Gastrointestinal Bacterial and Methanogenic Archaea Diversity Dynamics Associated with ...

12 Pages·2015·1.51 MB·English

Hindawi Publishing Corporation
International Journal of Microbiology
Volume 2014, Article ID 141909, 11 pages
http://dx.doi.org/10.1155/2014/141909
Research Article
Gastrointestinal Bacterial and Methanogenic
Archaea Diversity Dynamics Associated with
Condensed Tannin-Containing Pine Bark Diet in
Goats Using 16S rDNA Amplicon Pyrosequencing
ByengR.Min,1SandraSolaiman,1RaymonShange,1andJong-SuEun2
1DepartmentofAgriculturalandEnvironmentalSciences,TuskegeeUniversity,Tuskegee,AL,USA
2DepartmentofAnimal,Dairy,andVeterinarySciences,UtahStateUniversity,Logan,UT,USA
CorrespondenceshouldbeaddressedtoByengR.Min;[email protected]
Received16July2013;Revised4September2013;Accepted4September2013;Published2January2014
AcademicEditor:IsabelSa´-Correia
Copyright©2014ByengR.Minetal.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense,
whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
EighteenKiko-crossmeatgoats(𝑛=6)wereusedtocollectgastrointestinal(GI)bacteriaandmethanogenicarchaeafordiversity
measureswhenfedcondensedtannin-containingpinebark(PB).Threedietarytreatmentsweretested:controldiet(0%PBand30%
wheatstraw(WS);0.17%condensedtannins(CT)drymatter(DM));15%PBand15%WS(1.6%CTDM),and30%PBand0%WS
(3.2%CTDM).A16SrDNAbacterialtag-encodedFLXampliconpyrosequencingtechniquewasusedtocharacterizeandelucidate
changesinGIbacteriaandmethanogenicarchaeadiversityamongthediets.Proteobacteriawasthemostdominantphylumingoats
withmeanrelativeabundancevaluesrangingfrom39.7(30%PB)to46.5%(control)and47.1%(15%PB).Otherphylaindividually
accountedforfewerthan25%oftherelativeabundanceobserved.PredominantmethanogenswereMethanobrevibacter(75,72,
and49%),Methanosphaera(3.3,2.3,and3.4%),andMethanobacteriaceae(1.2,0.6,and0.7%)populationincontrol,15,and30%
PB,respectively.Amongmethanogens,Methanobrevibacterwaslinearlydecreased(𝑃=0.05)withincreasingPBsupplementation.
TheseresultsindicatethatfeedingPBselectivelyalteredbacteriaandmethanogenicarchaealpopulationsintheGItractofgoats.
1.Introduction Planttannins(condensed(CT)andhydrolysabletannins)
are polyphenolic compounds of relatively high molecular
Studiesongastrointestinal(GI)microorganismshavetradi- weightwiththecapacitytoformcomplexesmainlywithpro-
tionally depended on the use of anaerobic cultivation tech- teins due to the presence of a large number of phenolic
niques[1]whichonlycandetectanestimated11%ofthetotal hydroxyl groups [7]. They are ubiquitously spread in nutri-
bacterial populations present in the rumen [2]. Molecular tionallyimportantforagetrees,shrubs,legumes,cereals,and
methodologies developed over the past decade now enable tree barks. The effects of tannins on ruminant production
researchers to examine the diversity of the gut microflora havebeenextensivelypublishedinthepast,andamongthem
independentofculturalmethods.Bacterialdiversitieswithin beneficial effects of tannins on animal production system
the GI tract of humans [3, 4] and rumen of beef cattle have been much focused, with particular interest in their
[5, 6] have been investigated in recent years as a result of positiveeffectsonproteinmetabolism,preventionoffrothy
development of 16S rDNA-based analysis, yet similar data bloat,andmodificationonrumenmicrobialpopulation[7].
on the microbiomes of bacteria and methanogenic archaea Microbialpopulationchangesinthegutwhenanimalswere
diversity in the lower GI tract of meat goats are limited. In fed CT-containing diets were reported using a 16S PCR
addition,thereisnoclearinformationontheprevalenceof technique in sheep [8] and rat [9]. In addition, the use of
methanogenic and microbial populations in the hindgut of rapid sequencing technologies combined with molecular
animalsotherthanhumanandswine. methods is becoming a prevalent standard for evaluating
2 InternationalJournalofMicrobiology
Table1:Chemicalcompositionsofpinebark(PB),wheatstraw(WS),bermudagrasshay(BGH),andexperimentaldietsfedtoKikocrossbred
malegoatkids.
Ingredient,% Experimentaldiets1
Item SEM
PB WS BGH Control 15%PB 30%PB
Ingredientofthegrain/PBmix,%asis
GroundPB — — — 0 15.0 30.0 —
GroundWS — — — 30.0 15.0 0 —
Corn — — — 20.0 20.0 20.0 —
Soybeanmeal — — — 18.5 20.0 21.0 —
Soyhulls — — — 4.5 5.0 4.0 —
Alfalfameal — — — 5.0 3.0 3.0 —
Molasses — — — 6.0 6.0 6.0 —
Vitaminandmineralmix — — — 0.5 0.5 0.5 —
Salt — — — 0.5 0.5 0.5 —
NH Cl — — — 0.5 0.5 0.5 —
4
BGH — — — 15.0 15.0 15.0 —
Chemicalcomposition,%DM(n=3)
Drymatter 83.6 83.5 91.4 89.7 87.8 87.3 0.77
Crudeprotein 1.2 4.1 7.3 15.7 16.8 16.1 0.41
Neutraldetergentfiber 78.6 79.0 69.2 35.0 31.8 27.5 1.77
Aciddetergentfiber 72.1 49.2 37.3 23.7 23.2 23.6 1.42
Etherextract 1.65 0.42 1.51 2.3 2.6 2.5 0.25
Ash 2.3 2.0 4.8 6.4 6.2 5.9 0.31
NFC2 17.1 16.7 19.1 42.1 42.5 47.1 1.91
Aciddetergentlignin 21.3 8.01 6.3 5.9 9.9 12.4 0.85
Condensedtannins3 10.3 0.03 0.04 0.19 1.63 3.20 0.19
1
Control(0%PB),15%PB,and30%PBonanas-fedbasis.Exceptbermudagrasshay,allingredientswereincorporatedinthegrainmixes.
