Key message

  • Atractyloside, carboxyatractyloside, and atractyligenin characterized the C. gummifer extract.

  • Complete structural elucidation of carboxyatractyloside was reported.

  • The toxicity of all diterpenes was tested against the two-spotted spider mite, T. urticae.

  • Carboxyatractyloside exhibited the highest mortality and oviposition inhibition.

Introduction

Nowadays, the growing demographic increase is proportionally linked to the incessant growth in food needs, and to the huge demand for new agricultural spaces (Ray et al. 2013). In recent years, scientific research has aimed at the establishment of new technologies intended at increasing the yield of agricultural crops and quality of products, not forgetting the progressive decrease in land availability, water scarcity, and increasing climatic changes. Scientific development involves advances in the field of phytogenetics, searching for more resistant seeds with a higher yield, in the management of grazing, as well as about the development of biopesticides based on natural products (Pavela and Benelli 2016; Isman 2020).

The latter research field includes a focus on products of natural origin and semi-synthetic derivatives, which can play a fundamental role for managing insects and mites of economic importance (Stevenson et al. 2017; Benelli and Pavela 2018). Indeed, it is estimated that in 2012, natural products and related derivatives accounted for 50% of global sales of agrochemicals (Loso et al. 2017). Therefore, studying the modes of action of secondary metabolites has expanded their use in agricultural applications towards the control of harmful insects and mites (Jankowska et al. 2017). Of note, different researches reported the use of plants as interesting sources of secondary metabolites showing significant toxicity against a growing number of arthropod pests and vectors. Good examples are several plant species belonging to the Apiaceae (Basile et al. 2022; Badalamenti et al. 2021a; Spinozzi et al. 2021), Asparagaceae (Badalamenti et al. 2021b; 2022a), and Asteraceae families (Haris et al. 2022; Mssillou et al. 2022; Kavallieratos et al. 2022), among others (Stevenson et al. 2017; Isman 2020; Pavela et al. 2019a, 2020; Kavallieratos et al. 2021).

Chamaeleon gummifer Cass., i.e., the scientifically accepted name of the more common Atractylis gummifera L. (Worldfloraonline 2022), belongs to the Asteraceae family. It has a distribution in almost the entire Mediterranean basin (Spain, Portugal, South Italy, Greece, North Morocco, and Algeria). It is a thorny and perennial thistle with wide roots that can reach 40–50 cm, and a diameter of 8–10 cm. The stem, which extends even up to 1 m, is covered with very spiny leaves and, during the full summer days, with intense pink flowers (Vallejo et al. 2009; Bouabid et al. 2019a). This plant is commonly used in regional ethnopharmacology, for example, in Morocco (Kharchoufa et al. 2018); however, it is considered very toxic, and whose mortality depends on the dose used, the age of the subject who uses it, and the vegetal part utilized (roots, leaves, or flowers) (Bouabid et al. 2019a, 2019b). The species is traditionally used against colds, nausea, stomach pain, and dizziness, but also, in the form of an infusion, against blisters or bleeding, as a vermifuge or even as a purgative (Ahid et al. 2012; Hammich et al. 2013). The cases of poisoning due to this plant mention its use for traditional purposes and have taken into consideration the phytochemistry of aqueous or organic extracts, and above all the presence of highly toxic diterpenes such as atractyloside (1), carboxyatractiloside (2), and atractyligenin (3).

