Overexpression of BnKCS1-1,BnKCS1-2,and BnCER1-2 promotes cuticular wax production and increases drought tolerance in Brassica napus

Yanmei Wang, Shurong Jin, Yi Xu, Shuai Li, Shuangjuan Zhang, Zheng Yuan,Jiana Li, Yu Ni*

College of Agronomy and Biotechnology,Academy of Agricultural Sciences, Southwest University,Chongqing 400715,China

Keywords:Brassica napus Cuticular wax Drought tolerance Overexpression Transformation

ABSTRACT Higher amounts of cuticular wax in plants have been associated with improved plant stress tolerance and increased potential for industrial use. In this study, orthologs of KCS1 and CER1 in Arabidopsis, designated BnKCS1-1, BnKCS1-2, and BnCER1-2, were isolated from Brassica napus.Transcription of BnKCS1-1 and BnKCS1-2 in B.napus were induced by abscisic acid (ABA) and drought treatment, while transcription of BnCER1-2 was induced only by drought treatment. All three gene transcripts decreased significantly when plants were treated with methyl jasmonate (MeJA) or subjected to cold stress. Overexpression of BnKCS1-1, BnKCS1-2, and BnCER1-2 under the control of the CaMV35S promoter led to a significant increase in cuticular wax on transgenic B. napus leaves. BnKCS1-1 and BnKCS1-2 overexpression led to similar differences from non-transformed plants, with significantly higher levels of aldehydes (C29 and C30), alkanes (C28, C29, and C31) and secondary alcohols(C28 and C29),and a significantly lower level of C29 ketone.Overexpression of BnCER1-2 led to an increase in alkanes(C27,C28,C29,and C31),a decrease in secondary alcohols(C28 and C29),and insignificant changes in other wax components. Scanning electron microscopy revealed that overexpression of BnKCS1-1, BnKCS1-2, and BnCER1-2 in B. napus resulted in a higher density of wax crystals on the leaf surface than observed in non-transformed plants. Transgenic plants showed a reduced rate of water loss and increased drought tolerance compared to non-transformed plants. These results suggest that BnKCS1-1,BnKCS1-2, and BnCER1-2 gene products can modify the cuticular wax of B. napus. Changing cuticular waxes using transgenic approaches is a new strategy for genetic improvement of plant drought tolerance and provides an opportunity for development of B. napus as a surface-wax crop.

1. Introduction

Plant surface wax forms a hydrophobic layer covering aerial plant organs. Wax is deposited outside of the cuticle as epicuticular wax or within the cuticular matrix as intracuticular wax. These cuticular waxes are composed mainly of complex mixtures of very-long-chain fatty acids (VLCFAs) and their derivatives,including aldehydes, alkanes, secondary alcohols, ketones,primary alcohols, and wax esters [1-3]. Wax compounds are synthesized from C16and C18long-chain fatty acids by multiple elongases including 3-ketoacyl-CoA synthase (KCS), 3-ketoacyl reductase (KCR), enoyl-CoA reductase, and 3-hydroxyacyl-CoA dehydratase. The VLCFAs in the epidermal cells are then converted in the decarbonylation pathway to aldehydes,alkanes,secondary alcohols, and ketones and in the acyl reduction pathway to primary alcohols, which are further esterified with free fatty acids to give alkyl esters[2-4].

Plant cuticular waxes serve as a protective barrier against water loss, UV light, pathogens, and insects [4]. Recent studies have used genetic approaches to modify cuticular waxes to improve drought tolerance in plants. For example, ectopic expression of wax-associated genes and transcription factors in transgenic plants can increase wax deposition and confer increased tolerance to water deficiency stress in some species[5-9].

