To prepare test samples, 500 mL of Milli-Q water and 20 mL of a 10 g L⁻¹ sodium alginate (Na⁺-Alg) solution containing 200 mg of Na⁺-Alg were added to 300 g of each sediment sample (n=3) in a beaker, followed by stirring for 5 min. These samples were labeled “sediment + Na⁺-Alg 200 mg”. The resulting suspensions were stored overnight (12 h) at 4 °C to allow the added Na⁺-Alg to transform into naturally occurring alginate forms, such as magnesium alginate (Mg²⁺-Alg), Ca²⁺-Alg, and Fe³⁺-Alg gel. Although previous studies have not clearly identified the chemical forms of alginate in marine sediments, we hypothesized that ion exchange would occur between Na⁺-Alg and Mg²⁺, Ca²⁺ or Fe³⁺ ions, all of which are abundant in seawater and the pore water of ocean sediments (Masuzawa and Kitano, 1983; Klinkhammer, 1980). The rationale for the selective interaction of Fe³⁺ – but not Fe²⁺ – with alginate to form a gel is discussed in Sect. 3.2.1. After this treatment, alginate in the samples was presumed to exist as a mixture of gelled forms (Ca²⁺-Alg and Fe³⁺-Alg) and soluble salts (Na⁺-Alg and Mg²⁺-Alg).

To remove impurities such as soluble sugars, proteins, and pectin from the sediment sample, enzymatic treatment was employed. A total of 0.80 g of each of the following four enzymes was sequentially added to the sediment suspension: amylase (KOKULASE™, Mitsubishi Chemical, Tokyo, Japan); cellulase (SCLASE™C, Mitsubishi Chemical); protease (KOKULASE™P Granule, Mitsubishi Chemical); and pectinase (SCLASE™N, Mitsubishi Chemical). Each enzymatic reaction was performed for 30 min under optimized conditions according to the manufacturer's specifications: amylase at 27 °C and pH 6.0; cellulase at 50 °C and pH 6.0; protease at 50 °C and pH 5.5; and pectinase at 38 °C and pH 4.2. A hot stirrer was used to maintain the reaction temperature during stirring. The pH of each reaction suspension was adjusted as necessary by adding 1.0 mol L⁻¹ hydrochloric acid (HCl) and 3.0 mol L⁻¹ NaOH solution. Following this enzymatic treatment, alginate in the suspension sample was presumed to exist as a mixture of gelled forms (Ca²⁺-Alg and Fe³⁺-Alg) and soluble salts (Na⁺-Alg and Mg²⁺-Alg). Pectin in the sample was presumed to have been enzymatically degraded into monosaccharides, primarily galacturonic acid.

To convert soluble alginate (Na⁺-Alg and Mg²⁺-Alg) into Ca²⁺-Alg gel in the sediment suspension, 100 mL of 100 g L⁻¹ calcium chloride (CaCl₂, Wako, purity ≥ 95 %) solution was added to the sample and stirred for 5 min. Subsequently, approximately 15 mL of 5.0 mol L⁻¹ HCl was added to adjust the pH to 1.0, and the suspension was kept overnight at room temperature (25 °C). After the reaction, the resulting approximately 900 mL suspension was divided into eight 45 mL centrifuge tubes and centrifuged (3500 rpm, 10 min, 2380 × g). The supernatant from each tube was carefully removed, and 40 mL of 0.10 mol L⁻¹ HCl was added to each tube. The tubes were manually shaken to ensure thorough mixing, followed by centrifugation to remove the acidified supernatant. This acid treatment and the subsequent removal of the supernatant served to eliminate polyvalent cations (e.g., Ca²⁺ and Mg²⁺), acid-soluble phenolic compounds, and easily degradable polysaccharides such as fucoidans from the sediment sample (Bertagnolli et al., 2014). During this process, a portion of the Ca²⁺-Alg was presumed to have undergone ion exchange with H⁺, forming insoluble H⁺-Alg (McHugh et al., 2001). Following this step, alginate in the sediment sample was presumed to exist as a mixture of gelled forms (Ca²⁺-Alg and Fe³⁺-Alg) and protonated alginate (H⁺-Alg) in a gelatinous form. Pectin was presumed to have been removed with the supernatant, because depolymerized pectin is generally considered unable to form stable gel networks through Ca²⁺ crosslinking.

