLOX2 were separately transformed into Escherichia coli BL21 (DE3) and a fusion protein product was induced with 0.1 mM isopropylthio-b-D-galactoside (IPTG) at 18 C overnight. The E. coli cells were collected and resuspended in lysis buffer (20 mM Tris–HCl pH 8.0, 0.5 M NaCl). After sonication on ice, the supernatants from both before and after IPTG induction samples were subjected to electrophoresis on 15 % SDS–polyacrylamide gels. The induced HIS and HIS-LOX2 fusion proteins, present in the supernatant, were purified by His•Bind Kits (Novagen) and assayed for LOX activity. Alpha-linoleic acid (LA) and linolenic acid (LeA), were purchased from Sigma-Aldrich (Shanghai). Preparation of the substrates and standard enzymatic reactions were conducted as described earlier (Axelrod 1981) with some modifications. Aliquots of 40 ll of Tween-20 and 30 ll of substrate were added to 5 ml distilled water and mixed. The mixture became transparent with several drops of 2 M NaOH, followed by adjustment to 10 ml with water, and a final concentration of 10 mmol/L of LA or LeA. Both substrates were used immediately or stored in -20 C for later study. Each reaction mixture contained 10 ll purified enzyme (29.4 ng/ll), 6.6 ll preparation LA or LeA (10 mM) and 183.4 ll buffered solutions. The enzymatic activity of OsLOX2 was measured in buffered solutions at different various pH values

123

Transgenic Res (2014) 23:643–655

(3.0, 4.0, 5.0, 5.5, 6.0, 6.5, 7.0, 8.0 and 9.0) and different temperatures (20–50 C). Absorbance of the reaction mixture at 234 nm was detected automatically with a spectrophotometer (SpectraMax M3) every 20s following initiation of reactions. The value of OD234nm in the first 2 min increased linearly and the rates was calculated by (OD234nm2 min–OD234nm0 min)/2 min. Specific activity was finally represented by the HPOD products (e = 25 mM-1 cm-1) per milligram protein per minute via the formula ‘A = ebc’. The purified HIS protein was used as a control. Three replicates were performed for each reaction. Construction and transformation of overexpression and RNA interference plasmids Intact cDNA in T-LOX2 was isolated by BamH I and Hind III and subcloned into the pPZP2Ha3 (?) binary vector (Fuse et al. 2001), driven by the CaMV35S promoter. The resultant OE plasmid pPZP-LOX2 was introduced into Wuyunjing 7 by Agrobacteriummediated transformation (Hiei et al. 1994). To generate OsLOX2 gene knock-down transgenic lines, two RNAi constructs were produced: one was based on the artificial microRNA and the other was dependent on the formation of a hairpin structure. As described elsewhere (Warthmann et al. 2008), we first designed four primers on the website (http://wmd3.weigel world.org/cgi-bin/webapp.cgi), then taking plasmid PNW55 as template, three separate fragments were obtained by PCR amplification using primer sets G4368/L2-II, L2-I/L2-IV, and L2-III/G4369. Subsequent fusion PCR of the three PCR products with primers G-4368 and G-4369 resulted in one DNA fragment for cloning into the pGE vector. After sequencing, the artificial microRNA was subcloned into the pPZP2Ha3 (?) binary vector by BamH I and Hind III for transformation. In addition, a 535 bp fragment with additional restriction sites, viz. BamH I and Sac I in the 50 -end, and Sac I and Mlu I in the 30 -end, was amplified by primers RNAi-535F/R using T-LOX2 as template and then cloned into the pGE vector for sequencing. By Sac I digestion, this fragment was first cloned into pLHRNAi (Li et al. 2013) and the recombinant plasmid with a forward insertion was subsequently digested with Mlu I and BamH I for insertion of the 535 bp fragment in the reverse direction. Thus, a recombinant plasmid with two inverted

Transgenic Res (2014) 23:643–655

647

repeat PCR fragments was completed. Both the microRNAi and hairpin structure RNAi constructs were introduced into A. tumefaciens strain EHA105 and then used to infect Nipponbare callus. Plantlets regenerated from hygromycin-resistant calli were transplanted into the field. Transgenic plants were confirmed by both PCR and hygromycin resistance. Full-filled and healthy seeds from the positive homozygote T1 plant were selected for further analysis of germination rates and accelerated aging treatments.

seeds were put in oven for accelerated aging and *300 seeds per batch were taken out for germination test after 12, 15, 20, 25 and 30-day aging, respectively. A survival curve was made by the germination rate at different time points and the number of days required to decline the germination rate to 50 % (LD50) was estimated.

