J. Microbiol. Biotechnol. (2016), 26(12), 2127–2137 http://dx.doi.org/10.4014/jmb.1607.07025 Research Article Review jmb

Metagenomic and Proteomic Analyses of a Mangrove Microbial Community Following Green Macroalgae Enteromorpha prolifera Degradation S Yijing Wu1, Chao Zhao1,2*, Zheng Xiao1, Hetong Lin1, Lingwei Ruan3, and Bin Liu1,4

1College of Food Science, Fujian Agriculture and Forestry University, Fuzhou 350002, P.R. China 2Department of Chemistry, University of , Davis, CA 95616, USA 3Key Laboratory of Marine Biogenetic Resources, Third Institute of Oceanography, State Oceanic Administration, Xiamen 361005, P.R. China 4National Engineering Research Center of Juncao, Fuzhou 350002, P.R. China

Received: July 11, 2016 Revised: August 7, 2016 A mangrove microbial community was analyzed at the gene and protein levels using Accepted: August 10, 2016 metagenomic and proteomic methods with the green macroalgae Enteromorpha prolifera as the substrate. Total DNA was sequenced on the Illumina HiSeq 2000 PE-100 platform. Two- dimensional gel electrophoresis in combination with liquid chromatography tandem mass

First published online spectrometry was used for proteomic analysis. The metagenomic data revealed that the orders August 24, 2016 Pseudomonadales, Rhizobiales, and Sphingomonadales were the most prevalent in the

*Corresponding author mangrove microbial community. By monitoring changes at the functional level, proteomic Phone: +86-591-83530197; analyses detected ATP synthase and transporter proteins, which were expressed mainly by Fax: +86-591-83530197; members of the phyla Proteobacteria and Bacteroidetes. Members of the phylum E-mail: [email protected] Proteobacteria expressed a high number of sugar transporters and demonstrated specialized and efficient digestion of various glycans. A few glycoside hydrolases were detected in S upplementary data for this members of the phylum Firmicutes, which appeared to be the main cellulose-degrading paper are available on-line only at http://jmb.or.kr. bacteria. This is the first report of multiple “omics” analysis of E. prolifera degradation. These results support the fact that key enzymes of glycoside hydrolase family were expressed in pISSN 1017-7825, eISSN 1738-8872 large quantities, indicating the high metabolic activity of the community. Copyright© 2016 by The Korean Society for Microbiology Keywords: Enteromorpha prolifera, mangrove, degrading community, metagenomics, proteomics and Biotechnology

Introduction easily hydrolyzable carbohydrate and low lignin contents, enabling enzymes access to these substrates and achieve an Global energy demand is increasing, creating problems efficient hydrolysis rate [7, 33]. such as energy security, resources depletion, environmental Enteromorpha (Ulva) prolifera (Muell.) J. Agardh (, degradation, and climate change, which are expected to ), which is distributed worldwide in the intertidal increase over the next decades. Biomass is considered a zone of the sea, is the dominant species of green tide sustainable renewable energy source. The biofuel potential (Fig. 1). Enteromorpha prolifera (EP) caused the world’s in marine biomass is more than 100 EJ/year and significantly largest transregional macroalgal bloom in the Yellow Sea of higher than the values for land biomass (22 EJ/year) [7]. China. The massive green tide covered 600 km2, and more Third-generation algae-based biofuel feedstock, seaweed, than one million tons of green alga were cleaned from the and microalgae are considered to have potential for biofuel beach and coast [26]. In addition, Ireland, , Italy, and production. Algae biomass shows higher yields and Japan suffered from massive Ulva (Enteromorpha) green rates, higher photosynthetic efficiency, and lower land tides on the shoreline. EP dry matter is mainly composed of requirements [1]. Furthermore, macroalgae have greater approximately 45% water-soluble carbohydrate, 15% crude

December 2016 ⎪ Vol. 26⎪ No. 12 2128 Wu et al.

