Effect of exogenous 5-aminolevulinic acid on glucosinolate biosynthesis in rape（Brassica napus L.）seedlings

MAODZEKA Antony,LU Lingzhi,ZHAO Xinze,XU Ying,WU Dezhi,JIANG Lixi()

Abstract 5-aminolevulinic acid (ALA) has been used as a plant growth regulator and can affect physiochemical processes,including carbon fixation and nutrient assimilation.Here,we investigated the effect of ALA on sulfur metabolism and glucosinolate (GSL) biosynthesis in rape (Brassica napus L.) seedlings, which were treated with 0, 0.5, 1.0, 5.0, and 10.0 mg/L ALA supplemented in a Hoagland solution. After 28 d of treatment, the effect of ALA on thiol and GSL contents and transcriptional profile of associated genes was analyzed. Results showed that low ALA concentrations increased cysteine, GSL, and total soluble thiol contents,and upregulated expression of genes such as BnSULTR1.1,BnSULTR2.2,and BnAPK1 that regulate sulfur uptake and metabolism.ALA also increased the GSL content,particularly the aliphatic GSLs,due to the improvement of sulfur metabolism and assimilation to methionine. Other GSLs biosynthesis stages, such as desulfo-GSL glycosylation,were also significantly improved by the ALA applications as indicated by the increased expression of BnUGT74B1 and BnUGT74C1. High ALA concentrations negatively affected sulfur metabolism and GSL synthesis by inflicting photo-oxidative stress that damaged cellular components. Moderate ALA concentrations promoted sulfur acquisition,assimilation,and GSL biosynthesis.

Key words 5-aminolevulinic acid;sulfur metabolism;glucosinolate;glutathione;rape seedling

5-aminolevulinic acid (ALA) is a key precursor of porphyrin biosynthesis leading to chlorophyll and heme. ALA has been widely used as a plant growth regulator due to its ability to promote a wide range of plant morphophysiological characteristics. ALA has been shown to promote plant physiological processes such as carbon fixation, nitrogen metabolism,chlorophyll synthesis resulting in improved plant growth and yield[1-2]. The promotion of plant abiotic stress defense systems such as enhancing antioxidant activity has prompted the use of ALA as an abiotic stress ameliorator capable of improving crop performance under a wide range of stresses such as drought,heavy metal and salt stresses[3-5].

ALA was shown to promote sulfur acquisition through promoting expression of sulfate transportersSULTR1andSULTR2inArabidopsis[6]. ALA also improves the transcripts of sulfur assimilatory genes such as adenosine 5′-phosphosulfate reductases and serine acetyltransferase (SAT) which lead to concurrent increases in thiols such as glutathione (GSH) and cysteine[6]. GSH plays important roles in plants acting as a reservoir of nonprotein reduced sulfur, antioxidant and signaling molecule[7]. Improvement of sulfur acquisition and assimilation by ALA can affect various biochemical processes that are directly or indirectly linked to sulfur metabolism such as secondary metabolite synthesis that includes glucosinolate(GSL).GSLs may contain up to 30%of the plant sulfur reserves and consequently they form a tightly linked and regulated association with sulfur metabolism[8-9].

GSLs are sulfur rich in anionic metabolites mostly found in the Brassicales and fall into three subclasses depending on the nature of the precursor amino acids that are aliphatic (alanine, leucine,isoleucine, valine, and methionine), indolic (tryptophan)and aromatic (phenylalanine or tyrosine)[10]. GSLs in particular are important in plant herbivore or pathogen deterrence and are particularly important to humans due to their anticarcinogenic properties and flavor[11-12]. Various methods, including irradiation[13],exogenous chemical application[14-15]and fertilization[16-17]have been applied to elicit plant glucosinolate responses; these methods act by influencing biochemical processes directly or indirectly linked to GSL biosynthesis. Sulfur fertilization in particular, was found to improve GSL content by 10-fold or more in different plant species[8]. The versatile effects of ALA on various plant physiochemical processes such as sulfur metabolism may prompt its use in eliciting health-promoting phytocompounds like GSL.

Improvement of thiols such as GSH and cysteine by ALA may affect GSL biosynthesis since they play important roles in GSL formation.Several glutathioneS-transferases(GSTs)were identified via computational predictions, which facilitate integration of GSH withaci-nitro compounds during GSL core structure formation[12]. Cysteine is required for methionine biosynthesis, which is required for synthesis of aliphatic GSLs that account for 70%-97% of total GSL content ofBrassicatissues[9].SUPEROOT 1(SUR1) encoding an enzyme with cystine lyase activity convertsS-alkyl-thiohydroximates to thiohydroximates during GSL biosynthesis[18]. Sulfotransferases (SOTs)participate in the final step of GSL synthesis by facilitating sulfation of desulfo-GSLs[11].

