Introduction
Synthetic herbicides are known to cause many environmental problems as they accumulate in soil (Bhowmik and Inderjit, Reference Bhowmik and Inderjit2003; Shah et al., Reference Shah, Iqbal, Ullah, Yang, Yousaf, Fahad, Tanveer, Hassan, Tung, Wang, Khan and Wu2016). These herbicides are widely and continuously used in agriculture to deal with undesirable weeds, which in turn affects the balance of agricultural fields and neighboring ecosystems, changing soil physico-chemical and biological properties, and increasing the ecotoxicological risk for aquatic and terrestrial non-target organisms, including plants, animals, soil microorganisms and even humans (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b; Mehdizadeh et al., Reference Mehdizadeh, Mushtaq, Siddiqui, Ayadi, Kaur, Yeboah, Mazraedoost, Al-Taey and Tampubolon2021; Monteiro and Santos, Reference Monteiro and Santos2022). Besides, several weeds causing economically crop losses, such as Lolium rigidum, have evolved resistance to synthetic herbicides (Owen and Zelaya, Reference Owen and Zelaya2005; Taberner Palau et al., Reference Taberner Palau, Cirujeda Ranzenberger and Zaragoza Larios2007). Therefore, the demand for new environmentally safe, low risk alternatives, the so-called bioherbicides, is nowadays increasing. Bioherbicides are biologically based control agents for weeds (Soltys et al., Reference Soltys, Krasuska, Bogatek, Gniazdowsk, Price and Kelton2013) that can be used alone or together with other practices (integrated weed management) to reduce the use of synthetic herbicides (Scavo and Mauromicale, Reference Scavo and Mauromicale2020). Their main downside is that the process of development and commercialization takes many years and the costs associated are rather high (Kremer, Reference Kremer2005; Weaver et al., Reference Weaver, Lyn, Boyette, Hoagland, Upadhyaya and Blackshaw2007; Bailey, Reference Bailey and Dharam2014).
Some of the characteristics of a good bioherbicide comprise: (1) efficacy on target species, (2) no effect on crop species or other non-target plants growing around the application, (3) limited reproduction if it is a plant-based bioherbicide, (4) low persistence in soil, and therefore the least possible legacy effect on the ground (Bailey, Reference Bailey and Dharam2014). Plant-based bioherbicides have the following advantages regarding other bioherbicides: (1) many plant phytotoxic compounds are soluble in water, which avoids the use of a surfactant for its application, and (2) the mode of action of allelochemicals is similar to that of synthetic herbicides, but they are safer because of their low persistence and activity in soil (Soltys et al., Reference Soltys, Krasuska, Bogatek, Gniazdowsk, Price and Kelton2013). There is a consistent literature that compiles traditional agronomic uses of plants as bioherbicides (Araniti et al., Reference Araniti, Sorgonà, Lupini and Abenavoli2012; Pacanoski, Reference Pacanoski, Price, Kelton and Sarunaite2015; Puig et al., Reference Puig, Gonçalves, Valentão, Andrade, Reigosa and Pedrol2018; Souza-Alonso et al., Reference Souza-Alonso, Puig, Pedrol, Freitas, Rodríguez-Echeverría and Lorenzo2020), but it is necessary to explore the efficacy of each donor plant in each specific context and the best application techniques (e.g., aqueous extracts or mulches; pre-emergence or postemergence) (Soltys et al., Reference Soltys, Krasuska, Bogatek, Gniazdowsk, Price and Kelton2013; Marble, Reference Marble2015). Bioherbicides based on the allelopathic compounds naturally produced by plants can be cheap and environmentally friendly alternatives (Albuquerque et al., Reference Albuquerque, Santos, Lima, de Melo Filho, Nogueira, Câmara and Ramos2011; Soltys et al., Reference Soltys, Krasuska, Bogatek, Gniazdowsk, Price and Kelton2013). Particularly, cover crops and mulches are ecofriendly choices for sustainable agriculture that can be used not only for weed control but also as fertilizers to improve crop performance (Campiglia et al., Reference Campiglia, Mancinelli, Radicetti and Caporali2010).
Allelopathy can be defined as the ability of plants to inhibit the survival, growth, development and reproduction of other organisms through the release of chemical compounds present in their tissues (Lorenzo and González, Reference Lorenzo and González2010; Cheng and Cheng, Reference Cheng and Cheng2015). Several plant species can use allelopathy as a strategy to compete with other plants for environmental resources, but in some species this strategy is more relevant (Pisula and Meiners, Reference Pisula and Meiners2010). Additionally, some plant species can be more susceptible to the allelopathic compounds released to the environment by other plants (Medina-Villar et al., Reference Medina-Villar, Alonso, Castro-Díez and Pérez-Corona2017). Also, the quantity and composition of allelochemicals greatly vary among plant species, plant organs and specific contexts (Cappuccino and Arnason, Reference Cappuccino and Arnason2006; Pisula and Meiners, Reference Pisula and Meiners2010; Bauer et al., Reference Bauer, Shannon, Stoops and Reynolds2012; de las Heras et al., Reference de las Heras, Medina-Villar, Pérez-Corona and Vázquez-de-Aldana2020). Allelopathy is a phenomenon with ecological implications, as it influences the species distribution in a community (Hierro and Callaway, Reference Hierro and Callaway2021). Therefore, studying how species chemically interact with each other by means of allelopathy it is essential to better understand their role structuring plant communities in different ecosystems, including agroecosystems, but also, this knowledge can be applied to map adequate candidates to be used as bioherbicides for a more sustainable agriculture (Albuquerque et al., Reference Albuquerque, Santos, Lima, de Melo Filho, Nogueira, Câmara and Ramos2011; Hasan et al., Reference Hasan, Ahmad-Hamdani, Rosli and Hamdan2021; Lopes et al., Reference Lopes, Marques Morais, de Lacerda and da Araújo2022).
