- Email: [email protected]
A new ELISA for quantification of brassinosteroids in plants
Steroids xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
A new ELISA for quantification of brassinosteroids in plants Andrey G. Pradko b, Raisa P. Litvinovskaya a, Alina L. Sauchuk a, Svetlana V. Drach a, Alexander V. Baranovsky a, Vladimir N. Zhabinskii a, Tatyana V. Mirantsova b, Vladimir A. Khripach a,⇑ a b
Institute of Bioorganic Chemistry, National Academy of Sciences of Belarus, Kuprevich St., 5/2, 220141 Minsk, Belarus Hemma-Test Ltd., Masherova St., 11, 220029 Minsk, Belarus
a r t i c l e
i n f o
Article history: Received 12 August 2013 Received in revised form 1 August 2014 Accepted 22 August 2014 Available online xxxx Keywords: Brassinosteroids Phytohormones Synthesis ELISA HPLC Flores Chamomillae
a b s t r a c t Starting from (22R,23R)-2a,3a,22,23,26-pentahydroxy-5a-cholestan-6-one 26-hemisuccinate, conjugates of 28-norcastasterone with horse radish peroxidase and bovine serum albumin were prepared. The latter conjugate was injected into rabbits; produced polyclonal antibodies were used to quantitate 6-keto-brassinosteroids. The newly developed analytical system was used in combination with two other immunoenzymatic assays for brassinosteroids to determine individual compounds of this series. In addition, a direct method of brassinosteroid analysis was proposed. It has the advantage of requiring no sample pretreatment steps such as extraction with organic solvents and chromatography. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Brassinosteroids (BS) are low molecular bioregulators playing an important role in plant growth and development [1]. Content of natural BS in plant sources is very low (less than 105%) [2], and to achieve notable biological effects from their application, only a minute amount of exogenous BS is required. A pronounced plant growth stimulating and adaptogenic effect was observed on BS treatment of various plant species at a dose 5–20 mg/ha resulting in a positive effect on the quantity and quality of crops and tolerance against various biotic and abiotic stresses [3]. These features of BS have made them attractive as active ingredients for agrochemicals [4]. The most active and the most important BS from both scientific and practical point of view are those containing 6-oxo-7-oxa- or a 6-ketone function in the B-cycle [5–8] (Fig. 1). 24-Epibrassinolide 2 and 28-homobrassinolide 3 bearing a lactone moiety in the B-cycle have already found a practical use in agriculture [9,10].
Abbreviations: BS, brassinosteroids; BSA, bovine serum albumin; C.V., coefficient of variation; LOD, limit of detection; DCC, N,N0 -dicyclohexylcarbodiimide; DW, dry weight; HRP, horse radish peroxidase; NHS, N-hydroxysuccinimide; PBS, phosphate buffered saline; SD, standard deviation; SIM, selective ion monitoring; TMB, 3,30 ,5,50 -tetramethylbenzidine; Tris, tris(hydroxymethyl)aminomethane. ⇑ Corresponding author. Tel./fax: +375 172 678 647. E-mail address: [email protected] (V.A. Khripach).
It is accepted that compounds 5–8, which belong to 6-keto-BS subgroup, are not only biosynthetic precursors of the corresponding lactones 1–4, but they also play an independent role in plant growth and development [11]. There are reasons to believe that these highly active and easily accessible BS could also find their practical applications in agriculture. One of the preconditions to achieve their massive use is a simple, selective and highly sensitive method for their analysis. Based on our and others experience in BS research [12–20], the ELISA could be considered as a method of choice for the routine analyses. In this respect, the first task of the present investigation was to develop an ELISA for 6-keto-BS. A certain disadvantage of immunochemical methods is that they are group specific ones and cannot be applied for analysis of an individual compound. The obvious solution of the problem is a combined application of several immunoassays separately quantifying the group components and/or immunoassaying of chromatographic fractions. In this connection, the present study proposes quantification of individual BS of 28-homo-series (28-homocastasterone and 28-homobrassinolide) by a joint utilization of the assay for 6-keto-BS (IA1) elaborated in this study with previously developed test systems for B-lactone-BS (IA2) [21] and 28-homo-BS (IA3) [22] (Fig. 2). Although each of these three assays is a group specific, their groups are different, and this allows differentiation of the hormones that are interfering when the only one group is analyzed with a single ELISA.
http://dx.doi.org/10.1016/j.steroids.2014.08.022 0039-128X/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
2
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
Fig. 1. Structures of the most important natural brassinosteroids.
2. Experimental 2.1. Reagents and materials The chemicals and materials used in this study were purchased from following sources: DCC, NHS, HRP, dioxane, and Sephadex G-25 from Sigma–Aldrich (USA); BSA from Acros Organics (Germany); chloroform, cyclohexane from ECOS-1 (Russia); complete Freund’s adjuvant from ICN Biochemicals (France); methanol from Fisher Scientific UK Ltd. (UK); silica gel from Macherey–Nagel GmbH & Co KG (Germany). High binding capacity 96 flat-bottom wells microtiter plates (Maxisorp™) used acquired from Nalgene Nunc (Thermo Fisher Scientific, Inc., USA). The following buffers were used for the immunoassay: coating buffer 1 (0.05 M PBS, pH 7.4, containing 0.9% NaCl), coating buffer 2 (0.025 M PBS, pH 7.4, containing 0.1% BSA, 0.02% Tween™ 20), blocking buffer (0.025 PBS, pH 7.4, containing 0.1% BSA, 2% sucrose, 5% sorbitol), assay buffer (0.05 M Tris, pH 7.4, containing 0.9% NaCl, 0.1% BSA, 0.02% Tween™ 20) and washing buffer (1% NaCl, containing 0.02% Tween™ 20). The ready solution of TMB in the substrate buffer containing H2O2 (Enhanced K-Blue TMB Substrate, Neogen, USA) was used for staining in assay, 5% H2SO4 was used as a stopping reagent for quenching the color reaction. BS and their derivatives were synthesized in the Laboratory of Steroid Chemistry (Institute of Bioorganic Chemistry, NAS of Belarus); other steroids were purchased from Steraloids Inc. (USA). 2.2. HPLC Preparative high-performance liquid chromatography was carried out on a HPLC–Hitachi (UV-VIS Detector L-4250, Intelligent Pump L-6200A) with reversed phase using column RP18 (LiChroCART 250 mm 10 mm 10 lm). 70% Aqueous acetonitrile was used as mobile phase with flow rate 1.5 mL/min. UV detection was performed at a wavelength of 204 nm. 2.3. HPLC–MS The HPLC–MS experiments were carried out on a Accela chromatograph (Thermo Electron Corp., USA) linked to a LCQ Fleet 3D ion trap mass spectrometer equipped with an electrospray
interface (Thermo Fisher Scientific, USA) with reversed-phase column C18 HYPERSIL Gold (50 mm 2.1 mm 1.9 lm). The column thermostat was set at 25 °C and 75% acetonitrile in water was used as a mobile phase for 15 min at a flow rate of 100 lL/min. The column was equilibrated to initial conditions for 5 min. The ESI-MS parameters (positive mode) were as follows: spray voltage (4.5 kV), capillary temperature (325 °C), and capillary voltage 36 V. Nitrogen was used as sheath, auxiliary and sweep gas. The sheath, auxiliary and sweep gas flow rates were 30, 0 and 0 arbitrary units, respectively. Data were processed by Xcalibur software (version 2.0.6, Thermo Electron Corp., USA).
2.4. Synthesis of the immunogenic conjugate of 28-norcastasterone with BSA A solution of (22R,23R)-2a,3a,22,23-tetrahydroxy-5a-cholestan-6-one 26-hemisuccinate 9 (60 mg, 107 lmol, prepared according to [21]) and NHS (16 mg, 139 lmol) in anhydrous dioxane (10 mL) was added sequentially to a solution of DCC (26 mg, 126 lmol) in anhydrous dioxane (10 mL) at 8–10 °C. The reaction mixture was stirred for 30 min at this temperature and then for 6 h at room temperature. The precipitated dicyclohexylurea was removed by filtration, and obtained solution of ester 10 was added to a solution of BSA (120 mg in 20 mL of 0.1 M NaHCO3, pH 8.35). The mixture was kept for 20 h at room temperature. Then an excess of activated ester and dioxane was removed by dialysis against 0.05 M NaCl to give, after lyophilization, 160 mg of the conjugate 11, which was frozen and stored at 18 °C.