Besides their ecophysiological functions, plant cuticular waxes are also valuable raw materials for a variety of industrial applications such as lubricants, adhesives, coatings, sealants and impregnation materials [10]. However, to date,cuticular waxes have been commercially harvested from only a small number of plant species,such as carnauba palm,candelilla,and jojoba,and wax esters have been the main wax compounds used for industrial applications [10,11]. Owing to the variability of wax mixtures and poor agronomic properties of these plant species,their potential as wax-producing plants is limited. Thus, using genetic manipulation to develop crop species for surface wax production could be an alternative strategy for producing plant wax for industrial use.Currently,sugar cane is the only crop species from which cuticular wax is harvested as a byproduct, as wax is easily extracted from the press cake from sugar production[12].

Brassica napus L. is a globally important crop. Its seed oils are used for food and industrial products, such as biofuels,lubricants, hydraulic fluids, and plastics [13,14]. However,drought severely reduces its yields and planting distribution in many countries[15-17].Improving the stress tolerance of B.napus and widening its adaptive range would allow increased cultivation and stabilization of the oilseed supply.Tassone et al. [18] quantified the levels of 24 leaf cuticular wax chemical constituents in a diverse panel of 517 accessions representing B. napus seed stock center collections worldwide, and heritability analysis suggested that the phenotypic variation in wax traits was influenced primarily by genetic effects.This finding suggests that modifying leaf wax of B. napus by genetic regulation for breeding is feasible. Such modification could both improve the plant's stress tolerance and increase the potential of B.napus as a wax-producing industrial crop.

Great advances have been made in isolating and identifying the genes involved in cuticular wax biosynthesis,particularly in the model plant Arabidopsis thaliana [4]. The KCS gene family encodes 3-ketoacyl CoA synthase,controlling the amount and identity of VLCFAs produced in specific cell types and tissues. The KCS gene family consists of multiple homologs in plant species,including 21 paralogs in A.thaliana[19]. KCS1, KCS2/DAISY, KCS5/CER60, KCS6/CER6/CUT1, KCS9,and KCS20 are known to be involved in the synthesis of VLCFAs for cuticular waxes [20-25]. In particular, KCS1 was shown to have broad substrate specificity for saturated and mono-unsaturated C16-C24acyl-CoAs when expressed in yeast [26,27]. A series of ECERIFERUM (CER) genes such as CER1, CER2, CER3, and CER4 are involved in the final steps of wax monomer biosynthesis in Arabidopsis [28-30]. CER1 has been reported to control VLC alkane formation and play a key role in the alkane-forming pathway[31].

In the present study, we isolated B. napus genes orthologous to KCS1 and CER1, named BnKCS1-1, BnKCS1-2,and BnCER1-2, and characterized their expression patterns.We investigated the effects of these genes on cuticular wax components and structure by overexpressing them in B.napus and characterizing the effects of cuticular wax modifications on plant growth and development and on plant responses to soil water deficit. This study lays a foundation for future efforts to modify leaf waxes to increase the environmental stress tolerance of B.napus and to produce waxes with optimal chemical compositions for industrial use.

2. Materials and methods

2.1. Plant materials and growth conditions

Two B. napus cultivars with different wax traits,Zhongshuang11 (ZS11) and NoWAX, were used for gene expression analysis. ZS11 was further used for gene cloning and genetic transformation of B. napus. Seeds were germinated on wet filter paper in Petri dishes. Seven days after germination, plants to be subjected to hormone, NaCl, and cold treatments were transferred into black glass pots(5.5 cm × 8.0 cm) containing 1/10 MS solution, while plants to be subjected to drought treatments and transgenic plants were transferred into pots (6.5 cm × 8.0 cm) with 0.25 kg soil(soil:vermiculite 3:1). The plants were grown in a greenhouse at 22 °C/20 °C, with a 16 h light/8 h dark photoperiod and relative humidity of approximately 75%. The pots with soil were watered every three days and the solutions of the plants grown in 1/10 MS medium were changed every seven days.

