Phylogenetic and functional characterization of the microbiome of Sargassum seaweed waste

ABSTRACT In order to utilize seaweed waste as a feedstock in biorefinery, it is essential to characterize the microbial communities and their metabolic processes in seaweed decomposition. In this study, the phylogeny and functionality of the prokaryotic and fungal microbiome of Sargassum seaweed waste that has inundated the Barbados coast were assessed for the first time by targeting the 16S rRNA gene and internal transcribed spacer (ITS) region of ribosomal RNA, respectively, using the cultivation-independent Illumina next-generation sequencing approach. The resulting 67 032 and 91 085 high-quality paired-end reads of 16S rRNA gene and ITS were assigned to 734 and 179 amplicon sequence variants (ASVs), respectively. The predominant bacterial genera recorded were Muricauda, Aminobacterium, Mesorhizobium, Marinobacter, Reichenbachiella, Hyphomonas, Simiduia, Aquamicrobium, Oceanicola and Alcanivorax. The major genera of Archaea belonged to Methanosarcina, Methanogenium and Methanosaeta, and that of fungi to Sigmoidea, Tylopilus and Mucor. The metabolic functions of the Sargassum metagenome, analysed via PICRUSt, revealed the predominant metabolic pathways related to the biosynthesis of terpenoids and polyketides, the biosynthesis of other secondary metabolites, the metabolism of cofactors and vitamins, and the xenobiotic biodegradation. Furthermore, the prevalence of diverse alginate lyase-producing bacteria suggests that Sargassum waste is an ideal niche material for the isolation of potent microbial biocatalysts to sustainably depolymerize alginate and thereby produce third-generation biofuels and bioactive alginate oligosaccharides with novel industrial applications.


Introduction
The input of nutrients via anthropogenic sources, including agricultural runoff and wastewater discharge, and the impact of coastal upwelling, hurricane and precipitation have been identified as the causal agents of coastal eutrophication and the triggering of algal blooms (Gower & King, 2019;Mohapatra, Schambeau, Pannell, & Kiene, 2017;Smetacek & Zingone, 2013).Global climate and land-use changes have also been identified as the environmental stressors for the induction of seaweed blooms (Morand & Merceron, 2005;Oviatt, Huizenga, Rogers, & Miller, 2019).Seaweed blooms disrupt the coastal ecosystem functionality via alteration of the biogeochemical cycles (Aires, Muyzer, Serrão, & Engelen, 2019;Hervé, Lambourdière, René-Trouillefou, Devault, & Lopez, 2021;Smetacek & Zingone, 2013).The bloom-forming seaweed species are classified into Chlorophyta, Phaeophyceae (Ochrophyta) and Rhodophyta.Recently, two pelagic species of the brown seaweed Sargassum (S. fluitans and S. natans) have caused massive blooms in the subtropical and tropical Atlantic (Gower & King, 2019;Langin, 2018).These blooms spanned a distance of 8850 km from the African Gulf of Guinea to the Greater Caribbean Sea and the Gulf of Mexico in the year 2018, with an estimated weight of >20 million tons (Wang et al., 2019).The recurring bloom of the holopelagic species of Sargassum has been adversely affecting the regional economy because of the inundation of putrefying biomass (Sargassum waste) onto the pristine beaches, which consequently deters the tourists (Chávez et al., 2020).Additionally, the Sargassum inundation is a threat to the public and environmental health since it fosters production of toxic and corrosive hydrogen sulphide and attracts skin-irritating chironomid insects (Merle et al., 2021;Resiere et al., 2018).
In this paper, the phylogenetic diversity of the prokaryotic and fungal communities associated with the Sargassum waste piled up at two different beaches off Barbados coast were examined via the cultivationindependent Illumina sequencing approach by targeting the 16S rRNA gene and the internal transcribed spacer (ITS) region of ribosomal RNA, respectively.Additionally, the metabolic functions of the Sargassum waste metagenome were predicted based on 16S rRNA marker gene sequences' data (Langille et al., 2013).The alginate-depolymerizing efficacy of the identified prokaryotic species was also assessed via a carbohydrateactive enzymes (CAZy) database (Drula et al., 2022).This molecular analysis provides insight, not only into the phylogeny of microorganisms associated with the Sargassum waste piles but also their predicted functional profiles.This appears to be the first report characterizing the phylogeny and functionality of the microbiome of pelagic Sargassum seaweed waste inundating Barbados' coast.Knowledge of the microbiome composition and the metabolic functions of its components in Sargassum waste piles are essential in evaluating the microbiome's impact on coastal ecosystems and on efficient valorization of these renewable biomasses.Additionally, this information will assist in the isolation and identification of novel microbial strains for use in the fermentation of seaweed waste and production of bioactive compounds.

