10.1139@cjas 2019 0024
10.1139@cjas 2019 0024
10.1139@cjas 2019 0024
metatranscriptomic studies
Stephanie A. Terry1,2#, Ajay Badhan2#, Yuxi Wang2, Alexandre V. Chaves1 and Tim A.
McAllister2*
1School of Life and Environmental Sciences, Faculty of Science, University of Sydney, NSW,
Australia
2Agriculture and Agri-Food Canada, Lethbridge Research and Development Centre, 5403 1st
1
Page 2 of 42
ABSTRACT
Plant biomass is the most abundant renewable resource on the planet and the biopolymers of
protein. Plant cell walls are chemically heterogeneous structures that have evolved to provide
structural support and protection to the plant. Ruminants are the most efficient digesters of
lignocellulose owing to a rich array of bacteria, archaea, fungi and protozoa within the rumen
and lower digestive tract. Metagenomic and metatranscriptomic studies have enhanced the
current understanding of the composition, diversity and function of the rumen microbiome.
There is particular interest in identifying the carbohydrate active enzymes responsible for the
polysaccharide utilising loci in ruminal fibre degradation could provide insight into strategies to
majority of rumen microorganisms are still uncharacterised, and their ability to act
synergistically is still not understood. By advancing our current knowledge of rumen fibre
digestion, there may be opportunity to further improve the productive performance of ruminants
2
Page 3 of 42
INTRODUCTION
The sustainable and efficient conversion of lignocellulose by ruminants into protein (meat, milk
and fibre) is unparalleled. The ability of ruminants to utilise these forages plays an important
role in preserving permanent pastures as well as utilising feeds which do not compete for human
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
consumption (Kim and Dale, 2004). Low quality forages like crop residues and grasses consist
mainly of lignocellulosic polysaccharides. Despite the efficiency at which ruminants can obtain
nutrients from lignocellulosic biomass, less than 50% of energy in low quality forages is
digestible by cattle (McCartney et al. 2006), and as a result cattle are often fed cereal grains
during finishing. To further improve the conversion of low quality forage into energy, a complex
understanding of the rumen microbiota and its role in fibre digestion is required using advanced
technologies.
Can. J. Anim. Sci.
considerably enhanced our understanding of the rumen microbiome. Global initiatives including
current understanding of the rumen microbiome. For example, understanding the core
microbiome of the rumen across a range of host species could provide insight into management
and pectin (Ribeiro et al. 2016). These components make up the plant cell wall providing
structural integrity and enabling plants to resist physical or microbial damage. Lignin and
3
Page 4 of 42
cellulose have been identified as the major resistant cell wall components due to their cross
linkages with esterified xylan. Xylans have specific structural chemistry and their side-group
substitutions are interspersed, intertwined and covalently linked at various points with the
overlying sheath of lignin. This produces a coat around underlying strands of cellulose via
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
hydrogen bonding (Joseleau et al. 1992). The xylan layer, with its covalent linkage to lignin and
its non-covalent interaction with cellulose maintains the integrity of cellulose in situ and protects
Though complex, ruminants possess the unique ability to naturally degrade and obtain nutrients
from these structures. Mastication during eating and rumination mechanically damages plant
fibres, exposing inner structures and facilitating microbial colonisation (Beauchemin, 1992).
Subsequent enzyme activity is required to degrade plant cell walls, but lignocellulose is often
Can. J. Anim. Sci.
Specific enzymatic arrays of functionally diverse glycosyl hydrolases (GHs) are produced by the
rumen microbiota and are required to successfully degrade lignocellulosic biomass (Fig. 1).
Breakdown of cellulose involves the synergistic action of three classes of cellulolytic enzymes
including endo-β-1, 4-glucanase (EC 3.2.1.4; Glycosyl hydrolase family GH 5, 6, 7, 8, 9, 44, 45,
48), cellobiohydrolase (EC 3.2.1.91; GH 5, 6, 9, 48) and β-glucosidase (EC 3.2.1.21; GH 5, 9).
