Fucoidan Degradation by Marine Bacteria: Chemicals, Culture Methods, Genomics and Enzyme Analysis

This study investigated the degradation of brown algal fucoidans by marine bacterial communities and isolates. The experimental workflow included fucoidan enrichment, bacterial isolation, genome sequencing, carbohydrate-active enzyme annotation, monosaccharide quantification by LC–MS, co-culture experiments, community modelling and heterologous enzyme characterization.

Chemicals and Reagents for Fucoidan Analysis

Fucoidan from Fucus vesiculosus (Sigma-Aldrich, F8190, lot no. 0000485452) was used as the primary substrate. Additional F. vesiculosus fucoidans were obtained from Marinova (FVF2021547) and Biosynth (YF57714).

Fucoidans from other brown algal species were sourced as follows: Fucus serratus (Biosynth, YF09360), Fucus evanescens (OceanBasis and extracted as described previously⁶⁰), Cladosiphon okamuranus (Biosynth, YF146834), Durvillaea potatorum (Biosynth, YF157165), Ecklonia maxima (Biosynth, YF157166) and Laminaria hyperborea (TheFucoidanStore, LowEndo Fucoidan).

For monosaccharide derivatization, 1-phenyl-3-methyl-5-pyrazolone (PMP) was obtained from Sigma-Aldrich (M70800). Internal standards for LC–MS analysis included d-galactose-¹³C₆ (Sigma-Aldrich, 605379), d-mannose-¹³C₆ (Sigma-Aldrich, 592994) and PMP-[²H₅] (CAS 1228765-67-0), custom-synthesized by BOC Sciences (Shirley).

LC–MS-grade solvents and reagents included acetonitrile (Honeywell), methanol (Honeywell), ethanol (Sigma-Aldrich), formic acid (Sigma-Aldrich) and ammonium formate (Merck). Ultrapure water was produced using a Q-POD system (Merck). Unless otherwise specified, all other chemicals were analytical grade and obtained from Sigma-Aldrich.

Bacterial Growth Media

Three media, described in Supplementary Table 1, were used for bacterial enrichment, isolation on solid medium and routine cultivation of bacterial isolates in MBL medium⁵². Each medium was designed to mimic the ionic composition of coastal seawater and contained 340 mM NaCl, 15 mM MgCl₂, 6.75 mM KCl and 1 mM CaCl₂.

The pH was adjusted to 8.0 using bicarbonate in the enrichment and plate media or 50 mM HEPES in MBL medium. Nutrients included ammonium chloride, sodium phosphate, sodium sulfate, a trace metal mixture and a vitamin mixture⁵².

• Enrichment medium: 0.02% (w/v) F. vesiculosus fucoidan.
• Solid isolation medium: 2% (w/v) carrageenan (Sigma C1013), supplemented with acetate, citrate, xylose, galactose, mannose, glucose, fucose, cellobiose and tryptone at 0.002% (w/v) each.
• Verrucomicrobiota degraders: 0.2% (w/v) F. vesiculosus fucoidan in MBL medium.
• Other degraders and exploiters: 0.2% (w/v) l-fucose and 0.2% (w/v) F. vesiculosus fucoidan.
• Scavengers: acetate, citrate, xylose, galactose, mannose, glucose, fucose, cellobiose and tryptone at 0.02% (w/v) each.

Enrichment of Fucoidan-Degrading Marine Communities

Surface seawater was collected on 30 March 2019 from a rocky shoreline covered by the brown algae F. vesiculosus, Ascophyllum nodosum and Saccharina latissima. The sampling site was located near the Marine Science Center of Northeastern University at Canoe Beach, Nahant, Massachusetts, USA (42° 25′ 10.8732″ N, 70° 54′ 25.686″ W).

Seawater was pre-filtered through a 10-μm PTFE membrane filter (Millipore, JCWP04700) and diluted 1:100. Three replicate enrichment cultures, each containing 25 ml of diluted seawater, were established in 150-ml glass bottles containing 0.02% (w/v) F. vesiculosus fucoidan. Bottles were sealed with rubber stoppers and incubated at 20 °C in the dark with agitation at 50 rpm.

Microbial growth was monitored every 24–48 h by measuring OD₆₀₀. Fucoidan degradation was quantified using the phenol–sulfuric acid assay⁶¹. For each measurement, 200 μl of culture was mixed with 1 ml concentrated sulfuric acid and 200 μl of 5% (v/v) phenol. Samples were incubated at 50 °C for 20 min, and absorbance was measured at 490 nm. Fucoidan concentrations were calculated using an external fucose standard curve.

After 10 days, the enrichment cultures had degraded 40–60% of the initial fucoidan. Serial growth–dilution cycles were then initiated by transferring cultures 1:50 into fresh medium every 2 days for 12 cycles. At every second time point, 10 ml of culture was filtered through a 0.22-μm Sterivex filter (Millipore, SVGPB1010). DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen) for metagenomic sequencing. Final enrichment communities were stored at −80 °C in 15% (v/v) glycerol.

