Anaerobe xxx (2017) 1e9

Contents lists available at ScienceDirect

Anaerobe
journal homepage: www.elsevier.com/locate/anaerobe

Effect of bioaugmentation by cellulolytic bacteria enriched from sheep

rumen on methane production from wheat straw
E. Gozde Ozbayram a, Sabine Kleinsteuber b, *, Marcell Nikolausz b, Bahar Ince c,
Orhan Ince a
a
Department of Environmental Engineering, Faculty of Civil Engineering, Istanbul Technical University, Maslak, Istanbul, Turkey
b
Department of Environmental Microbiology, Helmholtz Centre for Environmental Research - UFZ, Leipzig, Germany
c
aziçi University, Bebek, Istanbul, Turkey
Institute of Environmental Sciences, Bog

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study was to determine the potential of bioaugmentation with cellulolytic rumen
Received 14 November 2016 microbiota to enhance the anaerobic digestion of lignocellulosic feedstock. An anaerobic cellulolytic
Received in revised form culture was enriched from sheep rumen fluid using wheat straw as substrate under mesophilic condi-
14 March 2017
tions. To investigate the effects of bioaugmentation on methane production from straw, the enrichment
Accepted 15 March 2017
Available online xxx
culture was added to batch reactors in proportions of 2% (Set-1) and 4% (Set-2) of the microbial cell
number of the standard inoculum slurry. The methane production in the bioaugmented reactors was
higher than in the control reactors. After 30 days of batch incubation, the average methane yield was 154
Keywords:
Anaerobic digestion
mLN CH4 g
1 VS in the control reactors. Addition of 2% enrichment culture did not enhance methane pro-
Biogas duction, whereas in Set-2 the methane yield was increased by 27%. The bacterial communities were
Lignocellulosic feedstock examined by 454 amplicon sequencing of 16S rRNA genes, while terminal restriction fragment length
454 amplicon sequencing polymorphism (T-RFLP) fingerprinting of mcrA genes was applied to analyze the methanogenic com-
T-RFLP munities. The results highlighted that relative abundances of Ruminococcaceae and Lachnospiraceae
mcrA increased during the enrichment. However, Cloacamonaceae, which were abundant in the standard
inoculum, dominated the bacterial communities of all batch reactors. T-RFLP profiles revealed that
Methanobacteriales were predominant in the rumen fluid, whereas the enrichment culture was domi-
nated by Methanosarcinales. In the batch rectors, the most abundant methanogens were affiliated to
Methanobacteriales and Methanomicrobiales. Our results suggest that bioaugmentation with sheep rumen
enrichment cultures can enhance the performance of digesters treating lignocellulosic feedstock.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
Anaerobic digestion (AD) is a proven sustainable technology to
recover energy from many biodegradable wastes such as livestock
manure, agricultural residues, food and brewery wastes, sewage
sludge, industrial and municipal wastewaters, wood industry res-
idues etc. [1,2]. Among these waste streams, agricultural residues
belong to the most attractive substrates for AD due to the high
abundance, high energy content and low costs [3]. As payback
periods are major issues for biogas production plants, efficient
* Corresponding author. Helmholtz Centre for Environmental Research e UFZ,
Permoserstr. 15, 04318 Leipzig, Germany.
E-mail addresses: gozbayram@itu.edu.tr (E.G. Ozbayram), sabine.kleinsteuber@
ufz.de (S. Kleinsteuber), marcell.nikolausz@ufz.de (M. Nikolausz), bahar.ince@
boun.edu.tr (B. Ince), inceor@itu.edu.tr (O. Ince).
methane production is of great concern [2]. Besides the many ad-
vantages of AD, the main challenge of this process is the hydrolysis
of lignocellulose-rich feedstock, which is generally considered as
the rate-limiting step. Thus, pretreatment technologies and/or
other enhancement strategies are frequently integrated with the
AD process.
The mutualistic interactions between different functional
groups including hydrolytic, acidogenic and acetogenic bacteria
and methanogenic archaea are crucial for efficient digestion [4].
Bioaugmentation is a promising option to introduce specific mi-
croorganisms into the autochthonous digester community to
improve certain stages of the AD process. Recently, researchers
have shown an increased interest in the bioaugmentation strategy
to enhance the hydrolysis rate and consequently methane pro-
duction from many plant residues, such as wheat straw [5,6], sweet
corn processing waste [7], corn straw [3] and cattail [8].

