Three-year-old P. sibiricum Red. were collected from Yingemen Town, Qingyuan County, Fushun City, China, and identified as P. sibiricum Red. by Professor Zhongyun Piao of Shenyang Agricultural University (Voucher specimen number PE 02089832). Rhizomes of uniform size (length: 8.0 cm–20.0 cm diameter: 1.0 cm–2.0 cm) were selected, cleaned, steamed (100 °C–105 °C) and dried in a drying oven (JINGHONG DHG-9140A; Shanghai, China) at 50 °C according to Table 1 specifications, labeled S0-S9. The epidermis of the rhizomes subjected to each cycle of drying remained dry while the interior was soft and tender. During processing, partial samples were collected after each cycle of steaming and drying, dried to water content≤8%, cut into small segments, freeze-dried, ground to a powder, sifted through a 60-mesh sieve, and stored at −80 °C.

Color measurements of P. sibiricum Red. powder was performed using a colorimeter (Chroma Meter CR-3; Minolta Corp., NJ, United States), which recorded lightness (L*), redness (a*), and yellowness (b*) values across samples subjected to different cycles of nine-steaming-nine-processing. Dried P. sibiricum Red. powder (2.5 g) was placed in a quartz dish (diameter: 3 cm) and the colourimeter probe placed below. Calibration was performed using a white standard plate. The colorimetric value (C) was calculated using Formula 1. C = a 2 + b 2 + L 2 1 / 2 (1)

P. sibiricum Red. powder (20.0 mg) was accurately weighed and added to 600 µL MeOH containing 2-Amino-3-(2-chloro-phenyl)-propionic acid (4 ppm), vortexed for 30 s, followed by the addition of 100 mg glass bead. The mixture was placed in a tissue grinder for 90 s at 60 Hz, extracted for 15 min using ultrasound at room temperature, and centrifuged for 10 min at 12,000 rpm and 4 °C. The supernatant was filtered through a 0.22 µm membrane and transferred into the detection bottle for Liquid chromatography-mass spectrometry (LC-MS) analysis.

LC-MS analyses were performed using an ultra-high performance liquid chromatography (UPLC) system (Vanquish, Thermo Fisher Scientific, United States) with an ACQUITY UPLC^® HSS T3 (2.1 × 150 mm, 1.8 µm) (Waters, Milford, MA, United States) column connected to Thermo Orbitrap Exploris 120 mass spectrometry detector (Thermo Fisher Scientific, United States). The column was maintained at 40 °C, with a flow rate of 0.25 mL/min and an injection volume of 2 μL. For LC-electrospray ionization (ESI) (+)-MS analysis, the mobile phases consisted of (B2) 0.1% formic acid in acetonitrile (v/v) and (A2) 0.1% formic acid in water (v/v). Separation was conducted under the following gradient: 0–1 min, 2% B2; 1–9 min, 2%–50% B2; 9–12 min, 50%–98% B2; 12–13.5 min, 98% B2; 13.5–14 min, 98%–2% B2; 14–20 min, 2% B2. For LC-ESI (−)-MS analysis, the analytes were separated using (B3) acetonitrile and (A3) ammonium formate (5 mM). Separation was conducted under the following gradient: 0–1 min, 2% B3; 1–9 min, 2%–50% B3; 9–12 min, 50%–98% B3; 12–13.5 min, 98% B3; 13.5–14 min, 98%–2% B3; 14–17 min, 2% B3. The ESI source conditions were set as follows: 30 arb of sheath gas, 10 arb of auxiliary gas, capillary temperature 325 °C, spray voltage 3.50 kV (positive) or −2.50 kV (negative), full MS resolution of 60, 000 with a primary ion scanning range of 100–1,000 m/z, secondary resolution of 15,000.

The filtered data were analyzed using Microsoft Excel 2019, SPSS 19.0, and SIMCA-P software for statistical tests. The heatmap was generated using Multiple Experiment Viewer (4.9.0). Differentially abundant metabolites (DAMs) were identified based on criteria of VIP ≥1, P value <0.05, and |log₂ (fold change (FC))| ≥1.