Collection of Sargassum waste
The samples were collected in quadruplicates from Sargassum waste inundating two beaches [Harrismith (HS;13.12ºN,59.42ºW)and Long Beach (LB; 13.06ºN, 59.49ºW)], off the coast of Barbados.The sample collection sites at both beaches were located away from the splash zone.The Sargassum waste samples were comprised of the decomposed biomasses of two pelagic species of Sargassum, S. fluitans and S. natans (Fig 1).Due to the decomposition of Sargassum biomasses, it was not possible to identify their morphotypes.After collection, samples were stored in sterile containers and transported to the laboratory on ice.Upon arrival, the samples were gently rinsed with autoclaved seawater to remove the dirt and were stored at −80 º C until further processing.

DNA extraction
One gram of frozen Sargassum biomass from each sample was homogenized with a sterile mortar and pestle.Then, 0.25 g of the homogenate was used for extraction of DNA via the SurePrep TM soil DNA isolation kit (Thermo Fisher Scientific, Fair Lawn, New Jersey, USA) according to the manufacturer's instructions.The extracted DNA was resuspended in the supplied elution buffer and stored at −80 º C. The DNA concentrations were quantified using QuantiFluor TM ONE dsDNA System (Promega, Madison, Wisconsin, USA) and Quantus™ Fluorometer (Promega, Madison, Wisconsin, USA).

Illumina amplicon sequencing
The DNA extracted from quadruplicate Sargassum samples from each beach was pooled into one representative sample, then sent in dry ice to the ZymoBIOMICS® targeted sequencing service (Zymo Research Service, Irvine, California, USA) for sequencing.The sequencing libraries for prokaryotic (Bacteria and Archaea) and fungal communities were prepared using Quick NGS Library Preparation Kit (Zymo Research, Irvine, California, USA) by targeting the V3-V4 region of the 16S rRNA gene and the ITS2 region of rRNA, respectively.The libraries were prepared using qPCR to minimize the chimera formation.The cleaning of the libraries was performed using the Select-a-Size DNA Clean and Concentrator™ (Zymo Research Service, Irvine, California, USA) and quantified using TapeStaion™ (Agilent Technologies, Santa Clara, California, USA) and Qubit™ (Thermo Fisher Scientific, Waltham, Washington, USA).The sequencing of the libraries was conducted by using an Illumina MiSeq platform with a v3 reagent kit (600 cycles) and a 10% PhiX spike-in.

Bioinformatics
The unique amplicon sequence variants (ASVs) were curated from the raw reads of each beach sample after removal of the potential sequencing errors and chimeric sequences via the DADA2 (v.1.26)pipeline (Callahan et al., 2016).Taxonomy assignments of unique ASVs were conducted using the UCLUST program (v. 1.2.22) integrated in QIIME v.1.9.1 (Caporaso et al., 2010) and Zymo Research's internally curated 16S database for prokaryotes (Zymo Research Service, Irvine, California, USA) and the UNITE database (Nilsson et al., 2018) for fungi as the references.Alpha-and betadiversity of the microbial communities were assessed via QIIME v.1.9.1 (Caporaso et al., 2011).

Functional predictions of the metagenome
The putative functions of the metagenome of Sargassum waste samples from each beach were predicted based on the 16S rRNA gene sequences dataset via the EzBiocloud 16S-based MTP pipeline with XOR algorithm and PICRUSt analysis (ChunLab, Inc., Seoul, South Korea).Functional classifications of the metagenome were performed according to the Kyoto Encyclopedia of Gene and Genome (KEGG) modules (i.e., functional units of gene sets in KEGG metabolic pathways).