The first compound isolated, and then structurally identified, from the roots of C. gummifer Cass. was atractyloside (Lefranc 1868). However, today these diterpenes have also been found in different plants such as Xanthium strumarium L. (Machado et al. 2021), Coffea arabica L. and C. robusta L. Linden (Gao et al. 2021; Hu et al. 2021), and Antennaria rosea subsp. confinis (Greene) R.J.Bayer (Xiao et al. 2019). Atractyloside and carboxitractyloside, extensively investigated for their biological activities (Todisco et al. 2015; Cho et al. 2017), act at the cellular level with inhibition of oxidative phosphorylation in the hepatocytes (Vignais et al. 1978). Essentially, these two secondary metabolites negatively affect the production of ATP in the cell, since, by blocking the transport of adenosine diphosphate (ADP) into the mitochondria, they inhibit the adenine nucleotide translocator (ANT) (Roux et al. 1996; Pebay-Peyroula et al. 2003). The aglycon atractyligenin, due to the absence of isovaleric unit and sugar moiety, resulted significantly less lethal than 1 and 2 (Vignais et al. 1978). Several studies reported different chemical modifications of atractyligenin, supported by excellent biological properties. These experiments included the photoinduced modification of methyl C-20 (Buscemi et al. 2001; 2003), enzymatic transformations of all OH– groups (Monsalve et al. 2005), and the preparation of anti-tumour derivatives (Rosselli et al. 2007; Cotugno et al. 2014). One of these compounds, 15-ketoatractyligenin methyl ester, showed a high activity against several tumour cell lines (Cotugno et al. 2014; Vasaturo et al. 2017; Badalamenti et al. 2022b).

In this work, the first complete structural elucidation of 2 by 1D-NMR (Nuclear Magnetic Resonance) and bidimentional techniques such as 1H-1H-COSY (Correlated Spectroscopy), NOESY (Nuclear Overhauser Effect Spectroscopy), HSQC (Heteronuclear Single Quantum Coherence), and HMBC (Heteronuclear Multiple Bond Correlation) spectroscopy and mass spectrometry (HPLC/ESI/Q-TOF) is presented. Furthermore, compounds 1–3 were evaluated for their acaricidal efficacy and oviposition inhibition activity against the two-spotted spider mite, Tetranychus urticae Koch (Acari: Tetranychidae), a polyphagous mite pest of high agricultural importance (Attia et al. 2013), which has been found resistant to a rather wide number of acaricides exerting their toxicity through different mechanisms of action (van Leuveen et al. 2010; Xu et al. 2018; Wu et al. 2019; Adesanya et al. 2021).

Materials and methods

Plant material

Chamaeleon gummifer roots were collected in Piana degli Albanesi, Sicily (Italy) at the beginning of May 2020. The authentication was carried out by Prof. Vincenzo Ilardi, University of Palermo, Italy. A voucher specimen (PAL MB-2020/84) has been deposited in STEBICEF Department, University of Palermo.

General procedures

Optical rotations, measured in a CH3OH solution, on a JASCO P-1010 digital polarimeter; 1H (Fig. S1) and 13C-NMR spectra (Fig. S2), for compounds 1–2, were recorded at 400/100 MHz in (CH3)2SO (dimethyl sulfoxide, DMSO-d6) unless otherwise noted, on Bruker spectrometers, using the residual solvent signal (δ = 2.50 ppm in 1H and δ = 39.51 ppm in 13C for DMSO) as reference. For the compound 3 proton and carbon spectra were recorded in CD3OD using the residual solvent signal (δ = 3.31 ppm in 1H and δ = 49.15 ppm in 13C for CD3OD) as reference. The sperimental procedures have been previously reported (Badalamenti et al. 2021b).

Extraction and isolation of atractyloside (1), carboxyatractyloside (2), and atractyligenin (3)

The dried roots of C. gummifer (≈ 4.5 kg) were grounded and extracted in CH3OH three times at room temperature (4 L × 3 times). After filtration and evaporation of solvent under reduced pressure, a raw methanolic extract was obtained (98.01 g). This extract, dissolved in distilled water, was partitioned into n-butanol (11.23 g). This latter layer was chromatographed on a silica gel column, eluting with dichloromethane/methanol (99:1 → 90:10 v/v), to give 6 different fractions Cg1-Cg6. The Cg4-Cg5 (4.32 g) fractions were re-chromatographed, using a mixture of dichloromethane/acetone/water (5.5/4/0.5, v/v/v) as eluent, on a silica gel, to give compounds 1 (611 mg) and 2 (453 mg). From the Cg1 fraction (619 mg), after the separation and the purification on silica column, compound 3 (374 mg) was obtained.

