Water-Soluble Saccharina latissima Polysaccharides and Relation of Their Structural Characteristics with In Vitro Immunostimulatory and Hypocholesterolemic Activities

Brown macroalgae are an important source of polysaccharides, mainly fucose-containing sulphated polysaccharides (FCSPs), associated with several biological activities. However, the structural diversity and structure–function relationships for their bioactivities are still undisclosed. Thus, the aim of this work was to characterize the chemical structure of water-soluble Saccharina latissima polysaccharides and evaluate their immunostimulatory and hypocholesterolemic activities, helping to pinpoint a structure–activity relationship. Alginate, laminarans (F1, neutral glucose-rich polysaccharides), and two fractions (F2 and F3) of FCSPs (negatively charged) were studied. Whereas F2 is rich in uronic acids (45 mol%) and fucose (29 mol%), F3 is rich in fucose (59 mol%) and galactose (21 mol%). These two fractions of FCSPs showed immunostimulatory activity on B lymphocytes, which could be associated with the presence of sulphate groups. Only F2 exhibited a significant effect in reductions in in vitro cholesterol’s bioaccessibility attributed to the sequestration of bile salts. Therefore, S. latissima FCSPs were shown to have potential as immunostimulatory and hypocholesterolemic functional ingredients, where their content in uronic acids and sulphation seem to be relevant for the bioactive and healthy properties.


Introduction
Macroalgae (also known as seaweed) represent a sustainable source of natural bioactive compounds with potential for the development of new products and biomaterials to improve human health. These include compounds, such as polysaccharides, proteins, lipids, and pigments, which have great potential for commercial exploitation in the food, nutraceutical, pharmaceutical, and cosmetic industries [1,2]. maintain the integrity of coastal marine ecosystems. Indeed, cultivation of S. latissima has been developed to increase stocks for commercial uses, namely with European funding (e.g., GENIALG project). However, a deeper knowledge on the S. latissima bioactive compounds is still needed to boost their added value and potential applications. Specifically, an understanding of the structures of its polysaccharides (mainly of FCSPs) and their bioactivities is far from being known.
In this context, this work aims to characterize the chemical structure of water-soluble Saccharina latissima polysaccharides and evaluate their immunostimulatory and hypocholesterolemic activities. S. latissima polysaccharides were selected for this work considering previous studies that reported bioactive properties of FCSPs and laminarans extracted from other brown macroalgae, namely of Saccharina genus (S. japonica and S. sculpera) [26][27][28][29]. In addition, unfractionated hot water extract of S. latissima showed immunomodulatory properties towards human THP-1-derived macrophages [30]. Herein, for the first time, the potential of fractionated water-soluble S. latissima polysaccharides for immunostimulatory and hypocholesterolemic purposes was assessed, helping to pinpoint a structureactivity relationship. For that, polysaccharides from cultivated S. latissima were extracted and fractionated using a green solvent-based procedure, including ethanol extraction of non-polysaccharide compounds, hot water extraction, alginate precipitation with calcium chloride, and anion exchange chromatography. In addition to the chemical characterization, polysaccharide-enriched fractions of S. latissima were tested for their in vitro lymphocyte stimulatory activity using BALB/c mice splenocytes. Their effects on cholesterol solubility were also evaluated using a simplified in vitro model composed of glycodeoxycholic acid (GDCA) bile salt.

