Nathani et al.

AMB Express 2013, 3:55

http://www.amb-express.com/content/3/1/55

ORIGINAL ARTICLE Open Access

Comparative evaluation of rumen metagenome

community using qPCR and MG-RAST
Neelam M Nathani1,3, Amrutlal K Patel1, Prakash S Dhamannapatil1, Ramesh K Kothari2, Krishna M Singh1
and Chaitanya G Joshi1*

Abstract
Microbial profiling of metagenome communities have been studied extensively using MG-RAST and other related
metagenome annotation databases. Although, database based taxonomic profiling provides snapshots of the
metagenome architecture, their reliability needs to be validated through more accurate methods. Here, we
performed qPCR based absolute quantitation of selected rumen microbes in the liquid and solid fraction of the
rumen fluid of river buffalo adapted to varying proportion of concentrate to green or dry roughages and compared
with the MG-RAST based annotation of the metagenomes sequences of 16S r-DNA amplicons and high throughput
shotgun sequencing. Animals were adapted to roughage-to-concentrate ratio in the proportion of 50:50, 75:25 and
100:00, respectively for six weeks. At the end of each treatment, rumen fluid was collected at 3 h post feeding.
qPCR revealed that the relative abundance of Prevotella bryantii was higher, followed by the two cellulolytic bacteria
Fibrobacter succinogens and Ruminococcus flavefaciens that accounted up to 1.33% and 0.78% of the total rumen
bacteria, respectively. While, Selenomonas ruminantium and archaea Methanomicrobiales were lower in microbial
population in the rumen of buffalo. There was no statistically significant difference between the enumerations
shown by qPCR and analysis of the shotgun sequencing data by MG-RAST except for Prevotella. These results
indicate the variations in abundance of different microbial species in buffalo rumen under varied feeding regimes
as well as in different fractions of rumen liquor, i.e. solid and the liquid. The results also present the reliability of
shotgun sequencing to describe metagenome and analysis/annotation by MG-RAST.
Keywords: Real time PCR; MG-RAST; Bubalus bubalis; Fibrobacter succinogens; Ruminococcus flavefaciens;
Prevotella bryantii; Selenomonas ruminantium; Methanomicrobiales

Introduction microscopic enumeration (Veal et al 2000). Rapid enu-

Culture dependent methods for enumerating bacterial meration of bacteria is also possible using a variety of
abundance are known to be biased since it is possible to molecular approaches (Hugenholtz et al. 1998; von
cultivate bacteria only if their metabolic and physiological Wintzingerode et al. 1997).
requirements can be fulfilled in vitro. Fluorescence-based Molecular methods have shown that the complexity
methods like the flow cytometry can also be used to of microbial communities is much greater and that the
enumerate bacteria (Veal et al. 2000). However, most majority of rumen microbes are still unknown (Leser et al.
bacteria are observed to be optically similar and thus 2002; Pryde et al. 1999) due to the failure of many bacteria
require artificial modification of the target bacteria to grow in a given culture medium (Huijsdens et al. 2002;
using fluorescent labeling (Attfield et al. 1999). Differences Langendijk et al. 1995). Quantitative molecular methods
in bacterial size and the presence of different contaminating could be more sensitive and selective than traditional
matrices (e.g. feed particles, soil) may also interfere in methods as they do not rely on the ability of bacteria to
grow. Primers with broad interspecies specificity have
been designed to amplify 16S rDNA by PCR and have
* Correspondence: [email protected]
1
Department of Animal Biotechnology, College of Veterinary Science & been used to determine bacterial abundance in complex
Animal Husbandry, Anand Agricultural University, Anand, Gujarat 388 001, communities (Klausegger et al. 1999; Marchesi et al. 1998;
India Suzuki et al. 2000). qPCR, such as the SYBR green
Full list of author information is available at the end of the article

© 2013 Nathani et al.; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction
in any medium, provided the original work is properly cited.
Nathani et al. AMB Express 2013, 3:55 Page 2 of 8
http://www.amb-express.com/content/3/1/55

