Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=labt20

Animal Biotechnology

ISSN: 1049-5398 (Print) 1532-2378 (Online) Journal homepage: https://www.tandfonline.com/loi/labt20

Exploration of the rumen microbial diversityand carbohydrate active enzyme profile of blackBengal goat using metagenomic approach

Prashant R. Suryawanshi, Chandan Badapanda, Krishna M. Singh & AnkitaRathore

To cite this article: Prashant R. Suryawanshi, Chandan Badapanda, Krishna M. Singh &Ankita Rathore (2019): Exploration of the rumen microbial diversity and carbohydrate activeenzyme profile of black Bengal goat using metagenomic approach, Animal Biotechnology, DOI:10.1080/10495398.2019.1609489

To link to this article: https://doi.org/10.1080/10495398.2019.1609489

Published online: 12 May 2019.

Submit your article to this journal

View Crossmark data

Exploration of the rumen microbial diversity and carbohydrate activeenzyme profile of black Bengal goat using metagenomic approach

Prashant R. Suryawanshia�, Chandan Badapandab, Krishna M. Singhc, and Ankita Rathoreb

aDepartment of Veterinary Microbiology, College of Veterinary Sciences & Animal Husbandry, Agartala, India; bBioinformatics Division,Xcelris Labs Limited, Ahmedabad, India; cMolecular Biology Department, Unipath Specialty Laboratory Ltd., Ahmedabad, India

ABSTRACTBlack Bengal goats possess a rich source of rumen microbiota that helps them to adapt forthe better utilization of plant biomaterial into energy and nutrients, a task largely performedby enzymes encoded by the rumen microbiota. Therefore the study was designed in orderto explore the taxonomic profile of rumen microbial communities and potential biomassdegradation enzymes present in the rumen of back Bengal goat using Illumina Nextseq-500platform. A total of 83.18 million high-quality reads were generated and bioinformatics ana-lysis was performed using various tools and subsequently, the predicted ORFs along withthe rRNA containing contigs were then uploaded to MG-RAST to analyze taxonomic andfunctional profiling. The results highlighted that Bacteriodetes (41.38–59.74%) were themost abundant phyla followed by Firmicutes (30.59–39.96%), Proteobacteria (5.07–7.61%),Euryarcheaota (0.71–7.41%), Actinobacteria (2.05–2.75%). Genes that encode glycosidehydrolases (GHs) had the highest number of CAZymes, and accounted for (39.73–37.88%) ofall CAZymes in goat rumen. The GT families were the second-most abundant in CAZymes(23.73–23.11%) and followed by Carbohydrate Binding module Domain (17.65–15.61%),Carbohydrate Esterase (12.90–11.95%). This study indicated that goat rumen had complexfunctional microorganisms produce numerous CAZymes, and that can be further effectivelyutilised for applied ruminant research and industry based applications.

KEYWORDSSequencing; Genomics;metagenome; black Bengalgoat; rumen microbiota;MGRAST; CAZymes

Introduction

Black Bengal goats make up a significant proportionof the domesticated animal species in the state ofTripura, India and goats are fed on open pasturelands. The FAOSTAT database from 2013 reportsover a 468 Million milk producing ruminants (buffalo,goat and sheep). In India, goats play a very crucialrole solely contributing for 3.5% of total milk produc-tion of 165 Million tonnes and 14.22% of total meatproduction of 74 Million tonnes of all Indian livestockpopulation in the year 2016–17.1 Therefore, under-standing how ruminants convert their plant fibresinto food products, such as milk, meat, energy, is ofobvious importance.

The rumen is a complex ecosystem that harbors awide variety of microorganisms, including bacteria,protozoa, archaea, and fungi.2 The symbiosis betweenanimal and microbe in the rumen allows for a

cooperative system in which both the host andmicrobes derive a benefit.3 The rumen is a capaciouspre-gastric fermentation chamber that sustains a richcommunity of microorganisms that rapidly colonizeand digest feed particles. Improvement in the abilityof the rumen microbiota to degrade plant cell wall isgenerally highly desirable and usually leads toimproved animal performance.4 Carbohydrate poly-mers in plants are indigestible to most animals butcan be hydrolyzed and fermented by mutualistic inter-actions between different functional groups includinghydrolytic, acidogenic, acetogenic bacteria, methano-genic archaea in the rumen for efficient digestion.5,6

The end products of this fermentation are volatilefatty acids (VFAs), which form a major metabolic fuelfor the ruminant, and microbial cells that are a majorsource of protein and amino acids when absorbed inthe lower digestive tract of animals.7

CONTACT Chandan Badapanda Chandan.bioinfo@gmail.com Bioinformatics Division, Xcelris Labs Limited, Old Premchand Nagar Road, Ahmedabad380015, Gujarat, India�Present Address: Department of Veterinary Microbiology, College of Veterinary and Animal Sciences, Parbhani, India.Color versions of one or more figures in the article can be found online at http://www.tandfonline.com/labt.� 2019 Taylor & Francis Group, LLC.