Germination test

LHP content was measured by FOX2 (ferrous oxidation-xylenol orange edition 2) method developed and validated by Long (unpublished data), referring to the FOX method (Bou et al. 2008). Dehulled seeds were polished by a laboratory rice polisher and the bran passing through an 80-mesh sieve was collected and immediately stored on ice for use. 0.2 g rice bran were extracted with 1 ml pre-chilled chloroform/methanol (2:1) solution containing 0.1% 2,6-Di-tert-butyl-4methylphenol (BHT) by a Mixer Mill (Retsch MM301, Germany) under a period of 1/28 s shaking for 3 9 2 min. After centrifugation at 1,5000g for 5 min, the supernatant was directly used for LHP determination. 100 ll resulting supernatant was added in 10 ll methanol and 890 ll FOX2 solution which is made up with 840 ll chloroform/methanol (3:7) solution containing 0.1 % BHT, 25 ll solution with 10 mM (NH4)2Fe(SO4)2 and 1 M H2SO4, 25 ll 4 mM xylenol orange. Another 100 ll supernatant of each sample combined with 10 ll 25 mM TPP methanol solution and 890 ll FOX2 solution was also prepared as negative control. The absorbance at 560 nm was measured after 40 min inoculation at 30 C. The absorbance caused by LHP was represented by the difference of A560 value between the reaction without and with triphenylphosphine (TTP). Concentration of LHP equivalent to that of H2O2 was calculated using a standard curve prepared with authentic H2O2 solution. Standard H2O2 solution was diluted from commercially available 30 % H2O2 stock solution with chloroform/methanol (2:1) containing 0.1 % BHT.

To test seed viability, 150 plump seeds from each individual were carefully selected and put on three layers of filter papers in three Petri dishes (50 seeds/ each) for germination test. The seeds were fully immersed in distilled water for 2 days at room temperature and then placed under conditions of 30 C, 12 h darkness/12 h light and with sufficient water for germination in a growth chamber for 7 days. The germination rate of each line was scored as the percentage of germinated seeds at 12 hourly intervals. TTC assay Seed viability was evaluated by triphenyltetrazolium chloride (TTC) reduction assay (Moore et al. 1973). 100 9 3 seeds were selected randomly and soaked in water for 12 h, cut in halves longitudinally, and then stained in 1 % TTC solution for 2 h (37 C) in the dark. Seed viability was presented by the percentage of deeply stained seeds. Accelerated aging Accelerated aging was performed as Siddique et al. (1989) indicated with some modifications. Seeds harvested 35 days after flowering and dried in an oven for 5 days at 50 C, the moisture content was brought to about 12 %. To accelerate aging of each line, a mesh bag with at least 300 healthy seeds was placed in a seed aging chamber, maintained at 40 C and approximately 80 % relative humidity for 10–30 days. Seed viability after aging was evaluated both by TTC assay and germination test. In order to determine a suitable number of aging days in formal test, seed survival curve was made in wild types (Wuyunjing 7 and Nipponbare) at first. At least 2,000

Measurement of lipid hydroperoxide (LHP) in rice seeds

Statistical analysis One-way analysis of variance (ANOVA) was used to compare the differences between the transgenic lines and the control followed by t tests. A P value of\0.01 was as adopted for statistical significance.

123

648

Transgenic Res (2014) 23:643–655

Results

for 2 days, expression of OsLOX2 was also dramatically increased to approximately 75-fold of that before immersion (Fig. 1c). After germination, the roots and leaves were divided for separate expression analyses. At the 4th day, OsLOX2 maintained high levels of expression, especially in roots, and was almost 180-fold higher than that in dry embryos (Fig. 1c). However, expression decreased from the 6th day and then kept a steady level similar to that in mature plants. Transcripts in seedling roots were more abundant than in leaves, consistent with expression levels in the reproductive growth phase. These results indicated that OsLOX2 may play an important role in embryos during seed development and germination.

OsLOX2 was highly expressed during embryo development and initial seed germination The OsLOX2 (L-2) gene was located on chromosome 3, corresponding to the putative LOX gene Os03g 0738600 (cDNA: NM_001057747), which has 4 exons and 3 introns. A total of 14 LOX genes were identified in rice (Umate 2011). Among them, OsLOX2 shared 77.39, 72.16, and 63.07 % amino acids identities with OsLOX3 (Os03g0700400), r9-LOX1 (Os03g0699700), and OsLOX1 (Os03g070 0700), respectively. In Nipponbare, OsLOX2 transcripts were detected in all tissues at 7 days after flowering (DAF) except expanded leaves (Fig. 1a). During seed formation, the developing embryos were collected every 3 days from 6 to 21 DAF. OsLOX2 was highly increased and reached a peak at 15 DAF (Fig. 1b). After imbibitions

No signal peptide was predicted in OsLOX2 via SignalP 4.1 Server (http://www.cbs.dtu.dk/services/ SignalP/). GFP alone driven by the 35S promoter was 10