The quality of P. sibiricum Red. subjected to nine cycles of steaming and drying was evaluated using PCA. The membership function value U(X _ij ) of each individual index was calculated according to Formula 2, and the correlation and statistical frequency distribution among indices were determined for PCA. The weight (W _i ) of each comprehensive index and the D-value (comprehensive evaluation value) were calculated according to Formulas 3, 4, respectively, to evaluate the quality of P. sibiricum Red. subjected to different cycles of steaming and drying. U X i j = X i j − X i j , ⁡ min / X i j , ⁡ max − X i j , ⁡ min   j = 1 , 2 , 3 … … , n (2) P i   / ∑ i = 1 n P i   i = 1 , 2 , 3 … … , n (3) D i = ∑ i = 1 n U   X i j × W i   i = 1 , 2 , 3 … … , n (4)

The experiments were conducted in triplicates and expressed and mean ± SD. Statistical analyses of experimental data were performed using IBM SPSS statistical software. Duncan’s tests were conducted to evaluate significant differences at the 5% level. Graphical representations were generated using Origin Lab software (Origin Lab, MA, United States).

The traditional Chinese medicinal technique, known as “nine-steaming-nine-drying”, is commonly employed to mitigate adverse reactions, increase bioactive components, and enhance pharmaceutical effects of P. sibiricum Red. (Yu et al., 2018; Guan et al., 2024). However, this process alters the external appearance (e.g., color, luster) and internal texture (e.g., hardness, moisture content) of the rhizomes, posing challenges to standardization (Liu et al., 2020; Lu, 2021; Yu et al., 2023). In this study, unprocessed P. sibiricum Red. (S0) exhibited a yellowish-white color with tongue-numbing properties, color, as illustrated in Figure 1, after the third steaming (S3), the rate of color change decreased. Following the fourth steaming (S4), the tongue-numbing sensation dissipated, and the sweetness intensified. After the seventh steaming (S7), bitter and sour flavor notes began to develop. Previous studies found that the tongue-numbing sensation might be related to the mucilage, positive hexanal, camphene, saponins and calcium oxalate needle crystals, and polysaccharides are the main components of mucilage, therefore, in this study, the disappeared tongue-numbing sensation might be related to the decreased polysaccharides (Figure 8A) and the cracking of the calcium oxalate needle crystals dispersing within the mucous cells (Chen, 2023). The bitter flavor is usually related to the alkaloids, terpenoids, glycosides, amino acids and peptides (Germán, 2024; Luo et al., 2025; Bureau et al., 2025). As shown in Figure 2A, this investigation revealed that during processing, total content of the main bitter substances including alkaloids, glycosides, some bitter amino acids and peptides exhibited an increasing trend. The sour flavor might be related to the organic acids, as the gluconic acid, and the increasing trend was also observed for sour substances in processed P. sibiricum Red. as the cycles of steaming and drying increased (Figure 2B; Yan et al., 2018; Yan et al., 2024). These findings suggest that moderate processing can effectively ensure the quality and flavor of P. sibiricum Red.

To quantify color changes during processing, a colorimeter was utilized to measure color changes in P. sibiricum Red. rhizomes subjected to different cycles of steaming and drying. L* represents color lightness, ranging from black (L* = 0) to white (L* = 100). The degree of blackening increased as the L* value decreased rapidly before the fourth steaming (S4), as shown in Figure 3A, after S4, the L* value stabilized between 20 and 24, indicating that five-steaming-five-drying maximally reduced product brightness. Positive and negative a* values indicated red and green, while positive and negative b* values denoted yellow and blue, respectively. The a* values of S0-S9 (Figure 3C), ranging from 36.13 to −14.08, peaked at the fifth steaming (S5), with browning intensity increasing as a* values rose, followed by a sharp decline. The elevated a* value may be attributed to browning during processing (Pan and D. Melton, 2007), the negative a* value revealed the red component had faded away, and the color had been changed to dark black. The b* values, ranging from 47.72 to −10.15 (Figure 3D), reached a maximum at the second steaming (S2), with color intensity increasing as b* values rose, then declined sharply, likely due to P. sibiricum Red.’ gradual transition from yellow to black. The C value, represents the purity or saturation of a color, it indicates the proportion of spectral colors, the higher the C value, the higher the purity or saturation of the color. The C value of S0-S5, as shown in Figure 3B, ranged from 50.54 to 30.69, then declined sharply, indicated that reasonable steaming and drying potentially maintained the product’s color saturation to the greatest extent. These findings offer a quantitative basis for implementing precise quality control during processing, improving upon empirical judgment by addressing the challenge of distinguishing appearance changes between consecutive steaming and drying cycles with the naked eye (Liu et al., 2020; Guan et al., 2024).