Exotic invasive plants arise as good candidates to study new bioherbicides because of different reasons. First, many invasive plant species are allelopathic, some of them with high allelopathic potential, such as Ailanthus altissima (Mill.) Swingle (Pisula and Meiners, Reference Pisula and Meiners2010; Chengxu et al., Reference Chengxu, Mingxing, Xuhui and Bo2011; Kalisz et al., Reference Kalisz, Kivlin and Bialic-Murphy2021). Second, there are numerous exotic invasive plant species causing serious ecological and socio-economic impacts worldwide, which need to be managed and eliminated for native ecosystem recovery (Pimentel et al., Reference Pimentel, Zuniga and Morrison2005; Pyšek et al., Reference Pyšek, Jarošík, Hulme, Pergl, Hejda, Schaffner and Vilà2012; Seebens et al., Reference Seebens, Blackburn, Dyer, Genovesi, Hulme, Jeschke and Essl2017). The measures to eliminate these species are extremely difficult and expensive and involve the extraction of a huge amount of biomass, which is frequently burned releasing CO2 to the atmosphere (Pimentel et al., Reference Pimentel, Zuniga and Morrison2005; Lovell et al., Reference Lovell, Stone and Fernandez2006; Villamagna and Murphy, Reference Villamagna and Murphy2010). Therefore, using the removed biomass of invasive species, instead of burning it, provides an environmentally friendly and cost-effective solution for weed management. Some studies verified the effectiveness of using invasive species, such as Parthenium hysterophorus L., Acacia dealbata Link., Eucalyptus globulus Labill. or Cytisus scoparius (L.) Link, for weed management in agroecosystems without endangering the culture species (Singh et al., Reference Singh, Batish, Pandher and Kohli2005; Marwat et al., Reference Marwat, Khan, Nawaz and Amin2008; Puig et al., Reference Puig, Álvarez-Iglesias, Reigosa and Pedrol2013; Souza-Alonso et al., Reference Souza-Alonso, Puig, Pedrol, Freitas, Rodríguez-Echeverría and Lorenzo2020; Pardo-Muras et al., Reference Pardo-Muras, Puig, Souto and Pedrol2020a, Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). The alterations of soil properties produced by using plant material from invasive species should also be considered (legacy effects), as for example, the possible increase in soil nitrogen let by the tissues from Fabaceae plants, able to fix N2 from the atmosphere (Elgersma et al., Reference Elgersma, Ehrenfeld, Yu and Vor2011; Grove et al., Reference Grove, Haubensak and Parker2012, Reference Grove, Parker and Haubensak2015; Von Holle et al., Reference Von Holle, Neill, Largay, Budreski, Ozimec, Clark and Lee2013). Besides, increases in soil nitrate may interfere with allelopathic effects by promoting seed germination (Duermeyer et al., Reference Duermeyer, Khodapanahi, Yan, Krapp, Rothstein and Nambara2018). From an ecological point of view, studying how an exotic invasive plant interacts with other plants (e.g., weeds, crop species) is interesting to better understand the bio-properties of the invasive species and its capability to impact plant diversity and agroecosystems.
A N2-fixing species, able to alter soil properties is Ulex europaeus L. (Fabaceae), commonly known as ‘gorse’ (Bateman and Vitousek, Reference Bateman and Vitousek2018). It is a native species from NW Europe and a serious invader in many world regions, such as Chile, Australia, Sri Lanka, Hawaii or in the west coast of USA (Clements et al., Reference Clements, Peterson and Prasad2001; Bateman and Vitousek, Reference Bateman and Vitousek2018). This species has been intentionally introduced in many countries as a livestock fodder plant or as a living fence, but it has also arrived accidentally by zoochory in animal's fur, being now an invasive species distributed worldwide (Norambuena and Piper, Reference Norambuena and Piper2000; Clements et al., Reference Clements, Peterson and Prasad2001). This plant species is able to affect soil properties in the invaded ecosystems, e.g., reducing soil pH (Bateman and Vitousek, Reference Bateman and Vitousek2018), and it has been classified as one of the 100 worst invasive species of the world by the International Union for Conservation of Nature (IUCN/SSC, 2000). For instance, in the case of Chile, if no action is taken, the United Nations Development Program (UNDP) has estimated the economic loss associated with the presence of U. europaeus in more than 49 millions of dollars in the next two decades (IUCN/SSC, 2000). Considering the ecological and socio-economic impact caused by U. europaeus (Galappaththi et al., Reference Galappaththi, de Silva and Clavijo Mccormick2022), the need to eradicate or reduce its density is evident.
To recover management costs, the use of the extracted biomass of U. europaeus as biofuel (Viana et al., Reference Viana, Vega-Nieva, Ortiz Torres, Lousada and Aranha2012), fertilizer (Galappaththi et al., Reference Galappaththi, de Silva and Clavijo Mccormick2022), promotor of secondary metabolism (Tighe-Neira et al., Reference Tighe-Neira, Díaz-Harris, Leonelli-Cantergiani, Iglesias-González, Martínez-Gutiérrez, Morales-Ulloa and Mejías-Lagos2016) in crops and bioherbicide (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souto and Pedrol2020a, Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b) has already been suggested. The ability of its compounds (volatile oxygenated monoterpenes and water-soluble phenolic compounds) to significantly inhibit germination and growth of Amaranthus retroflexus L. and Digitaria sanguinalis (L.) Scop. has been reported (Pardo-Muras et al., Reference Pardo-Muras, Puig, Lopez-Nogueira, Cavaleiro and Pedrol2018, Reference Pardo-Muras, Puig and Pedrol2019, Reference Pardo-Muras, Puig, Souto and Pedrol2020a, Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). Besides, one of the most prominent secondary compounds in U. europaeus, known as maackiain (Cappuccino and Arnason, Reference Cappuccino and Arnason2006), extracted from another Fabaceae species, Trifolium pratense L., showed phytotoxic effects on grass species (Liu et al., Reference Liu, Xu, Yan, Jin, Cui, Lu, Zhang and Qin2013). The efficacy and potential pre-emergence use of fresh plant material from flowering U. europaeus against weeds in maize crops was demonstrated in a previous pot experiment conducted by Pardo-Muras et al. (Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). However, different target species, experimental conditions and approaches may greatly change the outputs in allelopathic studies (Haugland and Brandsaeter, Reference Haugland and Brandsaeter1996; Kobayashi, Reference Kobayashi2004; Zhang et al., Reference Zhang, Liu, Yuan, Weber and van Kleunen2021).
