Introduction

Sclerotium rolfsii and Sclerotinia sclerotiorum cause root and stem rot. Both polyphagous phytopathogens are associated with significant losses in crops of economic importance. S. rolfsii is predominantly found in tropical and subtropical regions (^{Gholami et al., 2019}), whereas S. sclerotiorum presents a broader geographic distribution, affecting crops in temperate and subtropical areas (^{Ordóñez-Valencia et al., 2018}). The diseases produced by fungi that affect agricultural crop yields are frequent. There is a growing concern over the resistance of the strains to the fungicides used for the chemical control of fungi, so new products are being sought for the management of these phytopathogenic agents. The use of plant extracts as antifungal agents can be an alternative; different Datura species have shown an inhibiting effect against diverse phytopathogenic fungi, although the effect varies according to the concentration and the part of the plant from which the extract is obtained, as well as the types of solvents used for the process (Öz, 2017).

D. discolor extracts have been evaluated against different phytopathogens; methanolic and ethanolic leaf and stem extracts inhibited the in vitro development of Aspergillus flavus, Aspergillus niger, Penicillium chrysogenum, Penicillium expansum, Fusarium moniliforme and Fusarium poae (^{Tequida-Meneses et al., 2002}). In addition, the aqueous leaf extract obtained by infusion affected the growth of Colletotrichum gloeosporioides (^{Verdugo-Contreras et al., 2023}). Likewise, aqueous extracts (2, 4 and 6 % p/v) from roots, seeds and leaves obtained with high-pressure processing (HPP) displayed variable percentages of in vitro inhibition against S. rolfsii, S. sclerotiorum and C. gloeosporioides, with the 6 % leaf extract being the one that displayed the highest effectiveness against the three fungi (^{Urias-Lugo et al., 2024}). The effect of D. discolor aqueous extracts obtained by infusion against S. rolfsii and S. sclerotiorum has not been evaluated.

The biological activity of the species of the Datura genus has been attributed to the presence of secondary metabolites such as phenolic compounds, flavonoids and alkaloids, which perform antioxidant and antimicrobial activities (^{Céspedes-Méndez et al., 2021}). Despite D. discolor being widely distributed in Mexico, reports on the phytochemical characterization of this species are scarce (^{Verdugo-Contreras et al., 2023}). Withanolides have been reported (^{González et al., 2023}), as well as the presence of diverse groups of metabolites in raw leaf extracts, including alkaloids, steroids, terpenoids, glucosides, saponins, phenolic compounds and lipids (^{Ahanotu et al., 2024}), which highlights the bioactive and biotechnological potential of this little-studied species.

The aim of this study was to identify the main groups of phytochemical compounds found in the aqueous D. discolor extracts obtained by infusion and HPP, and to evaluate the in vitro antifungal potential of the aqueous D. discolor extracts obtained by infusion against S. rolfsii and S. sclerotiorum.

Obtaining the extracts. The D. discolor plants were gathered in the flowering stage and fruition in April 2024 in the municipality of Ahome, Sinaloa, Mexico (25°54-55’ N and 108-109°02-55’ W). Disinfection and pulverization were carried out following the descriptions by ^{Urias-Lugo et al. (2024}). The aqueous extracts were prepared at a ratio of 1:10 p/v. For the extraction by infusion, the procedure by ^{Bitwell et al. (2023}) was followed, and for the extraction by HPP, the one by Urias-Lugo et al. (2024). For both the infusion and the HPP methods, the mixtures were centrifuged (RC-5C Sorvall, Newtown, CT, USA) at 4000 rpm for 10 min, and the supernatants were extracted and stored at 4 °C until use.

Phytochemical sieving of Datura discolor. The main groups of bioactive compounds found in D. discolor aqueous extracts were identified with qualitative tests, as per reports by ^{Khan et al. (2019}). Color change tests were used for phenols, flavonoids, coumarins, tannins, terpenoids, anthraquinones, anthocyanins and betacyanins; precipitation was used for alkaloids and foam formation for saponins. Characterization was carried out using colorimetric and precipitation tests, specific to each type of compound, and the results were expressed on a semi-quantitative scale of relative concentration, ranging from absence (-) to abundant presence (++++).