Saccharides, including polysaccharides, oligosaccharides and monosaccharides, are the most important constituents of P. sibiricum Red. (Wang et al., 2017; Zhao et al., 2020; Zhao et al., 2020b). Polysaccharides exhibit blood sugar regulation and anti-cancer properties, representing significant medicinal components of P. sibiricum Red. (Jin et al., 2018; Chinese Pharmacopoeia Commission, 2020; Xiao et al., 2024; Wu et al., 2024). Research indicates that steaming treatment significantly affects both polysaccharides and their antioxidant properties (Fan et al., 2020; Wu et al., 2024; Zhang Q. H. et al., 2024). During processing, the polysaccharides showed a decreasing trend (Figure 9A). The content of total polysaccharides decreased significantly in S1−S3, decreased slowly in S4−S7, compared to the S0, the content of polysaccharides in S9 decreased by 47.38%.

The content of thirteen main oligosaccharides showed dynamic changes (Figure 9B), decreased slowly in S2, S4−S8, compared to the S0, whereas had no significant decreasing in S1, S3 S9; among the oligosaccharides, turanose rapidly degraded, the relative content decreased from 76.15% to 11.18% of total oligosaccharides, the stachyose decreased from 2.37% to 0.37%, while the trehalose increased from 20.09% to 65.66%, the cellobiose increased from 0.06% to 17.95%, these suggest that the turanose in the pcocessed P. sibiricum Red. can transform to smaller monosaccharides, or participate in other reactions. Meanwhile, fructose, galactose, and glucose emerge as the primary monosaccharides, with their contents increasing substantially to comprise 93.11%, 3.51%, and 1.10% of total monosaccharides, respectively, and among this monosaccharides, only the fructose had the greatest decline, this suggests that the changes that occur in the later processing may be related to fructose (Figures 9C,D). Given the noticeable increase in sweetness after processing, fructose serves as a key biomarker for measuring sweetness in processed P. sibiricum Red., which is consistent with previous studies (Jin et al., 2018; Zhang Q. H. et al., 2024; Shi et al., 2025). After processing, the reduction in the content of total polysaccharides and oligosaccharides, increased relative content of certain monosaccharides suggests the transformation of large molecular sugars into smaller monosaccharides. However, excessive steaming may intensify bitterness. This indicates that steaming and drying are beneficial for the conversion of the saccharides in P. sibiricum Red.

Amino acids and analogs represent key functional components that provide tonifying and strengthening effects while enhancing immune function. They serve as essential precursors for protein synthesis, enzyme formation, antibody production, and the formation of certain hormonal compounds in the body (Yu et al., 2019; Park et al., 2020). The analysis identified 88 amino acids, peptides, and analogs in unprocessed, steamed, and dried P. sibiricum Red., representing the most diverse group among all metabolites. As illustrated in Figure 10A, the identification revealed 29, 43, 26, 20, 12, 16, 12, 7, 10, and 51 DAMs of amino acids, peptides, and analogs in S0vsS1, S1vsS2, S2vsS3, S3vsS4, S4vsS5, S5vsS6, S6vsS7, S7vsS8, S8vsS9, S0vsS7, respectively. These data suggest a gradual reduction in changes, with more substantial differences observed in the initial five cycles of steaming and drying samples. Leucine, glutamic acid, proline, tyrosine, and valine have been identified as markers of unprocessed P. sibiricum Red. Post-processing analysis revealed amino acids predominantly existing as dipeptides or polypeptides, as demonstrated in Figures 10B–F, particularly in S0vsS7, where 47 amino acids increased and 4 decreased significantly. The metabolites showing the most substantial increase included S-Hexyl-glutathione, linatine, N-Acetylornithine, N2-Succinyl-L-glutamic acid 5-semialdehyde, tyrosine methylester, L-arogenate, (5-L-glutamyl)-L-glutamate, cyclopeptine, L-dopa, and gamma-glutamyltyramine, and others, primarily comprising dipeptides or polypeptides. Additionally, S7 exhibited elevated levels of pyroglutamic acid, leucine, tyrosine, proline, valine, glutamic acid, and arginine, confirming their status as distinguishing metabolites in the processed rhizomes. The most notable reductions occurred in L-serine, while other groups showed decreases in L-leucine, N-acetylaspartylglutamate, N-methyl-L-glutamic acid, saccharopine, N-acetyl-alpha-D-glucosamine, N-acetylleucine, L-glutamine, L-arogenate, N-a-acetylcitrulline, 2-amino-2-deoxy-D-gluconate, pyroglutamic acid, L-leucine, gamma-glutamyltyramine, predominantly comprising serine, leucine, glutamic acid, and their derivatives. These findings partially align with previous research (Park et al., 2020; Li et al., 2024). The results provide comprehensive insights into the differentiation of amino acids and their derivatives following steaming and drying processes. The variations in amino acid metabolite accumulation and degradation across different cycles of steamed and dried Polygonatum sibicum may correlate with steaming temperature, duration, drying temperature, and time (Chen N. Y. et al., 2024; Guan et al., 2024).

