The most crucial step in the separation of TCMPOs is extraction. Table 1 depicts the methods and techniques used to extract, isolate and purify the polysaccharides from TCM. Table 1 illustrates that the traditional and most popular laboratory technique for producing TCMPOs is extract within hot or boiling water. In order to get useful TCMPOs yields, this approach often requires a high extraction temperature (60∼100°C), long duration (1∼24 h), high energy, a high water-to-material proportion, as well as several extraction stages. However, due to its relative conventionality, ease of control, and affordability, this approach is still widely utilized today. Additionally, to increase the extraction efficiency, a variety of extractant and supported techniques have been employed, including treatment with dilute acid (23, 27–30) or alkali (31, 32), high pressure (18, 33–35), high-power ultrasound (36, 37). Dilute acid or dilute alkali are capable of improving the amount extracted of polysaccharides, however several studies have shown that this method tends to damage the structure for TCMPOs which thus may further affect their biological activity (12, 38). According to earlier studies, the extraction yields of Dictyophora indusiata polysaccharides prepared using various extraction methods ranged from 5.62 to 6.48% (22). Compared with simple water extraction, microwave, ultrasonic and pressurized extraction take a shorter time and have higher extraction efficiencies (Table 1). However, in view of the above facts, the heat and extractant-assisted extraction method of TCMPOs is excessively conventional and has many limitations. Hence, the development of novel technology is imperative. For example, microwave extraction, which can generate high-frequency concussion and evenly penetrate the material to destroy the cell wall (39) and promote the extraction of intracellular polysaccharides (38) in a short time, is energy-efficient and could achieve a better yield. In addition, supercritical CO₂ extraction is a productive extraction method that identifies unique characteristics of the extract at crucial stages (40). During this process, the use of appropriate entrainers or modifying agents such as ethanol and methanol (41, 42) improves the probability of obtaining TCMPOs.

After the TCM water extracts are obtained, the isolation and purification process need to be first carried out by removing the non-polysaccharide components. The polysaccharides obtained through ethanol precipitation has been partly cleansed through deproteination as well as decoloration (38, 43, 44) before further being isolated and purified via dialysis (16), ultrafiltration (29, 45), membrane separation (18, 21), ion-exchange chromatography (46), gel filtration chromatograph (45, 47, 48), along with other methods. Figure 1 provides a schematic representation of the TCMPOs extraction and purification procedure. To obtain the pure fractions of TCMPOs, elution was carried out using the proper running buffers, then collection, concentration, dialysis, as well as lyophilization. For example, the water-extracted crude polysaccharides from Sargassum fusiforme were separated through a DEAE Sepharose Fast Flow column, as well as three polysaccharide fractions, namely, Sf-1, Sf-2, and Sf-3, were obtained from elution with water, 0.5 M NaCl, 2 M NaCl, respectively (23, 27). Wu et al. (21) fractionated the Hirsutella sinensis water extract into four fractions (H1, H2, H3 and H4) on the basis of their Mw through membrane separation. Zha et al. (49) fractionated the L. japonica water extract into the three fractions LP1, LP2, and LP3 using the ethanol precipitation method at ultimate concentrations of 40, 60, and 80%, correspondingly. Different separation and purification methods will lead to different TCMPOs components. However, some research suggests that most TCMPOs obtained by the present extraction, isolation and purification techniques are still crude products, and the quality of TCMPOs is difficult to control (12, 50, 51). First, the region of origin, the portions of the plant that are collected, the time of year it is harvested, and the technique of processing all affect how medicinally effective the TCM is (52). Second, non-template synthesized polysaccharides are never pure compounds regardless of how many procedures have been applied for purification (12). The acquired TCMPOs are always bound within either narrow or broad Mw ranges (12, 50). It is also worth noting that the extraction methods, solvent, equipment, personal operation method and so on differed from lab to lab. Therefore, special attention must be paid to the whole extraction, isolation and purification procedures to obtain high-quality TCMPOs.