ANIMAL BIOTECHNOLOGYhttps://doi.org/10.1080/10495398.2019.1609489

Few studies were previously performed on goatseither at the 16S level or at the metagenomic level todecipher rumen microbiota at small scale.8–12 So faraccording to our knowledge, this is the first extensivereport on the exploration of rumen microbiota ana-lysis on black Bengal goat considering six rumen sam-ples, with age groups ranging from 3–12 months. Inour group, we have also developed and analyzed dif-ferent environmental samples through integratedmetagenomics or metatranscriptomics pipeline.13–16

The microbial world is considered the largest reser-voir of genes encoding enzymatic catalysts; however,this wealth of catalytic potential remains largely unex-plored because more than 99% of the microbes acrossdifferent habitats cannot be grown in the laboratory.17

Handelsman18 introduced the concept of‘metagenomics’, a culture-independent approachwhich involves genomic analysis of DNA extractedfrom its natural environment and is a more inclusivestrategy to access the total microbial genetic reservoiras compared to the culture-dependent approaches. Itdiffers from traditional genomic library constructionin that the cloned DNA does not originate from a sin-gle known microbe, but rather from the entire popu-lation in an environmental sample.

During recent times, the rumen metagenomicsstudies have revealed the vast diversity of fibrolyticenzymes, multiple domain proteins, and the complex-ity of microbial composition in the ecosystem.19,20

A full-length phytase gene, designated RPHY1 wasidentified from Prevotella species present in Mehsanibuffalo rumen and Phytases have been widely used asanimal feed supplements to increase the availability ofdigestible phosphorus, especially in monogastric ani-mals fed cereal grains.21 The rumen is a dynamicsconsortium of microorganisms that harbor the com-plex lignocellulosic degradation machinery for diges-tion of plant biomass. Evidence also suggests that themost important organisms and gene sets involved inthe most efficient hydrolysis of plant cell wall areassociated with the fiber portion of the rumendigesta.22 Predominant ruminal hemicellulose-digest-ing bacteria such as Butyrivibrio fibrisolvens and

Prevotella ruminicola degrade xylan and pectin andutilize the degraded soluble sugars.23 The density ofgenes for glycoside hydrolases in the camel rumenmetagenome was estimated to be 25 per Mbp ofassembled DNA, which is significantly greater thancow rumen. They also identified large sequences ofencoding scaffoldins, dockerins, and cohesins, indicat-ing the potential for cellulosome-mediated lignocellu-lose degradation and these sequences, associated withBacteroidetes and Firmicutes.24 Therefore, the studywas undertaken to explore the uncultured rumenmicrobiota at taxonomic level as well as to establishthe microbial gene function such as CAZymes presentin rumen of black Bengal goats that can be furthereffectively utilized for applied ruminant research andindustry application.

Materials and methods

Sample collection

The female healthy goats were selected for rumen sam-ples collection from G. B. Hospital A{23.859412�N}{91.291206�W} and Bus Motor stand{23.831457 N}{91.286778�W} local goat slaughter area in Agartala,West Tripura district. The rumen samples were col-lected in 100ml sterile sample container which wastransferred to an icebox and all the samples were keptat �20� C before metagenomes sequencing. A total ofsix rumen samples were collected from the femaleblack Bengal goats of different age groups starting from3 months to 1 year of age from Agartala, West Tripuradistrict of Tripura, India during November 2014 toJanuary 2015 is represented in Table 1. Black Bengalgoats are fed on open pasture lands in the state ofTripura, India. Common feeds and fodders of blackBengal goats in Tripura are tree/shrub leaves (Babul,neem, mango etc.), cereal fodder (maize, oat, etc.),green grasses (Para, Guinea, etc), leguminous fodder(cowpea), leguminous pasture (Stylosanthes), leaves ofdifferent vegetables (Cauli flower, cabbage leaves, etc.)and dry fodders (dry leaves of trees).

Table 1. Data of goat rumen samples collected from different places of west Tripura districts fromNovember 2014 to January 2015.Sample code Age Sex Weight Location

A 3 months Female 3.75 kg West Tripura, G. B. Hospital, A{23.859412�N} {91.291206�W}B 4 months Female 5.50 kg West Tripura, G. B. Hospital

A{23.859412�N} {91.291206�W}C 5 months Female 6.15 kg West Tripura, G. B. Hospital A{23.859412�N} {91.291206�W}D 6 months Female 7.35 kg West Tripura, Bus Motor stand{23.831457�N} {91.286778�W}E 9 months Female 9.50 kg West Tripura, Bus Motor stand{23.831457�N} {91.286778�W}H 12 months Female 10.50 kg West Tripura, Bus Motor stand{23.831457�N} {91.286778�W}

2 P. R. SURYAWANSHI ET AL.

Genomic DNA isolation

The DNeasy Power Soil Kit from MoBio was used toisolate microbial genomic DNA using InhibitorRemoval Technology (IRT). IRT eliminates humicacid content leaving highly pure DNA for downstreamapplications analysis. Cell lysis was done by mechan-ical and chemical methods. DNA was captured on asilica membrane in a spin column format followed bywashing and elution of DNA (as per the manufac-turer’s instructions, QIAGEN Catalogue number12888.100). DNA was subjected to qualitative andquantitative analysis. Quality was checked on 0.8%agarose gel (loaded 5 ll). The gel was run at 110V for30 mins. 1 ml of each sample was loaded in Nanodrop8000 for determining A260/280 ratio, and 1 ml of eachsample was used for determining concentration usingQubitVR 2.0 Fluorometer.

