Metabolomics of Withania somnifera L. extracts by an integrated LC-MS and NMR approach and evaluation of their tyrosinase inhibitory activity

The goal of this study was to evaluate how different environmentally friendly extraction methods and ethanol-water mixtures influence the chemical profile of Withania somnifera root extracts. Advanced LC-MS and NMR techniques were used to characterize the metabolites, with particular attention to pharmacologically active compounds such as withanolides and tropane alkaloids.

The study also assessed the tyrosinase inhibitory activity of the extracts, which is relevant for their potential use in treating neurological disorders. Results revealed significant variation in metabolite composition depending on the extraction technique and solvent, with maceration yielding the highest levels of withanolides and withanosides.

The original article

Metabolomics of Withania somnifera L. extracts by an integrated LC-MS and NMR approach and evaluation of their tyrosinase inhibitory activity

Luciana Maria Polcaro, Antonietta Cerulli, Milena Masullo, Sonia Piacente 

Journal of Pharmaceutical and Biomedical Analysis, Volume 253, 15 January 2025, 116520

https://doi.org/10.1016/j.jpba.2024.116520

licensed under CC-BY 4.0

Selected sections from the article follow. Formats and hyperlinks were adapted from the original.

Withania somnifera L., commonly known as Ashwagandha, winter cherry or poisonous gooseberry, is a plant belonging to the Solanaceae family [1]. The species name somnifera derives from Latin and means “inducing sleep” justifying its extensive use as neuroprotective. Instead, the name ashwagandha is a combination of “ashva” in Sanskrit, which means horse and “gandha”, which means perfume, reflecting that the roots have a strong smell like a horse [1]. In the Ayurvedic system, Withania is classified as "Rasayana" which means "tonic" as it acts mainly for body rejuvenation, defence against diseases, slowing down ageing and improving memory. W. somnifera has been used in Ayurveda by boiling fresh roots in milk or even crushing the fine powder roots, which are then mixed mainly with water, milk, or honey [2].

The roots represent the part of the plant most used in traditional preparations for the presence of bioactive compounds, mainly belonging to steroidal lactones called withanolides, isolated for the first time from W. somnifera [3]. Chemically, these are C₂₇ oxygenated ergostane-type steroids, having a γ-lactone in the side chain and a 2-en-1-one system in the ring [4]. They are reported to possess several activities, including anti-inflammatory [4], analgesic, anti-arthritic [4], anti-epileptic [5], anti-Alzheimer [6], antiparkinson [6], neuroprotective [6], antioxidant [7], immunomodulatory, anti-depressant [6], anti-platelet [8], and fibrinolytic [9].

This research aims at comparing different extraction methods to highlight the differences in the chemical composition and biological activity of the extracts in the perspective of obtaining a preparation as rich in withanolides as possible. Therefore, W. somnifera roots were submitted to different extraction protocols, including a conventional extraction technique represented by maceration and non-conventional extraction techniques such as ultrasound-assisted extraction (UAE) and, for the first time, solid-liquid dynamic extractions (SLDE-Naviglio).

Regarding the solvents, for all the protocols, mixtures of EtOH and H₂O at 50:50, 75:25 and 100:0 were used. The chemical profiles of all the extracts were analyzed using two different approaches: LC-MS, which allows the identification of secondary metabolites, and ¹H NMR, which has the advantage of detecting several primary metabolites. Moreover, to unambiguously assign the structures of compounds putatively detected by LC-MS, the W. somnifera root extract N_50 was fractionated by chromatographic steps to afford the purified compounds. The isolated compounds were characterized based on 1D and 2D NMR spectroscopy in combination with MS experiments. To highlight differences between extracts and understand how the extraction methods and the solvent mixtures can affect the chemical profile, both LC-MS and ¹H NMR data of the extracts were analyzed by Principal Component Analysis (PCA). Moreover, based on the use of this plant in neurological disorders and the evidence that in Parkinson’s disease neurons that contain the dark-brown cytoplasmic pigment neuromelanin are particularly susceptible to neurodegeneration [11], the tyrosinase inhibitory activity of the extracts was herein tested by spectrophotometric assay.

2. Material and methods

2.3. LC-ESI/QExactive/MS/MS analysis and multivariate data analysis of LC-MS data

All obtained extracts were analyzed using liquid chromatography coupled with electrospray ionization and a high-resolution mass spectrometer (QExactive: hybrid Quadrupole-Orbitrap Mass Spectrometer, Thermo Fischer, Waltham, MA, USA), operating in positive ion mode. LC-MS analysis was carried out on a Kinetex 2.6 µm C18 100 Å (100 × 2.1 mm) column (Phenomenex, Aschaffenburg, Germany), using a flow rate of 0.2 ml/min. A binary solvent system was used (eluent A: water with 0.1 % formic acid; eluent B: acetonitrile with 0.1 % formic acid. The HPLC gradient started at 5 % B, after 30 min % B was at 95 %, this percentage was maintained for another 5 min before returning to the starting percentage. For more details, see Supplementary material.

LC-ESI/QExactive/MS/MS chromatograms of all obtained extracts were subjected to a chemometric study by a multivariate statistical approach such as Principal Component Analysis (PCA) (for more details, see Supplementary material).