Most often, the analysis of monosaccharide composition and proportion demands cleavage for glycosidic bonds via acid hydrolysis, followed by derivatization, detection along with measurement through HPLC, GC, or GC/MS (38, 43, 44, 57, 60, 61). In the process of derivatization, 1-phenyl-3-methyl-5-pyrazolone (PMP) is one of the best precolumn derivatization reagents in the determination of monosaccharides because its derivatives are not easy to cleave and generate isomeric peaks (48, 62). It is noteworthy that high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) was increasingly used by investigators to detect the composition of monosaccharides because there is no need for derivatization (35, 63).

As shown in Table 2, most TCMPOs are mainly composed of glucose (Glc), galactose (Gal), mannose (Man), arabinose (Ara), rhamnose (Rha), glucuronic acid (GlcA), galacturonic acid (GalA), xylose (Xyl), and fucose (Fuc) with different molar ratios. It is noteworthy that G. lucidum polysaccharides were detected to contain glucosamine (GlcN) with a percentage of 1.1% (18). The diversity of monosaccharide composition and their molar ratio is associated with the TCM sources, geographical origins and harvest season. For example, monosaccharide composition differences were discovered for S. fusiforme harvested in Qingdao, Wenzhou, separately (27, 37). In addition, different monosaccharides were detected in the polysaccharides of S. fusiforme that were harvested in Qingdao in May, July, and August. The polysaccharides of S. fusiforme harvested in July had the highest content of Xyl and Rha (23, 28, 29, 37). The differences in monosaccharide composition in polysaccharide fractions are also closely relevant to the extraction and separation methods. For instance, DFP-H, DFP-M, DFP-U as well as DFP-P were obtained from D. indusiata with different extraction methods and possessed similar monosaccharide compositions but exhibited different molar ratios of constituent monosaccharides (22). LP1, LP2 along with LP3 were extracted and partitioned from L. japonica by the ethanol precipitation technique with various final concentrations. The dominant monosaccharides of LP1 were Ara, Man, Gal and Glu at a molar proportion for 2.63:1.46:1:0.47, whereas LP2 as well as LP3 comprised a total made up of Rha, Ara, Xyl, Man, Gal and Glu at molar proportions for 0.13:1:0.3:1.28:1.64:1.19 as well as 0.39:1:0.67:3.01:2.47:0.99, correspondingly (49).

Multiple analytical techniques, including osmometry, viscosity measurement, sedimentation, GPC, HPGPC, HPSEC, and HPLC, are employed to determine the Mw distribution of TCMPOs, of which HPGPC and HPSEC are the most commonly utilized methods (38, 39, 43, 44, 64). As mentioned above, polysaccharides are never pure compounds regardless of how many procedures have been applied for purification. The acquired TCMPOs are always within either narrow or broad Mw ranges (12). Therefore, only the average Mw of TCMPOs can be detected. In addition, the average Mw mainly depend on the separation method, such as the different final ethanol concentrations (49), running buffers (23, 27, 48), separation membranes (18). Different Mw ranging from ∼10³ to ∼10⁷ Da were discovered under a variety of experimental conditions and source materials, which are demonstrated in Tables 1, 2.

The core of backbone structural analysis of TCMPOs is the sequence of monosaccharides and the configuration, type as well as location of glycosidic linkage. Generally, the priority is to investigate the backbone of polysaccharides with its branch link sites, the ring type (pyran or furan), the linkage sequence and the absolute configuration (D- or L-type) of monosaccharide residues, the α- or β-isomeric form of each glycosidic bond and feature groups (sulfate group, acetyl group, phosphate group, methyl group and so on) that may connect (12, 20, 65). Due to the complex structure of polysaccharides, the degradation (acid hydrolysis, periodate oxidation, smith degradation, etc.) of polysaccharides is usually performed first, followed by instrumental analysis methods (FT-IR, NMR, GC–MS, ion mobility-mass spectrometry, etc.) to obtain structural information (12, 48, 66, 67).

