Research progress of polysaccharide detection methods in traditional Chinese medicine

Abstract

In recent years, with the advancement of analytical technology and the development of molecular biology, the research of traditional Chinese medicine (TCM) polysaccharides is becoming a new frontier and hot spot in life science research due to their effectiveness and non-toxicity. The quantitative determination of polysaccharides plays an extremely important role in the quality control of polysaccharide drugs. However, the structural complexities of natural polysaccharides present a major obstacle to their accurate quantification. The objective of this review is to provide a comprehensive guide for the accurate quantification of polysaccharides in TCM. In this paper, the quantitative methods of polysaccharides are divided into traditional quantitative methods and new methods, and summarized from the medicinal materials included in the Chinese Pharmacopoeia (2025 edition) and the relevant literature in the past ten years. The traditional methods for polysaccharide quantification include titration, colorimetry, high performance liquid chromatography (HPLC) and resonance light scattering (RLS). With the technological progress, new nondestructive testing methods such as near-infrared spectroscopy (NIR) and hyperspectral imaging (HSI) have gradually emerged. In addition, glycospectrometry, as another new method, can accurately quantify the structural characteristics and content of polysaccharides by combining positional enzyme digestion technology and HPLC. Moreover, this review elaborates on the principles and discusses the advantages and limitations of each method mentioned. This paper aims to provide a comprehensive reference for the accurate determination of polysaccharide content in TCM, enhance its practicability and precision, and promote its wider application in the quality control of TCM and Chinese patent medicine preparations.

1. Introduction

Traditional Chinese medicine (TCM) has a long history and a wide range of applications. Decoction followed by administration is the main form of TCM in clinical application (Zhao, Ma, & Li, 2018). Therefore, compounds with high water solubility and large polarity such as polysaccharides are one of the main active ingredients in TCM. Polysaccharides are widely existing biological macromolecules, which play a crucial role in maintaining life activities. In nature, polysaccharides can be found in almost all living organisms, including seeds of herbaceous plants, tissues of stems and leaves, body fluids of animals, and cell walls and extracellular fluids of bacteria, yeast and fungi (Singh, Kumar, & Sanghi, 2012).

The polysaccharides extracted from TCM are mainly heteropolysaccharides with complex structures composed of arabinose, rhamnose, glucose and other monosaccharides. There are two differences from other polysaccharides in quantitative detection. First, TCM polysaccharides require multiple steps to remove impurities (such as proteins, pigments, polyphenols, etc.), such as Sevage deprotein and H₂O₂ decolorization (Xia et al., 2025). For example, the crude polysaccharides from Scutellaria baicalensis Georgi stems and leaves (purity 43.2%) were deproteinized by the Sevag method, defatted with petroleum ether, decolorized with AB-8 macroporous resin, and dialyzed with a 3 × 10³ dialysis bag to remove small molecular impurities. They were then purified by diethyl amino ethyl (DEAE) seplife fast flow (FF) anion exchange column to obtain S. baicalensis stem and leaf polysaccharides 1 (SBSLP1, purity 95.4%, mass 623.4 mg) (Lin, 2025). However, other polysaccharides with simpler pretreatment steps and fewer interfering components usually do not require complex impurity removal. For instance, squid ink polysaccharides of animal origin can remove interference only by enzymatic extraction followed by deproteinization (Shi, Liu, Wu, Chen, Zhang, & Zhong, 2011). In addition, due to the structural diversity and lack of standard substances, TCM polysaccharides need to use alternative standards, such as glucose or dextran, and there is a lack of polysaccharide reference materials for qualitative and quantitative analysis (Zhang, 2015). Because the monosaccharide composition of dodder polysaccharide is different from that of glucose, the conversion factor of glucose to refined polysaccharide (f = 0.91) should be calculated to correct the quantitative deviation of glucose as the standard (Xu et al., 2011).

Polysaccharides utilized in TCM are recognized as active molecular entities, and their biological activity is intricately associated with various factors, including molecular weight, monosaccharide composition, physical and chemical properties, molecular conformation, and chemical modifications, such as sulfation and selenation. A large number of studies have shown that polysaccharides have a wide range of pharmacological effects, such as anti-tumor (Ma et al., 2019), anti-inflammatory (Zhang et al., 2019), antioxidant (Lin, Ji, Wang, Yin, & Peng, 2019), anti-aging (Feng et al., 2019), hypoglycemic (Chen et al., 2023), and immune regulation (Chen et al., 2021). Recent years have seen a notable surge in polysaccharide research, garnering interdisciplinary attention and driving substantial patent filings. Yet the focus on polysaccharide content in these patents remained ambiguous. Using the Baiten database, 3 480 patents were screened with “polysaccharide” as the keyword, 3 374 of which originated from China. Fig. 1 illustrates a 20-year trend in polysaccharide-content patent applications, peaking in 2016 with 331 filings-the highest annual total. Technical classification revealed 57.34% (1 954 patents) centered on human life-critical areas, highlighting their central innovative role. Quantitative analysis of polysaccharides in TCM has become a key quality control topic. Traditional techniques include titration, colorimetry, high performance liquid chromatography (HPLC) and resonance light scattering (RLS), while emerging methods like near-infrared spectroscopy (NIR) and saccharide mapping have emerged-each with distinct advantages and limitations in applicability.

Fig. 1.

Annual patent applications quantity of polysaccharide content in past 20 years.

This article synthesizes decade-long literature on TCM polysaccharide content determination, elucidating traditional methods’ strengths/weaknesses, emerging technologies’ characteristics, and future directions for quantitative detection (Fig. 2). Its significance lies in providing a scientific basis for selecting detection methods, enhancing quality control, and facilitating practical application in TCM formulations. This picture shows the classification of polysaccharide detection methods, distinguishing traditional methods from new ones with a circular layout. The “Traditional polysaccharide content determination methods” include:indirect iodine quantity method, phenol–sulfuric acid, anthrone-sulfuric acid, HPLC, high performance Gel permeation chromatography (HPGPC), high performance capillary electrophoresis (HPCE), high performance anion exchange chromatography (HPAEC), RLS. The “New methods” include: hyperspectral imaging (HSI), near-infrared spectroscopy (NIR), and saccharide mapping. This figure clearly presents the traditional and new methods for the determination of polysaccharide content, reflecting the diversity and development of detection technologies.

Fig. 2.

Qualitative and quantitative methods for determination of polysaccharide from TCM.

2. Traditional polysaccharide content detection method

The initial utilization of titration techniques for assessing polysaccharide content can be traced back to the early 20th century. The methodology (Wan et al., 2021) involves combining the sample with sulfuric acid and phenolphthalein in a boiling water bath, followed by titration to the endpoint using a potassium sulfate chromic oxide solution. The volume of the solution consumed during this process is then measured to calculate the polysaccharide content. A similar approach is employed for the determination of monosaccharides, albeit without the hydrolysis step. However, this method is relatively intricate, and variables such as the rate of titration and the vigor of shaking may introduce inaccuracies, leading to its diminished prevalence in contemporary practice. With advancements in technology, traditional techniques such as colorimetric analysis and HPLC have become more prevalent for the quantitative assessment of polysaccharides.