Transcriptional factors (TFs), such asSULFUR LIMITATION 1(SLIM1) andSULFUR DEFICIENCY INDUCED(SDI), mediate sulfur and glucosinolate metabolism through acting as upstream regulators of GSL biosynthesis under sulfur-limited conditions by downregulating the expression of key GSL biosynthetic genes and TFs[9].SLIM1can also upregulate myrosinases (MYRs) that break down GSLs and sulfur transporter (SULTRs) expression[19]. Increased sulfur availability increases GSL content even by up to 10-fold with more preference to aliphatic and aromatic GSLs[8]. MYRs or thioglucosidases that break down GSLs, have been shown to be significantly affected by the sulfate levels[20]. With such evidence,the importance of sulfur metabolism on GSL dynamics cannot be understated; the influence of ALA on plant GSL content should be investigated because ALA improves thiol metabolism.

The effects of ALA on plant secondary metabolism and the metabolome remain largely unexplored. ALA affects a wide range of plant physiochemical processes that include sulfur uptake and metabolism. Sulfur is an important macronutrient essential for normal plant growth and development and tightly linked to secondary metabolites such as GSLs.Brassica napusL. belonging to the Brassicales,is grown for seed oil (rapeseed) or as vegetable (leaf rape) since its leaves are laden with health-promoting phytocompounds such as GSLs. On this basis, we conducted this work to elucidate the potential role of ALA in the elicitation of the GSL contents of rape seedlings,which would be beneficial to humans due to their health-promoting and anticarcinogenic properties.

1 Materials and methods

1.1 Plant materials and growth conditions

Rape seeds (variety B4-Abba khel) were germinated in a float tray containing vermiculite and substrate (m∶m=1∶1) medium. Healthy two-week-old seedlings were then exposed to ALA treatments (0,0.5, 1.0, 5.0 and 10.0 mg/L) that were added to a supplementary basic Hoagland solution. The nutrient solution containing different ALA treatments was refreshed at three-day intervals and each treatment was replicated three times. Each float tray measured 30 cm×60 cm and contained 12 seedlings that constituted a single replicate. The plants were grown in a greenhouse with an ambient temperature of 25 ℃.After four weeks of continuous treatment with ALA, the plant materials were harvested, flash frozen in liquid nitrogen and then stored in an ultra-low temperature (-80 ℃) freezer for analysis of various biochemical parameters. Roots were washed with distilled water prior to flash freezing whereas leaf material was collected from the uppermost fully developed leaf. Measurements were done for various biochemical components including total soluble thiols,cysteine,and GSH.Enzymatic assays were conducted for SAT, cysteine synthase (CS), GST, phenylalanine ammonia lyase (PAL) and MYRs activities. The abundances of various transcripts of genes mediating thiol, glucosinolate and phenolic compound metabolism were also measured by real-time fluorescence quantitative polymerase chain reaction(RT-qPCR).

1.2 Quantification of thiolic compounds

Total soluble nonprotein thiols were assayed in accordance with a previously described method[21].Frozen leaf material was ground with a mill in 1.5 mL microtubes that were dipped in liquid nitrogen prior to grinding.With cooling, 1 mol/L HCl (500 µL) was then added and the tubes were vortexed and then centrifuged at 1.2×104gfor 10 min at 4 ℃. The supernatant (200 µL) was then added to 800 µL of neutralizing buffer (0.5 mol/L K2HPO4containing 25 µL of 10 mmol/L 5,5′-dithiobis-2-nitrobenzoic acid). The absorbance was then read at 412 nm. The amount of thiols were expressed in µmol/g as fresh mass.

Cysteine and GSH contents were assayed according to previous studies[22]. The thiolic compounds were extracted from 0.2 g frozen leaf tissue by homogenizing with 1 mL of extraction buffer [6.3 mmol/L diethylenetriaminepentaacetic acid (DTPA) containing 13 mmol/L trifluoroacetic acid(TFA)].The supernatants were collected after centrifuging at 1.3×104gfor 10 min and used for further analysis. Appropriate standards of cysteine and GSH were also prepared by dissolving in extraction buffer to final concentrations ranging from 0.001 to 0.01 mmol/L. Cysteine and GSH contents were analyzed by derivatization with monobromobimane (mBBr) and then followed by high performance liquid chromatography (HPLC)analysis. The HPLC system consisted of a Waters 1525 series binary pump equipped with a column heater, 2707 series autosampler, 2745 fluorescent detector and Empower 2 software (Waters Corporation,Milford, Massachusetts, USA). A Hypersil ODS2 column (5 µm particle size, 4.6 mm×150 mm) was used and the heater temperature was maintained at 40 ℃.Detection was done with excitation and emission wavelengths of 380 and 470 nm, respectively. The mobile phases included(A)99.9%acetonitrile(ACN)and (B) 89.9% water+10%ACN, and both containing 13 mmol TFA and the gradient program was set as described in a previous study[22].