role in the determination of microbial function in a variety of environments [45, 48-50]. The use of two-dimensional gel electrophoresis (2-DE) in combination with liquid chromatography (LC)-mass spectrometry (MS)/MS approaches has become widely used in recent studies. Recently, with the availability of extensive metagenomic sequences from various marine microbial communities, metaproteomics has attracted considerable attention in the marine biology field to answer a variety of questions. This method can also be used to determine specific features of mixed marine Fig. 1. Green tide of Enteromorpha prolifera in coastal waters of the Yellow Sea near , China, in August 2014. samples and overall microbial ecosystem function. The biodegradation of different terrestrial biomasses with microbial communities has been well-documented [3], fiber, and 14% crude protein [22]. E. prolifera is a high- but the EP-degrading microbial communities from mangrove valued biomass rich in complex carbohydrate and fiber ecosystems are largely unknown. A better understanding contents. One critical step in biofuel development is of the EP-degrading bacterial communities can facilitate determining the most favorable conditions for enzymatic the development of the algal biofuels industry. The use saccharification to hydrolyze the biomass to fermentable of metagenomics and proteomics may play a key role sugars. The saccharification of EP can be advantageous in determining the molecular mechanisms by which when EP is treated with a mixture of effective enzymes. microorganisms attack and degrade the green macroalgae However, the polysaccharides present in EP have not been (Fig. 2). Therefore, the purpose of this study was to investigate fully characterized [53] and the complex structures are an EP-degrading community from mangrove sediment difficult to degrade. Mangrove ecosystems are very using an Illumina HiSeq 2000 sequencing system and LC- important for both maintaining and improving biological MS/MS. Our results provide insights into the biodegradation environments [25]. The humid subtropical climate fosters of green macroalgae. microbial diversity in mangrove sediments, and the microorganisms that thrive in mangrove environments are Materials and Methods a valuable source of novel enzymes. Metagenomics is a powerful approach for determining Enrichment of EP-Degrading Microbial Community the structure, diversity, gene content, and functional Green macroalgae EP was collected from a beach (N36°03’, composition of microbial communities. It also provides E120°20’) in Qingdao, China (Fig. 1). The soil sediments were insight into the large number of uncultured microbes [9]. sampled from a typical mangrove protection zone in the Dongzhai Next-generation sequencing (NGS), known as high- Port of Haikou, China. The mangrove sediment samples were enriched with the EP powders as the sole source of carbon and throughput sequencing, plays a key role in the analysis of energy. An EP-degrading enrichment culture was obtained and shotgun sequencing-based metagenomics [16]. The four named DZ21. The community studied was capable of digesting commercially available NGS platforms (Illumina/Solexa, the green macroalgae. The EP-degrading samples after degradation, 454/Roche, ABI/SOLiD, and Helicos BioSciences) have coated with a gold sputter coating, were dried for scanning provided unprecedented opportunities for functional genomic electron microscopic (SEM) observation (JSM-6380, Japan). research based on their ability to generate hundreds of megabases of data [32, 36]. The Illumina/Solexa Genome Total DNA Isolation and Sequencing DNA on the Illumina Analyzer is the most widely used sequencing platform in HiSeq 2000 PE-100 Platform the field and currently dominates the NGS market [31]. Total microbial genomic DNA of the EP-degrading community Illumina constructed longer and more accurate contigs and was extracted as previously described [42]. The DNA concentration scaffolds than other platforms did, and is a more appropriate was determined with a NanoDrop 2000 Spectrophotometer and then analyzed by gel electrophoresis. A highly pure (A /A = approach for metagenomic surveys [27, 30, 37]. 260 280 1.8) and large amount of total genomic DNA (200-360 ng/μl) was Metaproteomics is a new field within the “omics” generated (Fig. S1). Total DNA from the bacterial community was sciences that attempts to identify all proteins expressed in a sheared to produce a DNA fragment library for whole-genome given ecosystem. Metaproteomic studies play an important shotgun sequencing according to the manufacturer’s instructions

J. Microbiol. Biotechnol. Omics Analysis of Enteromorpha prolifera-Degrading Community 2129

Fig. 2. Schematic overview of the application of metagenomics and proteomics to study the Enteromorpha prolifera-degrading microbial community. from the TruSeq DNA sample preparation kit (Illumina, USA). contigs were used for further analysis. DNA sequencing was performed on an Illumina HiSeq 2000 sequencer, which generated 100 bp paired-end (PE) raw reads. Gene Prediction in Non-Redundant Microbial Gene Sets of Metagenomes Statistics of Sequencing Data De Novo Assembly of Illumina MetaGeneMark was used to predict open reading frames GA Short Reads (ORFs) from the assembly contigs as well as merging contigs from The obtained raw sequences were submitted to NCBI’s Sequence the sample. Next, BLAST was used to align the predicted ORFs. A Read Archive (SRA). The metagenomic sequences were deposited pair of genes with more than 95% identity over 90% of the shorter at the NCBI SRA under GenBank Accession No. SRX590675. After ORF length was grouped together. Groups sharing genes were removal of adaptor sequences, ambiguous “N” nucleotides (ratio then merged, and the longest ORF in each merged group was of “N” > 10), and low-quality sequences (Q-value ≤ 5) [24], clean used to represent the group, whereas the other members of the reads were assembled by using the Short Oligonucleotide group were considered redundant. ORFs with lengths less than Alignment Program (SOAPdenovo) [23, 24]. Raw reads and 100 bp were discarded, and the qualified ORFs were translated

December 2016 ⎪ Vol. 26⎪ No. 12 2130 Wu et al.