Six-week-old plants were subjected to hormone and stress treatments. For hormone treatments, plants grown in MS liquid cultures were treated with 20 μmol L-1abscisic acid(ABA), 200 μmol L-1salicylic acid (SA), 200 μmol L-11-aminocyclopropane-1-carboxylic acid (ACC), or 100 μmol L-1methyl jasmonate (MeJA), and incubated for 6 h. For salt treatments, plants were submerged for 6 h in MS liquid cultures containing 100 mmol L-1NaCl. For cold treatments,plants in 1/10 MS media were exposed to 4 °C for 24 h. Plants growing in 1/10 MS solution were used as controls for hormone, NaCl, and cold treatments. For drought treatment,plants were subject to drought by halting watering for seven days, while plants grown in soil under well-watered conditions were treated as controls. All treatments were replicated three times. For RNA extraction, leaves were sampled directly after treatments. Leaves from six-week-old transgenic plants as well as non-transgenic plants were sampled to analyze gene transcript levels, water loss rates,and wax structure and composition.

2.2. Plasmid construction and plant transformation

Total RNA was isolated from B. napus (ZS11) leaves using total RNA extraction kit(TransGen,Beijing,China),reverse transcribed with a PrimeScript RT reagent Kit with gDNA Eraser (Takara,Beijing,China).With the cDNA as template,the full-length coding regions of BnKCS1-1,BnKCS1-2,and BnCER1 were amplified using gene-specific primers. The restriction enzyme sites BamH I and Xma I were added to the forward primers of BnKCS1 and BnCER1,respectively, and a Sac I site was added to the reverse primers.The PCR amplification conditions were as follows: 94 °C for 2 min, 30 cycles of 94 °C for 30 s, 57 °C for 30 s, and 72 °C for 2 min,and then 72 °C for 10 min.PCR products were cloned into pMD19-T vector(Takara,Beijing,China)for sequencing.

Verified by sequencing,the coding sequences of BnKCS1-1,BnKCS1-2, and BnCER1-2 were inserted into the binary vector pCAMBIA2301G under the control of the CaMV 35S promoter from pMD19-T vector using restriction sites BamH I/Sac I and Xma I/Sac I, respectively. The recombinant plasmids, named pC2301-BnKCS1-1, pC2301-BnKCS1-2, and pC2301-BnCER1-2,were transferred into Agrobacterium tumefaciens strain LBA4404 by freeze-thaw and transformed into B. napus ZS11 using the hypocotyl segment transformation method described by Cardoza et al. [32]. Transformed overexpression lines were selected on the basis of their resistance to kanamycin and also identified by PCR using primers apHA2F2/apHA2R2. T1 seedlings were checked for the presence of the transgene by PCR.The segregation ratio (3:1) suggested the integration of the transgene at a single locus. Homozygous lines were then obtained by screening seedlings germinated from seeds of T1 plants with β-glucuronidase (GUS). The expression of the transcript for the transgene was confirmed in T1, T2, and T3 plants. Seeds from positively identified homozygous lines were used to generate plants for all subsequent experiments.Primers used in the above experiments are listed in Table S1.

2.3. Gene expression analysis

Total RNA was isolated from roots, stems, leaves, flowers and siliques of ZS11, leaves of hormone, drought, cold and salt-treated ZS11,leaves of NoWax,and leaves of transgenic B.napus. cDNA was synthesized as described above. The quantitative RT-PCR reaction was performed on a Bio-Rad CFX96 Real-Time PCR Detection System using the SYBR Premix Ex Taq II (Takara, Beijing, China). qRT-PCR amplification was performed as follows: 95 °C for 30 s, 45 cycles of 95 °C 10 s, 54 °C 30 s for BnKCS1-1 and BnKCS1-2 or 57 °C 30 s for BnCER1-2, and 65 °C for 5 s. The reference gene actin7 was used to normalize the total RNA amount. Relative quantities of gene expression for sample comparison were calculated using the comparative Ct(2-ΔΔCt)method[33].All experiments were repeated using three biological and three technical replicates.Primers used for qRT-PCR are listed in Table S1.

