- Email: [email protected]
Bioherbicidal activity of Sinapis alba seed meal extracts
Industrial Crops & Products 115 (2018) 174–181
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Bioherbicidal activity of Sinapis alba seed meal extracts a,⁎
a
Matthew J. Morra , Inna E. Popova , Rick A. Boydston a b
T
b
Department of Soil & Water Systems, University of Idaho, 875 Perimeter Drive, MS 2340, Moscow, ID, 83844-2340, USA Agricultural Research Service, USDA, 24106 N. Bun Road, Prosser, WA, 99350, USA
A R T I C L E I N F O
A B S T R A C T
Keywords: Bioherbicides Mustard seed meal Phytotoxic plant extract Thiocyanate Glucosinolate
Although seed meal from yellow mustard (Sinapis alba L.) is a potential tool for controlling weeds as a consequence of produced phytotoxic products, use is limited by batch-to-batch variability and logistical constraints. Our objective was to develop an efficacious bioherbicide by extracting and identifying the active ingredients in S. alba seed meal that demonstrate phytotoxicity to greenhouse-grown Powell amaranth (Amaranthus powellii) and green foxtail (Setaria viridis). Companion bioassays with separate potential active ingredient solutions containing ionic thiocyanate (SCN−), 4-(hydroxymethyl)phenol (4-OH), or 2-(4-hydroxyphenyl)acetonitrile (Nitrile) at concentrations approximating those in the extract were performed. When applied pre- (PRE) or postemergence (POST), SCN− and extracts were the most active solutions on both weed species. The highest rate tested of SCN− of 2.8 kg ha−1 controlled Powell amaranth 98% and green foxtail 84% compared to the highest rate of extract (2.8 kg SCN− ha−1) that controlled Powell amaranth 97% and green foxtail 82%. POST application of the extract was less effective as compared to SCN− solutions, with SCN− showing 97% control of Powell amaranth and 71% control of green foxtail as compared to the extract displaying only 46% control of Powell amaranth and 23% control of green foxtail. Little or no herbicidal activity was observed on both weed species following PRE or POST application of 4-OH or Nitrile. Development of a bioherbicide based on extracting and concentrating SCN− from S. alba seed meal is feasible, especially if scale up activities focus on eliminating the need for alcoholic extractants and yield higher active ingredient products.
1. Introduction Various research groups have shown that yellow mustard (Sinapis alba L.) seed meal is phytotoxic (Boydston et al., 2008; Boydston et al., 2011; Handiseni et al., 2011; Shrestha et al., 2015; Wang et al., 2015; Webber et al., 2017; Yu and Morishita, 2014), leading to the proposed commercial utilization of the seed meal as a bioherbicide (Borek and Morra, 2005). The substitution of synthetic herbicides by natural compounds is essential in organic agriculture and potentially beneficial to human and environmental health. Adoption and large-scale use of mustard seed meal bioherbicides are limited by the bulky nature of the seed meals, variability in active ingredient concentration, and unanticipated impacts on the soil ecosystem as caused by the addition of large amounts of carbon and nitrogen (Popova et al., 2017). Increased efficacy and improved transportation and application logistics for a bioherbicide product can be achieved by extracting and concentrating the active ingredients from S. alba seed meal. Mustard plants contain compounds called glucosinolates that are enzymatically hydrolyzed by myrosinase (thioglucoside
glucohydrolase, EC 3.2.3.1) to form a variety of biologically active compounds (Brown and Morra, 1997; Rosa et al., 1997). S. alba seed meal contains the glucosinolate sinalbin (4-hydroxybenzyl glucosinolate; [(2S,3R,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)oxan-2-yl] 2-(4-hydroxyphenyl)-N-sulfooxyethanimidothioate) that when hydrolyzed produces 4-(isothiocyanatomethyl)phenol, an unstable molecule quickly transformed non-enzymatically to SCN−, 4-(hydroxymethyl) phenol (4-OH), or 2-(4-hydroxyphenyl)acetonitrile (Nitrile) (Fig. 1) (Borek and Morra, 2005). Concentrations of sinalbin as measured in seed meal produced by cold pressing S. alba seed range from 90 to 202 μmol g−1 meal with a mean value of 148 μmol g−1 (unpublished data), making the seed meal a valuable feedstock for herbicidal active ingredients. Based on a correlation of sinalbin and the observed phytotoxic response, as well as the historic use of SCN− as a herbicide (Ahlgren et al., 1951; Beekhuis, 1975; Brown and Morra, 1997; Stiehl and Bible, 1989), it has been speculated that SCN− is the responsible active ingredient (Borek and Morra, 2005); however, conclusive evidence demonstrating what compound or compounds in S. alba seed meal are responsible is lacking.
Abbreviations: PRE, preemergence; POST, postemergence; DAT, days after treatment; DAS, days after sowing; 4-OH, 4-(hydroxymethyl)phenol; Nitrile, 2-(4-hydroxyphenyl)acetonitrile ⁎ Corresponding author. E-mail address: [email protected] (M.J. Morra). https://doi.org/10.1016/j.indcrop.2018.02.027 Received 25 October 2017; Received in revised form 5 February 2018; Accepted 7 February 2018 0926-6690/ © 2018 Elsevier B.V. All rights reserved.
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 1. Pathway for sinalbin hydrolysis to form the three potentially bioactive compounds SCN−, 4-(hydroxymethyl)phenol, and 2-(4-hydroxyphenyl)acetonitrile.
S. alba (IdaGold variety) seed was obtained locally (Latah County, ID, USA). Seed was cold crushed to produce seed meal that contained approximately 15% residual oil (Peterson et al., 1983).
Santa Clara, CA, USA). Separation was performed using a Zorbax SB-Aq (50 mm × 4.6 mm, 3.5 μm) rapid resolution column with a Zorbax SBAq (12.5 mm × 4.6 mm, 5 μm) guard column (Agilent, Santa Clara, CA, USA) maintained at 30 °C. The injection volume was 5 μL. The mobile phase consisted of 0.1% formic acid in water (solvent A) and in acetonitrile (solvent B). The gradient program started with isocratic elution at a flow rate of 0.4 mL min−1 using 5% B for 3 min followed by a linear gradient to 70% B from 3 to 10 min 4-OH and Nitrile were quantified by monitoring absorbances from 220 to 230 nm. External calibration curves were used for quantification of all the analytes. Limits of detection were 0.11 for 4-OH and 0.21 mM for Nitrile. SCN− was analyzed as described previously (Popova and Morra, 2014). Briefly, quantification involved a Dionex Ion Analyzer (Dionex, Sunnyvale, CA, USA) fitted with a Dionex 4 × 210 mm anion-exchange IonPac AS16 column and analyte elution using sodium hydroxide (100 mM) as the mobile phase at a flow rate of 0.9 mL min−1. Detector stabilizer temperature was set at 30 °C with temperature compensation of 1.7% per °C, and anion suppressor current set to 300 mA.
2.2. Chemicals
2.4. Weed control efficacy
Our preliminary investigations indicated the possibility that 4-OH and Nitrile may contribute to observed phytotoxicity. For scale up of the extraction procedure and for registration of a commercial product as a biopesticide, it is imperative to definitively identify the phytotoxic active ingredient(s). Our objectives were to 1) extract the active ingredient(s) to produce an efficacious bioherbicide and 2) determine if SCN− is the primary active ingredient responsible for phytotoxicity. Our eventual goal is to develop protocols for active ingredient extraction from S. alba seed meal for the purpose of formulating commercially viable biopesticides. 2. Materials and methods 2.1. Materials
4-OH, Nitrile, and potassium thiocyanate (SCN−) standards were purchased from Sigma-Aldrich (St. Louis, MO, USA). Acetonitrile, water, methanol, and other solvents were of HPLC or LC/MS grade. Solvents and all other chemicals (at least of analytical grade) were purchased from Sigma–Aldrich or ThermoFisher (Pittsburgh, PA, USA).