-4

OsLOX2/ EF-1α (10 )

b 12 10 8 6 4 2 es

8 6 4 2 0

6

9

av

es cl

12

70 50

-4

18

21

Embryos Leaves Roots

60

OsLOX2/ EF-1α (10 )

15

Days after flowering

Le

ni Pa

em St

oo R

c

s

0 ts

-4

OsLOX2/ EF-1α (10 )

a

Subcellular localization of OsLOX2

40 30 20

6 4 2 0

0

2

4

6

8

10

12

Days after imbibition Fig. 1 Relative expression levels of OsLOX2 in various tissues and periods. a In roots, stems, panicles and leaves from Nipponbare at the reproductive growth stage (7 days after flowering). b In embryos throughout the seed development at the 6, 9, 12, 15, 18 and 21 days after flowering. c In germinating

123

embryos and new leaves and roots. Seeds were germinated after 2 days of soaking and grown for 10 days in 12 h light/12 h darkness. Data obtained by quantitative real-time RT-PCR were normalized against expression of the EF-1a mRNA

Transgenic Res (2014) 23:643–655

649

diffused in both the nucleolus and cytoplasm (Fig. 2a– d). OsLOX2-GFP fusion protein was located mainly in the cytoplasm not in the chloroplast, as indicated by the red auto-fluorescence signal of chloroplasts in response to UV (488 nm) with no overlapping with GFP signals (Fig. 2e–h). Thus, OsLOX2 acts in the cytoplasm, but not in the chloroplasts. Combined with a high sequence similarity to others LOXs, OsLOX2 belongs to the Type I LOX group. Enzymatic activity of OsLOX2 on a-linoleic acid (LA) and linolenic acid (LeA) We introduced intact OsLOX2 cDNA into pET30a (?) for in vitro expression. By SDS-PAGE, a unique band of approximately 110 kDa was observed; it consisted of OsLOX2 (about 97 kDa) and an 8 kDa 6 9 HIS tag (Supplemental Fig 1). The soluble proteins HIS and HIS-LOX2 fusion protein were obtained in the supernatant and purified by His•Bind Kits (Novagen) for enzymatic activity analysis (Supplemental Fig. 1). In different reaction systems with gradient pH values at 30 C, the HIS-LOX2 fusion protein showed the highest activity on a-linoleic acid (LA) at pH 6.0 GFP

a

and less prominently on linolenic acid (LeA) at pH 5.5 (Fig. 3a, b). This revealed that OsLOX2 can catalyze both LA and LeA within the optimal pH range of 5.5–6.0. Using LA as substrate, the optimal temperature range was from 20 to 50 C (Fig. 3c). In contrast, purified HIS protein showed negligible activity under all conditions (Fig. 3a, b, c). The dynamics of the same reaction system with sufficient a-linoleic acid (LA) and linolenic acid (LeA) were determined under the optimal conditions (pH 5.5, 30 C),. For the LA reaction, the HPOD content increased dramatically in the first 5 min and then maintained a steady level. By contrast, the product in LeA reaction changed more slowly (Fig. 3d). Interestingly, no such differences were observed for LOX3 reactions (data not shown). Overall, OsLOX2 can catalyze both LA and LeA in the range pH 5.5–6.0 within a wide range of temperatures, and has a preference for LA. The OsLOX2 gene promotes germination rates under the normal conditions We obtained two T2 transgenic lines homozygous for the exogenous OsLOX2 gene, designated ‘OE13’ and

Chloroplast

b

Bright field

c

Merged

d

10µm

e

f

g

h

10µm Fig. 2 Subcellular localization of OsLOX2 in rice protoplasts. Samples were visualized under a confocal microscope. The 35::GFP construct pA7 (a–d) and pA7-OsLOX2-GFP construct (e–h) were transiently expressed in protoplasts derived from

10-day-old rice seedlings. Images are GFP (a, e), chlorophyll autofluorescence (b, f), bright field (c, g) and the merged (d, h). Scale bar = 10 lm

123

650

Transgenic Res (2014) 23:643–655

b

LA-pET

LeA-pET

6 4 2 0

5

-1

8

2

3

4

5

6

7

8

LA-OsLOX2

14

9

d

LA-pET

3 2 1 0

2

3

40

10 8 6 4 2 20

25

30

35

40

45

50

55

Temperature (°C)

4

5

6

7

8

9

LOX2-LA pET-LA LOX2-LeA pET-LeA

-1

12

0 15

4

pH

10

HPOD (µmol mg )

-1

-1

HPOD(µmol mg min )

LeA-OsLOX2

-1

10

pH

c

6

HPOD(µmol mg min )

LA-OsLOX2

12

-1

-1

HPOD(µmol mg min )

a

30

20 10

0

0

50

100

150

200

250

300

Time (s)