http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
1075-9964/© 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
2 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9
Lignocellulosic compounds are degraded in the gut system of
animals by specific microorganisms excreting hydrolytic enzymes
such as cellulases, hemicellulases and lignin-degrading enzymes
[9]. Ruminants such as cattle, goat and sheep have the ability to
feed on plant fibers with the help of their intestinal symbionts [10].
The unique rumen microbiota can easily adapt to a different diet
and gain energy to sustain necessary metabolic functions [11].
Previous studies suggested that microorganisms in rumen fluid are
more efficient in lignocellulosic feedstock degradation, thanks to
their higher cellulolytic activities, than inocula from anaerobic di-
gesters [12,13]. As the production of volatile fatty acids (VFA) is a
key parameter to determine the efficiency of hydrolysis, reactors
inoculated with rumen fluid can reach higher VFA concentrations
during digestion [2]. Despite the fact that commercial enzyme
addition has a positive effect on the hydrolysis rate, using ligno-
cellulose degrading microorganisms is a sustainable and more
efficient strategy with regard to their ability to produce the
necessary enzymes depending on the given feedstock [14]. How-
ever, what is not yet clear in the literature is the fate of the rumen
bacteria in anaerobic digesters and the mechanisms how they
contribute to more efficient hydrolysis.
In the present study, we aimed to enrich cellulolytic microor-
ganisms from sheep rumen fluid and to determine their bio-
augmentation potential in biogas production from wheat straw. The
fate of the rumen microbiota and their potential role in the bio-
augmented digesters were investigated by microbial community
analysis.
2. Material and methods
2.1. Sampling
Rumen fluid was taken from a healthy non-medicated East-
Friesian milk sheep (10 years old, 85 kg). The sheep was cared for
and handled in a barn of the Veterinary Faculty of Leipzig Univer-
sity, Germany. Its diet consisted of basically pelleted grass, root
chips and hay. The rumen sample was collected via a rumen fistula
in accordance with the institutional animal care guidelines, and the
sample was immediately transferred to the laboratory for the
enrichment procedure. Aliquots of the sample were stored
at
20  C for DNA extraction.
2.2. Enrichment culture
The enrichment and cultivation procedure was performed ac-
cording to the protocol described by Porsch et al. [15] with some
modifications. Briefly, we used the modified DSMZ medium 1036
containing 0.5 g NH4Cl, 0.2 g KH2PO4, 0.1 g MgCl2  6H2O, 0.2 g KCl,
2.0 g NaCl, 0.2 g yeast extract, 1.0 mL trace element solution SL10
(DSMZ medium 320) and 0.5 mg resazurin dissolved in 850 mL
high-purity water. For making the medium anoxic, it was stirred for
30 min in an anaerobic chamber (98% N2, 2% H2) and then auto-
claved (20 min, 121  C). Wheat straw was used as complex carbon
and energy source. For 50 mL cultures, 0.5 g straw and 1 mL high-
purity water were added to 100 mL serum bottles, then autoclaved
(30 min, 121  C) and stored overnight in an anaerobic chamber for
changing the bottle's conditions to anoxic. In the anaerobic cham-
ber, high-purity water was added once again, and the bottles were
closed with butyl rubber stoppers and autoclaved for a second time
(30 min, 121  C). Then, sterile anoxic medium was filled into the
culture bottles. All bottles were flushed with a mixture of 75% N2
and 25% CO2 to adjust the pH to 7.5.
Before the first inoculation, pre-cultures were prepared from
freshly sampled rumen fluid under anaerobic conditions. For the
pre-cultures, rumen fluid was diluted 1:4 with water and 3 mL of
Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
the diluted sample was added to bottles and incubated at 37  C.
Two days later, fresh medium bottles were inoculated with 1 mL of
the pre-culture and incubated at 37  C. Sterile controls were pre-
pared in the same way as the bottles for the cultures except that
they were not inoculated. Biogas production, gas composition, VFA
concentration and pH were monitored as described by Porsch et al.
(2015). After the third transfer, cultures were up-scaled in 450 mL
batches with 4.5 g wheat straw to get enough inoculum for bio-
methane potential (BMP) analysis.
2.3. BMP assay
The potential effects of bioaugmentation with cellulolytic mixed
cultures to enhance AD of wheat straw was assessed in batch tests
at 37  C using the Automatic Methane Potential Test System
(AMPTS) II (Bioprocess Control). The amount of substrate and
inoculum applied in the experiment was calculated according to
the manufacturer's instructions and the inoculum/substrate ratio
was 2 on the basis of volatile solids (VS).
After the third transfer, the enrichment culture was scaled up
and used as a supplementary inoculum (bioaugmentation) in the
BMP experiment. The amount of enrichment culture added to the
reactors corresponded to approximately 2% and 4% of the microbial
cell number of the standard inoculum, which was taken from a
pilot-scale biogas plant treating cow manure and maize silage
under mesophilic conditions. The cell numbers were determined
according to Str€auber et al. [6]. Non-bioaugmented controls (straw
inoculated with standard inoculum) and blank reactors including
only standard inoculum without straw were also set up. Each set-
up tested in AMPTS II was run in triplicates, thus the experiment
comprised 12 batch reactors.
In the anaerobic chamber, 50 mL of anoxic sterile growth me-
dium was added to the controls and 50 mL of one of the concen-
trated microbial cultures to the experimental set-ups. The bottles
with wheat straw were closed and each reactor was connected with
the CO2 trap and counting units of the AMPTS. The headspace of
each reactor and its peripheral installation were flushed with
300 mL N2. The produced methane volume was measured online
and automatically corrected for the N2 present in the system and
normalized to standard conditions (p ¼ 101.325 kPa, T ¼ 273.15 K,
no humidity). The reactors were operated with enrichment cultures
and wheat straw for 1 day, and then standard inoculum was added
to the reactors and further operated for 29 days.
2.4. Microbial community analysis
Samples from sheep rumen fluid, the enrichment culture, the
standard inoculum and samples taken at the end of the BMP assays
were analyzed for their microbial community composition. DNA
was extracted from frozen cell pellets using the NucleoSpin® Soil
Kit (Macherey-Nagel) applying buffer SL1 and enhancer solution SX
according to the manufacturer's instructions. After checking the
quality by agarose electrophoresis and photometric quantification,
DNA samples were stored at
20  C until further analysis.
2.5. 454 pyrosequencing
Bacterial community composition was determined by 454
pyrosequencing of 16S rRNA amplicons as described by Ziganshin
et al. [16]. Polymerase chain reaction (PCR) amplification with 25
cycles was performed using the Phire Hot Start II DNA Polymerase
(Thermo Scientific) with the primers Bac27F (50 -GAG TTT GAT CMT
GGC TCA G-30 ) and Bac519R (50 -GWA TTA CCG CGG CKG CTG-30 )
specific for bacterial 16S rRNA gene fragments. A nested PCR with
10 cycles was performed using 454 fusion primers tagged with
 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9
multiplex identifier sequences. MinElute Gel Extraction Kit (Qia-