Sequence information
A total of 168 634 and 186 862 paired-end raw sequences were acquired through the targeting of the V3-V4 region of the 16S rRNA gene and ITS2 region of the rRNA, respectively, from the Sargassum waste samples collected from two different beaches, Long Beach (LB: 78 552 and 82 540) and Harrismith Beach (HS: 90 082 and 104 322).After quality filtering, including trimming, paired-end reads merging, removal of the chimeric sequences and size separation, a total of 67 032 and 91 085 sequences (LB: 27 899 and 40 354; HS: 39 133 and 50 731) with an average size of 414 bp and 320 bp for V3-V4 and ITS2 regions, respectively, were retained.The retained sequences of V3-V4 and ITS2 regions were assigned to 734 and 179 amplicon sequence variants (ASVs), respectively, via the DADA2 (v.1.26) pipeline.

Prokaryotic and fungal diversity in Sargassum seaweed waste
The rarefaction curves of the reads that were used to estimate the alpha diversity (richness: S obs ; diversity: Shannon and Simpson; evenness: Pielou's evenness) indices displayed a plateau for the prokaryotes (Supplementary fig.S1) and fungi (Supplementary fig.S2) in both the sites (LB and HS) indicating that the sequencing depths were adequate for characterizing the prokaryotic and fungal communities with respect to taxonomy and predictive functions.For prokaryotes, the values of the indices, S obs , Shannon, Simpson and Pielou's evenness were recorded as 482, 4.88, 0.98 and 0.79, respectively, for LB and 359, 4.03, 0.97 and 0.68, respectively, for HS.The S obs , Shannon, Simpson and Pielou's evenness values of fungal communities were estimated as 80, 3.48, 0.75 and 0.79, respectively, for LB and 101, 1.93, 0.49 and 0.42, respectively, for HS.The beta diversity of the prokaryotic and fungal communities of LB (0.62 and 0.06) and HS (0.62 and 0.06) sites indicated that there was no significant difference.Twenty-eight prokaryotic families were detected at >1% relative abundance of ASVs in at least one sample (Fig 3).The relative abundance of the unclassified prokaryotic populations at the family level was recorded as 30.1% and 16.2% for LB and HS, respectively.The dominant bacterial families were Flavobacteriaceae (LB: 5.65% and HS: 20%), Synergistaceae (LB: 4.51% and HS: 4.58%), Rhodobacteraceae (LB: 1.78% and HS: 6.04%), Spirochaetaceae (LB: 2.68% and HS: 4.4%), Rhodospirillaceae (LB: 4% and HS: 2.63%), Phycisphaeraceae (LB: 5.13% and HS: 1.18%) and Hyphomonadaceae (LB: 1.03% and HS: 1.89%).The archaeal families were Methanosarcinaceae (LB: 2.18% and HS: 3.4%) and Methanomicrobiaceae (LB: 1.09% and HS: 4.36%).The prokaryotic communities at the genus level (>1% relative abundance of ASVs in at least one sample) were classified into 18 genera, which accounts for 32.9% of the total number of the ASVs (Fig 4).About 47% of the ASVs were not classified at the genus level, indicating the presence of novel prokaryotic lineages in Sargassum waste.The 10 most abundant bacterial genera were Muricauda, Aminobacterium, Mesorhizobium, Marinobacter, Reichenbachiella, Hyphomonas, Simiduia, Aquamicrobium, Oceanicola and Alcanivorax, with the relative abundance ranging from 1% to 16.3%.The three most abundant archaeal genera were identified as Methanosarcina, Methanogenium and Methanosaeta with the relative abundance in the range 1-3%.

Functional predictions of Sargassum waste metagenome
The categorization of the predicted metagenome profiles at the levels of functional units of gene sets resulted in nine KEGG metabolic pathway modules (Suplementary table S1).Furthermore, these nine metabolic pathway modules consisted of 61 KEGG modules (Supplementary table S1).The metabolic pathways related to the biosynthesis of terpenoids and polyketides were predicted to be the most abundant (36.9%) in Sargassum waste.The other predicted metabolic pathways were related to amino acid metabolism, the biosynthesis of other secondary metabolites, the energy metabolisms, the metabolism of cofactors and vitamins, the xenobiotic biodegradation, the nucleotide metabolisms, the carbohydrate metabolism and the lipid metabolism with an average abundance of 16.6%, 16.3%, 11.9%, 11.5%, 5.71%, 0.7%, 0.35% and 0.11%, respectively.