Fig. 3 Determination of LOX activity of OsLOX2 using alinoleic acid (LA) and linolenic acid (LeA). The specific activities were determined spectrophotometrically by monitoring the increase of absorbance at 234 nm in the first 2 min and calculated by the formation of HPOD (e = 25 mM-1 cm-1) per milligram protein per minute in each enzyme reaction. a– b Specific activities at different pH levels were detected at 30 C in 0.1 M acetate buffers (pH 3.0–6.0), 0.1 M phosphate buffer (pH 6.0–8.0) and 0.1 M Tris–HCl buffer (pH 8.0–9.0), respectively; LA-OsLOX2 and LA-PET represent the activity

of HIS-LOX2 and HIS protein using LA as a substrate (a); LeAOsLOX2 and LeA-PET represent the activity of HIS-LOX2 and HIS protein using LeA as a substrate (b); At pH 6.0 (acetate buffer), the optimal temperature of HIS-LOX2 was assayed using LA as substrate (c); On the optimal condition (pH 5.5, 30 C), the dynamic changes of HPOD content per milligram protein were illustrated in OsLOX2 reaction systems with LA or LeA (d). Data are mean ± SD from three independent experiments

‘OE15’. Compared with the wild-type ‘Wuyunjing 7’, ‘OE13’ and ‘OE15’ had 2-fold higher expression of OsLOX2 during germination (Fig. 4a). After imbibition, the germination percentage of ‘OE13’ and ‘OE15’ rapidly increased to 20–30 % at 24 h, compared to\5 % for the wild type (Fig. 4b). At 36 h, the germination rates of the transgenic lines were 60–70 %, and nearly the same as the wild type. At 72 h after imbibition, both the wild type and transgenic lines were fully germinated. These results indicated that the OsLOX2 gene accelerates early stage germination. Ten transgenic lines in which endogenous OsLOX2 gene was knocked down by RNAi were identified (Fig. 5a). Among them, OsLOX2 gene in 4 mRNAi lines (interfered by artificial microRNA) declined to the basal level as in dry embryos, whereas 6 hRNAi

lines (interfered by hairpin structure RNA) showed OsLOX2 reduction in different degree (Fig. 5a). The OsLOX2 gene expression in ‘hRNAi20’ declined to one-third of the level of wild type (Fig. 5a). We also compared the germination ability of the wild type and its RNAi lines by assessing the germination rates at 36, 48, 60, 72 and 120 h after imbibition. In first 2 days, when 40 % of the wild-type seeds germinated, a significantly lower rate was detected in all RNAi lines (Fig. 5b). The retardation was lasted until 72 h after imbibition. All seeds eventually fully germinated. Interestingly, the more the OsLOX2 expression was suppressed, the slower the seeds germinated. Both the accelerated germination in overexpression lines and the retardation in RNAi lines demonstrated that the OsLOX2 gene plays a vital role in promoting early seed germination under normal conditions.

123

Transgenic Res (2014) 23:643–655

651

b

2.5

100

2

80

WT OE13

Germination rate (%)

Relative transcript abundance

a

1.5

1

0.5

0 WT

OE13

60

40

**

**

20

0

OE15

OE15

24

36

48

60

72

Hours after imbibition

90

d

Germination rate after 16 days aging

80 70

70

**

60 50

**

40 30

viable seeds (%)

Germination rate (%)

80

60

** 40 30 20

10

10

0

0

WT

OE13

OE15

**

50

20

e

Aging for 2 years in natural condition

0.45

µmol Equiv H2O2.g-1 bran

c

LHP content

**

0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05

WT

OE13

OE15

0

WT

OE15

Fig. 4 Comparison of OsLOX2 OE lines with wild-type WuYunJing 7. a Relative OsLOX2 gene expression in the embryo after 2 days imbibitions in WuYunJing 7 and homozygous transgenic lines ‘OE13’and ‘OE15’. Data obtained by quantitative real-time RT-PCR were normalized against expression of the EF-1a mRNA. b Germination rates under normal conditions were calculated at 24 h after imbibition and

subsequently every 12 h. c Germination rates after 16 days accelerated aging d Percentage of vigorous seeds after 16 days accelerated aging was evaluated by TTC reduction assay d lipid hydroperoxide (LHP) content in seeds of OE line 15 aged for 2 years in natural condition. ‘‘**’’ Significant differences by t tests at P \ 0.01 compared with the control. Values are mean ± SD of three replicates

Seed viability was dramatically affected by aging treatment in both over-expression (OE) and RNA interference (RNAi) lines

Fig 2). Based on the seed survival curve, the LD50 of Wuyunjing 7 and Nipponbare were about 16 days and 25 days, respectively. Therefore, we investigated the seed viability of ‘OE13’ and ‘OE15’ together with Wuyunjing 7 after 16 days of aging. Compared with the wild type, the OE lines had significantly decreased seed germination levels (Fig. 4c), a similar result was observed by TTC assay with seeds after aging for 2 years in natural condition (Fig. 4d). These results indicated that OsLOX2 gene promotes aging process and reduces seed viability in both accelerated aging and natural aging conditions. Moreover, an elevated