Here we suggest the use of different target species and experimental conditions as those used in the studies of Pardo-Muras et al. (Reference Pardo-Muras, Puig, Lopez-Nogueira, Cavaleiro and Pedrol2018, Reference Pardo-Muras, Puig and Pedrol2019, Reference Pardo-Muras, Puig, Souto and Pedrol2020a, Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). These conditions raise new and relevant insights to study the allelopathic potential of U. europaeus and its possible use as a bioherbicide. For instance, we used dried instead of fresh plant material because it is a useful way to conserve and store the surplus plant material before its use, preventing decomposition and keeping allelopathic potential (Bonanomi et al., Reference Bonanomi, Incerti, Abd El-Gawad, Cesarano, Sarker, Saulino, Lanzotti, Saracino, Rego and Mazzoleni2018; Gatto et al., Reference Gatto, Veiga, Higaki, Swiech, de Bona Sartor, Gribner and Miguel2021). We also used activated carbon (AC), as an allelopathy neutralizer, commonly used to discriminate if the effects of plants are driven by allelopathic compounds (Sturm et al., Reference Sturm, Peteinatos and Gerhards2018; Lorenzo et al., Reference Lorenzo, González and Ferrero2021). Moreover, as the use of U. europaeus materials is relevant in invaded areas, because there is already an enormous quantity of undesirable U. europaeus plants needed to be removed, the plants used in invaded areas need to be cut before seeding, and even before flowering, to avoid any possibility to spread seeds or add invasive seeds to the crop field, which could increase the invasion problem and affect the crop yield. Therefore, we were interested in assessing whether the vegetative part of U. europaeus is allelopathic. Harvesting and transporting big amount of biomass of an invasive plant species from its established habitat to crop fields would inevitably cost farmer extra money. Thus, in our opinion the best way to act, an also the most sustainable, is using the extracted plant material on crops close to geographical areas that are already invaded by U. europaeus.
Having into account all these considerations, in this study, we aimed to evaluate the suitability of U. europaeus plants as a bioherbicide in oat crops (Avena sativa L., commonly named ‘oat’) as a case study, testing the allelopathic potential of dry plant material (vegetative parts) from U. europaeus on the following target plants belonging to the same family (Poaceae) and subfamily (Pooideae): two common weeds of oat crops, Lolium multiflorum Lam. and L. rigidum Gaud. (commonly referred to as ‘ryegrass’), and on the crop species. We also aimed to evaluate the soil legacy effect of U. europaeus. We selected oat as the crop species because it is a very common crop in several countries where U. europaeus is a troublesome invader that reaches high densities and hinders crop practices (Quiroz et al., Reference Quiroz, Pauchard, Marticorena and Cavieres2009; Koch et al., Reference Koch, Jeschke, Overbeck and Kollmann2016; ODEPA, 2019). L. rigidum and L. multiflorum are native from Mediterranean Basin and widespread grain weeds (Romero and Fraga, Reference Romero and Fraga1990; Diez de Ulzurrun and Leaden, Reference Diez de Ulzurrun and Leaden2012), coexisting with U. europaeus (Moreno-Chacón et al., Reference Moreno-Chacón, Mardones, Viveros, Madriaza, Carrasco-Urra, Marticorena and Saldaña2018). The allelopathic potential of U. europaeus and its soil legacy effect has not been previously evaluated on A. sativa and Lolium weeds.
Based on previous studies showing the ability of U. europaeus to negatively affect different herb species, but not the crop species (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souto and Pedrol2020a, Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b), we could expect less effectiveness of U. europaeus on oat than on Lolium weeds, but also bigger seed size of the former supports our expectation (Liebman and Sundberg, Reference Liebman and Sundberg2006). To firmly consider U. europaeus as a proper bioherbicide, negative effects need to be demonstrated on the weeds but not on the crop species and the soil legacy effect should be the minimum, reducing possible effects to non-target species. If so, this study will be the starting point to consider the implementation of the U. europaeus as a bioherbicide in oat crops. Besides, exploring the biological properties of invasive plant material contributes to understand the ability of U. europaeus to affect coexistent herb species by means of allelopathy or by changing other soil properties.
Material and methods
We evaluated in a greenhouse pot experiment two possible techniques to apply U. europaeus as a bioherbicide in crop fields: (a) direct use of dried U. europaeus material (hereby named as ‘mulch’) preemergence added and (b) preemergence mulch combined with a subsequent postemergence application of aqueous extracts from the mulch. We also analyzed the legacy effect (2 months period), let by U. europaeus (mulch + extracts) in the soils, on the germination of the target species.
Plant material
The donor species whose phytotoxic potential was tested is U. europaeus, a shrubby legume species with entomophilous pollination, from humid acidic habitats (Rapoport et al., Reference Rapoport, Marzocca and Drausal2009). Branches of U. europaeus were collected from at least 20 randomly selected individuals in non-herbivorized native populations located in NW Spain, specifically in Orense (N 42°18′21″; O −8°7′7″) and Pontevedra (N 42°11′53.6″; O –8°39′53″) in late Spring 2018 and 2019. In Orense, the mean annual temperature is 13.1°C and total annual precipitation is 1224 mm whereas in Pontevedra it is 14.9°C and 1303 mm, respectively (average data of the years 1982–2012; www.climatedata.org). The collected branches were oven dried (60°C, >48 h).
The target species (i.e., the ones on which it is desired to test the bioherbicide effect of U. europaeus) are three therophyte grasses (Poaceae family): the crop species A. sativa (‘oat’) and the weeds: L. rigidum and L. multiflorum (Romero and Fraga, Reference Romero and Fraga1990; Michitte et al., Reference Michitte, De Prado, Espinoza, Pedro Ruiz-Santaella and Gauvrit2007). The fact that the three target species are grasses will be worthwhile to better assess the selective capacity of the allelopathy effect of U. europaeus on different species of the same family. This will also eliminate possible different responses due to taxonomic peculiarities of the species.