Fractionation and Characterization of S. latissima Polysaccharides
The hot-water-soluble polysaccharides of S. latissima were extracted from the alcoholinsoluble residue (AIR) obtained with 80% ethanol. Afterwards, alginate was separated via precipitation with CaCl 2 (Ppt_CaCl 2 ) and the polysaccharides present in the supernatant (Sn_CaCl 2 ) were further fractionated by anion-exchange chromatography ( Figure 1). The S. latissima biomass and all the fractions were characterized by total sugars content and monosaccharide composition (Table 1).    S. latissima biomass accounted for 42.0% (w/w) of total sugars. Uronic acids (UA, 37.3 mol%), mannose (Man, 29.9 mol%), and glucose (Glc, 25.6 mol%) were the main sugars, followed by fucose (Fuc, 4.2 mol%), galactose (Gal, 2.1 mol%), and xylose (Xyl, 0.9 mol%). Glc is mainly associated with the presence of glucose-rich polysaccharides (laminarans and cellulose), whereas Fuc and Gal are components of fucose-containing sulphated polysaccharides (FCSPs). UA should mostly derive from alginates but may also be constituents of FCSPs. Most of the Man detected may derive from free mannitol, which occurs naturally in Phaeophyta (including S. latissima) and has a recognized role in osmotic regulation [31]. Therefore, the sugars composition corroborated the presence of different polysaccharides in S. latissima, as reported for this species [10] or brown seaweed in general [11].
Mannitol and mannose were previously identified (by HPAEC-PAD) in dried biomass of S. latissima from Iceland in a proportion of 2:1, together accounting for about 3% [19]. In the same biomass (total sugars of 68%), UA represented about 47% of S. latissima biomass, including mannuronic acid (36%) and guluronic acid (9%), components of alginate, and glucuronic acid (2%) [19]. The differences in monosaccharide composition and content of our results compared with the results from the literature may be related to the geographical origin and harvesting time, as well as to different conditions of macroalgae processing (i.e., washing step) and experimental procedures (i.e., hydrolysis step). Specifically, S. latissima is known to have wide seasonal variation in the content of mannitol (0.5-24%) [10].
In this work, the extraction with ethanol was carried out to eliminate non-polysaccharide components, such as lipids [32], pigments [33], and mannitol [31], from macroalga biomass. The AIR, containing the S. latissima polysaccharides, was recovered with a yield of 56.3% (w/w of macroalga biomass), whereas supernatant (Sn_AIR), containing the nonpolysaccharide components, represented 38.2%. The AIR had a total sugars content of 53.4% and contained UA (59.8 mol%, Glc (26.0 mol%), and Fuc (6.8 mol%) as main sugars. The Sn_AIR comprised 24.9% of total sugars, of which Man (91.2 mol%) was predominant. The Man identified in Sn_AIR may derive from naturally occurring mannitol [31], which was extracted from macroalga biomass with 80% ethanol.
Hot-water extraction from AIR allowed us to obtain a residue (Res_H 2 O) with a yield of 58.7% (w/w of AIR). This residue was composed of 50.8% of sugars, of which UA (64.1 mol%) and Glc (25.2 mol%) were the most abundant. The hot-water extract (Ext_H 2 O; 53.5% of total sugars) revealed a composition of similar amounts of Glc (35.3 mol%) and Mar. Drugs 2023, 21, 183 5 of 15 UA (33.4 mol%), followed by Man (15.1 mol%) and Fuc (11.4 mol%). The higher percentage of Fuc found in Ext_H 2 O than in Res_H 2 O suggests that most FCSPs are soluble in water at 90 • C. This is also corroborated by the amount of Man in Ext_H 2 O, as Man is also described as a constituent of FCSPs. The Glc found in the hot-water-insoluble fraction (Res_H 2 O) was probably derived from cellulose present in S. latissima cell walls, whereas the Glc found in the soluble one (Ext_H 2 O) most likely derived from laminarans [10].