A review of the guidelines regarding the determination of polysaccharide content in TCM as outlined in the 2020 Edition of the Pharmacopoeia of the People's Republic of China (ChP), the 2024 Edition of the United States Pharmacopoeia (USP), the 11th Edition of the European Pharmacopoeia (EP), and the 18th Edition of the Japanese Pharmacopoeia (JP) reveals that the ChP specifies the determination of polysaccharide content in eight medicinal materials: Polygonati Odorati Rhizoma (Yuzhu in Chinese), Coriolus (Yunzhi in Chinese), Lycii Fructus (Gouqizi in Chinese), Dendrobii Officinalis Caulis (Tiepishihu in Chinese), Polygonati Rhizoma (Huangjing in Chinese), Rosae Laevigatae Fructus (Jingyingzi in Chinese), Ganoderma (Lingzhi in Chinese), and Mel (Fengmi in Chinese), as well as in the compound preparation Mishitong Capsule. In contrast, the USP only addresses the determination of polysaccharide content in Ganoderma, while no relevant regulations are present in the EP or JP (Table 1). Furthermore, there exists a substantial number of TCM products containing polysaccharides that require further investigation to establish detection standards for polysaccharides.

Table 1.

Summary of polysaccharide properties and determination methods.

No.	Traditional Chinese medicine names	Method of polysaccharide extraction	Reference substance	Method of determination	Wavelength of detection	Content requirements	Source
1	Polygonati Odorati Rhizoma	Water extracting-alcohol precipitating	Glucosum anhydricum	Phenol-sulfuric acid method	490 nm	Calculated as anhydrous glucose, polysaccharide content no less than 6.0%	Chp
2	Coriolus	Water extracting	Glucosum anhydricum	Indirect iodine quantity method	−	Calculated as anhydrous glucose, polysaccharide content (the total sugar content minus the monosaccharide content) no less than 3.2%	Chp
3	Lycii Fructus	Ether skim, alcohol and water double extracting	Glucosum anhydricum	Phenol-sulfuric acid method	490 nm	Calculated as anhydrous glucose, polysaccharide content no less than 1.8%	Chp
4	Dendrobii Officinalis Caulis	Water extracting-alcohol precipitating	Glucosum anhydricum	Phenol-sulfuric acid method	488 nm	Calculated as anhydrous glucose, polysaccharide content no less than 25.0%	Chp
5	Polygonati Rhizoma	Alcohol and water double extracting	Glucosum anhydricum	Anthrone-sulfuric acid method	582 nm	Calculated as anhydrous glucose, polysaccharide content no less than 7.0%	Chp
6	Rosae Laevigatae Fructus	Water extracting	Glucosum anhydricum	Phenol-sulfuric acid method	490 nm	Calculated as anhydrous glucose, polysaccharide content no less than 25.0%	Chp
7	Ganoderma	Water extracting-alcohol precipitating	Glucosum anhydricum	Anthrone-sulfuric acid method	625 nm	Calculated as anhydrous glucose, polysaccharide content no less than 0.90%	Chp
		Water extracting	Mannose, D-glucuronic acid, dextrose, galactose, and L-fucose.	HPLC-UV	250 nm	Calculated the sum of the percentages of mannose, d-glucuronic acid, dextrose, galactose, and l-fucose not be less than 0.7%	USP
8	Mel	Acetonitrile dissolving	Sucrose, maltose, fructose, glucose	HPLC-RID	−	Sucrose and maltose content may not be over 5.0%, the amount of fructose and glucose shall not be less than 60.0%, the ratio of fructose and glucose content shall not be less than 1.0	Chp
9	Mishitong Capsule	Alcohol extracting	Glucosum anhydricum	titrimetry	530 nm	Each capsule containing mistleleaf dry extract is measured by mistleleaf polysaccharide (calculated as anhydrous glucose) and shall not be less than 10.0 mg	Chp

Note: Chp:The 2020 Edition of the Pharmacopoeia of the People's Republic of China; USP: The 2024 Edition of the United States Pharmacopoeia. −: No specified detection wavelength in Chp or USP.

2.1. Colorimetric method

Colorimetric method including phenol–sulfuric acid and anthrone-sulfuric acid, it with UV–Vis commonly used in TCM polysaccharide content determination (Fig. 3).

Fig. 3.

Schematic representation of determination of polysaccharide content in TCM by colorimetric method.

This figure illustrates the process of determining the content of polysaccharides by the concentrated sulfuric acid method of phenol and the concentrated sulfuric acid method of anthrone. Initially, polysaccharides are hydrolyzed by sulfuric acid (H₂SO₄) into monosaccharides. Monosaccharides are further treated with sulfuric acid to obtain glycolaldehyde derivatives. This derivative reacts with phenol to form an orange-yellow complex, with a maximum absorption wavelength of 480 nm. This derivative reacts with anthrone to form a blue-green complex, with a maximum absorption wavelength of 620 nm. Finally, the absorbance was measured at the maximum absorption wavelength using a UV–Vis spectrophotometer, and quantitative analysis was conducted using a glucose standard curve. The content of polysaccharides was determined through these color reactions and instrumental detection.

2.1.1. Phenol-sulfuric acid colorimetric method

The phenol–sulfuric acid method is the predominant technique employed for the quantification of polysaccharides in TCM. In this method, a polysaccharide solution is subjected to a reaction with phenol and concentrated sulfuric acid, followed by the measurement of absorbance at 490 nm using a UV–Vis light detector after the completion of color development. The underlying principle involves the hydrolysis of polysaccharides, oligosaccharides, and disaccharides into monosaccharides by concentrated sulfuric acid, which subsequently undergo rapid dehydration to yield sugar aldehyde derivatives. These derivatives then react with phenol to produce an orange-yellow compound. The absorbance is recorded by the UV detector, allowing for the determination of polysaccharide content through the correlation between the absorbance and the concentration of a standard substance, specifically glucose solution. This method is applicable for the quantification of methylated sugars, pentose sugars, and polysaccharides. With advancements in technology, numerous novel methods for the quantitative analysis of polysaccharides have been developed, including NIR and hyperspectral imaging (HSI), as discussed subsequently. Notably, these methods often rely on the phenol–sulfuric acid method for the initial determination of polysaccharide content, which serves as a basis for subsequent modeling and predictive analyses. Consequently, the phenol–sulfuric acid method remains a critical approach for polysaccharide detection. In addition to the six types of Chinese medicinal herbs specified in the Pharmacopoeia, such as Polygonati Odorati Rhizoma, Lycii Fructus, Dendrobii Officinalis Caulis, Polygonati Rhizoma, Rosae Laevigatae Fructus, and Ganoderma, various other herbal materials, including Pseudostellariae Radix (Taizishen in Chinese), Poria (Fuling in Chinese), Arteisiae Argyi Folium (Aiye in Chinese), Ginseng Radix et Rhizoma (Renshen in Chinese), and Panacis Quinquefolia Radix (Xiyangshen in Chinese) have also had their polysaccharide content assessed using this method. This technique was employed to analyze the polysaccharide content in 28 batches of Pseudostellariae Radix extracts obtained from diverse provinces in China, including Guizhou, Fujian, Jiangsu, Anhui, Hebei, Shandong, and Shanxi, to evaluate the impact of varying cultivation conditions on polysaccharide levels. The findings indicated significant discrepancies in polysaccharide content among the Pseudostellariae Radix samples from different regions, with the highest polysaccharide content recorded in samples from Guizhou at 25.13%, while the lowest was observed in samples from Anhui at 10.14%. The average polysaccharide contents of the other samples were situated between the aforementioned extremes (Sha et al., 2023). A total of 71 batches of water-soluble polysaccharides derived from various regions of Poria were examined using this methodology in the existing literature. The lowest observed content was 0.342% in samples collected from Hubei, whereas the highest was recorded at 3.265% in samples from Yunnan; however, the polysaccharide content in Yunnan samples exhibited considerable variability, ranging from 1.037% to 3.265%. Due to the limited number of samples obtained from Shanxi, Guangxi, Guizhou, Guangdong, and Hubei, any comparisons involving these regions may not be representative. In contrast, the polysaccharide content in samples from Anhui varied from 0.983% to 2.592%, while samples from Hunan ranged from 1.369% to 3.015%. Minor differences in polysaccharide content were noted among samples from different regions (Yi, 2022).