1.3 Assay of enzymes involved in thiol metabolism

Fresh frozen leaf tissue was homogenized in different respective buffers (suited for assay of various enzymes) and centrifuged at 1.2×104gfor 10 min at 4 ℃. The protein content of the supernatant was measured according to the reference [23] and this was then used for the assays of different enzymes.

CS activity was assayed in accordance with a previously described method[24]. The reaction mixture(500 µL) contained 50 mmol/L phosphate buffer (pH 8.0),4 mmol/L Na2S,12.5 mmol/LO-acetylL-serine and the enzyme extract.After incubation at 30 ℃for 20 min, the reaction was terminated by adding 100µL of 7.5% trichloroacetic acid and the amount of cysteine synthesized was determined according to the reference [25]. A unit of CS activity was equivalent to 1 nmol/L of cysteine synthesized per minute and expressed per mg of protein.

SAT activity was assayed in accordance with previously described methods[21].The reaction mixture contained 63 mmol/L Tris-HCl(pH 7.6),1.25 mmol/L ethylenediaminetetraacetic acid disodium salt(Na2EDTA),1.25 mmol/L 5,5-dithio-2-nitrobenzoic acid (DTNB),0.1 mmol/L acetyl-CoA,1 mmol/LL-serine and 6µL of leaf extract. The reaction rate was measured at 412 nm with a spectrophotometer with a unit of enzyme defined as the amount of enzyme which catalyzes acetylation of 1 nmol ofL-serine per 5 min.

GST activity was assayed according to the method described in previous studies[26-27]. The reaction mixture contained 1.5 mmol/L GSH, 50 mmol/L potassium phosphate buffer (pH 6.5), 1 mmol/L 1-chloro-2, 4-dinitrobenzene (CDNB) and the enzyme solution.Addition of CDNB initiated the reaction and the increase in absorbance was measured at 340 nm using a spectrophotometer. The activity was calculated using the extinction coefficient of 9.6 mmol/cm.

1.4 Glucosinolate analysis

Glucosinolates were extracted and measured in accordance to previously described methods in a 96 well format with modifications[28]. GSLs were extracted from 0.01 g freeze-dried and milled leaf material with methanol containing sinigrin as an internal standard. Isolation of GSLs was done on a filter plate (No. MAHVN4550, Millipore, Tempe,Arizona, USA) containing 30 mg of DEAE Sephadex A25 followed by sulphatase treatment and then elution with 60% methanol and ddH2O using a vacuum manifold (WelVac 210, Rocker Scientific,Taiwan, China). Desulfo-glucosinolates separation and quantification were done on a Waters 1525 series binary pump equipped with a column heater, 2707 series autosampler,2998 series DAD detector(Waters Corporation,Milford,Massachusetts,USA)and Empower 2 software. A Hypersil C18 column (5 µm particle size, 4.6 mm×250 mm; Elite Analytical Instruments Co. Ltd., Dalian, China) was used with a heater temperature of 30 ℃.The injection volume was 45 µL and the absorbance was measured at 229 nm. Water(solvent A) and acetonitrile (solvent B) were used as the mobile phases and the gradient was as follows:100% A for 1 min, 95% A for 1 min, 90% A for 2 min,80%A for 5 min,70%A for 5 min,60%A for 2 min,50%A for 3 min,35%A for 2 min,20%A for 1 min, 90% A for 1 min, 100% B for 1 min and then 100%of A for 5 min.Quantification of glucosinolates was done with reference to the internal standard and using published response factors of individual GSLs[29].

1.5 MYR activity

MYR activity was measured in accordance to a method described in previous studies with slight modifications[30-31]. Leaf materials (500 mg) were homogenized with 5 mL of cooling 0.1 mol/L sodium phosphate buffer(pH 6.5).The mixture was centrifuged at 1×104gfor 15 min and the supernatant was used as the enzyme extract for assay of MYR activity. The MYR activity was assayed by incubating 500 µL of the extract with 500 µL of 0.25 mmol/L sinigrin at 37 ℃for 15 min and then the amount of glucose formed was determined with anthrone reagent.Protein content of extract was determined according to a method mentioned above[23]. One myrosinase unit was equivalent to 1 nmol glucose formed per minute and specific activity was expressed as units per mg of protein.