Studies have shown that the different fractions obtained from the same TCM may share the same glycosyl linkages, while the percentage of each glycosyl bond type varies significantly. For example, a structural investigation of two polysaccharide fractions (RRP1 and RRP2) from Rhodiola rosea indicated that RRP1 consisted of 1→6 or 1→ (50.63%), 1→2 or 1→4 (25.9%) as well as 1→3 (23.47%) glycosidic connections, while RRP2 consisted of 1→6 or 1→ (23.5%), 1→2 or 1→4 (60.0%) and 1→3 (16.5%) glycosidic connections (48). Furthermore, there were three polysaccharide fractions (LP1, LP2 and RRP2) from L. japonica, among which LP1 contained non-reducing terminal residues and (1→6)-linked (60%), (1→2)-/(1→4)-linked (35%) and (1→3)-linked (5%) glycosyl bonds, LP2 contained non-reducing terminal residues and (1→6)-linked (70%), (1→3)-linked (24%) and (1→2)-/(1→4)-linked (6%) glycosyl bonds, and LP3 contained non-reducing terminal residues and (1→6)-linked (24%), (1→3)-linked (59%) and (1→2)-/(1→4)-linked (17%) glycosyl bonds (49). Three kinds of maidong polysaccharides were examined for their structures, and these polysaccharides were all composed of Fruf-(2→,→2)-Fruf-(6→, →6)-Glcp-(1→ and →1, 2)-Fruf-(6→with molar ratios of 5.0:18.2:1.0:5.3 (Liriope spicata, LSP), 6.8:15.8:1.0:5.8 (Ophiopogon japonicus, OJP), as well as 8.3:12.3:1.0:3.9 (Liriope muscari, LMP) (53). In addition, the extraction and isolation method of TCMPOs may not affect their main chain structure. For instance, DFP-H, DFP-M, DFP-U, and DFP-P were obtained from D. indusiata with different extraction techniques, although variations in the kinds of glycosidic linkages and polysaccharide structures were not noticed (22). The structural characterization of TCMPOs is shown in Table 2, which indicates that most of the researchers actually did not further explore the structure of TCMPOs. The structure of TCMPOs is tightly related to biological activities (12, 57, 67), hence, it is particularly important to focus research on primary structure identification in the future.

Generally speaking, polysaccharides within aqueous solutions show a variety of chain conformations, including aggregates, random coils, as well as distinct helical forms (single, double, and triple helices) (43, 68–70). Some of the advanced technologies have been used to investigate the conformational features of polysaccharides, such as HPLC-static light scattering (HPLC-SLS) (70), HPLC-dynamic light scattering (HPLC-DLS) (70), viscosity test using the hypothesis of diluted polymer solutions (69), the Congo red test (28, 48), differential scanning calorimetry (DSC) (19), atomic force microscopy (AFM) (28, 43), circular dichroism (CD) (57, 69), fluorescence correlation spectroscopy (FCS) (69), transmission electron microscopy (TEM) (71), scanning electron microscopy (SEM) (28), and NMR spectroscopy (19, 72). For instance, research was done on the senior structures of two distinct polysaccharide fractions (RRP1 and RRP2) obtained from R. rosea, which were eluted using distilled water as well as 0.1 M NaCl solution, correspondingly (48), RRP1 was found to have a triple-helical conformation by the Congo red test, whereas RRP2 did not. RRP1’s surface included a modest lamellar or irregular dendritic structure, according to SEM tests, while RRP2’s surface topography was smooth and featured big wrinkles and drop-shaped bulges on the margins. AFM suggested that partial aggregation was observed in the polysaccharide chains of RRP1, and RRP2 had no molecular aggregations. These findings showed that the different elution solvents may have an impact on the advanced structure of TCMPOs. As shown in Table 2, only very few of the researchers have studied the senior structures of TCMPOs. The complexity and polymorphism of polysaccharides unquestionably bring major obstacles to their higher structural identification. Since the superior structures of polysaccharides may be more related to their bioactivities than to their primary structures (12, 51), further research and confirmation still need to be conducted in the future. Moreover, future technology development needs to concentrate on characterizing the high-level structural dynamic changes of polysaccharides (69).