2.1.2. Anthrone-sulfuric acid colorimetric method

The anthrone-sulfuric acid method is an experimental technique that involves dissolving anthrone in concentrated sulfuric acid, which is subsequently added to a polysaccharide solution in a single step. The resulting mixture is subjected to heating in a water bath, followed by cooling in an ice bath, after which the absorbance is measured at 620 nm using UV–Vis spectrophotometry. This method operates on principles akin to those of the phenol–sulfuric acid method, wherein polysaccharides undergo hydrolysis and dehydration, resulting in the formation of a blue-green compound upon reaction with anthrone. The quantification of polysaccharide content is achieved through colorimetric analysis. A notable characteristic of the anthrone-sulfuric acid method is its ability to detect a wide range of carbohydrates, encompassing not only pentoses and hexoses but also various oligosaccharides and polysaccharides. In a comparative analysis of the phenol–sulfuric acid and anthrone-sulfuric acid methods for quantifying polysaccharides in Ophiopogonis Radix (Maidong in Chinese), the phenol–sulfuric acid method yielded an average value of 42.96% with a relative standard deviation (RSD) of 1.73%, while the anthrone-sulfuric acid method produced an average value of 38.56% with an RSD of 3.40%. The comparison of RSD values indicates that the phenol–sulfuric acid method exhibits greater precision and reproducibility. In stability assessments conducted over a period of 0 to 120 min, the phenol–sulfuric acid method demonstrated a RSD of 0.46%, which was marginally higher than the 0.34% observed for the anthrone-sulfuric acid method, suggesting that the latter possesses slightly superior stability. Furthermore, in experiments evaluating sample recovery rates, both methods met accuracy requirements; however, the RSD for the phenol–sulfuric acid method was 0.99%, lower than the 2.77% recorded for the anthrone-sulfuric acid method. Collectively, the findings from all methodological evaluations indicate that the phenol–sulfuric acid method is more stable and reliable (Zhang & Liu, 2018).

2.1.3. Summary of colorimetric method of polysaccharides

Colorimetry is a common method for the determination of polysaccharides in traditional Chinese medicine. It is easy to operate and low cost, but its limitations cannot be ignored. The phenol–sulfuric acid and anthrone sulfuric acid methods are widely favored for their simplicity, rapidity, and reproducibility. Among them, the phenol–sulfuric acid method is more versatile, can accurately measure polysaccharides including methylated sugars and pentose sugars, and is less interfered by proteins. However, phenol solution is unstable and easy to be oxidized, which may affect the determination accuracy. The anthrone-sulfuric acid method is mainly suitable for plant samples, but its accuracy and sensitivity are low, and it requires strict control of temperature and time. It is worth noting that both methods are susceptible to interference by the sugar moiety of saponins and flavonoids (Zhang & Liu, 2018), which may lead to overestimation of polysaccharide content, especially in extracts with high levels of these compounds. Furthermore, during the calculation process, when glucose is usually used as the standard substance for the determination of polysaccharide content, the measured value always differs from the true value. Standard curves should be made respectively with various monosaccharide reference substances to obtain the slope k_i. From this, the correction coefficient k_i' of each monosaccharide, that is, their relative slopes with respect to glucose, should be calculated. Then, the correction coefficient of this polysaccharide should be calculated using the following formula: k' = ∑k_i' × w_i, where w_i is the mass percentage of the monosaccharide i that constitutes it. By applying this coefficient, the known composition of heteropolysaccharides can be determined based on glucose as the standard, and more accurate results can be obtained (Yin, Li & Ma, 2015). In addition, colored samples may also cause the readings to be higher. Meanwhile, the stability of the color reaction is also easily affected by factors such as environmental temperature, reaction time and reagent purity, and the experimental conditions need to be strictly controlled. In conclusion, the colorimetric method is more suitable for the analysis of samples with relatively single components or low content of interfering substances. In practical applications, the appropriate determination method should be selected based on the characteristics of the sample and the requirements for determination accuracy.

2.2. HPLC

HPLC is a widely employed analytical technique recognized for its efficiency and accuracy in the quantitative assessment of polysaccharide content (Ma et al., 2024). This method is characterized by its high separation capability and sensitivity, enabling effective separation and detection of polysaccharide components. However, due to the intricate structures and substantial molecular weights of polysaccharides, the selection of an appropriate column is crucial for the accurate determination of total polysaccharide content via HPLC. For instance, normal phase HPLC typically utilizes an acetonitrile–water mobile phase in conjunction with a silica gel stationary phase that is modified with amino groups. Conversely, in reversed-phase HPLC, columns filled with C₁₈ material are commonly used as the stationary phase. Other column types employed in this context include gel permeation chromatography (GPC), capillary electrochromatography (CE), and anion exchange chromatography (AEC). The role of detectors in HPLC is also critical, as they are responsible for the analysis and identification of compounds. In the quantitative detection of polysaccharides, various detectors are utilized, including UV-Vis, differential refractive index detectors (RID), evaporative light scattering detectors (ELSD), and electrochemical detectors (ECD). Given that polysaccharides lack luminescent groups, the detectors predominantly employed for their analysis are ELSD, RID, and multi-angle laser light scattering (MALLS). This section will elucidate the applications and characteristics of the commonly used detectors in the quantitative analysis of total polysaccharides, while also summarizing the advantages, disadvantages, and applicability of these detectors.

2.2.1. HPLC with a C₁₈ column

Pre-column derivatization is frequently necessary for the quantification of polysaccharide content utilizing C₁₈ as the stationary phase. The underlying principle involves the hydrolysis of the polysaccharide extract, followed by the application of 1-phenyl-3-methyl-5-pyrazolone (PMP) as a derivatization reagent to convert the monosaccharides present in the hydrolysate into derivatized products that exhibit UV absorbance. Subsequently, UV detection is employed for quantification (Fig. 4). This methodology is applicable for assessing the polysaccharide content in Ophiopogonis Radix. Prior to column chromatography, the samples undergo PMP derivatization. The Waters Symmet-C₁₈ column, utilizing an acetonitrile–phosphate buffer as the mobile phase, facilitates the detection of monosaccharide content in Ophiopogonis Radix at a wavelength of 250 nm. The polysaccharide content is determined based on the peak area and the cumulative percentage of mannose, glucose, and galactose. The average mass fraction of polysaccharide content across 11 batches of Ophiopogonis Radix was found to be 1.46%. This method is characterized by its simplicity and rapidity, making it suitable for the determination of polysaccharides in Ophiopogonis Radix (Sun et al., 2022).