2.5. 1H NMR analysis, data processing and multivariate data analysis of NMR data

For ¹H NMR analysis, each extract (4.4 mg) was dissolved in 537 μL of phosphate buffer (1 M KH₂PO₄, D₂O, and 2 mM NaN₃ to prevent microbial contamination). After that, 13 μL of TSP (trimethylsilyl propanoic acid) 1 mM were added as an internal standard and finally the mixture was placed into a 5 mm NMR tube [11]. All measurements were performed on a Bruker Avance III 600 Ascend NMR spectrometer (Bruker, Germany) operating at 600 MHz (for more details, see Supplementary materials). Compounds were identified and quantified using the Chenomx software (Chenomx Inc., AB, Canada) [13] (see Supplementary materials).

The PCA of NMR data was performed. The matrix was made up of the amount of each primary metabolite identified in W. somnifera extracts, and the observations included the nine extracts (see Supplementary Materials).

3. Results and discussion

3.1. LC-ESI/QExactive/MS/MS analysis of W. somnifera extracts

W. somnifera roots were extracted by a conventional extraction procedure (maceration) and by non-conventional extraction procedures (UAE and SLDE-Naviglio), using the same solvent mixtures of EtOH:H₂O (50:50, 75:25, 100:0). All the extracts were analyzed by HPLC coupled with electrospray ionization and hybrid Quadrupole-Orbitrap mass spectrometry (LC-ESI/QExactive/MS/MS) in positive ion mode (Figures S1-S9). Fig. 1 shows the LC-MS profile of extract N_50. The LC-MS profiles showed 42 main peaks. Their accurate masses, characteristic fragmentation patterns, retention times, and literature data allowed us to putatively identify compounds mainly belonging to the steroids class (Fig. 2) [15].

Journal of Pharmaceutical and Biomedical Analysis, Volume 253, 2025, 116520: Fig. 1. LC-ESI/Q-exactive/MS Base Peak profile of N_50 extract of W. somnifera roots in positive ion mode.

3.2. Multivariate statistical analysis of LC-ESI/QExactive/MS/MS profiles

To identify differences in the chemical profile among different extracts, LC-ESI/QExactive/MS/MS profiles of N_50, N_75, N_100, U_50, U_75, U_100, M_50, M_75 and M_100, in triplicate, were analyzed by multivariate statistical analysis [18]. The raw data were first filtered and normalized using MZ mine 3.38 software and then processed using MetaboAnalyst software by selecting “Log transformation” for data transformation, and “Autoscaling” for data scaling. The result of Principal Component Analysis (PCA) emphasized the significance and predictability of the model when the targeted approach was applied; in particular, PC1 contributed to 50.0 % of the variance, followed by PC2, which contributed to 38.9 %. These results allowed us to highlight differences between the extraction methods. As observed from the score plot, the extracts were grouped into three different clusters related to the extraction technique used. The first cluster, colored in red, gathered extracts obtained by macerations; the second area, in green, contained all the extracts obtained using the SLDE-Naviglio, and lastly, in the cluster colored in blue, were grouped all the ultrasound-assisted extracts (Fig. 3A). The metabolites responsible for the distribution of the extracts in the different areas of the score plot are shown in Fig. 3B (heatmap), from which it is possible to observe the abundance of each metabolite in each extract. The extracts obtained by maceration are the richest in withanolides, marker compounds of Ashwagandha, able to cross the blood-brain barrier promoting beta-amyloid clearance and improving synaptic function [19]. Moreover, among the detected metabolites in the LC-MS profiles, the N-feruloyltyramine is a known tyrosinase inhibitor and PCA analysis revealed its highest content in the Naviglio extracts [20].

Journal of Pharmaceutical and Biomedical Analysis, Volume 253, 2025, 116520: Fig. 3. Multivariate data analysis of specialized metabolites detected in W. somnifera extracts by LC-MS analysis: (A) PCA score scatter plot; (B) clustering result in the form of a heatmap.