Most of the sediment containing alginate gels in each tube was transferred to a single beaker using a spoon. To recover any remaining precipitate in tubes, all eight tubes were rinsed with 500 mL of 10 g L⁻¹ sodium bicarbonate (NaHCO₃, Wako, purity ≥ 99.5 %) solution, and the rinse solution was collected into a beaker. The resulting suspension in the beaker was heated to 60 °C with continuous stirring for 1.5 h, followed by reaction overnight at room temperature to convert the alginate gel to Na⁺-Alg. Following this step, alginate in the suspension sample was presumed to exist primarily as soluble Na⁺-Alg; however, if Fe³⁺-Alg was not soluble under the alkaline conditions (pH 8.0) of the NaHCO₃ solution, it may have remained in a gel form, as described in Sect. 3.2.1.

The pH of the suspension was adjusted to 5.0 by adding an appropriate volume of 1.0 mol L⁻¹ HCl. After centrifugation, the supernatant was collected into a beaker. Subsequently, 100 mL of 300 g L⁻¹ CaCl₂ solution was added to the supernatant, and the mixture was kept overnight at room temperature to convert soluble Na⁺-Alg to Ca²⁺-Alg gel. Following centrifugation, the supernatant was removed, and the resulting precipitate was retained in the tubes. Then, an excess amount of Ca²⁺ was added to completely react with the CO32- ions derived from the previously added NaHCO₃ solution, thereby ensuring efficient ion exchange between Na⁺-Alg and Ca²⁺. After this step, alginate in the precipitate sample was presumed to exist as Ca²⁺-Alg (gel form).

Twenty milliliters of 0.10 mol L⁻¹ HCl was added to each centrifuge tube containing the precipitates. The tubes were manually shaken to ensure thorough mixing, followed by centrifugation to remove the acidified supernatant. This acid-washing treatment was repeated until oligosaccharides were no longer detected in the supernatant. To verify the presence of oligosaccharides, the phenol-sulfuric acid colorimetric method was employed: 1.0 mL of the supernatant was transferred to a centrifuge tube, to which 1.0 mL of 50 g L⁻¹ phenol (Wako, purity ≥ 99 %) solution was added and mixed. Subsequently, 5.0 mL of H₂SO₄ was quickly added and mixed thoroughly. After 10 min, the appearance of a yellow color in the solution, indicating the presence of oligosaccharides, was monitored. If color was observed, an additional round of acid-washing treatment was performed. When the solution remained colorless, confirming the absence of detectable oligosaccharides, the acid-washing treatment was terminated. The precipitates from each centrifuge tube were then collected into a single beaker using a spoon. After this HCl treatment, alginate in the precipitate sample was presumed to exist as a mixture of Ca²⁺-Alg (gel form) and H⁺-Alg (gelatinous form).

One hundred milliliters of 10 g L⁻¹ NaHCO₃ solution was added to a beaker containing the precipitates. The suspension was then heated to 60 °C with continuous stirring for 1.5 h, followed by reaction overnight at room temperature to convert Ca²⁺-Alg to Na⁺-Alg. After centrifugation, the supernatant was collected into a clean beaker. Following this step, alginate in the solution sample was expected to exist as soluble Na⁺-Alg. This extract solution appeared dark brown, likely due to the presence of residual phenolic compounds (McHugh, 1987).

To remove non-targeted compounds soluble in organic solvents, such as pigmentation, fats, and phenolic compounds, 400 mL of ethanol (Wako, purity: 99.5 %) was added to the beaker containing the extract solution (Saji et al., 2022; Trica et al., 2019). The ethanol–water mixture was kept overnight at room temperature to precipitate alginate, because alginate is insoluble in ethanol (Gomez et al., 2009). The resulting precipitate was collected by filtration using a polycarbonate membrane filter (Nuclepore, Whatman, Maidstone, UK; pore size: 10 µm, diameter: 47 mm). The filter with the retained precipitate was transferred to a beaker and air-dried overnight at room temperature. Subsequently, 100 mL of 10 g L⁻¹ NaHCO₃ solution was added to the beaker containing the dried filter, and the mixture was stirred for 1.0 h. The solution was then kept overnight at room temperature to ensure complete dissolution of Na⁺-Alg. After centrifugation, the supernatant was recovered; however, it exhibited a slight brown coloration. To reduce the coloration and ensure compatibility with colorimetric analysis, the supernatant was diluted 20-fold with Milli-Q water and analyzed colorimetrically as described in Sect. 2.2. However, if trace amounts of Na⁺-Alg are soluble in ethanol, significant loss may occur during the extraction of trace levels of Na⁺-Alg from ocean sediment, as described in Sect. 3.2.2.

Bleaching treatment has been reported to effectively remove pigments from alginate extracts derived from dark brown algae, which are presumed to be rich in polyphenols, although direct quantification was not performed in that study (Andriamanantoanina and Rinaudo, 2010). Therefore, in the present study, bleaching treatment was applied to reduce the polyphenol content in the extract solution.