system, allows rapid detection and quantification of Overbeek et al. 2005). The server provides several methods
DNA (Heid et al. 1996). In addition, the in-built 96-well to analyze the different data types, including phylogenetics
format greatly increases the number of samples that can be and the ability to compare one or more metagenomes.
simultaneously analyzed. Conserved regions of 16S rDNA The objective of this work was to determine the bacterial
provide the means for detecting and enumerating complex quantification in rumen samples and to compare the qPCR
bacterial populations by qPCR. The final determination of results with those obtained by using MG-RAST server for
bacterial load by real-time PCR in a multi-species popula- metagenomic analysis. We hypothesized that bacteria,
tion may be influenced by the variation in the number of including total bacteria, and anaerobic bacteria, could be
rRNA operons in a given species (Farrelly et al. 1995). reproducibly quantified from rumen samples using qPCR,
However, in a variety of complex environmental, industrial and it would have comparable sensitivity and specificity to
and health-related situations in which multi-species popu- MG-RAST predicted quantities. We chose Ruminococcus
lations are sampled along with impurities, other method- flavefaciens, Fibrobacter succinogens, Prevotella bryantii,
ologies are likely to be far less sensitive or precise. This Selenomonas ruminantium and methanomicrobiales tar-
information is thus applied to determine the anaerobic gets as these are the most common microorganisms present
bacteria load in the community samples. Moreover, DNA- in the rumen microflora. The panel of anaerobic bacteria
based methods offer the option of storing samples until was chosen based on metagenomic shotgun sequencing
their analysis, which could be an important advantage. data from our laboratory as they are frequently present in
The complex symbiotic microbiota of the rumen is buffalo rumen metagenome samples.
responsible for breakdown of plant fiber, an ability the
ruminal host animal lacks. This microbiota is highly
responsive to changes in diet, age, antibiotic use, and Materials and methods
the health of the host animal and varies according to Sample collection
geographical location, season, and feeding regimen (Bryant The experiment for metagenomic analysis of ruminal mi-
1959; Tajima et al. 2001). Earlier rumen microbiology used crobes was carried out on 8 Mehsani buffalo reared at
cultivation-based techniques for monitoring such changes, Livestock Research Station, Sardarkrushinagar Dantiwada
but recent developments of more accurate molecular Agricultural University, Gujarat. Four animals were fed on
detection methods have brought new improvements in 50:50, 75:25 and 100:00 dry roughage to concentrate ratio.
the field (Tajima et al. 2001). The availability of isolates Another four animals of the same group were fed on 50:50,
representing the main groups of general diversity offers 75:25 and 100:00 green roughage to concentrate ratio over
clear advantages with the possibility of assigning the uniform feeding regime of six weeks. Samples were col-
functional role in the rumen based on the metabolic lected on the last day (42 days dietary period), 3 hours
properties of pure cultures. Detection probes or PCR post feeding using flexible stomach tube and vacuum
primers for rumen bacteria have been used for detection pump from rumen of the buffalo. The rumen fluid was
and quantification of the corresponding species in the then passed through muslin cloth, without pressure for
rumen by qPCR approach. High complexity of rumen removal of particulate matter, and the filtrate aliquoted
microflora has led researchers to find particular microbial as fractions of 200 μl in cryo vials was stored at −80°C
groups that could serve as an indicator of health promoting for further study. About 200 mg of solid fraction, after
microbiota. Studies have consistently reported a high preva- removing liquid by squeezing, was aliquoted in sterile
lence of anaerobic bacteria within the complex rumen tubes and then immediately frozen into liquid nitrogen
microbial community. (Additional file 1: Online Resource 1).
Recently, random community genomics, or metageno-
mics, where DNA is sequenced directly from environ-
mental samples has provided insights into microbial DNA extraction and PCR amplification of 16S RNA genes
communities. DNA is sequenced without cloning, using Total DNA was extracted separately by using commer-
next-generation sequencing techniques. Regardless of the cially available QIAmp DNA stool mini kit (Qiagen, USA).
sequencing approach used, first step in analysis of any The total DNA mixture was used as a template in PCR
metagenome involves comparing the sequences to known to amplify 16S rDNA. Species specific PCR primers
sequence databases. Subsequent analysis including phylo- were used for the amplification of target region of the
genetic comparisons, functional annotations, binning of 16S rRNA genes. The target DNA of total bacteria,
sequences, phylogenomic profiling, and metabolic recon- Fibrobacter succinogens, Prevotella bryantii, Ruminococcus
structions need to be computed. MG-RAST server is a flavefaciens, Selenomonas ruminantium and archaea were
freely available, fully automated open source system for amplified from the metagenomic DNA, as described
processing metagenome sequence data based on the SEED previously (Koike and Kobayashi 2001; Muyzer et al.
framework for comparative genomics (McNeil et al. 2007; 1993; Tajima et al. 2001).
Nathani et al. AMB Express 2013, 3:55 Page 3 of 8
http://www.amb-express.com/content/3/1/55