1.6 PAL activity

PAL activity was assayed in accordance with methods described in previous studies[32-33]. The reaction mixture containing 100 µL of the enzyme extract and 500µL of 0.1 mol/L sodium-borate buffer(pH 8.8) with 10 mmol/LL-phenylalanine was incubated at 37 ℃for 3 h and the reaction was stopped by addition of 0.05 mL 5 mol/L HCl. The amount oftrans-cinnamic acid (extracted with toluene) was estimated by measuring the absorbance at 290 nm. Enzyme activity was expressed as nmoltrans-cinnamic formed in per mg protein per minute.Protein content of the enzyme extract was assayed according to a method mentioned above[23].

1.7 RT-qPCR

Total RNA from freeze-dried root and leaf tissues was extracted using E.Z.N.A®Plant RNA Kit(R6827-01, Omega Bio-Tek, Norcross, Georgia, USA)according to manufacturer’s instructions. The residual DNA was then removed by digesting with DNase Ⅰat 37 ℃for 10 min and then the DNase was deactivated by incubation in a water bath at 75 ℃for 10 min. The extracted RNA was reversetranscribed with a ProScript First Strand cDNA Synthesis Kit (E6560S, NEB, Ipswich, Massachusetts,USA) using 1 µg of RNA template. Quantitative assays were done in triplicate on each cDNA template using Roche Light Cycler®480 (Roche Diagnostics,Mannheim, Germany). SYBR Green ⅠMaster in a 20 µL reaction volume and a Roche Light Cycler®4802. Sequence specific primers were designed to amplifyB. napushomologs obtained by searching the NCBI database (https://www.ncbi.nlm.nih.gov/) using targetArabidopsisgenes. Primers were designed with Primer Premier 5(http://www.premierbiosoft.com/index.html). The PCR amplification conditions included 1 cycle of 95 ℃for 5 min, followed by 45 cycles of 95 ℃for 10 s and 60 ℃for 15 s.Relative expression was calculated using the 2-△△CTmethod. Primers used in the study are listed in Table 1.

1.8 Statistical analysis

Analysis of variance was performed with oneway analysis of variance (ANOVA) followed by Duncan’s least significant difference (LSD) tests atP＜0.05 by using SAS 9.2 (SAS Institute). Principal component analysis (PCA) was done with Sigmaplot version 13.0 to visualize relationships amongst variables and detection, clustering, similarity and other trends.Graphical illustrations were prepared with Sigmaplot version 13.0 in conjunction with Microsoft Excel 2013.

2 Results

2.1 ALA mediated changes in thiol contents

Application of ALA significantly affected the thiol contents of rape seedlings (Fig. 1).ALA exerted variable effects on GSH content. Moderate doses (0.5and 1.0 mg/L ALA) promoted GSH content while further increases in ALA diminished the promotive effect or elicited negative effects (Fig. 1A). The treatment of plants with 0.5 and 1.0 mg/L ALA significantly increased GSH content to 0.84 and 0.85 mg/g which were equivalent to an increase of 26.43% and 28.6%compared with that of the control (0 mg/L). The GSH content did not significantly change in 5.0 mg/L ALA treatment, which was 0.68 mg/g. In 10.0 mg/L ALA treatment, the GSH content was significantly reduced to 0.51 mg/g, which was a 23.44% decrease compared with that of the control.

Table 1 Primers used in this study

ALA had a promotive effect on cysteine content as they increased significantly across all treatments(Fig. 1B). The highest increase in cysteine content was observed in the 1.0 mg/L ALA treatment which was 1.54 µg/g, an 86.77% increase in comparison with the control. The cysteine content in 5.0 mg/L ALA treatment was 1.34 µg/g, which was a 60.64%increase in comparison with that of the control. The lowest increase in cysteine content was observed in 10.0 mg/L ALA treatment, that is, 1.18 µg/g, which was a 41.47% increase compared with that of the control.

Application of ALA in moderate quantities (0.5 and 1.0 mg/L) resulted in an increase in the total soluble thiols, cysteine and GSH contents. Total soluble thiol contents were significantly increased in the 0.5,1.0 and 5.0 mg/L ALA treatments in comparison with the control(Fig.1C).Total soluble thiol contents in these treatments were 0.77, 0.89 and 0.81 µmol/g which corresponded to 15.1%, 33.12% and 21.67%increases,respectively.Application of 10.0 mg/L ALA increased total soluble thiol content to 0.72 µmol/g which was a 7.43%increase with respect to the control.

Fig.1 Effects of ALA on glutathione(GSH),cysteine and total soluble thiol contents

2.2 Influence of ALA on thiol metabolism

ALA had variable effects on various thiol metabolism enzymes that include SAT and CS, which are involved in cysteine biosynthesis and GST that facilitates integration of GSH with various components(Fig. 2). ALA promoted the activities of SAT across all treatments (Fig. 2A). The SAT activity in the 0.5 mg/L ALA treatment increased by 72.36%to 0.58 U/mg,whereas 1.0 mg/L ALA had the highest increase of 94.78%to 0.65 U/mg with respect to the control.