Three hundred and seventy-five milliliters of 24 mL L⁻¹ sodium hypochlorite (NaClO, Wako) solution was added to 75 mL of extract solution in a beaker and stirred for 20 min. Subsequently, 188 mL of Milli-Q water was added, followed by neutralization with 0.05 mol L⁻¹ HCl. Then, 20 mL of 300 g L⁻¹ CaCl₂ was added, and the mixture was kept overnight at room temperature to convert soluble Na⁺-Alg to Ca²⁺-Alg gel. The following day, the precipitate was collected by centrifugation and transferred to a clean beaker. Forty milliliters of 10 g L⁻¹ NaHCO₃ solution was added to the precipitate, and the suspension was heated to 60 °C with continuous stirring for 1.5 h, followed by reaction overnight at room temperature to convert Ca²⁺-Alg to Na⁺-Alg. Finally, the supernatant was collected by centrifugation. To reduce the coloration of the extract solution as much as possible, this bleaching process was repeated six times with appropriate adjustments to the reagent quantities (Table 1). After the final bleaching step, the volume of the sample was reduced to 15 mL. The same procedure was also applied to 75 mL of 10 µg mL⁻¹ Na⁺-Alg solution.

To prepare the buffer solution for adjusting pH and salinity, 1.0 mL of 1.0 mol L⁻¹ Tris-HCl (pH 7.5, NIPPON GENE), 1.0 mL of 1.0 mol L⁻¹ Tris-HCl (pH 8.0, NIPPON GENE), 3.0 mL of 5.0 mol L⁻¹ sodium chloride (NaCl, Wako, purity ≥ 99.5 %) solution, and 5.0 mL of Milli-Q water were combined in a tube and thoroughly mixed. For the enzymatic reaction, 5.0 mL of Milli-Q water, 0.80 mL of the prepared buffer solution, 2.0 mL of the bleached extract solution, and 0.40 mL of alginate lyase (HULK) were sequentially added to a centrifuge tube with gentle mixing after each addition. This enzymatic degradation experiment was performed in triplicate using the same sample (n=3). In parallel, a control experiment was performed without the addition of alginate lyase. In this case, 5.0 mL of Milli-Q water, 0.80 mL of the prepared buffer solution, 2.0 mL of the bleached extract solution, and 0.40 mL of Milli-Q water instead of alginate lyase were sequentially added to a centrifuge tube with gentle mixing after each addition. The control experiment was also conducted in triplicate (n=3).

The same enzymatic experiment was performed on the bleached Na⁺-Alg standard solution (n=1). In addition, a blank sample (n=1) was prepared as a control using water in place of the extract solution: specifically, 5.0 mL of Milli-Q water, 0.80 mL of the buffer solution, 2.0 mL of Milli-Q water, and 0.40 mL of alginate lyase (HULK) were sequentially added to a centrifuge tube with gentle mixing. The mixtures were centrifuged (3500 rpm, 1 min, 2380 × g) to remove residual water from the inner wall surface and subsequently incubated in an incubator (SLI-220, EYELA, Tokyo, Japan) at 30 °C for one week. After incubation, the pH of each mixture was adjusted to 4.0 using 1.0 mol L⁻¹ HCl. Subsequently, 10 mL of 300 g L⁻¹ CaCl₂ was added, and the mixtures were kept overnight at room temperature to convert Na⁺-Alg remaining in the mixture to Ca²⁺-Alg.

On the following day, the resulting precipitates were collected by filtration using polycarbonate membrane filters (Nuclepore, Whatman; pore size 0.40 µm, diameter 47 mm). At this stage, the oligomerized alginate produced by alginate lyase was expected to remain in the filtrate, because it does not form crosslinks in the presence of Ca²⁺. Each filter containing the precipitate was transferred into a separate tube, to which 5.0 mL of 20 g L⁻¹ NaHCO₃ solution was added. The tubes were then incubated in a 60 °C water bath for 1.5 h with gentle shaking every 10 min, followed by reaction overnight at room temperature to convert Ca²⁺-Alg to Na⁺-Alg. The supernatants were collected by centrifugation (3500 rpm, 10 min, 2380 × g), and analyzed colorimetrically as described in Sect. 2.2.