3.3. Chemical profile of W. somnifera extracts by 1H NMR analysis

To define the full metabolite profile of W. somnifera extracts and detect not only the specialized metabolites but also the primary metabolites, all the extracts were investigated by an approach based on ¹H NMR analysis [21] (Figures S10-S12, Supplementary Material). Fig. 4 shows the ¹H NMR spectrum of N_100, characterized by the presence of various classes of metabolites. A comprehensive inspection of all ¹H NMR spectra based on the characteristic signals allowed the identification of organic acids, phenolics, carbohydrates, amino acids and alkaloids. A total of 21 metabolites were identified using the Chenomx, Human Metabolome libraries Databases (HMDB) and 2D NMR [22]. The software Chenomx is based on a highly sophisticated targeted profiling technology that allows for the easy deconvolution of complex NMR spectra into their components, which can be quantified with good accuracy [23]. The concentration of the identified metabolites was determined with respect to the known concentration of the internal standard (TSP) [24]. The identified metabolites and their concentrations, expressed in mg/g of extract, are listed in Table 2. Based on Fig. 4, the most intense signals are observed in the region ranging from 3.00 to 5.00 ppm, representing the carbohydrate constituents (fructose, glucose and sucrose) [25]. In the high-frequency region, characteristic peaks of aromatic organic acids and phenolic compounds such as vanillic acid, formic acid, fumaric acid, hydroxyphenylacetate, benzoic acid and quinone along with aromatic aminoacids such as histidine were identified (Fig. 4). In the low frequency region dimethylglycine, dimethylamine, proline, senecioic acid, valine and leucine were identified. All the ¹H NMR chemical shifts and multiplicity (J in Hz) are reported in Table S1. Moreover, some metabolites reported for the activity on CNS such as benzoic acid, sarcosine, tropane alkaloids and tyrosine were detected. Benzoic acid is naturally present in berries, prunes, tea, some herbs and spices such as nutmeg, cinnamon, cassia, and cloves and possesses several neuroprotective properties; it can improve the positive symptoms of schizophrenia [26]. Sarcosine (also known as N-methylglycine, or monomethylglycine) naturally occurs in foods such as egg yolks, turkey and legumes. It can be bought as a dietary supplement, promoted as a brain health supplement because there is evidence of neuroprotective effects in schizophrenia, depression, Parkinson and other psychotic disease [27]. The biologically relevant tropane alkaloids are identified in many members of the Solanaceae family (Atropa belladonna, Datura spp., Brugmansia and Hyoscyamus), Tropane-like compounds such as cuscohygrine, 3α-tigloyloxytropane, tropine, pseudotropine typical of Solanaceae family have been reported to be constituents of ashwagandha [28], [29]. Tropane-type alkaloids are known as competitive antagonists of the muscarinic acetylcholine receptor, acting both on the central and peripheral nervous systems. Due to this activity, they were the first form of therapy to relieve the symptoms of Parkinson's disease, a neurodegenerative disease caused by the destruction of dopaminergic neurons located in the substantia nigra of the midbrain, resulting in imbalance between muscarinic cholinergic transmission and dopaminergic neurotransmission striatal [30]. Regarding tyrosine, in plants, it is an important biosynthetic precursor of another class of alkaloids with act on CNS too, such as morphine, papaverine, noscapine, codeine but in humans, tyrosine plays a fundamental role in the synthesis of neurotransmitters such as acetylcholine, dopamine and norepinephrine. A previous work has demonstrated that when tyrosine is taken orally, the plasma tyrosine concentration was raised high enough to cross the blood brain barrier, increasing its availability for catecholamine synthesis; it has been suggested that supplementing tyrosine may increase neurotransmitters production proving useful in the treatment of Parkinson [31].

Journal of Pharmaceutical and Biomedical Analysis, Volume 253, 2025, 116520: Fig. 4. 1H NMR spectrum of W. somnifera roots N_100 extract.

4. Conclusion

The growing interest in the prevention of pathologies led to the commercialization of many botanical-based remedies. This growth is particularly related to cognition supplements, which aid in improving mental health. The mental health concerns pertaining to Alzheimer’s and Parkinson’s syndrome have led to a wide acceptance of cognition supplements and driven its market.

W. somnifera is one of the most essential ethnomedicinal herbs in Ayurveda medicine due to its wide range of therapeutic actions, among which the treatment of multiple brain disorders. Noteworthy, no mutagenicity and genotoxicity have been reported for W. somnifera, and thus, the plant has been cleared as a safe use for the management of neurocognitive disorders [35].

Different extraction techniques were carried out with the aim of selecting an ecological and environmentally friendly strategy for extraction. LC-MS analysis of the extracts allowed us to putatively identify withanolides and withanosides, which were subsequently confirmed after isolation and structural characterization by NMR. Among the isolated metabolites, 1α,3β,20α_F-trihydroxy-20,22-witha-5,24-dienolide 3-O-β-glucopyranoside (18) has been isolated for the first time from a natural source in the present work.

The LC-MS data were processed through a chemometric approach using targeted PCA, which highlighted the highest quantity of withanolides in the extracts obtained using maceration. The analysis of the extracts by ¹H NMR, allowed the identification and quantification of other metabolites not revealed by the LC-MS analysis. The results showed that the N_100 had the highest presence of benzoic acid, sarcosine and tropane alkaloids derivatives.

Regarding the use of W. somnifera, the tyrosinase inhibitory activity of the extracts was tested by spectrophotometric assay, showing how N_100 extract had the highest inhibition of the tyrosinase enzyme. In conclusion, the results highlighted how the extracts obtained using non-conventional extraction technique (SLDE-Naviglio and UAE), are better, in terms of biological activity, than the extract obtained using a conventional extraction technique (maceration). Therefore, the biological activity could be ascribed not only to withanolides but also to the other compounds, such as benzoic acid, tropane alkaloids derivatives, sarcosine and tyrosine, herein identified. Consequently, considering the growth of the use of W. somnifera in the food supplements market, the SLDE-Naviglio, here performed for the first time, may represent the most efficient technique which allows to exhaust in a short time, solid matrices containing extractable substances, operating at low temperatures with a guarantee of the production of high-quality extracts, easy use, and low-energy consumption.