2.5. Synthesis of labeled antigen A solution of active N-hydroxysuccinimide ester 10 (4.5 mg, 6.6 lmol), prepared as described above, was added to a solution of HRP (9 mg) in distilled water (900 lL) with 0.1 M NaHCO3 (2– 3 drops, pH 8.35). The reaction mixture was stirred at 10 °C for 1.5 h and purified on a column with Sephadex eluting with distilled water. The yellow–brown fraction containing conjugate 12 was collected, diluted with glycerol (1:2), and stored in a freezer at 18 °C.
Fig. 2. Immunodominant fragments of BS-molecules specific for recognition by IA1, IA2, and IA3.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
2.6. Antisera production A group of eight rabbits was immunized with the conjugate of 28-norcastasterone with BSA 11. Every rabbit was subcutaneously injected into 4–6 points on its back with 1 mg conjugate, dissolved in a PBS (0.5 mL, pH 7.4) and emulsified in the equal volume of the complete Freund’s adjuvant. Intervals between the injections were 4–4.5 weeks. The immunization was continued for 5 months, and blood from the auricular vein was taken at regular intervals. The taken samples of blood serum were tested for the binding ability for 28-norcastasterone 8, and their titers (working dilutions) were determined. The antiserum A1 with the highest titer and sensitivity in the assay was used in this study. 2.7. Analysis of BS in plants by ELISA 2.7.1. Preparation of samples 2.7.1.1. Variant A. Dry flowers of horse gowan Flores Chamomillae (3 g) (bought at the drug store) were homogenized and extracted with methanol (30 mL) for 24 h at room temperature. The extract was filtered on a sintered glass filter; an aliquot (1 mL) was evaporated, dissolved in assay buffer (0.5 mL) and used for analysis. The rest of methanol extract was used in Variant B. 2.7.1.2. Variant B. The methanol extract from Variant A was evaporated and the residue was partitioned between cyclohexane (15 mL) and 80% aqueous MeOH (3 10 mL). The combined 80% aqueous MeOH extract was evaporated to dryness; the residue was dissolved in MeOH (1.5 mL) and subjected to TLC (20 20 cm) on silica gel, using a mixture of chloroform–methanol (88:12) as eluent. The BS-containing zone (Rf 0.4–0.6) was collected and eluted with MeOH, solvent was evaporated. The residue was dissolved in MeOH (3 ml) and an aliquote of 100 lL was evaporated, dissolved in assay buffer (0.5 mL) and used for analysis by ELISA. The rest of methanol solution was evaporated and the obtained sample was used in Variant C. 2.7.1.3. Variant C. The sample from Variant B was dissolved in 300 lL MeOH and separated by reversed-phase HPLC, using 70% aqueous MeCN as mobile phase with flow rate 1.5 mL/min. The solvents were degassed before using. Based on the retention times of 28-homobrassinolide (15 min) and 28-homocastasterone (16 min), the fraction (Rt 14–17 min) was collected and evaporated. The residue was dissolved in MeOH (3 mL) and an aliquote of 100 lL was evaporated, dissolved in assay buffer (0.5 mL) and used for analysis by ELISA. The remaining part of the collected fraction was evaporated and used for analysis by HPLC–MS.
3
and incubated at 4 °C overnight. The blocking solution was removed from the wells; the plates were dried at room temperature. The calibrating samples (0–100 nM) or the test samples of BS (50 lL in duplicates) and the HRP-conjugate (100 lL per well) were added to the wells of the plate with the immobilized antibodies. The plates were incubated for 1 h at 37 °C, the liquid was removed from the wells, and the plates were washed 4 times with 150 lL of washing buffer. The TMB solution was added to the wells (150 lL) and incubated for 15 min at 37 °C. The reaction was quenched by the addition of stopping reagent (50 lL per well), and optical absorbance was measured at 450 nm with a plate reader (F300TP, ‘‘VITYAS’’, Belarus). The analytical results were calculated by interpolation from the calibration curve and values were expressed in nmol/L or ng/g. Sigmoid curves for standards were linearized by the log–logit transformation: logit B/B0 = ln((B/B0)/ (100 B/B0)) [23,24]. 2.7.2.2. Method 2. The calibrating samples or the test samples of BS (150 lL per well) were placed in wells of the plate with the immobilized antibodies. The plates were incubated for 2 h at 37 °C, liquid was removed from the wells, and the plates were washed 4 times with 150 lL of washing buffer. The HRP-conjugate (150 lL) was added to the wells and incubated for 5 min at 37 °C. Then the liquid was removed from the wells, and the plates were washed as described above. The TMB solution in the substrate buffer (150 lL) was added to the wells and incubated for 20 min at 37 °C. The reaction was quenched by the addition of stopping reagent (50 lL in each well), and optical absorbance was measured at 450 nm. The analytical results were calculated by interpolation from the calibration curve as above. Validation procedures were carried out according to the methods previously published [23]. Accuracy was determined by adding known amounts of steroid standard dose (10 nmol/L for Method 1, 0.1 nmol/L for Method 2) to model samples. All samples were assayed by 6-keto-BS ELISA (IA1). Imprecision was estimated by replicate determinations of BS levels in model samples. To evaluate assay parallelism, serial dilutions of 3 model samples prepared according to Variant B were tested for Method 1 and samples prepared according to Variant D were tested for Method 2. 2.8. Analysis of BS in plants by HPLC–MS Calibrators and test samples in 0.2 ml glass inserts were derivatized by heating to 60 °C for 30 min with a solution of
2.7.1.4. Variant D. Dry flowers of horse gowan Flores Chamomillae (1 g) were homogenized in 20 mL of assay buffer. The buffer extract was centrifuged (3000 rpm, 20 min); the supernatant diluted with assay buffer (1:50) was used for analysis. 2.7.2. ELISA procedure ELISA test-systems IA1, IA2 [21] and IA3 [22] were used for determination of 6-keto-, B-lactone- and 28-homo-BS, respectively. 2.7.2.1. Method 1. Microtiter plates for ELISA were prepared as follows. The microtiter plates were coated with the sheep antibodies to rabbit IgG (2 lg/mL, 150 lL per well) in coating buffer 1. The plates were covered with parafilm to prevent evaporation. After overnight incubation at 4 °C, the coating solution was removed from the wells without washing. Then the plates were coated with the antiserum A1 (150 lL per well) in coating buffer 2 and incubated at 4 °C overnight. The coating solution was removed from the wells and blocking buffer was added to the wells (200 lL)
Scheme 1. Synthesis of immunogenic conjugate 11 and labeled antigen 12.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
4
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
3-(dansylamino)phenylboronic acid (20 lL, 5 mg/mL) in a mixture of pyridine and acetonitrile (1:19) [25]. The reaction mixture was then diluted ten-fold with acetonitrile and investigated by HPLC–MS without further purification. The volume of injected sample was 5 lL, the retention time was determined by the standard sample (9.80 min for authentic 28-homobrassinolide dansylaminophenylboronate). Quantification was performed by SIM of the molecular-related ions [M+H]+ (m/z 829). The samples were analyzed immediately after preparation. 3. Results and discussion The hemisuccinate 9 [21] was used as a starting material for the preparation of conjugates with proteins (Scheme 1). The conjugate 11 of 28-norcastasterone with BSA was obtained by the interaction of the activated N-hydroxysuccinimide ester 10 with BSA in aqueous dioxane solution followed by dialysis and lyophilization. The reaction of the ester 10 with a HRP solution in the presence of potassium bicarbonate gave, after chromatographic purification of the product on Sephadex, the conjugate 12, which was used as a labeled antigen. The 6-keto-BS specific antibodies were produced by injecting the antigen 11 into rabbits. Titers were determined to evaluate antiserum binding ability for 28-norcastasterone 8. The experiments were carried out with the antiserum A1. Its dilution 1:500,000 and dilution of HRP-conjugate 1:10,000 were found to be the best combination of immunoreagents for both Method 1 and Method 2 ELISA. The 28-norcastasterone calibration curve for IA1 was constructed in the concentration range of 1–100 nmol/L for Method 1, 0.01–5 nmol/L for Method 2 (Fig. 3). An analytical working range of the assay for Method 1 is 1–100 nmol/L (part of the curve between 10% and 60%), and for Method 2 is 0.01–1 nmol/L (part of the curve between 30% and 90%) for Method 2. The limit of detection (LOD) for Method 1 was found to be 0.3 nmol/L, for Method 2 – 0.01 nmol/L. The LOD observed in this study for Method 1 was close to those reported by us earlier [13,14,22]. The only other example of BS ELISA [18,19] based on rabbit antisera against 24-epicastasterone 6 showed LOD value of 1.5 nmol/L. Accuracy studies, parallelism and imprecision data were determined and all were found to be satisfactory. The recovery of added
28-norcastasterone ranged from 95% to 114% as shown in Table 1, with range of values being narrower for Method 1. The results of serial dilution of 3 model samples are shown in Table 2. They varied from 6% to 9% for Method 1 and from 7% to 14% for Method 2. The results of assay imprecision for 3 model samples are shown in Table 3. They ranged from 4% to 7% for Method 1 and from 10% to 15% for Method 2. Although Method 2 showed rather high coefficient of variation for both parallelism and imprecision experiments, nevertheless, taking into account low concentration level of its operating, it could be considered as an acceptable one. The specificity of 28-norcastasterone antiserum A1 was determined by competitive binding assay with various steroids. As seen from Table 4, the obtained antibodies were highly specific for the determination of 6-keto-BS, but showed low cross-reactivity with the synthetic BS-analog, (22S,23S)-24-epicastasterone 14, having non-natural stereochemistry of the diol group in the side chain. There was no cross reactivity detected with non-BS steroids like cholesterol 20, androstenolone 21, or synthetic intermediates 15–17. Only a small cross reactivity (0.3–1.6%) was observed for B-lactone-BS 1–3, 18. Having in hands this new immunoassay and earlier developed ELISAs for other important groups of natural BS, we could proceed further with analysis of chemically individual BS in plant material using hormones of 28-homo-series (28-homocastasterone and 28homobrassinolide) as an example. These BS are among the most physiologically active representatives of steroid plant hormones and, at the same time, they are very promising for medicinal and agricultural applications, which nowadays are just in the very early development stage. Apart from the elaborated in this study IA1, two other immunoassay systems (IA2 [21] and IA3 [22]) were used for analysis of the plant material. The antibodies of the test system IA2 specifically bind to steroids containing 2a,3a-diol,22R,23R-diol and 7-membered 7-oxalactone B-ring (B-lactone-BS), and they have low (2% to 8%) cross-reactivity with BS of 6-keto-series. The 28-norbrassinolide calibration curve for IA2 was constructed as reported earlier [21]. The assay IA3 was designed to measure steroidal compounds bearing 2a,3a-diol,22R,23R-diol and 24S-ethyl groups. It has low cross-reactivity with BS of 24S-methyl- (9–14%) and 24R-methyl-
Fig. 3. Standard curves for 6-keto-BS ELISA (IA1, Method 1 () and Method 2 (N)).