2.4. Cuticular wax analysis

The third leaf from the top was used to extract cuticular waxes.Before wax extraction, leaves were photographed for pixel counting using ImageJ [34] to determine surface areas. Each sample was extracted twice in CHCl3containing 5 μg tetracosane (Sigma Aldrich, Missouri, USA) as an internal standard for 30 s each time. The two extracts were combined and filtered through glass wool,dried under a nitrogen stream,and derivatized with 50 μL BSTFA (N,O-bis(trimethylsilyl)trifluoroacetamide)and 50 μL pyridine for 45 min at 70 °C.The derivatized extract was redissolved in 200 μL chloroform for qualitative and quantitative wax analysis.

GC analysis was performed with a 9790II gas chromatograph(Zhejiang Fuli Analytic Instruments Co.,China).The GC column was a DM-5 30 m × 0.32 mm × 0.25 μm capillary column(Dikma Technologies Inc., USA). For qualitative analysis, 2 μL of the derivatized wax mixture was injected on a GCMS-QP2010(Shimadzu, Japan) using a HP-5MS 30 m × 0.32 mm × 0.25 μm capillary column(Agilent Technologies,USA)and helium as the carrier gas. The injector and flame ionization detector (FID)temperatures were set at 300 °C and 320 °C, respectively. The oven temperature of the GC was programmed with a starting temperature of 80 °C and increased at 15 °C min-1to 260 °C,where the temperature remained 10 min.The temperature was then increased at 2 °C min-1to 290 °C, and further increased at 5 °C min-1to 320 °C, where it was held for 10 min. Compound identification was based on mass spectra.Total wax amount per unit leaf area was expressed as μg cm-2.Values were the means of four biological replicates(x ± SE).

2.5. Scanning electron microscopy

The second leaf from the top of 6-week-old non-transformed plants and transgenic plants were used for scanning electron microscopy (SEM) analyses. Samples were rinsed in distilled water and dried at room temperature in filter paper. Leaf segments were mounted onto stubs, sputter-coated with gold particles with a Polaron SC-500(Quorum Technologies,Laughton,UK),and viewed with a Quanta 200(FEI,Netherland)microscope.

2.6. Measurement of leaf water loss

For quantifying leaf water loss,plants were dark-acclimated for 6 h prior to measurement.Leaves from non-transformed plants and transgenic plants were excised and soaked in water for 60 min in the dark.The leaves were gently blotted dry to remove excess water,weighed every 30 min at room temperature,and finally dried in a 70 °C oven overnight,until the dry weight was constant. Total water content was calculated as the initial water-saturated weight minus the dry weight after the heat treatment. The water loss at each interval was expressed as a percentage of the water loss relative to the total water content.The values were means of five replicates(x ± SE).

2.7. Drought tolerance analysis of plants

Fig.1-Tissue-specific expression of BnKCS1-1,BnKCS1-2,and BnCER1-2.The expression levels were measured by quantitative PCR.The results are the means of three biological replicates,each including three technical replicates.

To assess drought tolerance of the transgenic rapeseed lines,6-week-old non-transformed plants and transgenic plants were deprived of water for 11 days and then re-watered once.Plants were observed and photographed seven days after deprivation of water and two days after re-watering.The wellwatered non-transformed plants and transgenic plants growing in the greenhouse were treated as controls.

3. Results

3.1. Expression profiling of BnKCS1-1, BnKCS1-2, and BnCER1-2 in B. napus

Using Arabidopsis KCS1 and CER1 genes as probes,two KCS1 and two CER1 genes, here named respectively BnKCS1-1,BnKCS1-2, BnCER1-1, and BnCER1-2, were cloned from B. napus.Sequence comparison showed that BnKCS1-1 and BnKCS1-2 shared 97% sequence identity at the nucleotide level. Ten amino acid differences between the two genes at the amino acid level were observed. Sequence analysis showed that BnCER1-1 was a pseudogene.Accordingly,BnKCS1-1(KY744241),BnKCS1-2(KY744242), and BnCER1-2 (KT795330) were chosen for subsequent analysis. An FAE1-CUT1-RppA conserved domain was found at A114-A403of BnKCS1-1 and BnKCS1-2,and a WAX2 Cterminal domain(pfam12076)was found at L453-P615of BnCER1-2(Figs.S1A,S2A).Phylogenetic analysis showed that these three genes fall into the dicotyledon group and share a clade with Arabidopsis KCS1 and CER1,respectively(Figs.S1B,S2B).