Precipitation with Calcium Chloride
As UA found in Ext_H 2 O may derive in part from alginate, calcium chloride (CaCl 2 ) was added to the Ext_H 2 O to separate this polysaccharide by precipitation. The precipitate recovered after centrifugation (Ppt_CaCl 2 ) had 62.1% of sugars, mostly UA (93.5 mol%), corroborating the presence of alginates. Further, minor amounts of neutral sugars, mainly Fuc (3.8 mol%), were identified in Ppt_CaCl 2 , possibly due to a small proportion of coprecipitated FCSPs. Glc (47.7 mol%) and Fuc (23.7 mol%) were the main sugars found in the supernatant (Sn_CaCl 2 ), probably due to the presence of laminarans and FCSPs, respectively. Total sugars (neutral and UA) accounted for 55.9% of the Sn_CaCl 2 fraction, whereas sulphate esters (as -SO 3 − ) represented 6.4% and proteins represented 11.2% ( Figure 2 and Supplementary Table S1).   Table  Supplementary Table S1 (for sulphates and proteins).

Anion-Exchange Chromatography
To fractionate the polysaccharides present in Sn_CaCl2, this fraction was subjec anion-exchange chromatography, recovering three fractions with increasing strength: F1 (eluted with 0.05 M HCl), F2 (eluted with 1 M NaCl), and F3 (eluted wit NaCl) (Figure 3). Fraction F1, eluted without ionic strength, had a total content of sugars of 7 mainly composed of Glc (97.7 mol%) ( Table 1). No sulphate was detected in this fr and protein was found in low amounts (2.6%) ( Figure 2). Considering the n character and sugars composition, it can be confirmed that F1 is a laminaran-enr fraction. The low amount of Man (2.3 mol%) found in F1 may derive from ma located at the reducing end of laminarans (classified as M chains) [10].
Fraction F2, eluted with the lowest ionic strength, was composed of 42.7% of s UA (44.7 mol%) and Fuc (28.7 mol%) were the main sugars, together with minor am of Gal (8.5 mol%), Glc (6.8 mol%), Xyl (5.7 mol%), and Man (5.6 mol%) ( Table 1). Su Fraction F1, eluted without ionic strength, had a total content of sugars of 76.2%, mainly composed of Glc (97.7 mol%) ( Table 1). No sulphate was detected in this fraction and protein was found in low amounts (2.6%) ( Figure 2). Considering the neutral character and sugars composition, it can be confirmed that F1 is a laminaran-enriched fraction. The low amount of Man (2.3 mol%) found in F1 may derive from mannitol located at the reducing end of laminarans (classified as M chains) [10].
The fraction F3, eluted with the strongest ionic strength, showed a total content of sugars of 42.6%. F3 was mainly composed of Fuc (59.1 mol%), Gal (20.8 mol%), and UA (12.0 mol%), containing also Glc (3.2 mol%), Xyl (3.0 mol%), and Man (2.0 mol%) ( Table 1). Sulphates accounted for 14.3% of F3 and a low content of protein was detected in this fraction (6.2%) ( Figure 2). F2 and F3 are not pure/homogeneous fractions of FCSPs, which is in accordance with previous studies on FCSPs of S. latissima that also used anion-exchange chromatography for fractionation [13,19]. F2 is possibly a complex mixture, containing mainly (β1→3)glucuronan chains [13]. F3 fraction probably contains mainly sulphated fucans, among other polysaccharide structures [13]. As described in the literature, FCSPs are complex heteropolysaccharides composed of several types of monosaccharides that, due to the presence of uronic acid residues and sulphate groups, are highly charged. Regarding the protein content found in F2 and F3, other studies also reported the presence of proteins in FCSP-enriched fractions of brown macroalgae (including S. latissima), suggesting that proteins are tightly associated with certain FCSPs [34,35].

