Phytotoxic Potential of Benzyl Salicylate and Benzyl Benzoate on Bioindicator and Invasive Species

The growing number of invasive species in agricultural areas reduces productivity and results in production losses. The need to discover new compounds with herbicidal activity increases as cases of resistance of invasive plants to herbicides rise. The aim of this study was to evaluate the phytotoxic potential of benzyl salicylate and benzyl benzoate upon the growth of Triticum aestivum coleoptiles and on the initial growth of Lactuca sativa, Lycopersicon esculentum, Allium cepa, T. aestivum, Euphorbia heterophylla, and Megathyrsus maximus. For the T. aestivum coleoptile bioassays, the treatments used the concentrations of 10 M, 3 × 10 M, 10 M, 3 × 10 M, and 10 M; while for the initial growth bioassays the concentrations of 10 M, 10 M, and 10 M were used. Both compounds presented a minimum of 89% growth inhibition on T. aestivum coleoptiles in all concentrations. Both compounds inhibited the growth of the root system and shoot of A. cepa and E. heterophylla at all concentrations. The species most affected by both compounds in all evaluated parameters was E. heterophylla. For the benzyl benzoate, the inhibition of the roots of E. heterophylla were statistically equivalent to those obtained with the herbicide. Regarding benzyl salicylate, the root inhibition in this species in the 10 M and 10 M treatments did not differ statistically from the herbicide in the same concentrations. Benzyl salicylate and benzyl benzoate are compounds that presented phytotoxic activity on E. heterophylla and for the first time the phytotoxic effect of these compounds on invasive species is reported.


Introduction
The increase in the number of invasive species in agricultural areas causes a reduction in the productivity of agricultural crops resulting in economic losses (Lamego et al., 2013). These plants, besides competing for space with the crop, also compete for nutrients, water, light (Galon et al., 2018), and may also host pests and diseases common to the crop (Alvino et al., 2011;Vasconcelos, Silva, & Lima, 2012). The need to discover the herbicidal potential of new compounds increases as cases of resistance to commercial products rise. Brazil is the fifth country with the largest number of herbicide-resistant species, with 50 plant species, behind the United States (165), Australia (95), Canada (68), and France (55) (Heap, 2020). These numbers increase more and more due to bad agricultural practices such as the recurrent use of the same herbicide or herbicides with the same mechanism of action in the same agricultural area. This leads to the selection of herbicide-resistant invasive plant biotypes for a particular mechanism of action. supplier (benzyl salicylate 98% from Sigma-Aldrich and benzyl benzoate 99% from Neon Commercial Ltda). All experiments were carried out in a climate-controlled chamber type B.O.D under controlled conditions.

Triticum aestivum Coleoptile Bioassays
The bioassays were performed with wheat diaspores (T. aestivum L: Poaceae), BRS 264 cultivar, pre-twinned in plastic boxes covered with 2 sheets of filter paper, moistened with distilled water (±12 mL) and stored in a B.O.D. for 72 hours at 24±1 °C in the absence of light.
After the time spent in the B.O.D. chamber, the coleoptiles were cut in a Van der Weij guillotine and the apical part (2 mm) was discarded as recommended by Hancock, Barlow, and Lacey (1964). After this stage, the coleoptiles were cut into 4 mm segments and used in bioassays (Macías et al., 2010). The assembly of this experiment was carried out in a closed environment with green safety light (Nitsch & Nitsch, 1956).
Three repetitions were used in each treatment, each one being composed of five coleoptiles of T. aestivum, kept in test tubes containing 2 mL of solution. The samples, properly identified, were kept at 25±1 °C, in the dark, and with a constant rotation of 1.4 rpm for 24 h, in order to guarantee the homogeneity of the contact between the plant material and the solution. After that period the coleoptiles were removed, photographed, and measured using the Image J® 1.8.0 software (Macías et al., 2010).
To evaluate the effect of the compounds on the initial growth of invasive plants, two invasive target species were used: the eudicotyledon wild poinsettia (E. heterophylla L. Euphorbiaceae) and the monocotyledon guinea grass (M. maximus J. Poaceae). In addition, four target bioindicator species were also employed, two eudicotyledons, lettuce (L. sativa L. Asteraceae) and tomato (L. esculentum L. Solanaceae); and two monocotyledons, wheat (T. aestivum L. Poaceae) and onion (A. cepa L. Amaryllidaceae). These species are considered indicators of phytotoxic activity because they present important characteristics such as rapid and uniform germination and sensitivity to phytotoxic substances even in low concentrations (Ferreira & Áquila, 2000;Souza Filho et al., 2010).
The diaspores were placed in plastic boxes for germination and moistened with distilled water. After the emergence of the primary root (2 mm) and the identification of positive gravitropism, the diaspores were considered germinated (Brazil, 2009). Then, the diaspores that germinated were transferred to transparent plastic boxes (10.5 × 6 × 4.5 cm) containing two sheets of filter paper moistened with 6 mL of each solution.
Each treatment contained four replicates with ten seedlings of the target species. The boxes with the respective treatments were identified, closed, and kept in a germination chamber at 27±1 °C, under a 12-hour photoperiod, according to Inoue et al. (2010). After eight days, the material was removed from the germination chamber, and measurements were taken of the length of the shoot and the main root (for eudicotyledons L. sativa, L. esculentum, and E. heterophylla) and of the shoot and the longest root (for monocotyledons T. aestivum, A. cepa, and M. maximus).
This step was performed with the aid of the Image J® 1.8.0 software. In addition, in the presence of abnormalities, they were photographed, counted, and characterized as described by Brazil (2009).

