The reference standard, including neferine, quercetin, and kaempferol, was dissolved in methanol at a concentration of 0.1 mg/mL. All solutions were stored at –20 °C until use.

Professor Xiangri Li, an expert in Chinese materia medica from Beijing University of Chinese Medicine, identified the lotus seeds as the dried mature nuts of Nelumbo nucifera Gaertn., a plant belonging to the genus Nelumbo (family Nymphaeaceae). The lotus seed coat (Fig. 1) has been deposited in the Laboratory of Chinese Materia Medica Processing, Beijing University of Chinese Medicine, and is available for inspection and verification by peers.

Fig. 1.

Processing pathways and morphological differences: Medicinal vs. Edible Lotus Seeds.

For positive ion mode analysis, 1.0 g of the powder was extracted with 0.3 mol$\cdot$L⁻¹ hydrochloric acid-methanol solution under 60 min ultrasonication, filtered, and 5 mL of the filtrate was evaporated to dryness in a water bath. For negative ion mode analysis, approximately 1.0 g of lotus seed coat powder was mixed with 70% methanol, ultrasonicated for 60 min, filtered through filter paper, and the filtrate was then passed through a 0.22 µm microporous membrane. The residue was reconstituted with methanol, transferred to a 5 mL volumetric flask, made up to volume with methanol, and filtered through a 0.22 µm microporous membrane.

A Vanquish UHPLC system (Thermo Fisher Scientific, San Jose, CA, USA) was employed for separation. Samples were separated on an Acquity HSS T3 column (2.1 $\times$ 100 mm, 1.8 µm, Waters, MA, USA). The chemical components of the lotus seed coat were analyzed in positive and negative ion modes. Chromatographic and mass spectrometric conditions are provided in Supplementary Materials (Section LC/MS Testing Conditions).

Data were processed with Xcalibur TM 4.1 software (Thermo Fisher). MS/MS data were further analyzed using an online LC/MS database (http://www.hmdb.ca; https://massbank.eu/MassBank/), and by comparing the MS/MS spectra with reference standards and the literature.

One gram of lotus seed coat powder was weighed and extracted with 20 mL of 70% methanol (v/v) for 60 min in an ultrasonic water bath (40 kHz, 500 W). The extracts were passed through filter paper and stored at 4 °C. The extraction rate of the lotus seed coat is approximately 22.4%.

Free radicals (atoms or groups with unpaired electrons) are closely related to aging. Common methods for measuring free radical scavenging capacity include the DPPH and ABTS assays. The DPPH solution employed in the assay was prepared at a concentration of 0.4 mmol/L, with 70% methanol serving as the diluent. For the standard group, 100 µL of ascorbic acid standard solution was added to 100 µL DPPH solution. For the sample group, 100 µL of lotus seed coat sample extract diluent was added to 100 µL of DPPH solution. The reaction system was maintained at room temperature under dark conditions for 30 minutes, after which the absorbance was measured at 517 nm using a microplate reader.

The ABTS$\cdot$⁺solution was prepared by mixing 2 mL of 7.4 mmol/L ABTS solution and 2 mL of 2.6 mmol/L K₂S₂O₈ solution and storing at room temperature for 12–16 h away from light. This was diluted with methanol. The absorbance of the ABTS solution was within the range of 0.70 $\pm$ 0.02 at 734 nm. Sample dilution (100 µL) or ascorbic acid standard solution (25–185 µg/mL) was pipetted into a 96-well cell plate, and 100 µL of ABTS$\cdot$⁺solution was added. After mixing well, the reaction was performed at room temperature for 6 min. Clearance was calculated with the following Eqn. 1:

(1)$$Clarance(\%)=\frac{A_o-A_x}{A_x}\times 100\%$$

A_χ: absorbance of the sample solution/standard solution after the reaction.

A₀: absorbance of the blank reaction.

The DPPH/ABTS clearance of the sample was calculated by fitting the equation to the standard curve, and then expressed as ascorbic acid equivalent: mg ascorbic acid eqg DW (mgAAE/g DW).

The cultivation and experimental manipulation of C. elegans were conducted in accordance with established protocols [15]. Freeze-dried powder of 70% methanol extract from the coat of lotus seeds was dissolved in water and added to the nematode growth medium (NGM) plate. Simultaneously, E. coli OP50 was spread on the NGM plate to serve as nourishment for C. elegans. In preliminary toxicological experiments, the lotus seed coat was found to have no toxic effect on C. elegans in the concentration range of 0.01~1 mg$\cdot$mL^-1. The aging experiment consisted of four groups: blank (without treatment, BG), low-dose lotus seed coat extract (0.04 mg/mL, LG), medium-dose lotus seed coat extract (0.2 mg/mL, MG), and high-dose lotus seed coat extract (1 mg/mL, HG).