Isolation and Identification of Fucoidan-Degrading Bacterial Strains

Cells from each enrichment culture were enumerated by light microscopy using a counting chamber. Aliquots containing 10², 10³ or 10⁴ cells were plated in triplicate on solid medium in 150-mm Petri dishes (VWR 391-0616). Plates were incubated for 14 days at ambient temperature in the dark.

In total, 768 colonies were selected and re-streaked at least three times until a single colony morphotype was observed. Colonies were lysed in 0.1% (v/v) Triton X-100 in TE buffer, followed by Sanger sequencing of the 16S rRNA gene. Pure isolates were cultivated in MBL medium containing carbon sources appropriate for their metabolic requirements and cryopreserved at −80 °C in 15% (v/v) glycerol.

To remove redundant isolates, 96 strains were selected based on their 16S rRNA sequences for genomic DNA extraction using the DNAadvance kit (Beckman Coulter) and draft genome sequencing. Pairwise average nucleotide identity (ANI) was calculated with OrthoANIu v1.2⁶², identifying 28 unique bacterial strains. The characterized degrader ‘Lentimonas’ sp. CC4 was also included as Verruco4³, resulting in a final collection of 29 bacterial strains.

Metagenome and Isolate Genome Sequencing

Sequencing was performed in collaboration with the BioMicroCenter at MIT. Enrichment metagenomes were sequenced using an Illumina NovaSeq6000 S4 flow cell with 150-nt paired-end reads. Metagenomes were assembled with SPAdes v3.13.0⁶³ using the parameters --meta --only-assembler -k 21,33,55,77.

Metagenome-assembled genomes (MAGs) were reconstructed using CONCOCT v1.0.0⁶⁴, MaxBin v2.2.7⁶⁵ and DAS Tool v1.1.2⁶⁶, all with default settings.

Draft genomes of bacterial isolates were sequenced on an Illumina NextSeq 500 platform using 150-nt paired-end reads at approximately 200× coverage. Assemblies were generated with SPAdes v3.13.0 using --only-assembler -k 21,33,55,77 --careful.

Closed genomes were generated for selected degraders—Flavo12, Flavo40, Flavo56, Flavo94, Gamma88, Verruco25 and Verruco69. Genomic DNA was extracted from 1 ml of culture using the DNeasy Blood & Tissue Kit. Long-read sequencing was conducted on Nanopore PromethION FLO-PRO002 flow cells. Hybrid assemblies combining Illumina short reads and Nanopore long reads were produced with Unicycler v0.4.8⁶⁷ using --keep 3 --mode normal --min_fasta_length 1000 --kmers 77. These assemblies yielded circularized genomes.

MAG completeness and contamination were evaluated using CheckM v1.1.2⁶⁸. A total of 73 medium- and high-quality genomes⁶⁹ were obtained and summarized in Supplementary Table 2. Taxonomic classification and phylogenetic reconstruction were performed with GTDB-Tk v2.0.0⁷⁰ (release 202).

Functional Annotation of Bacterial Genomes

Open reading frames were predicted and annotated using DRAM v1.2.4⁷¹. Carbohydrate-active enzymes (CAZymes) were identified with HMMER⁷² v3.3 using hidden Markov models from the dbCAN database⁷³ (dbCAN-HMMdb-V10).

HMM hits were validated using Diamond⁷⁴ v0.9.14.115 in blastp mode with the --more-sensitive option against the CAZy database, accessed in September 2022. Only matches with bit scores greater than 100 were retained.

Sulfatases were identified by hmmsearch against PF00884, the sulfatase domain, using an e-value threshold of <10⁻⁴ and a minimum alignment length of 100 amino acids. Sulfatases were classified into subfamilies using Diamond v0.9.14.115 against the SulfAtlas v1.2 database⁷⁵.

Classification of Degraders, Exploiters and Scavengers

Genomes were assigned to three metabolic roles—degraders, exploiters or scavengers—based on the presence or absence of enzymes involved in fucoidan and l-fucose metabolism.

Fucoidan Degraders

Degraders were defined as genomes encoding at least five enzymes from a curated set of ten fucoidanase families³. These included the glycoside hydrolase families GH29, GH95, GH141, GH107 and GH168, as well as sulfatase families S1_15, S1_16, S1_17, S1_22 and S1_25.

Fucoidan Exploiters

Exploiters lacked fucoidanases but encoded one of two alternative l-fucose catabolic pathways described in MetaCyc⁵⁷. At least 75% of the enzymes in the relevant pathway had to be present.

Pathway I included:

• K07248, lactaldehyde dehydrogenase;
• K02431, l-fucose mutarotase;
• K00879, l-fuculokinase;
• K01818, l-fucose/d-arabinose isomerase; and
• K01628, l-fuculose-phosphate aldolase.

Pathway II included:

• K18333, l-fucose dehydrogenase;
• K18334, l-fuconate dehydratase;
• K18335, 2-keto-3-deoxy-l-fuconate dehydrogenase;
• K07046, l-fuconolactonase;
• K18336, 2,4-didehydro-3-deoxy-l-rhamnonate hydrolase; and
• K01685, altronate hydrolase.