For the detection of total phenols, 1 % ferric chloride was added, and its reaction with the extract produced green, violet, blue or black colors as a positive indicator. The presence of flavonoids was confirmed with the addition of concentrated sulfuric acid, with a characteristic orange color observed. Coumarins were identified with the formation of a yellow color when the extract was mixed with 10 % sodium hydroxide. The test for persistent foam formation lasting 15 to 20 minutes helped detect saponins. For tannins, adding 5 % ferric chloride after heating the extract resulted in green or blue colors, which indicated the presence of condensed or hydrolysable tannins, respectively. Terpenoids were identified with the Salkowski test, showing a reddish-brown color after mixing the extract with chloroform and concentrated sulfuric acid. Anthraquinones were detected using the method by Bontrager, observing a pink color due to the acid treatment, extraction with chloroform and alkalinization. Anthocyanins and betacyanins were differentiated by the color change to blue or yellow, respectively, when bringing the extract to a boil with NaOH at 0.1 N. Finally, the alkaloids were detected with Wagner's reagent, where the appearance of turbidity or a reddish-brown or unconfirmed their presence.

Quantification of phytochemical compounds. For the quantification of total metabolites, the liquid or lyophilized extracts were used (VirTis 25EL, VirTis Co. U.S.A.).

Total phenolics. This was determined using the colorimetric method by ^{Singleton et al. (1999}). A total of 10 µL of diluted extract (1:20 v/v) were placed in a 96-well microplate, followed by the oxidation of phenols with 100 µL of Folin-Ciocalteu reagent (1:10 v/v in water). After 2 min, the reaction was neutralized with 90 µL of 10 % Na2CO3. Subsequently, the sample was incubated at 40 °C for 30 min in the absence of light and absorbance was measured at 765 nm using a microplate reader (Synergy HT, Winooski, VT, USA). Gallic acid was used as a standard (0-0.4 mg mL^-1), and the results were expressed as µg, equivalents of Gallic Acid per mL of extract (µg EAG mL^-1).

Total flavonoids. The flavonoid content was determined using the colorimetric method by Quettier et al. (2000). A total of 50 µL of diluted extract (1:10) were placed in a 96- well microplate and 100 µL of 1.5 % AlCl₃ were added. After incubating for 10 min at 25 °C, absorbance was measured at 403 nm using a microplate reader (Synergy HT, Winooski, VT, USA). Quercetin was used as a standard (0-0.15 mg mL^-1) and results were expressed as µg quercetin equivalents per mL of extract (µg EQ mL^-1).

Saponins. Quantification was carried out using the colorimetric method by ^{Hiai et al. (1976}), based on the formation of chromophore groups with vanillin and sulfuric acid. Twenty milligrams of freeze-dried D. discolor extract were dissolved in 80 % methanol, centrifuged at 4000 rpm for 2 min and the supernatant was collected. In a 96-well microplate, 20 µL of the extract were placed along with 8 % vanillin, maintaining the microplate in an ice bath for 2 min. Next, 200 µL of (cold) 72 % sulfuric acid were added, and it was stirred for 3 min. The sample was incubated at 60 °C for 10 min, cooled for 5 min and the absorbance was recorded at 470 nm (Synergy HT, Winooski, VT, USA). Diosgenin was used as the standard (0-0.4 mg mL^-1), and the results were expressed as µg diosgenin equivalents per mL of extract (µg ED mL^-1).

Total alkaloids. This was determined using the colorimetric method by ^{Shamsa et al. (2008}), with modifications. Four hundred milligrams of freeze-dried aqueous D. discolor leaf extract were defatted with 5 mL of hexane, followed by sonication (30 min) and centrifugation (4500 rpm/2 min), discarding the supernatant. The residual hexane was removed under a stream of nitrogen and the residue was resuspended in 5 mL of methanol, subjected to sonication and centrifugation, and the supernatant was collected. The methanol was evaporated in a rotary evaporator at 38 °C and the residue was resuspended in 1 mL of HCl 2 N, washed twice with 5 mL of chloroform and the pH was adjusted to neutral using NaOH 1 N. Then, 5 mL of phosphate buffer (2 M, pH 4.7) and 5 mL of bromocresol green solution (69.8 mg of BCG in 3 mL of NaOH 2 N and dilute to 1000 mL) were added. The formed complex was sequentially extracted with 1, 2, 3 and 4 mL of chloroform, stirring and collecting the lower phase and adjusting the final volume to 10 mL with chloroform. Finally, absorbance was measured at 470 using a microplate reader (Synergy HT, Winooski, VT, USA). An atropine calibration curve was constructed, and results were expressed as µg atropine equivalents per mL of extract (µg EA mL^-1).