gen) was used for purification of the PCR products after cutting the
respective band from the agarose gel. The purified fragments were
checked on a DNA1000 chip with the Agilent 2100 Bioanalyzer. The
amplicons were quantified with PicoGreen according to the GS
Junior Amplicon Library Preparation Method Manual (Roche). Then,
the amplicons were pooled and applied for emulsion PCR using the
Lib-L emPCR Kit (Roche). 454 pyrosequencing run was performed
on a picotiter plate using the GS Junior platform according to the
manufacturer's instructions (Roche).
The QIIME 1.9.0 Virtual Box release was used to process raw
reads following the procedure described by Kuczynski et al. [17],
Lucas et al. [18] and QIIME documentation (http://qiime.org/).
Reads with lengths between 150 and 590 bp were kept and the
quality threshold was set to 25. Read end sections of 50 bp were
trimmed if they were below this quality threshold. The USEARCH
pipeline was used in quality filtering and the clustering steps [19]. A
97% identity threshold was used for the determination of opera-
tional taxonomic units (OTU). Non-chimeric OTUs represented by
at least 5 reads were kept. The taxonomic classification was done
using the latest Greengenes reference OTU builds (release ‘gg 13 8’)
[20] and based on alignment with the Infernal algorithm [21].
Alpha diversity indices (Chao1, Simpson and Pielou) were calcu-
lated using the QIIME pipeline; Shannon entropy was calculated as
described by Lucas et al. [22].
The Krona graphs were prepared according to Ondov et al. [23]
to present the bacterial community composition of the samples on
phylum, class, order and family levels. Venn diagrams were pre-
pared according to Oliveros et al. [24] to show the shared and
unique OTUs between samples.
The sequence raw reads determined in this study were depos-
ited under the EMBL-EBI accession numbers PRJEB15307 and
PRJEB15600.
2.6. T-RFLP analysis
Bacterial and methanogenic communities were profiled by ter-
minal restriction fragment length polymorphism (T-RFLP) finger-
printing of bacterial 16S rRNA and mcrA amplicons according to the
protocol described previously by Stra€uber et al. [6]. Bacterial 16S
rRNA amplicons were analyzed based on digestion with the re-
striction enzymes MspI and RsaI (New England Biolabs) using the
standard MapMarker 1000 (Eurogentec). The restriction enzyme
BstNI (New England Biolabs) was applied for mcrA analysis and
GeneScan-500 ROX (Applied Biosystems) was used as a standard. T-
RFLP peaks in the range of 50e1000 bp (16S rRNA amplicons) and
50e500 bp (mcrA amplicons) were kept in the further analysis,
respectively. Taxonomic assignment of mcrA terminal restriction
fragments (T-RF) representing methanogenic community members
was done according to the T-RFLP database described by Bühligen
et al. [25].
3. Results & discussion
3.1. Enrichment of a cellulolytic methanogenic culture from sheep
rumen fluid
A cellulolytic enrichment culture was established from sheep
rumen fluid and transferred twice before analyzing its bio-
augmentation potential in anaerobic digesters treating wheat
straw. Biogas production, composition of the biogas, pH and VFA
production in each transfer are shown in Fig. 1. Biogas was pro-
duced mainly within the first two weeks of incubation and pro-
duction stopped at the end of the incubation period (30 days) in all
transfers. The cumulative biogas production for the first
Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
3
enrichment, second and third transfers was 1.7, 1.1 and 2.3 mLgas/
mLmedium, respectively (Fig. 1A). There was minor or no gas pro-
duction in the control bottles (non-inoculated), indicating that the
two-step autoclaving approach as previously carried out by Porsch
et al. [15] was sufficient to inactivate most of the microorganisms
colonizing the straw.
At the beginning of the cultivation, the gas composition of all
bottles was identical due to the same gas mixture used for flushing.
The methane content in the culture bottles increased over time. It
was 9% at the end of the first transfer and increased to 24% in the
second transfer (Fig. 1B). The highest methane content was recor-
ded at the end of the third transfer (31%). Thus, the enrichment
procedure enriched not only bacteria but also methanogens, and
their metabolic contribution increased during the transfers. The
CO2 concentration in the bottles also increased during the incu-
bation period. The highest value was recorded as 56% on day 30 in
the first enrichment, and the concentration was similar in the
second and third transfers (44 ± 2%).
The initial pH was 7.3 ± 0.2 in all culture bottles. During the
incubation, the pH decreased in all cultures due to VFA production.
However, the pH slightly increased in the second and third trans-
fers after day 14. This observation could be attributed to the con-
sumption of VFA in the cultures and correlated with the higher
methane production in the third transfer. At the last day of incu-
bation, the pH was measured as 6.2 to 6.6 in the inoculated bottles
and it stayed around 7.2 ± 0.1 in the controls (Fig. 1C).
The main VFAs detected were acetic, propionic and n-butyric
acid. The highest acetic acid concentration was recorded in the first
enrichment with up to 1.6 g/L (Fig. 1D), indicating that acidogenic
fermentation dominated in the first enrichment step and the ac-
tivity of methanogens was not yet sufficient to convert the acetate.
The highest acetic acid concentration in the second transfer was
measured at the end of the first week of incubation (725 mg/L). In
the third transfer, the acetic acid concentration was significantly
lower than in the first enrichment and second transfer. Accordingly,
biogas production was highest in the third transfer, indicating that
the conditions were favorable for methanogens and the activity of
acetoclastic methanogens was sufficient to convert the acetate
completely. As shown in Fig. 1E, no significant difference in propi-
onic acid concentrations was observed between the three batch
cultivations. At the end of each incubation propionic acid accu-
mulated in concentrations of 1e1.1 g/L. In addition, there was a
clear trend of increasing butyric acid concentrations (Fig. 1F). The
concentration was recorded as 337 ± 3 mg/L at the end of the in-
cubation period. Further, iso-valeric and iso-butyric acids were
determined in the enrichment cultures and their concentrations
gradually decreased over the transfers. The highest iso-butyric acid
concentration was measured in the first enrichment (34 mg/L) at
the end of the incubation. In the second and third transfers, it was
recorded as 28 and 21 mg/L, respectively. The same trend was
observed for iso-valeric acid, the concentration of which was
measured as 49-21 mg/L. The control bottles produced no VFA or
only low amounts of acetic and propionic acid (105 ± 15 mg/L on
day 30), maybe due to the activity of some spore-forming bacteria
that had survived the autoclaving procedure.
3.2. Bioaugmentation potential of the enrichment culture in AD of
wheat straw
The methane production enhancement potential of the enrich-
ment cultures was assessed in a batch experiment. The methane
yields calculated on the basis of VS content of wheat straw for the
bioaugmented batch cultures and control batches are shown in
Fig. 2. At the end of the 30 days operation period, the average
methane yield was 154 mLN CH4 g
1 VS in the control reactors
4 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9