Hemicellulose degradation, with its variable structure, is far more complex than that of cellulose
endoxylanase (EC 3.2.1.8; GH 5, 7, 8, 10, 11, 43), β-xylosidase (EC 3.2.1.37; GH 3, 39, 43, 52,
(EC 3.2.1.55; GH 3, 10, 43, 51, 54, 62), acetyl xylan esterases (EC 3.1.1.72; CE1, 2, 4, 7),
feruloyl esterases (EC 3.1.1.73; CE 1), and α-glucuronidases (EC 3.2.1.39; GH 67,115) are also
4
Page 5 of 42
essential in plant cell wall degradation as they act to remove the side chains on the xylan
backbone. Efficient digestion of the complex structural components within the plant cell wall
requires the coordinated interaction of all these enzymes, and an understanding of the microbes
involved in this process may enable the dynamics of plant cell wall degradation to be optimised.
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Given the ability of the rumen ecosystem to successfully digest plant cell walls, it is an ideal
natural environment from which to identify and mine novel proteins with the potential to
sequencing regions of 16S rRNA, mcrA, 18S RNA or ITS-1 to describe populations of both
bacteria and archaea, archaea, protozoa and fungal communities, respectively. Metagenomics
Can. J. Anim. Sci.
involves the sequencing of DNA extracted from a sample and can describe the composition of
the rumen microbial community, and convey the relative abundance of individuals within
microbial communities. However, it does not reflect the functional potential of a rumen
microbial community as the abundance of a gene does not necessarily coincide with its metabolic
role (Ribeiro et al. 2016). Metatranscriptomics, in addition to providing the compositional data,
can highlight the activity or the expressions of genes thus allowing for their association with a
The Global Rumen Census and the Hungate1000 project are global initiatives which aim to
define the rumen microbial community. The Global Rumen Census collected a total of 742
samples from over 32 different species of rumen and hind-gut fermenters across 35 countries
(Henderson et al., 2015). This study largely found that the microbial community of the
5
Page 6 of 42
herbivory-adapted forestomach was similar worldwide, with diet being the primary factor
The Hungate1000 sought to produce a large scale reference set of rumen microbial genome
sequences from cultivated rumen bacteria, methanogens, anaerobic fungi and ciliate protozoa
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
(Seshadri et al. 2018). Creevey et al. (2014) conducted a meta-analysis to focus the Hungate1000
project using information from culture collections, the scientific literature, and NCBI and RDP
databases, as well as 16S rRNA gene based surveys that further characterized the rumen bacterial
microbiome. Although the major taxonomic groups known to play key roles within the rumen
microbiome have been identified, the collection still does not represent the full diversity of the
rumen community (Seshadri et al. 2018). This is primarily due to the low sequence similarity
between 16S rRNA genes from rumen microbes to known isolates, with at least 77% of rumen
Can. J. Anim. Sci.
microorganisms having < 97% identity with 16S rRNA sequences of known bacteria (Koike et
al. 2010).
As these studies were based on amplicon-based methods, they only described the composition of
the rumen microbial community and consequently convey little information of the mechanisms
by which enzymes achieve the extraordinary high levels of ruminal fibre digestion. The recent
herbivore gut microbiomes, leading to the identification of several novel cellulolytic enzymes.
Though this shotgun sequencing approach lacks data related to gene expression, it can identify
microbes present and provide information on the functional and metabolic capabilities of the
metagenomic and metatranscriptomic approach be utilised not only to define the composition of
6
Page 7 of 42
microbial communities, but to also gain a deeper understanding of the complex interactions
FED RUMEN
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Ruminants do not produce the enzymes involved in fibre digestion, but harbour rich and dense
microbial consortia which produce an array of enzymes that digest complex plant
polysaccharides. The volatile fatty acids (VFA) produced as result of microbial fermentation of
feed, are the principal energy source utilised by the host. Despite intense research efforts to
understand fibre degradation in the rumen, much still remains unknown. Most of the information
al. 2013) with the contribution of non-cultured microorganisms to fibre digestion being largely
unknown. Whilst there has been recent comprehensive insight into the composition of the rumen
microbiota (Table 2) and the effect that diet, geographic location and host contribute (Henderson
et al., 2015), there is incomplete understanding of the mechanisms used for fibre digestion by
could contribute to the development of strategies to manipulate fibre digestion within ruminants,
Fosmid and bacterial artificial chromosome (BAC) libraries can provide insight into rumen
microbial composition; but these methods are based on functional enrichment and not microbial
activity. Fosmid library based metagenomic analysis of a cow fed Timothy hay identified
7
Page 8 of 42
Firmicutes as the predominant phylum with major genera including Clostridium, Ruminococcus,
BAC clones from yaks fed wheat straw revealed Bacteroidetes as the predominant phylum (Dai
et al. 2012). Bacteroidetes has been proposed to be the primary degrader of complex soluble
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
polysaccharides in plant cell walls, as members of this phylum have a high diversity of GHs and
polysaccharide lyases (PLs) (Kaoutari et al. 2013; Naas et al. 2014). Polysaccharide lyases are a
class of enzymes that act to cleave active glycosidic linkages present in acidic polysaccharides.