In vitro antifungal evaluation. The identity and phytopathogenicity of S. rolfsii and S. sclerotiorum have been described earlier; these pathogens are a part of the collection of the Environment and Health Research Unit of the UAdeO (^{Martínez-Álvarez et al., 2021}; ^{Martínez-Ereva, 2022}). The experiments were performed using infusion-derived extracts diluted to different concentrations (2, 4 and 6 %; v/v) in PDA (PDA; Bioxon, Cuautitlán Izcalli, Estado de México, México) in Petri dishes, following the methodology described by ^{Urias-Lugo et al. (2024}). The experiments were conducted twice in a completely randomized design with four replicates per treatment and Petri dish as the experiment unit.

Statistical analysis. All data were subject to a Shapiro-Wilk normality test. For the in vitro test against pathogens, a one-way ANOVA was performed and for the data of phytochemical compounds, a two-way ANOVA was performed, followed by a Tukey test (^{Little and Hills 1978}) with a value of P < 0.05. To allow values of zero in some treatments, the data were transformed using √x+1 as described by Gómez and Gómez (1984). Additionally, Student's unpaired t test (P < 0,05) was used to compare the alkaloid concentrations. The statistical analyses were carried out in the GraphPad Prism software, version 6.00 for Windows (GraphPad Software).

Phytochemical characterization. Phytochemical screening detected phenolic compounds, saponins, flavonoids, alkaloids, tannins, terpenoids, coumarins and betacyanins, but no anthraquinones or anthocyanins in any of the extracts (Table 1). The root and stem extracts presented a lower metabolite diversity and content. The seed and leaf extracts presented the same metabolite families, although their concentrations were different. The HPP extraction method displayed a higher phenol and saponin extraction/retention in leaves, and of alkaloids, saponins and tannins in the seed, whereas the infusion presented a higher content of flavonoids and coumarins in the seed. This analysis helped establish a preliminary phytochemical profile for D. discolor, with potential implications in its biological activity.

Based on the relative abundance observed in the preliminary screening and on the earlier evidence of its bioactivity, the phenolic compounds, flavonoids (^{Matías et al., 2020}), saponins (^{Barile et al., 2007}) and alkaloids (^{Singh et al., 2007}) were chosen for quantification, due to their well-known involvement in plant defense mechanisms and their inhibiting potential against fungal pathogens. In general, a difference in concentrations is observed between the plant tissue types and the extraction method (Figure 1 A-D): the content of phenolic compounds and flavonoids is greater in the leaf extract obtained with HPP (1370 µgEAG mL^-1 and 21.4 µgEQ mL^-1) than by infusion (586 µgEAG mL^-1 and 0.0 µgEQ mL^-1) (Figure 1 A and C), but the content of saponins and alkaloids is equal in leaf extract, regardless of the extraction method (Figure 1 B and D).

Figure 1 Quantification of phytochemical compounds from 6 % aqueous Datura discolor leaf extracts obtained by high-pressure processing (HPP) and infusion (INF). Gallic acid equivalents (EAG), Diosgenin equivalents (ED), Quercetin equivalents (EQ), Atropine equivalents (EA). 

Phenolic compounds are directly related to the antifungal action, they can act as fungicides or fungistatic (^{Matías et al., 2020}), they inhibit mycelial growth, in addition to acting on the emission of the germ tube of phytopathogenic fungi (^{Díaz-García et al., 2024}). Meanwhile, the flavonoids act directly on cytoplasmic granulation, the disordering of the cell content, the rupture of the plasmatic membrane and the inhibition of enzymes produced by the fungi during the penetration of the host (^{Knaak and Fuiza, 2010}). Phenolic compounds-both flavonoids and phenolic acids-have displayed antifungal activity against the phytopathogens Alternaria alternata, Rhizoctonia solani, Fusarium oxysporum, Botrytis cinerea and Phytophthora infestans (^{Wianowska et al., 2016}). Matías et al. (2020) related the antifungal potential of ethanolic and ethanolic D. inoxia leaf extracts against S. sclerotiorum with the abundance of alkaloids and phenolic compounds, mainly with total flavonoids.