into amino acid sequences using the NCBI’s Genetic Codes. loaded onto 7 cm immobilized pH gradient (IPG) strips, pH 4-7. Isoelectric focusing was then performed at 20°C with a current Gene Taxonomic Assignment and Gene Functional Classification setting of 50 mA/strip as follows: the IPG strip was rehydrated at The taxonomic classification of metagenomic reads was 30 V for 12 h, and then proteins were focused successively for 1 h performed using the BLAST-based tool against the NCBI non- at 500 V, 1 h at 1,000 V, and 8 h 20 min at 8,000 V. Next, the IPG redundant (nr) protein database [4] with an E-value cut-off of 1e- strip was equilibrated in DTT and iodoacetamide buffer. For the 5. Species richness and diversity analysis based on sequencing second dimension, the proteins were run on a 12.5% SDS- reads and assembled contigs were determined using MEGAN. polyacrylamide gel at a constant current of 17 mA. Protein spots The MEGAN program uses a lowest common ancestor algorithm were stained with Coomassie brilliant blue. All experiments were based on their blast bit scores to classify reads to indicated taxa conducted in triplicates. [13]. The lowest common ancestor parameters used were min- support 5, min-score 35.0, and top-percent 10 [24]. To investigate Gel Images, Statistical Analyses, and Protein in-Gel Digestion the gene profile characteristics of the EP-degrading community, Gel images were analyzed using Melanie 3.02 software (Genebio, all sequencing reads were functionally annotated against the Switzerland). Spot detection, spot measurement, spot matching, Clusters of Orthologous Groups of proteins (COGs) database [43]. and background subtraction were performed. Only protein spots COGs were identified according to BLASTX analysis with an E- displaying reproducible change patterns and showing significant value cut-off of 1e-5. All candidate regions with a query coverage difference in up- or down-regulation, were considered to be ≥50% were retained [39]. The final sequencing reads were differentially expressed proteins for further identification. Excised assigned to the COG functional categories based on their best protein spots from 2D gels were digested as described by Oda et BLAST hits. Gene annotation was predicted in the orthology al. [34]. databases of KEGG (Kyoto Encyclopedia of Genes and Genomes) and eggNOG (evolutionary genealogy of genes: Non-supervised LC-MS/MS Analysis and Database Search Orthologous Groups). The predicted and known protein functions The target protein spots were analyzed by LC-MS/MS. The of COGs were classified into 23 categories. Using the NC-IUBMB samples were then concentrated and desalted using a reversed- Enzyme Commission (E.C.) nomenclature, enzyme data and phase C18 precolumn, and further peptide separation was metabolic pathway information were obtained by downloading achieved using a nano-HPLC reversed-phase C18 column with an relevant maps from the KEGG database. UltiMate nano-HPLC system and a 50 min linear gradient of acetonitrile. The column outlet was directly coupled to a Q Exactive Bacterial Protein Extraction and Two-Dimensional Polyacrylamide Orbitrap (Thermo Fisher Scientific, USA) mass spectrometer Gel Electrophoresis (2D-PAGE) working in the regime of data-dependent MS to MS/MS switching. The time-course dynamics of cellulose degradation by The electrospray voltage and cone voltage were 1,500 V and 30 V, community DZ21 was determined and the crude fiber had clearly respectively. The obtained peptide data from the excised spots degraded after 7 days. The whole proteins of community DZ21 were analyzed automatically by database matching against the after treatment for 1 day (D1) and 7 days (D7) were analyzed by a NCBI nr protein database using Mascot ver. 2.3 (Matrix Science, combination of 2D-PAGE and LC-MS/MS. A total of 50 mg of UK) [35]. cells was dissolved in 5 ml of ice-cold 20 mM Tris·HCl buffer

(pH 8.0) consisting of 1 mM KCl, 2 mM MgCl2 and 10mM Statistical Analysis dithiothreitol (DTT), and then sonicated for 5 min with 2 sec ON The normalized integral values were statistically analyzed using and 3 sec OFF at 180 W, three times. Next, nuclease (v/v, 100:1) SPSS ver. 13.0 for Windows (SPSS, Inc., USA). One-way analysis was added to the sample followed by incubation at 37°C for of variance was used to determine the statistically significant 30 min. Subsequently, 10% (w/v) trichloroacetic acid in ice-cold differences in selected signals among the samples. A p-value of acetone solution was added at a ratio of 20:80 (v/v), and placed at less than 0.05 was considered to indicate statistical significance. -20°C for 2 h in order to improve protein precipitation. The pellet was centrifuged at 20,000 ×g for 45 min (4°C) and washed three Results times with ice-cold solution. The supernatants were removed and 500 μl of lysis buffer consisting of 1 mM phenylmethylsulfonyl Isolation and Characterization of EP-Degrading Microbial fluoride, 2 mM EDTA, and 10 mM DTT was added to the pellet. Community from Mangrove Sediments The suspension was sonicated for 5 min with 2 sec ON and 3 sec Samples showing degradation were observed by SEM. OFF at 180 W to improve protein solubilization. Finally, the The untreated EP exhibited a rigid and ordered structure samples were centrifuged at 20,000 ×g for 30 min at 4°C and the (Fig. 3A). After treatment with the EP-degrading community resulting supernatant was subjected to isoelectric focusing. The total protein concentration was measured using the Bradford DZ21 for 7 days, the EP was severely decayed and contained assay [5]. A volume of 125 μl, corresponding to 50 μg protein, was more holes, cracks, and erosion troughs on the surface

J. Microbiol. Biotechnol. Omics Analysis of Enteromorpha prolifera-Degrading Community 2131

Fig. 3. Scanning electron microscopic observation of Enteromorpha prolifera (EP) degradation. (A) Untreated EP; (B) after treatment with the EP-degrading community DZ21 for 7 days.