Library preparation and sequencing

The TruSeq Nano DNA HT Kit was used to constructlibraries from the isolated DNA (Illumina’s instruc-tions). Using a single enzymatic ‘tagmentation’ reac-tion, the Nextera transposome simultaneouslyfragmented and tagged the DNA with unique adaptersequences. Limited-cycle PCR was used to amplify thetagged DNA and add sequencing indexes. The ampli-fied library was analyzed in Bioanalyzer 2100 (AgilentTechnologies) using High Sensitivity (HS) DNA chipas per manufacturer’s instructions. Using this stream-lined workflow, 6 DNA libraries were prepared forsequencing on the NextSeq 500 desktop sequencer. All6 libraries were pooled together for cluster generationand sequencing. Libraries were loaded onto a reagentcartridge and clustered on the NextSeq 500 System. Apaired-end, 2� 150 bp sequencing run was performedusing the NextSeq 500 High-Output Kit.

The metagenomic sequencing for six black Bengalgoat sample data was done on the Illumina NextSeq500 platform with 2� 150bp chemistry. Table 2 repre-sents the Sequence details of six metagenomic samples

Metagenome assembly

The quality of raw reads generated from Nextseq500platform was checked using the fastqc tool. Adaptersequences, bases having quality score �20 and readshaving read length less than 40 were trimmed usingTrimmomatic (v0.36)25 to get a high-quality data. Denovo assembly of high-quality reads were performedto generate contigs using CLC Genomics Workbench6.0 at default parameters (Minimum contig length:500, Automatic word size: Yes, Perform scaffolding:Yes, Mismatch cost: 2, Insertion cost: 3, Deletion cost:3, Length fraction: 0.5, Similarity fraction: 0.8).

Gene prediction and gene annotation

Using prodigal (v2.6.3)26 coding genes were predicted.Blastx against NCBI ‘NR’ database was executedlocally considering E-value cut off of 1� 10�5.

rRNAs were predicted by aligning contigs againstRNA database (NCBI) using BlastN. Subsequently, thepredicted ORFs along with the rRNA containing con-tigs were then uploaded to MG-RAST web server(with default parameters) to analyze taxonomic andfunctional profiling of the six rumen metagenomicsamples of goats. The stand-alone analysis tool,MEGAN,27 assigns a read with hits in multiplegenomes to their lowest common ancestor (LCA) inthe NCBI taxonomy. KEGG analysis was achieved byusing MEGAN. dbCAN is a web server and DataBasefor automated Carbohydrate-active enzymeANnotation (http://csbl.bmb.uga.edu/dbCAN/) andwas used to identify carbohydrate-active enzymes inour samples with E-value cut off of 1� 10�5.

Results and discussion

Metagenome sequencing of 6 samples usingNextseq500 platform resulted into a total of 83.18 mil-lion high quality reads. The sequencing data alongwith NCBI-SRA ID, MGRAST accession number isprovided in Table 2. Based on the MG-RAST annota-tion, 259, 132, 87, 68, 81, 98 rRNA were predictedfrom Sample A, B, C, D, E, H respectively based onhits against the 16S rRNA gene sequence databases

Table 2. Sequence details of six metagenomic samples.Sample SRA ID MG-RAST accession HQ reads Initial sequences QC passed sequences rRNA sequences Post QC bp count Alpha diversity

A SRR6261387 4679740.3 18927521 120104 110758 259 126MB 349B SRR6261386 4679765.3 13125990 113623 97136 132 97MB 252C SRR6261389 4679770.3 12578820 93388 82002 87 82MB 248D SRR6261388 4679777.3 12749710 97106 83173 68 84MB 187E SRR6261385 4679778.3 13974374 109777 89419 81 88MB 235H SRR6261384 4679779.3 11824058 96809 80619 98 78MB 333

ANIMAL BIOTECHNOLOGY 3

while maximum reads from all samples were classifiedinto various taxonomic and functional category.

Taxonomic distribution in the goat rumenmicrobiome samples

The rumen microbiota samples from black Bengalgoat dataset harvest diverse organisms at domainlevel such as bacteria, archaea, eukaryote and virusfollowed by unclassified microbiota. In this 6 rumen

microbiome data-set, we observed that abundance ofbacteria (91.83–98.28%) was dominant as compared toarchaea (0.74–7.19%), Eukaryota (0.57–0.72%), virus(0.03–0.09%) and others (0.16–0.19%).

A total of 147391 (92.91%), 129479 (93.9%), 110952(97.95%), 116091(98.07%), 120912 (98.28%), 101544(91.83%) sequences were enriched with bacteria fromsample A, B, C, D, E, H, respectively. The secondlargest domain identified in 6 rumen microbiota sam-ple was archaea. A total of 9723 (6.1%), 7216 (5.2%),

Figure 1. Taxonomic domain distribution. A stacked bar chart of taxon abundance in (y-axis) of different domains distributedacross samples A, B, C, D, E, and H in (x-axis).

Figure 2. Taxonomic distribution of different phyla based on predicted proteins in six metagenomic samples.