2.4.1. Preemergence (PRE) assays Twenty-five Powell amaranth (Amaranthus powellii) or 50 green foxtail (Setaria viridis) seeds were planted 1- or 2-mm deep, respectively, in 10-cm containers filled with loamy sand soil (pH 7.0, 0.55% organic matter). Soil was maintained at field capacity by watering daily. The following day aqueous solutions of S. alba extract, potassium thiocyanate (SCN−), 4-OH, and Nitrile were applied separately to the surface of pots using a pneumatic bench sprayer equipped with a single flat fan nozzle (Teejiet 80015E, Spraying Systems Co., Wheaton, IL.) calibrated to deliver 234 L ha−1. The three rates of individual compounds tested approximated those applied in the three rates of S. alba extract. Extract concentrations as noted in all figures and tables refer to SCN− present in each respective extract to allow for direct comparison of phytotoxicity with solutions of synthetic SCN−. The rates were 50, 100, and 200 mM for SCN− and 4-OH, and 12.5, 25, and 50 mM for Nitrile. Concentrations of the compounds as applied with the described spray system correspond to application rates of: 0.7, 1.4, and 2.8 kg SCN− ha−1; 23.5, 47.0, and 94.0 kg extract ha−1; 1.45, 2.9, and 5.8 kg 4-OH ha−1; and 0.4, 0.8, and 1.6 kg Nitrile ha−1. A nontreated control that received only water was included. All solutions were made using distilled water. Due to the relatively high rate of S. alba extract required in solution, the 200 mM rate (94 kg ha−1) of extract was achieved by mixing a 100 mM rate (47 kg ha−1) and spraying pots twice. Treatments were replicated five times and each trial was repeated. Pots were placed in a greenhouse with a 14.5-h day length and maintained at 32/19 °C maximum/ minimum temperatures. Pots were watered overhead daily by lightly sprinkling using a hose attachment as needed to prevent pots from
2.3. Mustard meal extract preparation and analysis Mustard meals were homogenized and ground to a fine powder. Mustard meal was extracted with 30% (v/v) methanol at a 1:20 v/v ratio using an end-to-end shaker at room temperature for 2 h. Seed debris was separated by filtering, and filtrates were concentrated by rotary evaporation to remove most of the solvent. Concentrated extract was then freeze-dried to obtain a free flowing powder. For analysis of potentially biologically active compounds, 0.1 g of S. alba extract was dissolved in 5 mL of 73% aqueous methanol and agitated on an end-to-end shaker for 1 h. Solutions were centrifuged and the supernatant was analyzed for sinalbin, 4-OH, Nitrile, and SCN−. For analysis of 4-OH and Nitrile, extracts were diluted 20 times with 73% aqueous methanol. For analysis of SCN−, extracts were evaporated to dryness under a gentle stream of N2 and redissolved in water. All analyses were performed immediately prior to the first application, yielding results that accurately reflect extract concentrations at the time of efficacy trials based on preliminary studies indicating that the compounds of interest are stable within the time period of the experiments. 4-OH and Nitrile were analyzed using an Agilent 1200 Series HPLC system equipped with a diode array detection (DAD) system (Agilent, 175
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
3.2.2. Green foxtail PRE bioassays Green foxtail responded similarly to Powell amaranth, with seed meal extracts and SCN− reducing the number of green foxtail plants per pot (Fig. 2), final plant height (Fig. 3), and plant dry weight per pot (Fig. 4), and causing visual injury indicating control 21 DAS (Fig. 5). As with Powell amaranth, control increased and final number of living plants, plant height, and dry weight decreased as rate increased. Although control was incomplete, the highest rate tested (200 mM SCN−) of both the SCN− solution (2.8 kg ha−1) and extract (94 kg ha−1) controlled green foxtail 84 and 82%, respectively at 21 DAS (Fig. 5) and reduced final dry weights by 91 and 85% (Fig. 4), respectively. 4-OH and Nitrile had no significant effect on number of plants and in most cases final dry weight, and did not cause visual injury symptoms demonstrating control when applied PRE at the three rates tested (Figs. 2, 4, and 5). Nitrile tended to reduce final dry weight of green foxtail, but differences were not statistically significant from nontreated controls except for the lowest rate of 12.5 mM (0.4 kg ha−1) (Fig. 4). Although 4-OH did not reduce final plant height, Nitrile solutions applied PRE reduced final plant height of green foxtail by approximately 20% at all rates tested (Fig. 3).
drying out. Emerged live seedlings were recorded at 7, 14, and 21 days after sowing (DAS). Percent control was visually rated at 14 and 21 DAS on a scale of 0 = no injury to 100 = dead. At 21 DAS, average plant height per pot was recorded and all plants were cut off at the soil surface, oven dried at 60 °C, and final dry weight determined. 2.4.2. Postemergence (POST) assays Powell amaranth and green foxtail were sown in 10-cm containers filled with loamy sand soil as described above. Shortly after emergence, plants were thinned to five uniform seedlings per container. When Powell amaranth seedlings were 3–4 cm tall with 3 leaves and green foxtail were 2–4 cm tall with 3 leaves, solutions of S. alba extract, SCN−, 4-OH, and Nitrile were applied using a pneumatic bench sprayer as described previously. All solutions were made using distilled water and included a nonionic surfactant at 1% (v/v) (R-11, Wilbur-Ellis, San Francisco, CA). Each compound was tested at the same three rates tested in the PRE trial. A nontreated control that received only water containing 1% nonionic surfactant was included. Treatments were replicated four times and each trial was repeated. Pots were placed in a greenhouse with a 14.5-h photoperiod and maintained at 32/19 °C maximum/minimum temperatures. Following the spray application, plants did not receive any water on the foliage for 18 h. Pots were then watered overhead daily as needed to prevent pots from drying out. Live plants per pot and visual percent control were recorded on a scale of 0 = no injury to 100 = dead at 7 and 14 days after treatment (DAT). At 14 DAT, average plant height was recorded and all plants were cut off at the soil surface, oven dried at 60 °C, and final dry weight determined.
3.3. POST weed control efficacy trials 3.3.1. Powell amaranth POST bioassays Solutions of SCN− applied POST to Powell amaranth killed most plants at the highest rate of 200 mM (2.8 kg SCN− ha−1) (Fig. 6). No other treatments tested reduced the number of live plants per pot at 14 DAT. Seed meal extract solutions and SCN− applied POST reduced final plant height (Fig. 7) and plant dry weight per pot (Fig. 8) at the higher rates tested and caused visual injury indicating control at 14 DAT (Fig. 9), while the lowest rate tested for both compounds had little effect on any response variable measured. The 200 mM rate of SCN− resulted in 97% control, whereas the corresponding highest rate of mustard meal extract resulted in 46% control (mainly stunted growth of plants) (Fig. 9). 4-OH and Nitrile had no significant effect on number of plants, final plant height, or dry weight, and did not cause significant visual injury or control when applied POST at the rates tested (Figs. 6, 7, 8, and 9). The two higher rates of Nitrile caused some minor epinasty on leaves, but injury was insufficient to result in significant control.
2.4.3. Statistical analyses All data were subjected to analysis of variance (ANOVA) using the PROC GLIMMX procedure in SAS (Statistical Analysis Systems®, version 9.4, SAS Institute Inc., SAS Campus Drive, Cary, NC 27513, USA) to test for significance (P < 0.05) of treatments and other measured variables. Data from repeated trials were pooled for analysis when ANOVA indicated no significant treatment by trial interaction. PRE and POST trials were analyzed separately. Mean separation was conducted using Tukey-Kramer LSMEANS (P = .05).
3.3.2. Green foxtail POST bioassays None of compounds applied POST resulted in death of green foxtail plants at 14 DAT (Fig. 6). All rates of SCN− reduced green foxtail final height (Fig. 7) and dry weight (Fig. 8) compared to nontreated controls, and the highest rate tested resulted in 71% control (mainly stunted growth and leaf tip necrosis) (Fig. 9). The two highest rates of extract tested reduced green foxtail final dry weight by 28 and 54% (Fig. 8) and the highest rate reduced final plant height by 31% (Fig. 7). 4-OH and Nitrile had no significant effect on any response parameter measured (Figs. 6, 7, 8, and 9). The highest rate of Nitrile caused some minor leaf tip necrosis on lower leaves, but injury was insufficient to result in significant control (Fig. 9).
3. Results 3.1. Seed meal chemistry Freeze-dried S. alba seed meal extracts contained up to 20% total solids and 8.6% insoluble solids. The concentrations of SCN−, 4-OH, and Nitrile in the dried material were 441, 104, and 39 μmol g−1, respectively. The pH of the extract in water was 4.4. 3.2. PRE weed control efficacy
4. Discussion
3.2.1. Powell amaranth PRE bioassays Seed meal extracts and SCN− reduced the number of Powell amaranth plants per pot (Fig. 2), final plant height (Fig. 3), and plant dry weight per pot (Fig. 4), and caused visual injury indicating control 21 DAS (Fig. 5). Control increased and final number of living plants, plant height, and dry weight decreased as rate increased. The highest rate tested of both compounds, 200 mM SCN− (2.8 kg SCN− ha−1 and 94 kg extract ha−1), almost completely controlled Powell amaranth (Fig. 5). 4-OH and Nitrile had no significant effect on number of plants, final plant height, or dry weight, and did not cause any visual injury indicating control when applied PRE at the rates tested (Figs. 2, 3, 4, and 5).