2.5 A B 8 C
C
100
7.8

Relative gas composition [%]

(mLvolume/mLmedium)
Cumulative gas volume

2 80 7.6
7.4
60
1.5 7.2

pH
40 7
1 6.8
20
6.6
0.5 0 6.4
6.2
0 6
0 10 20 30 0 10 20 30
Time (d) CH4 N2 O2 CO2 H2 Time (d)

1600 D E 500 F
1200
1400
1000 400
Propionic acid (mg/L)

Butyric acid (mg/L)

1200
Acetic acid (mg/L)

1000 800 300

800 600
200
600
400
400
100
200
200
0 0 0
0 10 20 30 0 10 20 30 0 10 20 30
Time (d) Time (d) Time (d)

500
0
-20 30
1st Enrichment 2nd Transfer 3rd Transfer 1st Control 2nd Control 3rd Control

Fig. 1. Physiological data of the first enrichment cultures and two transfers (average values of duplicate cultures). A. Normalized semiquantitative cumulative gas production in
relation to the medium volume. B. Relative gas composition in the headspace after 30 days of incubation. C. pH values of the cultures. D. Acetic acid concentration in the liquid
phase. E. Propionic acid concentration in the liquid phase. F. Butyric acid concentration in the liquid phase.

220 2% Sheep Culture

4% Sheep Culture
200
Control
180
Methane yield (mLN/gVS)

160
140
120
100
80
60
40
20
0
0 5 10 15 20 25 30
Time (d)

Fig. 2. Methane yields from bioaugmented and control reactors. The data are the average of triplicate reactors. Error bars indicate the standard deviation.

inoculated with standard inoculum only. The methane production enrichment culture did not enhance the methane yield consider-
in the bioaugmented reactors (set up with an inoculum containing ably (166 mLN CH4 g
1VS ). During the first eight days, there was no
2% or 4% of the rumen enrichment culture) was generally higher significant difference between the methane yields of the batches
than in the control reactors. However, amendment with 2% amended with 2% and 4% enrichment culture. After day 8, the

Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9
difference in methane yield between these two set-ups started to
increase. Eventually, the methane yield of the batch cultures
amended with 4% rumen enrichment culture reached on average
196 mLN CH4 g
1 VS , representing an enhancement by 27%. Although
these methane yields are rather low, indicating that wheat straw is
not a suitable mono-substrate for biogas production, our study
shows the potential of the bioaugmentation approach to improve
methane yields from lignocellulose-rich feedstocks in general.
Further research should investigate how this applies to other fiber-
rich substrates that are commonly exploited in anaerobic digestion.
Differently from our study, Zhang et al. [3] used a pure culture of
Acetobacteroides hydrogenigenes in anaerobic digestion of corn
straw. They examined the bioaugmentation potential of 10% addi-
tion of this strain and found an increase of the methane yield by
19e23%. Acetobacteroides hydrogenigenes can hydrolyze pectin and
starch [26] but not cellulose. Thus, an increased VFA production
from cellulose can not be expected in cultures bioaugmented with
this species. In another recent study, a pure culture of Clostridium
cellulolyticum was used for the bioaugmentation of anaerobic di-
gesters treating wheat straw [5]. The amount of culture added to
the system was 25% of the inoculum (v/m). This bioaugmentation
study resulted in a methane yield increased by 7.6e13%. Taken
together, these results suggest that selected hydrolytic pure cul-
tures can have a clear positive effect on methane production.
However, the enhancement achieved in our experiment was higher
than in other studies and required a smaller amount of bio-
augmentation culture. The observed increase in methane yield
most probably was attributed to the synergistic relationships of the
community members derived from sheep rumen in our bio-
augmentation culture. The bioaugmentation effect observed in our
study was much more pronounced than that observed by Stra €uber
et al. [6] who used 1% of an alkali-tolerant cellulolytic enrichment
culture as a supplementary inoculum during AD of alkaline pre-
treated wheat straw. In that study, the authors observed a faster
methane production in the beginning of the BMP tests but no sig-
nificant enhancement of the cumulative methane yield. Possibly,
the allochthonous microbes of the alkali-tolerant enrichment cul-
ture could not establish within the complex AD community. We
assume that a threshold level is needed for the effective estab-
lishment of the bioaugmented microorganisms into a well-
functioning AD system. Moreover, we added the standard inoc-
ulum one day later than the enrichment culture. Maybe this gave an
advantage for the cells of the enrichment culture to attach to the
substrate surface without competition with microbes from the
standard inoculum.
3.3. Microbial community composition and dynamics
The bacterial communities of all samples were first pre-
screened by T-RFLP fingerprinting of 16S rRNA genes. The com-
munity patterns of the replicate batch cultures at the end of the
BMP tests were quite similar (Suppl. Fig. 1), therefore a more
Table 1
Summary of the estimated richness and evenness of the bacterial communities.
Sample No. of reads No. of OTUs
Sheep rumen 7800 337 (279)
Enrichment culture 14060 232 (177)
Standard inoculum 8138 255 (193)
4% Reactor 2 3551 214 (178)
4% Reactor 3 4189 218 (177)
Control 1 4196 211 (165)
Control 3 1944 173 (172)
The values in parentheses refer to randomly subsampled libraries generated from the lowest number of reads per sample.
Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
5
detailed analysis by amplicon sequencing was performed only with
few selected samples. Thus, the standard inoculum and the reactor
samples taken on the last day of operation of the 4% enrichment
culture added set-ups and non-bioaugmented control set-ups were
analyzed for better understanding of bacterial community dy-
namics and the fate of the inoculated microorganisms. The bacterial
community composition of the original rumen sample and the
enrichment culture derived from it were also investigated to reveal
the enrichment effect.
The summary of the estimated bacterial diversity is shown in
Table 1. Whereas the highest number of OTUs and highest values
for the Chao1, Shannon and Simpson indices were determined in
the sheep rumen sample, the bacterial diversity decreased during
the enrichment procedure. The bacterial diversity also decreased
under the reactor conditions compared to the standard inoculum
and the enrichment culture. High values of the Pielou's evenness
index indicate a more evenly distributed community. The highest
evenness was calculated for the sheep rumen sample whereas the
reactor samples were characterized by an uneven species
distribution.
Fig. 3 presents the bacterial community composition of the
samples at the phylum, class, order and family levels. It is apparent
from the Krona charts that the most abundant phyla in all samples
were Firmicutes and Bacteroidetes except in the AMPTS reactors,
which were dominated by the candidate phylum WWE1. The sheep
rumen sample comprised 15 phyla with Bacteroidetes being the
most abundant phylum (73.6% of all reads). Firmicutes (15.3%) was
the second most dominant phylum and only 1% of the reads could
not be assigned to any bacterial phylum (unclassified bacteria). The
rumen bacterial community profile obtained in this study is in line
with previous studies showing that rumen communities were
dominated by those phyla that are known to comprise species
involved in cellulose decomposition [29e31]. The enrichment
culture comprised 14 phyla and the proportions of representative
phyla were shifted as a result of the enrichment procedure. The
bacterial community was dominated by Firmicutes (46%) with the
most abundant families Lachnospiraceae (17%) and Ruminococca-
ceae (14%). In the rumen sample, the abundance of the family Vic-
tivallaceae (phylum Lentisphaerae) was lower than 1%, whereas 15%
of the reads from the enrichment culture were assigned to this
phylum. Members of this family were reported to play a role in the
fermentation of sugars such as cellobiose and glucose and to pro-
duce acetate and hydrogen in the presence of methanogens [30].
Moreover, 3.5% of the total reads could not be assigned to any
phylum. At the family level, the major proportion of sequences
(43%) was assigned to Prevotellaceae in the sheep rumen sample
while this family was found to be far less abundant in the enrich-
ment culture (<1%). Although most of the Prevotellaceae species are
linked to the breakdown of proteins and carbohydrates and some
species isolated from rumen environments show cellulolytic ac-
tivities like carboxymethyl cellulase and xylanase activities [31] we
could not enrich these bacteria. However, the abundance of
Chao1 Shannon Simpson Pielou's evenness
372 (319) 4.99 0.98 (0.98) 0.86
264 (219) 3.87 0.94 (0.95) 0.71
287 (250) 4.04 0.96 (0.95) 0.73
254 (251) 2.80 0.72 (0.73) 0.52
250 (213) 2.82 0.73 (0.74) 0.52
241 (232) 2.58 0.68 (0.68) 0.48
236 (233) 2.90 0.75 (0.75) 0.56
 6
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on