Flavonoids and polyphenols constitute essential active compounds with extensive applications in pharmaceuticals and functional foods (Wojdyło et al., 2009; Devkota et al., 2022; Zhang et al., 2022). Research has demonstrated that steaming treatments significantly enhance both the quantity of bioactive compounds, including flavonoids and polyphenols, and their antioxidant properties (Musilová et al., 2024; Ma et al., 2025; Ciudad Mulero et al., 2025). This investigation revealed that during processing, total flavonoid content exhibited an increasing trend (Figure 11A), peaking in S7, followed by a gradual decrease. A similar trend was observed for polyphenols (Figure 11B); compared to the S0, the content of polyphenols in S7 increased by 5.62 times. These findings suggest that processing causes a breakdown in the structural characteristics of some flavonoids and polyphenols, thereby increasing their accessibility for detection (Collins et al., 2024). Additionally, repeated high-temperature steaming may facilitate tissue cell fragmentation and covalent bond breakage, promote increased phenolic substance release (Mudgal and Singh, 2024; Ren et al., 2025). Furthermore, elevated temperatures can inactivate certain endogenous enzymes and prevent polyphenol oxidation, result in increased polyphenol content after steaming (Wojdyło et al., 2009; Liu R. L. et al., 2024; Xu L. et al., 2025; Xu N. et al., 2025).

Saponins demonstrate antibacterial, antitumor, and other therapeutic effects (Pan et al., 2024; Zhang Q. H. et al., 2024; Song et al., 2024). Analysis of saponin content in P. sibiricum Red. processed through multiple steaming and drying cycles reveals, as shown in Figure 11C, minimal variation during the initial two cycles. A significant increase occurred in S3 (P < 0.05), followed by gradual increases after S4. The mass fraction in S9 reached 74.06 ± 3.38 mg/g, representing a 1.58-fold increase compared to S0. Research suggests that steaming facilitates the conversion of steroidal saponins to glycosides and secondary glycosides, thereby elevated saponin content (Guan et al., 2024; Wang et al., 2024).

Previous studies have established that heat treatment affected sample color and processing product quality through enzymatic reactions and non-enzymatic reactions (Li et al., 2018; Suri et al., 2019; Ren et al., 2025). However, the steaming process for a long time (95 °C ∼ 105 °C) can cause the deactivation of polyphenol oxidase to inhibit the enzymatic browning. Therefore, it can be inferred that non-enzymatic browning was the main reason of the color change in P. sibiricum Red. subjected to nine cycles of steaming and drying. In this study, the content of polysaccharides and oligosaccharides decreased continuously (Figures 9A,B), the monosaccharides, especially the fructose, with their contents increased substantially in S1-S7, then rapidly decreased in S8 and S9 (Figures 9C,D), alongside an increase in nine main lipids in S3-S5 and decrease in S6-S9 (Figure 12A), significant decrease in 1-aminocyclopropanecarboxylic acid, gamma-aminobutyric acid, L-proline and L-leucine (Figure 12B), these findings indicate that steaming and drying induced the Maillard Reaction, where amino compounds (proteins, peptides, amines, and amino acids) reacted with carbonyl compounds (reducing sugars, lipids, aldehydes, ketones, polyphenols, ascorbic acids, and steroids) to produce various compounds including melanoids, ketones, aldehydes and heterocyclic compounds (Li et al., 2018; Li et al., 2018; Suri et al., 2019; Wan and Liu, 2019; Zhang Q. H. et al., 2024; Chen, 2023). 5-HMF, an intermediate product of the Maillard reaction, commonly occurs in traditional Chinese medicinal products and undergoes notable content changes during processing, storage, and compatibility. It demonstrates pharmacological effects, including antioxidation, improved learning and memory, antiallergy, neuroprotection, and anti-inflammation, while also finding applications in agricultural plant growth regulation and as a spice (Uchida et al., 2020; Joachim et al., 2022; Mahon et al., 2024).

During steaming and drying, 5-HMF content increased progressively until the seventh steaming (Figure 12C), then gradually decreased, where amino acids condense with reducing sugars to form Amadori/Heynes rearrangased after reaching its peak, suggesting that steaming and drying may promote the Maillard reactionement products, which further decompose into flavor compounds such as furfural and reduced ketones, as shown in Figure 12D, the content of aldehydes, ketones, carboxylic acids and derivatives increased significantly in S4-S8 compared to S0, nitrogen-containing heterocyclic compounds content increased progressively until the fourth steaming (Huang et al., 2024; Wang et al., 2024b). In this study, the caramelization reaction can not occur in P. sibiricum Red. subjected to nine cycles of steaming and drying because of the presence of amino acids, furthermore, only a small amount of ascorbic acid was detected in the processed P. sibiricum Red., but the dehydroascorbic acids or 2,3-Diketogulonic acid, as the intermediate products of ascorbic acid degradation, were not detected, in conclusion, it can be concluded that the browning mainly correlated with Maillard reaction.