2Nonfibercarbohydrates=100–CP–NDF–etherextract–ash.
3Condensedtannins(CT)arerelativetoapurifiedquebrachoCTstandard.
SEM:standarderrorofthemean.
the microbiome of animals [10]. However, there is still a 2.1. Experimental Animals and Diets. Eighteen Kiko cross-
lack of information on the microbial diversity in the gut of bred male goat kids (Capra hircus; initial body weight
ruminantssuchassheepandgoats.Todate,littleresearchhas (BW)=31.0±1.49kg)werestratifiedbyBWandrandomly
been focused on changes in microbial population diversity, assignedtooneof3experimentaltreatmentgroups(𝑛=6).
particularlymethanogenicarchaealpopulation,andnostudy Goatswereindividuallyhousedindoorsinpensofapproxi-
has measured the fecal microbial population in response mately1.2m2 withelevatedfloorsandwerefedgrainmixes
to feeding diets containing different dietary concentrations withdifferentamounts(0,15,and30%drymatter(DM)PB)
of CT-containing pine bark (PB). In our previous study, ofPBandlongbermudagrasshay(BGH;Cynodondactylon)
additionofPBingoatdietsimprovedgrowthperformance, at85:15onanas-fedbasis.Anadjustmentperiodof4weeks
reduced fecal egg counts, and favorably modified ruminal allowedgoatstobecomeacclimatedtopenlivingandroutine
fermentation[11].Inthepresentstudy,therefore,weutilizeda feedingandtoallottimeforproperdietadjustmentpriorto
bacterialtag-encodedFLX-titaniumampliconpyrosequenc- studyinitiation.Inwk1,allanimalswerefedadietwithout
ing (bTEFAP) method that is able to perform diversity PB(controldiet).Startingonwk2,intheadjustmentperiod,
analyses of fecal microbial populations. We evaluated the experimentaldietsweregraduallyfedtoanimalsinastepwise
fecal microbiomes from 18 Kiko-cross meat goats fed PB increasingfashion,andattheendofwk4,allanimalswerefed
containingdiets.Hence,ourprincipalobjectiveofthisstudy whole,preassignedexperimentaldiets(Table1).
wastomeasuretheeffectsofCT-containingPBsupplemen- Experimentaldiets(Table1)containeddifferentamounts
tationasafeedsupplementonGImethanogenicarchaeaand of the CT-containing ground PB replacing ground wheat
bacterialdiversitydynamicsinmeatgoats. straw (WS; Triticum aestivum). Experimental treatments
included the controldiet (0% PB), 15% PB, and 30% PB on
2.MaterialsandMethods anas-fedbasis.TheWScontainednegligibleCT(0.03%),but
concentrationsofneutraldetergentfiber(NDF)andnonfiber
Care and handling of all experimental animals were con- carbohydrates were similar between PB and WS [11]. The
ductedunderprotocolsapprovedbytheTuskegeeUniversity freshPBwasdonatedbyawood-processingcompany(West
InstitutionalAnimalCareandUseCommittee. Fraser Timber Co. Ltd., Opelika, AL) and air-dried under
InternationalJournalofMicrobiology 3
the shed before processing. Freshly dried PB (10.3%CT) optimization and pyrosequencing analysis. Bacterial tag-
was ground (Hammer Mill Model 1250; Lorenz MFG Co., encodedFLXampliconpyrosequencingPCRwascarriedout
Benson,MN)toapproximately3mmparticle.TheWS(0.03% accordingtoproceduredescribedpreviously[10,14].
CT)wasalsoground(3mm)andincorporatedinthegrain
mixportionofthediets.ConcentrationofCTin15%PBand
2.4. bTEFAP Sequencing PCR. The bTEFAP and data pro-
30%PBdietswas1.63%and3.20%DM,respectively(Table1).
cessingwereperformedasdescribedpreviously[10].AllDNA
The inclusion rate of PB was chosen, as we previously
samples were adjusted to 100ng/𝜇L. A 100ng (1𝜇L) aliquot
reported that beneficial effects of CT in the diet on sheep
of each sample’s DNA was used for a 50𝜇L PCR reaction.
performance may occur in the range of 2 to 4% CT of diet 󸀠
The 16S universal eubacterial primers 530F (5-GTG CCA
DM [7, 11]. Mixtures containing ground PB and WS were 󸀠
GCM GCN GCG G) and 1100R (5 -GGG TTN CGN TCG
commercially prepared at the local feed mill (Eclectic Feed
TTG)wereusedforamplifyingthe600bpregionof16SrRNA
Mill, Eclectic,AL). Experimentaldiets metrequirementsof
genes. HotStar Taq Plus Master Mix Kit (Qiagen, Valencia,
experimentalanimalsforgrowthandBWgainaccordingto ∘
CA)wasusedforPCRunderthefollowingconditions:94 C
the NRC [12]. Diets were isonitrogenous and isocaloric but ∘ ∘
for 3min followed by 32 cycles of 94 C for 30sec; 60 C for
differed in NDF, lignin, and CT concentrations; lignin and ∘ ∘
40secand72 Cfor1min;andafinalelongationstepat72 C
CTconcentrationsweregreaterin15%and30%PBrations,
for5min.AsecondaryPCRwasperformedforFLX(Roche,
whereas NDF concentration was decreased compared with
Nutley,NJ)ampliconsequencingunderthesamecondition
controldiet.
by using designed special fusion primers with different tag
sequences as: LinkerA-Tags-530F and LinkerB-1100R [10].
2.2. Sample Collection and Laboratory Analysis. After com- The use of a secondary PCR prevents amplification of any
pletion of adaptation period for 4wk, total fecal collection potential bias that might be caused by inclusion of tag and
wasperformedfor7dtoallgoats.Animalswerefedoncea linkersduringinitialtemplateamplificationreactions.After
dayat8:00handhadfreeaccesstowaterandatracemineral secondaryPCR,allampliconproductsfromdifferentsamples
saltblock.Grainmixesandhaywereofferedseparately,and weremixedinequalvolumesandpurifiedusingAgencourt
refusals were recorded daily. Amounts of feed offered were Ampurebeads(AgencourtBioscienceCorporation,MA).
adjustedeverydaytomaintainthepreferreddailyrefusalrate
of5to10%.