E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9

Fig. 3. Bacterial community composition according to 454 amplicon sequencing on phylum, class, order and family levels. A. Composition of the sheep rumen community, the enrichment culture used for bioaugmentation and the
standard inoculum. B. Community composition in the bioaugmented and control reactor samples after 30 days of incubation.
 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9 7

Ruminococcaceae anLachnospiraceae increased during the enrich-

ment procedure. The genus Ruminococcus was reported to be
involved in cellulose decomposition [32]. Some members of the
Lachnospiraceae have the ability to degrade xylan and some have
cellulolytic activity [33]. It is also known that members of the
Lachnospiraceae play a key role in butyric acid production [34].
Their high abundance in the enrichment culture is in accordance
with the butyric acid production observed during the enrichment
procedure (Fig. 1). Moreover, some members have the capability to
degrade plant polymers in gut environments [35]. Besides their
function in plant polymer degradation, members of the Lachno-
spiraceae and Ruminococcaceae have been described to play a role in
reductive acetogenesis, which is an alternative hydrogen sink in
rumen systems [36]. It is conceivable that acetogens were enriched
and that both competing hydrogen-consuming pathways (reduc-
tive acetogenesis and hydrogenotrophic methanogenesis) coex-
isted in the enrichment culture, whereas acetogens were
outcompeted in the AMPTS batch cultures.
The bacterial community of the enrichment culture was ex-