Pearson’s correlation analysis finds widespread application across disciplines, facilitating the identification of crucial evaluation indicators in comprehensive assessments. This method enables the selection of metrics highly relevant to evaluation objectives, thereby streamlining evaluation models and enhancing assessment accuracy and efficiency (Zhang X. Y. et al., 2024; Mitiku et al., 2025). The analysis revealed relationships between 27 primary quality traits and 4 morphological traits (Figure 13). The bigger the circle, the stronger the correlation. As shown by the color bar, from the deepest red equals to the strongest positive correlation to the deepest blue equal to the strongest negative correlation. The parameter L* showed positive correlations with amino acids, peptides, and analogs, C*, xylose, imidazopyrimidines, organic acids and derivatives, arabinose, lipids, b*, and polysaccharides, with particularly strong positive correlations with the latter six components. Negative correlations were observed with carboxylic acids and derivatives, total monosaccharides, fructose, keto acids and derivatives, flavonoids, polyphenols, saponins, nucleoside and nucleotide analogs, and benzene and substituted derivatives, with notably strong negative correlations with the latter five components. 5-HMF demonstrated positive correlations with 23 parameters, particularly strong with iditol, imidazopyrimidines, glucose, erythrulose, and C*, while showing negative correlations with carboxylic acids and derivatives, total monosaccharides, and fructose. This investigation specifically illuminates the relationship between chroma and internal chemical composition, offering novel insights into how steaming and drying processes affect P. sibiricum Red. ’s flavor and quality.

Nine-steaming-nine-drying represents a TCM processing technique that alters medicinal materials by reducing toxicity, enhancing efficacy, improving absorption, and adjusting medicinal properties (Yu et al., 2018; Guan et al., 2024). The quality evaluation of processed P. sibiricum Red. rhizomes requires multiple parameters for accurate assessment. A comprehensive evaluation methodology incorporating PCA, membership function analysis, and cluster analysis encompasses multiple evaluation indicators, thereby avoids the limitations of single-indicator assessment and enables objective quality evaluation of medicinal materials (Chen W. M. et al., 2024; Zhang Y. Q. et al., 2024). The quality analysis of P. sibiricum Red. processed through nine cycles of steaming and drying utilized PCA. This analysis categorized the quality traits of P. sibiricum Red. into six main components (Table 2). The first principal component, comprising saponins, nucleoside and nucleotide analogs, flavonoids, polyphenols, and benzene and substituted derivatives, demonstrated a load score exceeding 0.90 with a contribution rate of 53.769%. The second principal component, based on carboxylic acids and derivatives, showed a maximum load score of 0.63 with a contribution rate of 13.221%. The third principal component indicators, including organic acids and derivatives, amino acids, peptides and analogs, and arabinose, exhibited load scores above 0.60 with a variance contribution rate of 7.901%. The fourth principal component presented an eigenvalue of 1.759 and a variance contribution of 6.514%. The fifth and sixth principal components showed variance contributions of 5.393% and 4.628%, respectively, with imidazopyrimidines and carboxylic acids and derivatives demonstrating a maximum positive impact on these components.

Based on the U-value and index weight of comprehensive indices for each sample, the D-value comprehensive quality evaluation was conducted for P. sibiricum Red. materials subjected to different steaming and drying cycles (Table 3). Higher D-value indicates superior quality of processed P. sibiricum Red. S1 and S0 exhibited lower D-values, while increasing steaming and drying cycles corresponded to gradual D-value increases, reaching peak quality at S7 before the gentle decline. These findings suggest that moderate steaming and drying effectively reduces P. sibiricum Red. irritability while enhances beneficial properties. Analysis of primary active ingredient changes in P. sibiricum Red., particularly those associated with pharmacological activity, establishes optimal processing parameters: L* value below 30, colorimetric value (C value) below 10, and a value approximately 2, conditions under which P. sibiricum Red. main active ingredient content remained elevated. Thus, these comprehensive evaluation methods effectively reflect TCM quality and provide an objective assessment of Chinese medicinal materials and herbal slices, establishing a robust foundation for selecting optimal therapeutic processing methods.