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
5
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx Table 1 Accuracy data obtained for 6-keto-BS ELISA. Measured amount of 6-keto-BS (nmol/L)
Added amount of 28-norcastasterone (nmol/L)
Mean amount recovered (nmol/L)
Mean recovery (%)
Method 1 4.8 17.0 35.0 Overall mean ± 1SD
10 10 10
15 26 46
101.4 96.3 102.2 100.0 ± 3.2
Method 2 0.05 0.14 0.44 Overall mean ± 1SD
0.1 0.1 0.1
0.17 0.23 0.52
113.5 96.8 95.4 101.9 ± 10.1
Table 2 Parallelism studies for the 6-keto-BS ELISA. Sample No. Calculated value Method 1 Neat value Dilution factor 2 4 8 Mean ±1SD % C.V. Method 2 Neat value Dilution factor 2 4 8 Mean ±1SD % C.V.
1 nmol/L
2 nmol/L
3 nmol/L
13.0
29.0
22.0
11.6 12.8 14.4 13.0 1.2 8.9
32.0 28.0 28.0 29.3 1.9 6.5
20.0 24.0 23.2 22.3 1.7 7.8
0.32
0.40
0.80
0.36 0.28 0.38 0.34 0.04 13.20
0.36 0.50 0.42 0.42 0.06 14.00
0.89 0.85 0.75 0.82 0.06 7.40
Table 3 Imprecision data for the 6-keto-BS ELISA. Mean (nmol/L)
±1SD
C.V. (%)
Method 1 4.8 17 35
0.21 1.22 2.24
4.38 7.18 6.40
Method 2 0.05 0.14 0.44
0.007 0.014 0.055
14.50 9.65 12.60
(2–3%) series, but shows 100% cross reactivity with 28-homobrassinolide 3 and 28-homocastasterone 7. The 28-homobrassinolide calibration curve for IA3 was constructed in the concentration range of 1–100 nmol/L for Method 1, 0.01–5 nmol/L for Method 2 (Fig. 4). In both cases, analytical working ranges and LOD-parameters were similar to the corresponding values of IA1. At the same time, for the Method 1 they also coincided with previously described [22]. Homogenized dry flowers of horse gowan Flores Chamomillae were extracted and processed according to Variant C to give HPLC fractions containing a mixture of 28-homo-BS, which were collected, based on the retention times of 28-homobrassinolide 3 and 28-homocastasterone 7. Having in mind composition of the analyzed fraction which included only these two hormones, sum of the results of the assays IA1 and IA2 should be equal to that obtained for the IA3. Comparison of immune responses measured with the assays IA1, IA2 and IA3 are shown in Table 5. Indeed, sum of values of
3.99 ± 0.20 ng/g (obtained by IA1 and reflecting the level of 28homocastasterone 7 in the sample according to Variant C) and 20.35 ± 0.41 ng/g (acquired by IA2 and showing the level of 28homobrassinolide 3) were in a good agreement with the result 24.75 ± 1.06 ng/g found for the total level of 28-homo-BS 3 and 7 using assay IA3. At the same time, analysis by IA3 of the sample prepared according to Variant A gave total BS-amount of 61.25 ± 1.77 ng/g, and for Variant B – 37.34 ± 3.77 ng/g. To check accuracy of the obtained results, a more specific instrumental analysis was carried out using HPLC–MS. The major part of the HPLC-purified fraction (Variant C) containing 28-homo-BS was treated with an excess of 3-(dansylamino)phenylboronic acid and investigated by HPLC–MS. The obtained chromatogram (Fig. 5) was compared with a chromatogram of the authentic 28-homobrassinolide dansylaminophenylboronate (retention time 9.80 min, a peak of the protonated molecular ion at m/z 829 a.e. in the mass spectrum). According to this method, content of 28-homobrassinolide 3 was found to be 23.47 ± 3.20 ng/g that was comparable with the result 20.35 ± 0.41 ng/g showed by IA2 immunoassay with sample prepared according to Variant C. It should be noted that ELISA quantification of BS in plants requiring preliminary HPLC purification (Variant C), although being comparable in accuracy with HPLC–MS (Table 5, experiments 3, 4, 6) and having no preference in its laboriousness, nevertheless, can be considered as a practical one because it can be used as an alternative to HPLC–MS. Attempts to analyze plant extracts without such a pretreatment led to false positive responses (Table 5, experiments 1, 2, 7), probably due to the matrix effect [26]. To overcome these effects, we divided the analysis procedure into two steps (Method 2). In the first one, the diluted buffer extract (Variant D) was incubated in wells with antibodies to BS (BS–HRP conjugate had no contacts with sample). In the second step (after washing of wells), the BS–HRP conjugate was added and incubated. Subsequent procedures were carried out as described in Section 2.7.2. Dramatic increase in sensitivity for two-step procedure compared to one-step (see Figs. 3 and 4) could be explained by the fact that two-step method is eliminating contact between BS and HRP conjugate and the extract, so it is eliminating matrix effect. 28Homo-BS sum measured by Method 2 (22.50 ± 1.20 ng/g, Table 5, experiment 8) was much closer to that observed in HPLC–MS (23.47 ± 3.20 ng/g, Table 5, experiment 6) and in good agreement with data obtained in experiments 4 and 5 (Table 5). We also measured total amount of 6-keto-BS in the samples prepared according to Variant B (TLC, Method 1) and Variant D (no purification, Method 2). The obtained results were close for both measurements: 36.00 ± 1.88 ng/g for Method 1 (Table 5, experiment 9) and 38.33 ± 3.21 ng/g for Method 2 (Table 5, experiment 10) that showed applicability of Method 2 to measuring BS-content in plant material without pretreatment steps such as extractions and chromatographic purifications.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
6
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
Table 4 Cross reactivity of selected steroids with 28-norcastasterone antiserum. Structure
Compound
28-Norcastasterone, 8
Cross reactivity, % Method 1
Method 2
100
100
(22R,23R)-2a,3a,22,23,26-Pentahydroxy-5a-cholestan-6-one, 13
92.6
48.2
Castasterone, 5
83.0
129.4
24-Epicastasterone, 6
87.7
52.4
28-Homocastasterone, 7
85.0
67.2
(22S,23S)-24-Epicastasterone, 14
1.2
<0.01
(22R,23R)-22,23-Dihydroxy-3a,5-cyclo-5a-stigmast-6-one, 15
<0.01
<0.01
(22S,23S)-22,23- Dihydroxy-3a,5-cyclo-5a-stigmast-6-one, 16
<0.01
<0.01
(22R,23R)-22,23-Dihydroxystigmast-2-en-6-one, 17
0.19
<0.01
Brassinolide, 1
1.6
1.6
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
7
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx Table 4 (continued) Structure
Compound
Cross reactivity, % Method 1
Method 2
24-Epibrassinolide, 2
1.1
0.02
28-Homobrassinolide, 3
0.3
0.43
(22S,23S)-24-Epibrassinolide, 18
0.6
0.24
Ecdysterone, 19
<0.1
<0.01
Cholesterol, 20
<0.01
<0.01
Androstenolone, 21
<0.01
<0.01
Fig. 4. Standard curves for 28-homo-BS ELISA (IA3, Method 1 (N) and Method 2 (j)).