Fig. 4.

General flow chart of HPLC-UV measurement of polysaccharide content and monosaccharide composition. Chromatograms of PMP derivatives of monosaccharide standards and sample adapted from this article (Li et al., 2022). HPLC-UV: high-performance liquid chromatography-ultraviolet detector; PMP: 1-phenyl-3-methyl-5-pyrazolone.

2.2.2. HPGPC

This technique operates on the principle of separating samples based on their molecular weight through the use of gel particle size (Li, Zuo, & Zheng, 2020). Initially, this detection method was employed to ascertain the molecular weight of high molecular weight compounds, particularly proteins and peptides. It has since become widely adopted for the determination of polysaccharide molecular weights (Zhang et al., 2022). It is noteworthy that HPGPC is now capable of quantitatively detecting polysaccharides. In the process of determining polysaccharide content via the HPGPC method, an analytical framework for molecular weight and distribution analysis is first established, followed by the identification of characteristic chromatographic peaks. Subsequently, quality markers are isolated through ultrafiltration and other techniques to serve as standard substances, from which a standard curve is generated to facilitate content calculation. In instances where the molecular weight of the polysaccharide under investigation corresponds to that of an existing standard polysaccharide, various polysaccharides with known relative molecular weights may also be utilized directly as standards for measurement. Given the established relationship between the elution retention time and the relative molecular weights of polysaccharides on the gel column, a standard curve is initially constructed using polysaccharides of known relative molecular weights. The relative molecular weights of the samples can then be derived from this curve based on their retention times. By connecting the PL aquagel-OH 6 column and the PL aquagel-OH 40 column in series and employing a refractive index differential detector, this methodology achieves enhanced column efficiency compared to the use of a single column, thereby enabling effective quantification of Aloe (Luhui in Chinese) polysaccharide content. Reference materials of mannose and three kinds of dextrans with relative molecular weights of 11 600, 147 600, and 27 300 were prepared to create a standard storage solution. This solution was subsequently diluted to various concentrations for liquid phase detection, and a linear regression analysis was performed on the peak area (RIU) against content (μg). The findings indicate a strong linear correlation between the content (μg) and peak area (RIU) for the mannose and glucan reference materials within a specific range. For instance, the linear relationship for glucan with a molecular weight of 11 600 is represented by the equation Y = 14 727X − 2 835, with a linear range of 2.18 − 163.35 μg. Furthermore, the relative molecular mass did not influence the linear equation of content. Two primary peaks were identified in the sample, with retention time of approximately 20 min and 25 min. Based on the content ratio of these two chromatographic peaks and the molecular weight distribution ratio, the polysaccharide content in five batches of Aloe was determined to be between 0.10% and 0.16%, demonstrating the capability to measure both molecular weight and polysaccharide content (Yang, Jiang, Qian, Shen, & Wang, 2016).

Two TSK GMPWXL columns (300 mm × 7.8 mm, 10 μm) coupled with an ELSD were utilized to separate polysaccharides extracted from Dendrobii Officinalis Caulis across various species and geographical origins. The major peak (DOP), which represents the predominant polysaccharides in the chromatographic profile of Dendrobii Officinalis Caulis, was isolated from the HPGPC fingerprint of Dendrobii Officinalis Caulis using an ultracentrifuge filter, thereby serving as the reference standard. Furthermore, the polysaccharide content of Dendrobii Officinalis Caulis from different regions was quantitatively analyzed, revealing significant variability in quality. For example, among various batches of commercially processed Dendrobii Officinalis Caulis sourced from Hong Kong, the polysaccharide content exhibited a range from a minimum of 100.34 mg/g to a maximum of 365.03 mg/g, indicating a more than threefold difference. While this methodology enhances the precision of polysaccharide content determination, it necessitates the isolation of DOP as a reference substance, and the selection and separation of DOP are frequently constrained, thereby limiting the applicability of the HPGPC-ELSD method (Xu et al., 2014).

Currently, the TSK-GEL G4000PWXL gel chromatography column is employed for the analysis of standard polysaccharides, while the TSK-GEL G6000PWXL tandem combination column is utilized for the examination of TCM polysaccharide samples. This analytical approach is integrated with multi-angle laser light scattering (MALLS) and a RID to quantitatively assess the polysaccharides derived from three medicinal plants: Notoginseng Radixr et Rhizoma (Sanqi in Chinese), Panacis Quinquefolii Radix, and Ginseng Radix et Rhizoma. The quantification process relies on the linear refractive index response to standard samples, wherein the refractive index increment (dn/dc) serves as a specific parameter indicative of the polymer’s response to the RID. This parameter facilitates the direct calculation of polysaccharide content based on the concentration-specific refractive index increment equation. The determined polysaccharide contents for Notoginseng Radix et Rhizoma, Panacis Quinquefolii Radix, and Ginseng Radix et Rhizoma were found to be 120.29, 70.03, and 70.96 mg/g, respectively. Furthermore, this methodology, which demonstrates an average recovery rate ranging from 89.2% to 102.3%, exhibits greater accuracy compared to HPSEC-ELSD analysis utilizing glucose standards as a reference (with an average recovery rate of 36.8% to 76.4%) and the phenol–sulfuric acid method (with an average recovery rate of 40.3% to 88.9%) (Cheong, Wu, Zhao, Li, 2015).

2.2.3. HPLC with gel columns (HPAEC)

HPAEC, utilizing an anion-exchange column for separation, is a widely employed technique for analyzing the monosaccharide composition and concentration of polysaccharides derived from TCM. The HPAEC system, when coupled with an electrochemical detector (PAD), produces distinct current signals corresponding to various monosaccharides, enabling their quantification based on the intensity of these signals (Yu, Shang, Feng, & Wang, 2014). Similar to HPGPC, HPAEC also facilitates the indirect estimation of polysaccharide content in TCM. This process involves the hydrolysis of polysaccharides, followed by the separation and quantification of the resulting monosaccharides. Notably, the aldehyde uronic acid content in polysaccharides derived from Angelica Sinensis Radix (Danggui in Chinese) can be quantified using this methodology. The calculated proportion of galacturonic acid (GalA) in Angelica Sinensis Radix polysaccharide is determined to be 53.62%, exhibiting a relative error of 8.67% when compared to the colorimetric method, which reports a value of 58.27%. This method demonstrates commendable reproducibility (Sun, Tang, Wu, & Gu, 2008). Furthermore, in addition to quantifying polysaccharides from Ganoderma, this technique is applicable for assessing the monosaccharide composition and content of polysaccharides in Ganoderma spore powder (Yu et al., 2014). In recent years, people are more and more interested in the synergistic application of ion chromatography and mass spectrometry. Researchers have capitalized on the benefits of ion chromatography, which does not necessitate derivatization and offers effective separation, in conjunction with the high precision and selectivity afforded by mass spectrometry. A suppressor has been employed to integrate HPAEC with Quadrupole Orbitrap High-Resolution Mass Spectrometry (Q Orbitrap HRMS) for analytical purposes. By implementing a post-column online sensitization technique, researchers have successfully formed adducts of metal salts with monosaccharides, thereby enhancing the sensitivity of monosaccharide detection in mass spectrometry. This established method has been utilized for the compositional analysis of polysaccharides in Lonicera Japonica Flos (Jinyinhua in Chinese) and Loniceae Flos (Shanyinhua in Chinese), revealing that glucose and arabinose are the predominant components. The glucose content is found to range from 49.6% to 66.7%, while the arabinose content varies from 10.3% to 21.6%. This methodology is characterized by high sensitivity, excellent repeatability, ease of operation, and reliable accuracy, thereby providing a valuable reference for the quality assessment of TCM such as Lonicera Japonica Flos and Loniceae Flos (Jin, 2023).