A. sativa is cultivated in most temperate regions of the world (http://faostat.fao.org). L. rigidum is a very problematic weed, considered among the ten species with the highest resistance to herbicides including glyphosate (Lemerle et al., Reference Lemerle, Verbeek and Orchard2001; Soltys et al., Reference Soltys, Krasuska, Bogatek, Gniazdowsk, Price and Kelton2013). Multiple biotypes of L. multiflorum are also resistant to different herbicides of the families of ALS and ACCasa inhibitors and to glyphosate (Espinoza et al., Reference Espinoza, Díaz, Galdames, De Prado, Rodríguez and Ruiz2009; Diez de Ulzurrun and Leaden, Reference Diez de Ulzurrun and Leaden2012). Therefore, it seems an urgent need to find new effective herbicides to face weed resistance. If they are ecofriendly alternatives, we would reduce negative impacts in crops and ecosystems.
Seeds from L. rigidum and L. multiflorum were purchased in Semillas Silvestres (www.semillassilvestres.com) and those from A. sativa in Fitoagrícola (www.fitoagricola.net). As commercial seeds, their viability is ensured. They were all disinfected before use with consecutive 1-min baths of sodium hypochlorite (50%) and ethanol (69%) to reduce fungus proliferation during the germination bioassays. Finally, they were rinsed with plenty of deionized water. The allelopathic effect of U. europaeus was tested at different development stages of the target plants (germination, seedling emergence and plant growth).
Mulch and extract preparation
Mulch for the experiments was prepared using dried branches, thorny twigs and phyllodes from U. europaeus, that were cut in fractions (c.a. 2 cm) to simulate the crushing or chopping process traditionally carried out by an electric fodder cutter after U. europaeus remotion for agricultural application (Jamil et al., Reference Jamil, Cheema, Mushtaq, Farooq and Cheema2009; Atlan et al., Reference Atlan, Udo, Hornoy and Darrot2015; Khan et al., Reference Khan, Afridi, Hashim, Khattak, Ahmad, Wahid and Chauhan2016). Also, as a spiny shrub, U. europaeus would be difficult to farmers to handle if it is not cut. This fraction size of the mulch was used in similar agricultural studies (Jamil et al., Reference Jamil, Cheema, Mushtaq, Farooq and Cheema2009; Puig et al., Reference Puig, Álvarez-Iglesias, Reigosa and Pedrol2013; Stagnari et al., Reference Stagnari, Galieni, Speca, Cafiero and Pisante2014; Souza-Alonso et al., Reference Souza-Alonso, Puig, Pedrol, Freitas, Rodríguez-Echeverría and Lorenzo2020). Flowers and fruits were discarded to prepare the mulch, as the focus in this study was on the vegetative part. As explained before, for field studies in invaded areas, U. europaeus plants need to be collected out of their fructification and even floriation period to avoid reproduction of the mulch in the cultivar.
Aqueous extract was prepared with U. europaeus mulch at a concentration of 10% (10 g of U. europaeus mulch per 100 ml of deionized water), keeping the mixture or ‘tea’ stirring for 24 h at 90 rpm. Similar concentrations have been used in previous phytotoxicity studies (e.g., Singh and Sangeeta, Reference Singh and Sangeeta1991; Cheema and Khaliq, Reference Cheema and Khaliq2000; Jamil et al., Reference Jamil, Cheema, Mushtaq, Farooq and Cheema2009; Soltys et al., Reference Soltys, Krasuska, Bogatek, Gniazdowsk, Price and Kelton2013). After stirring time, with the help of a suction pump, the resulting liquid was filtered in a MILLIPORE Express® PLUS container with a 0.22 μm filter. The extract was preserved at 4°C until use the day posterior to the preparation.
Effect of U. europaeus mulch on target species
Thermoformed pots (11 × 11 × 12 cm; base with aeration) were filled in as follows: (1) only with commercial substrate, (2) with commercial substrate and AC—20 ml per liter of substrate (Callaway, Reference Callaway2003)—with a homogeneous mix of commercial substrate and mulch from U. europaeus—4 g of mulch per kg of substrate (Singh et al., Reference Singh, Batish, Pandher and Kohli2005)— and (4) with a homogeneous mix of AC-supplemented substrate and mulch from U. europaeus. The commercial substrate was a blend of commercial growth media (50% fine blond peat and 50% black peat; Projar Professional Seed Pro 5050; www.projar.es) supplemented with 10% (v/v) vermiculite to avoid desiccation. Therefore, this fully factorial design includes, for each target species, two factors with two levels each: (1) U. europaeus mulch (presence or absence; hereafter called as ‘Ulex’) and (2) ‘AC’ (presence or absence). A total of 120 pots (10 × 3 target species × 2 AC × 2 Ulex) were placed in a greenhouse at Real Jardín Botánico Alfonso XII (Complutense University of Madrid, Spain). Twenty disinfected seeds of each target species (A. sativa, L. rigidum y L. multiflorum) were sown (c.a. 2 cm deep) in each pot. Several studies used AC as a substance to reveal allelopathic effects because this substance can adsorb organic compounds released by plant species (Tian et al., Reference Tian, Feng and Chao2007; Yuan et al., Reference Yuan, Wang, Zhang, Tang, Tu, Hu, Yong and Chen2013; Del Fabbro and Prati, Reference Del Fabbro, Güsewell and Prati2014).
Pots were weekly watered enough to maintain optimal humidity conditions for plant growth, and they were frequently relocated in the greenhouse to homogenously distribute possible microenvironment effects or border effects among pots. Seedling emergence was daily monitored. For each target species and pot, the emergence percentage and emergence speed were then calculated. After 1 month since seed sown, plant height was measured and averaged per pot, and after 48 days since seed sown, we kept four plants per pot, the biggest ones, which ranged from 12 to 16 cm. The rest of the plants were harvested, oven-dried (48 h at 60°C) and weighted. To obtain the aerial biomass per plant, the weight was divided by the number of plants harvested in each pot. For each variable the number of replicates per treatment was 10 (10 pots).