Similarly, ALA treatments promoted the activities of CS across all treatments(Fig.2B).The CS activities were significantly increased by 34.34% and 51.14%equivalent to 7.03 and 7.91 U/mg in 0.5 and 1.0 mg/L ALA treatments, respectively. The application of 5.0 mg/L ALA increased the CS activity by 15.8% (6.06 U/mg), whereas the increase in 10.0 mg/L ALA was slightly higher, that is, it was increased by 19.89%.The promotive effect of ALA on CS activities were more pronounced in the moderate ALA treatments, 0.5 and 1.0 mg/L, compared with that in the high ALA treatments.

The application of ALA negatively affected the GST activities (Fig. 2C). The application of 0.5 and 10.0 mg/LALA was characterized by a slight insignificant decrease in the GST activity. In these treatments, the GST activities declined by 8.36% and 3.22%, which were equivalent to 0.027 and 0.028 U/mg, respectively.The GST activities in 1.0 and 5.0 mg/L ALA treatments were significantly reduced by 29.73% and 13.22%corresponding to 0.02 and 0.025 U/mg,respectively.

2.3 Effect of ALA on expression of thiol metabolic genes

Fig.2 Effects of ALA on serine acetyltransferase(SAT),cysteine synthase(CS)and glutathione S-transferase(GST)activities

ALA treatments enhanced the expression of genes involved in sulfur uptake and assimilation (Fig.3A-I). Expression ofSULFUR TRANSPORTER 1(BnSULTR1.1) andBnSULTR2.2that facilitating sulfur uptake, was improved under different ALA treatments asBnSULTR1.1transcripts and their expression increased by more than 3-fold in the moderate ALA treatments (0.5 and 1.0 mg/L). The expression ofBnSULTR2.2increased by 44.32% in the 10.0 mg/L ALA treatment although preceding treatments elicited slight insignificant responses (Fig.3A-B).ADENOSINE-5'-PHOSPHOSULFATE KINASE 1(BnAPK1) transcript involved in providing activated sulfate for synthesis of secondary metabolites, was also increased particularly in the 1.0 and 10.0 mg/L ALA treatments, which caused a 230.0% and 64.4%increase, respectively (Fig. 3C). Expression ofGLUTAMATE-CYSTEINE LIGASE 1(BnGSH1)involved in GSH biosynthesis, was promoted particularly in moderate ALA treatments (0.5 and 1.0 mg/L) resulting in a 2-fold increase in transcripts in both treatments. In contrast, changes in later treatments were insignificant(Fig. 3D).GLUTATHIONE S-TRANSFERASE F11(BnGSTF11) involved in conjugation of GSH to various compounds during stress alleviation or secondary metabolite synthesis,was significantly downregulated in the 1.0 and 5.0 mg/L ALA treatments,whereas other treatments induced minimal changes(Fig. 3E). Expression of genes involved in cysteine and methionine synthesis,SERINE ACETYLTRANSFERASE 1(BnSAT1) (except 5.0 mg/L ALA treatment) andMETHIONINE SYNTHESIS 1(BnMS1) were also significantly enhanced by ALA (Fig. 3F-G).BnSAT1transcript increased by 2-fold in the 1.0 mg/L ALA treatment and also by 3-fold in the 10.0 mg/L ALA treatment. However, no significant change was observed in the 5.0 mg/L treatment.BnMS1transcript were significantly increased in all ALA treatments,and the highest increase of 78.41% was observed in the 0.5 mg/L ALA treatment.

SULPHUR DEFICIENCY-INDUCED 1(BnSDI1)andSULFUR LIMITATION 1(BnSLIM1)that modulate sulfur metabolism under sulfur deficiency were downregulated under ALA treatments (Fig. 3H-I). The expression ofBnSDI1was significantly reduced under all ALA treatments, whereasBnSLIM1transcript was not significantly affected by moderate ALA treatments. High ALA treatments (5.0 and 10.0 mg/L)significantly decreased theBnSLIM1transcript by 32.89%and 35.87%,respectively.