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
8
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
Table 5 Level of 28-homo-BS in Flores Chamomillae.
a
Experiment number
Assaya
Variant of sample preparation
Sample composition
Method of analysis
Measured result, ng/g DW
Measured hormones
1 2 3 4 5 6 7 8 9 10
IA3 IA3 IA3 IA2 IA1 HPLC–MS IA3 IA3 IA1 IA1
A B C C C C D D B D
Crude BS 3+7 3+7 3+7 3+7 Crude Crude BS Crude
1 1 1 1 1 HPLC–MS 1 2 1 2
61.25 ± 1.77 37.34 ± 3.77 24.75 ± 1.06 20.35 ± 0.41 3.99 ± 0.20 23.47 ± 3.20 150.70 ± 2.07 22.50 ± 1.20 36.00 ± 1.88 38.33 ± 3.21
28-homo-BS 28-homo-BS 3+7 3 7 3 28-homo-BS 28-homo-BS 6-keto-BS 6-keto-BS
extract
extract extract extract
IA1: 6-keto-BS ELISA; IA2: 6-oxo-7-oxa-BS ELISA [21]; IA3: 28-homo-BS ELISA [22].
Fig. 5. HPLC–MS chromatograms of dansyl-3-aminophenylboronates of authentic 28-homobrassinolide (A) and 28-homobrassinolide from plant tissue (B).
4. Conclusion The new ELISA for measurement of 6-keto-BS (IA1 test system) was elaborated. The combination of this immunoenzyme assay with previously developed test systems for B-lactone-BS (IA2) and 28-homo-BS (IA3) enabled measuring the content of individual BS. The rapid and sensitive method for BS immunoassaying that does not require preliminary sample pretreatment steps (such as extraction and chromatography) was developed. An application of this method considerably facilitated the determination of brassinosteroid phytohormones in plant material. Acknowledgments The authors are grateful to Dr. Mark Kushnir and Dr. Igor Pavlov (ARUP Laboratories, Salt Lake City, UT, USA) for fruitful discussion and linguistic help. We are indebted to the Belarusian Foundation for
Fundamental Research for financial support (Project X14P-139). Finally, we thank the anonymous reviewers for their helpful and constructive comments that greatly contributed to improving this manuscript.
References [1] Khripach VA, Zhabinskii VN, de Groot A. Brassinosteroids. A new class of plant hormones. San Diego: Academic Press; 1999. [2] Bajguz A. Brassinosteroids – occurrence and chemical structures in plants. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 1–27. [3] Divi UK, Krishna P. Brassinosteroid: a biotechnological target for enhancing crop yield and stress tolerance. New Biotechnol 2009;26:131–6. [4] Khripach V, Zhabinskii V, de Groot A. Twenty years of brassinosteroids: steroidal plant hormones warrant better crops for the XXI century. Ann Bot 2000;86:441–7. [5] Thompson MJ, Meudt WJ, Mandava NB, Dutky SR, Lusby WR, Spaulding DW. Synthesis of brassinosteroids and relationship of structure to plant growthpromoting effects. Steroids 1982;39:89–105. [6] Takatsuto S, Yazawa N, Ikekawa N, Morishita T, Abe H. Synthesis of (24R)-28homobrassinolide analogs and structure activity relationships of
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
brassinosteroids in the rice-lamina inclination test. Phytochemistry 1983;22:1393–7. Takatsuto S, Yazawa N, Ikekawa N, Takematsu T, Takeuchi Y, Koguchi M. Structure activity relationship of brassinosteroids. Phytochemistry 1983;22:2437–41. Takatsuto S, Ikekawa N, Morishita T, Abe H. Structure activity relationship of brassinosteroids with respect to the A/B-ring functional groups. Chem Pharm Bull 1987;35:211–6. Kang YY, Guo SR. Role of brassinosteroids on horticultural crops. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 269–88. Khripach VA, Zhabinskii VN, Khripach NB. New practical aspects of brassinosteroids and results of their ten-year agricultural use in Russia and Belarus. In: Hayat S, Ahmad A, editors. Brassinosteroids. Dordrecht: Kluwer Academic Publishers; 2003. p. 189–230. Hategan L, Godza B, Szekeres M. Regulation of brassinosteroid metabolism. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 57–81. Khripach VA, Zhabinskii VN, Litvinovskaya RP. Immunoassays of brassinosteroids. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 375–92. Khripach VA, Sviridov OV, Pryadko AG, Litvinovskaya RP, Drach SV, Matveentsev VD, et al. Immunoenzymatic analysis of (24R)-brassinosteroids. Russ J Bioorg Chem 2007;33:347–53. Khripach VA, Litvinovskaya RP, Drach SV, Averkova MA, Zhabinskii VN, Sviridov OV, et al. Immunoenzymatic analysis of (24S)methylbrassinosteroids. Doklady NAN Belarusi 2009;53:74–7. Yokota T, Watanabe S, Ogino Y, Yamaguchi I, Takahashi N. Radioimmunoassay for brassinosteroids and its use for comparative-analysis of brassinosteroids in stems and seeds of Phaseolus vulgaris. J Plant Growth Regul 1990;9:151–9.
9
[16] Takatsuto S. Microanalysis of naturally occurring brassinosteroids. ACS Symp Ser 1991;474:107–20. [17] Taylor PE, Spuck K, Smith PM, Sasse JM, Yokota T, Griffiths PG, et al. Detection of brassinosteroids in pollen of Lolium perenne L. by immunocytochemistry. Planta 1993;189:91–100. [18] Swaczynová J, Šíša M, Hnilicˇková J, Kohout L, Strnad M. Synthesis, biological, immunological and anticancer properties of a new brassinosteroid ligand. Pol J Chem 2006;80:629–35. [19] Swaczynová J, Novák O, Hauserová E, Fuksová K, Šíša M, Kohout L, et al. New techniques for the estimation of naturally occurring brassinosteroids. J Plant Growth Regul 2007;26:1–14. [20] Khripach VA, Sviridov OV, Litvinovskaya RP, Pryadko AG, Drach SV, Zhabinskii VN. Analysis of brassinosteroids. Pol J Chem 2006;80:651–4. [21] Khripach V, Zhabinskii V, Antonchick A, Litvinovskaya R, Drach S, Sviridov O, et al. A new type of modified brassinosteroids for enzyme-linked immunosorbent assay. Nat Prod Commun 2008;3:735–48. [22] Khripach VA, Litvinovskaya RP, Raiman ME, Drach SV, Zhabinskii VN, Sviridov OV, et al. Synthesis and immunochemical determination of 28homobrassinosteroids. Vesti NAN Belarusi, ser khim navuk 2008:47–58. [23] Law B. Immunoassay. A practical guide. Taylor & Francis; 2005. p. 179–82. [24] Chard T. An introduction to radioimmunoassay and related techniques. Elsevier/North-Holland Biomedical Press; 1978. [25] Svatos A, Antonchick A, Schneider B. Determination of brassinosteroids in the sub-femtomolar range using dansyl-3-aminophenylboronate derivatization and electrospray mass spectrometry. Rapid Commun Mass Spectrom 2004;18:818–23. [26] Plimmer JR, Gammon DW, Ragsdale NN. Encyclopedia of agrochemicals. John Wiley & Sons, Inc.; 2003.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
A new ELISA for quantification of brassinosteroids in plants Andrey G. Pradko b, Raisa P. Litvinovskaya a, Alina L. Sauchuk a, Svetlana V. Drach a, Alexander V. Baranovsky a, Vladimir N. Zhabinskii a, Tatyana V. Mirantsova b, Vladimir A. Khripach a,⇑ a b
Institute of Bioorganic Chemistry, National Academy of Sciences of Belarus, Kuprevich St., 5/2, 220141 Minsk, Belarus Hemma-Test Ltd., Masherova St., 11, 220029 Minsk, Belarus
a r t i c l e
i n f o
Article history: Received 12 August 2013 Received in revised form 1 August 2014 Accepted 22 August 2014 Available online xxxx Keywords: Brassinosteroids Phytohormones Synthesis ELISA HPLC Flores Chamomillae
a b s t r a c t Starting from (22R,23R)-2a,3a,22,23,26-pentahydroxy-5a-cholestan-6-one 26-hemisuccinate, conjugates of 28-norcastasterone with horse radish peroxidase and bovine serum albumin were prepared. The latter conjugate was injected into rabbits; produced polyclonal antibodies were used to quantitate 6-keto-brassinosteroids. The newly developed analytical system was used in combination with two other immunoenzymatic assays for brassinosteroids to determine individual compounds of this series. In addition, a direct method of brassinosteroid analysis was proposed. It has the advantage of requiring no sample pretreatment steps such as extraction with organic solvents and chromatography. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Brassinosteroids (BS) are low molecular bioregulators playing an important role in plant growth and development [1]. Content of natural BS in plant sources is very low (less than 105%) [2], and to achieve notable biological effects from their application, only a minute amount of exogenous BS is required. A pronounced plant growth stimulating and adaptogenic effect was observed on BS treatment of various plant species at a dose 5–20 mg/ha resulting in a positive effect on the quantity and quality of crops and tolerance against various biotic and abiotic stresses [3]. These features of BS have made them attractive as active ingredients for agrochemicals [4]. The most active and the most important BS from both scientific and practical point of view are those containing 6-oxo-7-oxa- or a 6-ketone function in the B-cycle [5–8] (Fig. 1). 24-Epibrassinolide 2 and 28-homobrassinolide 3 bearing a lactone moiety in the B-cycle have already found a practical use in agriculture [9,10].