In order to investigate the influence of the OsLOX2 gene on seed storage, we applied accelerated aging to both the transgenic lines and wild types. After the accelerated aging for 0, 12, 15, 20, 25 and 30 days, wild type Wuyunjing 7 and Nipponbare, showed dramatically reduced levels of germination rate, indicating that aging treatment damages seed viability and hence affects germination rates (Supplemental

123

652

Transgenic Res (2014) 23:643–655

OsLOX2/EF-1α(10-3)

a

14.0

Relative expression in RNAi lines

12.0 10.0 8.0 6.0 4.0 2.0

b

100

5

6

Ai R m

m

R

N

N

Ai N R m

m

Ai

7

2 N R

N

Ai

Ai hR

N

Ai

16

7

9 hR

hR

N

N

Ai

Ai

23

2 Ai N

WT hRNAi20 hRNAi25 mRNAi2 mRNAi7 mRNAi6 mRNAi5

80

Germination rate (%)

hR

hR

hR

N

Ai

W

20

T

0.0

60

40 20 0

48

60

72

120

Hours after imbibition

d

Germination rate after 25 days aging

** **

Accelerated aging for 25 days 80

**

**

viable seeds (%)

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

**

**

70

µmol Equiv H2O2 ·g-1 bran

e

50 40 30 20

0.4

LHP content in seeds aging for 2 years in natural condition

0.3 0.2 0.1

WT

hRNAi20

hRNAi23

mRNAi2

mRNAi5

mRNAi7

i2 NA hR

i16 hR NA

NA

i7

i9 hR

NA

i20

hR

NA hR

WT

mR NA i5

NA i7 mR

hR NA i16

NA i7 hR

hR NA i9

0

0

123

**

**

**

60

10 WT hR NA i20

Germination rate

c

Transgenic Res (2014) 23:643–655 b Fig. 5 Comparison of OsLOX2 knockdown lines with wild-

type Nipponbare. a Relative OsLOX2 gene expression in the embryo after 2 days imbibitions in Nipponbare and homozygous knock-down lines (hRNAi produced by hairpin RNAi and mRNAi produced by artificial microRNA). Data obtained by quantitative real-time RT-PCR were normalized against expression of the EF-1a mRNA. b Germination rates under normal conditions were calculated at 48, 60, 72 and 120 h after imbibition. c Germination rates after 25 days accelerated aging. d Percentage of vigorous seeds after 25 days accelerated aging was evaluated by TTC reduction assay (d) lipid hydroperoxide (LHP) content in seeds of knockdown lines aged for 2 years in natural condition. ‘‘**’’ indicate significant differences by t tests at P \ 0.01 compared with the control. Values are mean ± SD of three replicates

LHP content was observed in OE15 after 2-year natural aging (Fig. 4e). The reduced seed longevity in OE lines may result from high expression of OsLOX2 gene and increased lipid peroxidation. Under normal conditions, RNAi seeds germinated more slowly than the wild type. Nevertheless, they were still vigorous and capable of complete germination (Fig. 5b). Four hRNAi lines exerted higher germination rate than the wild-type after 25 days accelerated aging (Fig. 5c), supported by the increased percentage of viable seeds in TTC assay (Fig. 5d). Interestingly, lower LHP content was found in ‘hRNAi20’ and ‘hRNAi23’ (Fig. 5e). A probable correlation between OsLOX2 gene expression, LHP content and seed longevity becomes more apparent. However, there are still some debates in mRNAi lines. When the mRNAi lines ‘mRNAi7’ and ‘mRNAi5’ were treated with 25 days accelerated aging, reduced seed viability was detected (Fig. 5c), though their LHP content was much lower than the wild-type (Fig. 5e). Except for the results from mRNAi lines, we concluded that the more OsLOX2 gene expressed, the much LHP was generated under natural storage condition, therefore the seed longevity was decreased; and vice versa. The puzzling behaviour of mRNAi lines will be discussed later.