4.1. Bioassay results PRE bioassays with both Powell amaranth and green foxtail demonstrated that seed meal extracts were phytotoxic and thus potentially valuable as bioherbicides. The similarity in plant responses to seed meal extracts and SCN−, and the lack of a phytotoxic response to 4-OH and Nitrile indicate that SCN− is most likely the primary active ingredient affecting the weeds tested. Our data thus indicate that 4-OH or Nitrile did not demonstrate phytotoxicities that function additively or synergistically when in combination with SCN−. Although 176
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 2. Plant count means and respective standard deviations determined 21 DAS following PRE treatment of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl) phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
PRE assays, additional components in the extract appeared to reduce efficacy relative to SCN− solutions. Although we did not identify or quantify these compounds in the extract other than the three potential active ingredients, additional compounds are obviously present in a crude extract as obtained here. A wide variety of potentially bioactive compounds are present in water extracts from S. alba seed including antioxidants and emulsifiers (Wu et al., 2015; Wu et al., 2016). Although other components may negatively affect efficacy, formulation techniques may be developed to take advantage of such compounds such as included emulsifiers.
significant phytotoxicities of either compound were not observed in both PRE and POST Powell amaranth and green foxtail bioassays, literature indicates that 4-OH inhibits carrot (Daucus carota) seed formation (Kobayashi et al., 2003) and the glucosinolate-derived nitriles 1cyano-2-hydroxy-3-butene and (3-methoxyphenyl)acetonitrile demonstrate phytotoxicity (Stevens et al., 2009; Vaughn and Berhow, 1998; Vaughn et al., 1996). We applied 4-OH and Nitrile at concentrations in a range similar to those present in S. alba extracts, thus higher concentrations may be required to demonstrate a phytotoxic response. It is also unknown whether other weeds than those tested here might show a response. In comparison, both SCN− solutions and the S. alba seed meal extract demonstrated less efficacy when applied POST to both Powell amaranth and green foxtail. Assays also indicated that in contrast to
4.2. Phytotoxicity of the active ingredient The phytotoxicity of SCN− is well known (Ahlgren et al., 1951; Fig. 3. Plant height means and respective standard deviations determined 21 DAS following PRE treatment of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl) phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
177
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 4. Plant dry weight means and respective standard deviations determined 21 DAS following PRE treatment of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
green foxtail a monocot. Although both responded similarly to applications of the individual compounds and the extract, it is possible that selectivity may occur such that some crop plants are tolerant to the active ingredient, SCN−. Stiehl and Bible (1989) tested the phytotoxicity of SCN− on 39 different crop species by measuring germination and shoot and root growth. Although germination was not affected in any of the species, 46% of the species showed growth inhibition when exposed to 0.5 mM solutions of KSCN. Species within the Apiaceae and Chenopodiaceae were least sensitive, whereas species in the Lamiaceae and Asteraceae were most sensitive to SCN− exposure. Species in the Brassicaceae, Fabaceae, Poaceae, Solanaceae, and Cucurbitaceae showed mixed results, with approximately 50% of the respective species in each family demonstrating sensitivity and 50% tolerance. Studies with S. alba seed meal confirm that differential tolerances exist
Beekhuis, 1975; Bissey and Butler, 1934; Harvey, 1931; Ju et al., 1983), with concentrations between 270–1366 kg SCN− ha−1 demonstrating complete kill of all plants for 4 months (Ahlgren et al., 1951; Harvey, 1931). Cotton defoliation was observed with solutions of SCN− sprayed directly on vegetative growth at a much lower rate of 8.6 kg SCN− ha−1 (Ahlgren et al., 1951). As demonstrated, the rates required for PRE control of Powell amaranth and green foxtail of only 0.7–2.8 kg SCN− ha−1 were substantially less than those previously reported for complete and extended phytotoxicity. The efficacy of lower application rates was facilitated by the fact that control was monitored in PRE assays and in a confined volume of soil that ensured root contact with the active ingredient. Decreased efficacy in POST assays confirms that SCN− rates were relatively low compared to those previously tested. Two representative weeds were tested, Powell amaranth a dicot and
Fig. 5. Control means and respective standard deviations of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract as determined by visually assessing plant injury 21 DAS following PRE treatment. Bars (means) within a species with the same letter are not significantly different according to Tukey-Kramer LSMEANS test at P = .05.
178
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 6. Live plant count means and respective standard deviations determined 14 DAT following POST application of SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract to Powell amaranth and green foxtail. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
among crops with lettuce (Latuca sativa) demonstrating extreme sensitivity (Rice et al., 2007) and carrots showing greater tolerance (Hansson et al., 2008; Snyder et al., 2009). Commercial utilization of extracts derived from S. alba seed meal may thus take advantage of these differential tolerances to control weeds in a crop that exhibits greater tolerance to SCN−. In addition to species-dependent selectivity, it may be possible to achieve selectivity through management strategies. One option is to take advantage of crop size by applying the phytotoxic extract to small weed seedlings in an established or slightly older crop after an initial weed control or suppression operation is performed. Similarly, weed control in perennial or transplanted crops may also be possible. The exact mode of action for the observed phytotoxicity of SCN− has not been determined; however, evidence suggests that SCN−
interferes with photosynthetic O2 release in cotton (Wu and Basler, 1969) and that interactions with iron cause chlorosis in cabbage (Brassica oleraeea) and beans (Phaseolus vulgaris) (Ju et al., 1983). In contrast, hydroponic exposure of rice seedlings to KSCN and NH4SCN solutions did not cause chlorosis or change total chlorophyll content, despite noted decreases in transpiration and growth (Yu and Zhang, 2013; Yu et al., 2012). The major effect observed was a positive correlation in antioxidative enzyme activities (superoxide dismutase, peroxidase, catalase, and ascorbate peroxidase) within the tissues exposed to SCN− (Yu and Zhang, 2013).
4.3. Commercialization considerations Despite the lack of clarity concerning the mode and mechanism of Fig. 7. Plant height means and respective standard deviations determined 14 DAT following POST application of SCN−, 4(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract to Powell amaranth and green foxtail. Bars (means) within a species with the same letter are not significantly different according to Tukey-Kramer LSMEANS test at P = .05.
179
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 8. Plant dry weight means and respective standard deviations determined 14 DAT following POST application of SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract to Powell amaranth and green foxtail. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
Fig. 9. Control means and respective standard deviations of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract as determined by visually assessing plant injury 21 DAT following POST application. Bars (means) within a species with the same letter are not significantly different according to Tukey-Kramer LSMEANS test at P = .05.
180
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
action for SCN−, our research substantiates the need to focus on SCN− as the active ingredient when developing an industrial process for S. alba seed meal extraction and formulation of an efficacious bioherbicide. Thus, scaled-up extraction procedures designed to provide larger extract volumes for commercialization of a bioherbicide may target extraction and concentration of SCN− without considering how extraction may affect 4-OH and Nitrile concentrations. Our procedure relied on the inclusion of 30% methanol to facilitate filtration of the extracts, but not inhibit myrosinase activity and consequent production of SCN−. There is a positive correlation of myrosinase deactivation with methanol concentration, with near complete inhibition occurring in a 73% (v/v) methanol solution (Popova et al., 2017). Utilization of more efficient filtration or centrifugation procedures may permit reduction or elimination of methanol in the extract solution, thereby decreasing extraction costs, the requirement for explosion proof equipment, and the need to recover and recycle spent methanol. Scale-up activities that incorporate these concepts are in progress in combination with methods to increase SCN− concentration in the final dried product. The product as envisioned here may be registered in U.S. by the Environmental Protection Agency (EPA) as a biopesticide potentially useful in high value or organic crops. A relatively low mammalian toxicity reported as an LD50 for rats of 750 mg kg−1 of body weight (Boening and Chew, 1999) will facilitate registration. Registration with EPA will also require consideration of SCN− environmental fate. Although fate studies of SCN− as derived from extracts have not been conducted, production and movement of SCN− as originating from S. alba seed meals have been quantified in field soils (Hansson et al., 2008). Results indicate relatively rapid dissipation of SCN− with concentrations never exceeding approximately 10 μmol kg−1 dry soil at soil depths of 25–35 cm when monitored for 44 d in soil amended with 2 t ha−1 of S. alba seed meal containing 157 μmol sinalbin g−1.