(Fig. 3B). The composition changes in EP treated with the approach resulted in 17,138 large contigs with an average community were determined (Fig. S1). Total dissolved length of 1,838 bases and an N50 of 4,175 bp. When mapping solids and total soluble sugar concentrations were reduced single-end and paired-end reads simultaneously, the by 88.4% and 68.9%, respectively. The biological treatment mapping rate was 62.5%. of EP for 7 days led to 42.1% degradation of crude fiber. The yield of reducing sugar was 0.26 g/g dry EP, indicating Microbial Community Composition high potential for industrial application. To assess the species richness of this system, we used rarefaction analysis based on reads assigned to the NCBI Sequencing and Metagenomic Assembly by SOAPdenovo taxonomy. The rarefaction curve was nearly horizontal at Isolation of total community DNA using a CTAB-based 100% of the total reads, indicating that the DZ21 library extraction method with a gel filtration purification procedure was close to saturation. The reads assigned to the kingdom yielded high-quality genomic DNA (Fig. S2). Clean “reads” Bacteria accounted for 55.02% of the total reads, whereas were screened from raw sequencing reads after removing 35.56% of the total reads showed no hits in the present of adaptors, low-quality reads, and ambiguous reads. database. Based on taxon identification analysis, the most Approximately 65.9 million (6.59 G bp) clean reads were prevalent bacterial taxa at the phylum level were obtained. Metagenomic Paired-end Illumina sequence Proteobacteria (54.26% of hit-bacterial reads), Bacteroidetes reads were assembled using SOAPdenovo software. This (0.35% of hit-bacterial reads), and Actinobacteria (0.19% of

Fig. 4. Microbial phyla (A) and genera (B) compositions of DZ21 based on the reads.

December 2016 ⎪ Vol. 26⎪ No. 12 2132 Wu et al.

hit-bacterial reads). A small number of reads was assigned classified as “function unknown (S),” “energy production and to Firmicutes, Deinococcus-Thermus, and Streptophyta conversion (C),” “amino acid transport and metabolism (Fig. 4A). At the genus rank, Novosphingobium (11.24% of hit- (E),” “carbohydrate transport and metabolism (G),” and bacterial reads) was the prevalent genus in this community. “inorganic ion transport and metabolism (P)” (Fig. 5). In addition, Sphingopyxis (7.92%) and Pseudomonas (3.37%) According to the BLASTX results of total contigs against were also present (Fig. 4B). Following the most predominant the CAZy database, we detected 22 homologous genes species Pseudomonas stutzeri (53.07%), the top five species classified into 13 different glycoside hydrolase (GH) were Sphingopyxis alaskensis (16.91%), Ochrobactrum intermedium families, mainly belonging to GH13, GH51, GH10, GH1, (12.56%), Ochrobactrum anthropi (3.42%), and Bordetella sp. GH53, and GH43, based on the contig aligned number (2.89%). (Table S1). In the GH family, enzymes involved in the hydrolysis of cellulosic materials were detected, including Gene Function Annotation and Classification cellulase (E.C. 3.2.1.4), beta-1,4-xylanase (E.C. 3.2.1.8), beta- Algae and land plants differ in cell wall composition. The glucosidase (E.C. 3.2.1.21), alpha-L-arabinofuranosidase key difference is the low concentration in the presence or (E.C. 3.2.1.55), and alpha-glucuronidase (E.C. 3.2.1.139) absence of lignin in macroalgal feedstocks. The chemical (Table S2). Most were found to belong to Pseudomonas. structures and contents of carbohydrates in green algae and land plants also differ significantly. From the Two-Dimensional Electrophoresis and Identified Proteins perspective of biomass conversion, appropriate enzymes The metaproteome profiles of the microbial community must be introduced to liberate biomass for hydrolysis. The were detected by two-dimensional gel electrophoresis composition of the enzymatic mixture largely depends on (Fig. 6). The use of pH 4–7 IPG strips resulted in well- the biomass origin. As a potential biomass material, EP has spread protein spots, which enabled accurate excision and practical values in the biomass energy, food, and medicinal image identification. All spots were separated and showed industries. To obtain a metabolic profile of community relative molecular masses of 19–99 kDa (Fig. 6). A total of DZ21, all contigs were annotated using the BLASTX tool 20 relevant protein spots were excised from the stained 2D based on the COG, KEGG, and CAZy databases. The genes gel and digested with trypsin, and the collected peptides annotated by the COG database were classified into 23 were analyzed by MS/MS. These protein spots were COG functional categories. A large number of contigs were identified and further characterized to determine their

Fig. 5. Clusters of orthologous groups (COG) functional classification of the community DZ21 metagenome. Of the 67,497 hits in the nr database, 14,190 unigenes with significant homologies in the COG database (E-value ≤ 1e-5) were classified into 23 COG categories.