Statistical Analysis
The data obtained were submitted to the Shapiro-Wilk normality test using the Past v.2.17c software (Hammer, Harper, & Ryan, 2001). Normal data were subjected to Analysis of Variance (ANOVA), followed by the Tukey test at 5% significance with the aid of the Sisvar software (Ferreira, 2011). Non-normal data were submitted to    Note. 1 * values reported in the literature  and 2 * values obtained in the present study.
Other authors used pure compounds extracted from different plant species and found their phytotoxic effects on the growth of T. aestivum coleoptiles (Anese, Jatobá, Grisi, Gualtieri, Santos, & Berlinck, 2015;Miranda et al., 2015;Jatobá et al., 2016;Silva et al., 2017). Thus, they confirmed that the initial growth bioassays of T. aestivum coleoptiles are a classic and efficient method to evaluate the phytotoxic activity of compounds belonging to the secondary plant metabolism (Hancock et al., 1964). Bioassays using T. aestivum coleoptiles have the advantage of being fast and sensitive to a wide variety of bioactive substances (Nepomuceno, 2011). In addition, they are widely used because they have proved to be able to evaluate the inhibition or growth stimulus when exposed to phytotoxic substances (Accarini, 2016).
In the initial growth bioassays of L. esculentum both compounds inhibited root growth at all concentrations. While shoot growth was significantly inhibited only by benzyl benzoate at concentrations of 10 -3 M and 10 -4 M. These results corroborated those of Candido et al. (2016) who tested the same compounds and observed root and shoot inhibition of this species at the highest concentration (10 -3 M). However, the data of these authors presented a notable decrease in root growth inhibition as the concentrations decreased, as well as stimulation of shoot growth. While the results of the present work remained uniform over the different concentrations.
In general, the four bioindicator species showed uniform inhibition values between treatments with the application of the same compound, even at the lowest concentrations. Corroborating data from the literature that tested different pure compounds from plant extracts. The authors used these bioindicator species and reported root and shoot inhibition mainly at the highest concentration tested. They also reported that there was a rapid decrease in inhibition when the compounds used were diluted, and in some cases, growth stimulus was observed (Miranda et al., 2015;Silva et al., 2017).
This may emphasize what was previously mentioned. The high purity of the compounds used in this work has interfered in this observed difference, since the data in the literature are related to compounds isolated from plant species. The methods employed to isolate these compounds may leave them with impurities (Leite, 2009). Compared to a synthetic compound with a high degree of purity, such as the ones employed in this study. Other factors may have contributed to the differences observed, such as the variety of the species used, the temperature at which the experiments were conducted, the length of the experiments, as well as the type of compound used. This shows that the response of the plant to secondary metabolites is a species-specific characteristic, in which some species being more sensitive to certain compounds or molecules than others (Ferreira & Áquila, 2000).
Among the invasive species, M. maximus was the least affected by both compounds, with the roots being the most affected parameter, in accordance with the results of Candido (2016). Nevertheless, this author, who used leaf extracts of O. pulchella containing benzyl salicylate and benzyl benzoate in their composition, observed a greater susceptibility of the M. maximus roots to these extracts. On the other hand, E. heterophylla obtained a high percentage of inhibition both in the root and in the shoot, being as efficient as the herbicide mainly in the root system. Confirming the results of Candido (2016), who observed the same inhibitory effect on this plant species, mainly in shoot.
In addition, the same author mentioned that the length reduction of E. heterophylla roots caused by leaf extracts of O. pulchella is due to the decrease in the elongation of metaxylem cells. The presence of benzyl salicylate and benzyl benzoate may have resulted in a lower absorption of water and minerals by the roots. Since the action of the allelochemicals is more evident on these structures, because they have a direct contact with the solution of allelochemicals (Lupini, Sorgonà, Miller, & Abenavoli, 2010). However, Carvalho, Carvalho, Abbade Neto, & Teixeira (2014) highlighted that this condition makes plants more sensitive, since their elongation depends on cell divisions. Once inhibited they affect the normal development of the plant, resulting in a reduction of the roots and stems of the seedlings (Bogatek, Gniazdowska, Zakrzewska, Oracz, & Gawronski, 2005;Kenany & Darier, 2013).  Vol. 12, No. 12;2020 20 Compounds with phytotoxic activity can inhibit cell elongation. This might be associated with changes in the concentration of plant hormones such as cytokinins and auxins, since these are of vital importance in the development of roots, vascular differentiation, and in the gravitropism of plants (Gatti, Ferreira, Arduin, & Perez, 2010;Grisi, Gualtieri, Anese, Pereira, & Forim, 2013).