Preparation of DNA standard for absolute quantification which the threshold line crosses the amplification curve).
by qPCR Upon completion of real-time PCR run, data were auto-
Plasmid DNA containing the respective target DNA matically analyzed for melt curve and quantification by
fragment was used as DNA standard in qPCR. The target 7500 system Sequence Detection Software (SDS).
DNA was amplified using the species-specific primer sets
(Table 1). After confirmation of the expected size amplifi-
Shotgun sequencing
cation on an agarose gel, the PCR products were excised
The DNA samples were subjected to sequencing based on
from the gel, purified using the QIAquick gel extraction
ion semiconductor technology using Ion Torrent PGM
kit (Qiagen, CA), and then ligated into pTZ57RT cloning
as per the manufacturer’s instructions. The steps in brief
vector (Fermentas, UK). The ligated products were trans-
included enzymatic fragmentation of DNA to obtain
formed to competent Escherichia coli DH5 alpha cells
fragments in the range of 280 to 300 bp size. The desired
by heat shock. Plasmids were purified from positive clones
size fragments were ligated with the library adaptors and
using a QIAprep spin miniprep kit (Qiagen, CA), and
subjected to emulsion PCR, followed by its recovery and
cloned in pTZ57RT cloning vector using InsTAclone
loading onto Ion Torrent PGM 316 chip.
PCR cloning kit (Fermentas, UK). The recombinants
were screened by colony PCR with respective primer
sets and plasmids from positive clones were purified 16 s r-DNA amplicon pyrosequencing
using a QIAprep spin miniprep kit. The concentration of Specific hypervariable regions in the 16 s rDNA were
the plasmid was determined with a Nanodrop spectropho- targeted for amplicon sequencing. The fusion primers were
tometer (Thermoscientific). Copy number of each standard designed so as to include the priming site for sequencing
plasmid was calculated using formula; Copy No/μl = Con- primer, MIDs, and target specific sequences. DNA from
centration of plasmids (gm/μl) × 6.022 x 1023 / length of liquid and solid fractions was used as template to generate
recombinant plasmid (bp) × 660, (660 = Molecular weight amplicons of targeted hypervariable region. Four sets
of one basepair, 6.022 × 1023 = Avogadro’s number). Ten- of primers viz. 8F and 534R, 517F and 926R, 917F and
fold dilution series ranging from 109 to 102 copies were 1386R and 1099F and 1541R were used to generate
prepared for each target. Quantitative PCR was performed amplicons. The amplification was performed using the
with ABI 7500 FAST real time PCR system (Applied emulsion PCR master mix (Roche diagnostics, USA) in
Biosystems, USA) using QuantiFast SYBR green PCR the presence of 3% DMSO. The expected size amplicons
mastermix (Qiagen, CA). The amplification reactions were were gel purified using QIAquick gel extraction kit.
performed in a total volume of 15 μL, containing 30 ng of The gel purified amplicons were quantified by Nanodrop
template DNA, 7.5 μL of 2X SYBR green master mix, 1000 spectrophotometer. The number of molecules of the
0.5 μL of each primer (10 pmol /μL) and sterile H2O. The amplicons was estimated by molar conversion formula.
cycling conditions consisted of initial denaturation step at The amplicons were diluted to 109 molecules per μl. Equal
95°C for 5 min, followed by 40 cycles of 95°C 15 s and 60°C amount of amplicons from four sets of primers (i.e. 8F &
for 1 min. The 10-fold dilution series of the standard 534R, 517F & 926R, 917F & 1386R and 1099F & 1541R)
plasmid for the respective target was run along with the for each sample (i.e. Rumen Liquid DNA and Rumen Solid
corresponding samples in triplicate. Fluorescence was DNA) were pooled and further diluted to 107 molecules/μl
measured once every cycle after the extension step using concentration. The amplicon library was then subjected
filters for SYBR Green (excitation at 492 nm and emission to emulsion PCR using the GS-FLX Titanium kit (Roche
at 530 nm). The normalized fluorescence data were diagnostics, USA). The recovery and enrichment of the
converted to a log scale and the threshold was determined beads and pyrosequencing was carried out as described
to calculate the threshold cycle value (Ct; the cycle at in the protocol.

Table 1 PCR primers for detection of rumen bacteria

Target bacterium Primer Product
Forward primer Reverse primer Size(bp)
Total bacteria CCTACGGGAGGCAGCAG ATTACCGCCGCTGTTGG 194
Prevotella bryantii ACTGCAGCGCGAACTGTCAGA ACCTTACGGTGGCAGTGTCTC 540
Fibrobacter succinogenes GGTATGGGATGAGCTTGC GCCTGCCCCTGAACTATC 445
Ruminococcus flavefaciens GGACGATAATTGACGGTACTT GCAATCYGAACTGGGACAAT 520
Selenomonas ruminantium TGCTAATACCGAATGTTG TCCTGCATCAAGAAAGA 513
Methanomicrobiales ATCGRTACGGGTTGTGGG CACCTAACGCRCATHGTTTAC 506
Nathani et al. AMB Express 2013, 3:55 Page 4 of 8
http://www.amb-express.com/content/3/1/55