2.5.bTEFAPFLXMassivelyParallelPyrosequencing. Inprep-
Duringthecollectionperiods,totalindividualfecalsam-
aration for FLX sequencing, the DNA fragments size and
plefromthemetabolismcratewascollectedat9:00am(24h
concentrationwereaccuratelymeasuredbyusingDNAchips
overnight collection) during 7d with two different experi-
under a Bio-Rad Experion Automated Electrophoresis Sta-
mentalperiods.Feedandortssamplesforindividualanimals
tion(Bio-RadLaboratories,CA)andaTBS-380Fluorometer
were collected daily during the collection period, dried at
∘ (TurnerBiosystems,CA).Asampleofdouble-strandedDNA
60 Cfor48h,groundtopassa1mmscreen(standardmodel
containing 9.6 × 106 molecules/𝜇L of average size 625bp
4; Arthur H. Thomas Co., Swedesboro, NJ), and stored for
was combined with 9.6 million DNA capture beads and
subsequentanalyses.Dailyportionsofgroundsampleswere
then amplified by emulsion PCR. After bead recovery and
composited for each animal and analyzed for DM, crude
bead enrichment, the bead-attached DNA was denatured
protein (CP), acid detergent lignin, ether extract, and ash
with NaOH, and sequencing primers were annealed. A
accordingtothemethodsdescribedbyAOAC[13].Nitrogen
two-region 454 sequencing run was performed on a 70 ×
for diet sample was determined using a Kjeldahl N, and
75GS PicoTiterPlate (PTP) by using a Genome Sequencer
CP was calculated by multiplying N by 6.25. The NDF
FLX System (Roche, Nutley, NJ). It should be noted that
andADFconcentrationsweresequentiallydeterminedusing
200/220 100 total samples were run within this same FLX 2-region
an ANKOM Fiber Analyzer (ANKOM Technology,
sequencingreaction.Theadditional79taggedsampleswere
Macedon, NY). Sodium sulfate heat stable amylases (Sigma
associatedwithunrelatedstudies.AllFLXrelatedprocedures
Aldrich Co., St. Louis, Mo) were used in the procedure
were performed according to manufacturer’s instructions
for NDF determination and pretreatment with heat stable
(GenomeSequencerFLXSystem).Acustomscriptwrittenin
amylase (Type XI-A from Bacillus subtilis; Sigma-Aldrich
manufacture’sinstructionwasutilizedtogenerateallpossible
Corporation, St. Louis, MO). Acetone (70%) extractable
combinations of 10-mer oligonucleotide tags with GC %
CT in grain mixes were determined using a butanol-HCL
between40and60%.Fromthispool20individualtagswere
colorimetricprocedure[7,11].
chosen(Table2).
Theresultantindividualsampleafterparsingthetagsinto
2.3. DNA Extraction. Genomic bacterial DNA was isolated individual FASTA files was assembled using CAP3. The ace
from0.3goffecalsamplesaccordingtothemethoddescribed files generated by CAP3 were then processed to generate
inthePowersoilDNAminiKit(PowerSoil,WestCarlsbad, a secondary FASTA file containing the tentative consensus
CA).ExtractedDNA(2𝜇L)wasquantifiedusingaNanodrop (TC) sequences of the assembly along with the number of
ND-1000spectrophotometer(NyxorBiotech,Paris,France) reads integrated into each consensus. The TC was required
and run on 0.8% agarose gel with 0.5M tris-borate-EDTA tohaveatleast3-foldcoverage.TheresultingTCFASTAfor
(TBE) buffer. The samples were then transported to the each sample was then evaluated using BLASTn [15] against
Research and Testing Laboratory (Lubbock, TX) for PCR a custom database derived from the RDP-II database [16]
4 InternationalJournalofMicrobiology
Table2:PrimersequencesutilizedforgoatssamplesduringbTEFAP.
󸀠 󸀠
Name Primersequence(5-3)
454-F30 GCCTCCCTCGCGCCATCAGCGCACTACGTGTGCCAGCMGCNGCGG
454-F31 GCCTCCCTCGCGCCATCAGCGCAGCTGTTGTGCCAGCMGCNGCGG
454-F32 GCCTCCCTCGCGCCATCAGCGCATACAGTGTGCCAGCMGCNGCGG
454-F33 GCCTCCCTCGCGCCATCAGCGCATCTATAGTGCCAGCMGCNGCGG
454-F34 GCCTCCCTCGCGCCATCAGCGCATTGGTGGTGCCAGCMGCNGCGG
454-F35 GCCTCCCTCGCGCCATCAGCGCCAGAAAAGTGCCAGCMGCNGCGG
454-F36 GCCTCCCTCGCGCCATCAGTGTGACGTACGTGCCAGCMGCNGCGG
454-F37 GCCTCCCTCGCGCCATCAGTGTGTGCATAGTGCCAGCMGCNGCGG
454-F38 GCCTCCCTCGCGCCATCAGTGTGTCCTCAGTGCCAGCMGCNGCGG
454-F39 GCCTCCCTCGCGCCATCAGTGTGCATCACGTGCCAGCMGCNGCGG
454-F40 GCCTCCCTCGCGCCATCAGTGTGCCTAGAGTGCCAGCMGCNGCGG
454-F41 GCCTCCCTCGCGCCATCAGTGTACATAGTGTGCCAGCMGCNGCGG
454-F42 GCCTCCCTCGCGCCATCAGTGTACATTGAGTGCCAGCMGCNGCGG
454-F43 GCCTCCCTCGCGCCATCAGTGTACATTGTGTGCCAGCMGCNGCGG
454-F44 GCCTCCCTCGCGCCATCAGTGTACCAACAGTGCCAGCMGCNGCGG
454-F45 GCCTCCCTCGCGCCATCAGTGTACCAACTGTGCCAGCMGCNGCGG
454-F46 GCCTCCCTCGCGCCATCAGTGTACCAATCGTGCCAGCMGCNGCGG
454-F47 GCCTCCCTCGCGCCATCAGTGTACCAGATGTGCCAGCMGCNGCGG
454-F48 GCCTCCCTCGCGCCATCAGTGTACCCATAGTGCCAGCMGCNGCGG
454-F49 GCCTCCCTCGCGCCATCAGTGTACAGGGTGTGCCAGCMGCNGCGG
454-F50 GCCTCCCTCGCGCCATCAGTGTACCTATCGTGCCAGCMGCNGCGG
linkerB-1100R GCCTTGCCAGCCCGCTCAGGGGTTNCGNTCGTTG
4000 1800
1600
3500
d
e 1400
ct 3000
edi Us 1200
ved/pr 22050000 ed OT 1000
obser 1500 bserv 680000
TUs 1000 O 400
O
500 200
0
0
Sobs Chao ACE 0 2000 4000 6000 8000 10000
Sequences sampled
0%PB 0%PB
15%PB 15%PB
30%PB 30%PB
(a) (b)
Figure 1: Bacterial richness/diversity estimator (a) and rarefaction curves (b) presented as calculated by MOTHUR at a level of 3%
dissimilarity.0%PB(control),15%PB,and30%PBonanas-fedbasis.