Abbreviations: BS, brassinosteroids; BSA, bovine serum albumin; C.V., coefficient of variation; LOD, limit of detection; DCC, N,N0 -dicyclohexylcarbodiimide; DW, dry weight; HRP, horse radish peroxidase; NHS, N-hydroxysuccinimide; PBS, phosphate buffered saline; SD, standard deviation; SIM, selective ion monitoring; TMB, 3,30 ,5,50 -tetramethylbenzidine; Tris, tris(hydroxymethyl)aminomethane. ⇑ Corresponding author. Tel./fax: +375 172 678 647. E-mail address: [email protected] (V.A. Khripach).
It is accepted that compounds 5–8, which belong to 6-keto-BS subgroup, are not only biosynthetic precursors of the corresponding lactones 1–4, but they also play an independent role in plant growth and development [11]. There are reasons to believe that these highly active and easily accessible BS could also find their practical applications in agriculture. One of the preconditions to achieve their massive use is a simple, selective and highly sensitive method for their analysis. Based on our and others experience in BS research [12–20], the ELISA could be considered as a method of choice for the routine analyses. In this respect, the first task of the present investigation was to develop an ELISA for 6-keto-BS. A certain disadvantage of immunochemical methods is that they are group specific ones and cannot be applied for analysis of an individual compound. The obvious solution of the problem is a combined application of several immunoassays separately quantifying the group components and/or immunoassaying of chromatographic fractions. In this connection, the present study proposes quantification of individual BS of 28-homo-series (28-homocastasterone and 28-homobrassinolide) by a joint utilization of the assay for 6-keto-BS (IA1) elaborated in this study with previously developed test systems for B-lactone-BS (IA2) [21] and 28-homo-BS (IA3) [22] (Fig. 2). Although each of these three assays is a group specific, their groups are different, and this allows differentiation of the hormones that are interfering when the only one group is analyzed with a single ELISA.
http://dx.doi.org/10.1016/j.steroids.2014.08.022 0039-128X/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
2
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
Fig. 1. Structures of the most important natural brassinosteroids.
2. Experimental 2.1. Reagents and materials The chemicals and materials used in this study were purchased from following sources: DCC, NHS, HRP, dioxane, and Sephadex G-25 from Sigma–Aldrich (USA); BSA from Acros Organics (Germany); chloroform, cyclohexane from ECOS-1 (Russia); complete Freund’s adjuvant from ICN Biochemicals (France); methanol from Fisher Scientific UK Ltd. (UK); silica gel from Macherey–Nagel GmbH & Co KG (Germany). High binding capacity 96 flat-bottom wells microtiter plates (Maxisorp™) used acquired from Nalgene Nunc (Thermo Fisher Scientific, Inc., USA). The following buffers were used for the immunoassay: coating buffer 1 (0.05 M PBS, pH 7.4, containing 0.9% NaCl), coating buffer 2 (0.025 M PBS, pH 7.4, containing 0.1% BSA, 0.02% Tween™ 20), blocking buffer (0.025 PBS, pH 7.4, containing 0.1% BSA, 2% sucrose, 5% sorbitol), assay buffer (0.05 M Tris, pH 7.4, containing 0.9% NaCl, 0.1% BSA, 0.02% Tween™ 20) and washing buffer (1% NaCl, containing 0.02% Tween™ 20). The ready solution of TMB in the substrate buffer containing H2O2 (Enhanced K-Blue TMB Substrate, Neogen, USA) was used for staining in assay, 5% H2SO4 was used as a stopping reagent for quenching the color reaction. BS and their derivatives were synthesized in the Laboratory of Steroid Chemistry (Institute of Bioorganic Chemistry, NAS of Belarus); other steroids were purchased from Steraloids Inc. (USA). 2.2. HPLC Preparative high-performance liquid chromatography was carried out on a HPLC–Hitachi (UV-VIS Detector L-4250, Intelligent Pump L-6200A) with reversed phase using column RP18 (LiChroCART 250 mm 10 mm 10 lm). 70% Aqueous acetonitrile was used as mobile phase with flow rate 1.5 mL/min. UV detection was performed at a wavelength of 204 nm. 2.3. HPLC–MS The HPLC–MS experiments were carried out on a Accela chromatograph (Thermo Electron Corp., USA) linked to a LCQ Fleet 3D ion trap mass spectrometer equipped with an electrospray
interface (Thermo Fisher Scientific, USA) with reversed-phase column C18 HYPERSIL Gold (50 mm 2.1 mm 1.9 lm). The column thermostat was set at 25 °C and 75% acetonitrile in water was used as a mobile phase for 15 min at a flow rate of 100 lL/min. The column was equilibrated to initial conditions for 5 min. The ESI-MS parameters (positive mode) were as follows: spray voltage (4.5 kV), capillary temperature (325 °C), and capillary voltage 36 V. Nitrogen was used as sheath, auxiliary and sweep gas. The sheath, auxiliary and sweep gas flow rates were 30, 0 and 0 arbitrary units, respectively. Data were processed by Xcalibur software (version 2.0.6, Thermo Electron Corp., USA).
2.4. Synthesis of the immunogenic conjugate of 28-norcastasterone with BSA A solution of (22R,23R)-2a,3a,22,23-tetrahydroxy-5a-cholestan-6-one 26-hemisuccinate 9 (60 mg, 107 lmol, prepared according to [21]) and NHS (16 mg, 139 lmol) in anhydrous dioxane (10 mL) was added sequentially to a solution of DCC (26 mg, 126 lmol) in anhydrous dioxane (10 mL) at 8–10 °C. The reaction mixture was stirred for 30 min at this temperature and then for 6 h at room temperature. The precipitated dicyclohexylurea was removed by filtration, and obtained solution of ester 10 was added to a solution of BSA (120 mg in 20 mL of 0.1 M NaHCO3, pH 8.35). The mixture was kept for 20 h at room temperature. Then an excess of activated ester and dioxane was removed by dialysis against 0.05 M NaCl to give, after lyophilization, 160 mg of the conjugate 11, which was frozen and stored at 18 °C.
2.5. Synthesis of labeled antigen A solution of active N-hydroxysuccinimide ester 10 (4.5 mg, 6.6 lmol), prepared as described above, was added to a solution of HRP (9 mg) in distilled water (900 lL) with 0.1 M NaHCO3 (2– 3 drops, pH 8.35). The reaction mixture was stirred at 10 °C for 1.5 h and purified on a column with Sephadex eluting with distilled water. The yellow–brown fraction containing conjugate 12 was collected, diluted with glycerol (1:2), and stored in a freezer at 18 °C.