Fucoidan Scavengers

Scavengers were defined as genomes lacking all enzymes associated with fucoidan degradation and l-fucose catabolism.

Identification of Fucoidan Polysaccharide Utilization Loci

Candidate polysaccharide utilization loci (PULs) were identified in the closed genomes of cultured fucoidan degraders using a sliding 12-gene window. Regions containing at least four CAZyme genes were flagged. Regions encoding two or more known fucoidanases—GH29, GH95, GH141, GH107, GH168, S1_15, S1_16, S1_17, S1_22 or S1_25—were retained as putative fucoidan PULs.

Candidate loci were manually curated to exclude regions likely to target other polysaccharides. Five loci from Flavobacteriia containing carrageenases from GH16, GH82, GH150 and GH167 were removed. The final curated dataset contained 34 fucoidan-associated PULs spanning 54 CAZyme families, as reported in Supplementary Table 3.

Comparison of Fucoidan-Degrading Enzyme Repertoires

For each degrader, the fucoidanase repertoire included:

1. All CAZymes and sulfatases located within fucoidan PULs; and
2. Additional chromosomal genes from enzyme families predicted to target sulfated fucose or rare fucoidan-associated sugars.

The sulfated-fucose-associated families were GH29, GH95, GH141, GH107, GH168, S1_15, S1_16, S1_17, S1_22 and S1_25. Families predicted to target rare sugars included GH30, GH31, GH36, GH39, GH92, GH97, GH115 and GH120.

For the characterized degrader³ Lentimonas sp. CC4, only enzymes upregulated at the protein level during growth on F. vesiculosus fucoidan were included, specifically those from co-expression clusters 2–6.

Fucoidan-associated CAZymes were clustered using MMseqs2 v13.45111⁷⁶ with easy-cluster --min-seq-id 0.6 -c 0.5 --cov-mode 0. Enzyme homologues were defined using a threshold of at least 60% amino acid identity and at least 50% alignment coverage for both query and target sequences.

These clusters were used to compare enzyme-family abundance and repertoire composition across isolates. Pairwise similarities were calculated using the Jaccard index.

To evaluate how enzyme diversity scaled with community size, a rarefaction analysis was performed. In each of 10,000 iterations, n degrader strains (n = 1–29) were randomly selected, the number of unique enzyme clusters was counted and expected total richness was estimated using the Chao1 estimator.

Functional Annotation of Fucoidan-Active CAZymes

The activities of fucoidan-associated enzymes from isolated degraders were inferred by comparison with characterized CAZymes in the CAZy database, accessed in March 2026, using Diamond v0.9.14.115 in blastp mode with the --more-sensitive option.

For each protein domain, the highest-scoring hit was used to assign a putative function and EC number when available. Proteins containing multiple catalytic domains were annotated at the domain level.

Annotation confidence was classified as:

• High: at least 70% identity and 70% coverage;
• Medium: at least 30% identity and 50% coverage; and
• Family-level: assignments below the medium-confidence thresholds.

Enzymes were assigned to broad activity classes when the inferred EC number, CAZyme family or closest characterized homologue indicated compatibility with known fucoidan linkage chemistry. These classes included activities targeting fucose, galactose, mannose, xylose and glucuronic acid.

• GH107, GH141 and GH168: endo-acting fucoidanases;
• GH29 and GH95: exo-α-l-fucosidases;
• GH36 and GH97: α-galactosidases;
• GH2: β-galactosidases;
• GH92: α-mannosidases;
• GH115: α-glucuronidases; and
• GH31, GH39, GH120 and GH30: α-/β-xylosidases.

Enzyme families without evidence of activity against known fucoidan residues were classified as hypothetical. Inferred activities, EC numbers and confidence levels are provided in Supplementary Table 3.

Phylogenetic validation of representative enzyme families was performed using MAFFT with the L-INS-i strategy (--localpair --maxiterate 100) for sequence alignment, trimAl (-gt 0.1) for trimming and FastTree using the LG + Γ model for phylogenetic inference.

Fucoidan Degradation Assays in Monoculture

Substrate utilization and fucoidan degradation were assessed for all 29 isolates using defined carbon-source growth assays. Strains were revived from glycerol stocks in 3 ml MBL medium for 3–6 days using carbon sources matched to their metabolic requirements.

Cultures were washed in carbon-free MBL medium and inoculated into 200 μl of fresh medium containing one of the following carbon sources at 0.2% (w/v): l-fucose, d-galactose, d-mannose, d-xylose, d-glucuronic acid or F. vesiculosus fucoidan.

All cultures were grown in biological triplicate (n = 3) at 20 °C with orbital shaking at 200 rpm in flat-bottom polystyrene 96-well plates (Corning). Growth was monitored by OD₆₀₀ for 5 days using a Tecan Sunrise plate reader.

For fucoidan degradation measurements, cultures were sampled at the final time point during early stationary phase. Cells were removed by centrifugation at 2,200 rpm for 10 min, and cell-free supernatants were analysed following acid hydrolysis and LC–MS quantification.