andGenBankwebsiteathttp://www.ncbi.nlm.nih.gov/.The package(SPSSInc.,v17.0,Chicago,IL).Relativeabundance
sequences contained within the curated 16S database were data are presented as percentages/proportions, but prior to
both >1200bp and considered as high quality based upon subjectiontoGLM,theyweretransformedusingthearcsine
RDP-IIstandards. functionfornormaldistributionpriortoanalysis.Packageof
NCSS(NCSS,2007,v7.1.2,Kaysville,UT)wasusedforcluster
analysisthroughwhichdoubledendrogramsweregenerated
2.6. Data Processing and Statistical Analysis. All statistical through use of the Manhattan distance method with no
analyses (Figures 1 to 6) were performed using the SPSS scaling and the unweighted pair technique. Raw sequences
InternationalJournalofMicrobiology 5
6000 1200
d 5000 1000
e
ct
edi 4000 Us 800
ved/pr 3000 ed OT 600
Us obser 2000 Observ 400
T
O 1000 200
0 0
Sobs Chao ACE 0 5000 10000 15000
Richness estimators Sequences samples
0%PB 0%PB
15%PB 15%PB
30%PB 30%PB
(a) (b)
Figure2:Fecalarchaearichness/diversityestimator(a)andrarefactioncurves(b)presentedascalculatedbyMOTHURatalevelof3%
dissimilarity.0%PB(control),15%PB,and30%PBonanas-fedbasis.
weresubmittedtotheNCBISequenceReadArchive(SRA). 16S database were those considered as high quality based
In addition, quantification of major hindgut bacterial and upon Greengenes [20] standards and which had complete
archaea populations (Tables 3 and 4) was analyzed by the taxonomicinformationwithintheirannotations.Clustersat
GLM procedure of the SAS (SAS Inst., Cary, NC) in a 3%and5%werethenutilizedtogeneraterarefactioncurves
completely randomized design with the factors examined andthediversityindicesACE[21,22]aswellasunweighted
beingthreelevelsofPBsupplementationinthediets.Linear and weighted UniFrac for Principle Coordinate Analysis
and quadratic effects were determined utilizing orthogonal (PCOA)plotsandaVenndiagram.
polynomialcontrastsforequallyspacedtreatments.Animals
were the experimental unit and treated as a random effect.
There were no treatment × period interactions (𝑃 > 0.10), 3.ResultsandDiscussion
henceonlythemaineffectsarereportedformajormicrobial
Despiteseveraldecadesofstudiesdemonstratingtheroleof
populationsintheresultsection.Resultsarereportedasleast
theruminalandGImicrobialdiversityinruminantsassoci-
squaremeans.
atedwithbacteriaandarchaeapopulations,theresponseof
Quality trimmed sequences were provided with the
the microbialconsortiumto feeding variousCT-containing
sequencingservicesbytheResearchandTestingLaboratory
diets remains largely unknown. The principal objective of
(Lubbock,TX)[16].Tagswhichdidnothave100%homology
this study was to assess the effects of CT-containing PB
to the original sample tag designation were not included
supplementation as a feed replacement on gastrointestinal
in data analysis. Sequences which were less than 250bp
methanogenic archaea and bacterial diversity dynamics in
after quality trimming were not also considered. The B2C2
meat goats. The most significant findings in the present
software[17],whichisdescribedandfreelyavailablefromthe
study was a decreased predominant fecal methanogenic
ResearchandTestingLaboratory,wasusedtodepletesamples
genera Methanobrevibacter population (comprising up to
of definite chimeras. Further processing and operational
75%ofthearchaeapopulation)asthelevelofPBsupplement
taxonomic units (OTUs) based analyses were then carried
increasedinthediet.Thebacterialdistributionshowedthat
out using the MOTHUR v.1.19.4 [18] suite of algorithms
proteobacteria was the most dominant phyla with mean
for sequence processing and diversity analysis, including
relativeabundancevaluesrangingfrom46.5%incontrolto
commands for identifying/consolidating unique sequences,
47.1%in15%PBandto39.7%in30%PBdiets.
filtering, multiple sequence alignment, generating distance
matrices,andclusteringofsequencesintoOTU.Theresulting
clusterswereassessedat3%and5%dissimilaritytoprovide 3.1.DietComposition. Ingredientsandchemicalcomposition
thedataneededfordiversityanalysis.Baseduponrarefaction of experimental diet, PB, WS, and BGH are presented in
[19],itisexpectedthat0%dissimilarityinsequencesprovides Table1. Goats were provided diets that met all animals’
dramatic overestimation of the species present in a sample. requirements for growth and gain according to NRC [12].