In Vitro Lymphocyte Stimulatory Activity
The selected fractions Sn_CaCl 2 , Ppt_CaCl 2 , F1, F2, and F3, containing different proportions of water-soluble polysaccharides from S. latissima (alginate, laminarans, and FCSPs), were incubated with murine splenocytes to evaluate in vitro lymphocyte stimulatory activity. For all the fractions and concentrations tested (25, 100, and 250 µg/mL), the cell viability was not significantly decreased when compared to the negative control with  Figure S1).
The percentage of T cells activated in the presence of polysaccharide-enriched fractions from S. latissima ranged from 1.3 to 4.2% for all tested concentrations, inferred by the expression of the early activation marker CD69 on the surface of CD3 + cells (Supplementary Figure S2). Only the Sn_CaCl 2 fraction (mixture of FCSPs and laminarans) at 250 µg/mL induced significant T-cell activation when compared to the negative control (4.2% vs. 1.4%) but negligible when compared to concanavalin A (ConA) used as a positive control of T-cell activation (84.4%) (Supplementary Figure S2).
On the other hand, CD19 + cells (B cells) were significantly stimulated by all fractions, except for F1 (laminaran-enriched fraction). The B-cell activation occurred in a dosedependent manner. The percentage of B cells expressing CD69 (8.0% in negative control) increased upon incubation with the different polysaccharide concentrations used, ranging from 25 to 250 µg/mL, as follows: 58.1% to 84.4% for Sn_CaCl 2 (containing FCSPs and laminarans); 17.3% to 39.9% for Ppt_CaCl 2 (mainly alginates); 61.8% to 76.9% for F2 (moderate charged FCSPs); and 59.7% to 72.4% for F3 (high charged FCSPs) ( Figure 4).  In addition, the percentage of B cells activated by incubation with S. latissima fractions was evaluated in the presence of polymyxin B (PB) to assess possible contamination of the samples with bacterial endotoxin (LPS). PB is known to bind LPS and inhibit LPS-induced B-cell activation [36,37]. Indeed, the treatment of cells with PB reduced B-cell activation induced by LPS (used as positive control) from 95.0% to 14.6%. PB also reduced the extent of B-cell activation in cultures stimulated with S. latissima fractions, indicating that LPS contamination cannot be completely excluded. However, even in the presence of PB and for the three concentrations tested, the percentage of B cells activated by incubation with fractions containing FCSPs (Sn_CaCl2, F2, and F3) remained significantly higher than the negative control (7.3%). This effect was also dose-dependent, increasing from 25 to 250 µg/mL of polysaccharides, as follows: 15.5% to 30.6% (Sn_CaCl2), 17.7% to 32.6% (F2), and 14.0% to 36.7% (F3) (Figure 4). Moreover, the activation percentages found for cells stimulated with 100 and 250 µg/mL of Sn_CaCl2, F2, and F3 in the presence of PB were significantly higher than those observed for LPS stimulation with PB.
Considering that F1 (only containing laminarans) did not exhibit immunostimulatory activity on B lymphocytes, the activity of Sn_CaCl2, F2, and F3 can be directly associated with the presence of FCSPs. These results are in line with a previous study comparing fucoidans and laminarans extracted from other Laminariaceae species (Laminaria japonica), where fucoidans exhibited a stronger immune activation ability [29]. The co-precipitation of FCSPs in Ppt_CaCl2 (enriched in alginates) may contribute to the immunostimulatory activity on B lymphocytes observed in this fraction. However, studies have shown that alginates (including oligomer derivatives) themselves have In addition, the percentage of B cells activated by incubation with S. latissima fractions was evaluated in the presence of polymyxin B (PB) to assess possible contamination of the samples with bacterial endotoxin (LPS). PB is known to bind LPS and inhibit LPS-induced B-cell activation [36,37]. Indeed, the treatment of cells with PB reduced B-cell activation induced by LPS (used as positive control) from 95.0% to 14.6%. PB also reduced the extent of B-cell activation in cultures stimulated with S. latissima fractions, indicating that LPS contamination cannot be completely excluded. However, even in the presence of PB and for the three concentrations tested, the percentage of B cells activated by incubation with fractions containing FCSPs (Sn_CaCl 2 , F2, and F3) remained significantly higher than the negative control (7.3%). This effect was also dose-dependent, increasing from 25 to 250 µg/mL of polysaccharides, as follows: 15.5% to 30.6% (Sn_CaCl 2 ), 17.7% to 32.6% (F2), and 14.0% to 36.7% (F3) (Figure 4). Moreover, the activation percentages found for cells stimulated with 100 and 250 µg/mL of Sn_CaCl 2 , F2, and F3 in the presence of PB were significantly higher than those observed for LPS stimulation with PB.
Considering that F1 (only containing laminarans) did not exhibit immunostimulatory activity on B lymphocytes, the activity of Sn_CaCl 2 , F2, and F3 can be directly associated with the presence of FCSPs. These results are in line with a previous study comparing fucoidans and laminarans extracted from other Laminariaceae species (Laminaria japonica), where fucoidans exhibited a stronger immune activation ability [29]. The co-precipitation of FCSPs in Ppt_CaCl 2 (enriched in alginates) may contribute to the immunostimulatory activity on B lymphocytes observed in this fraction. However, studies have shown that alginates (including oligomer derivatives) themselves have immunostimulating effects, namely by inducing B cells to express CD69 [38]. Even so, the B-cell activation induced with Ppt_CaCl 2 was significantly lower than that observed for Sn_CaCl 2 , F2, or F3 fractions, suggesting that FCSPs have a higher activation capacity than alginates. The presence of sulphate esters in the polysaccharides, as occurs in FCSPs, has been reported to be relevant for immunostimulatory activity [39,40]. On the other hand, co-extracted laminarans in Sn_CaCl 2 had no negative impact on B-cell activation promoted by FCSPs. The possibility of eliminating the fractionation step of the Sn_CaCl 2 may be advantageous considering potential applications, namely in the formulation of functional foods or nutraceuticals.
Regarding the importance of lymphocyte activation, it is worth noting that B cells are found along the intestinal tract in Peyer's patches. This is a possible route for direct B-cell activation by FCSPs, as hypothesized for other polysaccharides [41]. Indeed, intestinal immunomodulating activity via Peyer's patch cells was reported for fucoidans extracted from two brown seaweed species (Sargassum crassifolium and Padina australis) [42]. Another hypothesis is indirect B-cell activation promoted by cytokines produced by polysaccharidestimulated enterocytes or phagocytes [41].