Time-Resolved Fucoidan Degradation

Eight degraders were cultivated in 4 ml MBL medium containing F. vesiculosus fucoidan in 24-deep-well plates (Eppendorf) sealed with breathable lids (Kuhner). Samples were collected throughout growth. Supernatants were analysed for fucoidan-derived monomers released by acid hydrolysis and free extracellular monosaccharides.

Acid Hydrolysis of Fucoidan in Culture Supernatants

Acid hydrolysis was used to cleave the glycosidic bonds of residual fucoidan in culture supernatants and release constituent monosaccharides. Sampled supernatants (5 μl) were mixed with 45 μl double-distilled water and 50 μl of 2 M HCl.

The HCl solution contained 15 μM each of d-galactose-¹³C₆ and d-mannose-¹³C₆. These compounds served as processing internal standards to correct for technical variability during sample preparation.

PCR plates were sealed with plastic strips (Thermo Fisher Scientific AB0600 and AB0784). Hydrolysis was performed for 24 h at 100 °C in an oven using a custom clamping device to prevent leakage. Hydrolysed samples were neutralized by adding 4 M NaOH.

PMP Derivatization of Monosaccharides

Acid hydrolysates, consisting of 10 μl sample and 15 μl double-distilled water, or free-monosaccharide samples (25 μl), were derivatized with 75 μl of 0.1 M PMP in 2:1 methanol:double-distilled water containing 0.4 M ammonium hydroxide. Reactions were conducted for 100 min at 70 °C according to a published protocol⁴⁷.

Absolute quantification was performed using an external calibration curve containing glucuronic acid, xylose, fucose, galactose and mannose at concentrations from 1 mM to 200 nM. Standards were prepared in a matrix identical to the samples.

Following derivatization, samples and standards were neutralized with 2 M HCl and diluted 1:50 in 0.1% (v/v) formic acid in double-distilled water containing 50 nM injection internal standards. These standards corrected for changes in ionization efficiency at the mass-spectrometer ion source and consisted of glucuronic acid, xylose, fucose, galactose and mannose derivatized with heavy-labelled PMP-[²H₅].

Targeted LC–MS Analysis of PMP-Derivatized Sugars

PMP derivatives were analysed using a SCIEX qTRAP5500 mass spectrometer coupled to an Agilent 1290 Infinity II LC system. Chromatographic separation used a Waters CORTECS UPLC C18 column, 90 Å, 1.6 μm, 2.1 mm × 50 mm, with a guard column and 0.2-μm inline filter.

Mobile phase A consisted of 10 mM NH₄ formate in double-distilled water with 0.1% (v/v) formic acid. Mobile phase B consisted of 100% acetonitrile with 0.1% (v/v) formic acid.

PMP derivatives were separated using an initial isocratic flow of 13% buffer B for 30 s, followed by a binary gradient from 13% to 36% buffer B over 2 min. A 30-s wash at 100% buffer B was followed by 30 s of re-equilibration. The flow rate was 0.5 ml min⁻¹, with a constant pressure of approximately 430 bar.

A diverter valve directed the first 1.5 min of chromatography to waste. The column compartment was maintained at 40 °C and the autosampler at 10 °C. Electrospray ionization settings were 625 °C, curtain gas 30, collision gas medium, ion spray voltage 5,500 V and ion source gases 1 and 2 at 90 arbitrary units.

Data were acquired in positive mode using multiple reaction monitoring (MRM) with previously optimized transitions and collision energies⁴⁶. For example, the galactose derivative had an exact Q1 mass of 511.2 m/z and was fragmented at a collision energy of 35 V to produce a quantifier ion at 175.0 m/z and a diagnostic fragment at 217.2 m/z. All MRM transitions and retention times are listed in Supplementary Table 4.

Each analytical run typically included 150–250 randomized samples. The highest-concentration standard mix was used for quality control. QC samples and water blanks were injected every 15 samples to monitor analytical consistency and carryover. Calibration curves were prepared in triplicate, and each sample was analysed in technical duplicate. Guard columns and inline filters were replaced every 500–1,000 injections.

Absolute Quantification of Fucoidan-Derived Monosaccharides

Chromatographic data were processed in Skyline⁷⁷ v24.1.0.414. Peak areas were integrated using default settings and exported for downstream analysis in Python. A reproducible workflow is available at https://github.com/EnvSysMicroLAB/Sichert2025_Fucoidan.

Peak areas for target compounds and ¹³C-labelled processing standards were first corrected using injection internal standards. The corrected ¹³C processing standards were then used to normalize target signals. Absolute monomer concentrations were calculated by linear regression against external calibration curves. Technical duplicates were averaged to obtain final concentrations.

Quantification and Statistical Analysis of Monomer Degradation

Fucoidan degradation was quantified by comparing concentrations of fucose, glucuronic acid, galactose, mannose and xylose in culture supernatants with those in an uninoculated medium control processed in parallel.