Combined effect of mulch and extract from U. europaeus on target species
After 48 days since the seed sown, with the help of an aerosol vaporizer, we added to the pots with the remaining four plants the following: (1) extract from U. europaeus (10%) in the pots with the presence of U. europaeus mulch and (2) deionized water in the pots without the mulch. We applied 2 ml per day and pot (in two consecutive days), keeping the extract cold (4°C) from one day to the next (Tighe-Neira et al., Reference Tighe-Neira, Díaz-Harris, Leonelli-Cantergiani, Iglesias-González, Martínez-Gutiérrez, Morales-Ulloa and Mejías-Lagos2016). After 28 days since the vaporization of the extract, i.e., 76 days from the seed sown, plants were harvested, separating the above- and belowground biomass. Roots were washed to remove the residual substrate. Plant aboveground and belowground biomass were dried in the oven at 60°C for 48 h and weighed. We divided the plant weight by the number of remained plants (1–4) in each pot (number of replicates per treatment = 10). Most of the four remained plants survived, except for three plants of L. multiflorum in one pot and one plant of L. rigidum in other pot. Therefore, we did not consider pertinent analyzing differences in seedling survival among treatments.
Soil legacy effects
The substrates where the species grew were collected after plant harvest and kept in a freezer at −20°C until its use. Treatments to test mulch and extract legacy effect left on substrate were: (1) substrate, (2) substrate + AC, (3) substrate + Ulex (mulch and extract) and (4) substrate + Ulex (mulch and extract) + AC. In Petri dishes (Ø 10 cm) we added 40 ml of substrate of each treatment substrate (5 replicates), which was composed by an equal mixture of two randomly selected replicates of the same treatment. On the surface of each substrate, 20 seeds of each target species were placed, always placing each target species in the soil where it had grown. A total of 20 Petri dishes by target species: 5 × 2 Ulex (with or without mulch + extract) × 2 AC treatments (with and without). The substrate was moistened with 4 ml of deionized water. Petri dishes were kept in a germination chamber, in the dark, at 20°C. The number of germinated seeds was recorded each day until no germination of any seed was observed after 3 days. During this time, dishes were moistened with deionized water as required. For each species, the germination percentage (%G) and germination speed (GE) were then calculated.
Variable's calculation and statistical analysis
Final percentage of emerged and germinated plants (final number of emerged or germinated seedling × 100/total number of sown seeds) were calculated. Emergence speed (ES) and germination speed (GS) were as ES or GS = [N1 + N2/2 + N3/3 + … + Nn/n] × 100, where N1, N2, N3, Nn, are the proportions of emerged seedlings or germinated seeds at 1, 2, 3, …, n days (Wardle et al., Reference Wardle, Ahmed and Nicholson1991). ES and GS range from 0 (if no seedlings emerge or germinate, respectively, at the end of the study period) to 100 (if all seedlings emerge or germinate, respectively, on the first day). To check variables normality and homoscedasticity, Kolmogorov–Smirnov and Levene tests were done and when required, transformation of the variables (Arcsen √variable/100) was done. For each target species (L. multiflorum, L. rigidum and A sativa), a two-way ANOVA was used to assess differences in emergence (%E), germination (%G), ES, GS, height and biomass among treatments: AC (presence–absence) and Ulex (presence–absence) and the interactions among them. After ANOVA, Least Significance Difference (LSD) post hoc test was used for multiple comparisons between factor levels. A permutational multivariate analysis of variance (PERMANOVA) using distance matrices was implemented to assess the influence of Ulex and CA accounting for all the variables: emergence, ES, height and biomass of the target species. All statistical analyses were performed using StatSoft Statistica software except for PERMANOVA that was performed in R software 3.4.3 (R Core Team, 2022) using ‘adonis2’ function (vegan package).
Results
Effect of U. europaeus mulch on target species
Both Ulex and AC significantly affected biomass of L. rigidum (Table 1). The interaction between U. europaeus mulch and AC (Ulex × AC) significantly affected the percentage of emergence (%E) and the emergence speed (ES) of L. rigidum and the height and aboveground biomass of L. multiflorum (Table 1). U. europaeus mulch decreased the %E, ES and height of L. rigidum, but the effect was absent in the presence of AC (Fig. 1), indicating that these effects were likely caused by allelopathic compounds possibly released by U. europaeus mulch. In the case of L. multiflorum, U. europaeus mulch also decreased height, but differences were only significant in the presence of AC (Fig. 1). U. europaeus mulch increased while AC decreased L. rigidum biomass (Fig. 1). In the absence of U. europaeus mulch and AC, A. sativa and L. multiflorum developed less biomass (Fig. 1).
Fig. 1. Mean values (±SE, N = 10) of (a) the percentage of emergence, (b) emergence speed (ES) and (c) height and (d) aboveground biomass of the target species (Avena sativa, Lolium multiflorum y Lolium rigidum) grown in the presence or absence of U. europaeus mulch (Ulex and No Ulex, respectively) and with or without activated carbon (AC) (grey and white bars, respectively). Different letters stand for statistically significant differences among treatments (Ulex × AC) at P < 0.05 (LSD test).
Table 1. Summary results of the two-way ANOVA assessing the effects of mulch from U. europaeus (Ulex), activated carbon (AC) and their interactions on the percentage of emergence (%E), emergence speed (ES) and height of each target species (Avena sativa, Lolium multiflorum and Lolium rigidum)
Significance level (P): *** <0.001; ** <0.01; * <0.05.
Combined effect of mulch and extract from U. europaeus on target species
The combined effect of U. europaeus mulch and extract (Ulex) significantly affected belowground and total biomass of L. rigidum and L. multiflorum (Table 2). Specifically, under the combined effect of U. europaeus mulch and extract, weed species (L. rigidum and L. multiflorum) reached more biomass, while A. sativa reached lower biomass (Fig. 2). The effect on A. sativa disappeared in the presence of AC (Fig. 2), indicating that allelopathic compounds of U. europaeus could be implicated. The interaction term Ulex × AC significantly affected A. sativa biomass (Table 2).
Fig. 2. Mean values (±SE, N = 10) of (a) aboveground biomass, (b) belowground biomass (ES) and (c) total biomass of the target species (Avena sativa, Lolium multiflorum y Lolium rigidum) grown in the presence or absence of U. europaeus mulch and extract (Ulex and No Ulex, respectively) and with or without activated carbon (AC) (grey and white bars, respectively). Different letters stand for statistically significant differences among treatments (Ulex × AC) at P < 0.05 (LSD test).