Cell growth in plants is dependent on the mitotic process. When cell divisions are compromised during germination, most seedlings die or present abnormalities (Imatomi, Novaes, & Gualtieri, 2013). Some allelochemicals, besides affecting seedling growth, can completely inhibit the formation of normal seedlings (Juchem, Lauxen, Silva, Denardin, & Sobral, 2013). Causing morphological changes such as root atrophy (Formigheiri et al., 2018), little or no root development, necrotic apexes, and hypocotyl winding, which can impair the emergence of the seedlings from the soil (Denardin et al., 2018).
The compounds benzyl salicylate and benzyl benzoate induced the appearance of several abnormalities and dead seedlings in the species tested with a higher incidence in E. heterophylla. Corroborating the results obtained by Candido (2016), who observed several abnormalities in this species, with the highest percentage of anomalies found in the root system, using O. pulchella leaf extracts.
Allelochemicals can cause changes in the plant growth and development such as cellular alterations, changes in the enzymatic activities of the phenylpropanoid pathway, reduction of leaf expansion and root elongation, reduction of the photosynthetic rate, disintegration of the cap, increase in the diameter of the vascular cylinder, and premature lignification of the metaxylem and cell wall. This may impair the nutrient absorption ability, the expansion of cells, and the ability to sustain plant growth (Soltys et al., 2011;Granã et al., 2013;Ferro et al., 2015;Bido et al., 2018).
The high percentage of abnormalities found in E. heterophylla in the present study, mainly in the root system, may reveal the action of both compounds on this species. Also, some of the abnormalities observed may seriously compromise the development of the plant. The growth of the root system is characterized by a high metabolic activity and for this reason, the roots are particularly more susceptible to stress by allelochemicals. Some studies have indicated that the impairment of root system growth by the action of allelochemicals is associated with premature lignification of cell walls (Politycka & Mielcarz, 2007;Bubhna et al., 2011).
Abnormal seedlings are those with a damaged or absent primordial structure, that present fragile development or are so deteriorated that their normal development is compromised (Brazil, 2009). The evaluation of abnormalities in seedlings is a fundamental parameter in detecting the phytotoxic effects of allelochemicals. Since they can induce the appearance of abnormalities, thus inhibiting their normal development (Ferreira & Aquila, 2000).
In general, the four bioindicator species and the invasive M. maximus presented a low percentage of abnormalities, as well as a low or zero percentage of dead seedlings, as a result of the compounds used. No pattern was observed regarding the presence of abnormalities between monocots and eudicots, with E. heterophylla being the species most sensitive to treatments with both compounds. Candido (2016) found abnormalities in the seedlings of M. maximus and E. heterophylla, when subjected to the action of O. Pulchella extracts. Favaretto (2018) described some abnormalities such as root necrosis, absence of secondary roots, and short and thick roots in L. sativa seedlings, when treated with different concentrations of catechin, caffeic acid, and vanillic acid. Likewise, Melo et al. (2017) found abnormalities in this species when treated with different concentrations of Curcuma zedoaria essential oil. However, the same authors reported that L. esculentum seedlings did not show changes in the percentage of abnormal seedlings when exposed to the same treatments with this essential oil. This information shows that tolerance or resistance to secondary metabolites is a species-specific characteristic, in which some species are more sensitive to certain compounds or molecules than others (Ferreira & Áquila, 2000), as observed in the results of the present study.
Information on the mode of action of some allelochemicals on plants is scarce. Since these substances affect several functions and cause many side effects, making them difficult to distinguish from other effects (Goldfarb, Pimentel & Pimentel, 2009). Phytotoxic substances may or may not inhibit germination, besides inducing the appearance of abnormal seedlings. Therefore, the evaluation of seedling abnormalities is an important tool for the identification of phytotoxic substances (Ferreira & Áquila, 2000;Alves, Oliveira, França, Alves, & Pereira, 2011).
Oliveira Jr., Pereira, Muller, and Matias (2014), mentioned that some compounds may be toxic for one species and not have the same effect on another one, even acting as a stimulant (Almeida, 1988). It is important to mention that results obtained in the laboratory involving phytotoxic activity may be different under natural jas.ccsenet.org Journal of Agricultural Science Vol. 12, No. 12; conditions, since the simultaneous occurrence of biotic and abiotic factors may interfere with the results (Formagio et al., 2010).

Conclusions
This study provided knowledge on the effect of the tested compounds upon plant species, reporting for the first time their effect on invasive species. The results obtained showed that the benzyl salicylate and benzyl benzoate compounds have phytotoxic potential. This effect was more evident in E. heterophylla since it was the most affected species according to all the evaluated parameters, and in some parameters, the inhibitory effects of both compounds have proved to be as efficient as the herbicide, therefore confirming the phytotoxic potential of both compounds.