In Vitro Hypocholesterolemic Effect
The effect of polysaccharide-enriched fractions obtained from S. latissima (Sn_CaCl 2 , F1, F2, and F3) in the sequestration of GDCA bile salt and in reductions in cholesterol solubilized in GDCA micelles was evaluated by quantitative NMR and compared with cationic resin colestipol used as a positive control (Supplementary Figure S3).
The F2 fraction, containing FCSPs eluted with a lower ionic strength in the anionexchange chromatography, significantly decreased the amount of GDCA in solution when compared to the negative control (GDCA with cholesterol), whereas no significant differences were observed for Sn_CaCl 2 (containing neutral laminarans and charged FCSPs), F1 (laminarans), and F3 (containing highly charged FCSPs) (Figure 5a). In addition, cholesterol solubility decreased significantly in the presence of fraction F2 when compared to the negative control (Figure 5b), being coincident with a decrease in bile salt concentration. Indeed, the quantity of solubilized cholesterol in the presence of F2 was proportional to the amount of GDCA in solution (Figure 5a,b). A chemical shift to lower ppm was also noticed in bile salt resonances, being evidence of the interaction between bile salt and the polysaccharide (Supplementary Figure S3). This corroborates that the mechanism behind reductions in cholesterol solubilization is the sequestration of bile salts by moderately charged FCSPs. Considering its composition, the F2 fraction is rich in uronic acids but has a lower content of fucose and sulphate groups in comparison with F3. Therefore, uronic acids (contrarily to sulphate groups) may have an important role on GDCA sequestration. Other types of polysaccharides primarily composed of uronic acids have been reported to interact with bile acids [43,44], namely pectins, with a degree of methylesterification of 62% [43]. Considering that GDCA is negatively charged at the intestinal lumen pH, the interactions behind the negatively charged polysaccharides and bile salts should be mostly due to hydrophobic interactions, possible involving the fucose (a deoxy sugar) present in FCSPs.
GDCA sequestration. Other types of polysaccharides primarily composed of uronic acids have been reported to interact with bile acids [43,44], namely pectins, with a degree of methylesterification of 62% [43]. Considering that GDCA is negatively charged at the intestinal lumen pH, the interactions behind the negatively charged polysaccharides and bile salts should be mostly due to hydrophobic interactions, possible involving the fucose (a deoxy sugar) present in FCSPs. In the present work, laminarans (F1) from S. latissima showed no effect, either in the sequestration of GDCA or cholesterol solubility. Although other β-glucans from cereals (barley and oat) and mushroom have been known to sequestrate bile salts and decrease cholesterol solubility, their structure is different from brown macroalgae laminarans Cereal β-glucans have (β1→3)-and (β1→4)-Glc linkages. Both laminarans and mushroom β-glucans are composed of (β1→3) and (β1→6)-Glc, but laminarans have a reported lower In the present work, laminarans (F1) from S. latissima showed no effect, either in the sequestration of GDCA or cholesterol solubility. Although other β-glucans from cereals (barley and oat) and mushroom have been known to sequestrate bile salts and decrease cholesterol solubility, their structure is different from brown macroalgae laminarans. Cereal β-glucans have (β1→3)-and (β1→4)-Glc linkages. Both laminarans and mushroom βglucans are composed of (β1→3) and (β1→6)-Glc, but laminarans have a reported lower average molecular weight [44]. This structural difference could explain the absence of a hypocholesterolemic effect of laminarans.
Like colestipol, a cationic resin used as a cholesterol-lowering drug, FCSPs from the F2 fraction seem to have the capacity (albeit to a lesser effect) to sequester bile salts, reducing their concentration in the intestinal lumen, lowering the cholesterol solubility, and limiting cholesterol absorption through the intestine. Considering other brown macroalgae, the fraction of Sargassum zhangii containing moderately charged FCSPs (eluted with 1 M NaCl) showed the best ability to bind bile acids and reduce the content of intracellular total cholesterol in HepG2 cells compared to those eluted with 0.5 M or 2 M NaCl [45]. These data also corroborate the higher potential of moderately charged FSCPs as hypocholesterolemic agents able to decrease cholesterol absorption, which is especially relevant considering the prevalence of the high cholesterol levels in human blood, a well-established risk factor for cardiovascular diseases [46].