Fig. 4. Venn diagram of the bacterial communities showing the number of shared and unique OTUs.
pected to include only the bacteria present in the sheep rumen.
However, only 109 OTUs were shared between these two samples,
representing 47% of the total number of OTUs in the enrichment
culture community (Suppl. Fig. 2). Thus, our sequencing effort was
not sufficient to detect rare members of the rumen community that
became dominant during the enrichment process.
The bacterial community of the standard inoculum, which was
taken from a pilot-scale biogas plant treating cow manure and
maize silage under mesophilic conditions, showed a similar bac-
terial community profile as typical AD. It contained 21 phyla, in
which Bacteroidetes and Firmicutes were equally represented in the
community (accounting for 34 ± 0.5% of all reads). Differently from
the other two samples, 12.6% of the reads were assigned to the
candidate phylum WWE1 (“Cloacimonetes”). The bacterial com-
munity profiles of the batch reactor samples were quite similar.
What was different compared to the other samples is that WWE1
was the most abundant phylum in the reactor samples (64 ± 3.3%).
Similarly, in a study carried out by Lucas et al. [18], the candidate
phylum WWE1was determined as a core phylum accounting for
22e32% of the total reads in reactor samples taken from a full scale
AD treating maize silage and other energy crops but no manure.
This result reported by Lucas and colleagues is close to our study, in
which the feedstock was similar (i.e., fiber-rich plant material). The
WWE1 phylum was frequently reported in anaerobic environments
such as AD [18,37,38], and discussed to be linked to cellulose hy-
drolysis and/or the fermentation of substrates resulting from cel-
lulose hydrolysis [39]. Another study carried out by Pelletier et al.
[40] showed that a member of WWE1 was involved in amino acid
fermentation.
The shared and unique OTUs between the inocula (standard
inoculum and rumen enrichment culture) as well as the AMPTS
batch cultures are presented in Fig. 4. As shown in the Venn dia-
grams, six core OTUs (Bacteroidaceae, Marinilabiaceae, MBA08,
Clostridiales, Caldicoprobacter and Candidatus Cloacamonas) were
determined across all samples. Whereas the enrichment culture
contained the highest numbers of unique OTUs, the bioaugmented
reactors showed the lowest OTU numbers. After 30 days of AMPTS
operation, the bacterial communities of the reactors and the stan-
dard inoculum shared high numbers of OTUs. Some of the OTUs in
the bioaugmented reactors such as the families Ruminococcaceae
and Christensenellaceae, the genus Pyramidobacter and unclassified
members of the Firmicutes originated only from the enrichment
culture. It is known that members of the Ruminococcaceae are
involved in cellulose degradation [41]. Christensenellaceae is a
relatively new family proposed by Morotomi et al. [42] to the order
Clostridiales and comprises species fermenting sugars and

Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
8 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9

Fig. 5. Methanogenic community composition of all samples according to T-RFLP profiles of mcrA amplicons digested with BstNI. Bioaugmented and control reactors were sampled
after 30 days of incubation.

producing VFA such as acetate and butyrate. The genus Pyr-
amidobacter, which belongs to the Synergistetes phylum, is
described as saccharolytic bacteria producing acetate [43]. The
enhancement of methane production appears to be linked to these
OTUs originating from the bioaugmentation culture. Furthermore, a
high number of OTUs were shared between all reactor samples and
thus originated from the standard inoculum.
The community composition of methanogens was analyzed by
T-RFLP analysis of mcrA genes. As shown in Fig. 5, the sheep rumen
sample had the least diverse methanogenic community and
harbored only hydrogenotrophic methanogens, dominated by a
sequence type affiliated to the Methanobacteriales (T-RF 467). This
sequence type was only of minor abundance in all other samples
and is probably an ecotype specifically adapted to the rumen
environment. In contrast, the most abundant T-RF in the enrich-
ment culture and the standard inoculum was affiliated to the genus
Methanosarcina indicating that acetoclastic methanogenesis was
favored in these communities. The absence of acetoclastic metha-
nogens in the rumen sample is not surprising as VFA are assimilated
by the host animal in the rumen system. Nevertheless, Meth-
anosarcina emerged and became abundant during the enrichment
procedure, showing that it was only below the detection limit of the
T-RFLP method in the rumen sample.
The methanogenic communities of the AMPTS reactors were
quite similar to each other and more diverse than the rumen and
inocula samples. In contrast to the enrichment culture and standard
inoculum, the final reactor communities at the end of the BMP tests
were dominated by hydrogenotrophic methanogens. These
comprised the genera Methanoculleus (T-RF 94/95) and Meth-
anobacterium (T-RF 123) as well as a T-RF type assigned to the
Methanobacteriales or Methanomassiliicoccales (T-RF 470) e the T-
RF cannot be clearly assigned with the applied restriction enzyme
[25]. Notably, this T-RF was detected in all reactors and the sheep
rumen sample, but neither in the standard inoculum nor in the
enrichment culture. It seems possible that the standard inoculum
and/or the enrichment culture still contained members of this
group (T-RF 470) below the detection limit of the T-RFLP method.
Obviously, members of this group had a selective advantage and
became dominant during the batch conditions.
The relative abundance of a member of the Methanobacteriaceae
with T-RF 464 in the control reactors was similar as in the standard
inoculum. However, it was quite dominant in the bioaugmented
reactors, indicating that this sequence type originated from the
enrichment culture and successfully established in the batch
cultures.
4. Conclusion
This study evaluated the application of cellulolytic microbiota
enriched from sheep rumen as a bioaugmentation culture and its
effect on methane production from wheat straw. The enrichment
procedure resulted in an increase in the abundance of Lachnospir-
aceae and Ruminococcaceae compared to the original rumen com-
munity. BMP assays showed that the addition of 4% enrichment
culture promoted the methane production significantly and thus
enhanced the AD system performance. Our results suggest that
bioaugmentation with sheep rumen enrichment cultures can be an
alternative strategy to enhance digester performance treating
lignocellulosic feedstocks. However, the amount of introduced
microorganisms should be above a certain threshold in order to
effectively compete with the autochthonous microbial community.
Acknowledgements
Emine Gozde Ozbayram was supported by the Research
Fellowship Programme of the Scientific and Technological Research
Council of Turkey (grant no. 2214A). The authors would like to
acknowledge Birke Brumme (UFZ) for technical assistance in
anaerobic cultivation and Ute Lohse (UFZ) for pyrosequencing. The
authors also thank Rico Lucas (UFZ) and Canan Karakoc (UFZ) for
their kind help in sequence data analysis.

Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
 E.G. Ozbayram et al. / Anaerobe xxx (2017) 1e9
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.anaerobe.2017.03.013.
References

[1] M. Cater, L. Fanedl, spela Malovrh, R. Marinsek Logar, Biogas production from
brewery spent grain enhanced by bioaugmentation with hydrolytic anaerobic
bacteria, Bioresour. Technol. 186 (2015) 261e269.
[2] C. Sawatdeenarunat, K.C. Surendra, D. Takara, H. Oechsner, S.K. Khanal,
Anaerobic digestion of lignocellulosic biomass: challenges and opportunities,
Bioresour. Technol. 178 (2015) 178e186.
[3] J. Zhang, R.B. Guo, Y.L. Qiu, J.T. Qiao, X.Z. Yuan, X.S. Shi, C.S. Wang, Bio-
augmentation with an acetate-type fermentation bacterium Acetobacteroides
hydrogenigenes improves methane production from corn straw, Bioresour.
Technol. 179 (2015) 306e313.
[4] J.J. Werner, D. Knights, M.L. Garcia, N.B. Scalfone, S. Smith, K. Yarasheski,
T.A. Cummings, A.R. Beers, R. Knight, L.T. Angenent, Bacterial community
structures are unique and resilient in full-scale bioenergy systems, Proc. Natl.
Acad. Sci. U. S. A. 108 (10) (2011) 4158e4163.
[5] X. Peng, R. Arag~ ao, I. Achu, J. Liu, Impact of bioaugmentation on biochemical
methane potential for wheat straw with addition of Clostridium cellulolyti-
cum, Bioresour. Technol. 152 (2014) 567e571.
[6] H. Stra €uber, F. Bühligen, S. Kleinsteuber, M. Nikolausz, K. Porsch, Improved
anaerobic fermentation of wheat straw by alkaline pre-treatment and addi-
tion of alkali-tolerant microorganisms, Bioengineering 2 (2) (2015) 66e93.
[7] A. Martin-Ryals, L. Schideman, P. Li, H. Wilkinson, R. Wagner, Improving
anaerobic digestion of a cellulosic waste via routine bioaugmentation with
cellulolytic microorganisms, Bioresour. Technol. 189 (2015) 62e70.
[8] V.N. Nkemka, B. Gilroyed, J. Yanke, R. Gruninger, D. Vedres, T. McAllister,
X. Hao, Bioaugmentation with an anaerobic fungus in a two-stage process for
biohydrogen and biogas production using corn silage and cattail, Bioresour.
Technol. 185 (2015) 79e88.
[9] A. Lazuka, L. Auer, S. Bozonnet, D.P. Morgavi, M. O'Donohue, G. Hernandez-
Raquet, Efficient anaerobic transformation of raw wheat straw by a robust
cow rumen-derived microbial consortium, Bioresour. Technol. 196 (2015)
241e249.
[10] A. Bayane , S.R. Guiot, Animal digestive strategies versus anaerobic digestion
bioprocesses for biogas production from lignocellulosic biomass, Rev. Environ.
Sci. Biotechnol. 10 (1) (2011) 43e62.
[11] D.P. Morgavi, E. Rathahao-paris, M. Popova, J. Boccard, K.F. Nielsen, H. Boudra,
Rumen Microbial Communities Influence Metabolic Phenotypes in Lambs vol.
6, 2015, pp. 1e13.
[12] Z.-H. Hu, G. Wang, H.-Q. Yu, Anaerobic degradation of cellulose by rumen
microorganisms at various pH values, Biochem. Eng. J. 21 (1) (2004) 59e62.
[13] Z.B. Yue, W.W. Li, H.Q. Yu, Application of rumen microorganisms for anaerobic
bioconversion of lignocellulosic biomass, Bioresour. Technol. 128 (2013)
738e744.