Discussion The OsLOX2 gene is essential for rice seed germination Seed germination is characterized by the mobilization of storage lipids as a carbon source for the seedling (Gerhardt 1992). During seed maturation, storage

653

lipids are deposited in lipid bodies, which are intracellular droplets of triacylglycerols surrounded by a phospholipids monolayer (Huang 1996). A set of proteins detectable in the phospholipid monolayer participate in the degradation of lipid bodies during the early stages of germination, including LOX proteins (Kindl 1997). In cucumber, the specific linoleate 13-LOX is not expressed in resting seeds, but is dramatically induced during the early stages of germination (Feussner et al. 1996; Ho¨hne et al. 1996). By immunocytochemical studies, a specific localization of this enzyme in the membrane of lipid bodies was detected during the early stages of germination (Feussner et al. 1996). Unsaturated storage lipids are oxygenated by LOX to their corresponding hydroperoxy derivatives. Among them, 13-HPOD is mainly detected, not only in cucumber, but also in sunflower, marigold and flax seedlings (Feussner et al. 1995b, 1997a). In the storage lipid fraction, the amount of LA residues falls from the first day of germination and the amount of the corresponding 13-HPOD transiently increases, paralleling by the occurrence of LOX protein at the lipid body membrane. A model for the breakdown of storage lipids by LOX protein during germination of oilseeds was proposed (Feussner et al. 1997b). L-2 in rice seed, previously identified as a LOX, corresponds to the isozymes LOX-2, but its physiological function remained unknown (Ohta et al. 1992). Consistent with the highest enzyme activity of LOX-2 among the three LOX proteins in seed during germination (Wang et al. 2008), our data showed that OsLOX2 was strongly induced during the initial stages of germination. Moreover, recombinant HIS-LOX2 protein exhibited substrate specificity for LA rather than linolenic acid. When OsLOX2 expression was enhanced in transgenic lines, a faster germination rate was observed. In contrast, germination was retarded with decreased OsLOX2. Taken together, OsLOX2 accelerates initiation of seed germination most probably by promoting oxygenation of unsaturated storage lipids, especially LA. OsLOX2 affects seed deterioration during storage Deterioration during storage negatively affects rice seed viability. Study of seed longevity under conventional storage conditions is a time consuming task. Accelerated aging has been widely utilized to speed

123

654

the process, such as in Arabidopsis, cucumber, onion and soybean, as well as rice (Salama and Pearce 1993; Shin et al. 2009; Stewart and Bewley 1980). The question is what changes during storage and why does it affect seed viability? Oxidative stress, lipid peroxidation and degradation, and respiration have been proposed as causes of the seed aging (McDonald 1999; Shin et al. 2009; Wilson and McDonald 1986; Zou et al. 2002). Wang et al. (2012) carried out lipid profiling analyses of rice seeds during accelerated aging and found that the levels of main membrane lipids phosphatidylcholine (PC), phosphatidylethanolamine (PE) and phosphatidylglycerol (PG) declined, whereas phosphatidic acid (PA), a hydrolysis product of phospholipase D (PLD) increased. These results suggested that PLD-mediated lipid hydrolysis is responsible for seed deterioration in rice. Furthermore, OsLOX2 (Loc_Os08g39850) was induced significantly during seed aging (Wang et al. 2012). Zhang et al. (2007) compared rice varieties with or without LOX-1, 2 and found that germination rates without LOX-1, 2 changed slowly after an accelerated aging experiment, whereas those with LOX-1, 2 changed quickly, suggesting that lipid peroxidation also plays critical roles during seed longevity under storage condition. To investigate whether OsLOX2 affects seed viability during storage, we investigated the germination rate after aging treatments. Compared with the wild type, OsLOX2 overexpression lines had more LHP content and the percentage of vigorous seeds was significantly decreased; meanwhile, hRNAi lines with reduced OsLOX2 gene expression had less LHP content and increased the seed longevity, indicating that OsLOX2 may hasten aging process during storage. However, in mRNAi lines, not only the germination rate before aging, but also the seed viability after aging was dropped. These results may caused by extremely low OsLOX2 expression level. It has been reported that distinct LOX proteins, including LOX-1, LOX-2 and LOX-3 were characterized in mature rice seeds (Ida et al. 1983). LOX protein may accumulate in seeds and prepare for initiation of germination. Although the OsLOX2 gene was down-regulated in mRNAi lines, the seeds eventually germinated before the aging treatment. However, a trace amount of OsLOX2 protein may store in seeds and it may degrade during the accelerated aging condition, and then affect the initiation of germination after aging. Since it is difficult to measure the OsLOX2 activity in crude extraction, an improved

123

Transgenic Res (2014) 23:643–655

method is needed. Future studies will need to address this question. Anyway, appropriate down regulation of the OsLOX2 gene by hRNAi can enhance the seed longevity and be applied in safe storage of rice. Acknowledgments We wish to thank Prof. Mingliang Xu for critical review of the draft manuscript. We also gratefully acknowledge Xi Liu for excellent field work. This project was financially supported by the National Transformation Science and Technology Program (2013ZX08001-006).

