**2. Results**

#### *2.1. Chemical Structures of Sargassum fusiforme Polysaccharides*

The physicochemical properties of the two *S. fusiforme* polysaccharides prepared by water extraction (SfW) and acid extraction (SfA) are shown in Table 1. The results indicated that the chemical structures of SfW and SfA were quite similar. For example, they had comparable contents of total sugar (70.0% vs. 62.4%) and sulfate group (28.5% vs. 31.3%), and their average molecular weights were also very close. Moreover, the monosaccharide compositions of the two polysaccharides were very similar. Both of them were mainly composed of glucose, fucose, and galactose with small amounts of mannose, glucuronic acid, and xylose, but the detailed molar ratio had a slight difference. For example, the content of xylose in SfW was lower than that in SfA (0.14 vs. 0.03), while glucose was more abundant in SfA (1.05 vs. 1.13).


**Table 1.** Physicochemical properties of *Sargassum fusiforme* polysaccharides.

\* Man, mannose; GlcA, glucuronic acid; Glc, glucose; Gal, galactose; Xyl, xylose; Fuc, fucose.

Here, 1H NMR spectra of SfW and SfA are shown in Figure S1. The resonance signals of the two polysaccharides at 3.0–5.5 ppm were ascribed to the typical distribution of 1H NMR signals of the polysaccharides [18]. The unresolved peaks at 5.3–5.5 ppm were assigned to the anomeric protons of α-L-fucopyranosyl units [19]. The resonance signals of the two polysaccharides at 3.3–4.5 ppm were apportioned to the ring protons H-2 to H-5, but the pattern was different from each other, and the chemical shifts at 1.1 and 1.4 ppm were assigned to methyl groups of fucose units [20]. In addition, due to the complex and heterogeneous structure of sulfated polysaccharides, the broadening and overlapping of 1H NMR peaks makes it difficult to completely describe their structural characteristics.

#### *2.2. Effects of SfW and SfA on HFD-Induced Metabolic Disorders*

The effects of SfW and SfA on HFD-induced metabolic disorders were evaluated after 16 weeks of co-treatment with HFD and polysaccharides. As shown in Figure 1A, 16-week HFD feeding significantly increased the body weight of the mice compared to that of the blank group, and co-treatment with SfW and SfA exacerbated body weight gain. The fasting blood glucose level significantly increased after 4 weeks of HFD-feeding, and only treatment with SfW at the fourth week reversed the increase (*p* < 0.05) (Figure 1B). In OGTT, the blood glucose reached the maximal level at 30 min after the dextrose gavage, and the blood glucose level of the control group was significantly higher than that of the blank group (*p* < 0.001). The polysaccharides administration could not attenuate the HFD-induced glucose intolerance (Figure 1C). Treatments with SfW and SfA did not protect HFD-induced insulin resistance and epididymal fat weight gain (Figure 1D–F).

**Figure 1.** Effects of SfW and SfA on ( **A**) body weight, (**B**) fasting blood glucose, ( **C**) OGTT, ( **D**) ITT, (**E**) liver weight, and (**F**) epididymal fat weight in HFD-fed mice. Values are mean ± SD (*n* = 5–6). \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001 vs. Blank. ### *p* < 0.001 vs. Control. Data sets in ( **A**–**D**) were analyzed using unpaired two-tailed Student's *t*-test. Data sets in (**E**–**F**) were analyzed using one-way ANOVA followed by a Turkey's test.

The effects of 16 weeks of HFD and polysaccharide administration on gu<sup>t</sup> microbiota in fecal samples and cecal contents of mice were analyzed by 16S rRNA high-throughput sequencing. As shown in Figure 2A–B, HFD significantly decreased both the Chao1 and Simpson's indices of cecal microbiota but showed negligible effect on the Simpson's index of fecal microbiota. Oral administration of SfW and SfA did not significantly improve the decrease in α-diversity of fecal and cecal microbiota. The unsupervised principal components analysis (PCA) plot at the phylum level showed that PC1 and PC2 were able to explain 56% and 41.3% of the variation, respectively, and exhibited significant distinction between the cecal and fecal microbiota of the blank group (Figure 2C). Polysaccharide administration showed no significant regulatory effect on the dysbiosis of cecal and fecal microbiota at the phylum level (Figure 2C). In detail, HFD significantly increased the relative abundance of Actinobacteria and decreased the abundance of Bacteroidetes and Verrucomicrobia in both the cecal and fecal microbiota but only enriched Firmicutes and Proteobacteria in fecal microbiota. SfW only presented a regulatory effect on Proteobacteria in fecal microbiota (Figure 2H,J).