& Sons New York. Beekhuis, H., 1975. Technology and industrial applications. In: Newman, A. (Ed.), Chemistry and Biochemistry of Thiocyanic Acid and Its Derivatives. Academic Press, London, pp. 222–225. Bissey, R., Butler, O., 1934. Effect of applications of sodium chlorate and ammonium thiocyanate on subsequent sowings of wheat. J. Am. Soc. Agron. 26, 838–846. Boening, D.W., Chew, C.M., 1999. A critical review: general toxicity and environmental fate of three aqueous cyanide ions and associated ligands. Water Soil Air Poll. 109, 67–79. Borek, V., Morra, M.J., 2005. Ionic thiocyanate (SCN−) production from 4-hydroxybenzyl glucosinolate contained in Sinapis alba seed meal. J. Agric. Food Chem. 53, 8650–8654. Boydston, R.A., Anderson, T., Vaughn, S.F., 2008. Mustard (Sinapis alba) seed meal suppresses weeds in container-grown ornamentals. HortScience 43, 800–803. Boydston, R.A., Morra, M.J., Borek, V., Clayton, L., Vaughn, S.F., 2011. Onion and weed response to mustard (Sinapis alba) seed meal. Weed Sci. 59, 546–552. Brown, P.D., Morra, M.J., 1997. Control of soil-borne plant pests using glucosinolatecontaining plants. Adv. Agron. 61, 167–231. Handiseni, M., Brown, J., Zemetra, R., Mazzola, M., 2011. Herbicidal activity of Brassicaceae seed meal on wild oat (Avena fatua), Italian ryegrass (Lolium multiflorum), redroot pigweed (Amaranthus retroflexus), and prickly lettuce (Lactuca serriola). Weed Technol. 25, 127–134. Hansson, D., Morra, M.J., Borek, V., Snyder, A.J., Johnson-Maynard, J.L., Thill, D.C., 2008. Ionic thiocyanate (SCN−) production, fate, and phytotoxicity in soil amended with Brassicaceae seed meals. J. Agric. Food Chem. 56, 3912–3917. Harvey, R.B., 1931. Ammonium thiocyanate as a weed eradicant. J. Am. Soc. Agron. 23, 944–946. Ju, H.Y., Bible, B.B., Chong, C., 1983. Influence of ionic thiocyanate on growth of cabbage, bean, and tobacco. J. Chem. Ecol. 8, 1255–1262. Kobayashi, T., Higashi, K., Kamada, H., 2003. 4-Hydroxybenzyl alcohol accumulates in flowers and developing fruits of carrot and inhibits seed formation. J. Plant Physiol. 160, 713–716. Peterson, C.L., Auld, D.L., Thompson, J.C., 1983. Experiments with vegetable oil expression. T. ASAE 26, 1298–1302. Popova, I.E., Morra, M.J., 2014. Simultaneous quantification of sinigrin, sinalbin, and anionic glucosinolate hydrolysis products in Brassica juncea and Sinapis alba seed extracts using ion chromatography. J. Agric. Food Chem. 62, 10687–10693. Popova, I.E., Dubie, J.S., Morra, M.J., 2017. Optimization of hydrolysis conditions for release of biopesticides from glucosinolates in Brassica juncea and Sinapis alba seed meal extracts. Ind. Crops Prod. 97, 354–359. Rice, A.R., Johnson-Maynard, J.L., Thill, D.C., Morra, M.J., 2007. Vegetable crop emergence and weed control following amendment with different Brassicaceae seed meals. Renew. Agric. Food Syst. 22, 204–212. Rosa, E.A.S., Heaney, R.K., Fenwick, G.R., Portas, C.A.M., 1997. Glucosinolates in crop plants. In: Janick, J. (Ed.), Horticultural Reviews. John Wiley and Sons, New York, pp. 99–215. Shrestha, A., Rodriguez, A., Pasakdee, S., Banuelos, G., 2015. Comparative efficacy of white mustard (Sinapis alba L.) and soybean (Glycine max L. Merr.) seed meals as bioherbicides in organic broccoli (Brassica oleracea Var Botrytis) and spinach (Spinacea oleracea) production. Commun. Soil Sci. Plant Anal. 46, 33–46. Snyder, A., Morra, M.J., Johnson-Maynard, J., Thill, D.C., 2009. Seed meals from Brassicaceae oilseed crops as soil amendments: influence on carrot growth, microbial biomass nitrogen, and nitrogen mineralization. HortScience 44, 354–361. Stevens, J.F., Reed, R.L., Alber, S., Pritchett, L., Machado, S., 2009. Herbicidal activity of glucosinolate degradation products in fermented meadowfoam (Limnanthes alba) seed meal. J. Agric. Food Chem. 57, 1821–1826. Stiehl, B., Bible, B.B., 1989. Reaction of crop species to thiocyanate ion toxicity. HortScience 24, 99–101. Vaughn, S.F., Berhow, M.A., 1998. 1-cyano-2-hydroxy-3-butene, a phytotoxin from crambe (Crambe abyssinica) seedmeal. J. Chem. Ecol. 24, 1117–1126. Vaughn, S.F., Boydston, R.A., MallorySmith, C.A., 1996. Isolation and identification of (3methoxyphenyl)acetonitrile as a phytotoxin from meadowfoam (Limnanthes alba) seedmeal. J. Chem. Ecol. 22, 1939–1949. Wang, X., Gu, M.M., Niu, G.H., Baumann, P.A., 2015. Herbicidal activity of mustard seed meal (Sinapis alba ‘IdaGold' and Brassica juncea ‘Pacific Gold') on weed emergence. Ind. Crop. Prod. 77, 1004–1013. Webber III, C.L., White, P.M.J., Boydston, R.A., Shrefler, J.W., 2017. Impact of mustard seed meal applications on direct-seeded cucurbits and weed control. J. Agric. Sci. 9, 1–10. Wu, Y.F., Basler, E., 1969. Effects of ammonium thiocyanate on carbohydrate metabolism in the cotton plant. Weed Sci. 17, 362–365. Wu, Y., Eskin, N.A.M., Cui, W., Pokharel, B., 2015. Emulsifying properties of water soluble yellow mustard mucilage: a comparative study with gum Arabic and citrus pectin. Food Hydrocolloids 47, 191–196. Wu, Y., Hui, A., Eskin, N.A.M., Cui, S.W., 2016. Water-soluble yellow mustard mucilage: a novel ingredient with potent antioxidant properties. Int. J. Biol. Macromol. 91, 710–715. Yu, J.L., Morishita, D.W., 2014. Response of seven weed species to corn gluten meal and white mustard (Sinapis alba) seed meal rates. Weed Technol. 28, 259–265. Yu, X.Z., Zhang, F.Z., 2013. Effects of exogenous thiocyanate on mineral nutrients, antioxidative responses and free amino acids in rice seedlings. Ecotoxicology 22, 752–760. Yu, X.Z., Zhang, F.Z., Li, F., 2012. Phytotoxicity of thiocyanate to rice seedlings. Bull. Environ. Contam. Toxicol. 88, 703–706.