Extraction and Fractionation of S. latissima Polysaccharides
The macroalga S. latissima was cultivated by the Scottish Association for Marine Science (Oban, UK) at the Port a' Bhuiltin seaweed farm (56.4886 • N, −5.4698 • E) and harvested in May 2017 [32]. Freeze-dried and milled biomass was used for extraction and fractionation of S. latissima polysaccharides, as represented in Figure 1.

Alcohol-Insoluble Residue (AIR) Preparation
S. latissima biomass (20 g) was suspended in distilled water (160 mL) and kept under stirring for 10 min at room temperature. Then, absolute ethanol (640 mL) was added, and the mixture was left for 15 min at 80 • C. The solution was filtered with a funnel with porous plate and a 110 nm filter under vacuum. The alcohol-insoluble residue (AIR) was washed with ethanol and acetone and left in the hood overnight to dry. The filtrate (Sn_AIR) was also recovered, concentrated on rotary evaporator at 40 • C to remove ethanol, and freeze-dried.

Hot Water Extraction
To extract hot-water-soluble polysaccharides, AIR (5 g) was suspended in distilled water (560 mL) and then kept at 90 • C for 1 h under stirring. An extract (Ext_H 2 O) and a residue (Res_H 2 O) were recovered by filtration under vacuum. Res_H 2 O and an aliquot of the Ext_H 2 O (1 mL) were dialyzed (cut-off 12 kDa) against distilled water and freeze-dried for further analysis.