“Rare” monomers were defined as the combined concentration of the four non-fucose sugars, whereas “total” monomers represented the sum of all five measured monosaccharides. Degradation of monosaccharide i by strain or strain combination j was calculated as:

$$ {f}_{i,j}=100\times \frac{{[{\rm{S}}]}_{i,{\rm{control}}}-{[{\rm{S}}]}_{i,j}}{{[{\rm{S}}]}_{i,\text{control}}} $$

where [S]_i,control is the concentration in the uninoculated control, [S]_i,j is the concentration following growth of strain j and f_i,j is the relative monomer degradation.

For each strain–monomer combination, a one-sided t-test was used to determine whether the observed decrease exceeded abiotic background variation. P values were corrected using the Benjamini–Hochberg false discovery rate (FDR) procedure with α = 0.01. Degradation was considered significant when both the raw P value and FDR-adjusted q value were <0.01.

Seven distinct datasets were generated, with monosaccharide concentrations and statistical analyses reported in Supplementary Table 4. Unless otherwise specified, Pearson correlation coefficients and associated P values were calculated from replicate-level data from three biological replicates without prior aggregation into means.

High-Throughput Bacterial Community Experiments

Four high-throughput growth experiments were conducted to evaluate biological interactions affecting fucoidan degradation. All experiments included biological triplicates (n = 3) and were incubated at 20 °C with orbital shaking at 200 rpm.

Glycerol stocks were revived in 3 ml MBL medium for 3–6 days using carbon sources matched to the metabolic requirements of each isolate. Cultures were diluted 1:5 into fresh medium in 96-deep-well plates containing 1.8 ml per well and a 4-mm glass bead (Sigma-Aldrich, Z143936). Cultures were incubated for 18 h until exponential phase, corresponding to OD₆₀₀ values of 0.15–0.3.

Exponential-phase cells were washed in carbon-free MBL medium and used to inoculate experimental cultures at an initial OD₆₀₀ of 0.01 per strain. Cultures were incubated for up to 5 days in the same deep-well plate format and sealed with breathable lids (Kuhner, 104098 and 104106).

At defined time points, 100 μl of culture was collected for OD₆₀₀ measurements using a Tecan Sunrise plate reader. Samples were centrifuged at 2,200 rpm for 10 min, and supernatants were stored at −20 °C for subsequent analysis.

Verrucomicrobiota-Centred Pairwise Co-Cultures

To assess the effects of community members on primary fucoidan degraders, each of the 29 isolates was co-cultured at a 1:1 ratio, normalized by OD₆₀₀, with one of three primary degraders: V4, V25 or V69. This produced 87 distinct pairwise combinations.

Co-cultures were incubated for 5 days, after which fucoidan degradation was measured. Growth dynamics were evaluated using:

1. Total biomass: the mean of the two highest OD₆₀₀ values recorded during the time course; and
2. Apparent growth rate: the inverse of the time required to reach an OD₆₀₀ threshold of 0.15 (1/t_th).

Aggregate formation in V69 co-cultures with F12, F40, F56 and G88 interfered with OD₆₀₀-based biomass estimates. In these cases, bacterial biomass was estimated indirectly by extracting DNA from 100 μl of culture using the DNAdvance Kit (Beckman Coulter), followed by absolute DNA quantification with the Femto Bacterial DNA Quantification Kit according to the manufacturer’s instructions.

Potential synergistic or antagonistic effects were identified by comparing biomass formation and fucoidan degradation in co-cultures with the corresponding monocultures. Biomass differences were considered significant when both the P value and FDR-adjusted value were <0.01 using the Benjamini–Hochberg correction with α = 0.01.

Degradation differences were considered significant when both the P value and FDR-adjusted value were <0.05 using α = 0.05 and the absolute log₂-transformed fold change exceeded 0.1.

Strain-Specific Quantification by qPCR

For strain-specific abundance measurements, 100 μl of co-culture was collected and subjected to two freeze–thaw cycles. DNA was extracted using the DNAdvance Kit (Beckman Coulter, A48705).

Strain-specific qPCR primers targeted approximately 100-bp regions of single-copy marker genes. Reactions were performed in 20 μl volumes using GoTaq qPCR Master Mix (Promega) on a QuantStudio 3 Real-Time PCR System.

Primer efficiencies ranged from 95–105%. Primer specificity and quantitative recovery were evaluated using defined mock communities containing all strains at known input ratios. Recoveries ranged from 90–127% (Supplementary Table 6).

Absolute abundances were determined by comparing Ct values with strain-specific standard curves generated from serial dilutions of genomic DNA ranging from 10 ng μl⁻¹ to 10 pg μl⁻¹. The resulting DNA concentrations were used to compare monoculture and co-culture abundances and classify interaction outcomes⁴⁸.

Testing Synergism Beyond Resource Partitioning

A null model was developed to evaluate synergistic interactions in pairwise fucoidan degradation. The model estimated expected co-culture performance from monoculture data under the assumption that strains had fully overlapping substrate preferences and lacked metabolic complementarity.