5. Conclusions It is possible to produce a powder derived from S. alba seed meal extracts that contains SCN− as the active ingredient. The powder may be reconstituted in water and applied as a spray in the form of a PRE bioherbicide. POST control of weeds may be possible with the extract, but higher application rates or improved formulations are required. Testing of the extract on additional weeds and crops is necessary to take advantage of the differential tolerance to SCN− among species and provide label information required for EPA registration. Field research to test extracts on crops grown under real-world conditions is being conducted in collaboration with business partners. Commercialization will benefit from optimizing extraction to produce higher SCN− concentrations in the product and eliminate the use of alcohol in the extracting solution. Also relevant but not discussed here, is the need to consider how the extracted and now detoxified seed meal may be used in other products such as animal feeds. Conflicts of interest The authors have no conflicts of interest to declare. Acknowledgements This project was supported by the Agriculture and Food Research Initiative competitive grant 2011-67009-20094 from the USDA National Institute of Food and Agriculture. The authors thank Bernardo Chaves-Cordoba for statistical analysis of the data and Treva Anderson for technical assistance. References Ahlgren, G.H., Klingman, G.C., Wolf, D.E., 1951. Principles of Weed Control. John Wiley
181
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Bioherbicidal activity of Sinapis alba seed meal extracts a,⁎
a
Matthew J. Morra , Inna E. Popova , Rick A. Boydston a b
T
b
Department of Soil & Water Systems, University of Idaho, 875 Perimeter Drive, MS 2340, Moscow, ID, 83844-2340, USA Agricultural Research Service, USDA, 24106 N. Bun Road, Prosser, WA, 99350, USA
A R T I C L E I N F O
A B S T R A C T
Keywords: Bioherbicides Mustard seed meal Phytotoxic plant extract Thiocyanate Glucosinolate
Although seed meal from yellow mustard (Sinapis alba L.) is a potential tool for controlling weeds as a consequence of produced phytotoxic products, use is limited by batch-to-batch variability and logistical constraints. Our objective was to develop an efficacious bioherbicide by extracting and identifying the active ingredients in S. alba seed meal that demonstrate phytotoxicity to greenhouse-grown Powell amaranth (Amaranthus powellii) and green foxtail (Setaria viridis). Companion bioassays with separate potential active ingredient solutions containing ionic thiocyanate (SCN−), 4-(hydroxymethyl)phenol (4-OH), or 2-(4-hydroxyphenyl)acetonitrile (Nitrile) at concentrations approximating those in the extract were performed. When applied pre- (PRE) or postemergence (POST), SCN− and extracts were the most active solutions on both weed species. The highest rate tested of SCN− of 2.8 kg ha−1 controlled Powell amaranth 98% and green foxtail 84% compared to the highest rate of extract (2.8 kg SCN− ha−1) that controlled Powell amaranth 97% and green foxtail 82%. POST application of the extract was less effective as compared to SCN− solutions, with SCN− showing 97% control of Powell amaranth and 71% control of green foxtail as compared to the extract displaying only 46% control of Powell amaranth and 23% control of green foxtail. Little or no herbicidal activity was observed on both weed species following PRE or POST application of 4-OH or Nitrile. Development of a bioherbicide based on extracting and concentrating SCN− from S. alba seed meal is feasible, especially if scale up activities focus on eliminating the need for alcoholic extractants and yield higher active ingredient products.
1. Introduction Various research groups have shown that yellow mustard (Sinapis alba L.) seed meal is phytotoxic (Boydston et al., 2008; Boydston et al., 2011; Handiseni et al., 2011; Shrestha et al., 2015; Wang et al., 2015; Webber et al., 2017; Yu and Morishita, 2014), leading to the proposed commercial utilization of the seed meal as a bioherbicide (Borek and Morra, 2005). The substitution of synthetic herbicides by natural compounds is essential in organic agriculture and potentially beneficial to human and environmental health. Adoption and large-scale use of mustard seed meal bioherbicides are limited by the bulky nature of the seed meals, variability in active ingredient concentration, and unanticipated impacts on the soil ecosystem as caused by the addition of large amounts of carbon and nitrogen (Popova et al., 2017). Increased efficacy and improved transportation and application logistics for a bioherbicide product can be achieved by extracting and concentrating the active ingredients from S. alba seed meal. Mustard plants contain compounds called glucosinolates that are enzymatically hydrolyzed by myrosinase (thioglucoside
glucohydrolase, EC 3.2.3.1) to form a variety of biologically active compounds (Brown and Morra, 1997; Rosa et al., 1997). S. alba seed meal contains the glucosinolate sinalbin (4-hydroxybenzyl glucosinolate; [(2S,3R,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)oxan-2-yl] 2-(4-hydroxyphenyl)-N-sulfooxyethanimidothioate) that when hydrolyzed produces 4-(isothiocyanatomethyl)phenol, an unstable molecule quickly transformed non-enzymatically to SCN−, 4-(hydroxymethyl) phenol (4-OH), or 2-(4-hydroxyphenyl)acetonitrile (Nitrile) (Fig. 1) (Borek and Morra, 2005). Concentrations of sinalbin as measured in seed meal produced by cold pressing S. alba seed range from 90 to 202 μmol g−1 meal with a mean value of 148 μmol g−1 (unpublished data), making the seed meal a valuable feedstock for herbicidal active ingredients. Based on a correlation of sinalbin and the observed phytotoxic response, as well as the historic use of SCN− as a herbicide (Ahlgren et al., 1951; Beekhuis, 1975; Brown and Morra, 1997; Stiehl and Bible, 1989), it has been speculated that SCN− is the responsible active ingredient (Borek and Morra, 2005); however, conclusive evidence demonstrating what compound or compounds in S. alba seed meal are responsible is lacking.
Abbreviations: PRE, preemergence; POST, postemergence; DAT, days after treatment; DAS, days after sowing; 4-OH, 4-(hydroxymethyl)phenol; Nitrile, 2-(4-hydroxyphenyl)acetonitrile ⁎ Corresponding author. E-mail address: [email protected] (M.J. Morra). https://doi.org/10.1016/j.indcrop.2018.02.027 Received 25 October 2017; Received in revised form 5 February 2018; Accepted 7 February 2018 0926-6690/ © 2018 Elsevier B.V. All rights reserved.
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 1. Pathway for sinalbin hydrolysis to form the three potentially bioactive compounds SCN−, 4-(hydroxymethyl)phenol, and 2-(4-hydroxyphenyl)acetonitrile.
S. alba (IdaGold variety) seed was obtained locally (Latah County, ID, USA). Seed was cold crushed to produce seed meal that contained approximately 15% residual oil (Peterson et al., 1983).
Santa Clara, CA, USA). Separation was performed using a Zorbax SB-Aq (50 mm × 4.6 mm, 3.5 μm) rapid resolution column with a Zorbax SBAq (12.5 mm × 4.6 mm, 5 μm) guard column (Agilent, Santa Clara, CA, USA) maintained at 30 °C. The injection volume was 5 μL. The mobile phase consisted of 0.1% formic acid in water (solvent A) and in acetonitrile (solvent B). The gradient program started with isocratic elution at a flow rate of 0.4 mL min−1 using 5% B for 3 min followed by a linear gradient to 70% B from 3 to 10 min 4-OH and Nitrile were quantified by monitoring absorbances from 220 to 230 nm. External calibration curves were used for quantification of all the analytes. Limits of detection were 0.11 for 4-OH and 0.21 mM for Nitrile. SCN− was analyzed as described previously (Popova and Morra, 2014). Briefly, quantification involved a Dionex Ion Analyzer (Dionex, Sunnyvale, CA, USA) fitted with a Dionex 4 × 210 mm anion-exchange IonPac AS16 column and analyte elution using sodium hydroxide (100 mM) as the mobile phase at a flow rate of 0.9 mL min−1. Detector stabilizer temperature was set at 30 °C with temperature compensation of 1.7% per °C, and anion suppressor current set to 300 mA.
2.2. Chemicals
2.4. Weed control efficacy
Our preliminary investigations indicated the possibility that 4-OH and Nitrile may contribute to observed phytotoxicity. For scale up of the extraction procedure and for registration of a commercial product as a biopesticide, it is imperative to definitively identify the phytotoxic active ingredient(s). Our objectives were to 1) extract the active ingredient(s) to produce an efficacious bioherbicide and 2) determine if SCN− is the primary active ingredient responsible for phytotoxicity. Our eventual goal is to develop protocols for active ingredient extraction from S. alba seed meal for the purpose of formulating commercially viable biopesticides. 2. Materials and methods 2.1. Materials
4-OH, Nitrile, and potassium thiocyanate (SCN−) standards were purchased from Sigma-Aldrich (St. Louis, MO, USA). Acetonitrile, water, methanol, and other solvents were of HPLC or LC/MS grade. Solvents and all other chemicals (at least of analytical grade) were purchased from Sigma–Aldrich or ThermoFisher (Pittsburgh, PA, USA).