Theresultingsequenceswerethenevaluatedusingtheclas- TotalCTconcentrationinthePBandWSwas10.3and0.03%
sify.seqsalgorithm(Bayesianmethod)inMOTHURagainst DM,respectively.However,grainmixescontained0.19,1.63,
adatabasederivedfromtheGreengenessetusingabootstrap and3.20%CTDMforthe0,15,and30%PBdiet,respectively.
cutoff of 65%. The sequences contained within the curated All the experimental treatments provided similar nutrient
6 InternationalJournalofMicrobiology
100 4.0 5.4 2.9 90
10.6 9.0
14.4
32.0
80 42.8 75 80.09
42.3
8.2
5.8 69.90
%)
Relative abundance ( 4600 41232612.....36538 41134742.....72151 32997...971 %ative abundance () 4650 59.98
Rel 30
20 21.8 36.1
28.9
30.1 15
21.2 21.0
0
0% 15% 30%
Diet (% pine bark) 0
0% 15% 30%
Sphingobacteria Clostridia Diet (% pine bark)
Bacilli Bacteroidia
Gammaproteobacteria Proteobacteria Figure 4: Relative abundance of major taxonomic group (Eur-
Flavobacteria Firmicutes yarchaeota)acrosspinebark(PB)dietarysupplementationingrow-
Betaproteobacteria Bacteroidetes ingmeatgoats.Phylaincludedinthisfigurehadrelativeabundance
valuesconsistentlygreaterthan1%.Valuespresentedarethemean
Figure 3: Relative abundance of major taxonomic groups across
percent.0%=0%PB(control),15%=15%PB,and30%=30%PB
pine bark (PB) dietary supplementation in growing meat goats.
onanas-fedbasis.
Phylaincludedinthisfigurehadrelativeabundancevaluesconsis-
tentlygreaterthan1%,aswellastheabundantclassesofthephylum
Proteobacteria.Valuespresentedarethemeanpercentage.0%=0%
PB(control),15%=15%PB,and30%=30%PBonanas-fedbasis. Smithet al.[9]reportedthatfecalbacterialpopulationwas
affectedbythepresenceofCTintheratgastrointestinaltract.
When considering richness observations and estimates
profiles,exceptCTandaciddetergentligninthatwerehigher for archaeal community in the study, results demonstrate
in15and30%PBration.Inourpreviousstudy,additionofPB a vastly different trend (Figures 2(a) and 2(b)). Mean val-
ingoatdietsimprovedADGandfavorablymodifiedruminal ues for observed richness were consistently higher in the
fermentation[11]. control treatment for all of the richness calculators. The
trend observed of OTUs the highest mean value for the
control (181), followed by the 15% diet (164), while the
3.2.RichnessandDiversityEstimates. Intherichnessanddi-
30% diet remained the lowest (135; Figure2(a)). The same
versity data in bacteria and archaea was presented in
trend was observed in the Chao and ACE estimates, as
Figures 1 and 2, respectively. For bacteria, the maximum
wellastheOTUsobservedinrarefaction.Furthermore,the
OTUs observed across the study were detected in control
Chao estimate appears to show the greatest coverage of the
group, although the highest mean of bacterial richness was
rarefaction peak values. The microbial community in the
detectedinthetreatmentgroupfedthe15%PBdiet(782.6;
30%PBsupplementedgroupconsistentlyshowedthelowest
Figure1(a)).MaximumOTUswerepredictedforCHAOand
values for richness in bacteria (Figures 1(a) and 1(b)) and
ACE estimates with similar trends for the same sample as
well. The trend of impact of PB in diet (15% > 30% > 0%) archaea(Figures2(a)and2(b)).
IthasbeenreportedthatthePBcontainedhighlevelof
was found in richness observations (sobs) that showed that
CT [11] and polyphenolic compounds [25]. This chemical
the highest concentration of PB in the goat diet resulted in
disturbance creates a microbial community with distinct
the lowest amount of bacteria detected in the feces while
abioticandbioticcharacteristics,whichimpactsthegrowth
thecontroland15%dietshowedverylittledifference.When
andstabilityofgastrointestinalmicrobialcommunity[26].
calculating richness with Chao and ACE, the distinction
between the treatments is more explicit as the 30% diet
remained the lowest, followed by the control and then the 3.3. Relative Abundance of Bacterial and Archaea Phyla.
15%diet.Incalculatingrarefactioncurvesrichnessestimates Bacterial(Figure3)andarchaeal(Figure4)communitycom-
was found to be within Figure1(b). It was hypothesized positions of the feces were examined at descending levels
thatbacterialcommunityrichnessunderCT-containingPB ofbiologicalclassificationtoweretodeterminetheeffectof
supplementation would differ due to inhibition of bacterial PB supplementation on community membership. Detailed
growth[23]andreducefiberdigestibilityintherumen[24]. phylogenic analyses grouped the fecal bacteria associated
InternationalJournalofMicrobiology 7
bacterial sequences into 20 phyla (including unknown). 9
The relative abundances of the 10 most abundant phyla are
presentedinFigure3.Thebacterialdistributionshowedthat
Proteobacteria was the most dominant phylum with mean 7.5
relative abundance values ranging from 46.5% in control
to 47.1 in 15% PB and to 39.7% in 30% PB diets. They
include a wide variety of pathogens, such as Escherichia, %) 6
Salmonella, Vibrio, Helicobacter, and many other notable ce (
n
a
genera[27].Othersarefree-living,andincludemanyofthe d
bacteria responsible for nitrogen fixation. Other dominant bun 4.5
a
phyla were Gammaproteobacteria (representing 32.0–42.8% ve
of the bacterial sequences in the samples) and Firmicutes ati
el 3
and Bacteroidetes (which represented 21.0 to 36.1% of the R
bacterial sequences in each sample). The remaining phyla
accounted for fewer than 25% of the relative abundance
1.5
observed and were designated as minor. Of these groups,
Gammaproteobacteria(𝑃 < 0.05),Flavobacteria(𝑃 < 0.01),
Proteobacteria (𝑃 < 0.05), and Bacteroidetes (𝑃 < 0.05) 0
werelinearlydecreased(𝑃 < 0.05)withincreasingPBdiets. 1 2 3 4 5 6 7 8 10 11 12 13 14 15 16 17 18
However, Clostridia (𝑃 < 0.01) and Firmicutes (𝑃 < 0.05) Sample number
Taxa
werelinearlyincreasedwithincreasingPBsupplementation.
Desulfurococcaceae (genus unknown) Archaeoglobus
ThishasbeenconfirmedbythefindingsthatFirmicutes,Bac-
Methanobacteriaceae (genus unknown) Mc18
teroidetes, Actinobacteria, and Proteobacteria were reported
Methanosphaera Thermofilum
tobedominantbacterialphylainthehumangut[28].