4 P. R. SURYAWANSHI ET AL.

1514 (1.33%), 1282 (1.08%), 917 (0.74%), 7951(7.19%) sequences were enriched with archea for sam-ple A, B, C, D, E, and H respectively. At the domainlevel, bacteria were the most abundant followed byarchaea, Eukaryota, viruses, and others. The

taxonomic distribution of different domains at thegene sequence level is shown in Fig. 1. The top 10most abundant genus found across all six sample inorder were Prevotella, Bacteroidetes, Clostridium,Methanobrevibacter, Eubacterium, Ruminococcus,

Table 3. Functional genes present in the goat rumen samples (values are in percentage).Functional genes A (%) B (%) C (%) D (%) E (%) H (%)

Clustering-based subsystems 16.95 16.91 17.07 17.00 17.20 16.89Carbohydrates 11.08 10.64 11.57 11.22 11.05 10.58Miscellaneous 8.54 9.20 9.16 9.26 9.42 8.61Cell-wall and capsule 6.92 7.40 6.97 7.48 7.46 6.88Protein metabolism 6.91 7.00 6.80 6.62 6.49 7.23Amino acids and derivatives 6.36 6.41 6.58 6.33 6.28 6.29DNA metabolism 6.32 5.97 6.33 6.52 6.20 6.43Cofactors, vitamins, prosthetic groups, pigments 5.64 5.93 5.29 5.36 5.41 5.77RNA metabolism 5.46 5.37 4.88 5.01 5.25 5.67Membrane transport 3.15 2.89 2.74 2.64 2.69 2.94Virulence, disease and defense 2.84 2.47 2.50 2.56 2.62 2.72Fatty acids, lipids, and isoprenoids 2.43 2.40 2.58 2.39 2.53 2.45Nucleosides and nucleotides 2.41 2.63 2.54 2.48 2.50 2.60Phages, prophages, transposable elements, plasmids 2.12 2.01 2.30 2.56 2.62 2.13Respiration 2.08 2.09 1.80 1.76 1.77 2.16Cell division and cell cycle 1.79 1.68 1.90 1.77 1.75 1.83Stress response 1.69 1.87 1.88 1.81 1.77 1.70Sulfur metabolism 1.29 1.09 1.14 1.20 1.07 1.19Regulation and cell signaling 1.14 1.21 1.22 1.30 1.25 1.26Iron acquisition and metabolism 1.01 1.06 1.16 1.13 1.04 0.92Motility and chemotaxis 0.87 0.81 0.73 0.76 0.85 0.83Metabolism of aromatic compounds 0.59 0.54 0.53 0.52 0.59 0.57Nitrogen metabolism 0.58 0.63 0.55 0.57 0.57 0.58Phosphorus metabolism 0.58 0.53 0.59 0.54 0.48 0.54Dormancy and sporulation 0.51 0.53 0.40 0.52 0.48 0.46Potassium metabolism 0.42 0.40 0.46 0.37 0.40 0.41Secondary metabolism 0.22 0.25 0.23 0.25 0.20 0.27Photosynthesis 0.10 0.08 0.09 0.06 0.07 0.08

Figure 3. KEGG pathway analysis of six Black Bengal goats using MEGAN.

ANIMAL BIOTECHNOLOGY 5

Butyrivibrio, Parabacteroides, Alistipes, and Roseburia.Among six rumen samples, the most abundantspecies identified were Prevotella ruminocola,Methanobrevibacter smithii, Bacteroides vulgatus,Bacteroides fragalis, Clostridium proteoclastium,

Ruminococcus albus, Eubacterium rectal and parabac-teroides distasonis, etc.

At phyla level Bacteriodetes (41.38–59.74%) werethe most abundant phyla followed by Firmicutes(30.59–39.96%), Proteobacteria (5.07–7.61%),

Figure 4. (a) Enzyme distribution of abundant CAZy classes (CBM, CE, GH, GT) of six metagenomic samples through dbCAN data-base. (b) less abundant CAZy classes (AA, Cohesin, Dockerin, PL, SLH) of six metagenomic samples by the help of dbCAN databaseis represented.

6 P. R. SURYAWANSHI ET AL.

Euryarcheaota (0.71–7.41%), Actinobacteria(2.05–2.75%), Spirochaetes (0.40–0.70%), Fusobacteria(0.37–0.68%), Chloroflexi (0.30–0.67%) and Chlorobi(0.37–0.49%). From the phyla distribution, it can beinferred that Bacteriodetes and Firmicutes are themost abundant phyla across six samples. Again sampleD was having the highest percentage of Bacteriodetes(59.74%) among all six samples whereas sample A washaving the highest percentage of Firmicutes (39.96%).A total of 9 different phyla were observed in the 6goat rumen microbiome dataset and distribution ofthese phyla is provided in Fig. 2. Our study findingswere consistent to the findings of Wang et al.28

wherein they reported that the Bacteroidetes,Proteobacteria, and Firmicutes were identified as thedominant phyla in goats’ rumen and also observedthat the abundance of Proteobacteria continuallydecreased, while Bacteroidetes continually increasedwith age. In our study similar drop in the abundanceof Proteobacteria was observed but the abundance of

Firmicutes was more in comparison to Proteobacteria.This may be due to the age of the goats used in ourstudy that was from 3 months to 1 year of age. Majorfiber degrading bacteria belonging to the bacteriagenera Clostridium, Ruminococcus, Eubacterium,Butyrivibrio, Roseburia, Caldicellulosiruptor,Rdodospirillum, and Treponema were identified in cat-tle rumen29 and we also identified similar bacterialgenera profile in black Bengal goat samples, might beinvolved in the degradation of plant fibre.