If Fe³⁺-Alg, formed via ion exchange between alginate and Fe³⁺, is present in a sediment sample, the gel may not dissolve in the NaHCO₃ solution and could remain in the sediment even after alkaline treatment (Step 3), leading to reduced recovery. To address this issue, we investigated the optimal conditions for the simultaneous dissolution of both Fe³⁺-Alg and Ca²⁺-Alg gels, which are presumed to be the predominant forms of alginate present prior to the alkaline treatment. Disodium dihydrogen ethylenediamine tetraacetic acid dihydrate (EDTA-2Na) is a potent chelator that forms more stable complexes with Fe³⁺ than the binding affinity of Fe³⁺ to alginate (Berner and Hood, 1983). Because the stability of EDTA-metal complexes depends on pH (Ueno, 1959), we investigated the dissolution behavior of Fe³⁺-Alg and Ca²⁺-Alg in EDTA-2Na solutions adjusted to pH 4.0 and 5.0, and compared the results with those obtained using NaHCO₃ solution at pH 8.0.

Fe³⁺-Alg gel was prepared by dropping 20 mL of 10 g L⁻¹ Na⁺-Alg solution into 20 mL of 100 g L⁻¹ iron (III) chloride hexahydrate (FeCl₃, Wako, purity ≥ 99 %) solution in a beaker. To ensure complete reaction between Na⁺-Alg and Fe³⁺ ions, the resulting Fe³⁺-Alg / Na⁺-Alg suspension was repeatedly filtered through a 100 µm stainless steel mesh using a silicone spatula. The reddish Fe³⁺-Alg gel retained on the mesh was collected and rinsed three times with Milli-Q water. For comparison, Ca²⁺-Alg gel was similarly prepared by dropping 20 mL of 10 g L⁻¹ Na⁺-Alg solution into 20 mL of 100 g L⁻¹ CaCl₂ solution. Each gel type (Fe³⁺-Alg and Ca²⁺-Alg) was divided into two portions. One half of each was added to 500 mL of 10 g L⁻¹ NaHCO₃ solution and heated to 60 °C with continuous stirring for 1.5 h. The other halves were added to 500 mL of 80 g L⁻¹ EDTA-2Na (Kanto Chemical, Tokyo, Japan, purity > 99.5 %) solution, which had an initial pH of 4.0, and stirred for 1.5 h. These suspensions were then kept overnight at room temperature. The presence or absence of remaining gel was assessed visually. Subsequently, the pH of the EDTA-2Na solution was adjusted to 5.0 by adding 3.0 mol L⁻¹ NaOH solution. After stirring for an additional 1.5 h, the mixtures were again kept overnight at room temperature, and the presence of gel was assessed visually.

Most of the Fe³⁺-Alg did not dissolve in NaHCO₃ (pH 8.0) solution, but it did dissolve completely in EDTA-2Na solution at pH 4.0 and pH 5.0 (Table 3). Ca²⁺-Alg, in contrast, was soluble in both NaHCO₃ (pH 8.0) and EDTA-2Na (pH 5.0) solutions, but it remained insoluble in EDTA-2Na at pH 4.0. Based on these observations, EDTA-2Na solution at pH 5.0 was selected as the optimal condition for the simultaneous dissolution of Fe³⁺-Alg and Ca²⁺-Alg in ocean sediment samples.

In the field of food chemistry, Na⁺-Alg is generally considered to be insoluble in ethanol–water mixtures; therefore, precipitation of Na⁺-Alg in a 4:1 (v/v) ethanol–water solution is commonly employed as the final step in extraction and purification. However, we were concerned that trace amounts of Na⁺-Alg might dissolve in a 4:1 (v/v) ethanol–water solution. If such minor solubility exists, it could result in significant loss during the extraction of trace levels of Na⁺-Alg from ocean sediment.

To evaluate the dissolution behavior, we prepared two Na⁺-Alg solutions (0.50 g L⁻¹ and 2.0 g L⁻¹; 100 mL each) and added 400 mL of ethanol (99.5 %) to each. The mixtures were kept overnight at room temperature to allow precipitation. Afterward, each mixture was filtered through a polycarbonate membrane filter (Nuclepore, Whatman; pore size: 0.22 µm, diameter: 47 mm). The filtrates were collected and evaporated to complete dryness. The resulting residues were re-dissolved in 20mL of Milli-Q water. Subsequently, 0.50 mL of each re-dissolved sample was transferred to a separate tube and diluted 40-fold with Milli-Q water. The concentration of Na⁺-Alg in each diluted sample was then determined using the m-hydroxydiphenyl colorimetric method.

We observed that the 2.0 g L⁻¹ Na⁺-Alg solution was partially soluble in ethanol, with approximately 66 % dissolution, whereas the 0.50 g L⁻¹ Na⁺-Alg solution was completely dissolved (Table 4). These results indicate that trace amounts of alginate are soluble in ethanol. Consequently, a significant loss of alginate could occur during ethanol precipitation, particularly when extracting trace levels of alginate from ocean sediment. Therefore, the ethanol precipitation process was considered unsuitable for the extraction of alginate from sediment samples.