2.2.4. HPCE

In 1981, researcher Jorgenson pioneered the development of a capillary electrophoresis column by employing high voltage to facilitate separation within a capillary tube featuring an inner diameter of 75 μm. HPCE operates under the influence of a high-voltage electric field and represents a form of liquid-phase separation technology that utilizes capillary tubes as separation channels. This method achieves separation based on the variances in mobility and distribution behavior among the components of a sample. HPCE is extensively utilized across diverse fields due to its advantages, which include rapid processing, efficiency, high sensitivity, minimal sample requirements, and reducing contamination risks. The technique is primarily applied in the analysis of monosaccharide composition in polysaccharides (Ma et al., 2017, Yang et al., 2007) and in the quantitative assessment of microbial polysaccharides (Ding & Fang, 1999), However, there is a notable scarcity of quantitative research concerning TCM polysaccharides.

2.2.5. RLS

In recent years, RLS, as an emerging spectral analysis method, has gained increasingly wide applications in the analysis of biological macromolecules and ionic associations due to its advantages of high sensitivity and good selectivity (Huang & Li, 2003). It can achieve quantitative determination of nucleic acids, proteins, inorganic ions and sugar substances. The method of experimental operation and colorimetric method has similarities: In sodium hydroxide (NaOH) and cetyl pyridine chloride (CPC) to join in the mixed system of polysaccharide solution, room temperature incubation after 10 min, by fluorescence spectrophotometer for spectral scanning solution. The core principle is based on the phenomenon of light scattering: when light interacts with particles in a medium, if the particle diameter is much smaller than the wavelength of the incident light, molecular scattered light mainly in the form of Rayleigh scattering is mainly produced at this time (Huang & Li, 2003). According to the theory of RLS, the intensity of scattered light has a linear relationship with the concentration of scattered particles. By measuring the intensity of the scattered light of the system at the maximum resonant scattering wavelength, the concentration of the polysaccharide solution can be calculated based on the above linear relationship. This technology through a nanoscale particle scattering signal changes, for quantitative analysis of polysaccharide material in the complex system provides efficient and accurate means. This technology enables the quantitative analysis of polysaccharide components in complex systems by monitoring the scattering signal variations of nanoscale particles, thereby providing an efficient and accurate analytical tool. Owing to its scientific principles and facile operation, it exhibits significant application potential in fields such as life sciences and analytical chemistry. Utilizing this approach, the temporal variations in polysaccharide content within the leaf residue of Taxus Chinensis (Hongshushan in Chinese) were assessed. Notably, the highest polysaccharide concentration was observed in samples collected in early September, with collections occurring bi-monthly. When comparing the RLS method to colorimetric techniques, it was noted that the presence of minor quantities of proteins and phenolic compounds in the samples could contribute to elevated readings in the anthrone-sulfuric acid method. Both the RLS and phenol–sulfuric acid methods are susceptible to interference from polyphenols that share structural similarities with sugars; however, the RLS method requires significantly lower sample concentrations, thereby minimizing the impact of polyphenols on the results. Consequently, the quantification obtained through the RLS method is more representative of the actual polysaccharide content in the samples (Cheong, Wu, Zhao, & Li, 2015).

2.2.6. Summary of HPLC analysis of polysaccharides

In the HPLC analysis of polysaccharides, choosing the appropriate detector and chromatographic column is crucial for obtaining accurate results. Although HPLC has high resolution and sensitivity, its application is limited by the complex properties of polysaccharides. For example, the characteristics of high molecular weight and low polarity of polysaccharides may lead to clogging of the chromatographic column or decrease in separation efficiency, and the chromatographic column needs to be replaced frequently (Li, Li, Zuo, Tian, & Zheng, 2020). The selection of detectors needs to be balanced based on the characteristics of the sample and the analytical target. Although UV and fluorescence detectors can provide high sensitivity, due to the lack of inherent chromophores or fluorophores in polysaccharides, chemical derivatization with reagents such as PMP or 2-aminobenzoic acid is usually required. This process not only introduces additional complexity but may also alter the structural integrity of sugar. Therefore, these detectors have advantages only when derivatization is feasible and the sensitivity requirement is high. ELSD is widely adopted due to its universal applicability and stability. It detects analyte particles through light scattering after solvent evaporation, which is compatible with gradient elution and is less affected by temperature (Jalaludin & Kim, 2021). However, ELSD requires the use of non-volatile analytes and volatile mobile phases, and its nonlinear response curve may increase the difficulty of quantitative analysis. This method can effectively detect monosaccharides, oligosaccharides and polysaccharides, including natural starch. In contrast, although the RID is widely used in carbohydrate analysis, its sensitivity is relatively low and it is easily affected by fluctuations in the composition of the mobile phase, temperature and pressure. It is only suitable for analysis scenarios that do not require high sensitivity. Furthermore, although ion chromatography (IC) and gel permeation chromatography (GPC) can respectively analyze the composition of polysaccharides and monosaccharides and the molecular weight distribution, they rely on specific standards and the maintenance cost of the instruments is relatively high (Cheong et al., 2015). In conclusion, each detection technology has its unique advantages and limitations. In practical applications, the analysis requirements, sample characteristics and method feasibility need to be comprehensively considered.

3. New polysaccharide content detection method

3.1. NIR

Conventional analytical techniques, including colorimetric methods, HPLC, and gas chromatography-mass spectrometry (GC–MS), while demonstrating high levels of accuracy, reliability, and sensitivity, often necessitate costly instrumentation, intricate sample preparation, and protracted testing procedures. These factors render them impractical for large-scale rapid testing applications. In recent years, NIR spectroscopy has gained prominence in the identification and quantification of active constituents in TCM. This method offers several advantages, including rapid analysis, high efficiency, straightforward operation, minimal environmental impact, non-destructive testing, and the capability for simultaneous multi-component analysis (Chen, Li, & Fan, 2016). The principle of NIR is based on the Lambert-Beer law, which uses the overtones and combined vibration characteristics of molecular vibrations to achieve quantitative analysis by analyzing the absorption of light at wavelengths between 950 nm and 1 650 nm (Wang, Zhang, Adhikari, & Zhang, 2023). Due to the serious overlap of NIR absorption bands, it is necessary to eliminate noise and interference by spectral preprocessing [such as Savitzky-Golay smoothing, orthogonal signal correction (OSC)], and then use variable selection methods [such as selectivity ratio (sRatio), importance of projection variable (VIP), genetic algorithm (GA)] to screen characteristic wavelengths. Finally, multiple linear regression (MLR), partial least squares regression (PLSR), back propagation neural network (BPNN) and other machine learning algorithms were combined to establish quantitative models of spectra and chemical components. The core of the model is to extract effective chemical information from complex spectra by multivariate calibration and deconvolution techniques (Cheng et al., 2021).