Data analysis studied in the present study in comparison to the other

The sequence data was processed by 454 run processing methods used for quantification (viz. qPCR and shotgun
software to remove the adaptor sequences and to filter out sequencing). On performing One-way ANOVA it was
the low quality sequence reads. The reads corresponding found that the variation between the results (in percentage)
to each amplicon were separated using the Perl script of real time PCR and shotgun sequencing was insignificant
based on MIDs. The high quality reads for each of the for all bacterial species except for the Prevotella bryantii
samples were analyzed by MG-RAST for taxonomic population.
profiling of the metagenomic data. For Fibrobacter succinogens, the proportion observed
was slightly lesser by quantification using qPCR as com-
Statistical analysis pared to the values obtained by the MG-RAST analysis of
One way ANOVA was carried out using SPSS (Statistical the shotgun sequences (Table 2). In spite of the difference
Package for Social Sciences) to check the statistical sig- in values, Fibrobacter abundance was more in the solid
nificance of the results. Spearmen’s correlation test was fraction of fluid in animals fed on green roughage
performed to determine the strength of relationships while in animals fed on dry roughage it was more in
between the qPCR and MG-RAST analysis for samples the liquid fraction of the fluid. One of the other major
sequenced using shotgun (shotgun sequencing/MG-RAST) cellulolytic bacteria usually found in the rumen sam-
and amplicon sequencing (amplicon sequencing/MG- ples, Ruminococcus flavefaciens, was observed to be
RAST), respectively. After performing the statistical more in the solid and liquid fraction of the rumen samples
analysis using the proportions of bacterial species obtained in animals fed with dry roughage compared to green
for all the 48 samples, the mean and standard deviations of roughage (Table 3). While the Prevotella bryantii species
green and dry liquid as well as solid samples of four animals showed dominant prevalence among the four bacterial
were calculated for interpretation of results. species quantified by us, and was in higher proportion
in the liquid fraction of both green and dry fed animals
Results of all three roughage combinations as shown by qPCR and
Real time PCR MG-RAST results (Figure 1). Comparative abundance ana-
CT values decreased at least three fold on each dilution lysis of the two cellulolytic bacterial species F. succinogens
between 109 and 102 molecules/μl which were used as and R. flavefaciens obtained by qPCR with the taxonomic
standards. In the reaction for all standard, nearly perfect distribution as predicted by MG-RAST on the basis of
linear regressions, intercept and slope were obtained ribosomal RNA genes annotation showed ambiguous
between threshold cycle and quantities of standard for all results (Additional file 3: Online Resource 3 and Additional
targets and data generated from the reaction were used file 4: Online Resource 4).
for further analysis. We also quantified the firmicutes Selenomonas ruminan-
tium bacteria and archaea methanomicrobiales in the
Enumeration of 4 bacterial species; an archaea and total
bacterial population Table 2 Comparison of Fibrobacter succinogens
Rumen fluid of buffalo, fed on varied concentrations of percentage by qPCR and MG-RAST analysis for same
green and dry roughage, and fractionated into liquid and samples processed by shotgun and amplicon sequencing
solid, were compared for bacterial abundance by qPCR Sample Fibrobacter succinogens (% ± SD)
and the results were compared with the MG-RAST based qPCR Shotgun Amplicon
analysis of the same samples sequenced using shotgun 50% roughage GL 0.33 ± 0.005 0.90 ± 0.001 0.33 ± 0.002
and amplicon sequencing methods. The total bacteria in DL 0.35 ± 0.001 0.70 ± 0.000 0.70 ± 0.004
the rumen were estimated to be >106 (detailed values in GS 0.20 ± 0.001 0.43 ± 0.0005 0.35 ± 0.001
Additional file 2: Online Resource 2). Statistical analysis
DS 0.11 ± 0.0007 0.38 ± 0.0005 0.37 ± 0.001
showed that the proportion of bacterial species detected
by qPCR was quite similar to that shown by shotgun 75% roughage GL 0.40 ± 0.003 1.23 ± 0.005 3.00 ± 0.000
sequencing, while amplicon sequencing showed significant DL 0.38 ± 0.003 1.63 ± 0.011 1.75 ± 0.009
variation compared to the qPCR results except for GS 0.10 ± 0.0004 0.50 ± 0.0008 0.70 ± 0.000
Ruminococcus flavefaciens. A positive correlation was ob- DS 0.02 ± 0.000 0.40 ± 0.0008 0.28 ± 0.002
served between qPCR and shotgun sequencing (Spearman 100% roughage GL 0.20 ± 0.002 0.53 ± 0.001 0.55 ± 0.311
rank correlation; r = 0.772, p < 0.01) whereas negative cor-
DL 0.78 ± 0.003 2.00 ± 0.000 3.00 ± 0.000
relation was seen between PCR and amplicon sequencing
analysis (Spearman rank correlation; r = −0.382, p < 0.01). GS 0.06 ± 0.0005 0.38 ± 0.002 0.18 ± 0.112
Majority of samples processed by amplicon sequencing DS 0.01 ± 0.000 0.28 ± 0.0005 0.09 ± 0.020
showed much higher proportions of the bacterial species GL = Green liquid, DL = Dry liquid, GS = Green solid, DS = Dry solid.
Nathani et al. AMB Express 2013, 3:55 Page 5 of 8
http://www.amb-express.com/content/3/1/55

Table 3 Comparison of Ruminococcus flavefaciens Table 4 Comparison of Selenomonas ruminantium