We modified two key aspects of the alginate extraction protocol. First, an EDTA-2Na treatment step was introduced to dissolve Fe³⁺-Alg and Ca²⁺-Alg gels into the liquid phase. Second, the ethanol precipitation step was omitted from the final purification process to avoid significant loss of alginate. The modified extraction protocol, in which Steps 1, 2, 4, and 5 remain identical to those described in the Methods section, is presented below. A complete overview of the updated protocol for alginate extraction from ocean sediment is provided in supplementary material (Fig. S1).

We prepared nine sediment samples with a wet weight of 300 g using muddy sediment rich in organic material collected from Funka Bay (Fig. 1) in March 2024. The test samples (n=3×3) were treated by the modified alginate extraction method as follows (Steps 1–7).

To prepare the test samples, 500 mL of Milli-Q water and 20mL of Na⁺-Alg solutions at concentrations of 10, 5.0, and 2.5 g L⁻¹ – containing 200, 100, and 50 mg of Na⁺-Alg, respectively – were added to 300g of sediment sample (n=3) in separate beakers, followed by stirring for 5 min. These samples were labeled as “sediment + Na⁺-Alg 200 mg,” “sediment + Na⁺-Alg 100 mg,” and “sediment + Na⁺-Alg 50 mg,” respectively. The resulting suspensions were stored overnight at 4 °C to allow the added Na⁺-Alg to transform into Mg²⁺-Alg, Ca²⁺-Alg gel, and Fe³⁺-Alg gel. After this treatment, alginate in the samples was presumed to exist as a mixture of gelled forms (Ca²⁺-Alg and Fe³⁺-Alg) and soluble salts (Na⁺-Alg and Mg²⁺-Alg).

To prepare 80 g L⁻¹ EDTA-2Na solution, 40 g of EDTA-2Na was dissolved in 500 mL of Milli-Q water. The pH of the resulting solution was 4.0 without any adjustment. Most of the sediment containing alginate gel from each tube was transferred into a single beaker using a spoon. To recover any remaining precipitate in tubes, all eight tubes were rinsed with 500 mL of 80 g L⁻¹ EDTA-2Na solution (pH 4.0), and the rinse solution was collected into a beaker. The resulting suspension in the beaker was stirred for 1.5 h, followed by reaction overnight at room temperature to convert Fe³⁺-Alg to Na⁺-Alg via chelation. Subsequently, the pH of the suspension was adjusted to 5.0 by adding 3.0 mol L⁻¹ NaOH solution, followed by stirring for another 1.5h and reaction overnight at room temperature to convert Ca²⁺-Alg into Na⁺-Alg through ion exchange. Following this step, alginate in the suspension sample was presumed to exist primarily as soluble Na⁺-Alg.

An appropriate volume (20–100 mL) of 10 g L⁻¹ NaHCO₃ solution, adjusted based on the estimated amount of alginate, was added to a beaker containing the precipitates to keep the alginate concentration within the range of the calibration curve (Fig. 2). For instance, 100 mL of 10 g L⁻¹ NaHCO₃ solution was added to the sample that had been supplemented with alginate during the preparation step, because it was presumed to contain a large amount of Ca²⁺-Alg. Conversely, for the sample without added alginate, which was presumed to contain a trace amount of Ca²⁺-Alg, only 20 mL of 10 g L⁻¹ NaHCO₃ solution was used. The suspension was then heated to 60 °C with continuous stirring for 1.5 h, followed by reaction overnight at room temperature to convert Ca²⁺-Alg to Na⁺-Alg. After centrifugation, the supernatant was collected into a clean bottle. Following this step, alginate in the solution sample was expected to exist as soluble Na⁺-Alg.

The extract solution appeared dark brown, likely due to the presence of residual phenolic compounds (McHugh, 1987). To reduce the coloration and ensure compatibility with colorimetric analysis, the extract solution was diluted 20–100-fold with Milli-Q water and analyzed colorimetrically as described in Sect. 2.2.

Table 5 shows the yield of alginate extracted from the sediment sample and the corresponding spike recovery based on the modified extraction protocol. We achieved higher spike recoveries (64.7 %–82.6 %) by using the modified extraction protocol than by the conventional protocol (recovery 38.7 %; Table 2). It should be noted that these efficiencies reflect the recovery of alginate naturally present in the sediment samples; therefore, the reported values may be slightly overestimated.