Precipitation with Calcium Chloride
For alginate precipitation, CaCl 2 2% (3.8 g) was added to the Ext_H 2 O recovered by filtration (with a volume of 380 mL) and stirred for 25 min at room temperature. The solution was kept for 2 h at 4 • C and then centrifuged at 15,000 rpm for 20 min at 4 • C. The resulting precipitate (Ppt_CaCl 2 ) and supernatant (Sn_CaCl 2 ) were dialyzed and freeze-dried.  Each fraction (F1, F2 and F3) was dialyzed and freeze-dried. A solution of 4 M NaCl was used for column cleaning. The elution profile was obtained by recovery of the fractions (1.9 mL), which were assessed by the colorimetric phenol-sulfuric acid method [47].

Neutral Sugars Analysis and Uronic Acid Determination
Polysaccharide fractions (2 mg) were subjected to pre-hydrolysis for 3 h at room temperature with 72% H 2 SO 4 (w/w), followed by hydrolysis at 100 • C with 1 M H 2 SO 4 for 2.5 h. Neutral sugars were determined by converting the hydrolyzed sugars in alditol acetates, as described previously [48], using 2-desoxyglucose as an internal standard. Alditol acetates were analyzed on a GC-FID (Perkin-Elmer Clarus 400, Waltham, MA, USA) equipped with DB-225 column (30 m of length, 0.25 mm of internal diameter and 0.15 µm of film thickness; Agilent J&W GC columns, Santa Clara, CA, USA), operating with injector temperature of 220 • C, detector temperature of 230 • C, and hydrogen flow rate of 1.7 mL/min. The oven was programmed as follows: 200 • C (held for 1 min), 40 • C/min to 220 • C (held for 7 min), and 20 • C/min to 230 • C (held for 1 min).
For uronic acid (UA) determination, the polysaccharide fractions were hydrolyzed for 1 h with 1 M H 2 SO 4 at 100 • C. Uronic acids were estimated by colorimetry using phenolsulfuric acid method using D-galacturonic acid as standard, as previously detailed [48].

Sulphate and Protein Content
Sulphur (S) and nitrogen (N) contents of the fractions Sn_CaCl 2 , F1, F2, and F3 (2 mg per replicate) were determined by elemental analysis (Leco Truspec-Micro CHNS 630-200-200 elemental analyser) [48]. Considering the determined sulphur content, the content of sulphate groups (calculated as -SO 3 − ) was estimated. The protein content was obtained using nitrogen-protein conversion factor of 4.37, a specific factor determined for S. latissima [49].

In Vitro Lymphocyte Stimulatory Activity
BALB/c mice were purchased from Charles River (Barcelona, Spain) and kept at i3S animal facilities. Spleen cell suspensions were obtained from naïve mice included in the project licensed by the competent national authority (Direção Geral de Alimentação e Veterinária, Lisbon, Portugal) with the reference number 001879/2021-01-06. The preparation of spleen lymphocytes for stimulation assays and analysis by flow cytometry were conducted as previously described [37,50]. Spleens were aseptically removed and splenocyte suspensions were obtained by mechanically disrupting the organ in Hanks' balanced salt solution (HBSS, Sigma, St. Louis, MO, USA) and filtering through 100 µm cell strainers. Splenocytes were resuspended in ammonium-chloride-potassium (ACK) lysing buffer for 3 min, to lyse erythrocytes, washed with HBSS, and resuspended in RPMI-1640 medium