J. Microbiol. Biotechnol. Omics Analysis of Enteromorpha prolifera-Degrading Community 2133

Fig. 6. Two-dimensional gel electrophoresis of proteins extracted from D1 (A) and D7 (B). The proteins were separated on a 7 cm immobilized pH gradient strip with a linear gradient of pH 4–7. Numbered circles indicate spots showing an expression level change between D1 and D7. The proteins are listed in Table 1.

correlations with bioinformatic databases. The peptide of biomass materials [53]. However, little information is mass peaks were compared with those in the NCBI database. available regarding macroalgae degradation in natural The protein identification data, including GenBank ID, environments, particularly mangrove environments. molecular weight, isoelectric point value, Mascot score, Novosphingobium species are an important source of cell wall and function are listed in Table 1. All digested spots polysaccharide-degrading enzymes such as rhamnosidase present in the gel were evaluated in triplicates. [14]. Moreover, the water-soluble polysaccharide of EP contained mainly rhamnose, galactose, and xylose [52]. In Discussion addition, Novosphingobium was reported to affect lignin biodegradation [6]. Most Sphingopyxis species are capable During the erosion process, mangrove-degrading bacteria of expressing lignocellulolytic glycoside hydrolases [40, affixed themselves to the surface of the matrix and colonized 41]. There are previous reports of Pseudomonas strains with on the surface to form a strong and compact network over fiber- and polysaccharide-degradation abilities. Pseudomonas the exterior surface, after which the bacteria secreted sp. has been reported to degrade lignin [38]. Bacterial extracellular enzymes to degrade EP. This may be the main laccase from P. stutzeri has been employed in lignocellulosic cause of EP degradation. Deswal et al. [8] employed crude fibers [20]. Additionally, Maalej et al. [28, 29] demonstrated enzymes for saccharification of straws, and the enzymolysis the production and biochemical characterization of a high reducing sugar yields of wheat straw and rice straw as maltotetraose-producing amylase from P. stutzeri. Ochrobactrum substrates were 0.214 and 0.157 g/g dry mass, respectively. species (α-Proteobacteria) were found to be present in the As expected, the biomass conversion rate was higher gut of the wood-feeding higher termite Zootermopsis because of the low contents of hemicellulose and lignin. angusticollis [47]. There are also reports of Ochrobactrum NGS technology offers a new and rapid approach for strains with lignin degradation ability [44]. Bordetella is genome-wide analysis. To characterize microbial community beneficial in oxygen consumption and pH neutralization in DZ21, including its genome structure, gene content, the cellulose-degrading community [17, 18]. Bordetella is metabolic capabilities, and role of specific organisms for EP generally considered to be independent of the substrates degradation, bacterial sequences within the metagenomic derived from cellulose degradation, such as saccharides. data were identified using the Illumina HiSeq 2000 Our results are similar to those of previous reports and sequencing technology. The microbial composition of DZ21 indicate that bacteria of Sphingopyxis and Ochrobactrum are was analyzed based on mapped reads using MEGAN. It is key EP degraders and function together with other degrading well known that one bacterial community shows greater bacteria in the mangrove community. Prospecting macroalgae efficiency compared with a single strain in the biodegradation as biomass for bioconversion is limited primarily by the

December 2016 ⎪ Vol. 26⎪ No. 12 2134 Wu et al.

Table 1. Proteins originating from microorganisms identified by MALDI TOF/TOF MS. Mascot Gene ID Matching Theoretical Theoretical Expression Spot Protein score Best-matched organism in NCBI peptides MW kDA pI level (MS/MS) 108 ATP synthase subunit beta 119386525 117 2 Paracoccus denitrificans PD1222 50.34 4.78 ↓ 298 ATP synthase subunit beta 77964181 192 5 Clostridium paradoxum 50.46 4.99 ↑ 311 Elongation factor complex 1942721 97 2 Escherichia coli 42.18 5.22 ↓ Ef-TuEF-Ts 320 Hypothetical protein 85709719 273 15 Erythrobacter sp. NAP1 26.50 5.03 ↑ NAP1_12578 328 Superoxide dismutase 85374874 285 15 Erythrobacter litoralis 22.34 4.89 ↑ HTCC2594 335 Elongation factor Tu 85374468 94 3 Erythrobacter litoralis 43.06 4.80 ↑ HTCC2594 364 H(+)-transporting ATP 89889622 170 8 Flavobacteria bacterium BBFL7 54.12 4.81 ↓ synthase beta chain 369 Transcriptional regulatory 86144202 364 17 Leeuwenhoekiella blandensis 27.89 5.25 ↑ protein MED217 395 Enoyl-(acyl-carrier-protein) 305667240 794 36 Maribacter sp. HTCC2170 29.59 5.69 ↑ reductase 429 ATP synthase subunit beta 15604633 252 14 Rickettsia prowazekii Madrid E 51.23 4.85 ↓ 456 N,N’-diacetylchitobiose 10834609 53 2 Vibrio furnissii 89.67 5.09 ↓ phosphorylase 457 Putative enoyl-(acyl- 86144197 247 6 Leeuwenhoekiella blandensis 30.17 5.61 ↑ carrier-protein) reductase MED217 489 N,N’-diacetylchitobiose 10834609 48 1 Vibrio furnissii 89.67 5.09 ↓ phosphorylase 589 Translation elongation 85703985 789 34 Roseovarius sp. 217 42.97 5.15 ↓ factor Tu 592 Elongation factor EF-Tu 114704463 820 56 Fulvimarina pelagi HTCC2506 42.62 5.16 ↓ protein 638 Translation elongation factor 85709060 595 32 Erythrobacter sp. NAP1 42.33 4.80 ↓ 692 Translation elongation 118591186 173 15 Stappia aggregata IAM 12614 43.04 4.92 ↓ factor Tu 711 OmpA/MotB 103487471 98 2 Sphingopyxis alaskensis RB2256 39.08 4.49 ↑ 722 Elongation factor Tu 260063564 1222 63 Robiginitalea biformata 43.21 4.98 ↑ HTCC2501 788 F0F1 ATP synthase 119503555 2480 91 Gamma proteobacterium 50.12 4.71 ↓ subunit beta HTCC2080