Table 2. Summary results of the two-way ANOVA assessing the combined effects of mulch and extract from U. europaeus (Ulex), activated carbon (AC) and their interactions on the above- and belowground and total biomass of the target species (Avena sativa, Lolium multiflorum and Lolium rigidum)
Significance level: *** <0.001; ** <0.01; * <0.05.
General effect of U. europaeus and AC
Accounting for all dependent variables, only in L. rigidum we found a significant effect produced by the interaction Ulex × CA (Table 3), indicating that L. rigidum was the most sensitive species to the treatments, but the direction of the treatment effects changed depending on the variable (Figs. 1 and 2).
Table 3. Summary results of the permutational multivariate analysis of variance using distance matrices, assessing the general effect of U. europaeus (Ulex), activated carbon (AC) and their interactions accounting for all dependent variables: percentage of emergence (% E), emergence speed (ES), height, above- and belowground and total biomass of the target species of each target species (Avena sativa, Lolium multiflorum and Lolium rigidum) on the (Avena sativa, Lolium multiflorum and Lolium rigidum)
Significance level: *** <0.001; ** <0.01; * <0.05.
Soil legacy effect
No soil legacy effects were observed in L. rigidum and L. multiflorum (Table 4 and Fig. 3). The interaction effect of 2-month conditioned substrates by AC and mulch and extracts from U. europaeus (Ulex) significantly affected the germination percentage (%G) of A. sativa, but any of the factors significantly affected germination speed (GS) of the target species (Table 4). Only the presence of CA in the absence of Ulex had a negative legacy effect on A. sativa (Fig. 3).
Fig. 3. Soil legacy effects. Mean values (±SE, N = 5) of germination percentage of the target species (Avena sativa, Lolium multiflorum y Lolium rigidum) submitted to substrates conditioned during 2 months by the following treatments: presence or absence of U. europaeus mulch and extract (Ulex and No Ulex, respectively) and presence or absence of activated carbon (AC) (grey and white, respectively). Different letters stand for statistically significant differences at P < 0.05 (LSD test).
Table 4. Soil legacy effects
Summary results of the two-way ANOVA assessing the soil legacy effects (2-months) left by mulch and extract from U. europaeus (Ulex), activated carbon (AC) and their interactions on the percentage of germination (%G) and germination speed (GS) of the target species (Avena sativa, Lolium multiflorum and Lolium rigidum)
Significance level: *** <0.001; ** <0.01; * <0.05.
Discussion
Effects of U. europaeus
The present study provides useful information on the effect of U. europaeus mulch and extract on the target species; this is a starting point for testing the feasibility of using U. europaeus as a bioherbicide in sustainable agriculture. Beginning with U. europaeus mulch, we found that it exerted a negative effect on the emergence and height of the weed L. rigidum, but not on the performance of the other target species. Thus, our previous expectations were not completely fulfilled because we expected that the germination of both weeds would be more affected than that of the crop species due to the bigger size of the later (Liebman and Sundberg, Reference Liebman and Sundberg2006) and due to previous studies reporting greater phytotoxic effects of U. europaeus amendments and extracts on weed species than on maize (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souto and Pedrol2020a, Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). The negative effect on both emergence and height of L. rigidum was only detected in the absence of AC, indicating that it can be attributable to the presence of phytotoxic compounds likely released by U. europaeus mulch in the soil. From an ecological perspective, our results indicate that U. europaeus is able to allelopathically hinder the germination and establishment of L. rigidum, which can be advantageous for U. europaeus to reduce competitors for its own seeds in the field.
Although significant, the effects on L. rigidum were rather small. Besides, the height of the weed species L. multiflorum significantly decreased in the presence of U. europaeus mulch, but the effect was also small and only detected in the presence of AC. This may indicate that some compounds released by U. europaeus hinder the growth in height of L. multiflorum in conditions of elevated carbon. Following the results of mulch effects, we could say that U. europaeus mulch can be used as a pre-emergence bioherbicide in oat crops, being safe for A. sativa and showing a small but significant effect on L. rigidum. However, we cannot certainly report this because the mulch also increased the aboveground biomass of weeds in a greater or lesser extent. This result unadvised the use of U. europaeus mulch to control Lolium weeds. Future investigations would be to focus on how to increase the effectiveness of U. europaeus mulch on Lolium weeds, for instance by increasing the quantity added or by adding the mulch only on the topsoil, creating a physical barrier for weed germination (Facelli and Pickett, Reference Facelli and Pickett1991).
Previous studies found that U. europaeus amendments affected weed species (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). However, the effects greatly varied among target species and dependent variables measured (emergence, biomass, height). Similar to our results, in the presence of U. europaeus amendments, the biomass of the weed Digitaria sanguinalis (Poaceae) increased, as well as other Monocotyledon species (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). Contrary, U. europaeus amendments decreased the emergence and height of the weed Amaranthus retroflexus (Amaranthaceae). Several studies also reported different responses of plant species to allelopathic and fertilizer effects of plant residues (Sturm et al., Reference Sturm, Peteinatos and Gerhards2018; Little et al., Reference Little, Ditommaso, Westbrook, Ketterings and Mohler2021). In the case of U. europaeus aqueous extracts, the effects also varied among target species, negatively affecting A. retroflexus but not D. sanguinalis or Zea mays L. (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souto and Pedrol2020a). Similarly, in this study, the effect of U. europaeus aqueous extract, added post-emergence, was negative for the biomass of A. sativa but positive for Lolium species. The sensitiveness of A. sativa to the water-soluble allelopathic compounds of U. europaeus can explain this result, as the negative effect was reduced in the presence of AC. Therefore, the postemergence use of U. europaeus aqueous in oat crops is disadvised but could contemplate its use in areas where A. sativa is an undesired species, mainly in wheat crops (Rapoport et al., Reference Rapoport, Marzocca and Drausal2009).