The trend with respect to the one dominant archaea Figure5:The6mostabundantarchaeagenera(Methanobrevibac-
communitygroupcanbeseeninFigure4,inwhichtheclass ter,Desulfurococcaceae,Methanosphaera,Archaeoglobus,Mc18,and
of Euryarchaeota population was linearly decreased (𝑃 < Thermofilum) in the hindgut of various levels of pine bark (PB)
0.01) with increasing PB diets. Collectively, the observed dietsfromacommoncohortof17meatgoats.Clusteringinthe𝑌
directionisindicativeofabundance,notphylogeneticsimilarity.0%
changes in frequency of occurrence depict dramatic shifts
PB(control)=sample(ortag)number1,2,3,4,5,and6;15%PB=
in fecal microbial communities associated with changes in
tagnumber7,8,10,11,and12;30%PB=tagnumber13,14,15,16,17,
diet in goats. It has been classified that Euryarchaeota, one
and18.
of the meanwhile four kingdoms (or phyla) of the archaeal
domain, consists of the strict anaerobic methanogens, the
extreme halophiles, the hyperthermophilic Archaeoglobales
3.4.DiversityandAbundanceofFecalBacteriaandArchaea.
andThermococcales,andthecellwallfreeThermoplasmatales
Foreaseofpresentationandinterpretation,wepresentpreva-
[29,30].
lentbacterialgenera(Figure5)andarchaealgenera(Figure6)
Tannins are phenolic plant secondary compounds and
communitybasedonacutoffvalueof0.9%ofrelativecom-
are widely distributed through the plant kingdom which
munityabundanceforinclusioninadoubledendogramclus-
affect animal performance and gut microorganisms. Tan-
ter analysis of individual animal microbial diversity within
nins exist primarily in CT and hydrolysable tannins [7].
andamongdietsinFigures5and6.Overall,animalsclustered
The hydrolysable tannins occur mainly in fruit pods and
relativelywellwithindietandanimals.However,15%PBdiet
plant galls and unlike CT their degradation products are
ofone(tagnumber12)animalclusteredmoreclosedtothe
absorbed from the small intestine of animals [7] and are
controldiet(0%PB).Weattribute“misclassification”ofthis
potentiallytoxictoruminants.In contrast,CT,unlikemost
animal to greater relative abundance of Xanthomonadales,
compounds,isonlypartiallyabsorbedinthesmallintestine
Clostridiales, Pseudomonadales, and Flavobacteriales and
in ruminants [31]. Most of the dietary CT reach the colon
lower abundance (%) of Mycoplasmatales and Bacteroidales
where simple phenolics and phenols (e.g., phenylacetic,
comparedwithanimalswhoreceivedPBdiet.Similardiver-
mono- and dihydroxyphenylacetic acids, and protocate-
sitytrendsinarchaealpopulationswereobservedonthesame
cuic acid) are produced by breakdown due to microbial
diets.Forarchaea,microbialarchaeaclusteredvarybetween
activity [32]. Ultimately, these simple phenolics can be
diets, but we classified these animals to greater relative
further metabolized to nonaromatics, such as short-chain
abundance of Methanobacteriales and lower abundance of
fatty acids, lactate, succinate, ethanol, and the gases (CO2
ArchaeoglobusandThermofilumbetweendiets.
and H2) by the colonic microorganisms [28]. Therefore,
the bacterial characterization of the gut microorganisms
is surely important, but ultimately the enzymatic activ- 3.5. Quantitative Analyses. So far, only a limited number
ity of its microorganisms becomes vital to map func- of bacterial species have been identified as being involved
tional metabolic reactions and describe the interaction in the metabolism of CT and polyphenols, and little is
between host and its microbe to understand metabolism known about the microbial populations in the hindgut of
[28]. thegoat.Themetabolicactivitiesofthebacteriaandarchaea
8 InternationalJournalofMicrobiology
Col1
RA Distance
55.78 2.00
0.77 1.50
0.03 1.00
0.00 0.50
0.00 0.00
Xanthomonadales
Clostridiales
Pseudomonadales
Flavobacteriales
Bacillales
Enterobacteriales
Burkholderiales
Sphingobacteriales
Lactobacillales
Actinomycetales
Spirochaetales
Rhodobacterales
Caulobacterales
Rhizobiales
Sphingomonadales
Oceanospirillales
Aeromonadales
Erysipelotrichales
Cw040
Kiloniellales
Verrucomicrobiales
Rickettsiales
Coriobacteriales
Roseiflexales
Pasteurellales
Desulfuromonadales
es Deinococcales
abl Fusobacteriales
Vari LPeirnetlilsuplahlaeesrales
Acholeplasmatales
Fibrobacterales
Synechococcales
Spartobacteriales
Chromatiales
Bifidobacteriales
Pseudanabaenales
Gemmatimonadales
Gn09
Anaeroplasmatales
Nitrospirales
Haloplasmatales
Alteromonadales
Rhodospirillales
Turicibacterales
Chroococcales
Methylomirabiliales
Methylococcales
Campylobacterales
Ws3candidatedivision
Desulfovibrionales
Mycoplasmatales
e
c Bacteroidales
n
Dista3.502.631.750.880.00 1 4 12 3 2 5 17 10 7 8 11 15 18 16 13 14 6
Figure6:Thermaldoubledendrogramofthe53mostabundantbacterialgenerainthehindgutofvariouslevelsofpinebark(PB)diets
fromacommoncohortof17meatgoats.Clusteringinthe𝑌directionisindicativeofabundance,notphylogeneticsimilarity.RA=relative
abundance;0%PB(control)=tagnumber1,2,3,4,5,and6;15%PB=tagnumber7,8,10,11,and12;30%PB=tagnumber13,14,15,16,17,
and18.
in the gut are very important in determining the healthy therelativeabundanceobserved.Ofthesegroups,Flavobac-
functioning and methane gas emissions from the gut. For terium gelidilacus (𝑃 < 0.02) and Myroides odoratimimus
furtherquantificationofmajorhindgutbacterialandarchaeal (𝑃 = 0.09) were decreased (𝑃 < 0.05) with increasing
speciesdiversitypopulationsinmeatgoats(𝑛 = 6)arepre- dietary PB concentration. However, Bacteroides capillosus
sented in Tables 3 and 4, respectively. For bacterial species, (𝑃 < 0.02), Clostridium orbiscindens (𝑃 < 0.03), and
Stenotrophomonaskoreensiswasthemostdominantbacterial Oscillospiraguilliermondii(𝑃=0.06)werelinearlyincreased
species with mean relative abundance values ranging from with increasing PB concentration. This finding agrees with
23.9% (control) to 9.9% (15% PB) and 17.2% (30% PB). The the results of a metabolic finger print patterns study of a
remainingbacterialspeciesaccountedforfewerthan10%of rat fed CT extract from Acacia angustissima. Condensed
InternationalJournalofMicrobiology 9
Table3:Quantificationofmajorhindgutbacterialdiversitypresentinmeatgoat(n=6)asanalyzedusingbTEFAPpyrosequencingmethod1.