availability of tractable microorganisms that can metabolize dependent. Based on previous studies, most Pseudomonas alginate [46]. Enzymatic treatment of algae has been species are capable of fermenting various carbohydrates. reported for Laminaria and was successfully saccharified However, the involvement of Pseudomonas species in EP using an enzymatic mixture. degradation has not been reported and their function in EP The annotation and classification of functional genes is degradation remains unknown. Although genetically very important for understanding the mechanism of identical, the species possess significant discrepancies and community DZ21 degradation. The types and contents of can adapt under different fermentation conditions [21]. The heteropolysaccharides from algae are largely species- bottleneck of biofuel production is the effective conversion

J. Microbiol. Biotechnol. Omics Analysis of Enteromorpha prolifera-Degrading Community 2135

of algal feedstocks into glucose [2]. In our study, important & JK2014014). The project was also supported by FAFU enzymes of the metabolic pathways involved in biodegradation grants XJQ201417 and 612014043. were detected in association with Pseudomonas in KEGG pathway analysis, such as xylan 1,4-beta-xylosidase (E.C. References 3.2.1.37), alpha-N-arabinofuranosidase (E.C. 3.2.1.55), pectate lyase (E.C. 3.1.1.11), cellulase (E.C. 3.2.1.4), beta-glucosidase 1. Adenle AA, Haslam GE, Lee L. 2013. Global assessment of (E.C. 3.2.1.21), and beta-mannanase (E.C. 3.2.1.4), indicating research and development for algae biofuel production and that Pseudomonas species likely contribute to EP degradation its potential role for sustainable development in developing during the fermentation (Table S2). countries. Energ. Policy 61: 182-195. In the analysis of the obtained proteome, the proteins 2. Balat M. 2011. Production of bioethanol from lignocellulosic materials via the biochemical pathway: a review. Energy identified in the metaproteome of community DZ21 had Convers. Manage. 52: 858-875. diverse functional activities related to glycolysis, protein 3.Barnard D, Casanueva A, Tuffin M, Cowan D. 2010. biosynthesis, oxidation-reduction, and ATP hydrolysis Extremophiles in biofuel synthesis. Environ. Technol. 31: 871- coupled to proton transport. The results were similar to 888. those of previous studies [11, 12]. Among the proteins 4. Benson D, Karsch-Mizrachi I, Lipman D, Ostell J, Wheeler encoded by housekeeping genes, such as ATP synthases, D. 2006. GenBank. Nucleic Acids Res. 34: D16-D20. elongation factors, ribosomal proteins, or aminoacyl-tRNA 5. Bradford MM. 1976. A rapid and sensitive method for the synthetases, as well as proteins involved in carbohydrate quantitation of microgram quantities of protein utilizing the metabolism [15, 19, 54], most species homologies were principle of protein-dye binding. Anal. Biochem. 72: 248-254. affiliated with Sphingopyxis or other genera of the 6. Chen Y, Chai L, Tang C, Yang Z, Zheng Y, Shi Y, Zhang H. Proteobacteria. Elongation factor Tu plays an important 2012. Kraft lignin biodegradation by Novosphingobium sp. B- role in the elongation cycle of protein synthesis [10, 51]. 7 and analysis of the degradation process. Bioresour. Technol. 123: 682-685. Although the proteome analysis of community DZ21 is 7. Chynoweth DP, Owens JM, Legrand R. 2001. Renewable limited to species with high cell numbers or large amounts methane from anaerobic digestion of biomass. Renew. Energy of biomass, this method generated valuable data that 22: 1-8. provided insight into E. prolifera degradation. 8. Deswal D, Khasa YP, Kuhad RC. 2011. Optimization of In summary, this is the first report of multiple “omics” cellulase production by a brown rot fungus Fomitopsis sp. analyses of marine macroalgae E. prolifera degradation. RCK2010 under solid state fermentation. Bioresour. Technol. Microalgae and brown seaweed have been evaluated for 102: 6065-6072. biofuel production because of their higher photosynthetic 9. Deutschbauer AM, Chivian D, Arkin AP. 2006. Genomics efficiency, biomass yield, and growth rates. However, for environmental microbiology. Curr. Opin. Biotechnol. 17: despite the high contents of carbohydrate and crude fiber 229-235. in E. prolifera, its utilization as an energy source remains 10. Eargle J, Black AA, Sethi A, Trabuco LG, Luthey-Schulten Z. low worldwide. In this study, we conducted an in-depth 2008. Dynamics of recognition between tRNA and elongation factor Tu. J. Mol. Biol. 377: 1382-1405. analysis of the structure and function of and E. prolifera- 11. Hanreich A, Schimpf U, Zakrzewski M, Schlüter A, Benndorf degrading microbial community. Our results provide a D, Heyer R, et al. 2013. Metagenome and metaproteome foundation for determining the mechanism of functional analyses of microbial communities in mesophilic biogas- genes for degradation and provide a theoretical foundation producing anaerobic batch fermentations indicate concerted and technical support for constructing genetic engineering plant carbohydrate degradation. Syst. Appl. Microbiol. 36: bacteria containing key enzymes. It can also suggest new 330-338. methods of resolving marine ecological security issues 12. Heyer R, Kohrs F, Benndorf D, Rapp E, Kausmann R, caused by the marine macroalgae E. prolifera. Heiermann M, et al. 2013. Metaproteome analysis of the microbial communities in agricultural biogas plants. Nat. Acknowledgments Biotechnol. 30: 614-622. 13. Huson DH, Auch AF, Qi J, Schuster SC. 2007. MEGAN This work was financially supported by the National analysis of metagenomic data. Genome Res. 17: 377-386. 14. Izzo V, Tedesco P, Notomista E, Pagnotta E, Donato AD, Natural Science Foundation of China (41306181) and Trincone A, Tramice A. 2014. a-Rhamnosidase activity in the Science Foundation of Fujian Province, China (2016J06009 marine isolate Novosphingobium sp. PP1Y and its use in the