C. elegans at the L4 stage were chosen and transferred onto NGM plates (3 plates per group, 30 strips per plate). To maintain a consistent drug concentration, C. elegans were transferred to new NGM plates the following day, and the number of deceased individuals was recorded. The blank group was used as the control, and survival time was calculated from the transfer time until all C. elegans had died, which marked the end of the experiment. The experiment was repeated three times. The lifespan of each group of C. elegans was recorded and used to calculate the average, median, and maximum lifespans. Evaluation criteria for the death of C. elegans were no movement or swallowing action, and no response after light touch.

All experiments used synchronized L4-stage C. elegans under uniform conditions. For the lipofuscin assay, worms were transferred to fresh NGM plates (3 plates/group, 30 worms/plate) and cultured for 5 days, then rinsed three times with M9 buffer, anesthetized, and mounted on slides. Blue fluorescence intensity was then recorded via an inverted microscope and analyzed with ImageJ 1.54f (National Institutes of Health, Bethesda, MD, USA). For the ROS assay, worms were cultured as above, then rinsed, centrifuged (supernatant discarded), and incubated with 200 µL of 100 µM DCFH-DA (Solarbio, ID31309, Beijing, China) for 120 min. Green fluorescence intensity was then recorded via an inverted microscope and analyzed with ImageJ. For the SOD/CAT activity and MDA content assay, worms were transferred to drug-supplemented NGM plates and cultured for 5 days, then rinsed, centrifuged, repeatedly washed with M9 buffer, and stored at –80 °C. For assessment, the samples were ground, centrifuged, and the supernatant analyzed using kits (A001-3-2, A003-1-2, A007-2-1, Nanjing Jiancheng, Nanjing, China) as per the manufacturer’s instructions.

Targets for the lotus seed coat compounds were predicted using two databases: SEA (https://sea.bkslab.org/) and SwissTargetPrediction (http://www.swisstargetprediction.ch/). The chemical composition of the lotus seed coat was determined based on in-house experimental results [10]. Candidate targets were standardized to UniProtKB IDs and corresponding gene symbols via the UniProtKB database. Aging-related targets were retrieved from differential genes identified using the Genecards database (https://www.genecards.org/). Common targets were defined as the intersection of compound-predicted targets and aging-related targets [16].

To construct the PPI network, UniProt IDs of common targets were uploaded to the STRING database. The species was set as Homo sapiens, and the interaction confidence score threshold was adjusted to 0.90 (highest confidence). The resulting PPI network was visualized using Cytoscape 3.9.1 (Cytoscape Consortium, Seattle, WA, USA), followed by topological analysis via the CytoHubba and MCODE plug-ins to identify key nodes and functional modules.

Metascape (http://metascape.org/) is a robust platform enabling users to apply currently popular bioinformatics analysis methods to bulk gene and protein analysis for knowledge of gene or protein function. The gene names for the above vital targets were entered into the Metascape website, and species were limited to humans. This enabled Gene Ontology (GO) functional annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis using the Metascape website.

Molecular docking was performed using Autodock Vina 1.2.5 (The Scripps Research Institute, La Jolla, CA, USA) to assess the binding affinity of small molecules to the key protein. Subsequently, three representative small molecule-protein complexes were selected for molecular dynamics (MD) simulations to evaluate complex stability. The main procedures were carried out as follows:

① The AMBER14 force field was employed to optimize the energy of the protein-ligand system. A periodic boundary simulation box was constructed (expanded by 10 Å), filled with TIP3P water molecules (density: 0.997 g$\cdot$L⁻¹), and supplemented with 0.9% NaCl to mimic the physiological ionic environment.

② The simulation was conducted over a duration of 50,000 ps (50 ns) at a temperature of 310 K (NVT ensemble with Langevin temperature control) to simulate physiological conditions, with a pH of 7.4 (physiological condition). The initial structure was optimized using the default steepest descent method (500 steps).

③ Key output parameters included: Root Mean Square Deviation (RMSD) to assess overall conformational stability; Root Mean Square Fluctuation (RMSF) to analyze local dynamic fluctuations of residues in the active site; Radius of Gyration (Rg) to measure protein compactness; binding energy to evaluate the energy released upon ligand-protein binding; and Free Energy Landscape (FEL, also referred to as free energy surface) to investigate processes such as protein folding and molecular motion, thus helping to understand molecular stability and dynamic behavior across different states.