2.4.1. Preemergence (PRE) assays Twenty-five Powell amaranth (Amaranthus powellii) or 50 green foxtail (Setaria viridis) seeds were planted 1- or 2-mm deep, respectively, in 10-cm containers filled with loamy sand soil (pH 7.0, 0.55% organic matter). Soil was maintained at field capacity by watering daily. The following day aqueous solutions of S. alba extract, potassium thiocyanate (SCN−), 4-OH, and Nitrile were applied separately to the surface of pots using a pneumatic bench sprayer equipped with a single flat fan nozzle (Teejiet 80015E, Spraying Systems Co., Wheaton, IL.) calibrated to deliver 234 L ha−1. The three rates of individual compounds tested approximated those applied in the three rates of S. alba extract. Extract concentrations as noted in all figures and tables refer to SCN− present in each respective extract to allow for direct comparison of phytotoxicity with solutions of synthetic SCN−. The rates were 50, 100, and 200 mM for SCN− and 4-OH, and 12.5, 25, and 50 mM for Nitrile. Concentrations of the compounds as applied with the described spray system correspond to application rates of: 0.7, 1.4, and 2.8 kg SCN− ha−1; 23.5, 47.0, and 94.0 kg extract ha−1; 1.45, 2.9, and 5.8 kg 4-OH ha−1; and 0.4, 0.8, and 1.6 kg Nitrile ha−1. A nontreated control that received only water was included. All solutions were made using distilled water. Due to the relatively high rate of S. alba extract required in solution, the 200 mM rate (94 kg ha−1) of extract was achieved by mixing a 100 mM rate (47 kg ha−1) and spraying pots twice. Treatments were replicated five times and each trial was repeated. Pots were placed in a greenhouse with a 14.5-h day length and maintained at 32/19 °C maximum/ minimum temperatures. Pots were watered overhead daily by lightly sprinkling using a hose attachment as needed to prevent pots from
2.3. Mustard meal extract preparation and analysis Mustard meals were homogenized and ground to a fine powder. Mustard meal was extracted with 30% (v/v) methanol at a 1:20 v/v ratio using an end-to-end shaker at room temperature for 2 h. Seed debris was separated by filtering, and filtrates were concentrated by rotary evaporation to remove most of the solvent. Concentrated extract was then freeze-dried to obtain a free flowing powder. For analysis of potentially biologically active compounds, 0.1 g of S. alba extract was dissolved in 5 mL of 73% aqueous methanol and agitated on an end-to-end shaker for 1 h. Solutions were centrifuged and the supernatant was analyzed for sinalbin, 4-OH, Nitrile, and SCN−. For analysis of 4-OH and Nitrile, extracts were diluted 20 times with 73% aqueous methanol. For analysis of SCN−, extracts were evaporated to dryness under a gentle stream of N2 and redissolved in water. All analyses were performed immediately prior to the first application, yielding results that accurately reflect extract concentrations at the time of efficacy trials based on preliminary studies indicating that the compounds of interest are stable within the time period of the experiments. 4-OH and Nitrile were analyzed using an Agilent 1200 Series HPLC system equipped with a diode array detection (DAD) system (Agilent, 175
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
3.2.2. Green foxtail PRE bioassays Green foxtail responded similarly to Powell amaranth, with seed meal extracts and SCN− reducing the number of green foxtail plants per pot (Fig. 2), final plant height (Fig. 3), and plant dry weight per pot (Fig. 4), and causing visual injury indicating control 21 DAS (Fig. 5). As with Powell amaranth, control increased and final number of living plants, plant height, and dry weight decreased as rate increased. Although control was incomplete, the highest rate tested (200 mM SCN−) of both the SCN− solution (2.8 kg ha−1) and extract (94 kg ha−1) controlled green foxtail 84 and 82%, respectively at 21 DAS (Fig. 5) and reduced final dry weights by 91 and 85% (Fig. 4), respectively. 4-OH and Nitrile had no significant effect on number of plants and in most cases final dry weight, and did not cause visual injury symptoms demonstrating control when applied PRE at the three rates tested (Figs. 2, 4, and 5). Nitrile tended to reduce final dry weight of green foxtail, but differences were not statistically significant from nontreated controls except for the lowest rate of 12.5 mM (0.4 kg ha−1) (Fig. 4). Although 4-OH did not reduce final plant height, Nitrile solutions applied PRE reduced final plant height of green foxtail by approximately 20% at all rates tested (Fig. 3).
drying out. Emerged live seedlings were recorded at 7, 14, and 21 days after sowing (DAS). Percent control was visually rated at 14 and 21 DAS on a scale of 0 = no injury to 100 = dead. At 21 DAS, average plant height per pot was recorded and all plants were cut off at the soil surface, oven dried at 60 °C, and final dry weight determined. 2.4.2. Postemergence (POST) assays Powell amaranth and green foxtail were sown in 10-cm containers filled with loamy sand soil as described above. Shortly after emergence, plants were thinned to five uniform seedlings per container. When Powell amaranth seedlings were 3–4 cm tall with 3 leaves and green foxtail were 2–4 cm tall with 3 leaves, solutions of S. alba extract, SCN−, 4-OH, and Nitrile were applied using a pneumatic bench sprayer as described previously. All solutions were made using distilled water and included a nonionic surfactant at 1% (v/v) (R-11, Wilbur-Ellis, San Francisco, CA). Each compound was tested at the same three rates tested in the PRE trial. A nontreated control that received only water containing 1% nonionic surfactant was included. Treatments were replicated four times and each trial was repeated. Pots were placed in a greenhouse with a 14.5-h photoperiod and maintained at 32/19 °C maximum/minimum temperatures. Following the spray application, plants did not receive any water on the foliage for 18 h. Pots were then watered overhead daily as needed to prevent pots from drying out. Live plants per pot and visual percent control were recorded on a scale of 0 = no injury to 100 = dead at 7 and 14 days after treatment (DAT). At 14 DAT, average plant height was recorded and all plants were cut off at the soil surface, oven dried at 60 °C, and final dry weight determined.
3.3. POST weed control efficacy trials 3.3.1. Powell amaranth POST bioassays Solutions of SCN− applied POST to Powell amaranth killed most plants at the highest rate of 200 mM (2.8 kg SCN− ha−1) (Fig. 6). No other treatments tested reduced the number of live plants per pot at 14 DAT. Seed meal extract solutions and SCN− applied POST reduced final plant height (Fig. 7) and plant dry weight per pot (Fig. 8) at the higher rates tested and caused visual injury indicating control at 14 DAT (Fig. 9), while the lowest rate tested for both compounds had little effect on any response variable measured. The 200 mM rate of SCN− resulted in 97% control, whereas the corresponding highest rate of mustard meal extract resulted in 46% control (mainly stunted growth of plants) (Fig. 9). 4-OH and Nitrile had no significant effect on number of plants, final plant height, or dry weight, and did not cause significant visual injury or control when applied POST at the rates tested (Figs. 6, 7, 8, and 9). The two higher rates of Nitrile caused some minor epinasty on leaves, but injury was insufficient to result in significant control.
2.4.3. Statistical analyses All data were subjected to analysis of variance (ANOVA) using the PROC GLIMMX procedure in SAS (Statistical Analysis Systems®, version 9.4, SAS Institute Inc., SAS Campus Drive, Cary, NC 27513, USA) to test for significance (P < 0.05) of treatments and other measured variables. Data from repeated trials were pooled for analysis when ANOVA indicated no significant treatment by trial interaction. PRE and POST trials were analyzed separately. Mean separation was conducted using Tukey-Kramer LSMEANS (P = .05).
3.3.2. Green foxtail POST bioassays None of compounds applied POST resulted in death of green foxtail plants at 14 DAT (Fig. 6). All rates of SCN− reduced green foxtail final height (Fig. 7) and dry weight (Fig. 8) compared to nontreated controls, and the highest rate tested resulted in 71% control (mainly stunted growth and leaf tip necrosis) (Fig. 9). The two highest rates of extract tested reduced green foxtail final dry weight by 28 and 54% (Fig. 8) and the highest rate reduced final plant height by 31% (Fig. 7). 4-OH and Nitrile had no significant effect on any response parameter measured (Figs. 6, 7, 8, and 9). The highest rate of Nitrile caused some minor leaf tip necrosis on lower leaves, but injury was insufficient to result in significant control (Fig. 9).
3. Results 3.1. Seed meal chemistry Freeze-dried S. alba seed meal extracts contained up to 20% total solids and 8.6% insoluble solids. The concentrations of SCN−, 4-OH, and Nitrile in the dried material were 441, 104, and 39 μmol g−1, respectively. The pH of the extract in water was 4.4. 3.2. PRE weed control efficacy
4. Discussion
3.2.1. Powell amaranth PRE bioassays Seed meal extracts and SCN− reduced the number of Powell amaranth plants per pot (Fig. 2), final plant height (Fig. 3), and plant dry weight per pot (Fig. 4), and caused visual injury indicating control 21 DAS (Fig. 5). Control increased and final number of living plants, plant height, and dry weight decreased as rate increased. The highest rate tested of both compounds, 200 mM SCN− (2.8 kg SCN− ha−1 and 94 kg extract ha−1), almost completely controlled Powell amaranth (Fig. 5). 4-OH and Nitrile had no significant effect on number of plants, final plant height, or dry weight, and did not cause any visual injury indicating control when applied PRE at the rates tested (Figs. 2, 3, 4, and 5).