percentage by real-time PCR and MG-RAST analysis for percentage by qPCR and MG-RAST analysis for same
same samples processed by shotgun and amplicon samples processed by shotgun and amplicon sequencing
sequencing Selenomonas ruminantium (% ± SD)
Ruminococcus flavefaciens (% ± SD) Sample qPCR Shotgun Amplicon
Sample qPCR Shotgun Amplicon 50% roughage GL 0.063 ± 0.0003 0.003 ± 0.000 0.400 ± 0.003
50% roughage GL 0.45 ± 0.002 0.18 ± 0.0009 0.35 ± 0.002 DL 0.040 ± 0.0001 0.005 ± 0.000 0.430 ± 0.001
DL 0.45 ± 0.002 0.10 ± 0.000 0.30 ± 0.0008 GS 0.006 ± 0.000 0.004 ± 0.000 0.750 ± 0.002
GS 1.33 ± 0.006 0.50 ± 0.0008 0.98 ± 0.0005 DS 0.008 ± 0.000 0.003 ± 0.000 1.000 ± 0.000
DS 0.58 ± 0.004 0.53 ± 0.0009 1.07 ± 0.008 75% roughage GL 0.090 ± 0.000 0.004 ± 0.000 0.800 ± 0.000
75% roughage GL 0.40 ± 0.002 0.20 ± 0.000 0.30 ± 0.000 DL 0.005 ± 0.000 0.002 ± 0.000 0.800 ± 0.000
DL 0.48 ± 0.001 0.23 ± 0.0005 0.40 ± 0.000 GS 0.050 ± 0.000 0.003 ± 0.000 2.000 ± 0.000
GS 0.55 ± 0.004 0.35 ± 0.001 1.00 ± 0.000 DS 0.001 ± 0.000 0.004 ± 0.000 1.000 ± 0.000
DS 0.77 ± 0.004 0.38 ± 0.0005 1.00 ± 0.000 100% roughage GL 0.002 ± 0.001 0.004 ± 0.000 0.770 ± 0.001
100% roughage GL 0.20 ± 0.010 0.15 ± 0.0005 0.35 ± 0.001 DL 0.002 ± 0.001 0.002 ± 0.000 0.800 ± 0.0005
DL 0.25 ± 0.002 0.25 ± 0.0005 0.75 ± 0.001 GS 0.005 ± 0.002 0.002 ± 0.000 2.000 ± 0.007
GS 0.24 ± 0.0008 0.30 ± 0.000 0.85 ± 0.008 DS 0.002 ± 0.0005 0.001 ± 0.000 1.000 ± 0.001
DS 0.63 ± 0.005 0.60 ± 0.0008 1.33 ± 0.005 GL = Green liquid, DL = Dry liquid, GS = Green solid, DS = Dry solid.
GL = Green liquid, DL = Dry liquid, GS = Green solid, DS = Dry solid.

buffalo rumen samples and the percentage of both were in observed that there was no statistically significant differ-
accordance with those predicted by the shotgun sequen- ence between the results from molecular approach using
cing/MG-RAST (Tables 4 and 5). These results showed real time PCR and results obtained by the MG-RAST
that the annotations by MG-RAST were showing much based taxonomic analysis of the shotgun metagenome
higher abundance of these species as compared to the sequence data, except for Prevotella.
values obtained from qPCR results. Molecular-based analysis using real time quantification
method have revealed that cultured bacterial species rep-
Discussion resent only a small fraction of the total bacterial diversity
In the present study, we report the enumeration of (Janssen 2006; Stevenson and Weimer 2007). qPCR is now
rumen bacterial species using qPCR and its comparison increasingly being used as a molecular method for enumer-
with the MG-RAST based microbial profiling of rumen ating bacterial load in complex microbiota like rumen
metagenome by shotgun and amplicon sequencing. We (Klieve et al. 2003; Tajima et al. 2001). The enumeration by

Figure 1 P. bryantii proportion as shown by qPCR and MG-RAST based analysis of shotgun and amplicon sequencing data. GL50,
75 and 100: Rumen Liquid fraction from 50%, 75% and 100% green roughage fed animals, respectively. GS50, 75 and 100: Rumen Solid
fraction from 50%, 75% and 100% green roughage fed animals, respectively. DL50, 75 and 100: Rumen Liquid fraction from 50%, 75% and
100% Dry roughage fed animals, respectively. DS50, 75 and 100: Rumen Solid fraction from 50%, 75% and 100% Dry roughage fed
animals, respectively.
Nathani et al. AMB Express 2013, 3:55 Page 6 of 8
http://www.amb-express.com/content/3/1/55