The machine learning algorithm uses support vector machine (SVM), extreme learning machine (ELM), decision tree (DT), random forest (RF), principal components regression (PCR), and partial least squares discriminant analysis (PLS-DA), which is a widely utilized regression learning technique. The researchers employed these six distinct algorithms to develop quantitative models, selecting the most effective algorithm based on the impact of the polysaccharide content from Anoectochilus roxburghii (Wall.) Lindl. on the predictive outcomes. The polysaccharide content, quantified using the phenol–sulfuric acid method, served as the reference value for NIR quantitative analysis, while the root mean square error of cross-validation (RMSECV) was utilized to assess the model’s quality (Tang, Chen & Li, 2018). The final evaluation metric for the NIR quantitative model is the residual prediction deviation (RPD). An RPD value exceeding eight indicates a high level of prediction accuracy for the model. The findings of this research demonstrate that the partial least squares (PLS) constructed quantitative model for polysaccharides exhibits the lowest root mean square error of cross-validation (RMSECV). Specifically, the root mean square error of calibration (RMSEC) for this model is 0.625, while the root mean square error of prediction (RMSEP) is 0.767. The RPD value is calculated to be 8.467, signifying that the model attains the highest accuracy, thereby facilitating the prediction of polysaccharide content in A. Roxburghii (Zhang et al., 2023).

The polysaccharide content of Dendrobium huoshanense C. Z. Tang et S. J. Cheng was quantified using the anthrone sulfate method, which served as the reference standard for NIR quantitative analysis. A model for determining the polysaccharide content of D. huoshanense was constructed employing the PLS chemical analysis technique. A comparative assessment was performed to analyze the accuracy and reliability of the predicted values generated by the NIR model in relation to those obtained through the traditional detection method. The RSD of the anthrone sulfuric acid method was determined to be less than 2.5%, indicating that the NIR quantitative method established exhibits remarkable precision and stability. To further validate the applicability of the developed NIR model, an additional 20 batches of D. huoshanense were analyzed, and the predicted polysaccharide contents were compared with their actual measurements. The results indicated that the difference in polysaccharide content measurements between the two methodologies was less than 1.6%. Consequently, the NIR method developed in this study is proficient in rapidly and accurately predicting the polysaccharide content of D. huoshanense (Hao et al., 2018).

Based on NIR combined with chemometrics methods, a rapid and non-destructive quantitative analysis model of Poria polysaccharides was established. The study took the polysaccharide content of Poria determined by the phenol–sulfuric acid method as the reference standard. The polysaccharide content of 71 batches of Poria samples from different origins was determined to be between 0.34% and 3.26%. By collecting the near-infrared spectral data of Poria samples and combining spectral preprocessing to construct a quantitative model, the R²c and R²P of the quantitative model were both greater than 0.95. The RPD is 4.86, which fully proves that there is a high linear correlation between spectral characteristics and the content of target components, and the optimized model has high prediction accuracy and stability (Yi, Hua, Sun, Guan, & Chen, 2020).

This methodology can be employed to assess and forecast the polysaccharide content of Cistanches Herba (Roucongrong in Chinese). In this study, researchers gathered 71 batches of Cistanches Herba polysaccharide samples and utilized the phenol–sulfuric acid method to ascertain the polysaccharide content, which served as the reference value. Subsequently, machine learning techniques were applied to establish a correlation between the spectral data and the reference values. The samples were partitioned into calibration and prediction sets in a 4:1 ratio, utilizing the Kennard-Stone algorithm. The model underwent optimization through various preprocessing techniques, with the final results indicating that the RF algorithm outperformed the other three machine learning models. The determination coefficient R²c for the calibration set was recorded at 0.976 3%, while the RMSEC was 0.352 7%. A predictive model for the polysaccharide content of Cistanches Herba has been established, which can significantly enhance the quality control of this species. However, there are still some limitations in the practical application of this method. Firstly, the absorbance of NIR spectra is not only affected by the chemical composition of the sample, but also closely related to the physical characteristics of the sample, such as moisture content, particle size distribution, density and surface roughness. High moisture content may cause the spectral signal to be masked by the absorption peaks of water molecules, which in turn interferes with the identification of characteristic peaks of polysaccharides (Xie, Wang, Zhao, & Zhao, 2023). In addition, the uneven particle size of the sample may lead to enhanced light scattering effect and reduce the prediction accuracy of the model (Hao et al., 2018). Therefore, samples should be strictly dried and homogenized before NIR analysis, and correction models for different physical states should be established. Second, the performance of the NIR model is highly dependent on the representation of the training data and the choice of the stoichiometric algorithm. Insufficient sample diversity or over-fitting of the model may lead to significant deviations in the prediction results during external validation (Tang et al., 2018). Although machine learning algorithms (such as PLS and RF) can be used to optimize the robustness of the model, its universality still needs to be further verified.

3.2. HSI

HSI needs to be used in conjunction with NIR. Unlike NIR, HSI technology does not require the herbal materials to be powdered, allowing for non-destructive testing. The HSI system is equipped with a near-infrared hyperspectral camera and a specially designed dark box. The camera is located at the top of the dark box, and four tungsten halogen light sources are evenly distributed at the four corners of the box. The sample stage can be raised or lowered to maintain a distance of 50 cm from the camera. The spectral collection is in diffuse reflection mode, and a black-and-white background is collected for calibration before the actual data collection (Gowen et al., 2007). The HSI and NIR methodologies exhibit similarities in that the polysaccharide content, quantified through the phenol concentrated sulfuric acid method, serves as the reference value. Subsequently, the relationship between reflectance and wavelength is established through software processing and modeling, ultimately leading to the calculation of polysaccharide content (Lu & Peng, 2006).

As shown in Fig. 5, it takes Ganoderma as the research object and presents the development process of the determination method from traditional to modern and from cumbersome to convenient and non-destructive through three technical means, as follows: The “UV–Vis” spectrophotometer and the total polysaccharide content analysis chart are shown at the bottom of the figure. This method is for Ganoderma extract. The Ganoderma needs to be extracted and processed. After obtaining the extract, the total polysaccharide content is detected by UV-Vis spectroscopy. This method is rather cumbersome to operate, has a complex extraction process, is destructive to the sample, and involves many steps and takes a long time. However, it was an important means used in the early stage to analyze the polysaccharide content of Ganoderma providing a data basis for subsequent methods. The NIR spectrometer and Ganoderma powder are shown on the left of the picture. This method grinds Ganoderma into powder and analyzes it through NIR scanning. Compared with the traditional ultraviolet method, it does not require a complex extraction process, directly detects powdered samples, is more convenient to operate, has a faster analysis speed, and causes relatively less damage to the samples. The near-infrared spectroscopy method can rapidly achieve qualitative or quantitative analysis of the components of Ganoderma powder by obtaining its spectral information, reflecting the development of detection technology towards a more convenient direction. The test of Ganoderma is shown on the right side of the figure. This technology utilizes a spectrometer, light source, computer and image acquisition software to directly take pictures of complete Ganoderma, conducts hyperspectral imaging analysis, and extracts the average spectrum of the region of interest (ROI). This method emerges with the development of technology and has significant advantages in non-destructive testing. It does not require any pretreatment of Ganoderma (such as extraction or crushing), and can directly conduct rapid and comprehensive analysis on complete samples. It can obtain both image information and spectral information of the samples, achieving a combination of visualization and spectral analysis. It represents a more advanced, efficient and non-destructive testing trend.