4.1. Bioassay results PRE bioassays with both Powell amaranth and green foxtail demonstrated that seed meal extracts were phytotoxic and thus potentially valuable as bioherbicides. The similarity in plant responses to seed meal extracts and SCN−, and the lack of a phytotoxic response to 4-OH and Nitrile indicate that SCN− is most likely the primary active ingredient affecting the weeds tested. Our data thus indicate that 4-OH or Nitrile did not demonstrate phytotoxicities that function additively or synergistically when in combination with SCN−. Although 176
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 2. Plant count means and respective standard deviations determined 21 DAS following PRE treatment of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl) phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
PRE assays, additional components in the extract appeared to reduce efficacy relative to SCN− solutions. Although we did not identify or quantify these compounds in the extract other than the three potential active ingredients, additional compounds are obviously present in a crude extract as obtained here. A wide variety of potentially bioactive compounds are present in water extracts from S. alba seed including antioxidants and emulsifiers (Wu et al., 2015; Wu et al., 2016). Although other components may negatively affect efficacy, formulation techniques may be developed to take advantage of such compounds such as included emulsifiers.
significant phytotoxicities of either compound were not observed in both PRE and POST Powell amaranth and green foxtail bioassays, literature indicates that 4-OH inhibits carrot (Daucus carota) seed formation (Kobayashi et al., 2003) and the glucosinolate-derived nitriles 1cyano-2-hydroxy-3-butene and (3-methoxyphenyl)acetonitrile demonstrate phytotoxicity (Stevens et al., 2009; Vaughn and Berhow, 1998; Vaughn et al., 1996). We applied 4-OH and Nitrile at concentrations in a range similar to those present in S. alba extracts, thus higher concentrations may be required to demonstrate a phytotoxic response. It is also unknown whether other weeds than those tested here might show a response. In comparison, both SCN− solutions and the S. alba seed meal extract demonstrated less efficacy when applied POST to both Powell amaranth and green foxtail. Assays also indicated that in contrast to
4.2. Phytotoxicity of the active ingredient The phytotoxicity of SCN− is well known (Ahlgren et al., 1951; Fig. 3. Plant height means and respective standard deviations determined 21 DAS following PRE treatment of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl) phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
177
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 4. Plant dry weight means and respective standard deviations determined 21 DAS following PRE treatment of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
green foxtail a monocot. Although both responded similarly to applications of the individual compounds and the extract, it is possible that selectivity may occur such that some crop plants are tolerant to the active ingredient, SCN−. Stiehl and Bible (1989) tested the phytotoxicity of SCN− on 39 different crop species by measuring germination and shoot and root growth. Although germination was not affected in any of the species, 46% of the species showed growth inhibition when exposed to 0.5 mM solutions of KSCN. Species within the Apiaceae and Chenopodiaceae were least sensitive, whereas species in the Lamiaceae and Asteraceae were most sensitive to SCN− exposure. Species in the Brassicaceae, Fabaceae, Poaceae, Solanaceae, and Cucurbitaceae showed mixed results, with approximately 50% of the respective species in each family demonstrating sensitivity and 50% tolerance. Studies with S. alba seed meal confirm that differential tolerances exist
Beekhuis, 1975; Bissey and Butler, 1934; Harvey, 1931; Ju et al., 1983), with concentrations between 270–1366 kg SCN− ha−1 demonstrating complete kill of all plants for 4 months (Ahlgren et al., 1951; Harvey, 1931). Cotton defoliation was observed with solutions of SCN− sprayed directly on vegetative growth at a much lower rate of 8.6 kg SCN− ha−1 (Ahlgren et al., 1951). As demonstrated, the rates required for PRE control of Powell amaranth and green foxtail of only 0.7–2.8 kg SCN− ha−1 were substantially less than those previously reported for complete and extended phytotoxicity. The efficacy of lower application rates was facilitated by the fact that control was monitored in PRE assays and in a confined volume of soil that ensured root contact with the active ingredient. Decreased efficacy in POST assays confirms that SCN− rates were relatively low compared to those previously tested. Two representative weeds were tested, Powell amaranth a dicot and
Fig. 5. Control means and respective standard deviations of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract as determined by visually assessing plant injury 21 DAS following PRE treatment. Bars (means) within a species with the same letter are not significantly different according to Tukey-Kramer LSMEANS test at P = .05.
178
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 6. Live plant count means and respective standard deviations determined 14 DAT following POST application of SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract to Powell amaranth and green foxtail. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
among crops with lettuce (Latuca sativa) demonstrating extreme sensitivity (Rice et al., 2007) and carrots showing greater tolerance (Hansson et al., 2008; Snyder et al., 2009). Commercial utilization of extracts derived from S. alba seed meal may thus take advantage of these differential tolerances to control weeds in a crop that exhibits greater tolerance to SCN−. In addition to species-dependent selectivity, it may be possible to achieve selectivity through management strategies. One option is to take advantage of crop size by applying the phytotoxic extract to small weed seedlings in an established or slightly older crop after an initial weed control or suppression operation is performed. Similarly, weed control in perennial or transplanted crops may also be possible. The exact mode of action for the observed phytotoxicity of SCN− has not been determined; however, evidence suggests that SCN−
interferes with photosynthetic O2 release in cotton (Wu and Basler, 1969) and that interactions with iron cause chlorosis in cabbage (Brassica oleraeea) and beans (Phaseolus vulgaris) (Ju et al., 1983). In contrast, hydroponic exposure of rice seedlings to KSCN and NH4SCN solutions did not cause chlorosis or change total chlorophyll content, despite noted decreases in transpiration and growth (Yu and Zhang, 2013; Yu et al., 2012). The major effect observed was a positive correlation in antioxidative enzyme activities (superoxide dismutase, peroxidase, catalase, and ascorbate peroxidase) within the tissues exposed to SCN− (Yu and Zhang, 2013).
4.3. Commercialization considerations Despite the lack of clarity concerning the mode and mechanism of Fig. 7. Plant height means and respective standard deviations determined 14 DAT following POST application of SCN−, 4(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract to Powell amaranth and green foxtail. Bars (means) within a species with the same letter are not significantly different according to Tukey-Kramer LSMEANS test at P = .05.
179
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
Fig. 8. Plant dry weight means and respective standard deviations determined 14 DAT following POST application of SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract to Powell amaranth and green foxtail. Bars (means) within a species with the same letter are not significantly different according to TukeyKramer LSMEANS test at P = .05.
Fig. 9. Control means and respective standard deviations of Powell amaranth and green foxtail with SCN−, 4-(hydroxymethyl)phenol, 2-(4-hydroxyphenyl)acetonitrile, or a S. alba seed meal extract as determined by visually assessing plant injury 21 DAT following POST application. Bars (means) within a species with the same letter are not significantly different according to Tukey-Kramer LSMEANS test at P = .05.
180
Industrial Crops & Products 115 (2018) 174–181
M.J. Morra et al.
action for SCN−, our research substantiates the need to focus on SCN− as the active ingredient when developing an industrial process for S. alba seed meal extraction and formulation of an efficacious bioherbicide. Thus, scaled-up extraction procedures designed to provide larger extract volumes for commercialization of a bioherbicide may target extraction and concentration of SCN− without considering how extraction may affect 4-OH and Nitrile concentrations. Our procedure relied on the inclusion of 30% methanol to facilitate filtration of the extracts, but not inhibit myrosinase activity and consequent production of SCN−. There is a positive correlation of myrosinase deactivation with methanol concentration, with near complete inhibition occurring in a 73% (v/v) methanol solution (Popova et al., 2017). Utilization of more efficient filtration or centrifugation procedures may permit reduction or elimination of methanol in the extract solution, thereby decreasing extraction costs, the requirement for explosion proof equipment, and the need to recover and recycle spent methanol. Scale-up activities that incorporate these concepts are in progress in combination with methods to increase SCN− concentration in the final dried product. The product as envisioned here may be registered in U.S. by the Environmental Protection Agency (EPA) as a biopesticide potentially useful in high value or organic crops. A relatively low mammalian toxicity reported as an LD50 for rats of 750 mg kg−1 of body weight (Boening and Chew, 1999) will facilitate registration. Registration with EPA will also require consideration of SCN− environmental fate. Although fate studies of SCN− as derived from extracts have not been conducted, production and movement of SCN− as originating from S. alba seed meals have been quantified in field soils (Hansson et al., 2008). Results indicate relatively rapid dissipation of SCN− with concentrations never exceeding approximately 10 μmol kg−1 dry soil at soil depths of 25–35 cm when monitored for 44 d in soil amended with 2 t ha−1 of S. alba seed meal containing 157 μmol sinalbin g−1.