Diets2 Pvalues3
Item SEM
Control 15%PB 30%PB L Q
Acinetobacterhaemolyticus 1.1 1.2 0.2 0.06 0.32 0.54
Acinetobacterlwoffii 7.0 6.5 0.3 3.70 0.22 0.56
Acinetobacterradioresistens 1.4 4.2 1.1 0.92 0.80 0.03
Acinetobacterrhizosphaerae 1.1 1.5 0.9 0.47 0.76 0.39
Acinetobacterschindleri 3.4 8.5 5.3 0.82 0.46 0.11
Bacteroidescapillosus 0.5 1.1 2.7 0.60 0.02 0.52
Clostridiumorbiscindens 0.02 1.4 2.3 0.48 0.03 0.92
Comamonasaquatica 2.4 3.9 4.7 1.43 0.25 0.88
Enterobacterhormaechei 0.8 3.3 2.0 0.95 0.42 0.16
Escherichiaalbertii 0.6 1.5 0.9 0.53 0.73 0.25
Escherichiafergusonii 0.6 1.9 1.2 0.60 0.49 0.18
Flavobacteriumcolumnare 0.4 2.0 0.5 0.83 0.91 0.18
Flavobacteriumgelidilacus 5.81 1.2 1.7 1.14 0.02 1.10
Flavobacteriumsuccinicans 1.7 0.2 1.3 1.80 0.58 0.15
Myroidesodoratimimus 7.8 4.5 1.2 2.58 0.09 0.99
Oscillospiraguilliermondii 0.99 1.2 3.0 0.69 0.06 0.38
Sphingobacteriumfaecium 1.2 4.4 0.53 1.09 0.83 0.16
Sphingobacteriummizutaii 1.8 0.6 1.4 0.41 0.40 0.07
Stenotrophomonaskoreensis 23.9 9.9 17.2 7.40 0.53 0.28
Rummeliibacilluspycnus 8.54 9.9 6.9 0.35 0.75 0.63
1
Datawerepresentedwithacutoffvalueof0.9%.
2Control(0%PB),15%PB,and30%PBonanas-fedbasis.
3Polynomialcontrastsforequallyspacedtreatments.
L:lineareffect;Q:quadraticeffect.
Table4:Quantificationofhindgutmethanogenicarchaeadiversitypresentinmeatgoat(n=6)asanalyzedusingbTEFAPpyrosequencing
method1.
Diets2 Pvalues3
Item SEM
Control 15%PB 30%PB L Q
Methanobrevibacterspp. 74.8 71.7 49.6 8.84 0.05 0.43
Methanosphaeraspp. 1.18 0.61 0.74 0.45 0.51 0.57
Methanobacteriaceaespp. 3.35 2.31 3.44 1.01 0.95 0.44
1
Datawerepresentedwithacutoffvalueof0.9%.
2Control(0%PB),15%PB,and30%PBonanas-fedbasis.
3Polynomialcontrastsforequallyspacedtreatments.L:lineareffect;Q:quadraticeffect.
tannin extracts from A. angustissima altered fecal bacterial concentratedgreenteawerenecessaryfortheretobeagreat
populations in the gastrointestinal tract, resulting in a shift decreaseinClostridiumspeciesandClostridiumperfringens.
inthepredominantbacteriatowardstannin-resistantgram- Othergroupsofbacteria,includingtheBacteroidaceae,were
negativeEnterobacteriaceaeandBacteroidetes[9].Incontrast, not affected [29, 30]. However, Moco et al. [28] reported
the gastrointestinal microbial population was significantly that diet containing polyphenols (e.g., +catechin) increased
different in cattle, dominated by Prevotella (18.2% of total thegrowthoftheClostridiumcoccoides,Eubacteriumrectale,
population) in the rumen and Clostridium (19.7% of total Bifidobacterium spp., and Escherichia coli, but significantly
population)inthefeces[29].Interestingly,themajorityofthe decreasedthegrowthoftheClostridiumhistolyticumgroup.
bacteriainthehumangutrecordedbelongstotheClostridia In chickens the levels of cecal lactobacilli increased signifi-
group,whichisalargecomponentofthegutmicrobiota[28]. cantly, while the levels of Enterobacteriaceae decreased [33,
Condensedtanninsfromgreenteacatechins,whichare 34].AllthesestudiesconfirmedthatCTcauseashiftinthe
monomeric polyphenols, have also been shown to cause a bacterial population in the intestinal tract, and a consistent
shift in bacteria populations. Polyphenolics from green tea resultisthatgram-positivebacterialgroupsaredecreased.
affect gastrointestinalbacteria in humans[28, 30], chickens Therelativeabundancesofthe4mostabundant(>0.9%)
[33, 34], and pigs [35]. In the human study, 4 weeks of archaea species are presented in Table4. Predomi-
consuming a polyphenolic content equivalent to 10 cups of nant hindgut archaeal species among methanogens were
10 InternationalJournalofMicrobiology
Methanobrevibacter(75,72,and49%),Methanosphaera(3.3, [5] D.W.Pitta,W.E.Pinchak,S.E.Dowdetal.,“Rumenbacterial
2.3,and3.4%),andMethanobacteriaceae(1.2,0.6,and0.7%) diversity dynamics associated with changing from bermuda-
population in control, 15, and 30% PB, respectively, and grasshaytograzedwinterwheatdiets,”MicrobialEcology,vol.
theywerelinearlydecreasedorincreasedwithincreasingPB 59,no.3,pp.511–522,2010.
concentration(𝑃 < 0.05).TheeffectofCTinforagesupon [6] A.N.Hristov,T.R.Callaway,C.Lee,andS.E.Dowd,“Rumen
interactions between rumen bacteria and plant protein has bacterial, archaeal, and fungal diversity of dairy cows in re-
beenproposed[7].Ithasbeensuggestedthattheproteolysis sponsetolauricormyristicacidsingestion,”JournalofAnimal
ofsolubleproteinsintherumenisaffected,primarilybycell- Science,vol.90,pp.4449–4457,2012.