It is generally accepted that the biological activities for polysaccharides are closely correlated with their Mw. According to related reports, TCMPOs with relatively high Mw usually demonstrate greater antiobesogenic activity than those with lower Mw. For example, varied Mw polysaccharides from H. sinensis oral administration to high fat diet (HFD)-induced obese rats, the high-Mw polysaccharides in fraction H1 (>300 kDa) showed greater antiobesogenic effects than the low-Mw polysaccharides in fraction H2 (10–300 kDa) as well as H3 (<10 kDa) (21). Administration of fraction H1 decreased body weight by ∼50% after 12 weeks and reduced the visceral fat and homeostatic model assessment-insulin resistance (HOMA-IR) index contrasted to the HFD control group, whereas fractions H2 and H3 only reduced the visceral fat pad mass and produced no influence on body weight or HOMA-IR index. In another study, the high-Mw polysaccharide PSF (≥100 kDa) from Polygonatum kingianum displayed better regulation of lipid metabolism via enhancing insulin sensitivity and more tightly controlling the make-up and activities of the gut microbiota (73). Comparable outcomes were also observed within the investigation of G. lucidum polysaccharide (>300 kDa) (18) and Pseudostellaria heterophylla polysaccharide (50∼210 kDa) (74).

Nevertheless, the association between the Mw of TCMPOs and their antiobesogenic activity is not simply a positive or negative correlation, and there are special cases that put forward different views (22, 75, 76). For instance, the antiobesity effects of four polysaccharides (DFP-H, DFP-M, DFP-U, and DFP-P) isolated from D. indusiata were investigated. The Mw size of the four polysaccharides was arranged as follows: DFP-H > DFP-P > DFP-M > DFP-U, whereas DFP-P showed the highest cholesterol/bile acids (BAs) attaching abilities as well as strongest inhibiting efficiency upon pancreatic lipase (22). Furthermore, it is worth noting that both DFP-P as well as DFP-H displayed significantly greater in vitro attaching characteristics, which includes fat, cholesterol as well as BAs, and lipase inhibiting properties compared to DFP-M and DFP-U, which may be because the relatively high Mw could reduce its water solubility, resulting in a decrease in hydrophilicity, thus enhancing the affinity for fat/cholesterol by increasing its surface aggregation (22, 76). The aforementioned data suggests that polysaccharides’ antiobesity effect is nevertheless enhanced by a reasonably high Mw. This may be explained by the fact that polysaccharides’ structural stability can only be preserved by a sufficiently high Mw (63). However, based on previous studies, maintaining a conformation with high biological activity of polysaccharides is not negatively impacted by either an excessive or modest Mw (77, 78), and the correlation between Mw and the antiobesogenic activity of TCMPOs should be further investigated.

The biofunctional activities of TCMPOs could be affected by their backbones and conformational properties. Prior studies have demonstrated that the structural properties of polysaccharides, including types of glycosidic bonds, monosaccharide compositions and molar ratios, degrees of branching (DBs) in the backbone, are closely related to their biological activity (59, 79, 80). Studies have found that β-glycosidic linkages usually exhibit stronger antiobesity activity than α-glycosidic linkages, causing the α-amylase in human small intestine which is specific for α-glycosidic linkages (55, 75). Polysaccharides rich in β-glycosidic linkages can remain undigested and are sequentially fermented by colonic microflora, thus playing a relatively important role in weight loss (55). DFP-H, DFP-M, DFP-U, and DFP-P obtained from D. indusiata possessed similar monosaccharide compositions but exhibited different molar ratios. Findings indicated that DFP-P and DFP-H were better at binding fat than DFP-U and DFP-M, which might be related to the higher ratios of galactose and glucose (22). Two polysaccharide components (TSP-2 and TSP-1) were extracted through Toona sinensis leaves. TSP-1 demonstrated more hepatoprotective action than TSP-2, which may be related to TSP-1’s higher galactose and glucose levels as well as its higher branching residue concentration (81). It is worth mentioning that the DB in the backbone is not positively correlated with activity; only the suitable DB is advantageous to the arrangement of the triple helical configuration (79).