Several studies have investigated the establishment of the rumen microbiome using amplicon-
based sequencing methods. Wu et al. (2012) described a core rumen microbiome that comprised
approximately 99% of 16S rRNA sequences. Li et al. (2015) compared microbial populations
among bovine, caprine and ovine hosts and found that whilst the rumen harboured a similar core
microbiome, there was significant variation in the abundance of the 15 families that comprised of
The Global Rumen Census found that despite differences in feeding strategies and diets, a
similar core rumen microbiome was found across ruminants over a wide geographical range
(Henderson et al., 2015). The most abundant genera within the rumen were Prevotella
8
Page 9 of 42
cellulose degrader, was abundant in forage fed cattle, whereas microbiota from concentrate fed
Seshadri et al. (2018) overlaid 16S rRNA gene sequences from the 501 Hungate1000 genomes
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
onto the 16S rRNA gene amplicon data set of the Global Rumen Census project. The Hungate
collection comprised 9 phyla, 48 families and 82 genera, and is the most comprehensive
collection of rumen bacterial genome sequences to date. In agreement with Henderson et al.
(2015), Firmicutes and Bacteroidetes accounted for the majority of the Hungate 1000 genome
sequences.
Stewart et al. (2018) presented 913 draft bacterial and archaeal genomes assembled from over
800 Gb of rumen metagenomic sequence data using both metagenomic binning and Hi-C- based
Can. J. Anim. Sci.
proximity guided assemblies. It was estimated that genomic read classification was improved by
five-fold as compared to other publicly available rumen databases, likely due to the use of
metagenomic binning which improved the classification rate without the need to culture. Hess et
al. (2011) sequenced and analysed 268 Gb of metagenomic DNA from fibre adherent microbes.
They also assembled 15 uncultured microbial genomes, which were validated by single cell
genome sequencing. Similarly, Pope et al. (2012), Svartström et al. (2017) and Gharechahi and
Salekdeh (2018) used metagenomics to identify the microbiome within the rumen contents of
reindeer, moose and camels, respectively. They reported that like the cow rumen, Bacteroidetes
and Firmicutes were the dominant phyla. However, unlike in cows, pectin digesting Spirochaetes
were the dominant phyla within the rumen of moose and camels (Svartström et al. 2017).
Although not a true ruminant, the camel microbiome was likely similar to the moose as a result
9
Page 10 of 42
With recent advancement in bioinformatics techniques like metagenomic binning and Hi-C-
based proximity guided assembly, near complete microbial genomes have been assembled
directly from rumen metagenomic sequencing data (Hess et al. 2011, Svartström et al. 2017,
Parks et al. 2017, Stewart et al. 2018). Additionally, inclusion of rumen uncultured genomes and
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Hungate 1000 genomes have substantially improved the coverage of rumen specific microbial
genomes in these public databases (Svartström et al. 2017). This full representation of the rumen
microbiome can allow for the association of feed efficiency, methane production and other
production attributes with particular microbial community structures. It can also provoke focus
on particular microbial niches that may be altered for improved ruminal fibre degradation.