Table 5 Comparison of Methanomicrobiales percentage results also showed the dominance of these species, in
obtained by qPCR and MG-RAST analysis for same particular P. bryantii in the liquid fraction of the
samples processed by shotgun and amplicon sequencing rumen samples as observed by both, qPCR and shotgun
Methanomicrobiales (% ± SD) sequencing based taxonomic analysis. P. bryantii is a
Sample qPCR Shotgun gram-negative bacterium that utilizes the soluble polysac-
50% roughage GL 0.05 ± 0.0002 0.08 ± 0.000 charides, namely xylans justifying its higher abundance in
DL 0.38 ± 0.005 0.03 ± 0.000 the liquid fraction as compared to the solid fractions of
rumen samples. This type of activity, polysaccharide
GS 0.02 ± 0.008 0.13 ± 0.0005
utilization, plays an important role in most complex gut
DS 0.01 ± 0.0002 0.23 ± 0.003
niches.
75% roughage GL 0.20 ± 0.001 0.15 ± 0.0006 Cellulolytic bacteria Fibrobacter succinogens and
DL 0.60 ± 0.0006 0.10 ± 0.000 Ruminoccocus flavefaciens showed greater abundance
GS 0.01 ± 0.000 0.09 ± 0.000 than that of other species of bacteria like the Selenomonas
DS 0.002 ± 0.000 0.09 ± 0.000 ruminantium. High prevalence of cellulolytic bacteria of
the genus Fibrobacter and Ruminoccocus is representative
100% roughage GL 0.06 ± 0.0005 0.07 ± 0.0001
of its already reported importance as the major cellulolytic
DL 0.16 ± 0.001 0.15 ± 0.0006
species in the rumen ecosystem. These species are found
GS 0.002 ± 0.000 0.09 ± 0.0001 to play major role in improving the fiber metabolizing
DS 0.04 ± 0.0004 0.10 ± 0.000 ability of rumen (Shi and Weimer 1996). The population
GL = Green liquid, DL = Dry liquid, GS = Green solid, DS = Dry solid. of F. succinogens was higher as compared to R. flavefaciens
across the feeding treatments for majority of samples.
Such results were also observed earlier for the abundance
qPCR is calculated with standard curves for different of cellulolytic species in the rumen of buffalo (Wanapat
bacterial species. The abundance of bacterial species is
and Cherdthong 2009).
calculated as a fraction of the total ruminal bacterial
In annotation, similarity search based on the prediction
population. of 16S rRNA gene is found to be convenient due to highly
The evolution of high-throughput sequencing technolo-
conserved gene sequences but similar to results observed
gies and their subsequent improvements in generating
by us; earlier reports have also shown that the annotations
reads have enabled the utilization of different bioinformatic based on ribosomal genes are often found to be ambigu-
tools for studying communities of complex flora. Access
ous (Lin et al. 2008). Few reasons for such ambiguity in-
to large amounts of data gives a better account of sample
clude the impact of partial sequences in databases due
diversity and at a reasonable cost. Any metagenomics data to the difficulty in covering both ends of the gene with
generated thus requires proper storage and analysis, one
universal primers commonly used for determining 16S
of the currently used servers for the same includes the
rRNA gene sequences. Such sequences compete with
MG-RAST. Such freely available public resources for the complete rRNA sequences in sequence similarity
analysis of metagenome sequence data have surpassed
searches. The anti Shine Dalgarno sequence, one of the
the primary bottlenecks in metagenome sequence ana-
most conserved features in 16S rRNA genes is evidently
lysis like the availability of high-performance computing critical to translation efficiency (Lin et al. 2008; McCarthy
for annotating the data.
and Brimacombe 1994). Anti-SD sequence being located
at the 3′ end, it is likely to be excluded from 16S rDNAs
Enumeration of bacterial and archaeal species obtained using molecular approaches. In their study of
Rumen fluid samples were collected at 3 h post feeding 16S rRNA genes, (Lin et al. 2008) observed that at least
and the cell populations were calculated for the same, 67 out of 252 different species of the complete microbial
since a 3 h post feed would allow sufficient cell prolifera- genomes were annotated inaccurately and the main cause
tions and adherence of few species on feed particles and for such annotation was found to be the 16S rRNA partial
microbes. Total bacteria population was similar among sequences.
treatments and ranged in 5.0 × 106 to 6.0 × 107 copies μL-1 S. ruminantium, gram negative grows on substrates like
rumen content. Such results were also obtained by (Pilajun fructose, glucose, mannose, cellobiose, maltose, sucrose,
and Wanapat 2012; Vinh et al. 2011). and salicin, producing end products from glucose like
Prevotella species have been observed to be dominant lactate, acetate, propionate, and formate. On the basis
in rumen suggesting their importance in metabolism of of the sugar fermentation characteristics, Selenomonas
dietary feed (Bekele et al. 2010; Wood et al. 1998). Among ruminantium has been reported to be one of the major
them P. ruminocola and P. bryantii are representative species of Selenomonas genus. Similar results were also
species of sugar fermenting microbes in the rumen. Our predicted by MG-RAST analysis which showed > 90%
Nathani et al. AMB Express 2013, 3:55 Page 7 of 8
http://www.amb-express.com/content/3/1/55

abundance of Selenomonas genus in all samples at all Author details

1
concentrations. Department of Animal Biotechnology, College of Veterinary Science &
Animal Husbandry, Anand Agricultural University, Anand, Gujarat 388 001,
Ruminal methane is formed by the action of methano- India. 2Department of Microbiology, Christ College, Vidhya Niketan, P.B.
gens present in ruminants and are capable of using No.05, Rajkot-5, Gujarat, India. 3Department of Biosciences, Saurashtra
H2/CO2; can multiply easily and are observed in high University, Rajkot-5, Gujarat, India.