The conformational features (spherical, random coil, double-helix, triple-helix, worm-like, rod-like) of polysaccharides may be more related to their bioactivities than the primary structures. Polysaccharides with triple-helix conformations exhibit more potent biological effects, including anti-inflammatory, antidiabetic and immunomodulatory effects, than those with other structures, and some of them, such as lentinan and schizophyllan, have been used in clinical therapeutics (70, 82). As previously mentioned, two polysaccharide fractions (RRP1 and RRP2) isolated through R. rosea were identified with different conformational features, and subsequent research demonstrated that the triple-helix structure enabled fraction RRP1 to exhibit more prominent antioxidant and hepatoprotective properties than RRP2 (48). Although the above structure-bioactivity interaction analyses of TCMPOs were not aimed at understanding their antiobesity activities, the anti-inflammation, antioxidant and hepatoprotective activities were considered as the potential requirements for the treatment of obesity (83, 84). It can be inferred that the conformational features of TCMPOs will also affect their antiobesity activity.

The functional group contents of polysaccharides are also linked to their bioactivities, like uronic acid, sulfuric acid and aminohexose (38, 85). Uronic acid residues, in particular, have the ability to modify the physiochemical characteristics and solubility of the corresponding polysaccharide conjugates, hence augmenting their biological activities. In a recent study, five fractions (MFPs-30-60, MFPs-30-80, MFPs-90-40, MFPs-90-60 as well as MFPs-90-80) had been purified from Fructus Mori, and in vitro studies revealed that MFPs-90-40, MFPs-90-60, and MFPs-30-60 showed a better capacity to bind BAs than MFPs-90-80 and MFPs-30-80, which was attributed to the increased uronic acid concentration and indicated that the uronic acid content was a significant signal representing the polysaccharide fractions’ hypolipidemic effects (86). In addition to the functional groups originally existing in polysaccharides, structural modifications, such as sulfation, acetylation, deacetylation, phosphorylation, acetylation, hydroxy-methylation, selenylation as well as complexation with zinc or iron, are of great significance to the physical and biological activities of polysaccharides. For example, the sulfation of G. lucidum polysaccharide (GLP) improved its antioxidant activities and BA-binding capacities (87). According to a different study, selenium modification may strengthen Chinese angelica polysaccharide’s hepatoprotective as well as antioxidant properties (88). Overall, the structure-bioactivity analyses of TCMPOs should be strengthened to better clarify the structural basis of their antiobesogenic activity.

Polysaccharides intake mainly regulates appetite and satiety through physical effects and chemical humoral stimulation effects (6), hence limiting the amount of food and energy consumed to stop the development of obesity. As shown in Figure 2, with bulking and viscosity-producing capabilities, TCMPOs could contribute to postprandial satiety by inducing gastric distension and delaying gastric emptying (89). Moreover, the appropriate ingestion of TCMPOs (such as Amorphophallus konjac polysaccharides) that, when hydrated, form viscous colloidal dispersions impacts several aspects of gastrointestinal (GI) function, including slowing down the transit of the small intestine (90, 91), inhibiting dietary enzymes from adhering to their substrates (especially fats and carbohydrates) (89), and creating an absorbent barrier layer within the small intestine by interacting with the mucosa (89, 91–93). This prolongation of the intestinal phase of nutrient processing and absorption may enhance satiety and help regulate food intake (89, 94).