The inferred compounds are shown in Supplementary Materials and were derived from references and online databases [11, 12, 17]. The main alkaloid components in lotus seed coat are classified into aporphine-type, mono-benzylisoquinoline-type, and bis-benzylisoquinoline-type alkaloids. Cleavage of bis-benzylisoquinoline alkaloids primarily involves the fragmentation of two benzyl groups [12, 18]. Typically, $\alpha$-cleavage occurs at the N-atom, leading to the loss of neutral fragments of CH₂=N-CH₃ (m/z 43) or NH₂CH₃ (m/z 31). The isoquinoline moiety fragments undergo further cleavage and rearrangement to form biphenyl compounds [19]. Taking neferine as an example, compound 14 (Supplementary Table 1) has a retention time of 29.78 min. In positive ion mode, its quasi-molecular ion is m/z 625.3265 [M+H]⁺. The secondary mass spectrometry fragment ions include m/z 594.2845 [M+H–NH₂CH₃]⁺, 489.2375 [M+H–C₆H₄(OCH₃)CH₂–CH₃]⁺, and 206.1174 [M+H–C₆H₄(OCH₃)CH₂–CH₃–C₁₀H₁₁NO– C₆H₄O₂CH₂]⁺. Combined with the reference standard, this compound was identified as neferine, with the molecular formula C₃₈H₄₄N₂O₆.

In general, flavonoid glycosides, with the exception of carbon glycosides, undergo initial cleavage at the glycosidic bond to generate fragment ions of flavone aglycones that have lost their glycosyl groups [20]. Subsequently, these fragment ions undergo cleavage like that of flavonoids. The most prominent cleavage pathway for flavonoids is Retro-Diels-Alder (RDA) cleavage of the C ring, with positions 1 and 3 of the C ring being commonly cleaved to generate an ion containing the m/z 151 of the A ring. Upon comparison with a reference standard, compounds 38 and 41 (Supplementary Table 2) were identified as quercetin and kaempferol [21].

Proanthocyanidins are a class of polymers consisting of flavan-3-ols condensed through C-C bonds. They typically have catechins or epicatechins as the basic unit, with the primary mode of cleavage being rupture of the covalent bonds between the constituent units [22, 23]. Like flavonoids, the primary mode of cleavage of flavan-3-ols is also RDA cleavage of the C ring. The molecular ion peak is m/z 577.1351, and secondary fragment ions include m/z 425.0881, m/z 407.0772, m/z 289.0718, and m/z 161.0243.

Most phenolic acids contain carboxyl or phenolic hydroxyl groups, readily losing neutral molecules such as H₂O, COOH, and CO₂ under high-energy collisions [24]. Ferulic and caffeic acids readily lose acyl groups. In the case of ferulic acid, for example, the molecular ion peak is m/z 193.0504 in negative ion mode, and the secondary fragment ions are m/z 178.0272, m/z 149.0608, and m/z 134.0373.

The fatty acids contained in the lotus seed coat are mostly straight-chain unsaturated fatty acids. Hydroxy-octadecenoic acid was the most abundant among the fatty acids identified in this experiment. The cleavage pathway of the fatty acids is generally straightforward, with easy loss of H₂O or CO fragments, and firm molecular ionic peaks [11].

Based on experimental data, we calculated the average, median, and maximum lifespan of C. elegans. The C. elegans lifespan curve is shown in Fig. 4A. Compared with the blank group, various doses of lotus seed coat extract were found to increase the average lifespan of C. elegans. The average lifespan in the blank group was 14.31 days, compared to 15.11, 15.97, and 15.82 days in the low-, medium-, and high-dose groups of lotus seed coat extract, respectively (Fig. 4A). The results were significantly different (p 0.05), indicating that lotus seed coat can effectively prolong the healthy lifespan of C. elegans.

Fig. 4.

The effect of lotus seed coat extract on the aging of C. elegans. (A–E) The effects of lotus seed coat extract on the lifespan, lipofuscin content, reactive oxygen species (ROS) levels, antioxidant enzyme (SOD/CAT) activities, and Malondialdehyde (MDA) content in C. elegans. (^*p 0.05, ^*⁣**p 0.001 vs. BG group.)

The lipofuscin content of C. elegans accumulates with age, with its autofluorescence properties used as a sign of aging [26]. Lotus seed coat extract significantly reduced the lipofuscin level in the nematode body (Fig. 4B and Fig. 5). The level was 49.99%, 59.58% and 70.12% lower in the low-, medium-, and high-dose groups compared to the blank group, respectively (p 0.001).

Fig. 5.

Lipofuscin fluorescence image of C. elegans. Scale bar = 100 µm.