Fig. 5.

Taking Ganoderma polysaccharide content detection as an example, three methods of UV–Vis, NIR, and HSI were used to determine the content of Ganoderma polysaccharide in three forms.

Total polysaccharides contents, HSI image and mean spectrum of ROI from this article (Liu et al., 2022). NIR spectra from this article (Zhang et al., 2023). The three techniques in the figure are arranged in a circular pattern, ranging from the traditional complex extraction liquid detection (UV–Vis) to the relatively simple powder detection (NIR), and then to the advanced non-destructive complete detection (hyperspectral imaging). This clearly presents the evolution of Ganoderma detection technology from old to new, and from cumbersome to convenient and non-destructive, reflecting the innovation of analytical methods brought about by technological progress. Make the detection more efficient, more accurate and more friendly to samples.

HSI represents a novel platform technology characterized by numerous continuous bands corresponding to each spatial location of the subject under investigation. A hyperspectral image, referred to as a hypercube, constitutes a three-dimensional data structure encompassing two spatial dimensions and one wavelength dimension (Lu & Chen, 1999). Therefore, each pixel in a hyperspectral image contains the spectrum for that specific location. The resulting spectrum is like a fingerprint and can be used to characterize the composition of that specific pixel. HSI technology began in 1985 and was initially applied to earth remote sensing technologies on aircraft and spacecraft (Goetz, Vane, Sokomon, & Rock, 1985), but it has recently become a powerful analytical tool for non-destructive food quantitative analysis. This technology was used to measure the content of pectin polysaccharides in food, establishing a quantitative model related to the Vis-NIR HSI data of whole mulberries and their pectin polysaccharide content. Researchers predicted the contents of dilute alkali-soluble pectin (DASP), water-soluble pectin (WSP), chelator-soluble pectin (CSP), and total soluble pectin (TSP). The experiment analyzed two varieties of mulberries and found that DASP and TSP had better predictive results in mulberries stored at room temperature. The pectin content ranged from DASP 23.525 to 91.787 g/kg, WSP 17.333 to 117.443 g/kg, CSP 22.914 to 135.522 g/kg, and TSP 63.773 to 344.752 g/kg. The study indicates that HSI is a promising alternative method for the rapid and non-destructive measurement of polysaccharide chemistry (Yang et al., 2021). It is worth noting that HSI detection methods have a prominent advantage of being non-destructive. Compared to traditional polysaccharide detection methods (such as colorimetry), it can quickly and non-invasively determine the polysaccharide content in Ganoderma, and it is also suitable for analyzing the changes in polysaccharide content during the growth process of TCM. The study collected a total of 280 Ganoderma samples from four different growth stages. Through the phenol–sulfuric acid method, the average polysaccharide content for the samples from the four periods was found to be 1.14% ± 0.12%, 0.77% ± 0.09%, 0.61% ± 0.11%, and 0.52% ± 0.07%, indicating a relatively low content that gradually decreases. The Ganoderma lucidum samples were illuminated using a diffuse halogen light source and scanned with a hyperspectral camera to capture visible and near-infrared spectral data. Comparisons revealed significant differences in the visible spectrum (VIS) spectral reflectance curves, with coefficient of determination of calibration (R²_cal) and coefficient of determination of validation (R²_val) being 0.841 and 0.900, respectively. A regression model for determining the polysaccharide content of Ganoderma stablished using NIR spectral data showed better results, with R²_cal and R²_val being 0.886 and 0.924, respectively. This indicates that both VIS and NIR spectral data can accurately predict the polysaccharide content of Ganoderma fruiting bodies, with NIR showing superior performance. It provides a feasible method to improve the quality and economic value of Ganoderma. It has the potential for high-throughput detection (Liu et al., 2022). HSI was used in the determination of polysaccharide content in Lilii Bulbus (Baihe in Chinese). First, 500 batches of lily samples were collected from multiple regions and treated by natural sun-drying and sulfur fumigation. Images at 350−2 550 nm were collected using a specific imaging system, and spectral data were obtained through processing such as correction and band combination. The polysaccharide content was determined by the method of the kit (YX-W-ZDT, HEPENGBIO, Shanghai, China). The principle was the colorimetric method. Standard solutions of different concentrations were prepared to construct standard curves and calculate the content. The support vector machine (SVM), convolutional neural network (CNN), long short-term memory network (LSTM), and CNN-LSTM (CLSTM) models were used for prediction. CLSTM performed the best, with R² reaching 0.769, MAE being 19.63, and RMSE being 25.23. Compared with other models, it has obvious advantages and shows efficient and accurate application effects in the determination of lily polysaccharide content (Zhang et al., 2025).

Although HSI technology combines spatial and spectral information and can achieve non-destructive analysis of traditional Chinese medicine samples, its limitations also need to be noted. Firstly, hyperspectral data have high dimensions and a large amount of information, and need to rely on complex preprocessing algorithms (such as denoising and baseline correction) and feature band screening; otherwise, it is easy to cause model overfitting (Liu et al., 2022). Secondly, the surface reflectivity of the sample is significantly affected by light intensity, angle and background interference. Data need to be collected in a standardized dark box environment, which limits its real-time application in the field or production site (Li, Xu, You, & Lu, 2021). In addition, HSI has a high requirement for sample uniformity. If there is local spoilage or contamination on the surface of the medicinal materials, it may lead to deviation in the prediction results.

3.3. Saccharide mapping

The prevalent methodologies for quantifying total sugar content, specifically the phenol–sulfuric acid and anthrone-sulfuric acid techniques, necessitate the hydrolysis of polysaccharides into monosaccharides. This process does not account for the monosaccharides that may already be present in the sample, resulting in an overestimation of the total sugar content. Similarly, HPLC and GC also derive polysaccharide content from the total monosaccharides produced through complete acid hydrolysis. While these techniques exhibit greater sensitivity and stability compared to colorimetric methods, the optimization of acid hydrolysis conditions can be quite complex. In contrast, saccharide mapping, which employs enzyme digestion technology in conjunction with high-performance size-exclusion chromatography coupled with multi-angle laser light scattering and refractive index detection (HPSEC-MALLS-RID), facilitates both qualitative and quantitative analyses of polysaccharides based on their active structural characteristics (Zhu, Chen, He, Li, & Zhao, 2025). Enzymatic hydrolysis offers several advantages, including enhanced selectivity, high specificity, and mild reaction conditions. The HPSEC-MALLS-RID approach effectively addresses the challenges associated with the intricate structures and diverse types of natural polysaccharides, which often complicate the acquisition of corresponding polysaccharide standards in practical applications. The methodology involves first establishing a characteristic profile of the prepolysaccharide for enzymatic hydrolysis via HPSEC, followed by the hydrolysis of the target polysaccharide using localized glycosidases. The resulting enzymatic products are then separated, and MALLS is employed to ascertain the molecular weight, particle size, and solution chain conformation of the polysaccharide, while simultaneously identifying the polysaccharide itself. A correlation equation relating polysaccharide concentration to the specific refractive index increment of the polysaccharide is utilized for quantitative analysis, enabling precise determination of the proportions of various components within mixed polysaccharide reference materials. As shown in Fig. 6, it illustrates schematic diagram of the saccharide mapping.