[14] M. Cater, M. Zorec, R. Marinsek Logar, Methods for improving anaerobic
lignocellulosic substrates degradation for enhanced biogas production,
Springer Sci. Rev. 2 (1e2) (2014) 51e61.
[15] K. Porsch, B. Wirth, E.M. To th, F. Schattenberg, M. Nikolausz, Characterization
of Wheat Straw-Degrading Anaerobic Alkali-Tolerant Mixed Cultures from
Soda Lake Sediments by Molecular and Cultivation Techniques, 2015.
[16] A.M. Ziganshin, J. Liebetrau, J. Pro €ter, S. Kleinsteuber, Microbial community
structure and dynamics during anaerobic digestion of various agricultural
waste materials, Appl. Microbiol. Biotechnol. 97 (11) (2013) 5161e5174.
[17] J. Justin Kuczynski, Jesse Stombaugh, William Anton Walters,
Antonio Gonza lez, R.K. Gregory Caporaso, Using QIIME to analyze 16S rRNA
gene sequences from Microbial Communities, Curr. Protoc. Bioinforma.
(2011).
[18] R. Lucas, A. Kuchenbuch, I. Fetzer, H. Harms, S. Kleinsteuber, Long-term
monitoring reveals stable and remarkably similar microbial communities in
parallel full-scale biogas reactors digesting energy crops, FEMS Microbiol.
Ecol. 91 (3) (2015) p. fiv004-fiv004.
[19] R.C. Edgar, Supplementary Material - search and clustering orders of magni-
tude faster than BLAST, Bioinformatics 26 (19) (2010) 2460e2461.
[20] D. McDonald, M.N. Price, J. Goodrich, E.P. Nawrocki, T.Z. DeSantis, A. Probst,
G.L. Andersen, R. Knight, P. Hugenholtz, An improved Greengenes taxonomy
with explicit ranks for ecological and evolutionary analyses of bacteria and
archaea, ISME J. 6 (3) (2012) 610e618.
Please cite this article in press as: E.G. Ozbayram, et al., Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on
methane production from wheat straw, Anaerobe (2017), http://dx.doi.org/10.1016/j.anaerobe.2017.03.013
9
[21] E.P. Nawrocki, D.L. Kolbe, S.R. Eddy, Infernal 1.0: inference of RNA alignments,
Bioinformatics, 25 (10) (2009) 1335e1337.
[22] R. Lucas, J. Groeneveld, H. Harms, K. Johst, K. Frank, S. Kleinsteuber, A critical
evaluation of ecological indices for the comparative analysis of microbial
communities based on molecular datasets, FEMS Microbiol. Ecol. (2016) 1e15.
October 2016.
[23] B.D. Ondov, N.H. Bergman, A.M. Phillippy, Interactive Metagenomic Visuali-
zation in a Web Browser Interactive Metagenomic Visualization in a Web
Browser vol. 385, 2011.
[24] J.C. Oliveros. VENNY. An Interactive Tool for Comparing Lists with Venn Dia-
grams, 2007. http://bioinfogp.cnb.csic.es/tools/venny/index.html.
[25] F. Bühligen, R. Lucas, M. Nikolausz, S. Kleinsteuber, Anaerobe A T-RFLP data-
base for the rapid pro fi ling of methanogenic communities in anaerobic di-
gesters, Anaerobe 39 (2016) 114e116.
[26] X.L. Su, Q. Tian, J. Zhang, X.Z. Yuan, X.S. Shi, R.B. Guo, Y.L. Qiu, Acetobacteroides
hydrogenigenes gen. nov., Sp. nov., an anaerobic hydrogen-producing bacte-
rium in the family Rikenellaceae isolated from a reed swamp, Int. J. Syst. Evol.
Microbiol. 64 (2014) (2014) 2986e2991.
[29] D.W. Pitta, S. Kumar, B. Veiccharelli, N. Parmar, B. Reddy, C.G. Joshi, Bacterial
diversity associated with feeding dry forage at different dietary concentra-
tions in the rumen contents of Mehshana buffalo (Bubalus bubalis) using 16S
pyrotags, Anaerobe 25 (2014) 31e41.
[30] E.G. Zoetendal, C.M. Plugge, A.D.L. Akkermans, W.M. de Vos, Victivallis
vadensis gen. nov., sp. nov., a sugar-fermenting anaerobe from human faeces,
Int. J. Syst. Evol. Microbiol. 53 (1) (2003) 211e215.
[31] T. Nyonyo, T. Shinkai, M. Mitsumori, Improved culturability of cellulolytic
rumen bacteria and phylogenetic diversity of culturable cellulolytic and
xylanolytic bacteria newly isolated from the bovine rumen, FEMS Microbiol.
Ecol. 88 (3) (2014) 528e537.
[32] L. Wang, A. Hatem, U.V. Catalyurek, M. Morrison, Z. Yu, Metagenomic insights
into the carbohydrate-active enzymes carried by the microorganisms
adhering to solid digesta in the rumen of cows, PLoS One 8 (11) (2013).
[33] H.J. Flint, K.P. Scott, S.H. Duncan, P. Louis, E. Forano, Microbial Degradation of
Complex Carbohydrates in the Gut © 2012 Landes Bioscience. Do Not
Distribute © 2012 Landes Bioscience. Do Not Distribute, 2012, pp. 289e306.
[34] C.J. Meehan, R.G. Beiko, A phylogenomic view of ecological specialization in
the lachnospiraceae, a family of digestive tract-associated bacteria, Genome
Biol. Evol. 6 (3) (2014) 703e713.
[35] C.-E. Wegner, W. Liesack, Microbial community dynamics during the early
stages of plant polymer breakdown in paddy soil, Environ. Microbiol. 49
(2015) p. n/a-n/a.
[36] E.J. Gagen, J. Padmanabha, S.E. Denman, C.S. McSweeney, Hydrogenotrophic
culture enrichment reveals rumen Lachnospiraceae and Ruminococcaceae
acetogens and hydrogen-responsive Bacteroidetes from pasture-fed cattle,
FEMS Microbiol. Lett. 362 (14) (2015) 1e8.
[37] X. Goux, M. Calusinska, S. Lemaigre, M. Marynowska, M. Klocke, T. Udelhoven,
E. Benizri, P. Delfosse, Microbial community dynamics in replicate anaerobic
digesters exposed sequentially to increasing organic loading rate, acidosis,
and process recovery, Biotechnol. Biofuels 8 (1) (2015) 122.
[38] D. Rivie re, V. Desvignes, E. Pelletier, S. Chaussonnerie, S. Guermazi,
J. Weissenbach, T. Li, P. Camacho, A. Sghir, Towards the definition of a core of
microorganisms involved in anaerobic digestion of sludge, ISME J. 3 (6) (2009)
700e714.
[39] R.D. Limam, R. Chouari, L. Maze as, T. Di Wu, T. Li, J. Grossin-Debattista,
J.L. Guerquin-Kern, M. Saidi, A. Landoulsi, A. Sghir, T. Bouchez, Members of the
uncultured bacterial candidate division WWE1 are implicated in anaerobic
digestion of cellulose, Microbiologyopen 3 (2) (2014) 157e167.
[40] E. Pelletier, A. Kreimeyer, S. Bocs, Z. Rouy, G. Gyapay, R. Chouari, D. Rivie re,
A. Ganesan, P. Daegelen, A. Sghir, G.N. Cohen, C. Me digue, J. Weissenbach,
D. Le Paslier, Candidatus Cloacamonas acidaminovorans': genome sequence
reconstruction provides a first glimpse of a new bacterial division, J. Bacteriol.
190 (7) (2008) 2572e2579.
[41] M. Morrison, J. Miron, Adhesion to cellulose by Ruminococcus albus: a com-
bination of cellulosomes and Pil-proteins? FEMS Microbiol. Lett. 185 (2)
(2000) 109e115.
[42] M. Morotomi, F. Nagai, Y. Watanabe, Description of Christensenella minuta
gen. nov., sp. nov., isolated from human faeces, which forms a distinct branch
in the order Clostridiales, and proposal of Christensenellaceae fam. nov, Int. J.
Syst. Evol. Microbiol. 62 (1) (2011) 144e149.
[43] J. Downes, S.R. Vartoukian, F.E. Dewhirst, J. Izard, T. Chen, W.H. Yu,
I.C. Sutcliffe, W.G. Wade, Pyramidobacter piscolens gen. nov., sp. nov., a
member of the phylum ‘Synergistetes’ isolated from the human oral cavity,
Int. J. Syst. Evol. Microbiol. 59 (5) (2009) 972e980.