Fig. 2. Immunodominant fragments of BS-molecules specific for recognition by IA1, IA2, and IA3.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
2.6. Antisera production A group of eight rabbits was immunized with the conjugate of 28-norcastasterone with BSA 11. Every rabbit was subcutaneously injected into 4–6 points on its back with 1 mg conjugate, dissolved in a PBS (0.5 mL, pH 7.4) and emulsified in the equal volume of the complete Freund’s adjuvant. Intervals between the injections were 4–4.5 weeks. The immunization was continued for 5 months, and blood from the auricular vein was taken at regular intervals. The taken samples of blood serum were tested for the binding ability for 28-norcastasterone 8, and their titers (working dilutions) were determined. The antiserum A1 with the highest titer and sensitivity in the assay was used in this study. 2.7. Analysis of BS in plants by ELISA 2.7.1. Preparation of samples 2.7.1.1. Variant A. Dry flowers of horse gowan Flores Chamomillae (3 g) (bought at the drug store) were homogenized and extracted with methanol (30 mL) for 24 h at room temperature. The extract was filtered on a sintered glass filter; an aliquot (1 mL) was evaporated, dissolved in assay buffer (0.5 mL) and used for analysis. The rest of methanol extract was used in Variant B. 2.7.1.2. Variant B. The methanol extract from Variant A was evaporated and the residue was partitioned between cyclohexane (15 mL) and 80% aqueous MeOH (3 10 mL). The combined 80% aqueous MeOH extract was evaporated to dryness; the residue was dissolved in MeOH (1.5 mL) and subjected to TLC (20 20 cm) on silica gel, using a mixture of chloroform–methanol (88:12) as eluent. The BS-containing zone (Rf 0.4–0.6) was collected and eluted with MeOH, solvent was evaporated. The residue was dissolved in MeOH (3 ml) and an aliquote of 100 lL was evaporated, dissolved in assay buffer (0.5 mL) and used for analysis by ELISA. The rest of methanol solution was evaporated and the obtained sample was used in Variant C. 2.7.1.3. Variant C. The sample from Variant B was dissolved in 300 lL MeOH and separated by reversed-phase HPLC, using 70% aqueous MeCN as mobile phase with flow rate 1.5 mL/min. The solvents were degassed before using. Based on the retention times of 28-homobrassinolide (15 min) and 28-homocastasterone (16 min), the fraction (Rt 14–17 min) was collected and evaporated. The residue was dissolved in MeOH (3 mL) and an aliquote of 100 lL was evaporated, dissolved in assay buffer (0.5 mL) and used for analysis by ELISA. The remaining part of the collected fraction was evaporated and used for analysis by HPLC–MS.
3
and incubated at 4 °C overnight. The blocking solution was removed from the wells; the plates were dried at room temperature. The calibrating samples (0–100 nM) or the test samples of BS (50 lL in duplicates) and the HRP-conjugate (100 lL per well) were added to the wells of the plate with the immobilized antibodies. The plates were incubated for 1 h at 37 °C, the liquid was removed from the wells, and the plates were washed 4 times with 150 lL of washing buffer. The TMB solution was added to the wells (150 lL) and incubated for 15 min at 37 °C. The reaction was quenched by the addition of stopping reagent (50 lL per well), and optical absorbance was measured at 450 nm with a plate reader (F300TP, ‘‘VITYAS’’, Belarus). The analytical results were calculated by interpolation from the calibration curve and values were expressed in nmol/L or ng/g. Sigmoid curves for standards were linearized by the log–logit transformation: logit B/B0 = ln((B/B0)/ (100 B/B0)) [23,24]. 2.7.2.2. Method 2. The calibrating samples or the test samples of BS (150 lL per well) were placed in wells of the plate with the immobilized antibodies. The plates were incubated for 2 h at 37 °C, liquid was removed from the wells, and the plates were washed 4 times with 150 lL of washing buffer. The HRP-conjugate (150 lL) was added to the wells and incubated for 5 min at 37 °C. Then the liquid was removed from the wells, and the plates were washed as described above. The TMB solution in the substrate buffer (150 lL) was added to the wells and incubated for 20 min at 37 °C. The reaction was quenched by the addition of stopping reagent (50 lL in each well), and optical absorbance was measured at 450 nm. The analytical results were calculated by interpolation from the calibration curve as above. Validation procedures were carried out according to the methods previously published [23]. Accuracy was determined by adding known amounts of steroid standard dose (10 nmol/L for Method 1, 0.1 nmol/L for Method 2) to model samples. All samples were assayed by 6-keto-BS ELISA (IA1). Imprecision was estimated by replicate determinations of BS levels in model samples. To evaluate assay parallelism, serial dilutions of 3 model samples prepared according to Variant B were tested for Method 1 and samples prepared according to Variant D were tested for Method 2. 2.8. Analysis of BS in plants by HPLC–MS Calibrators and test samples in 0.2 ml glass inserts were derivatized by heating to 60 °C for 30 min with a solution of
2.7.1.4. Variant D. Dry flowers of horse gowan Flores Chamomillae (1 g) were homogenized in 20 mL of assay buffer. The buffer extract was centrifuged (3000 rpm, 20 min); the supernatant diluted with assay buffer (1:50) was used for analysis. 2.7.2. ELISA procedure ELISA test-systems IA1, IA2 [21] and IA3 [22] were used for determination of 6-keto-, B-lactone- and 28-homo-BS, respectively. 2.7.2.1. Method 1. Microtiter plates for ELISA were prepared as follows. The microtiter plates were coated with the sheep antibodies to rabbit IgG (2 lg/mL, 150 lL per well) in coating buffer 1. The plates were covered with parafilm to prevent evaporation. After overnight incubation at 4 °C, the coating solution was removed from the wells without washing. Then the plates were coated with the antiserum A1 (150 lL per well) in coating buffer 2 and incubated at 4 °C overnight. The coating solution was removed from the wells and blocking buffer was added to the wells (200 lL)
Scheme 1. Synthesis of immunogenic conjugate 11 and labeled antigen 12.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
4
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
3-(dansylamino)phenylboronic acid (20 lL, 5 mg/mL) in a mixture of pyridine and acetonitrile (1:19) [25]. The reaction mixture was then diluted ten-fold with acetonitrile and investigated by HPLC–MS without further purification. The volume of injected sample was 5 lL, the retention time was determined by the standard sample (9.80 min for authentic 28-homobrassinolide dansylaminophenylboronate). Quantification was performed by SIM of the molecular-related ions [M+H]+ (m/z 829). The samples were analyzed immediately after preparation. 3. Results and discussion The hemisuccinate 9 [21] was used as a starting material for the preparation of conjugates with proteins (Scheme 1). The conjugate 11 of 28-norcastasterone with BSA was obtained by the interaction of the activated N-hydroxysuccinimide ester 10 with BSA in aqueous dioxane solution followed by dialysis and lyophilization. The reaction of the ester 10 with a HRP solution in the presence of potassium bicarbonate gave, after chromatographic purification of the product on Sephadex, the conjugate 12, which was used as a labeled antigen. The 6-keto-BS specific antibodies were produced by injecting the antigen 11 into rabbits. Titers were determined to evaluate antiserum binding ability for 28-norcastasterone 8. The experiments were carried out with the antiserum A1. Its dilution 1:500,000 and dilution of HRP-conjugate 1:10,000 were found to be the best combination of immunoreagents for both Method 1 and Method 2 ELISA. The 28-norcastasterone calibration curve for IA1 was constructed in the concentration range of 1–100 nmol/L for Method 1, 0.01–5 nmol/L for Method 2 (Fig. 3). An analytical working range of the assay for Method 1 is 1–100 nmol/L (part of the curve between 10% and 60%), and for Method 2 is 0.01–1 nmol/L (part of the curve between 30% and 90%) for Method 2. The limit of detection (LOD) for Method 1 was found to be 0.3 nmol/L, for Method 2 – 0.01 nmol/L. The LOD observed in this study for Method 1 was close to those reported by us earlier [13,14,22]. The only other example of BS ELISA [18,19] based on rabbit antisera against 24-epicastasterone 6 showed LOD value of 1.5 nmol/L. Accuracy studies, parallelism and imprecision data were determined and all were found to be satisfactory. The recovery of added
28-norcastasterone ranged from 95% to 114% as shown in Table 1, with range of values being narrower for Method 1. The results of serial dilution of 3 model samples are shown in Table 2. They varied from 6% to 9% for Method 1 and from 7% to 14% for Method 2. The results of assay imprecision for 3 model samples are shown in Table 3. They ranged from 4% to 7% for Method 1 and from 10% to 15% for Method 2. Although Method 2 showed rather high coefficient of variation for both parallelism and imprecision experiments, nevertheless, taking into account low concentration level of its operating, it could be considered as an acceptable one. The specificity of 28-norcastasterone antiserum A1 was determined by competitive binding assay with various steroids. As seen from Table 4, the obtained antibodies were highly specific for the determination of 6-keto-BS, but showed low cross-reactivity with the synthetic BS-analog, (22S,23S)-24-epicastasterone 14, having non-natural stereochemistry of the diol group in the side chain. There was no cross reactivity detected with non-BS steroids like cholesterol 20, androstenolone 21, or synthetic intermediates 15–17. Only a small cross reactivity (0.3–1.6%) was observed for B-lactone-BS 1–3, 18. Having in hands this new immunoassay and earlier developed ELISAs for other important groups of natural BS, we could proceed further with analysis of chemically individual BS in plant material using hormones of 28-homo-series (28-homocastasterone and 28homobrassinolide) as an example. These BS are among the most physiologically active representatives of steroid plant hormones and, at the same time, they are very promising for medicinal and agricultural applications, which nowadays are just in the very early development stage. Apart from the elaborated in this study IA1, two other immunoassay systems (IA2 [21] and IA3 [22]) were used for analysis of the plant material. The antibodies of the test system IA2 specifically bind to steroids containing 2a,3a-diol,22R,23R-diol and 7-membered 7-oxalactone B-ring (B-lactone-BS), and they have low (2% to 8%) cross-reactivity with BS of 6-keto-series. The 28-norbrassinolide calibration curve for IA2 was constructed as reported earlier [21]. The assay IA3 was designed to measure steroidal compounds bearing 2a,3a-diol,22R,23R-diol and 24S-ethyl groups. It has low cross-reactivity with BS of 24S-methyl- (9–14%) and 24R-methyl-
Fig. 3. Standard curves for 6-keto-BS ELISA (IA1, Method 1 () and Method 2 (N)).