Fig.2-Expression of BnKCS1-1,BnKCS1-2,and BnCER1-2 in B.napus with different wax traits.(A)Morphological characteristics of rapeseed cultivar ZS11(high coverage) and NoWax(low coverage).(B) Epicuticular wax crystallization patterns on leaf surfaces of ZS11 and NoWax visualized using scanning electron microscopy.Bar = 10 μm.(C)Cuticular wax coverage and compositions in ZS11 and NoWax leaves.Means of four biological triplicates were compared by Student's t-test(*P ＜0.01).(D)qRT-PCR analyses of BnKCS1-1,BnKCS1-2,and BnCER1-2 in leaves from ZS11 and NoWax.Means of three biological triplicates,each with three technical replicates, were compared by Student's t-test(*P ＜0.05).

To investigate the tissue distribution and specificity of BnKCS1-1,BnKCS1-2,and BnCER1-2 in B.napus,their expression in some major organs was characterized by qRT-PCR. Total RNAs were extracted from roots, stems, leaves, flowers and siliques of ZS11. BnKCS1-1, BnKCS1-2, and BnCER1-2 transcripts were detected in all tested organs with different expression levels.The highest expression of each of these three genes was in leaves and the lowest in roots (Fig. 1). The transcript abundances of BnKCS1-1, BnKCS1-2, and BnCER1-2 were significantly higher in ZS11 than in NoWax, which exhibits a glossy phenotype,reduced wax crystals,and reduced wax amount on leaves in comparison with ZS11 (Fig. 2), suggesting that these three genes are involved in cuticular wax production in B.napus.

The expression of BnKCS1-1,BnKCS1-2,and BnCER1-2 in ZS11 under salinity, drought, and cold stress conditions was measured.As shown in Fig.3,the expression of BnKCS1-1,BnKCS1-2,and BnCER1-2 were upregulated significantly by drought stress and downregulated significantly by cold stress. Under salinity conditions, BnKCS1-1 and BnKCS1-2 transcription tended to decrease slightly, while BnCER1-2 increased significantly. All three gene transcripts decreased significantly under MeJA treatment, while BnKCS1-1 and BnKCS1-2 transcripts were significantly elevated after exposure to ABA. There was no significant change for the three genes under SA treatment,for BnKCS1-1 under ACC,or for BnCER1-2 under ABA(Fig.3).

3.2. Transcript levels of target genes in transgenic B. napus

To obtain BnKCS1-1, BnKCS1-2, and BnCER1-2 overexpression transgenic lines, the full-length coding sequence of BnKCS1-1, BnKCS1-2, and BnCER1-2 under the control of the CaMV35S promoter was transformed into the B.napus cultivar ZS11. For each overexpression construct, six independent transgenic lines were confirmed by GUS staining and kanamycin screening separately.In these overexpression lines,the transcript level of BnKCS1-1, BnKCS1-2, and BnCER1-2 significantly increased relative to that in non-transformed plants,with an increase of 7- to 27-fold for BnKCS1-1 (Fig. 4A), 17- to 64-fold for BnKCS1-2 (Fig. 4B), and 7- to 53-fold for BnCER1-2(Fig.4C).Although the gene transcript levels changed,no clear impact on plant agronomic traits was observed in these transgenic overexpression lines(Table S2).

Fig.3- Effects of exogenous hormones and abiotic stress on the expression of BnKCS1-1,BnKCS1-2,and BnCER1-2.The expression levels were examined by qRT-PCR.Means of three biological triplicates,each with three technical replicates,were compared by Student's t-test(*P ＜0.05).CK,untreated plants as control;SA,salicylic acid;MeJA,methyl jasmonate;ACC,1-aminocyclopropane-1-carboxylic acid;ABA,abscisic acid.