The positive effects of U. europaeus mulch and aqueous extract on Lolium biomass could be due to nitrogen and other nutrients released during mulch decomposition or extracted in the water. Among Fabaceous plants, U. europaeus has been identified to produce a voluminous amount of fixed nitrogen through its ability of rapid symbiotic nitrogen fixation in nodules. Common gorse has an annual rate of 100–200 kg ha−1 nitrogen accumulation during the rapid dry-matter accumulation period (Galappaththi et al., Reference Galappaththi, de Silva and Clavijo Mccormick2022). In fact, U. europaeus has traditionally been used as a natural agricultural fertilizer due to its nutritive effect (Atlan et al., Reference Atlan, Udo, Hornoy and Darrot2015). Soil nitrogen not only favor plant growth, but also plant germination (Duermeyer et al., Reference Duermeyer, Khodapanahi, Yan, Krapp, Rothstein and Nambara2018), which can explain the small negative effects or the absence of them that we found on the germination of the target species exposed to U. europaeus mulch. The increase in other soil nutrients, such as phosphorus, after addition of U. europaeus amendments (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b), may also explain the positive effect we found in this study. Therefore, allelopathic effects of U. europaeus may be counteracted by positive fertilizer effects.
From an ecological point of view, our results indicate that in areas out of the native range of U. europaeus, the phenomenon known as invasional meltdown can occur, where an exotic invasive species (in this case U. europaeus) facilitates the establishment of other exotic species (in this case, Lolium species) (Simberloff and Von Holle, Reference Simberloff and Von Holle1999). This phenomenon is common in Fabaceae species (e.g., Von Holle et al., Reference Von Holle, Joseph, Largay and Lohnes2006) and there are, in fact, invaded areas (e.g., Chile, Australia, Brazil) where U. europaeus coexist with L. rigidum, L. multiflorum and other Lolium species (Gilfedder and Kirkpatrick, Reference Gilfedder and Kirkpatrick1996; Koch et al., Reference Koch, Jeschke, Overbeck and Kollmann2016; Moreno-Chacón et al., Reference Moreno-Chacón, Mardones, Viveros, Madriaza, Carrasco-Urra, Marticorena and Saldaña2018). However, this possibility needs to be experimentally tested.
The species L. rigidum was the most sensitive species to U. europaeus treatments, but its response was ontogenetically dependent, negative at germination stage and positive during seedling growth. Besides, the species-specific effects produced by U. europaeus highlights the ability of this species to structure plant communities through changes in the soil environment. However, here we have just studied the effect of U. europaeus through its plant residues. Testing other components, such as plant root exudates, should be considered for a more realistic understanding of U. europaeus allelopathy in future studies.
Side effect of AC
Regarding the isolated effect of AC (i.e., in the absence of U. europaeus mulch), it varied among target plant species. We found positive effects on the biomass of A. sativa and L. multiflorum but negative effects on the biomass of L. rigidum. The fertilizing capacity of AC itself, increasing plant biomass has been previously reported (Gómez-Aparicio and Canham, Reference Gómez-Aparicio and Canham2008; Lau et al., Reference Lau, Puliafico, Kopshever, Steltzer, Jarvis, Schwarzländer and Hufbauer2008). Additionally, AC may display other side effects, such as the reduction of mycorrhiza infection or the stimulation of microbial community (Weißhuhn and Prati, Reference Weißhuhn and Prati2009). Also, a clear negative effect of AC on plant biomass and survival was also reported in a previous study (Yuan et al., Reference Yuan, Li, Yu, Oduor and van Kleunen2021). Although AC was proven to be valid in allelopathic studies by blocking allelopathic effects (e.g., Sturm et al., Reference Sturm, Peteinatos and Gerhards2018; Kheirabadi et al., Reference Kheirabadi, Azizi, Taghizadeh and Fujii2020; Lorenzo et al., Reference Lorenzo, González and Ferrero2021), experimental designs as the one used in this study must allow the identification of possible isolated and interactive effects of AC on target plants. As AC may have species-specific effects on plant performance through changes in chemical and biological properties of soils, it is essential to account for these effects in allelopathy studies.
Soil legacy effects
No negative soil legacy effects left by U. europaeus mulch and extracts were detected in this study, indicating the environmental safety of their use. Therefore, the possibility of allelopathic compounds having unwanted effects on the local flora is limited. However, further research is needed to understand soil legacy effects of U. europaeus, not only related to soil allelopathy, but also regarding to soil chemical and biological modifications commonly produced by Fabaceae species (Grove et al., Reference Grove, Haubensak and Parker2012, Reference Grove, Parker and Haubensak2015; Von Holle et al., Reference Von Holle, Neill, Largay, Budreski, Ozimec, Clark and Lee2013). The absence of soil legacy effects can be due to the degradation or inactivation of allelochemical compounds of U. europaeus in soil, or because the negative effect of allelochemicals was counteracted by positive effect of soil nitrogen on germination (Kobayashi, Reference Kobayashi2004; Duermeyer et al., Reference Duermeyer, Khodapanahi, Yan, Krapp, Rothstein and Nambara2018), but this need to be further investigated.
The use of U. europaeus mulch could be considered adequate as a fertilizer to prepare the soil 2 months before sown because it can add nutrients to the soil and the allelopathic effects may not persist after 2 months, but research is needed to increase its effectiveness as a bioherbicide on the emergence of weeds. For instance, U. europaeus mulch can be used in combination with other conventional agricultural practices or to reduce the quantity of herbicide in crops. Good farming practices demand a sensible use of herbicides to avoid over-dependence on a single control measure; herbicide rotation and integration with other measures are recommended to augment system stability (Fernández-Quintanilla et al., Reference Fernández-Quintanilla, Dorado, Leguizamon and Navarrete2007). Our way to test the allelopathic potential of U. europaeus was intended to be easy, cheap, natural and sustainable, in order to seek for a practical methodology that farmers can effortlessly apply, avoiding the use of chemicals and expensive and complex pretreatments, but we are aware that other methods can be more effective to extract allelochemicals (see Pardo-Muras et al., Reference Pardo-Muras, Puig, Lopez-Nogueira, Cavaleiro and Pedrol2018, Reference Pardo-Muras, Puig and Pedrol2019, Reference Pardo-Muras, Puig, Souto and Pedrol2020a).