The composition of the rumen microbiome and the role it plays in fibre digestion can be better
described using metatranscriptomics, as this technique reflects the expression of genes that
encode fibrolytic enzymes. Metatranscriptomic analysis of samples collected from a cow fed
corn stover (Dai et al. 2015) reported Firmicutes, Bacteroidetes, Spirochaetes, Proteobacteria,
Actinobacteria and Fibrobacteres as the most abundant bacterial phyla and Prevotella,
Roseburia as the most abundant genera (Fig. 2). Comtet-Marre et al. (2017) found that bacteria
represented 77.5% of rumen ssu rRNA reads with Firmicutes, Bacteroidetes, Fibrobacteres,
Proteobacteria, Spirochaetae and Lentisphaerae representing the only phyla with abundances
>1%. Interestingly, the majority of the total cellulose targeted GHs identified reads belonging to
the genera Ruminococcus (Firmicutes) with a high proportion of cellulase reads associated with
Fibrobacter (Fibrobacteres) (Fig. 2B; Dai et al. 2015; Comtet-Marre et al. 2017). Despite its
10
Page 11 of 42
high overall abundance Bacteroidetes was found to not play a dominant role in cellulose
degradation (Fig. 2; Dai et al. 2015). Shinkai et al. (2017) obtained a metatranscriptomic
assembly from fibre associating microbiota in Holstein cows fed timothy hay. Transcripts of GH
families were assigned to phyla Bacteroidetes, Firmicutes, and Fibrobacteres and families
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Whilst most metagenomic and metatranscriptomic studies have focused on bacterial populations,
protozoa and fungi are both significant contributors to ruminal fibre digestion. Anaerobic fungi
have been shown to act synergistically with bacteria, disrupting the plant cell wall via invasive
rhizoidal growth followed by enzymatic activity (Gruninger et al. 2014; Ribeiro et al. 2016).
Can. J. Anim. Sci.
Comtet-Marre et al. (2017) found that archaea and eukaryota represented 0.7 and 21.8% of total
ssu reads, respectively (Comtet-Marre et al. 2017). Dai et al. (2015) also found significant
eukaryal reads associated with protozoa (3-6% total non-rRNA reads) and fungi (0.6-1% of total
non-rRNA reads). Fungal cellulases accounted for ~ 9% of the total GH cellulases in the rumen
with Piromyces and Neocallimastix accounting for the largest proportion of fungal cellulases
(Dai et al. 2015; Comtet-Marre et al. 2017). Interestingly, Neocallimastigaceae produced the
largest share of cellulose transcripts of all organisms as reported by Sollinger et al. (2017).
Contrasting information exists regarding the contribution of protozoa to fibre degradation, but is
clear that they possess the functional GHs required for plant cell wall degradation (Findley et al.
2011). Protozoa (most of reads associated with Epidinium and Polyplastron) also contributed
significantly to the cellulase pool in a study by Dai et al. (2015; Fig. 2A), where the majority of
11
Page 12 of 42
(Prevotella). Fibrobacter also contributed significantly towards hemicellulase related reads (Fig.
2C; Dai et al. 2015; Comtet-Marre et al. 2017). Bacteria, fungi and protozoa that contributed
significantly to the pool of cellulases and hemicellulases were found to make a minor
contribution towards the gene pool involved in the degradation of oligosaccharides (Fig. 2D; Dai
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
et al. 2015). Sollinger et al. (2018) reported that eukaryotes accounted for 25.1% of ssu reads,
with >70% of these being associated with ciliates. Ciliates produced both hemicellulose and
cellulose transcripts in abundance and those associated with the degradation of starch to a lesser
degree. Using combined metatranscriptomics and metabolomics, Sollinger et al. (2018) found
that the microbiome was unexpectedly stable during feed digestion, with an increase in overall
microbial abundance rather than a change in population. Most cellulase transcripts stemmed
from the bacterial families Fibrobacter and Clostridiales, with the eukaryotic families
Can. J. Anim. Sci.
al. 2018).