Discussion
Recent industrial interests in sustainable bioconversion of seaweed waste in biorefineries necessitate the characterization of their microbiome.In this study, Sargassum waste samples were collected from two different beaches in Barbados, and the taxonomic composition and functional potentials of the microbiome were assessed.The resulting alpha diversity indices indicated a high diversity of prokaryotic and fungal communities associated with Sargassum waste.The high proportion of shared ASVs and occurrence of similar predominant prokaryotic and fungal taxa suggested that the microbiome of Sargassum waste, sampled from two different beaches, were relatively similar.Furthermore, the beta diversity indices of both the sites indicated that there were no significant differences among the resulting phylogenetic lineages, as well as abundance of prokaryotic and fungal communities.The resulting ASVs of prokaryotic (734) and fungal (179) communities were classified into 12 (predominantly Proteobacteria, followed by Bacteroidetes, Planctomycetes, Synergistetes, Chloroflexi, Spirochaetae, Actinobacteria, Chlamydiae, Verrucomicrobia, Lentisphaerae, Firmicutes and Atribacteria) and five phyla (predominantly Basidiomycota, followed by Ascomycota, Zygomycota, Chytridiomycota and Glomeromycota), respectively.
The alginate lyase-producing efficacy of the prokaryotic microbiome of Sargassum waste was also evaluated.Alginate lyases have potential applications in agriculture, bioenergy, food, medical diagnostic and pharmaceutical industries (Cheng, Jiang, Xu, Liu, & Mao, 2020;Mohapatra, 2020).The results revealed that the potent alginate lyases producers belonged to the members of five phyla, Bacteroides, Proteobacteria, Planctomycetes, Verrucomicrobia and Chlorobi.Previous findings have also documented Proteobacteria and Bacteroidetes as the major phyla of alginatedepolymerizing bacteria associated with the brown algae Ascophyllum nodosum (Martin et al., 2015) and Laminaria sp (Dong et al., 2012).
In conclusion, a next-generation sequencing approach was used in this study to investigate the taxonomical compositions of prokaryotic and fungal communities along with the functions of the Sargassum waste metagenome.This cultivation-independent study demonstrated the prevalence of phylogenetically diverse phyla of bacteria, archaea and fungal communities in Sargassum waste.The results of metagenome functional analysis indicate that Sargassum waste is a promising niche material for the isolation of potent microbial strains for the production of novel antibiotic, anticancer, antitumor, immunosuppressant, antiparasite and antioxidant compounds and for the bioremediation of xenobiotics.Additionally, these microbial associates have the potential to be used as novel biocatalysts for efficient and ecofriendly depolymerization of alginate and may therefore be useful for sustainable bioconversion of Sargassum waste into thirdgeneration biofuels and novel biologically active alginate oligosaccharides of industrial importance.

Figure 1 .
Figure 1.An image showing the Sargassum waste used in this study.

Figure 2 .
Figure 2. The prokaryotic community composition of Sargassum waste at phylum level.LB, Long Beach; HS, Harrismith beach.The DNA extracted from quadruplicate Sargassum samples from each beach was pooled into one representative sample, and then used to prepare the Illumina sequencing libraries targeting the V3-V4 region of the 16S rRNA gene.

Figure 3 .
Figure 3.The prokaryotic community composition of Sargassum waste at family level.LB, Long Beach; HS, Harrismith beach.The DNA extracted from quadruplicate Sargassum samples from each beach was pooled into one representative sample, and then used to prepare the Illumina sequencing libraries targeting the V3-V4 region of the 16S rRNA gene.

Figure 4 .
Figure 4.The prokaryotic community composition of Sargassum waste at genus level.LB, Long Beach; HS, Harrismith beach.The DNA extracted from quadruplicate Sargassum samples from each beach was pooled into one representative sample, and then used to prepare the Illumina sequencing libraries targeting the V3-V4 region of the 16S rRNA gene.

Figure 5 .
Figure 5.The fungal community composition of Sargassum waste at phylum level.LB, Long Beach; HS, Harrismith beach.The DNA extracted from quadruplicate Sargassum samples from each beach was pooled into one representative sample and then used to prepare the Illumina sequencing libraries targeting the ITS2 region of the rRNA.

Figure 6 .
Figure6.The phylogenetic relatedness of alginate lyases-producing prokaryotic communities associated with Sargassum waste.ASV, amplicon sequence variant.The cladogram was drawn using maximum-likelihood algorithm.The bootstrap percentage values of 1000 replications are specified at the nodes.The scale denotes 0.05 substitutions per nucleotide position.Numbers inside the parentheses denote the relative abundance (%) of the ASVs.