Antifungal evaluation. The inhibition of the aqueous D. discolor extracts for S. rolfsii varied between 2 and 46 %. The highest inhibition was observed with the 6 % leaf extract (Table 2, Figure 2). Regarding the effect against S. sclerotiorum, the inhibition presented by the leaf extract (4 and 8 %) was not significant (Table 2, Figure 2). In the in vitro evaluation, the stem extract was discarded, given the low percentage of extraction in relation to the one for leaf (10 vs 55 %).

Regarding the antifungal activity of Datura species against the pathogens used in this study, Jaben et al. (2014) and Jaben et al. (2022), reported a reduction in growth between 69 and 94 % and between 29 and 88 % of S. rolfsii respectively, with methanolic D. metel fruit and leaf extracts. Higher activity may be due to the species of Datura and to the extraction solvent used. On the other hand, the aqueous D. discolor leaf extract obtained by HPP inhibited up to 100 % of the in vitro mycelial growth of S. rolfsii (^{Urias-Lugo et al., 2024}), possibly due to the extraction process used.

Regarding S. sclerotiorum, ^{Urias-Lugo et al. (2024}) reported that the aqueous extracts obtained by HPP inhibited up to 56 % of this fungus and ^{Matías et al. (2020}) observed that the aqueous D. inoxia leaf extracts obtained by constant static maceration for seven days inhibited the growth of S. sclerotiorum by up to 94 %. The effectiveness observed against the pathogen by the aqueous extracts can be attributed to differences in the extraction processes and in the type of bioactive compounds extracted in each one of the species.

Figure 2 Effect of an aqueous Datura discolor leaf extract obtained by infusion. Datura discolor (2, 4 and 6 %) and Control, respectively. A-D Sclerotium rolfsii, E-H Sclerotinia sclerotiorum. 

The phytochemical compound extraction processes that are based on thermal treatments can lead to the loss and degradation of thermolabile compounds with biological activity (^{Zhang et al., 2018}), whereas HPP provides a high impact on the cell structure, destroying cell walls and other structural barriers, which enhances the recovery of bioactive compounts (^{Le-Tan et al., 2022}). This may explain the effectiveness observed of the aqueous D. discolor extracts between the methods of infusion and HPP against S. rolfsii and S. sclerotiorum. The extraction processes used display differences in the phytochemical constituents of the aqueous Datura discolor extracts obtained by infusion and HPP, which exert a direct influence on the antimicrobial activity, since solubility, stability and synergy of the secondary metabolites extracted determine the degree of inhibition of the pathogens.

The results reported by ^{Urias-Lugo et al. (2024}), along with those from this study suggest the participation of flavonoids in the inhibition observed against S. sclerotiourm and S. rolfsii by the aqueous D. discolor leaf extracts obtained with HPP. In this study, the presence of flavonoids in the leaf extract obtained with HPP was only found with the quantitative method (Figure 1C). This could explain the lack of inhibition observed of the aqueous D. discolor extract obtained by infusion against S. sclerotiorum. Likewise, the extracts prepared by infusion also displayed inhibiting activity against S. rolfsii, but with a lower efficiency than the extracts obtained by HPP, indicating the possible presence of other bioactive compounds with antifungal effect, but with a lower potency or different action mechanism in comparison with the flavonoids extracted using HPP.

The results of this study do not help establish definitive conclusions regarding the secondary metabolites responsible for the antifungal activity observed, therefore the possibility of other compounds, not analyzed, found in the aqueous D. discolor extracts contributing individually or synergically to the inhibition of S. rolfsii and other evaluated fungi is not ruled out. This phytochemical complexity highlights the need for additional studies, aimed at the identification and characterization of the active compounds involved and their possible interactions.

Further studies should be conducted under greenhouse conditions, spraying leaves or the soil to observe the ability of control of the aqueous D. discolor extracts obtained by HPP and infusion, in order to confirm the reproducibility of the results observed under controlled conditions.