More significantly, TCMPOs have the ability to control the release of gut hormones as well as peptides that influence appetite along with food intake, including leptin, adiponectin, glucagon-like peptide 1 (GLP-1), and peptide YY (PYY). A study demonstrated that Holothuria leucospilota polysaccharide (HLP) supplement outcomes with an amelioration in the levels of leptin and adiponectin, with the levels of leptin concentrations in the typical range exhibit benefits for preventing appetite, improving fatty acid oxidation, as well as decreasing body fat, but excess levels of leptin could lead to insulin resistance (95). Most notably, HLP effectively reversed the overexpression of hepatic acetyl-CoA carboxylase (ACC) and CD36 to prevent fat buildup within the liver and prevent adipocytes from secreting too much leptin (96). Another study illustrated that P. cyrtonema polysaccharides could activate the T1R2/T1R3-mediated cAMP signaling pathway to stimulate GLP-1 production and decrease appetite (97). Furthermore, some investigations have demonstrated that the microbial fermentation of TCMPOs to form short-chain fatty acids (SCFAs) may encourage the production of GLP-1 and PYY (discussed in detail later), leading to elevated insulin and reduced glucagon release as well as appetite suppression (98, 99). The sole known orexigenic or hunger hormone, ghrelin, has been linked to the management of long-term energy balance as well as has the ability to increase appetite and meal initiation (4). Cholecystokinin (CCK), another anorexigenic intestinal hormone, is released in response to meal consumption and is involved in satiety (4). However, the potential effects of TCMPOs on ghrelin and CCK responses and the TCMPOs-mediated ghrelin/CCK relation to hunger and satiety ratings have rarely been reported and still require further research.

Numerous studies point to the interaction among intestinal permeability as well as TCMPOs as one mechanism connecting obesity and related diseases (Figure 3; Table 4). Oral treatment with TCMPOs beneficially maintained intestinal barrier integrity primarily by promoting intercellular tight junctions and strengthening the intestinal mucous layer. As shown in Figure 3 and Table 4, various studies have reported that TCMPOs, such as L. japonica polysaccharides and S. fusiforme polysaccharides, could elevate the expression of tight junction proteins (TJPs), including zonula occludens-1 (ZO-1), claudin-1 and occludin, to regulate the tight junction structure, thus protecting the integrity of the gut epithelium (18, 30, 32, 161, 162). Additionally, L. japonica polysaccharide (LJP61A) administration enhanced the mRNA levels of Kruppel-like factor 4 (Klf4, a marker of goblet cells) and mucin 2 (Muc2, a protein secreted by goblet cells) in HFD-fed mice, indicating that LJP61A may be able to counteract the HFD-induced reduction in goblet cells (137). Specifically, it has been reported that LJP61A may also elevate the relative abundance of A. muciniphila, which is reported that it may enhance the expression of regenerating islet-derived protein 3 γ (RegIIIγ) in response to gram-positive bacteria in the intestine and aid in its own survival, thereby promoting the development of the intestinal mucus layer (137, 163). Lipopolysaccharide (LPS), a cell wall component of gram-negative pathogenic bacteria, can enter the circulation and trigger a variety of transcription factors associated with inflammation, resulting in the initiation of chronic low-grade inflammation that is known as a causal factor of obesity and hepatic steatosis (32, 164). Oral treatment with TCMPOs could improve the intestinal barrier to prevent LPS from entering the circulation, hence attenuating HFD-LPS-induced inflammatory reactions and exerting the antiobesogenic effects of TCMPOs. For example, D. indusiata polysaccharides (DIPs) could maintain intestinal integrity and alleviate HFD-induced obesity by significantly reducing the levels of LPS and proinflammatory cytokines (tumor necrosis factor-α, TNF-α; interleukin-1β, IL-1β; interleukin-6, IL-6; monocyte chemoattractant protein-1, MCP-1) and enhancing the secretion of anti-inflammatory cytokines (interleukin-4, IL-4; interleukin-10, IL-10) (161, 162). It is worth noting that SCFAs, generated by microbial fermentation of TCMPOs, are considered vital energy sources for intestinal epithelial cells and can promote the development of the intestinal barrier (165).