By examining both the microbial composition and the enzymes they produce, we can associate
specific species of bacteria, archaea, fungi and protozoa with their possible role in the
degradation of fibre. Recently metagenomics has been applied to link microbiome and host
phenotypes. Relative abundance of microbial genes data from metagenomic analysis has been
used to predict methane emissions (Ross et al. 2013; Shi et al. 2014; Kamke et al. 2016; Zhang et
al. 2016; Roehe et al.2016) and feed conversion efficiencies (Shabat et al. 2016; Li and Guan
rumen microbiome, microbial fermentation and how these correlate to host phenotype variation
(Li et al. 2017). However, despite the diverse range of microbial studies it should be recognised
that estimation of relevant species abundance may not reflect the true microbial diversity within
12
Page 13 of 42
the rumen as the databases that define phylogeny are still incomplete. As these databases
continue to become more robust, the function of unidentified members of the rumen community
To understand the role of the rumen microbiota in fibre digestion, it is important to identify
enzymes involved in the synthesis, degradation and recognition of complex carbohydrates. Types
carbohydrate esterases (CEs), enzymes with auxiliary activities (AAs) and carbohydrate-binding
modules (CBMs). The GHs are the most diverse enzymes within the rumen microbiome and a
Can. J. Anim. Sci.
studies is shown in Table 3. However, it must be kept in mind that these are based on sequence
comparisons and not by isolation of proteins or the characterisation of enzyme activity. It is well
Among the 501 bacteria in the Hungate catalogue, GH13, GH43, GH3 and GH2 families were
reported as the most abundant CAZymes. Families GH5 and GH9 have been reported to be the
most diverse and predominant family of cellulases in the rumen (Brulc et al. 2009; Dai et al.
2012; Hess et al. 2011; Wang et al. 2013). In the moose (Svartström et al. 2017) and Indian dairy
cows (Jose et al. 2017), GH9 was underrepresented as compared to other ruminant species.
13
Page 14 of 42
Interestingly, all rumen metagenome studies have found a scarcity of the GH48 family
(cellobiohydrolase), despite reports suggesting this family are among the most important for the
hydrolysis of crystalline cellulose (Zhang et al. 2010). Within bovine rumen metagenome
studies, family GH6 (a family associated with endoglucanase activity) was only reported in
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
metagenomic studies by Wang et al. (2013) and Ciric et al. (2014) and not in earlier
metagenomic studies.
Endoxylanases (GH10 and GH11) were the most abundant families among hemicellulases. The
yak rumen microbiome (Dai et al.. 2012) was reported to be rich in GH67s including α-
glucuronidase (EC 3.2.1.139) and xylan α-1,2-glucuronidase (EC 3.2.1.131), while the bovine
rumen microbiome lacked GH67 CAZymes in some earlier metagenomic studies (Brulc et al.
Can. J. Anim. Sci.
2009; Wang et al. 2013). Enzymes in the GH67 family are involved in the breakdown of
The GH78 family, primarily represented by α-L-rhamnosidase (EC 3.2.1.40), catalyse the
greater relative abundance in the bovine and reindeer rumen microbiome as compared to the yak
rumen. Among oligosaccharidases, GH2, GH3 and GH43 have been reported to be most
Close interaction between hydrolytic enzymes and target insoluble substrates is required for
effective saccharification. Cellulose binding modules (CBM) exhibit a high affinity for cellulose
their degradation (Badhan et al. 2007a). Rumen metagenomic studies have shown an under
representation of CBMs, which may reflect the inability of pyrosequencing to detect the smaller
14
Page 15 of 42
modules within short reads (Brulc et al. 2009). However, GH genes recovered from BAC clones
generated from yak rumen contents were also reported to lack CBMs. Furthermore, none of the
GH family proteins retrieved from active yak BAC clones (Dai et al. 2012) or the bovine
metagenome (Hess et al. 2011) showed dockerin or cohesion modules. Likewise, in spite of
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
extensive metagenomic sequencing, very little information about the eukaryotic components of
rumen microbiome has been generated. A small fraction of the genes coding for GH families
were attributed to eukaryotes, even though it is widely assumed that rumen fungi and protozoa
act synergistically with bacteria to digest fibre (Qi et al. 2011; Dai et al. 2015).