Predicted gene functions

Analysis of the subsystem classification throughMGRAST results indicated the presence of function-ally characterized protein-coding genes (Table 3).The highest proportion of gene fragments assigned toknown functions was associated with clustering basedsubsystems ranging from 16.89 to 17.2%, followed bygenes associated with carbohydrate metabolism

Figure 5. Gycosyl hydrolase family distribution across six samples. The color codes range from shades of red to green whereshades of green represents low values and shades of red represents high values.

ANIMAL BIOTECHNOLOGY 7

(10.58–11.22%), miscellaneous (11.10–12.16%), Cellwall and Capsule (6.49–7.23%) and Protein metabol-ism (6.49–7.23%).

KEGG pathway analysis through MEGAN

The focus of this work is to report on new featuresof MEGAN that allow the functional analysis ofmultiple metagenomes based on the KEGG pathways.From the analysis, it is observed that KEGG analysisis broadly classified into six major categories suchas metabolism, genetic information processing,

environmental information processing, cellular proc-esses, organismal systems, and human diseases. In themetabolism category, the maximum annotation wasobtained across six black Bengal goat samples with8092 hits followed by genetic information processingcategory with 1850 hits and Fig. 3 represents theKEGG pathway analysis by MEGAN.

Identifying CAZymes from goat rumen metagenomeRumen fluid is an excellent source for identifyingenzymes responsible for the degradation of the

Figure 6. Carbohydrate binding modules family distribution across six samples. The X-axis represents different CBM sub-familiesand Y-axis represents the corresponding ORFs (Open Reading Frames) encoding each sub-families. The most abundantly foundCBM modules are depicted in (a), and less abundant CBM modules are represented in (b).

8 P. R. SURYAWANSHI ET AL.

lignocellulosic system. Total sequences obtained fromeach rumen sample were subjected for annotationagainst the dbCAN database as described in themethod section. Nine major categories of enzymeswere identified from dbCAN i.e. Glycoside Hydrolases(GH), Glycosyl Transferase (GT), CarbohydrateBinding module Domain (CBM), CarbohydrateEsterase (CE), Dockerin, Polysaccharide Lyases (PL),S-layer Homology domain (SLH), Auxillary Activity(AA) and Cohesin as represented in Fig. 4a and b.GH are a prominent group of enzymes that hydrolyzethe glycosidic bond among the carbohydrate mole-cules and a diverse family of GH catalytic modules(cellulases, Endo-hemicelluleases, Cell WallElongation, Debranching, Oligosaccharide degrading)were identified probably playing significant roles inthe goat rumen. Candidate sequence that belong to

the Glycosyl hydrolase families GH2, GH3, GH31,GH92, GH97 were most abundant, followed byCarbohydrate Glycosyl Transferase (GT) families(23.11–23.73%), Carbohydrate Binding moduleDomain (CBM) family (15.61–17.65%) andCarbohydrate Esterase (CE) (11.95–12.72%). Figure 5represents in details about the different GH familiesacross 6 rumen microbiota sample. In line with ourGH families identified in goat rumen microbiota,GH3, GH2, GH92 were abundantly found in buffalorumen.30 The major activity of GH3 involve broadsubstrate specificities such as b-D-glucosidases, a-L-arabinofuranosidases, b-D-xylopyranosidases, and N-acetyla-b-D-glucosaminidases and broad specificitiesare possible due to the presence of monosaccharideresidue, linkage position and chain length of a sub-strate.31,32 GH2 is majorly involved in activity with a

Figure 7. Comparison of predicted carbohydrate-active genes glycosyl transferase (GT) in six cellulosic metagenomic samples. (a)represents abundantly found GT classes i.e. GT2, GT4, and (b) represents less abundantly found GT classes across rumen metage-nomic samples.

ANIMAL BIOTECHNOLOGY 9

substrate such as b-D-galactosidases, exo-b-glucosami-nidase, b-D-mannosidases, b-glucuronidases andGH43 shows b-Xylosidase, xylanase, galactan, b-1-3-xylosidase and a-L-arabinofuranosidase (http://www.cazy.org/). Similar GH enzyme identification in theguts was reported in earlier research publication ofarthropods,33 hindgut in a wood-feeding termite30 andas well as in rumen of buffalo.16

A total of 52 subfamily of Carbohydrate-bindingmodules (CBM) were identified across six samples ofgoat rumen samples and the most abundant class ofCBM identified in our were CBM32, CBM50, CBM61,CBM67, etc. as depicted in Fig. 6a and b. CBMs areknown to target the enzyme to distinct regions on asaccharide substrate (reducing end, non-reducing end,internal polysaccharide chains), depending on the archi-tecture of its binding site and increase the concentrationof enzyme in close proximity to its saccharide substrate,

increase the hydrolytic activities of the enzymes againstinsoluble and soluble substrates which leads to morerapid and efficient substrate degradation.34 The occur-rence of CBMs in our study is in accordance with thefinding that microbiome of ruminants render cellulo-lytic bacteria associated with cellulosome complexes.35

Glycosyl transferases (GT) are ubiquitous enzymesthat catalyze the attachment of sugars to glycone.36 Inbuffalo rumen metagenome analysis, the most abun-dant glycosyl transferase families identified wereGT51, GT2, and GT 35.16 In line with this, we haveidentified 45 subfamilies of GT subfamilies across 6black Bengal goat rumen sample wherein abundanceof GT2, GT4, GT51, GT28, and GT83 were indecreasing order. Figure 7a represents the distributionof abundant GT subfamilies across 6 rumen micro-biota samples and Fig. 7b represents the distributionof less abundantly found GT families.