December 2016 ⎪ Vol. 26⎪ No. 12 2136 Wu et al.

bioconversion of flavonoids. J. Mol. Catal. B Enzym. 105: 95- stutzeri AS22. Biomed. Res. Int. 2014: 156438. 103. 29. Maalej H, Hmidet N, Ghorbel-Bellaaj O, Nasri M. 2013. 15. Jehmlich N, Kleinsteuber S, Vogt C, Benndorf D, Harms H, Purification and biochemical characterization of a detergent Schmidt F, et al. 2010. Phylogenetic and proteomic analysis stable a-amylase from Pseudomonas stutzeri AS22. Biotechnol. of an anaerobic toluene-degrading community. J. Appl. Bioprocess Eng. 18: 1-10. Microbiol. 109: 1937-1945. 30. Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, 16. Jia B, Xuan L, Cai K, Hu Z, Ma L, Wei C. 2013. NeSSM: a Bemben LA, et al. 2005. Genome sequencing in microfabricated next-generation sequencing simulator for metagenomics. high-density picolitre reactors. Nature 437: 376-380. PLoS One 8: e75448. 31. Metzker ML. 2010. Sequencing technologies - the next 17. Kato S, Haruta S, Cui ZJ, Ishii M, Igarashi Y. 2004. Effective generation. Nat. Rev. Genet. 11: 31-46. cellulose degradation by a mixed-culture system composed 32. Morozova O, Marra MA. 2008. Applications of next- of a cellulolytic Clostridium and aerobic non-cellulolytic generation sequencing technologies in functional genomics. bacteria. FEMS Microbiol. Ecol. 51: 133-142. Genomics 92: 255-264. 18. Kato S, Haruta S, Cui ZJ, Ishii M, Igarashi Y. 2005. Stable 33. Nkemka VN, Murto M. 2010. Evaluation of biogas production coexistence of five bacterial strains as a cellulose-degrading from seaweed in batch tests and in UASB reactors combined community. Appl. Environ. Microbiol. 71: 7099-7106. with the removal of heavy metals. J. Environ. Manage. 91: 19. Kohrs F, Heyer R, Magnussen A, Benndorf D, Muth T, 1573-1579. Behne A, et al. 2014. Sample prefractionation with liquid 34. Oda K, Kakizono D, Yamada O, Iefuji H, Akita O, Iwashita isoelectric focusing enables in depth microbial metaproteome K. 2006. Proteomic analysis of extracellular proteins from analysis of mesophilic and thermophilic biogas plants. Anaerobe Aspergillus oryzae grown under submerged and solid-state 29: 59-67. culture conditions. Appl. Environ. Microbiol. 72: 3448-3457. 20. Kumar A, Vanamala A, Kumar R. 2005. Exploration of 35. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS. 1999. bacterial laccase in Pseudomonas stutzeri and its application Probability-based protein identification by searching sequence in bleaching the wood pulp. FEBS J. 272(s1)(N6-008P). databases using mass spectrometry data. Electrophoresis 20: 21. Li A, Chu Y, Wang X, Ren L, Yu J, Liu X, et al. 2013. A 3551-3567. pyrosequencing-based metagenomic study of methane-producing 36. Quail MA, Kozarewa I, Smith F, Scally A, Stephens PJ, microbial community in solid-state biogas reactor. Biotechnol. Durbin R, et al. 2008. A large genome center’s improvements Biofuels 6: 3. to the Illumina sequencing system. Nat. Methods 5: 1005-1010. 22. Li DM, Chen LM, Zhao JS, Zhang XW, Wang QY, Wang 37. Quince C, Lanzén A, Curtis TP, Davenport RJ, Hall N, Head HX, Ye NH. 2010. Evaluation of the pyrolytic and kinetic IM, et al. 2009. Accurate determination of microbial diversity characteristics of Enteromorpha prolifera as a source of from 454 pyrosequencing data. Nat. Methods 6: 639-641. renewable bio-fuel from the Yellow Sea of China. Chem. 38. Sharafi-Yazdi MK, Azimi C, Khalili MB. 2001. Isolation and Eng. Res. Des. 88: 647-652. identification of bacteria present in the activated sludge 23. Li R, Yu C, Li Y, Lam TW, Yiu SM, Kristiansen K, Wang J. unit, in the treatment of industrial waste water. Iranian J. 2009. SOAP2: an improved ultrafast tool for short read Publ. Health 30: 91-94. alignment. Bioinformatics 25: 1966-1967. 39. She R, Chu JS, Wang K, Pei J, Chen N. 2009. GenBlastA: 24. Li X, Cui Z, Liu Y, Song C, Shi G. 2013. Transcriptome enabling BLAST to identify homologous gene sequences. analysis and discovery of genes involved in immune pathways Genome Res. 19: 143-149. from hepatopancreas of microbial challenged mitten crab 40. Shin KC, Oh DK. 2014. Characterization of a novel Eriocheir sinensis. PLoS One 8: e68233. recombinant β-glucosidase from Sphingopyxis alaskensis that 25. Liao QY, Li J, Zhang JH, Li M, Lu Y, Xu RL. 2009. An specifically hydrolyzes the outer glucose at the C-3 position ecological analysis of soil sarcodina at Dongzhaigang mangrove in protopanaxadiol-type ginsenosides. J. Biotechnol. 172: 30-37. in Hainan Island, China. Eur. J. Soil Biol. 45: 214-219. 41. Simmons CW, Reddy AP, D’Haeseleer P, Khudyakov J, 26. Liu DY, Keesing JK, Xing QG, Shi P. 2009 World’s largest Billis K, Pati A, et al. 2014. Metatranscriptomic analysis of macroalgal bloom caused by expansion of seaweed aquaculture lignocellulolytic microbial communities involved in high- in China. Mar. Pollut. Bull. 58: 888-895. solids decomposition of rice straw. Biotechnol. Biofuels 7: 495. 27. Luo C, Tsementzi D, Kyrpides N, Read T, Konstantinidis 42. Syn CK, Swarup S. 2000. A scalable protocol for the KT. 2012. Direct comparisons of Illumina vs. Roche 454 isolation of large-sized genomic DNA within an hour from sequencing technologies on the same microbial community several bacteria. Anal. Biochem. 278: 86-90. DNA sample. PLoS One 7: e30087. 43. Tatusov RL, Galperin MY, Natale DA, Koonin EV. 2000. The 28. Maalej H, Ben Ayed H, Ghorbel-Bellaaj O, Nasri M, Hmidet N. COG database: a tool for genome-scale analysis of protein 2014. Production and biochemical characterization of a high functions and evolution. Nucleic Acids Res. 28: 33-36. maltotetraose (G4) producing amylase from Pseudomonas 44. Taylor CR, Hardiman EM, Ahmad M, Sainsbury PD, Norris