**Figure 2.** Effects of SfW and SfA on the (**A**) Chao1 diversity and (**B**) Shannon's diversity indices of cecal and fecal microbiota in HFD-fed mice. (**C**) PCA and (**D**) bar plots of cecal and fecal microbiota at the phylum level. The relative abundance of (**E**) Actinobacteria, (**F**) Bacteroidetes, (**G**) Firmicutes, (**H**) Proteobacteria, (**I**) Verrucomicrobia, and (**J**) Tenericutes in cecal and fecal microbiota. Values are mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001 vs. the corresponding blank group. Data sets in (**A**), (**B**), (**E**–**J**) were analyzed using one-way ANOVA followed by a Turkey's test. C-, cecal microbiota; F-, fecal microbiota.

The effects of SfW and SfA on the cecal and fecal microbiota in HFD-fed mice were also investigated at the genus level. As shown in Figure 3A, bacteria in these cecal and fecal samples mainly consisted of 27 genera, including *Coriobacteriaceae* and *S24-7*. However, the detailed composition of them between the cecal and fecal microbiota and between the blank and control groups were different. The far distance between the F-Blank group and F-SfW or F-SfA group in the clustering scheme suggested that HFD-induced dysbiosis of cecal microbiota could be more significantly regulated by SfW and SfA. The PCA plot at the genus level further confirmed the compositional difference between the cecal and fecal samples (Figure 3B). HFD mainly altered the relative abundance of *Coriobacteriaceae*, *S24-7*, *Ruminococcus*, *Clostridiales*, *Oscillospira*, *Ruminococcaceae*, and *Akkermansia* in both the cecal and fecal microbiota but only enriched *Oscillospira* in fecal microbiota (Figure 3C–I). Notably, oral administration of SfW and SfA could partially alleviate the increase of *Clostridiales* and *Ruminococcaceae* in fecal microbiota (Figure 3F,H).

**Figure 3.** Effects of SfW and SfA on the cecal and fecal microbiota in HFD-fed mice. (**A**) Heatmap and (**B**) PCA plots of cecal and fecal microbiota at the genus level. The relative abundance of (**C**) *Coriobacteriaceae*, (**D**) *S24-7*, (**E**) *Ruminococcus*, (**F**) *Clostridiales*, (**G**) *Oscillospira*, (**H**) *Ruminococcaceae*, and (**I**) *Akkermansia* in cecal and fecal microbiota. Values are mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001 vs. the corresponding blank group. # *p* < 0.05 vs. the corresponding control group. Data sets in (**C**–**I**) were analyzed using one-way ANOVA followed by a Turkey's test. C-, cecal microbiota; F-, fecal microbiota.

#### *2.3. Effects of SfW and SfA on the Abundance of Genes Encoding Carbohydrate-Metabolizing Enzymes in Cecal and Fecal Microbiota*

The effects of *S. fusiforme* polysaccharides on the abundance of genes encoding carbohydrate-metabolizing enzymes in the cecal and fecal microbiota of HFD-fed mice were investigated using PICRUSt2 based on the 16S rRNA gene sequencing data. The results demonstrated that the HFD significantly decreased the abundance of genes encoding α-fucosidase (Figure 4A) and β-glucuronidase (Figure 4G), and increased that of monosaccharide-transporting ATPase (Figure 4B) and β-fructofur-anosidase (Figure 4E) in both the cecal and fecal microbiota, but the alterations of genes encoding α-galactosidase (Figure 4D) and β-glucosidase (Figure 4F) were only observed in fecal microbiota, and that of β-mannosidase (Figure 4H) was only presented in cecal microbiota. The administration of SfW and SfA mainly regulated the abundance of genes encoding monosaccharidetransporting ATPase, α-galactosidase, β-fructofuranosidase, and β-glucosidase with the latter showed more significant potency. For example, SfA alleviated the increase of genes encoding monosaccharide-transporting ATPase and β-glucosidase (Figure 4B–F).

**Figure 4.** Effects of SfW and SfA on functional gene of cecal and fecal microbiota in HFD-fed mice. (**A**) α-Fucosidase, (**B**) monosaccharide-transporting ATPase, (**C**) aldose 1-epimerase, (**D**) α-galactosidase, (**E**) β-fructofuranosidase, (**F**) βglucosidase, (**G**) β-glucuronidase, and (**H**) β-mannosidase. Values are mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001 vs. the corresponding blank group. ## *p* < 0.01 vs. the corresponding control group. Data sets were analyzed using one-way ANOVA followed by a Turkey's test. C-, cecal microbiota; F-, fecal microbiota.