The low abundance of eukaryotic DNA in the rumen metagenome, inefficient sample preparation
methods and the biased bioinformatics analysis for annotation of novel eukaryotic gene
Can. J. Anim. Sci.
sequences in rumen databases are a few of the factors that likely contribute to the lack of
information about anaerobic eukaryotes within rumen metagenomes. The low GC content DNA
from fungi, protozoa and the frequent occurrence of microsatellites often leads to fragmented
assembly, thereby avoiding detection when using the Illumina short read data (Ross et al. 2013;
Edwards et al. 2017). The long-read sequence technology, Single Molecule Real Time (SMRT)
final assemblies with low contig number and superior scaffold length (Edwards et al. 2017).
Long sequencing technology in combination with improved high molecular weight DNA
(>10kb) isolation has been successfully implemented most recently to sequence the genome of
the fungal species N. californiae, Pir. Finnis and A. robustus (Youssef et al. 2013; Lam et al.
15
Page 16 of 42
families (Qi et al. 2011; Sollinger et al. 2017) including a large number of novel putative GH6,
GH48 and swollenin genes. Metatranscriptomic studies suggested that exoglucanases are highly
crystalline cellulose (Qi et al. 2011). Similarly, Dai et al. (2015) undertook pyrosequencing of
total (prokaryotic and eukaryotic) rumen cDNA libraries from the bovine rumen and reported
GH9 and GH48 families were the predominant cellulases produced by rumen anaerobic fungi
and protozoa (Dai et al. 2015; Sollinger et al. 2017). Interestingly the GH48 family, which was
reported to be of low abundance in the rumen by earlier metagenomics studies, was found to be
abundant when examined from a metatranscriptomic perspective (Qi et al. 2011; Dai et al. 2015;
Can. J. Anim. Sci.
Comtet-Marre et al. 2017; Li and Guan. 2017). Most of the GH48 enzymes were associated with
importance of both fungi and bacteria in the saccharification of crystalline cellulose (Dai et al.
2015). The family GH6 was not detected in most metagenomic studies, however it was found in
Transcripts targeting xylan, mannan and pectin degradation were also reported with GH8, GH10,
GH11, GH26, GH28, GH51, GH53, GH67 and GH78 being the prominent hemicellulose
degrading families. Transcripts responsible for oligosaccharide degradation (GH1, GH2, GH3,
GH29, GH35, GH38, GH39, GH42, GH43 and GH94 family) were also abundant. Although the
rumen has been shown to possess a rich and diverse repertoire of GH families,
metatranscriptomic and metagenome studies to date suggest a scarcity of GH7, GH44 (cellulase),
GH12, GH52 and GH62 (xyloglucanase and arabinofuranosidase). Moreover, the total tract
16
Page 17 of 42
indigestible residues obtained from bovine faeces were shown to be rich in undigested esterified
arabinoxylan, suggesting ruminal arabinofuranosidase and acetyl xylan esterase as rate limiting
activities (Badhan et al. 2015). Similarly, supplementing rumen mixed enzymes with exogenous
aerobic GH7, acetyl xylan esterase and arabinofuranosidase activity has been shown to enhance
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
the saccharification efficiency of mixed rumen enzymes (Badhan et al. 2014a, 2014b).
The metatranscriptome from muskoxen (Qi et al. 2011) and the bovine rumen (Dai et al. 2015;
Shinkai et al. 2016) identified several multiple domain enzymes with CBMs being most
predominant. A eukaryotic gene expression study in muskoxen identified CBM10 as the most
abundant accessory module associated with many rumen fungal glycoside hydrolases, while
CBM1, CBM6, CBM13 and CBM18 were other major cellulose binding modules found in the
Can. J. Anim. Sci.
eukaryotic transcriptome. Shinkai et al. (2016) identified CBM6 as the most abundant in the
bovine rumen, followed by CBM4, CBM9, CBM11, and CBM2. Dai et al. (2015) identified a
greater diversity of CBM families including CBM2, CBM3, CBM4/9/16/22, CBM6, CBM10,
CBM11, and CBM13 associated with plant cell wall degrading catalytic domains of both
loci (PULs) (i.e., SusC, SusD, SusD-like, SusD-like_2, or SusD-like_3) linked with fibrolytic
genes were also reported in metatranscriptomic analyses. PULs are genomic regions which
encode for the binding, transport and depolymerisation of specific glycan structures (Seshadri et
al. 2018).