associatedproteinasesontherumenbacteria.Furthermore, [7] B. R. Min, T. N. Barry, G. T. Attwood, and W. C. McNabb,
CT in the diets have been shown to induce changes in “Theeffectofcondensedtanninsonthenutritionandhealthof
morphology of several species of rumen bacteria [7]. ruminantsfedfreshtemperateforages:areview,”AnimalFeed
InhibitionofrumenmicroorganismsbyCTisprobablydue ScienceandTechnology,vol.106,no.1-4,pp.3–19,2003.
toCT:substrate(microorganisms)interactions[7].Oneofthe [8] B.R.Min,G.T.Attwood,K.Reillyetal.,“Lotuscorniculatus
possibleexplanationsforthedecreaseinMethanobrevibacter condensedtanninsdecreaseinvivopopulationsofproteolytic
spp. prevalence is assumed to be linked to the substantial bacteriaandaffectnitrogenmetabolismintherumenofsheep,”
relationshipbetweenthesearchaeaandprotozoa[6]andthe CanadianJournalofMicrobiology,vol.48,no.10,pp.911–921,
protozoa and subsequent reduction in reducing equivalent 2002.
(H2) cross-feeding between archaea and protozoa [36]. [9] A.H.Smith,J.A.Imlay,andR.I.Mackie,“Increasingtheox-
Condensedtanninsareknowntodecreaseprotozoalnumber idative stress response allows Escherichia coli to overcome
[37],andthedecreaseinmethaneproductioncouldalsobe inhibitoryeffectsofcondensedtannins,”AppliedandEnviron-
mediatedthroughdecreaseinprotozoalnumber.Tavendale mentalMicrobiology,vol.69,no.6,pp.3406–3411,2003.
et al. [38] suggested two models of impacting methano- [10] S. E. Dowd, T. R. Callaway, R. D. Wolcott et al., “Evaluation
genesis: first, directly affecting activity or population of of the bacterial diversity in the feces of cattle using 16S
methanogens, resulting in lower methane emission, and, rDNA bacterial tag-encoded FLX amplicon pyrosequencing
second,indirectlybyreducedhydrogenproductionbylow- (bTEFAP),”BMCMicrobiology,vol.8,articleno.125,2008.
ering feed degradation. However, the possible mechanisms [11] B.R.Min,S.Solaiman,N.Gurungetal.,“Effectsofpinebark
and effects of CT-containing diets on gastrointestinal supplementation on performance, rumen fermentation, and
methanogenesis and its mode of action are not clearly carcasscharacteristicsofKikocrossbredmalegoats,”Journalof
understood.Moreinvivostudieswithawiderangeoftannins AnimalScience,vol.90,pp.3556–3567,2012.
sources (types) need to be conducted to evaluate the full [12] National Research Council (NRC), Nutrient Requirement of
potentialoftannins.Inaddition,thereisalackofknowledge Sheep,Goats,CervidsandCamelids,AcademyPress,Washing-
of a systematic and preventive process control from the ton,DC,USA,2007.
anthropogenicemissionsofCH4,H2,andvolatilefattyacids [13] AOAC, Official Methods of Analysis, Association of Official
anditsrelationshipwithmicrobialcommunity. Analytical Chemists, Gaithersburg, Md, USA, 16th edition,
1998.
ConflictofInterests [14] S. E. Dowd, K. Killinger-Mann, M. Brashears, and J. Fralick,
“Evaluationofgeneexpressioninasingleantibioticexposure-
Theauthorsdeclarethattheyhavenoconflictofinterests. derivedisolateofSalmonellaentericaTyphimurium14028pos-
sessingresistancetomultipleantibiotics,”FoodbornePathogens
andDisease,vol.5,no.2,pp.205–221,2008.
Acknowledgment
[15] S.F.Altschul,W.Gish,W.Miller,E.W.Myers,andD.J.Lipman,
“Basiclocalalignmentsearchtool,”JournalofMolecularBiology,
ThisworkwassupportedbytheJ.W.CoverCaprineResearch
vol.215,no.3,pp.403–410,1990.
andEducationCenter,TuskegeeUniversity.
[16] J.R.Cole,B.Chai,R.J.Farrisetal.,“TheRibosomalDatabase
Project (RDP-II): sequences and tools for high-throughput
References rRNAanalysis,”NucleicAcidsResearch,vol.33,pp.D294–D296,
2005.
[1] D.O.KrauseandJ.B.Russell,“Howmanyruminalbacteriaare
[17] V.Gontcharova,E.Youn,R.D.Wolcottetal.,“BlackBoxChi-
there?”JournalofDairyScience,vol.79,no.8,pp.1467–1475,
mera Check (B2C2): a windows-based software for batch
1996.
distinctbacterialcommunitiesinagroecosystem,”OpenMicro-
[2] J.E.Edwards,N.R.McEwan,A.J.Travis,andR.J.Wallace,“16S biologicalJournal,vol.4,pp.47–52,2010.
rDNA library-based analysis of ruminal bacterial diversity,”
[18] P. D. Schloss, S. L. Westcott, T. Ryabin et al., “Introduc-
AntonievanLeeuwenhoek,vol.86,no.3,pp.263–281,2004.
ingmothur:open-source,platform-independent,community-
[3] P.B.Eckburg,E.M.Bik,C.N.Bernsteinetal.,“Microbiology: supported software for describing and comparing microbial
diversityofthehumanintestinalmicrobialflora,”Science,vol. communities,”AppliedandEnvironmentalMicrobiology,vol.75,
308,no.5728,pp.1635–1638,2005. no.23,pp.7537–7541,2009.
[4] R. E. Ley, D. A. Peterson, and J. I. Gordon, “Ecological and [19] L.F.W.Roesch,R.R.Fulthorpe,A.Rivaetal.,“Pyrosequencing
evolutionaryforcesshapingmicrobialdiversityinthehuman enumerates and contrasts soil microbial diversity,” The ISME
intestine,”Cell,vol.124,no.4,pp.837–848,2006. Journal,vol.1,no.4,pp.283–290,2007.