2.4 ALA induced changes in GSL contents

The effects of different ALA treatments on GSL profiles are summarized in Table 2. A total of 11 GSLs were detected and were dominated by the aliphatic GSLs (＞80%), particularly progoitrin (PRO)and glucobrassicanapin (GBN). ALA had varying effects on total GSL contents and on individual GSL.Application of ALA in moderate amounts(0.5 and 1.0 mg/L) had promotive effects on total glucosinolate content as it increased by 22.6%and 20.44%to 29.49 and 28.97 µmol/L compared with that of the control,whereas 5.0 and 10.0 mg/L ALA reduced the total GSL contents by 10.63% and 22.03% to 21.50 and 18.75 µmol/L, respectively. The proportion of aliphatic GSLs increased across treatments from 82%in the control to 87.35%, 87.26%, 94.33% and 94.35% in different respective ALA treatments. In moderate ALA treatments, aliphatic GSLs exhibited the greatest increase as they increased by 29.12% and 26.7%to 25.76 and 25.28µmol/L,respectively.

A single aromatic GSL, namely gluconasturtiin(GNT),was identified and exhibited the same pattern;its quantity increased by about 17.4% in both 0.5 and 1.0 mg/L treatments to 1.62 and 1.71 µmol/L,respectively.High ALA treatments(5.0 and 10.0 mg/L ALA) negatively impacted both aliphatic and aromatic GSL contents as they were significantly reduced under these treatments. However, the contents of glucoiberin (GB) and epiprogoitrin (EPI) remained higher or comparable to control levels even under these treatments.

The indolic GSLs, including 1-methoxyglucobrassicin(1-MeGBS)and 4-methoxyglucobrassicin(4-MeGBS), exhibited contrasting behavior in response to ALA treatments. Glucobrassicin (GBS) increased by 10.75%and 5.38%in the moderate ALA treatments,respectively, but decreased under high ALA treatments,thereby showing the same pattern as the majority of the GSLs. 1-MeGBS and its stereoisomer 4-MeGBS decreased across all ALA treatments. Consequently,total indolic GSLs decreased in all ALA treatments.

Fig.3 Effects of ALA on relative expression levels of thiol metabolic genes

PCA was done to illustrate the changes in various GSL components under different ALA treatments (Fig.4A-B). PCA identified principle components PC1 and PC2 that accounted for 56.98% and 19.94% of the variance,respectively.PC1 was able to show GSL contents based on different treatments. Among the treatments, 0.5 and 1.0 mg/L ALA were distinct from 5.0 and 10.0 mg/L ALA treatments. PC1 loadings were negative for EPI (-1.28) but positive for all other GSLs. 4-methoglucobrassicin, GBS, and PRO had the highest PC1 loading coefficients. PC2 showed contrast between EPI, GB, gluconapoleiferin (GNL),GBN, gluconapin (GNA) and PRO which had positive coefficients with 4-MeGBS, GNT, GBS,glucoalyssin (GAL) and 1-MeGBS which had negative coefficients.

2.5 Mediation of ALA in GSL metabolism

ALA had significant effects on genes that mediate various steps of GSL biosynthesis (Fig. 5A).Expression ofBnMYB28, a major TF modulating methionine derived or aliphatic GSL biosynthesis was enhanced under different ALA treatments. The highest increase was observed in the 0.5 and 10.0 mg/L treatments, which were about 12- and 16-fold,respectively.Similarly,BnMAM1,involved in methionine chain elongation in the synthesis of aliphatic GSLs was significantly upregulated by approximately 17-fold in the moderate ALA treatments and by 6- to 7-fold in the high ALA treatments.

Table 2 Effect of 5-aminolevulinic acid on GSLs content of B.napus leaves(dry mass) µmol/L

Fig.4 Component loadings (A) and score plot (B) from PCA analysis of leaf GSLs contents under different ALA treatments

BnMYB51andBnMYB34, TFs that modulate indolic GSL biosynthesis, were unaffected by moderate ALA treatments (0.5 and 1.0 mg/L) but were downregulated under high ALA treatments (5.0 and 10.0 mg/L). Among those genes involved in aldoxime formation,BnCYP79A2andBnCYP79B2, participating in phenylalanine and tryptophan derived aldoxime formation, were not significantly affected by moderate ALA.BnCYP79F1, which mediates methionine derived aldoximes, was upregulated only in the 1.0 mg/L ALA treatment.The expression ofBnCYP83A1,which converts methionine derived aldoximes to correspondingaci-nitro compounds,was also significantly upregulated in the 0.5 and 1.0 mg/L ALA treatments.BnSUR1, withC-Slyase activity that facilitates thiohydroximate formation, was upregulated in the moderate ALA treatments with a significant increase of transcripts in the 0.5 mg/L ALA treatment. Similarly,transcript ofBnSOT18, a sulfotransferase (SOT), was upregulated by moderate ALA with an 82% increase in transcripts in the 1.0 mg/L ALA treatments. The expression ofBnUGT74B1andBnUGT74C1,which are actively involved in glycosylation of thiohydroximates to desulfo-GSLs, was significantly upregulated by moderated ALA treatments.High ALA treatments had negative effects on expression of various genes involved in the GSL core structure formation exceptBnMYB28,BnUGT74C1, andBnMAM1because their transcript abundance in these treatments remained relatively higher than the control.