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
5
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx Table 1 Accuracy data obtained for 6-keto-BS ELISA. Measured amount of 6-keto-BS (nmol/L)
Added amount of 28-norcastasterone (nmol/L)
Mean amount recovered (nmol/L)
Mean recovery (%)
Method 1 4.8 17.0 35.0 Overall mean ± 1SD
10 10 10
15 26 46
101.4 96.3 102.2 100.0 ± 3.2
Method 2 0.05 0.14 0.44 Overall mean ± 1SD
0.1 0.1 0.1
0.17 0.23 0.52
113.5 96.8 95.4 101.9 ± 10.1
Table 2 Parallelism studies for the 6-keto-BS ELISA. Sample No. Calculated value Method 1 Neat value Dilution factor 2 4 8 Mean ±1SD % C.V. Method 2 Neat value Dilution factor 2 4 8 Mean ±1SD % C.V.
1 nmol/L
2 nmol/L
3 nmol/L
13.0
29.0
22.0
11.6 12.8 14.4 13.0 1.2 8.9
32.0 28.0 28.0 29.3 1.9 6.5
20.0 24.0 23.2 22.3 1.7 7.8
0.32
0.40
0.80
0.36 0.28 0.38 0.34 0.04 13.20
0.36 0.50 0.42 0.42 0.06 14.00
0.89 0.85 0.75 0.82 0.06 7.40
Table 3 Imprecision data for the 6-keto-BS ELISA. Mean (nmol/L)
±1SD
C.V. (%)
Method 1 4.8 17 35
0.21 1.22 2.24
4.38 7.18 6.40
Method 2 0.05 0.14 0.44
0.007 0.014 0.055
14.50 9.65 12.60
(2–3%) series, but shows 100% cross reactivity with 28-homobrassinolide 3 and 28-homocastasterone 7. The 28-homobrassinolide calibration curve for IA3 was constructed in the concentration range of 1–100 nmol/L for Method 1, 0.01–5 nmol/L for Method 2 (Fig. 4). In both cases, analytical working ranges and LOD-parameters were similar to the corresponding values of IA1. At the same time, for the Method 1 they also coincided with previously described [22]. Homogenized dry flowers of horse gowan Flores Chamomillae were extracted and processed according to Variant C to give HPLC fractions containing a mixture of 28-homo-BS, which were collected, based on the retention times of 28-homobrassinolide 3 and 28-homocastasterone 7. Having in mind composition of the analyzed fraction which included only these two hormones, sum of the results of the assays IA1 and IA2 should be equal to that obtained for the IA3. Comparison of immune responses measured with the assays IA1, IA2 and IA3 are shown in Table 5. Indeed, sum of values of
3.99 ± 0.20 ng/g (obtained by IA1 and reflecting the level of 28homocastasterone 7 in the sample according to Variant C) and 20.35 ± 0.41 ng/g (acquired by IA2 and showing the level of 28homobrassinolide 3) were in a good agreement with the result 24.75 ± 1.06 ng/g found for the total level of 28-homo-BS 3 and 7 using assay IA3. At the same time, analysis by IA3 of the sample prepared according to Variant A gave total BS-amount of 61.25 ± 1.77 ng/g, and for Variant B – 37.34 ± 3.77 ng/g. To check accuracy of the obtained results, a more specific instrumental analysis was carried out using HPLC–MS. The major part of the HPLC-purified fraction (Variant C) containing 28-homo-BS was treated with an excess of 3-(dansylamino)phenylboronic acid and investigated by HPLC–MS. The obtained chromatogram (Fig. 5) was compared with a chromatogram of the authentic 28-homobrassinolide dansylaminophenylboronate (retention time 9.80 min, a peak of the protonated molecular ion at m/z 829 a.e. in the mass spectrum). According to this method, content of 28-homobrassinolide 3 was found to be 23.47 ± 3.20 ng/g that was comparable with the result 20.35 ± 0.41 ng/g showed by IA2 immunoassay with sample prepared according to Variant C. It should be noted that ELISA quantification of BS in plants requiring preliminary HPLC purification (Variant C), although being comparable in accuracy with HPLC–MS (Table 5, experiments 3, 4, 6) and having no preference in its laboriousness, nevertheless, can be considered as a practical one because it can be used as an alternative to HPLC–MS. Attempts to analyze plant extracts without such a pretreatment led to false positive responses (Table 5, experiments 1, 2, 7), probably due to the matrix effect [26]. To overcome these effects, we divided the analysis procedure into two steps (Method 2). In the first one, the diluted buffer extract (Variant D) was incubated in wells with antibodies to BS (BS–HRP conjugate had no contacts with sample). In the second step (after washing of wells), the BS–HRP conjugate was added and incubated. Subsequent procedures were carried out as described in Section 2.7.2. Dramatic increase in sensitivity for two-step procedure compared to one-step (see Figs. 3 and 4) could be explained by the fact that two-step method is eliminating contact between BS and HRP conjugate and the extract, so it is eliminating matrix effect. 28Homo-BS sum measured by Method 2 (22.50 ± 1.20 ng/g, Table 5, experiment 8) was much closer to that observed in HPLC–MS (23.47 ± 3.20 ng/g, Table 5, experiment 6) and in good agreement with data obtained in experiments 4 and 5 (Table 5). We also measured total amount of 6-keto-BS in the samples prepared according to Variant B (TLC, Method 1) and Variant D (no purification, Method 2). The obtained results were close for both measurements: 36.00 ± 1.88 ng/g for Method 1 (Table 5, experiment 9) and 38.33 ± 3.21 ng/g for Method 2 (Table 5, experiment 10) that showed applicability of Method 2 to measuring BS-content in plant material without pretreatment steps such as extractions and chromatographic purifications.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
6
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
Table 4 Cross reactivity of selected steroids with 28-norcastasterone antiserum. Structure
Compound
28-Norcastasterone, 8
Cross reactivity, % Method 1
Method 2
100
100
(22R,23R)-2a,3a,22,23,26-Pentahydroxy-5a-cholestan-6-one, 13
92.6
48.2
Castasterone, 5
83.0
129.4
24-Epicastasterone, 6
87.7
52.4
28-Homocastasterone, 7
85.0
67.2
(22S,23S)-24-Epicastasterone, 14
1.2
<0.01
(22R,23R)-22,23-Dihydroxy-3a,5-cyclo-5a-stigmast-6-one, 15
<0.01
<0.01
(22S,23S)-22,23- Dihydroxy-3a,5-cyclo-5a-stigmast-6-one, 16
<0.01
<0.01
(22R,23R)-22,23-Dihydroxystigmast-2-en-6-one, 17
0.19
<0.01
Brassinolide, 1
1.6
1.6
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
7
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx Table 4 (continued) Structure
Compound
Cross reactivity, % Method 1
Method 2
24-Epibrassinolide, 2
1.1
0.02
28-Homobrassinolide, 3
0.3
0.43
(22S,23S)-24-Epibrassinolide, 18
0.6
0.24
Ecdysterone, 19
<0.1
<0.01
Cholesterol, 20
<0.01
<0.01
Androstenolone, 21
<0.01
<0.01
Fig. 4. Standard curves for 28-homo-BS ELISA (IA3, Method 1 (N) and Method 2 (j)).