Atractyloside (1)

Colorless amorphous solid; [α\({]}_{\mathrm{D}}^{20}=\) − 51.7° (c 0.2, H2O). 1H-NMR (DMSO-d6, 400 MHz) δ 2.17 (1H, m, H-1α), 0.62 (1H, t, J = 12.0 Hz, H-1β), 4.03 (1H, m, H-2α), 2.21 (1H, dd, J = 2.2, 1.7 Hz, H-3β), 1.04 (1H, ddd, 5.4, 5.2, 4.6 Hz, H-3α), 2.60 (1H, bs, H-4), 1.33 (1H, m, H-5β), 1.74 (1H, m, H-6α), 1.50 (1H, m, H-6β), 1.52 (1H, m, H-7β), 1.37 (1H, m, H-7α), 1.00 (1H, d, J = 8.0 Hz, H-9β), 1.52 (1H, m, H-11a), 1.28 (1H, m, H-11b), 1.54 (1H, m, H-12α), 1.39 (1H, m, H-12β), 2.58 (1H, bs, H-13), 1.75 (1H, d, J = 11.0 Hz, H-14α), 1.30 (1H, m, H-14β), 3.61 (1H, m, H-15β), 5.07 (1H, bs, H-17a), 4.97 (1H, bs, H-17b), 0.90 (3H, d, J = 2.6 Hz, CH3-20), 4.55 (1H, d, J = 7.9 Hz, H-1ʹ), 4.66 (1H, t, J = 7.9 Hz, H-2ʹ), 4.32 (1H, t, J = 8.0 Hz, H-3ʹ), 4.00 (1H, t, J = 8.0 Hz, H-4ʹ), 3.43 (1H, m, H-5ʹ), 3.66 (1H, m, H1-6ʹ), 3.60 (1H, m, H2-6ʹ), 2.12 (1H, d, J = 2.30 Hz, H-2ʹʹ), 2.14 (1H, d, J = 2.0 Hz, H-2ʹʹ), 1.96 (1H, sept, J = 6.5 Hz, H-3ʹʹ), 0.87 (6H, d, J = 2.3 Hz, H-4ʹʹ, H-5ʹʹ). 13C-NMR (DMSO-d6, 100 MHz) δ 47.0 (C-1), 72.3 (C-2), 34.1 (C-3), 45.3 (C-4), 48.1 (C-5), 25.0 (C-6), 35.1 (C-7), 47.2 (C-8), 52.5 (C-9), 42.6 (C-10), 17.2 (C-11), 32.0 (C-12), 41.6 (C-13), 35.9 (C-14), 81.0 (C-15), 159.9 (C-16), 107.4 (C-17), 176.0 (C-19), 15.9 (C-20), 98.6 (C-1ʹ), 72.1 (C-2ʹ), 77.0 (C-3ʹ), 73.4 (C-4ʹ), 75.9 (C-5ʹ), 61.8 (C-6ʹ), 171.0 (C-1″), 42.6 (C-2″), 24.5 (C-3″), 22.13 and 22–15 (C-4″-C-5ʹʹʹ). ESIMS (‒) m/z 725.2134 [M‒H]‒ (calcd. for C30H45O16S2, m/z 725.2149). Its physical and spectroscopic data agreed with those reported in the literature (Brucoli et al. 2012).

Carboxyatractyloside (2)

Yellow-white amorphous powder; nauseating smell; [α\({]}_{\mathrm{D}}^{25}=\) − 46.5° (c 0.1, MeOH). For 1H- and 13C-NMR data, see Table 1. ESI–MS ( +) m/z 793.2033 [M + Na]+ (calcd. for 793.2023).