Fig.5 ALA mediated responses in GSLs biosynthetic genes

Most of the aliphatic GSLs had positive correlation with expression levels of their respective biosynthetic genes. GNA and GBN exhibited the highest positive correlation with respective biosynthetic genes (Fig. 6).EPI and GAL had the lowest correlation coefficients characterized by negative and positive correlations with respective biosynthetic genes.Amongst the indolic GSLs, GBS had the highest significant correlation coefficients with respective biosynthetic genes. GNT,the only detected aromatic GSL, was positively correlated with different respective biosynthetic genes with a significant correlation withBnSUR1.

2.6 Effect of ALA on MYR and PAL activities

MYRs are directly involved in the breakdown of GSLs to different products whereas PAL converts phenylalanine, which is also as precursor of aromatic GSLs, totrans-cinnamic acid and then facilitates synthesis of phenolic compounds. Effects of ALA on MYR and PAL activities are summarized in Fig. 7AB. Moderate ALA treatments (0.5 and 1.0 mg/L) had no significant effects on MYR activity, although a slight decrease was observed. High ALA treatments(5.0 and 10.0 mg/L) significantly increased MYR activity by 77.49%and 35.87%(49.14 and 37.62 U/mg,respectively) compared with those of the control (Fig.7A).The relative transcript ofBnTGG2, a MYR gene,corroborated with biochemical data characterized by insignificant changes in transcript abundance in moderate ALA treatments and enhanced expression levels in the higher ALA treatments(Fig.7C).

The PAL activity was significantly increased by 45.09%and 54.64%(18.23 and 19.43 U/mg,respectively)in the moderate ALA treatments (Fig. 7B). High ALA treatments had antagonistic effects, considering that PAL activity was significantly decreased in 5.0 and 10.0 mg/L ALA treatments. The transcript abundance ofBnPAL1, a PAL gene, also corroborated with biochemical data characterized by high transcript abundance in moderate ALA treatments and lower expression levels in the later treatments(Fig.7D-E).

Fig.6 Heat maps indicating correlation of GSLs biosynthetic genes and different GSLs present in rape leaves under different ALA treatments

3 Discussion

Elicitation is a competent way of improving the content of health-promoting bioactive compounds in food crops. Phytocompounds such as ascorbate,flavonoids, tocopherols, and GSLs are essential for maintenance of human wellbeing because of their anticarcinogenic and promotive effect on the cardiovascular system. The application of ALA had a diverse range of effects onB. napusseedlings,including thiol metabolism and also on GSL contents.

In this study, application of ALA improved plant thiol status by enhancing sulfur uptake and assimilation. ALA increased expression of sulfate transportersBnSULTR1.1andBnSULTR2.2, which could have resulted in increased uptake of sulfate as observed by other researchers[6]. The assimilation of sulfate into compounds such as 3′-phosphoadenosine 5′-phosphosulfate (PAPS), cysteine, methionine and GSH was also promoted as indicated by expression ofBnAPK1,BnSAT1,BnMS1, andBnGSH1involved in the synthesis of these compounds.The increase in CS activity could increase the incorporation of inorganic sulfide into cysteine. The inherent accumulation of these products could avail additional resources for other ensuing processes, such as synthesis of secondary metabolites like GSLs.

Although ALA promoted the total plant thiol status, high ALA treatments diminished the promotive effect as indicated by the reduced total soluble thiols and GSH contents. High ALA treatments were shown to inflict photooxidative stress, which could damage cellular components such as membranes and enzymes that facilitate essential processes, including thiol metabolism[34-35]. As such, the activities of thiol metabolic enzymes,such as CS and SAT,decreased.

Fig.7 Effect of ALA on MYR activity (A), PAL activity (B), BnTGG2 expression (C), BnPAL1 expression (D) and BnTGG2 and BnPAL1 transcript abundance(E)

Our results showed that moderate ALA treatments promoted GSL contents inB. napusseedlings, although a particular bias favored aliphatic GSLs. ALA enhanced expression of genes at various stages of GSL biosynthesis.For instance,the expression ofBnMYB28, which is a TF that modulates aliphatic GSL biosynthesis, and the expression ofBnMAM1,BnCYP79F1andBnCYP83A1, which are critical in core structure formation were improved[11].BnMAM1transcript was increased, indicating that the increase in aliphatic GSLs was mainly due to the enhanced incorporation of methionine into GSL synthesis. The aromatic GSL gluconasturtiin, formed from the precursor phenylalanine, was also significantly increased. Phenolic compound synthesis requires phenylalanine, the same precursor for aromatic GSL biosynthesis[36]. The PAL activity and associated transcript (BnPAL1) increased under ALA treatments suggesting that ALA promoted both pathways, and the increase in aromatic GSLs was not a result of impairment of phenolic compound synthesis.