& Sons New York. Beekhuis, H., 1975. Technology and industrial applications. In: Newman, A. (Ed.), Chemistry and Biochemistry of Thiocyanic Acid and Its Derivatives. Academic Press, London, pp. 222–225. Bissey, R., Butler, O., 1934. Effect of applications of sodium chlorate and ammonium thiocyanate on subsequent sowings of wheat. J. Am. Soc. Agron. 26, 838–846. Boening, D.W., Chew, C.M., 1999. A critical review: general toxicity and environmental fate of three aqueous cyanide ions and associated ligands. Water Soil Air Poll. 109, 67–79. Borek, V., Morra, M.J., 2005. Ionic thiocyanate (SCN−) production from 4-hydroxybenzyl glucosinolate contained in Sinapis alba seed meal. J. Agric. Food Chem. 53, 8650–8654. Boydston, R.A., Anderson, T., Vaughn, S.F., 2008. Mustard (Sinapis alba) seed meal suppresses weeds in container-grown ornamentals. HortScience 43, 800–803. Boydston, R.A., Morra, M.J., Borek, V., Clayton, L., Vaughn, S.F., 2011. Onion and weed response to mustard (Sinapis alba) seed meal. Weed Sci. 59, 546–552. Brown, P.D., Morra, M.J., 1997. Control of soil-borne plant pests using glucosinolatecontaining plants. Adv. Agron. 61, 167–231. Handiseni, M., Brown, J., Zemetra, R., Mazzola, M., 2011. Herbicidal activity of Brassicaceae seed meal on wild oat (Avena fatua), Italian ryegrass (Lolium multiflorum), redroot pigweed (Amaranthus retroflexus), and prickly lettuce (Lactuca serriola). Weed Technol. 25, 127–134. Hansson, D., Morra, M.J., Borek, V., Snyder, A.J., Johnson-Maynard, J.L., Thill, D.C., 2008. Ionic thiocyanate (SCN−) production, fate, and phytotoxicity in soil amended with Brassicaceae seed meals. J. Agric. Food Chem. 56, 3912–3917. Harvey, R.B., 1931. Ammonium thiocyanate as a weed eradicant. J. Am. Soc. Agron. 23, 944–946. Ju, H.Y., Bible, B.B., Chong, C., 1983. Influence of ionic thiocyanate on growth of cabbage, bean, and tobacco. J. Chem. Ecol. 8, 1255–1262. Kobayashi, T., Higashi, K., Kamada, H., 2003. 4-Hydroxybenzyl alcohol accumulates in flowers and developing fruits of carrot and inhibits seed formation. J. Plant Physiol. 160, 713–716. Peterson, C.L., Auld, D.L., Thompson, J.C., 1983. Experiments with vegetable oil expression. T. ASAE 26, 1298–1302. Popova, I.E., Morra, M.J., 2014. Simultaneous quantification of sinigrin, sinalbin, and anionic glucosinolate hydrolysis products in Brassica juncea and Sinapis alba seed extracts using ion chromatography. J. Agric. Food Chem. 62, 10687–10693. Popova, I.E., Dubie, J.S., Morra, M.J., 2017. Optimization of hydrolysis conditions for release of biopesticides from glucosinolates in Brassica juncea and Sinapis alba seed meal extracts. Ind. Crops Prod. 97, 354–359. Rice, A.R., Johnson-Maynard, J.L., Thill, D.C., Morra, M.J., 2007. Vegetable crop emergence and weed control following amendment with different Brassicaceae seed meals. Renew. Agric. Food Syst. 22, 204–212. Rosa, E.A.S., Heaney, R.K., Fenwick, G.R., Portas, C.A.M., 1997. Glucosinolates in crop plants. In: Janick, J. (Ed.), Horticultural Reviews. John Wiley and Sons, New York, pp. 99–215. Shrestha, A., Rodriguez, A., Pasakdee, S., Banuelos, G., 2015. Comparative efficacy of white mustard (Sinapis alba L.) and soybean (Glycine max L. Merr.) seed meals as bioherbicides in organic broccoli (Brassica oleracea Var Botrytis) and spinach (Spinacea oleracea) production. Commun. Soil Sci. Plant Anal. 46, 33–46. Snyder, A., Morra, M.J., Johnson-Maynard, J., Thill, D.C., 2009. Seed meals from Brassicaceae oilseed crops as soil amendments: influence on carrot growth, microbial biomass nitrogen, and nitrogen mineralization. HortScience 44, 354–361. Stevens, J.F., Reed, R.L., Alber, S., Pritchett, L., Machado, S., 2009. Herbicidal activity of glucosinolate degradation products in fermented meadowfoam (Limnanthes alba) seed meal. J. Agric. Food Chem. 57, 1821–1826. Stiehl, B., Bible, B.B., 1989. Reaction of crop species to thiocyanate ion toxicity. HortScience 24, 99–101. Vaughn, S.F., Berhow, M.A., 1998. 1-cyano-2-hydroxy-3-butene, a phytotoxin from crambe (Crambe abyssinica) seedmeal. J. Chem. Ecol. 24, 1117–1126. Vaughn, S.F., Boydston, R.A., MallorySmith, C.A., 1996. Isolation and identification of (3methoxyphenyl)acetonitrile as a phytotoxin from meadowfoam (Limnanthes alba) seedmeal. J. Chem. Ecol. 22, 1939–1949. Wang, X., Gu, M.M., Niu, G.H., Baumann, P.A., 2015. Herbicidal activity of mustard seed meal (Sinapis alba ‘IdaGold' and Brassica juncea ‘Pacific Gold') on weed emergence. Ind. Crop. Prod. 77, 1004–1013. Webber III, C.L., White, P.M.J., Boydston, R.A., Shrefler, J.W., 2017. Impact of mustard seed meal applications on direct-seeded cucurbits and weed control. J. Agric. Sci. 9, 1–10. Wu, Y.F., Basler, E., 1969. Effects of ammonium thiocyanate on carbohydrate metabolism in the cotton plant. Weed Sci. 17, 362–365. Wu, Y., Eskin, N.A.M., Cui, W., Pokharel, B., 2015. Emulsifying properties of water soluble yellow mustard mucilage: a comparative study with gum Arabic and citrus pectin. Food Hydrocolloids 47, 191–196. Wu, Y., Hui, A., Eskin, N.A.M., Cui, S.W., 2016. Water-soluble yellow mustard mucilage: a novel ingredient with potent antioxidant properties. Int. J. Biol. Macromol. 91, 710–715. Yu, J.L., Morishita, D.W., 2014. Response of seven weed species to corn gluten meal and white mustard (Sinapis alba) seed meal rates. Weed Technol. 28, 259–265. Yu, X.Z., Zhang, F.Z., 2013. Effects of exogenous thiocyanate on mineral nutrients, antioxidative responses and free amino acids in rice seedlings. Ecotoxicology 22, 752–760. Yu, X.Z., Zhang, F.Z., Li, F., 2012. Phytotoxicity of thiocyanate to rice seedlings. Bull. Environ. Contam. Toxicol. 88, 703–706.
5. Conclusions It is possible to produce a powder derived from S. alba seed meal extracts that contains SCN− as the active ingredient. The powder may be reconstituted in water and applied as a spray in the form of a PRE bioherbicide. POST control of weeds may be possible with the extract, but higher application rates or improved formulations are required. Testing of the extract on additional weeds and crops is necessary to take advantage of the differential tolerance to SCN− among species and provide label information required for EPA registration. Field research to test extracts on crops grown under real-world conditions is being conducted in collaboration with business partners. Commercialization will benefit from optimizing extraction to produce higher SCN− concentrations in the product and eliminate the use of alcohol in the extracting solution. Also relevant but not discussed here, is the need to consider how the extracted and now detoxified seed meal may be used in other products such as animal feeds. Conflicts of interest The authors have no conflicts of interest to declare. Acknowledgements This project was supported by the Agriculture and Food Research Initiative competitive grant 2011-67009-20094 from the USDA National Institute of Food and Agriculture. The authors thank Bernardo Chaves-Cordoba for statistical analysis of the data and Treva Anderson for technical assistance. References Ahlgren, G.H., Klingman, G.C., Wolf, D.E., 1951. Principles of Weed Control. John Wiley
181