Future research
The effect of U. europaeus mulch in the emergence and height of the weed L. rigidum and in other weed species need to be further investigated in more realistic conditions in the field. If we intend to use extracted plant material locally and suddenly from the invasive U. europaeus, there will not be a need to dry the biomass. In that case, using fresh instead of dried U. europaeus can be more effective (Pardo-Muras et al., Reference Pardo-Muras, Puig, Souza-Alonso and Pedrol2020b). For our study, we preserved the branches by oven drying them at 60°C before use, similar to previous studies (Singh and Sangeeta, Reference Singh and Sangeeta1991; Singh and Thapar, Reference Singh and Thapar2003; Gnanavel and Kathiresan, Reference Gnanavel and Kathiresan2007; Saeed et al., Reference Saeed, Ashfaq and Gul2011; Khan et al., Reference Khan, Afridi, Hashim, Khattak, Ahmad, Wahid and Chauhan2016). It is known that many allelopathic compounds can be degraded with the heat but there are also many that are thermostable (Gil et al., Reference Gil, Hong, Duan and Eom2022). For instance, the allelopathic potential of Pinus koraniensis Siebold & Zucc. even increased when dried at 90°C (Gil et al., Reference Gil, Hong, Duan and Eom2022) and allelopathic properties of Mentha pulgium L. were not altered when dried at 50°C (Ahmed et al., Reference Ahmed, Ayoub, Chaima and Hanaa2018). Even autoclave temperatures did not reduce the toxicity of A. altissima leaves (Heisey, Reference Heisey1990).
Given the variety of biotic and abiotic factors affecting the allelopathic interactions in natural soils, testing our experimental results in farmland would be advisable (Callaway, Reference Callaway2003). Similar studies found drastic differences between in vitro and in vivo experiments. For example, under experimental laboratory conditions, the germination of the weed Phalaris minor Retz. was reduced by 100%, while under natural field experimental conditions only 16%, when applying mulch from the leguminous Sesbania aculeata (Willd.) Pers. (Om et al., Reference Om, Dhiman, Kumar and Kumar2002).
Another aspect to be considered is if U. europaeus plants from the invaded range of distribution could have the same phytotoxic effectiveness on weed species as plants from the native range. In this study we tested the allelopathic effects of U. europaeus plants from the native range of distribution, but the effects of plants from the invaded range (where there is a need to eliminate the species) may differ. The comparison of the production of allelopathic compounds in plant species between native and invaded ranges has been scarcely examined and the data are not conclusive. For example, as noted by Lankau et al. (Reference Lankau, Nuzzo, Spyreas and Davis2009), the production of phytotoxic agents (glucosinolates) in Alliaria petiolata (M. Bieb.) Cavara & Grande was reduced throughout the invasion of chronosequence. Conversely, the species Solidago canadensis L. produced a greater amount of allelopathic compounds in the invaded than in the native range of distribution (Abhilasha et al., Reference Abhilasha, Quintana, Vivanco and Joshi2008). In the case of Ageratina adenophora (Spreng.) R. M. King & H. Rob., the concentration of some volatile compounds increased while others decreased in the invaded range (Inderjit et al., Reference Inderjit, Crocoll, Bajpai, Kaur, Feng, Silva, Treviño Carreón, ValienteBanuet, Gershenzon and Callaway2011). In the case of U. europaeus, Hornoy et al. (Reference Hornoy, Atlan, Tarayre, Dugravot and Wink2012) did not find differences in defensive chemicals (quinolizidine alkaloids) between native and invaded regions. Further research is needed to evaluate whether U. europaeus individuals from invaded regions are more allelopathic than those from native populations.
Despite our study failed to find the effectiveness of U. europaeus as bioherbicide in oat crops, this invasive plant has promising potential for further studies to focus on the development of a bioherbicide, because allelopathy can change depending on different contexts (soil, climate, target species, etc.) and experimental designs (Haugland and Brandsaeter, Reference Haugland and Brandsaeter1996; Kobayashi, Reference Kobayashi2004; Medina-Villar et al., Reference Medina-Villar, Alonso, Castro-Díez and Pérez-Corona2017; Zhang et al., Reference Zhang, Liu, Yuan, Weber and van Kleunen2021). Fertilizer is another possible use of U. europaeus. Regarding to management, invasive species are a controversial topic. There are polarized opinions whether plant eradications are feasible, the extent to which stakeholders should influence management decisions, and whether utilization of invasive species is an effective control approach. Innovative ideas based on rigorous scientific research should help improve consensus on how to approach invasive species management (Shackleton et al., Reference Shackleton, Vimercati, Probert, Bacher, Kull and Novoa2022).
Conclusions
Our study showed that U. europaeus cannot be used as a bioherbicide in oat crops, at least using the methodology we applied, because it favored the growth of the weeds L. rigidum and L. multiflorum and hindered the growth of crop species. However, some allelopathic effects of U. europaeus mulch on L. rigidum open the way for further investigations on the bio-properties of this invasive species and on how to increase its effectiveness as a bioherbicide. Our results also emphasized the use of U. europaeus as a fertilizer on crops. The use of U. europaeus in crops can be considered safe for germination of plants, as no soil legacy effect was detected. Given the intense degree of U. europaeus invasion, the need to remove this species and the great presence of oat crops in different areas of the world, we greatly recommend the utilization of U. europaeus residues in the invaded areas in a way that compensate the cost of control and elimination practices of this undesired species. For the moment, the use of U. europaeus residues as a bioherbicide on different crops needs further research. Our results also indicated the ability of U. europaeus to structure plant communities, being able to allelopathically hinder the germination and height of L. rigidum and facilitate the growth of both Lolium weeds.
Acknowledgements
This study was supported by the grants: REMEDINAL-TE (Regional Government of Madrid, S2018/EMT-4338), Complutense University of Madrid and Banco Santander (GR105/18) and FONDECYT/CONICYT 2018 No. 3180289 (Chile). MB was supported by UCM Scholarships for Collaboration in Departments and Institutes (2019).
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Open access