abundance in rumen (Zinder 1993). Contradictorily, in Received: 10 May 2013 Accepted: 10 August 2013
our results, methanogenic microbiales were found in Published: 11 September 2013
lesser proportion. The probable reason could be their
prevalence in a symbiotic interaction with protozoa.
References
Protozoa population will be low at 3 h post fed sampling Attfield P, Gunasekera T, Boyd A, Deere D, Veal D (1999) Applications of flow
time thereby subsequently reducing the proportion of cytometry to microbiology of food and beverage industries. Australas
Biotechnol 9:159–166
methanomicrobiales (Sahoo et al. 2000). Dilution of the
Bekele AZ, Koike S, Kobayashi Y (2010) Genetic diversity and diet specificity of
feed particles at 3 h post feed might have an impact on ruminal Prevotella revealed by 16S rRNA gene-based analysis. FEMS Microbiol
the growth of rumen methanogens as they are slow Lett 305(1):49–57
growers. Bryant MP (1959) Bacterial species of the rumen. Bacteriol Rev 23(3):125–153
Farrelly V, Rainey FA, Stackebrandt E (1995) Effect of genome size and rrn gene
The present study reveals reliability of database based copy number on PCR amplification of 16S rRNA genes from a mixture of
taxonomic profiling of rumen metagenome by validating bacterial species. Appl Environ Microbiol 61(7):2798–2801
through the already established molecular approach like Heid CA, Stevens J, Livak KJ, Williams PM (1996) Real time quantitative PCR.
Genome Res 6(10):986–994
the qPCR. As observed from the one way ANOVA analysis, Hugenholtz P, Goebel BM, Pace NR (1998) Impact of culture-independent studies
there is concordance between the qPCR and shotgun on the emerging phylogenetic view of bacterial diversity. J Bacteriol
sequencing whereas the qPCR and amplicon sequencing 180(18):4765–4774
Huijsdens XW, Linskens RK, Mak M, Meuwissen SG, Vandenbroucke-Grauls CM,
showed significant variation. The results are further Savelkoul PH (2002) Quantification of bacteria adherent to gastrointestinal
supported by the Spearman’s rank correlation test that mucosa by real-time PCR. J Clin Microbiol 40(12):4423–4427
shows significant positive correlation between the qPCR Janssen PH (2006) Identifying the dominant soil bacterial taxa in libraries of 16S
rRNA and 16S rRNA genes. Appl Environ Microbiol 72(3):1719–1728
and shotgun sequencing. Such techniques can also be Klausegger A, Hell M, Berger A, Zinober K, Baier S, Jones N, Sperl W, Kofler B
applied to check the abundance of bacterial species in (1999) Gram type-specific broad-range PCR amplification for rapid detection
other complex microbial communities. Results reflect the of 62 pathogenic bacteria. J Clin Microbiol 37(2):464–466
Klieve AV, Hennessy D, Ouwerkerk D, Forster RJ, Mackie RI, Attwood GT (2003)
ability of the microbes to adapt to the conditions and Establishing populations of Megasphaera elsdenii YE34 and Butyrivibrio
compete for the nutrients available; further signifying fibrisolvens YE44 in the rumen of cattle fed high grain diets. J Appl Microbiol
its importance to the overall functions of the complex 95:621–630
Koike S, Kobayashi Y (2001) Development and use of competitive PCR assays for
community. Knowledge about the proportion of ruminal the rumen cellulolytic bacteria: fibrobacter succinogenes, Ruminococcus
microbes inhabiting the rumen of a particular animal albus and Ruminococcus flavefaciens. FEMS Microbiol Lett 204(2):361–366
could also be used to manipulate the feeding regimes for Langendijk PS, Schut F, Jansen GJ, Raangs GC, Kamphuis GR, Wilkinson MH,
Welling GW (1995) Quantitative fluorescence in situ hybridization of
ruminants. Bifidobacterium spp. with genus-specific 16S rRNA-targeted probes and its
application in fecal samples. Appl Environ Microbiol 61(8):3069–3075
Leser TD, Amenuvor JZ, Jensen TK, Lindecrona RH, Boye M, Moller K
Additional files (2002) Culture-independent analysis of gut bacteria: the pig
gastrointestinal tract microbiota revisited. Appl Environ Microbiol
Additional file 1: Online resource 1. Overview of the pipeline used 68(2):673–690
for animal experiments and further evaluation of bacterial populations in Lin YH, Chang BC, Chiang PW, Tang SL (2008) Questionable 16S ribosomal RNA
the complex rumen metagenomic community. gene annotations are frequent in completed microbial genomes.
Additional file 2: Online resource 2. Total rumen bacterial population Gene 416(1–2):44–47
enumerated by real time PCR. Marchesi JR, Sato T, Weightman AJ, Martin TA, Fry JC, Hiom SJ, Dymock D, Wade
WG (1998) Design and evaluation of useful bacterium-specific PCR primers
Additional file 3: Online resource 3. Comparison of Fibrobacter
that amplify genes coding for bacterial 16S rRNA. Appl Environ Microbiol
succinogens percentage by qPCR and MG-RAST analysis based on
64(2):795–799
ribosomal RNA genes.
McCarthy JE, Brimacombe R (1994) Prokaryotic translation: the interactive
Additional file 4: Online resource 4. Comparison of R. flavefaciens pathway leading to initiation. Trends Genet 10(11):402–407
percentage by qPCR and MG-RAST analysis based on ribosomal RNA genes. McNeil LK, Reich C, Aziz RK, Bartels D, Cohoon M, Disz T, Edwards RA,
Gerdes S, Hwang K, Kubal M, Margaryan GR, Meyer F, Mihalo W,
Olsen GJ, Olson R, Osterman A, Paarmann D, Paczian T, Parrello B,
Competing interests Pusch GD, Rodionov DA, Shi X, Vassieva O, Vonstein V, Zagnitko O,
The authors declare that they have no competing interests. Xia F, Zinner J, Overbeek R, Stevens R (2007) The national microbial
pathogen database resource (NMPDR): a genomics platform based
Acknowledgements on subsystem annotation. Nucleic Acids Res 35(Database issue):
Authors are thankful to supporting staff at Livestock Research Station, S.K. D347–D353
Nagar, Gujarat, India for providing assistance in rumen sample collection. Muyzer G, de Waal EC, Uitterlinden AG (1993) Profiling of complex microbial
This work was supported by Niche area of excellence project “Metagenomic populations by denaturing gradient gel electrophoresis analysis of
analysis of ruminal microbes” funded by Indian Council of Agricultural polymerase chain reaction-amplified genes coding for 16S rRNA. Appl
Research, New Delhi, India. Environ Microbiol 59(3):695–700
Nathani et al. AMB Express 2013, 3:55 Page 8 of 8
http://www.amb-express.com/content/3/1/55