In Vitro Assessment of Hypocholesterolemic Effect
To assess the potential of S. latissima polysaccharides in reductions in cholesterol's solubility, mixtures of selected fractions (Sn_CaCl 2 , F1, F2, and F3) were tested using an in vitro intestinal model composed of a 50 mM bile salt sodium glycodeoxycholate (GDCA) and 3.5 mM [4-13 C]Cholesterol, as described previously [51,52]. The solutions were prepared in aqueous buffer, containing 10 mM Tris-HCl (pH 7.4), 0.15 M NaCl, 1 mM EDTA, and 0.02% sodium azide (NaN 3 ) in deuterated water (D 2 O) containing 3-(Trimethylsilyl)propionic-2,2,3,3-d4 acid sodium salt (TSP). Cationic resin colestipol was used as positive control at a similar concentration as the S. latissima extracts (5 mg/mL). Before measurements, the mixtures were left under stirring at 100 rpm and 37 • C for 24 h. 13 C NMR spectra were acquired at 37 • C using a 90 • pulse, with a 25,252 Hz spectral width, acquisition time of 1.3 s, relaxation delay of 5 s, and 2040 acquisition scans. Proton decoupling was accomplished by using a WALTZ-16 decoupling sequence. Nuclear Overhauser Enhancement (NOE) was obtained through the comparation between 13 C spectra with full proton decoupling and with proton decoupling only during acquisition [52]. 1 H NMR spectra were acquired with a 90 • pulse, a 7500 Hz spectral width, acquisition time of 1 s, relaxation delay of 5 s, and 128 acquisition scans. These experiments were acquired in 500 MHz NMR Bruker spectrometer and spectra were treated using MestreNova 6.1.1 (Mestrelab Research, Santiago de Compostela, Spain). With this methodology, due to their size, polysaccharides with GDCA or cholesterol aggregates and cholesterol precipitated as crystals are not observed by liquid state NMR, because the motion of these aggregates leads to a line-width broadening resulting from spin-spin relaxation (T2) phenomena. Therefore, bile salt sequestration and cholesterol emulsified in GDCA micelles (small aggregates) were determined by quantitative 13 C NMR using the area of TSP (10 mM) resonating at 0 ppm, as internal standard for quantification. The area of carbon 4 from 13 C enriched cholesterol, which resonates at 44.4 ppm, was used for its quantification, normalized by 13 C enrichment factor (1.109/99.8). Regarding bile salt GDCA, several resonances CH 3 , CH 2 , CH, and C areas were used for quantification by 13 C NMR (assignments are shown in Supplementary Table S2). In Supplementary Figure S3, the resonance CH position 5 of GDCA (45.0 ppm) is highlighted in the insert. All the areas obtained in 13 C NMR spectrum were corrected for the NOE effect, which was used to decrease the length of the 13 C NMR experiment per assay, without compromising the quantitative outcome of the experiment. Crosscheck of the results obtained by 13 C NMR with NOE correction factors was conducted by comparison with the results obtained by quantitative 1 H NMR experiments in the case of GDCA. The same approach was not addressed for cholesterol because no distinctive resonance from cholesterol was assigned by 1 H NMR.

Statistical Analysis
Statistical analysis was performed using Minitab 17. Data from in vitro assays were analyzed by one-way ANOVA, followed by Tukey's test. The confidence level was set at 95% (α = 0.05).

Conclusions
The consumption of macroalgae has long been related with health benefits, which can be related to the presence of structurally different polysaccharides. In this work, two fractions of fucose-containing sulphated polysaccharides (FCSPs) of Saccharina latissima with distinct charge and composition revealed similar immunomodulatory activity on B lymphocytes. The presence of sulphate esters in the structure of FCSPs seems to be a key feature for its immunomodulatory activity. However, for the bile salt sequestration, the content in uronic acids in FCSPs seems to be more relevant. Only moderately charged FCSPs, which have a higher content in uronic acids but lower fucose and sulphate content, showed hypocholesterolemic potential, exhibiting the capacity to sequester bile salts and reduce cholesterol solubility. On other hand, laminarans of S. latissima showed no immunostimulatory or hypocholesterolemic effect. Overall, fucose-containing sulphated polysaccharides of brown macroalga Saccharina latissima showed immunostimulatory and hypocholesterolemic activities, having potential to be used as functional ingredients.