References Axelrod B (1981) Lipoxygenase from soybeans. Methods enzymol 71:441–451 Beckman K, Ingram D (1994) The inhibition of the hypersensitive response of potato-tuber tissues by cytokinins-similarities between senescence and plant defense responses. Physiol Mol Plant Pathol 44:33–50 Blee E (2002) Impact of phyto-oxylipins in plant defense. Trends Plant Sci 7:315–321 Blee E, Joyard J (1996) Envelope membranes from spinach chloroplasts are a site of metabolism of fatty acid hydroperoxides. Plant Physiol 110:445–454 Bou R, Codony R, Tres A, Decker EA, Guardiola F (2008) Determination of hydroperoxides in foods and biological samples by the ferrous oxidation–xylenol orange method: a review of the factors that influence the method’s performance. Anal Biochem 377:1–15 Brash A (1999) Lipoxygenases: occurrence, functions, catalysis, and acquisition of substrate. J Biol Chem 274:23679–23682 Feussner I, Kindl H (1994) Particulate and soluble lipoxygenase isoenzymes-comparison of molecular and enzymatic properties. Planta 194:22–28 Feussner I, Wasternack C (2002) The lipoxygenase pathway. Annu Rev Plant Biol 53:275–297 Feussner I, Hause B, Voros K, Parthier B, Wasternack C (1995a) Jasmonate-induced lipoxygenase forms are localized in chloroplasts of barley leaves (Hordeum-vulgare cv salome). Plant J 7:949–957 Feussner I, Wasternack C, Kindl H, Ku¨hn H (1995b) Lipoxygenase-catalyzed oxygenation of storage lipids is implicated in lipid mobilization during germination. Proc Natl Acad Sci 92:11849–11853 Feussner I, Hause B, Nellen A, Wasternack C, Kindl H (1996) Lipid-body lipoxygenase is expressed in cotyledons during germination prior to other lipoxygenase forms. Planta 198:288–293 Feussner I, Balkenhohl T, Porzel A, Kuhn H, Wasternack C (1997a) Structural elucidation of oxygenated storage lipids in cucumber cotyledons—implication of lipid body lipoxygenase in lipid mobilization during germination. J Biol Chem 272:21635–21641 Feussner I, Ku¨hn H, Wasternack C (1997b) Do specific linoleate 13-lipoxygenases initiate b-oxidation? FEBS Lett 406:1–5 Feussner I, Kuhn H, Wasternack C (2001) Lipoxygenasedependent degradation of storage lipids. Trends Plant Sci 6:268–273

Transgenic Res (2014) 23:643–655 Fuse T, Sasaki T, Yano M (2001) Ti-plasmid vectors useful for functional analysis of rice genes. Plant Biotechnol 18:219–222 Gardner H (1995) Biological roles and biochemistry of the lipoxygenase pathway. HortScience 30:197–205 Gerhardt B (1992) Fatty acid degradation in plants. Prog Lipid Res 31:417–446 Grayburn W, Schneider G, Hamiltonkemp T, Bookjans G, Ali K, HildebrandI D (1991) Soybean leaves contain multiple lipoxygenases. Plant Physiol 95:1214–1218 Heitz T, Bergey D, Ryan C (1997) A gene encoding a chloroplast-targeted lipoxygenase in tomato leaves is transiently induced by wounding, systemin, and methyl jasmonate. Plant Physiol 114:1085–1093 Hiei Y, Ohta S, Komari T, Kumashiro T (1994) Efficient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T-DNA. Plant J 6:271–282 Ho¨hne M, Nellen A, Schwennesen K, Kindl H (1996) Lipid body lipoxygenase characterized by protein fragmentation, cDNA sequence and very early expression of the enzyme during germination of cucumber seeds. Eur J Biochem 241:6–11 Huang A (1996) Oleosins and oil bodies in seeds and other organs. Plant Physiol 110:1055 Ida S, Masaki Y, Morita Y (1983) The isolation of multiple forms and product specificity of rice lipoxygenase. Agric Biol Chem 47:637–641 Kindl H (1997) The oxygen-dependent modification of triacylglycerols and phospholipids, the different way of initiating lipid body mobilization. Zeitschrift fu¨r Naturforschung C A journal of biosciences 52:1–8 Kuhn H, Thiele B (1999) The diversity of the lipoxygenase family—many sequence data but little information on biological significance. FEBS Lett 449:7–11 Li H, Jiang L, Youn JH, Sun W, Cheng Z, Jin T, Ma X, Guo X, Wang J, Zhang X, Wu F, Wu C, Kim SK, Wan J (2013) A comprehensive genetic study reveals a crucial role of CYP90D2/D2 in regulating plant architecture in rice (Oryza sativa). New Phytol 200:1076–1088 Long Q, Zhang W, Wang P, Shen W, Zhou T, Liu N, Wang R, Jiang L, Huang J, Wang Y (2013) Molecular genetic characterization of rice seed lipoxygenase 3 and assessment of its effects on seed longevity. J Plant Biol 56:232–242 McDonald M (1999) Seed deterioration: physiology, repair and assessment. Seed Sci Technol 27:177–237 Mizuno K, Iida T, Takano A, Yokoyama M, Fujimura T (2003) A new 9-lipoxygenase cDNA from developing rice seeds. Plant Cell Physiol 44:1168–1175 Moore R (1973) Tetrazolium staining for assessing seed quality. Seed Ecol 11–23 Nellen A, Rojahn B, Kindl H (1995) Lipoxygenase forms located at the plant plasma-membrane. Zeitschrift Fur Naturforschung C A Journal of Biosciences 50:29–36 Ohta H, Shirano Y, Tanaka K, Morita Y, Shibata D (1992) cDNA Cloning of rice lipoxygenase L-2 and characterization using an active enzyme expressed from the cDNA in Escherichia-coli. Eur J Biochem 206:331–336 Oliw E (2002) Plant and fungal lipoxygenases. Prostaglandins Other Lipid Mediators 68–9:313–323 Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Res 29:e45–e45