Metagenomic sequencing tends to favour the most numerically abundant genes with gene
abundance not necessarily reflecting the degree of gene expression. The fact that the gene pool
observed in the metagenome reflects the most abundant genes is evident from the comparative
17
Page 18 of 42
GH profile analysis reported by Qi et al. (2011). The GH profile of the rumen metagenome
studies clustered closely to the GH profile of Prevotella ruminicola, one of the most predominant
bacterial species within the rumen, with no known active role in the degradation of recalcitrant
cellulose (Qi et al. 2011). In contrast, the GH profile of the metatranscriptome, clustered closely
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
with the GH profile of cellulolytic bacteria and fungi (Qi et al. 2011).
Whilst these studies revealed differences in abundance of some GH classes in different ruminant
species, these noted differences are most likely a result of differences in diet composition.
Experimental differences must also be taken into account when evaluating this collation of
research results where detection of particular domains and motifs may depend on the methods of
DNA / RNA extraction, sequencing method and depth, data processing procedures and the types
Can. J. Anim. Sci.
can provide a more in depth understanding of the gut microbiome and the mechanisms that
contribute to the efficiency of ruminal fibre digestion. Gene expression analysis to study
CAZymes is critical for enzyme discovery and improved understanding of the rate limiting
enzymes involved in fibre degradation within the rumen. Such information is vital for
development of efficient enzyme technologies for enhanced utilisation of low quality forage fibre
in the rumen than can improve the cost effectiveness and sustainability of cattle production.
WALL BREAKDOWN
Cellulosomes
18
Page 19 of 42
Cellulosomes are cell associated systems in which a small module (dockerin) attached to glycan-
cleaving enzymes gathers various catalytic units onto a cognate cohesion of repeats found on
large scaffolding proteins (Seshadri et al. 2018). Most of the cellulase and xylanase activity of
rumen bacteria use these systems (Miyazaki et al. 1997), however cellulosomes have only been
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
reported in a small number of bacterial species, specifically within the family Ruminococcaceae.
However, the abundance of CBM domains associated with anaerobic fungi suggests that
cellulosomes play a predominant role in enabling these eukaryotes to digest fibre (Gruninger et
al. 2018).
Recently Bensoussan et al. (2017), adopted bioinformatics screening for cellulosome modules
cellulosome components across different taxonomical lines in rumen. This study suggested broad
Interestingly dockerin-containing proteins from Bacteroidetes specifically for the Prevotella and
components in this phylum. They also identified 15 rumen uncultured genomes that potentially
involving six Ruminococcal strains (Ruminococcus flavefaciens (strain FD-1, 007c and 17),
Ruminococcus albus (strain 7,8 and SY3) and Fibrobacter succinogenes (Dassa et al. 2014).
Genetic elements for elaborate cellulosome systems with multiple cohesion and dockerin
containing proteins were observed in genomes of all strains of R. flavefaciens, with all three
19
Page 20 of 42
strains of R. albus presenting as cohesion-deficient and to encode mainly dockerins and a unique
Considering the absence of multiple cohesin-bearing scaffoldins and many CAZymes with
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
dockerin between R. flavifaciens and R. albus to produce a putative hybrid cellulosome has been
proposed (Dassa et al., 2014). Similarly a global transcriptomic analysis (Christopherson et al.,
2014) suggested utilisation of a CBM37 by R. albus as a tool for a coordinated response to fibre
anchoring of cellulolytic and associated enzymes has been suggested based on 1) a high copy
number for CBM-37 in R. albus, 2) the ability of CBM37 to attach enzymes directly to the
Can. J. Anim. Sci.