Figure 8. Comparison of predicted carbohydrate-active genes carbohydrate esterases in six cellulosic metagenomes. The mostabundant CE, such as CE1 and CE10, is represented in (a) and the less abundant CE groups are provided in (b).

10 P. R. SURYAWANSHI ET AL.

Sequences assigned to Carbohydrate esterase (CE-11.95–12.9%) belong to CE1, CE10, CE4, CE3, CE6,CE7, CE12, and Fig. 8a and b represent the distribu-tion of 15 CE-subfamily across the 6 rumen goat sam-ple. Carbohydrate esterases (CEs) are involved incatalyze deacylation of polysaccharides. Singh et al.,16

identified xylan esterase (CE1, 4, 6, 7) and pectinmethylesterase (CE8) in the buffalo rumen microbiota,probably works on the side chains of hemicelluloseand pectin respectively for a further breakdown of thelarge molecules into smaller for better accessibility.The presence of CE1, CE10, CE4, CE3, CE6, CE7,CE12 indicates that these are helpful on the similarway for breakdown of larger molecules into smallerfor better accessibility.

Also, sequences assigned to Pectate lyases (PL) inour 6 rumen samples were assigned to PL9, PL1-2,

PL11, PL22, PL1, PL12 and detail of the same is givenin Fig. 9a and b. PLs are known to degrade highlynegatively charged non-methylated or low-esterifiedsubstrates and cleave a-1,4 linkages between galactur-onosyl residues mainly pectin. Pectin is the majorcarbohydrate present in legumes and in some non-for-age plants and it constitutes 10 to 20% of total carbo-hydrate in the grass as well as a major component ofthe plant cell wall.

Many genes encoding cellulose and carbohydrate-binding module have been reported in buffalorumen.37,38 Large-scale metagenomic sequencing ofhindgut bacteria of a wood-feeding higher termiterevealed that GHF5 was predominant in all identifiedGH families.30 The most common activities of GH3include b-D-glucosidases, a-L-arabinofuranosidases,

Figure 9. Comparison of predicted carbohydrate-active genes pectate lyases (PL) in six cellulosic metagenomes. The most abun-dant PL groups are presented in (a) and other less abundant identified PL in six samples are represented in (b).

ANIMAL BIOTECHNOLOGY 11

b-D-xylopyranosidases, and N-acetyl-b-D-glucosami-nidases studied using protein modeling.32

Recently, Do et al., reported that carbohydrateesterases, polysaccharide lyases, glycoside hydrolase,hemicellulases, carbohydrate binding module andother lignocellulolytic enzymes in Vietnamese nativegoats’ rumen using enzymes using Illumina sequenc-ing, which is in agreement with our observation inthe rumen of black Bengal goat.9

The CBM (Carbohydrate binding module) domainhelps in binding of a CAZyme to its carbohydratesubstrate, thereby facilitating the enzyme’s activity.With respect to the enzyme binding cellulose, the pre-dominant CBM was CBM6 (450 genes) followed byCBM2 (91 genes) distributed across six black Bengalgoat; among the xylan-binding domains, members ofthe CBM4 (451 genes), CBM9 (51 genes), CBM16(158 genes), CBM22(93 genes); among the enzymesbinding to starch, the predominant CBM was CBM20(360 genes), CBM34 (87 genes); similarly among thegenes encoding proteins binding to galactose, CBM32(1061 genes), CBM51(108 genes);the chitin bindingenzymes were distributed between CBM5(9genes),CBM12(22 genes), CBM14(6 genes).

The most frequent CBM modules identified incamel rumen’s microbiome were CBM20, CBM37,CBM56, and CBM61 encoding genes,24 which is inline with our findings with black Bengal goat. Themost abundant CBM was CBM50, which is also com-mon in the biogas fermenter and anaerobic digestermicrobiomes.24 We have also identified CBM50encoding for 908 genes which are distributed acrosssix black Bengal goat.

Lignocellulosic-degrading enzymes are not onlyrelevant for biofuel production but they have alsogained use in several industries, including biopulpingof wood, treatment of animal feeds to increase digest-ibility, juice processing for improved clarification,detergent industry, flour handling modification forbaking and as fiber softeners in textile prepar-ation.17,39–41 Therefore, the interest in cellulases/hemi-cellulases is having wider industrial application andidentified enyzmes in our study can be further investi-gated in the near future.

Microbial enzymes detected in the rumen of blackBengal goats have plenty of applications which havebeen proved earlier as glycoside hydrolase have beenefficiently used as dough in the baking industry to actas bread anti-staling agents, esterases played importantroles in the food and pharmaceutical industries. In thefood industry, esterases are used in fat and oil modifi-cation and in the fruit juices and alcoholic beverages

industries to produce certain flavors and fragrances.42

Therefore, the rumen microbial enzymes can furtherbe explored for its industrial applications.