The ROS content in the body of C. elegans increases with the aging process. Excessive accumulation of ROS in the body can activate aging-related signaling pathways, inducing cell cycle arrest and the acquisition of a senescence-associated secretory phenotype (SASP) in cells, thereby accelerating the aging process [27]. Lotus seed coat extract significantly reduced the ROS level in C. elegans, as shown in Fig. 4C and Fig. 6. Lipofuscin levels in C. elegans were reduced by 28.92%, 40.87%, and 56.12% in the low-, medium-, and high-dose groups, respectively, compared to the blank group (p 0.001).

Fig. 6.

ROS fluorescence image of C. elegans. Scale bar = 100 µm.

SOD and CAT are the main antioxidant enzymes in C. elegans and can eliminate and reduce excessive free radicals, thereby delaying aging. MDA is one of the essential biomarkers of the oxidative stress response and can also reflect the aging status [28]. The results for SOD and CAT activity in the C. elegans body are shown in Fig. 4D. Lotus seed coat extract significantly improved the antioxidant enzyme activity in the C. elegans body in a dose-dependent manner (p 0.001). The results for MDA content in the nematode body are shown in Fig. 4E. Both the medium- and high-dose groups significantly reduced MDA content in the C. elegans body (p 0.05), with a non-significant decrease observed in the low-dose group.

Common targets were first imported into the STRING database. Those without interactions were removed, and high-confidence targets with a threshold value $>$0.900 were screened. The results were analyzed topologically using Cytoscape 3.7.2 software (Cytoscape Consortium, Seattle, WA, USA) to obtain an initial PPI network of 152 nodes and 580 edges. The target network was clustered using the MCODE plug-in, and targets with similar functions or closely related nodes were classified into clusters of the same topology. The Cytohubba plug-in was subsequently used for core target selection (Fig. 7B,C).

In the PPI network, the degree of freedom for each node represents the number of interactions with other nodes. The higher the value for the degree of freedom, the higher the importance of that node. The first 10 core genes predicted by this analysis were SRC, PIK3R1, PIK3CA, AKT1, EP300, ESR1, EGFR, IL6, TNF, and FGF2.

GO and KEGG pathway enrichment analyses were performed to analyze the biological functions and pathways involved in the 225 common targets. GO functional annotation enrichment analysis focused on three aspects: biological processes (BP), cellular composition (CC), and molecular functions (MF). The analysis identified 2016 BP, including regulation of hormone levels, regulation of nitrogenous compounds in cells, regulation of enzyme activity, and regulation of the MAPK cascade reaction; 122 CC including the cytoplasmic perinuclear region, cell bodies, transcriptional regulator complexes, mitochondrial membranes, and lysosomes; and 194 MF including oxidoreductase activity, transcription factors, protein kinase activity, and growth factors (Fig. 7D). GO functional annotation enrichment analysis revealed that the anti-aging effect of lotus seed coat is mediated mainly through biological processes related to hormone levels, enzyme activity, oxidative stress, immunity, inflammation, and blood.

KEGG pathway enrichment analysis indicated the anti-aging effect of lotus seed coat was mainly related to autophagy and to the phosphatidylinositol 3-kinase/protein kinase B pathway (PI3K–Akt), mitogen-activated protein kinase (MAPK), hypoxia-inducible factor 1 (HIF-1), forkhead box O (FOXO), tumor necrosis factor (TNF), and mammalian target of rapamycin (mTOR) ignaling pathways (Fig. 7E).

Neferine was identified as the primary alkaloid component in the lotus seed coat [29]. Analysis by high-performance liquid chromatography (HPLC) revealed that its content in the lotus seed coat was 0.033 mg/g (Supplementary Fig. 1). We next selected the top six key proteins in terms of degree value for molecular docking analysis with neferine. The binding energy of each protein to neferine was found to be 0, indicating a strong affinity with each (Fig. 8). Since phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA), histone acetyltransferase p300 (EP300), and RAC-alpha serine/threonine-protein kinase (AKT1) ranked as the top three proteins for affinity and are also associated with the PI3K–Akt signaling pathway, these were selected for molecular dynamics simulations. The results showed that each complex tended to stabilize within 50 ns, with the PIK3CA-neferine complex being the most stable (Fig. 9).

Fig. 8.

Binding modes and affinities of Neferine with six core proteins.

Fig. 9.

Molecular dynamics simulation results for Neferine with PIK3CA, EP300, and AKT. PIK3CA, phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit alpha; EP300, histone acetyltransferase p300; AKT1, RAC-alpha serine/threonine-protein kinase.