655 Rosahl S (1996) Lipoxygenases in plants—their role in development and stress response. Zeitschrift Fur Naturforschung C A Journal of Biosciences 51:123–138 Royo J, Vancanneyt G, Perez A, Sanz C, Stormann K, Rosahl S, SanchezSerrano J (1996) Characterization of three potato lipoxygenases with distinct enzymatic activities and different organ-specific and wound-regulated expression patterns. J Biol Chem 271:21012–21019 Salama AM, Pearce RS (1993) Ageing of cucumber and onion seeds: phospholipase D, lipoxygenase activity and changes in phospholipid content. J Exp Bot 44:1253–1265 Shibata D, Slusarenko A, Casey R, Hildebrand D, Bell E (1994) Lipoxygenases. Plant Mol Biol Report 12:S41–S42 Shin J-H, Kim S-R, An G (2009) Rice aldehyde dehydrogenase7 is needed for seed maturation and viability. Plant Physiol 149:905–915 Shirasawa K, Takeuchi Y, Ebitani T, Suzuki Y (2008) Identification of gene for rice (Oryza sativa) seed lipoxygenase-3 involved in the generation of stale flavor and development of SNP markers for lipoxygenase-3 deficiency. Breed Sci 58:169–176 Siddique SB, Seshu D, Pardee WD (1988) Rice cultivars variability in tolerance for accelerated aging of seed [Philippines]. IRRI research paper series Siedow J (1991) Plant lipoxygenase-structure and function. Annu Rev Plant Physiol Plant Mol Biol 42:145–188 Stephenson L, Bunker T, Dubbs W, Grimes H (1998) Specific soybean lipoxygenases localize to discrete subcellular compartments and their mRNAs are differentially regulated by source-sink status. Plant Physiol 116:923–933 Stewart RR, Bewley JD (1980) Lipid peroxidation associated with accelerated aging of soybean axes. Plant Physiol 65:245–248 Takano K (1993) Advances in cereal chemistry and technology in Japan. Cereal Foods World 38:695–698 Umate P (2011) Genome-wide analysis of lipoxygenase gene family in Arabidopsis and rice. Plant Signal Behav 6:335–338 Vertucci CW, Roos EE (1990) Theoretical basis of protocols for seed storage. Plant Physiol 94:1019–1023 Wang R, Shen W, Liu L, Jiang L, Liu Y, Su N, Wan J (2008) A novel lipoxygenase gene from developing rice seeds confers dual position specificity and responds to wounding and insect attack. Plant Mol Biol 66:401–414 Wang F, Wang R, Jing W, Zhang W (2012) Quantitative dissection of lipid degradation in rice seeds during accelerated aging. Plant Growth Regul 66:49–58 Warthmann N, Chen H, Ossowski S, Weigel D, Herve´ P (2008) Highly specific gene silencing by artificial miRNAs in rice. PLoS ONE 3:e1829 Wilson D Jr, McDonald M Jr (1986) The lipid peroxidation model of seed ageing. Seed Sci Technol 14:269–300 Zhang Y, Yu Z, Lu Y, Wang Y, She D, Song M, Wu Y (2007) Effect of the absence of lipoxygenase isoenzymes on the storage characteristics of rice grains. J Stored Prod Res 43:87–91 Zhang Y, Su J, Duan S, Ao Y, Dai J, Liu J, Wang P, Li Y, Liu B, Feng D, Wang J, Wang H (2011) A highly efficient rice green tissue protoplast system for transient gene expression and studying light/chloroplast-related processes. Plant Methods 7:30 Zou Y, Kim AR, Kim JE, Choi JS, Chung HY (2002) Peroxynitrite scavenging activity of sinapic acid (3, 5-dimethoxy-4-hydroxycinnamic acid) isolated from Brassica juncea. J Agric Food Chem 50:5884–5890

123