Table 1 1H- and 13C-NMR data of compound 2
Full size table

Atractyligenin (3)

White amorphous solid; [α \({]}_{\mathrm{D}}^{25}=\) − 146.3° (c 0.1, EtOH). 1H-NMR (CD3OD-d4, 400 MHz) δ 2.18 (1H, m, H-1α), 0.71 (1H, m, H-1β), 4.17 (1H, m, H-2α), 2.38 (1H, m, H-3β), 1.24 (1H, m, H-3α), 2.63 (1H, bs, H-4), 1.52 (1H, m, H-5β), 1.83 (1H, m, H-6α), 1.68 (1H, m, H-6β), 1.61 (1H, m, H-7β), 1.50 (1H, m, H-7α), 1.03 (1H, d, J = 7.5 Hz, H-9β), 1.70–1.62 (2H, m, H2-11), 1.70–1.62 (2H, m, H2-12), 2.69 (1H, bs, H-13), 1.88 (1H, m, H-14α), 1.40 (1H, dd, J = 4.9, 4.7 Hz, H-14β), 3.80 (1H, s, H-15β), 5.19 (1H, bs, H-17a), 5.06 (1H, bs, H-17b), 1.00 (3H, s, CH3-20). 13C-NMR (CD3OD-d4, 100 MHz) δ 50.3 (C-1), 64.9 (C-2), 38.2 (C-3), 45.0 (C-4), 50.0 (C-5), 26.4 (C-6), 36.1 (C-7), 48.0 (C-8), 54.5 (C-9), 41.6 (C-10), 18.8 (C-11), 33.5 (C-12), 43.7 (C-13), 37.0 (C-14), 83.8 (C-15), 160.2 (C-16), 109.0 (C-17), 179.5 (C-19), 17.2 (C-20). ESIMS ( +) m/z 321.2054 [M + H]+ (calcd. for C19H28O4, m/z 321.2060). Its physical and spectroscopic data agreed with those reported in the literature (Brucoli et al. 2012).

Two-spotted spider mite rearing

Tetranychus urticae adults used in our experiments were obtained from a laboratory mass-rearing at the Crop Research Institute (Prague, Czech Republic). Mites were reared on Phaseolus vulgaris L. plants (> 20 generations) in growth chambers under 25 ± 2 °C and 12:12 h (light:darkness) photoperiod.

Acaricidal experiments

The toxicity of carboxyatractyloside, atractyligenin, and atractyloside, measured as mortality at the 2nd and 5th day of exposure, was determined by tarsal application on T. urticae adults (Pavela 2015). Trials were carried out on blackberry leaf discs, Rubus fruticosus L., sized 1 cm2. Carboxyatractyloside, atractyligenin, and atractyloside were formulated in acetone, and distributed using an automatic pipette. In our trials, 10 µL of each compound diluted in acetone were applied uniformly onto the leaf discs. Tested doses were 12.5, 25, 50 and 100 µg cm−2. After application, the discs were placed in Petri dishes (5 cm) with an agar layer 0.3 cm thick on the bottom. Control discs were treated with acetone. Acetone was let evaporating and then a small brush was employed to move 10 T. urticae females (1–2 days old) on each leaf disc. Petri discs were placed in a growth chamber (16:8 (L:D), 25 °C) and mortality was checked after 2nd and 5th day post-application (Pavela et al. 2017).

A further bioassay was developed to assess the possible inhibitory effect of carboxyatractyloside, atractyligenin, and atractyloside on T. urticae fertility and fecundity. The three above-mentioned molecules were formulated at 12.5, 25, 50 and 100 µg cm−2 and subsenquently applied on the R. fruticosus leaf disks as described above. In each replicate, 10 T. urticae females (1–2 days old) were released on each leaf disc in a Petri dish. All material was moved in a growth chamber [16:8 (L:D), 25 °C] for 24 h. After 24 h, the females were removed from the leaf disks, and the eggs were counted. Eggs were incubated for 5 days at 25 °C, and the number of newly hatched larvae was checked every day (Pavela et al. 2017). Data were presented as: (i) egg hatching (%), (ii) hatching inhibition (%), i.e., how many percent fewer larvae hatched compared to the control; (iii) natality inhibition (%), i.e., by what percentage the F1 population was reduced compared to the control (i.e. reduction of oviposition + reduced hatching). Each experiment was replicated 5 times over different days and with different compound solutions, to account for any variability.