The application of ALA improved various stages of GSL core structure formation (Fig. 8). Genes such asBnAPK1,BnSUR1, andBnSOT18link thiol and GSL metabolism as they facilitate the export of products from sulfur metabolism into GSL synthesis[10].The improved thiol metabolism by low ALA, such as promotion ofBnAPK1expression,could have led to a successive increase in PAPS; a complementary increase in expression ofBnSOT18, a PAPS: desulfo-GSL sulfotransferase, could have also increased efficiency in the final step of GSL core structure formation. The abundance ofBnSUR1transcript, that possessesC-Slyase activity, was also significantly increased indicating efficiency in formation of thiohydroximates.BnGSTF11is putatively involved in catalyzing the integration of GSH withaci-nitro compounds to formS-alkyl-thiohydroximates[37]. In the present study, ALA treatments did not improve expression ofBnGSTF11. The glycosylation of thiohydroximic acids to form desulfo-GSLs was also increased by ALA as relative expression ofBnUGT74B1andBnUGT74C1, which catalyze this process, increased significantly. ALA promotes photosynthetic activity which could enhance primary assimilate production and inherently avail additional resources, such as UDP-glucose,for GSL biosynthesis.

Fig.8 Possible mechanisms by which ALA influences changes in GSL profiles in rape seedlings

Indolic GSLs exhibited contrasting responses to ALA treatments because glucobrassicanapin increased under moderate ALA treatments whilst 1-MeGBS and its stereoisomer 4-MeGBS were decreased significantly. Similar observations were found in treated pak choi, in which sulfur promotes aliphatic and aromatic GSLs but not indolic GSLs[38]. Indolic GSLs are related to indole acetic acid(IAA)formation,and indole-3-acetaldoxime(IAOx)is required for their synthesis[39].CYP79B2andCYP79B3catalyzes the conversion of tryptophan to IAOx, and the overexpression of these genes is associated with increase of IAA levels[40]. Factors determining how much of IAOx is channeled to either IAA or indole GSLs synthesis might explain these observations, and a thorough investigation will render an in-depth understanding of this process. ALA is a welldocumented and potent growth regulator known to affect developmental processes[1-2,41]. ALA could have skewed the conversion of IAOx to auxin to suite its role in growth regulation. Indolic GSLs breakdown also promotes IAA synthesis which results in improved root growth, particularly under sulfur deficient conditions[9]. However, sulfur deficiency would be an unlikely reason for the observed decrease in the content of indolic GSLs because thiolic contents of rape seedlings were improved by ALA. GSL regulatory networks are very complex,and further research is recommended to understand the underlying mechanisms.

GSL contents were significantly reduced under high ALA treatments.BnMYB28,BnMYB34andBnMYB51, which are TFs that are involved in modulation of various stages of GSL synthesis, were significantly reduced in these treatment[42]. Various other key genes involved in GSL core structure formation, that includeBnCYP83A1andBnUGT74B1,were also downregulated under high ALA treatments likely resulting in the reduced GSL contents[11]. This result might be due to oxidative stress inflicted by high ALA treatments, which hindered essential biochemical processes that damaged cellular components,thereby impairing various metabolic pathways[34-35].This mechanism could divert most thiols to the synthesis of stress alleviatory compounds, such as GSH, which play a major role in stress-mitigating mechanisms.

MYRs have been shown to catalyze breakdown of GSLs under conditions of stress in order to unlock sulfur, which can also be used in other critical stress mitigating mechanisms[9]. In the present study, the MYR activity was particularly enhanced under high ALA treatments.BnSDI1andBnSLIM1activate myrosinases, regulate sulfur metabolism and also downregulate GSL biosynthesis genes and TFs under sulfur deficient conditions[9]. However, these genes were downregulated under all ALA treatments indicating that GSL breakdown was initiated primarily not as a result of sulfur shortage. The decrease in GSL contents under these treatments was largely due to an impairment of the GSL biosynthesis pathway although myrosinases were also activated.

4 Conclusions

The improved sulfur acquisition and metabolism by moderate quantities of exogenous ALA results in improved thiol contents of rape seedlings. This improvement promotes GSL biosynthesis, which is closely linked to sulfur metabolism.We conclude that exogenous ALA can be a feasible way of improving health-promoting GSLs in leaf rape and other brassicaceous crops commonly used as vegetables,which could be beneficial to human health.