In conclusion, we have explored the rumen micro-bial diversity of black Bengal goats using culture-inde-pendent metagenomic approach and the differentCAZymes secreted by the diverse group of microor-ganisms. The diverse group of enzymes secreted byrumen microbes can be efficiently be utilized in vari-ous industrialized applications.

Compliance with ethical standards

The authors hereby declare that this Project was submittedto the funding agency with prior approval from the AnimalEthics and Biosafety Committee of College of VeterinarySciences & Animal Husbandry, Agartala Tripura.

Acknowledgments

The author acknowledges National Bureau of AgriculturallyImportant Microorganisms (NBAIM) for sanction of thisproject grant with file number NBAIM/AMAAS/2014-15/62.

Disclosure statement

No potential conflict of interest was reported bythe authors.

Funding

This work was funded by the Application ofMicroorganisms in Agriculture and Allied Sector (AMAAS)project, Indian Council of Agricultural Research (ICAR),New Delhi, India

References

1. Basic animal husbandry and Fishery statistics. 2017.AHS series-18 Government of India, Ministry ofAgriculture and Farmers welfare. Department ofAnimal Husbandry and Dairying, Department of fish-eries, Krishi Bhavan, New Delhi website. dahd.nic.in/Division/statistics/animal-husbandry-statistics-division

2. Hespell RB, Akin DE, Dehority BA. GastrointestinalMicrobiology. New York, NY: Chapman and Hall;1997:59–186.

3. Hungate RE. Symposium on ‘Nutritional implicationsof microbial action in the non-ruminal alimentarytract’. Microbes of nutritional importance in the ali-mentary tract. Proc Nutr Soc. 1984;43(1):1–11.

4. Fox DG, Barry MC, Pitt RE, Roseler DK, Stone WC.Application of the Cornell Net Carbohydrate andProtein model for cattle consuming forages. J AnimSci. 1995;73(1):267–277.

5. Ozbayram EG, Kleinsteuber S, Nikolausz M, Ince B,Ince O. Effect of bioaugmentation by cellulolytic

12 P. R. SURYAWANSHI ET AL.

bacteria enriched from sheep rumen on methane pro-duction from wheat straw. Anaerobe 2017;46:122–130.

6. Mackie RI. Mutualistic fermentative digestion in thegastrointestinal tract: diversity and evolution. IntegrComp Biol. 2002;42(2):319–326.

7. Tagang A, Ishaku KP, Abdullahi A. Volatile fattyacids production in ruminants and the role of mono-carboxylate transporters: a review. African JBiotechnol. 2010;9:6229–6232.

8. Han X, Yang Y, Yan H, Wang X, Qu L, Chen Y.Rumen bacterial diversity of 80 to 110-day-old goatsusing 16S rRNA sequencing. PLoS One. 2015;10(2):e0117811.

9. Do TH, Dao TK, Nguyen KHV, et al. Metagenomicanalysis of bacterial community structure and diver-sity of lignocellulolytic bacteria in Vietnamese nativegoat rumen. Asian-Australas J Anim Sci. 2018;31(5):738–747.

10. Cremonesi P, Conte G, Severgnini M, et al.Evaluation of the effects of different diets on micro-biome diversity and fatty acid composition of rumenliquor in dairy goat. Animal. 2018;12(9):1856–1866.

11. Lim S, Seo J, Choi H, et al. Metagenome analysis ofprotein domain collocation within cellulase genes ofgoat rumen microbes. Asian Australas J Anim Sci.2013; 26(8):1144–1151.

12. Cunha IS, Barreto CC, Costa OY, et al. Bacteria andArchaea community structure in the rumen micro-biome of goats (Capra hircus) from the semiaridregion of Brazil. Anaerobe 2011;17(3):118–124.

13. Badapanda C, Metha S. Advancing our understandingof the oxygen minimum zone microbial communitiesby an integrated metatranscriptomics approach. MetaGene. 2017;14:85–90.

14. Rani R, Badapanda C. Analysis of the metatranscrip-tome of microbial communities by comparison of dif-ferent assembly tools reveals improved functionalannotation. Anatomy Physiol Biochem. 2017;3:1–7.

15. Badapanda C, Rani C, Sahoo GC. Advancing ourunderstanding of the soil microbial communitiesusing QIIME software: a 16S data analysis pipeline.J Appl Biotechnol Bioeng. 2017;4:610–616.

16. Singh KM, Reddy B, Patel D, et al. High potentialsource for biomass degradation enzyme discovery andenvironmental aspects revealed through metagenomicsof Indian buffalo rumen. Biomed Res Int. 2014;2014:1.

17. Duque E, Daddaoua A, Cordero BF, et al. Ruminalmetagenomic libraries as a source of relevant hemicel-lulolytic enzymes for biofuel production. MicrobBiotechnol. 2018;11(4):781–787.

18. Handelsman J. Metagenomics: application of genom-ics to uncultured microorganisms. Microbiol Mol BiolRev. 2004;68(4):669–685.

19. Brulc JM, Antonopoulos DA, Miller ME, et al. Gene-centric metagenomics of the fiber-adherent bovinerumen microbiome reveals forage specific glycosidehydrolases. PNAS. 2009;106(6):1948–1953.