Fig. 6.

Schematic diagram of saccharide mapping.

This technology has been effectively applied to the quality control of various traditional Chinese medicine polysaccharides and their products. When using this technology combined with the refractive index increment (dn/dc) method to determine the content of water-soluble polysaccharides in Lycii Fructus, after defatting samples with methanol and microwave-assisted extraction (900 W, 7 min), separation is performed via TSK-Gel G5000PW and G3000PW columns connected in series (0.9% NaCl mobile phase, 0.5 mL/min), with content calculated using dn/dc = 0.148 mL/g. The data show that the total contents of polysaccharide components (peaks 1 − 3) in Lycii Fructus from different regions vary: the average total polysaccharide content in samples from Ningxia is (677.5 ± 27.1) μg/mg, from Qinghai is (641.1 ± 33.5) μg/mg, from Inner Mongolia is (655.6 ± 58.4) μg/mg, from Xinjiang is (646.7 ± 24.7) μg/mg, and from Gansu is (676.4 ± 26.3) μg/mg. Notably, there is a significant difference in total content between Ningxia and Qinghai (P < 0.01), indicating that this method can accurately determine polysaccharide content and reflect regional differences (Wu et al., 2016). Similarly, this technology can be used to determine the polysaccharide content in Ganoderma dietary supplements. The specific procedure is as follows: after defatting samples with ethanol, polysaccharides are extracted via microwave-assisted extraction, followed by α-amylase digestion to remove interfering substances such as starch. Separation is performed using TSK-Gel G5000PW and G3000PW columns in series (mobile phase: 0.9% NaCl, flow rate: 0.5 mL/min, 35 °C), with simultaneous detection by MALLS and RID detectors, and content calculated using dn/dc = 0.15 mL/g. The data show that only five out of 19 batches of samples detected characteristic Ganoderma polysaccharide: the total polysaccharide content of the standard substance GL20 was 2.15%, while the total content in samples GL01, GL09, GL13 − GL15 ranged from 2.81% to 8.28%. In contrast, 14 batches of samples such as GL03 and GL04 had contents below 1.5% or were not detected. Through precise analysis of molecular weight distribution and content, this technology revealed that over 70% of Ganoderma products in the U.S. market had ingredients that did not match the labels, confirming its core application value in quality consistency evaluation, authenticity identification, and qualitative/quantitative analysis of polysaccharide components in dietary supplements (Wu et al., 2017).

Utilizing this methodology, the total polysaccharide content and the concentration of active polysaccharides (defined as components with a molecular weight exceeding 1 × 10⁴) in Ophiocordyceps (Dongchongxiacao in Chinese) were assessed under various cultivation conditions. The findings indicated that distinct cultivation conditions can selectively enhance the levels of active polysaccharides in Ophiocordyceps. Specifically, the concentration of active polysaccharides in samples cultivated with supplemental glucose reached a maximum of 26.3 mg/g. The total polysaccharide content across different cultivation conditions varied between 25.8 and 64.4 mg/g, exhibiting no significant correlation with the specific cultivation conditions employed (Wang et al., 2015). In contrast to traditional methodologies, this approach eliminates the need for polysaccharide hydrolysis, reference materials, and the preparation of standard curves, thereby offering improved accuracy and the capability to simultaneously quantify polysaccharides and their various components. In the quality benchmarking between domestic lentinan injection and imported products, the glucose profiling method has been put into practical use. By optimizing the production process of domestic lentinan injection with the help of glycography 4.0 technology, the proportion of its helical structure was increased to be comparable to that of imported products, and the clinical efficacy was significantly improved. It is expected to play a key role in the quality control and process optimization of more polysaccharide drugs in the future (Zhang et al., 2023).

4. Conclusion and future perspectives

Carbohydrates represent ubiquitous components in TCM, with polysaccharides standing out due to their large molecular weights, structural complexity, and poor stability, which collectively pose significant challenges for their qualitative and quantitative analysis. Despite these obstacles, TCM polysaccharides have garnered substantial attention for their remarkable bioactivities, including immunomodulation, antioxidant effects, antitumor properties, and antiviral capabilities, coupled with their favorable safety profiles. These attributes have driven continuous exploration of polysaccharide applications, underscoring the critical need for accurate quantification methods.

As pivotal bioactive constituents of TCM, polysaccharides demand precise detection technologies to ensure scientific rigor and effectiveness in quality control. This review systematically evaluates conventional analytical approaches (e.g., titration, colorimetry, and HPLC) alongside emerging techniques (INR, HSI, and glycomics analysis), elucidating their principles, merits, and limitations. While traditional methods offer cost-effectiveness, they suffer from interference by impurities. Modern technologies exhibit enhanced sensitivity but are often constrained by operational complexity and high instrumentation costs. Recent advancements in nondestructive detection (e.g., NIR, HSI) and enzyme-assisted glycomics coupled with chromatography have opened new avenues for high-throughput and precise polysaccharide analysis. Nevertheless, the structural heterogeneity of polysaccharides and sample diversity necessitate integrated strategies rather than reliance on single-method approaches.

Future breakthroughs are anticipated through multidisciplinary convergence, standardized protocols, and intelligent analytics. For example, the following three ways of change: (a) AI-driven analytical optimization: Machine learning algorithms, particularly deep learning architectures, could enable adaptive modeling systems for NIR/HSI spectral data, dynamically refining prediction models to enhance analytical efficiency. Integration of multi-source data (growth environments, processing parameters) may establish predictive databases to unravel polysaccharide content-efficacy correlations; (b) Advanced sensing platforms: Nanomaterial-based biosensors utilizing fluorescence or electrochemical mechanisms could achieve trace-level polysaccharide detection and real-time monitoring, facilitating rapid field screening and inline quality control during production; (c) Multidimensional characterization: Synergistic coupling of HPLC-MALLS, glycomics, and mass spectrometry, augmented by chemometrics, may resolve polysaccharide molecular weight distributions, monosaccharide compositions, and modification patterns, thereby constructing comprehensive multidimensional fingerprint libraries. Notably, transcending the conventional “total polysaccharide content” paradigm, activity-guided fractionation techniques (cell-based assays, molecular docking) could identify bioactive polysaccharide fractions with specific therapeutic effects, establishing a tripartite “structure–activity-content” quality control framework. For instance, glycomics-based identification of antitumor polysaccharide markers, combined with machine learning-optimized extraction protocols, could catalyze the transition from crude quantification to precision quality management.

In conclusion, advancing polysaccharide detection requires interdisciplinary integration of chemistry, biology, informatics, and engineering. Through technological innovation and standardization, the development of efficient, intelligent, and sustainable analytical systems will solidify the scientific foundation for TCM modernization and global recognition.

CRediT authorship contribution statement

Lanying Zhang: Conceptualization, Formal analysis, Writing – original draft, Visualization, Writing – review & editing. Xinrui Wang: Conceptualization, Writing – review & editing. Jingze Zhang: Conceptualization, Writing – review & editing. Dailin Liu: Conceptualization, Writing – review & editing. Gang Bai: Conceptualization, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.