3.3. Cuticular wax structure in transgenic B. napus

To determine whether the epicuticular wax crystals were altered by upregulation of target genes, BnKCS1-1, BnKCS1-2,and BnCER1-2 overexpressing plants were further investigated by SEM. On the overexpressing B. napus leaf surfaces, wax crystals were much denser than those on non-transformed plants (Fig. 5). Some variations in the epicuticular wax crystallization pattern were observed on the leaf surfaces of transgenic lines (Fig. 5). For example, in the BnCER1-2 overexpression line,the leaf surface was covered with denser rod-shaped wax crystals.

3.4. Cuticular wax composition in transgenic B. napus

To evaluate the effect of BnKCS1-1, BnKCS1-2, and BnCER1-2 overexpression on wax production in B. napus, leaf waxes of two independent transgenic lines for each construct were extracted and quantified using GC-FID. As shown in Fig. 6,total amounts of leaf wax in the BnKCS1-1, BnKCS1-2, and BnCER1-2 overexpression lines were respectively 30%-46%,32%-76%,and 10%-11%higher than those in non-transformed plants.The relative changes of cuticular waxes were similar in BnKCS1-1 and BnKCS1-2 overexpressing lines, showing significant increases in aldehydes (C29and C30), alkanes (C28, C29,and C31), and secondary alcohols (C28and C29), and a significant decrease in C29ketone(Fig.7A,B).The overexpression of BnKCS1-1 and BnKCS1-2 promoted mainly the accumulation of long-chain fatty acids, primary alcohols, and aldehydes. For example, overexpression of BnKCS1-1 and BnKCS1-2 significantly increased the abundance of C30fatty acid,C28and C30primary alcohols,and C29and C30aldehydes,but significantly reduced the abundance of C24fatty acid, C26primary alcohol,and C26and C28aldehyde(Fig.7A,B).

The overexpression of BnCER1-2 led to an increase in alkanes, a decrease in secondary alcohols, and insignificant changes in other wax components (Fig. 6C). Compared to BnKCS1-1 and BnKCS1-2 overexpressing lines, the BnCER1-2 line promoted the accumulation of shorter-chain fatty acids,primary alcohols, and aldehydes. For example, overexpression of BnCER1-2 significantly increased the abundance of C24fatty acid, C26primary alcohol, and C28aldehyde, but significantly reduced the abundance of C26and C30fatty acids,C30primary alcohol,and C30aldehyde(Fig.7C).

3.5. Water loss and assessment of drought tolerance of transgenic plants

For each overexpression construct, two independent transgenic lines with high target gene expression and wax coverage were selected for further characterization. The BnKCS1-1,BnKCS1-2,and BnCER1-2 overexpression lines displayed lower rates of water loss than non-transformed plants(Fig.8A).The leaf water loss rate after 150 min was 28% for nontransformed plants and respectively 20%, 18%, and 22% for overexpression lines BnKCS1-1,BnKCS1-2,and BnCER1-2.

Fig.5-Epicuticular wax crystal structure on leaf surfaces of a non-transformed ZS11 plant and overexpression transgenic lines detected by scanning electron microscopy. Bar = 10 μm.

Fig.6- Cuticular wax compositions of BnKCS1-1,BnKCS1-2,and BnCER1-2 overexpression lines in B. napus.(A)BnKCS1-1 overexpression lines;(B) BnKCS1-2 overexpression lines;(C)BnCER1-2 overexpression lines.Values are means of four biological replicates plus or minus standard errors and asterisks indicate significant differences from non-transformed plants by Student's t-test(P ＜0.05).

Six-week-old transgenic plants and non-transformed plants were further subjected to water deprivation for 11 days, and then re-watered. Both non-transformed plants and transgenic plants were etiolated seven days after water deprivation, with non-transformed plants wilting severely.Two days after re-watering, non-transformed plants did not recover from dehydration, whereas the BnKCS1-1, BnKCS1-2,and BnCER1-2 overexpression lines regrew and showed at least one leaf recovering green color (Fig. 8B). This result showed that the BnKCS1-1, BnKCS1-2, and BnCER1-2 overexpression lines sustained growth better under soil water deficit conditions than non-transformed plants.