Under this model, each monomer could be degraded only to the level achieved by the better-performing monoculture. The synergism score was calculated as:

$$ \sigma (i,j)={f}_{\text{total}}(i,j)-\sum _{m\in \{{\rm{Fuc}},\text{Man},\text{Gal},\text{Xyl},\text{GlcA}\}}\max ({f}_{m}(i),{f}_{m}(j)) $$

Here, f(i,j) is the total fucoidan fraction degraded by strains i and j in co-culture, while f_m(i) and f_m(j) are the fractions of monomer m degraded by each strain in monoculture.

All terms were expressed as stoichiometrically weighted fractions rather than raw percentages. Each monomer-specific contribution was scaled according to its relative abundance in the fucoidan composition. This conservative approach prevented redundant contributions from metabolically similar strains from inflating the expected degradation.

Positive σ values indicated synergism through functional complementarity, whereas negative values indicated antagonism. A two-sided t-test was used to determine whether observed co-culture degradation differed significantly from the null expectation. P values were corrected using the Benjamini–Hochberg procedure. Synergism scores were considered significant when P < 0.01, q < 0.01 and the absolute effect size exceeded 5 (|σ| > 5).

This framework can also be extended to communities containing more than two members by using the maximum monomer degradation measured in the best-performing drop-out community as the null expectation.

Cross-Feeding of Residual Fucoidan

Partially degraded fucoidans were recovered from V4, V25 and V69 cultures grown to late stationary phase in 1 l of MBL medium containing 0.2% (w/v) F. vesiculosus fucoidan.

Sterile-filtered supernatants were concentrated on ice using an Amicon stirred ultrafiltration cell (EMD Millipore, UFSC20001) equipped with a 1-kDa cellulose membrane (EMD Millipore, PLAC06210) and nitrogen gas pressure. Desalting was performed by adding double-distilled water followed by additional concentration. Concentrated residual fucoidans were lyophilized.

To determine whether secondary degraders could use residual substrates, F12, F40, F56, F94, V25, V69 and V4 were cultivated in 1.5 ml MBL medium containing 0.1% (w/v) of each residual fucoidan. Cultures were incubated for 4 days, and degradation was measured from final-time-point supernatants relative to the initial substrate concentration.

Additional degradation by secondary degrader i on residual fucoidan from primary degrader j was compared with co-culture performance using:

$$ {f}_{{\text{total},\text{residual}}_{j}}(i)={f}_{\text{total},\text{untreated}}(i,j)-{f}_{\text{total},\text{untreated}}(j) $$

Here, f_{total,residualj}(i) denotes degradation by secondary degrader i on residuals produced by primary degrader j. The term f_{total,untreated}(i,j) represents total degradation by the co-culture on untreated fucoidan, and f_{total,untreated}(j) represents degradation by primary degrader j alone.

Full Combinatorial Fucoidan-Degrader Communities

All 127 possible combinations of seven degraders—F12, F40, F56, F94, V25, V69 and V4—were constructed, together with a no-cell negative control.

Strains were mixed at equal optical density, with each strain contributing an initial OD₆₀₀ of 0.01. Communities were inoculated into 1.8 ml MBL medium containing 0.2% (w/v) F. vesiculosus fucoidan as the sole carbon source. Cultures were incubated for 5 days with three biological replicates, after which supernatants were collected for fucoidan degradation analysis.

Trait-Based Modelling of Community Fucoidan Degradation

A nonlinear trait-based model was developed to predict degradation from community composition. Each bacterial strain was assumed to contribute independently to the degradation of two fucoidan-derived monomer pools: fucose and rare sugars.

Each strain i was assigned two degradation-capacity parameters, w_i^Fuc and w_i^Rare. For a community containing m strains and represented by a presence/absence vector x_i ∈ {0,1}, effective degradation capacity was calculated as:

$$ {C}^{\text{Fuc}}=\mathop{\sum }\limits_{i=1}^{m}{w}_{i}^{\text{Fuc}}\times {x}_{i} $$

To represent the saturating behaviour observed experimentally, a Hill function was applied:

$$ {D}^{\text{Fuc}}=\frac{{({C}^{\text{Fuc}})}^{n}}{{K}_{{\rm{m}}}^{n}+{({C}^{\text{Fuc}})}^{n}} $$

$$ {D}^{\text{Rare}}=\frac{{({C}^{\text{Rare}})}^{n}}{{K}_{{\rm{m}}}^{n}+{({C}^{\text{Rare}})}^{n}} $$

Here, K_mⁿ is the half-saturation constant and n is the Hill coefficient. Both parameters were shared across the two monomer classes. D^Fuc and D^Rare represent the predicted fraction of each monomer pool degraded by the community.

Experimental degradation data from 127 communities were used to fit one degradation-capacity parameter for each strain and monomer pool, along with the shared parameters K_mⁿ and n. Parameters were inferred by minimizing the root mean squared deviation between predicted and observed degradation. Optimization was performed in Julia using LsqFit.jl.

Initial and final parameter values and their effects on model fit are shown in Extended Data Fig. 9a–c. Goodness of fit was evaluated by visual comparison and R² statistics (Extended Data Fig. 9d). All code and data are available at https://github.com/EnvSysMicroLAB/Sichert2025_Fucoidan.