20. Hess M, Sczyrba A, Egan R, et al. Metagenomic dis-covery of biomass-degrading genes and genomes fromcow rumen. Science 2011;331(6016):463–467.

21. Mootapally CS, Nathani NM, Patel AK, Jakhesara SJ,Joshi CG. Mining of ruminant microbial phytase

(RPHY1) from metagenomic data of Mehsani buffalobreed: identification, gene cloning, and characteriza-tion. J Mol Microbiol Biotechnol. 2016;26(4):252–260.

22. Forsberg W, Cheng KJ, White BA. Polysaccharidedegradation in the rumen and large intestine. In:Mackie RI, White BA, eds. GastrointestinalMicrobiology. New York, NY: Chapman & Hall; 1997:319–379.

23. Flint HJ, Bayer EA, Rincon MT, Lamed R, White BA.Polysaccharide utilization by gut bacteria: potentialfor new insights from genomic analysis. Nat RevMicro. 2008;6(2):121–131.

24. Gharechahi J, Salekdeh GH. A metagenomic analysisof the camel rumen’s microbiome identifies the majormicrobes responsible for lignocellulose degradationand fermentation. Biotechnol Biofuels 2018;11:216.

25. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flex-ible trimmer for Illumina sequence data.Bioinformatics. 2014;30(15):2114–2120.

26. Hyatt D, Chen GL, Locascio PF, Land ML, LarimerFW, Hauser LJ. Prodigal: prokaryotic gene recognitionand translation initiation site identification. BMCBioinformatics. 2010;11:119.

27. Huson DH, Auch AF, Qi J, Schuster SC. MEGANanalysis of metagenomic data. Genome Res. 2007;17(3):377–386.

28. Wang L, Xu Q, Kong F, et al. Exploring the goatrumen microbiome from seven days to two years.PLoS One. 2016;11(5):e0154354.

29. Pandit RJ, Hinsu AT, Patel SH, et al. Microbiota com-position, gene pool and its expression in Gir cattle(Bos indicus) rumen under different forage diets usingmetagenomic and metatranscriptomic approaches.Syst Appl Microbiol. 2018;41(4):374–385.

30. Warnecke F, Luginb€uhl P, Ivanova N, et al.Metagenomic and functional analysis of hindgutmicrobiota of a wood-feeding higher termite. Nature2007;450(7169):560–565.

31. Lee RC, Hrmova M, Burton RA, Lahnstein J, FincherGB. Bifunctional family 3 glycoside hydrolases frombarley with alpha -L-arabinofuranosidase and beta -D-xylosidase activity. Characterization, primary struc-tures, and COOH-terminal processing. J Biol Chem.2003;278(7):5377–5387.

32. Harvey AJ, Hrmova M, De Gori R, Varghese JN,Fincher GB. Comparative modeling of the three-dimensional structures of family 3 glycoside hydro-lases. Proteins 2000;41(2):257–269.

33. Bashir Z, Kondapalli VK, Adlakha N, et al. Diversityand functional significance of cellulolytic microbesliving in termite, pill-bug and stem-borer guts. SciRep. 2013;3:2558.

34. Bolam DN, Ciruela A, McQueen-Mason S, et al.Pseudomonas cellulose-binding domains mediate theireffects by increasing enzyme substrate proximity.Biochem J. 1998;331(3):775–781.

35. Pope PB, Mackenzie AK, Gregor I, et al.Metagenomics of the Svalbard reindeer rumen micro-biome reveals abundance of polysaccharide utilizationloci. PLoS One. 2012;7(6):e38571.

ANIMAL BIOTECHNOLOGY 13

36. Lairson LL, Henrissat B, Davies GJ, Withers SG.Glycosyltransferases: structures, functions, and mecha-nisms. Annu Rev Biochem. 2008;77(1):521–555.

37. Duan CJ, Xian L, Zhao GC, et al. Isolation and partialcharacterization of novel genes encoding acidic cellu-lases from metagenomes of buffalo rumens. J ApplMicrobiol. 2009;107(1):245–256.

38. Duan CJ, Liu JL, Wu X, Tang JL, Feng JX. Novelcarbohydrate-binding module identified in a ruminalmetagenomic endoglucanase. Appl Environ Microbiol.2010;76(14):4867–4870.

39. Rashamuse KJ, Visser DF, Hennessy F, et al.Characterisation of two bifunctional cellulase-xylanaseenzymes isolated from a bovine rumen metagenomelibrary. Curr Microbiol. 2013;66(2):145–151.

40. Cai S, Li J, Hu FZ, et al. Cellulosilyticum ruminicola,a newly described rumen bacterium that possessesredundant fibrolytic-protein-encoding genes anddegrades lignocellulose with multiple carbohydrate-borne fibrolytic enzymes. Appl Environ Microbiol.2010;76(12):3818–3824.

41. Wang L, Hatem A, Catalyurek UV, Morrison M, YuZ. Metagenomic insights into the carbohydrate-activeenzymes carried by the microorganisms adhering tosolid digesta in the rumen of cows. PLoS One. 2013;8(11):e78507.

42. Coughlan LM, Cotter PD, Hill C, Alvarez-Ord�o~nez A.Biotechnological applications of functional metage-nomics in the food and pharmaceutical industries.Front Microbiol. 2015;6:672.

14 P. R. SURYAWANSHI ET AL.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具