4. Discussion

4.1.BnKCS1-1,BnKCS1-2,and BnCER1-2 affect cuticular wax chemical profiles in transgenic plants

The members of KCS gene family have diverse catalytic activities and different substrate specificities [27,35]. KCS1 appears to be involved in the production of C26and longer VLCFAs for cuticular waxes [21]. When it was expressed in yeast, KCS1 showed broad substrate specificity for saturated and mono-unsaturated C16-C24acyl-CoAs [26,27]. In this study,the increase in total leaf wax accumulation per surface area by 30%-46% and 32%-76% in BnKCS1-1 and BnKCS1-2 transgenic B. napus lines may be attributed mainly to aldehydes (C29and C30), alkanes (C28, C29, and C31), and secondary alcohols (C28and C29) (Fig. 6A, B). It suggests that BnKCS1-1 and BnKCS1-2 in B. napus may be involved in the production of C28and longer VLCFAs for cuticular waxes.The changes in wax compound classes by BnKCS1-1 and BnKCS1-2 in transgenic lines were also accompanied by changes in chain-length profiles. In comparison with non-transformed plants,transgenic lines showed longer carbon-chain distributions of aldehydes,alkanes,and primary alcohols,and shorter carbon-chain distributions of secondary alcohols (Fig. 7A, B).The increased wax crystal density observed on the surface of transgenic B. napus leaves relative to non-transformed plants(Fig. 5) supports the notion that BnKCS1-1 and BnKCS1-2 influence wax production.

Fig.7-Levels of component species within each wax compound class from BnKCS1-1,BnKCS1-2,and BnCER1-2 overexpression lines in B.napus.Values are means of four biological replicates plus or minus standard errors and asterisks indicate significant differences from non-transformed plants by Student's t-test(P ＜0.05).

It has been reported[28,31]that AtCER1 plays an important role in wax VLC alkane biosynthesis of Arabidopsis. In our study,overexpression of BnCER1-2 led to a significant increase of total wax, particularly an increase in alkanes(Figs.6C, 7C),indicating that BnCER1-2 plays a role in wax VLC alkanes biosynthesis in B. napus. A higher density of wax crystals on the surface of BnCER1-2 transgenic B. napus leaves than on those of non-transformed plants also suggests that BnCER1-2 is involved in wax biosynthesis (Fig. 5). Secondary alcohols generated in the subsequent steps of the alkane biosynthesis pathway were reduced significantly in BnCER1-2-overexpressing B. napus lines. Opposite trends for changes in alkanes and secondary alcohols were observed in CER1 overexpression in Arabidopsis [31]. This contrast might be due to the limited activity of alkane hydroxylase.Compared to BnKCS1-1 and BnKCS1-2 overexpression,BnCER1-2 overexpression promoted the accumulation of short-chain fatty acids,primary alcohols, and aldehydes (Fig. 7), suggesting different roles for BnKCS1 and BnCER1-2 in chain-length distribution.

Wax mixtures derived from different plant sources have unique chemical compositions that determine their physical properties and thereby their potential applications in industry. At present, industrial utilization of waxes from plant surfaces is restricted by low wax coverage, poor agronomic properties,and a lack of diversity in wax mixtures from limited plant species. To produce plant surface wax for direct industrial use in the future, wax coverage and composition of temperate crop species could be genetically manipulated in a controlled way. Using a leaf area index of 5, B.napus ZS11 fields contain approximately 7 kg of leaf wax per hectare. In comparison, transgenic leaf waxes per hectare exceed this value only by 1-3 kg.Although this quantity is too low for industrial exploitation, cuticular wax modification in BnKCS1-1,BnKCS1-2,and BnCER1-2 transgenic rapeseed plants reveals potential target genes for biotechnological wax production from crops. Multigene overexpression might increase plant wax biosynthesis. In recent studies [36,37],multigene overexpression was the optimum strategy for increasing the biosynthesis of plant secondary metabolites.