Degradation of Diverse Brown Algal Fucoidans

Degrader activity was evaluated across a panel of eight additional brown algal fucoidans. The panel included fucoidans from six brown algal species selected for differences in composition and structure, as well as two F. vesiculosus fucoidan variants obtained from different vendors.

The two vendor-specific preparations were included to capture potential compositional and structural differences caused by sampling time, sampling season and extraction method^6,33.

For community experiments, degraders were grouped to minimize redundancy and maximize functional complementarity:

• Group A: V25, V4 and F56;
• Group B: F12, F40 and F94; and
• Group C: V69.

The groups were tested individually, in all pairwise combinations and as a triplet, producing seven community configurations. Each community received equal OD₆₀₀ contributions from its member strains and was inoculated into 1.8 ml MBL medium containing 0.2% (w/v) of one of the eight fucoidans.

Cultures were incubated for 5 days with three biological replicates. Fucoidan degradation was quantified by full monosaccharide analysis, and monomer-specific depletion was calculated using equation (1). The resulting 63 community–substrate combinations were used for out-of-sample model validation.

To account for differences in monosaccharide composition, each monomer’s contribution to total degradation was weighted by its relative abundance in the corresponding polysaccharide.

Extended LC–MS Monosaccharide Profiling

Potential co-extracted polysaccharides were assessed using full monosaccharide profiling with an extended 10-min LC–MS method capable of resolving 21 monosaccharides. Acid hydrolysis, PMP derivatization, internal controls and data processing were performed as described above.

Chromatographic separation differed only in the gradient. The method began with a 2-min isocratic hold at 15% buffer B, followed by a linear gradient from 15% to 20% buffer B over 5.5 min. Buffer B was increased rapidly to 100% at 7.5 min, followed by a 1-min wash at 100% buffer B and re-equilibration to the starting conditions until 9.5 min. The flow rate was 0.5 ml min⁻¹, and the column temperature was 50 °C.

The standard mixture contained fucose, galactose, xylose, mannose, glucuronic acid, glucose, mannuronic acid, guluronic acid, rhamnose, glucosamine, galactosamine, gulose, allose, idose, galacturonic acid, lyxose, ribose, arabinose, iduronic acid, talose and altrose. Standards were prepared under matrix-matched conditions across concentrations ranging from 100 nM to 500 μM.

Monosaccharide concentrations were converted to anhydro-corrected masses to approximate polysaccharide yields. Sulfate content was inferred from literature-reported weight fractions using the fucoidan-associated monosaccharide pool. This enabled quantitative mass-balance estimates for total hydrolysable carbohydrates and unaccounted fractions, as summarized in Supplementary Table 9.

Heterologous Expression and Purification of Fucoidan-Active Enzymes

The coding sequences of 17 glycoside hydrolases were codon optimized, synthesized and cloned into the pET-28a(+) expression vector containing an N-terminal 6×His tag. Native signal peptides were omitted. Constructs were produced by Twist Bioscience.

Plasmids were transformed into competent BL21(DE3) Escherichia coli cells (New England Biolabs) following the manufacturer’s instructions. Expression strains were grown in 200 ml Luria–Bertani medium containing kanamycin at 37 °C until OD₆₀₀ reached 0.8.

Protein expression was induced with 0.1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG), followed by incubation at 12 °C for 16 h. Cells were harvested by centrifugation at 3,000g for 20 min at 4 °C and lysed with B-PER Bacterial Protein Extraction Reagent (Thermo Fisher Scientific).

Soluble proteins were purified by immobilized metal affinity chromatography using His GraviTrap TALON columns (Cytiva, 29-0005-94). Columns were equilibrated with lysis buffer before sample loading.

Non-specific proteins were removed using two 10-ml washes with IMAC 20 buffer and two 10-ml washes with IMAC 40 buffer:

• IMAC 20 buffer: 50 mM Tris-HCl, 500 mM NaCl, 5% glycerol and 20 mM imidazole, pH 8.0.
• IMAC 40 buffer: 50 mM Tris-HCl, 500 mM NaCl, 5% glycerol and 40 mM imidazole, pH 8.0.
• IMAC 200 elution buffer: 50 mM Tris-HCl, 500 mM NaCl, 5% glycerol and 200 mM imidazole, pH 8.0.

His-tagged enzymes were eluted with 2 ml IMAC 200 buffer. Expression, purity and expected molecular weights were assessed by SDS-PAGE and Coomassie staining.

Nine of the 17 proteins were successfully expressed in the soluble fraction:

• V25|GH97_A;
• V25|GH36;
• F56|GH39;
• F56|GH130;
• F56|GH92_C;
• F56|GH92_E;
• F56|GH115_C;
• V69|GH97_A; and
• V69|GH97_B.

Purified proteins were dialysed overnight at 4 °C against 50 mM Tris-HCl and 500 mM NaCl, pH 8.0. Final protein concentrations were measured using the broad-range Qubit Protein BR Assay (Thermo Fisher Scientific). Full construct names, accession numbers, sequences, molecular weights, signal peptide predictions and solubility data are provided in Supplementary Table 7.