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
8
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
Table 5 Level of 28-homo-BS in Flores Chamomillae.
a
Experiment number
Assaya
Variant of sample preparation
Sample composition
Method of analysis
Measured result, ng/g DW
Measured hormones
1 2 3 4 5 6 7 8 9 10
IA3 IA3 IA3 IA2 IA1 HPLC–MS IA3 IA3 IA1 IA1
A B C C C C D D B D
Crude BS 3+7 3+7 3+7 3+7 Crude Crude BS Crude
1 1 1 1 1 HPLC–MS 1 2 1 2
61.25 ± 1.77 37.34 ± 3.77 24.75 ± 1.06 20.35 ± 0.41 3.99 ± 0.20 23.47 ± 3.20 150.70 ± 2.07 22.50 ± 1.20 36.00 ± 1.88 38.33 ± 3.21
28-homo-BS 28-homo-BS 3+7 3 7 3 28-homo-BS 28-homo-BS 6-keto-BS 6-keto-BS
extract
extract extract extract
IA1: 6-keto-BS ELISA; IA2: 6-oxo-7-oxa-BS ELISA [21]; IA3: 28-homo-BS ELISA [22].
Fig. 5. HPLC–MS chromatograms of dansyl-3-aminophenylboronates of authentic 28-homobrassinolide (A) and 28-homobrassinolide from plant tissue (B).
4. Conclusion The new ELISA for measurement of 6-keto-BS (IA1 test system) was elaborated. The combination of this immunoenzyme assay with previously developed test systems for B-lactone-BS (IA2) and 28-homo-BS (IA3) enabled measuring the content of individual BS. The rapid and sensitive method for BS immunoassaying that does not require preliminary sample pretreatment steps (such as extraction and chromatography) was developed. An application of this method considerably facilitated the determination of brassinosteroid phytohormones in plant material. Acknowledgments The authors are grateful to Dr. Mark Kushnir and Dr. Igor Pavlov (ARUP Laboratories, Salt Lake City, UT, USA) for fruitful discussion and linguistic help. We are indebted to the Belarusian Foundation for
Fundamental Research for financial support (Project X14P-139). Finally, we thank the anonymous reviewers for their helpful and constructive comments that greatly contributed to improving this manuscript.
References [1] Khripach VA, Zhabinskii VN, de Groot A. Brassinosteroids. A new class of plant hormones. San Diego: Academic Press; 1999. [2] Bajguz A. Brassinosteroids – occurrence and chemical structures in plants. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 1–27. [3] Divi UK, Krishna P. Brassinosteroid: a biotechnological target for enhancing crop yield and stress tolerance. New Biotechnol 2009;26:131–6. [4] Khripach V, Zhabinskii V, de Groot A. Twenty years of brassinosteroids: steroidal plant hormones warrant better crops for the XXI century. Ann Bot 2000;86:441–7. [5] Thompson MJ, Meudt WJ, Mandava NB, Dutky SR, Lusby WR, Spaulding DW. Synthesis of brassinosteroids and relationship of structure to plant growthpromoting effects. Steroids 1982;39:89–105. [6] Takatsuto S, Yazawa N, Ikekawa N, Morishita T, Abe H. Synthesis of (24R)-28homobrassinolide analogs and structure activity relationships of
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022
A.G. Pradko et al. / Steroids xxx (2014) xxx–xxx
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
brassinosteroids in the rice-lamina inclination test. Phytochemistry 1983;22:1393–7. Takatsuto S, Yazawa N, Ikekawa N, Takematsu T, Takeuchi Y, Koguchi M. Structure activity relationship of brassinosteroids. Phytochemistry 1983;22:2437–41. Takatsuto S, Ikekawa N, Morishita T, Abe H. Structure activity relationship of brassinosteroids with respect to the A/B-ring functional groups. Chem Pharm Bull 1987;35:211–6. Kang YY, Guo SR. Role of brassinosteroids on horticultural crops. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 269–88. Khripach VA, Zhabinskii VN, Khripach NB. New practical aspects of brassinosteroids and results of their ten-year agricultural use in Russia and Belarus. In: Hayat S, Ahmad A, editors. Brassinosteroids. Dordrecht: Kluwer Academic Publishers; 2003. p. 189–230. Hategan L, Godza B, Szekeres M. Regulation of brassinosteroid metabolism. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 57–81. Khripach VA, Zhabinskii VN, Litvinovskaya RP. Immunoassays of brassinosteroids. In: Hayat S, Ahmad A, editors. Brassinosteroids: a class of plant hormone. Dordrecht: Springer; 2011. p. 375–92. Khripach VA, Sviridov OV, Pryadko AG, Litvinovskaya RP, Drach SV, Matveentsev VD, et al. Immunoenzymatic analysis of (24R)-brassinosteroids. Russ J Bioorg Chem 2007;33:347–53. Khripach VA, Litvinovskaya RP, Drach SV, Averkova MA, Zhabinskii VN, Sviridov OV, et al. Immunoenzymatic analysis of (24S)methylbrassinosteroids. Doklady NAN Belarusi 2009;53:74–7. Yokota T, Watanabe S, Ogino Y, Yamaguchi I, Takahashi N. Radioimmunoassay for brassinosteroids and its use for comparative-analysis of brassinosteroids in stems and seeds of Phaseolus vulgaris. J Plant Growth Regul 1990;9:151–9.
9
[16] Takatsuto S. Microanalysis of naturally occurring brassinosteroids. ACS Symp Ser 1991;474:107–20. [17] Taylor PE, Spuck K, Smith PM, Sasse JM, Yokota T, Griffiths PG, et al. Detection of brassinosteroids in pollen of Lolium perenne L. by immunocytochemistry. Planta 1993;189:91–100. [18] Swaczynová J, Šíša M, Hnilicˇková J, Kohout L, Strnad M. Synthesis, biological, immunological and anticancer properties of a new brassinosteroid ligand. Pol J Chem 2006;80:629–35. [19] Swaczynová J, Novák O, Hauserová E, Fuksová K, Šíša M, Kohout L, et al. New techniques for the estimation of naturally occurring brassinosteroids. J Plant Growth Regul 2007;26:1–14. [20] Khripach VA, Sviridov OV, Litvinovskaya RP, Pryadko AG, Drach SV, Zhabinskii VN. Analysis of brassinosteroids. Pol J Chem 2006;80:651–4. [21] Khripach V, Zhabinskii V, Antonchick A, Litvinovskaya R, Drach S, Sviridov O, et al. A new type of modified brassinosteroids for enzyme-linked immunosorbent assay. Nat Prod Commun 2008;3:735–48. [22] Khripach VA, Litvinovskaya RP, Raiman ME, Drach SV, Zhabinskii VN, Sviridov OV, et al. Synthesis and immunochemical determination of 28homobrassinosteroids. Vesti NAN Belarusi, ser khim navuk 2008:47–58. [23] Law B. Immunoassay. A practical guide. Taylor & Francis; 2005. p. 179–82. [24] Chard T. An introduction to radioimmunoassay and related techniques. Elsevier/North-Holland Biomedical Press; 1978. [25] Svatos A, Antonchick A, Schneider B. Determination of brassinosteroids in the sub-femtomolar range using dansyl-3-aminophenylboronate derivatization and electrospray mass spectrometry. Rapid Commun Mass Spectrom 2004;18:818–23. [26] Plimmer JR, Gammon DW, Ragsdale NN. Encyclopedia of agrochemicals. John Wiley & Sons, Inc.; 2003.
Please cite this article in press as: Pradko AG et al. A new ELISA for quantification of brassinosteroids in plants. Steroids (2014), http://dx.doi.org/10.1016/ j.steroids.2014.08.022