Overbeek R, Begley T, Butler RM, Choudhuri JV, Chuang HY, Cohoon M, de Crecy-
Lagard V, Diaz N, Disz T, Edwards R, Fonstein M, Frank ED, Gerdes S, Glass
EM, Goesmann A, Hanson A, Iwata-Reuyl D, Jensen R, Jamshidi N, Krause L,
Kubal M, Larsen N, Linke B, McHardy AC, Meyer F, Neuweger H, Olsen G,
Olson R, Osterman A, Portnoy V, Pusch GD, Rodionov DA, Ruckert C, Steiner
J, Stevens R, Thiele I, Vassieva O, Ye Y, Zagnitko O, Vonstein V (2005) The
subsystems approach to genome annotation and its use in the project to
annotate 1000 genomes. Nucleic Acids Res 33(17):5691–5702
Pilajun R, Wanapat M (2012) Microbial population in the rumen of swamp buffalo
(Bubalus bubalis) as influenced by coconut oil and mangosteen peel
supplementation. J Anim Physiol Anim Nutr (Berl) 97(3):439–445
Pryde SE, Richardson AJ, Stewart CS, Flint HJ (1999) Molecular analysis of the
microbial diversity present in the colonic wall, colonic lumen, and cecal
lumen of a pig. Appl Environ Microbiol 65(12):5372–5377
Sahoo B, Saraswat ML, Haque N, Khan MY (2000) Energy balance and methane
production in sheep fed chemically treated wheat straw. Small Ruminal
Res 35:13–19
Shi Y, Weimer PJ (1996) Utilization of individual cellodextrins by three predominant
ruminal cellulolytic bacteria. Appl Environ Microbiol 62(3):1084–1088
Stevenson DM, Weimer PJ (2007) Dominance of Prevotella and low abundance
of classical ruminal bacterial species in the bovine rumen revealed by relative
quantification real-time PCR. Appl Microbiol Biotechnol 75(1):165–174
Suzuki MT, Taylor LT, DeLong EF (2000) Quantitative analysis of small-subunit
rRNA genes in mixed microbial populations via 5′-nuclease assays.
Appl Environ Microbiol 66(11):4605–4614
Tajima K, Aminov RI, Nagamine T, Matsui H, Nakamura M, Benno Y (2001)
Diet-dependent shifts in the bacterial population of the rumen revealed with
real-time PCR. Appl Environ Microbiol 67(6):2766–2774
Veal DA, Deere D, Ferrari B, Piper J, Attfield PV (2000) Fluorescence staining and
flow cytometry for monitoring microbial cells. J Immunol Methods
243(1–2):191–210
Vinh NT, Wanapat M, Khejornsart P, Kongmun P (2011) Studies of diversity of
rumen microorganisms and fermentation in swamp buffalo fed different
diets. J Anim Vet Adv 10(4):406–414
von Wintzingerode F, Gobel UB, Stackebrandt E (1997) Determination of
microbial diversity in environmental samples: pitfalls of PCR-based rRNA
analysis. FEMS Microbiol Rev 21(3):213–229
Wanapat M, Cherdthong A (2009) Use of real-time PCR technique in studying
rumen cellulolytic bacteria population as affected by level of roughage in
swamp buffalo. Curr Microbiol 58(4):294–299
Wood J, Scott KP, Avgustin G, Newbold CJ, Flint HJ (1998) Estimation of the
relative abundance of different Bacteroides and Prevotella ribotypes in gut
samples by restriction enzyme profiling of PCR-amplified 16S rRNA gene
sequences. Appl Environ Microbiol 64(10):3683–3689
Zinder SH (1993) Physiological ecology of methanogens. In: Ferry JG (ed)
Methanogenesis: ecology, physiology, biochemistry and genetics. Springer,
US, pp 128–206

doi:10.1186/2191-0855-3-55
Cite this article as: Nathani et al.: Comparative evaluation of rumen
metagenome community using qPCR and MG-RAST. AMB Express
2013 3:55.

Submit your manuscript to a

journal and benefit from:
7 Convenient online submission
7 Rigorous peer review
7 Immediate publication on acceptance
7 Open access: articles freely available online
7 High visibility within the field
7 Retaining the copyright to your article

Submit your next manuscript at 7 springeropen.com