Enzyme Activity and Kinetic Assays

All enzyme assays were conducted in MBL medium containing 50 mM phosphate buffer, pH 8.0, at 20 °C. The final enzyme concentration was 5 nM.

Initial activity screens used pNP-labelled substrate analogues at 1 mM, including:

• pNP-α-d-galactopyranoside;
• pNP-α-d-mannopyranoside;
• pNP-α-d-xylopyranoside;
• pNP-β-d-galactopyranoside;
• pNP-β-d-mannopyranoside; and
• pNP-β-d-xylopyranoside.

Enzyme activity was determined by measuring the release of para-nitrophenol at 410 nm using a microplate reader.

Michaelis–Menten kinetics for F56|GH39 and V25|GH36 were determined from initial rates measured across substrate concentrations of 0.1–100 mM. Initial velocities were calculated from the linear phase of product formation. K_m and V_max values were estimated by nonlinear least-squares fitting to the Michaelis–Menten equation.

Native-substrate activity was assessed by incubating enzymes with 0.2% (w/v) F. vesiculosus fucoidan and 0.1% (w/v) residual fucoidan recovered from V4, V25 and V69 cultures. Reactions were incubated for 24 h, and released monosaccharides were quantified by PMP derivatization followed by LC–MS.

Global Distribution of Fucoidan-Degrading Isolates

The distribution of fucoidan-degrading isolates across ocean environments was assessed using global rRNA gene and metagenome-based databases.

First, 16S rRNA genes were extracted from isolate genomes using pyBarrnap v0.5.1 with evalue = 10−6, lencutoff = 0.8 and reject = 0.25. Sequences were compared with the MicrobeAtlas database⁷⁹, and matches were retained when sequence identity was at least 99%.

Isolate genomes were also compared with species in the mOTUs database⁵⁵, a global, species-resolved collection of isolate and metagenome-derived genomes. Comparisons were performed using the mOTUs profiler classification function, which aligns ten single-copy marker genes to representative genomes.

Because the mOTUs database contains relatively few macroalgal microbiome samples, isolates were additionally compared with two macroalgae-associated datasets.

The first dataset comprised amplicon sequence variants from an annual sampling campaign of macroalgal thalli along the Roscoff coastline⁸⁰. ASVs were aligned to full-length isolate 16S rRNA genes and retained when sequence identity was at least 99%.

The second dataset contained isolate genomes and MAGs from epiphytic macroalgal microbiomes in a coastal region of China⁸¹. Genome comparisons were performed using species-level clustering with dRep v3.5.0 (−comp 50 −con 10 −sa 0.95 −nc 0.3) and a 95% ANI threshold.

Information on the locations where matched references were detected was combined across databases and published datasets. Global distributions were visualized in R v4.3.3 using the ggplot2, rnaturalearth and sf packages.

Global Diversity and Dynamics of Fucoidan Degraders and Exploiters

Species in the mOTUs database were analysed to investigate the global distribution of fucoidan degraders and evaluate the potential ecological relevance of synergistic degradation. Species-representative genomes were retrieved and restricted to genomes derived from isolates or ocean metagenomes.

For initial screening, representative genomes were annotated against dbCAN v14⁷³ using HMMsearch⁷². Candidate genomes were retained when they contained at least one fucoidanase GH family, using a horizontal coverage threshold of >0.25 and an e-value threshold of <1 × 10⁻¹⁰. This produced 3,692 candidate genomes.

Candidate genomes underwent comprehensive annotation using Diamond⁷⁴ searches against CAZyDB⁸² and SulfAtlas⁷⁵, together with HMMsearch against KEGG⁸³ and PFAM.

Degrader species, or mOTUs, were defined as genomes encoding at least five fucoidan-targeting glycoside hydrolase and sulfatase families. Fucoidan degradation capacity was estimated from the number of fucoidanase genes per genome. Fucoidan-associated PULs were annotated using the procedure described above.

To provide ecological context, degrader species were profiled across more than 11,000 ocean metagenomes using the mOTUs profiler. Samples containing at least one degrader mOTU were retained, resulting in 12,347 samples.

For each sample, the diversity of detected degrader species and their relative abundance were calculated. Relative abundance was defined as the proportion of genomes represented by each species within the microbial community.

Estimating the Potential for Synergistic Fucoidan Degradation

Potential synergistic degradation was assessed by measuring variation in the fucoidan-degradation capacity of co-occurring degrader species using two complementary approaches.

First, degrader PULs were compared according to the composition of enzymes targeting the sulfated fucose backbone versus rare-sugar monomers. Species were divided into quartiles based on the proportion of fucose-targeting enzymes in their PULs.

Potential synergistic degradation was defined as the co-occurrence of a top-quartile and bottom-quartile degrader species. This represented the co-occurrence of species with PULs specialized for fucose degradation and rare-sugar monomer degradation, respectively.

As a complementary analysis, the median and standard deviation of fucoidanase gene content among co-occurring degrader species were calculated for each ocean metagenome.

Reporting Summary

Additional information on the research design is available in the Nature Portfolio Reporting Summary linked to this article.

Source: www.nature.com