J. Microbiol. Biotechnol. Omics Analysis of Enteromorpha prolifera-Degrading Community 2137

PR, Bugg TD. 2012. Isolation of bacterial strains able to to a mixed community of prokaryotic microorganisms. metabolize lignin from screening of environmental samples. Environ. Microbiol. 6: 911-920. J. Appl. Microbiol. 113: 521-530. 50. Wilmes P, Wexler M, Bond PL. 2008. Metaproteomics 45. Toyoda A, Iio W, Mitsumori M, Minato H. 2009. Isolation provides functional insight into activated sludge wastewater and identification of cellulose-binding proteins from sheep treatment. PLoS One 3: e1778. rumen contents. Appl. Environ. Microbiol. 75: 1667-1673. 51. Yu H, Chan YL, Wool IG. 2009. The identification of the 46. Wargacki AJ, Leonard E, Win MN, Regitsky DD, Santos CN, determinants of the cyclic, sequential binding of elongation Kim PB, et al. 2012. An engineered microbial platform for factors Tu and G to the ribosome. J. Mol. Biol. 386: 802-813. direct biofuel production from brown macroalgae. Science 52. Zhao C. 2014. Purification and composition analysis of 335: 308-313. polysaccharide from edible seaweed Enteromorpha prolifera and 47. Wenzel M, Schönig I, Berchtold M, Kämpfer P, König H. polysaccharides depolymerized enzymes from microorganisms. 2002. Aerobic and facultatively anaerobic cellulolytic bacteria Res. J. Biotechnol. 9: 30-36. from the gut of the termite Zootermopsis angusticollis. J. Appl. 53. Zhao C, Ruan LW. 2011. Biodegradation of Enteromorpha Microbiol. 92: 32-40. prolifera by mangrove degrading micro-community with 48. Williams MA, Taylor EB, Mula HP. 2010. Metaproteomic physical-chemical pretreatment. Appl. Microbiol. Biotechnol. characterization of a soil microbial community following 92: 709-716. carbon amendment. Soil Biol. Biochem. 42: 1148-1156. 54. Zhou J, Liu L, Shi Z, Du G, Chen J. 2009. ATP in current 49. Wilmes P, Bond PL. 2004. The application of two-dimensional biotechnology: regulation, applications and perspectives. polyacrylamide gel electrophoresis and downstream analyses Biotechnol. Adv. 27: 94-101.

December 2016 ⎪ Vol. 26⎪ No. 12