Statistical analysis

In T. urticae experiments, mortality rates post-exposure to carboxyatractyloside, atractyligenin, and atractyloside were corrected through the Abbott (1925) formula, then analyzed by ANOVA followed by Tukey’s HSD test (p < 0.05). Percentages were arcsine square-root transformed before analysis; the software BioStat v5.0 was used for all analyses.

Results and discussions

Chemical analyses

The three ent-kaurene metabolites (1–3) (Fig. 1) were isolated from the methanol extract of C. gummifer roots by different chromatographic columns. Compounds 1 and 3, extensively described chemically and biologically in the literature, presented spectroscopic data in agreement with those by of Brucoli et al. (2012). In this study, the complete spectroscopic and stereochemical investigation of compound 2 was reported for the first time using 1D- and 2D-NMR, polarimetric and mass spectrometry (HPLC–MS) analyses.

Fig. 1
figure 1

Structures of atractyloside (1), carboxyatractyloside (2), and atractyligenin (3)

Full size image

Compound 2 was obtained, after several chromatographic columns, as a yellowish powder with an unpleasant odor. The HPLC–MS spectrum showed a molecular ion peak at m/z 793.2033 [M + Na]+ (calcd. for 793.2023), in agreement with a molecular formula of C31H46O18S2Na. The proton and carbon spectrum (Fig. S1 and Fig. S2) and the 1H- and 13C-NMR data (Table 1), showed signals for an ent-kaurenic skeleton, a tetracyclic diterpene, characterized by the presence of a double bond in position C-16/C-17 [δC = 159.44 (C-16) and δC = 107.54 (C-17)], two carbonyl functions (C-18 and C-19), with the same carbonic value of chemical shift (δC = 170.90 ppm), linked in position 4, an axial hydroxyl group in position 15 [δH = 3.59 (H-15)] and, finally, an angular methyl (C-20), bonded to carbon C-10, at δC = 16.38 ppm (δH = 0.94, s, 3H, H-20). Furthermore, signals for a glucoside derivative were observed. In fact, using the COSY, HSQC, and HMBC couplings, the entire structure of carboxyatractyloside (2) was determined. The correlation spot presented in the HMBC spectrum, between the anomeric proton H1ʹ (δH = 4.54 ppm, d) and the aglycone C-2 (δC = 73.42 ppm), clearly indicated the exact link between the sugar moiety and the terpenoid skeleton (Fig. 2). Finally, characteristic signals of isovaleric acid, a branched-chain saturated fatty acid, bound to C-2ʹ of the glycosidic unit (correlation spot, in the HMBC spectrum, between proton H-2ʹ and C-1ʹʹ) were found to be quite distinguishable in the proton spectrum: two terminal methyls, doublets, at δH 0.86–0.88 ppm and the unmistakable nonet for H-3 (δH = 1.95 ppm). For the exact stereochemistry, the NOESY correlation between the H-2α with the methyl protons (3H-20α) confirmed the β-glycosidic bond between the sugar portion and the aglycone (Fig. 2).

Fig. 2
figure 2

Chemical structure of carboxyatractiloside (2). The blue and red arrows indicate the main NOESY and HMBC correlations, respectively, fundamental for the determination of the stereochemistry and the bonds between the different portions

Full size image

Acaricidal activity against T. urticae mites

Considering the key agricultural importance of polyphagous T. urticae populations (Wybouw et al. 2018), and their fast-growing resistance to several currently employed acaricides (Wu et al. 2019; Adesanya et al. 2021; Alsay and Ay 2022), the development of novel and reliable products in the Integrated Pest Management (IPM) scenario is a field of research interest. A growing number of botanical-based formulations are being considered for T. urticae management (Pavela et al. 2017, 2018; Benelli et al. 2017). Herein, all the three tested substances, i.e., carboxyatractyloside, atractyligenin, and atractyloside, showed significant toxicity against T. urticae (Tables 2, 3 and 4), in terms of mite mortality, inhibition of oviposition, lower hatchability of eggs, and overall inhibition of natality. However, differences in efficacy between the three molecules were found. Comparing the lowest tested dose, i.e., 12.5 µg cm−2, carboxyatractyloside (2) appears to be the most effective compound, leading to mortality higher than 63% on the 5th day post-application, inhibiting oviposition by more than 70%, and inhibiting the egg hatching by 33.3%. Overall, natality was reduced by 80.1% (Table 2). A significant efficiency was also found for atractyligenin, where testing 12.5 µg cm−2 led to mortality of 50.5% on the 5th day post-application and a total natality of 73.2% (Table 3). The least effective substance was atractyloside, which, when tested at 12.5 µg cm−2 achieved only 34.6% mortality and natality reduction of 45.2% (Table 4).