Fig.8- Rates of water loss from BnKCS1-1,BnKCS1-2,and BnCER1-2 overexpression plants and non-transformed plants and observation of their morphology under drought stress.(A)Water-loss rates of detached leaves from non-transformed plants and transgenic plants.Water loss is shown as percentage of total water content,calculated as the difference between initial water-saturated weights minus dry weight for each leaf.(B)Non-transformed plants and transgenic plants under drought stress.“Control” plants were 7-week-old plants grown under well-watered conditions. “Drought” plants were 7-week-old plants that had been deprived of water for seven days.“Rewatering” shows“Drought” plants at 7 weeks plus 6 days after 11 days of water deprivation and re-watering for 2 days.

4.2. BnKCS1-1, BnKCS1-2, and BnCER1-2 promote droughtstress tolerance in B.napus

Wax biosynthesis can be modulated by hormone and abiotic stresses, such as ABA, water deficiency, and osmotic stress[38]. In the present study, BnKCS1-1, BnKCS1-2, and BnCER1-2 transcripts increased significantly in plants subjected to drought stress (Fig. 2), suggesting that these three genes are strongly involved in plant response to water deficit. Previous studies [23,39] have suggested that ABA is necessary to activate several cuticle-associated genes in response to water deficit. Our results showed that ABA induced the expression of BnKCS1-1 and BnKCS1-2 but exerted no effect on the expression of BnCER1-2(Fig.2),indicating that BnKCS1-1 and BnKCS1-2 might act in an ABA-dependent way to involve in drought stress response. Seo et al. [5] also reported that ABA-mediated MYB96 transcription factor bound to the promoters of KCS and KCR genes to modulate cuticular wax biosynthesis,thus conferring drought resistance on the plant.

Cuticle permeability to water plays a major role in plant adaptation to drought, allowing the plant to conserve tissuestored water and thereby delay the onset of physiological stress [4]. In the present study, BnKCS1-1, BnKCS1-2, and BnCER1-2 overexpressing lines showed reduced water loss rates and increased drought tolerance (Fig. 8). These changes might be related to changes in wax composition profiles,chain-length distribution, or packing arrangements of individual wax molecules [40]. Grncarevic and Radler [41] speculated that alkanes, primary alcohols, and aldehydes confer more efficient resistance to water movement through artificial membranes than do fatty acids. In BnKCS1-1, BnKCS1-2,and BnCER1-2 transgenic plants,the alkane-forming pathway of wax biosynthesis was most affected, although the chainlength distribution of wax compounds from the acyl reduction pathway was modified. For example, BnKCS1-2 and BnKCS1-2 overexpression led to an increase of 39%-88% in alkanes and 86%-152% in secondary alcohols, resulting in an increase of 30%-76%in total wax(Fig.6A,B).In BnCER1-2 transgenic lines,the amount of alkanes increased by 23%-34%, resulting in around 10% of total wax increase (Fig. 6C). This finding indicates that alkanes play a major role in reduced cuticle permeability and reduced susceptibility to soil water deficit.Compared to other wax constituents, Wax VLC alkanes have been proposed [31,38] to contribute the most to cuticle impermeability and to play an important role in plant response to water deficiency.

5. Conclusions

The present study has provided evidence that BnKCS1-1,BnKCS1-2,and BnCER1-2 are involved in cuticular wax production in B. napus. Their overexpression can modify amounts,compositions, and structures of cuticular wax and thereby increase tolerance to drought stress in transgenic B. napus.Cuticular wax modification in transgenic plants also suggests potential targets for biotechnological wax production in B.napus as a surface-wax crop.

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2019.04.006.

Declaration of Competing Interest

Authors declare that there are no conflicts of interest.

Acknowledgments

We thank Dr. Y. J. Guo at Southwest University for GC-MS analysis and Dr.R.Welti at Kansas State University for critical reading of the manuscript. This work was supported by the National Science Foundation of China (31771694, 31670407),the Chongqing Basic and Advanced Research Project(cstc2018jcyjAX0263, cstc2016jcyjA0170), the Fundamental Research Funds for the Central Universities (XDJK2017B028)and the China Agriculture Research System(CARS-12).