Table 2 Toxicity of carboxyatractyloside on Tetranychus urticae mites
Full size table
Table 3 Toxicity of atractyligenin on Tetranychus urticae mites
Full size table
Table 4 Toxicity of atractyloside on Tetranychus urticae mites
Full size table

As outlined above, T. urticae is considered one of the most dangerous crop pests, characterized by the rapid development of acaricide-resistant populations (Raworth 2001). It is therefore very important to focus on new acaricidal substances with novel modes of action, which could become a suitable alternative to existing active ingredients. In the current study, we tested natural ent-kaurene diterpenoids isolated from C. gummifer against T. urticae. Of the substances that we evaluated through acaricidal tests, carboxyatractyloside and atractyligenin showed promising acaricidal efficiency, when at the lowest dose we tested, 12.5 µg cm−2, more than 50% mortality of adults was detected on the 5th day after application, and for carboxyatractyloside, at the same time it was detected greater than 80% inhibition of natality. The dose tested here, i.e., 12.5 µg cm−2, corresponds approximately to an application concentration of 0.125%. In this scenario, about 0.2% could be considered as a lethal concentration, then the tested substances carboxyatractyloside and atractyligenin are more effective than some essential oils. For example, Afify et al. (2012) tested EOs from Matricaria chamomilla L., Origanum majorana L. and Eucalyptus sp. against T. urticae, for which the LC50 were found to be 0.65, 1.84, and 2.18%, respectively, and for eggs 1.17, 6.26, and 7.33%, respectively. The LD50 of these compounds were also lower than that found for the root extracts of Saponaria officinalis L., for which an LC50 of 1.18% was estimated, based on which new botanical acaricides were developed (Pavela 2017).

On the other hand, the efficacy reported in our study was lower than that of Onosma visianii Clem. root extract, showing a LD50 of 2.6 µg cm−2 at the 5th day post-application (Sut et al. 2017), as well as than the Drimia pancration (Steinh.) J.C.Manning & Goldblatt extract, for which the LD50 was 8.5 µg cm−2 (Badalamenti et al. 2022a, b).

The substances tested here show a promising effect on the mortality and fertility reduction of T. urticae, which deepens with time after application. Extracts or compounds that could reduce the fertility of T. urticae are important since this pest multiplies and matures quickly. Therefore, it is necessary to reduce its population density as quickly as possible to minimize the damage caused by the sucking of these phytophagous mites, as well as their vector activity of numerous plant pathogens. Further tests will be required regarding the effectiveness of the compounds proposed in this study on non-target organisms to estimate their environmental safety, as well as the persistence of the effect or the possibility of a synergistic increase in the effectiveness of the substances carboxyatractyloside and atractyligenin.

Conclusions

In this study on the root extract of C. gummifer by mean of 1D- and 2D-NMR, NOESY, and MS spectra, the full stereochemical structure of carboxyatractyloside (2) was revealed for the first time. This diterpene, together with the other two compounds, atractyloside and atractyligenin (1–3), were evaluated for their potential activity as an acaricide against T. urticae. Overall, compounds 1–3, with special reference to carboxyatractyloside, represent valuable candidates to develop novel acaricides for crop protection. However, additional research efforts are still needed to include them in highly stable formulations, such as nanoemulsions for field use (Pavoni et al. 2019; Pavela et al. 2019b). Further research on the possible non-target effects of these compounds on pollinators and mite biocontrol agents is ongoing.