bacterial cell surface (Ezer et al. 2008), 3) the distribution of CBM37 in many R. albus enzymes;
orthologues for these CAZymes in R. flavefacins were observed to contain dockerins and 4) all
CBM37 (Dassa et al. 2014). Interestingly Fibrobacter succinogenes, another dominant fibrolytic
bacteria, lacks cellulosomes, but rather exclusively encodes for some GH families like GH45
GH116 (β-glucosidase and β-xylosidase, PL-14, CE-6 and CMB), which appear to be absent in
R. flavefaciens and R. albus strains (Dassa et al. 2014). Contributing exclusive families of
CAZymes by microbes likely confers a competitive advantage to the bacterium along with
20
Page 21 of 42
A PUL is a set of physically-linked genes expressing cell bond enzyme activities that enable a
bacterium to respond to a specific glycan by binding to it, degrading it and importing the
reported putative PUL-encoded carbohydrate active enzymes (Pope et al. 2010; Hess et al. 2011;
Qi et al. 2011; Dai et al. 2015; Seshadri et al. 2018). Specifically starch utilisation systems (Sus),
a name given to a single gene cluster consisting of eight genes responsible for a multiprotein
Seshadri et al. (2018), recently deposited PULs derived from 64 Bacteroidetes genomes from
Hungate catalogue to PULDB database. Similarly Stewart et al. (2018), predicted 1743 putative
PUL from the Prevotellaceae family. Sequencing of longer contigs revealed that plant cell wall
degrading genes from Bacteroidetes, Firmicutes and Prevotella often cluster together, suggesting
that these genes are co-expressed and function synergistically (Dai et al. 2012; Wang et al.
2013). Furthermore, SusD (starch utilisation system) and SusC-like (which encode an integral-
membrane protein) genes were observed to be located upstream of cellulolytic gene clusters in
Bacteroidetes. SusC/SusD-like genes code for proteins that play a role in transporting
The PUL reported by Wang et al. (2013) and Dai et al. (2012) included transcriptional regulators
(from family LacI, AraC, MerR and XRE.), nutrient or ion transporters (ABC transporters and
21
Page 22 of 42
functional screening of a fosmid library generated from DNA derived from rumen microbiota
(enriched on Rhodes grass) identified several unique Sus-like PULs from Bacteroidetes (Pope et
al. 2010; Rosewarne et al. 2014). These PULs included a putative SusC-like protein homolog of
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
TonB-dependent receptor family and a protein similar to those of the SusD-like family. A
somewhat different genomic organization has been reported for Firmicutes, as GH gene clusters
lack the SusC/SusD-like system and ABC transporters (ATP binding), or the PTSII genes
alternative sugar transport system and differences in the preferred oligosaccharides of Firmicutes
A highly conserved and xylan specific inducible gene cluster (Xus; xylan utilisation system) of
xylanolytic Bacteroidetes has been described by Dodd et al. (2011). The predicted model for
Xus in Prevotella bryantii B14 included two tandem repeats of SusC/SusD homologues (XusA,
XusB, XusC, XusD), followed by an outer membrane bound XusE, encoding a hypothetical
protein and a xyn10C gene encoding an endoxylanase. The model predicted that XusE, Xyl10C
and XusB/D proteins bind to extracellular xylan and degrade it to oligosaccharides through
endoxylanase activity of Xyl10C. The xylo-oligomers produced are in turn transported across the
outer membrane by a XusB/D protein into the periplasm through the TonB-dependent receptor of
XusA and XusC. With internalization of the xylo-oligomer, an unknown signalling cascade
results in the induction of xylan utilisation genes including members of the GH3, GH5, GH 8,
GH10, GH43, GH51 and GH67 families (Miyazaki et al. 2005; Dodd et al. 2011).
22
Page 23 of 42
genes has been reported in aerobic fungi as high molecular weight polymeric forms of cell wall
components like xylan and cellulose are unable to directly enter the fungal cell. Researchers have
suggested that low molecular-weight hydrolysis products of xylan and cellulose produced as
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
result of constitutively expressed activity, enter the cell and induce the expression of hydrolytic
enzymes which synergistically achieve saccharification (Haltrich et al. 1996; Badhan et al.
2007b).
A better understanding of the saccharification strategies adopted by rumen microbes and their
transcriptional regulation in anaerobic bacteria and fungi could help develop strategies to
enhance the conversion of recalcitrant plant cell walls This knowledge gap will undoubtedly
Can. J. Anim. Sci.
continue to shrink with the growing number of whole genome sequences, enabling comparative
analysis of cellulosome, PULs and regulatory mechanisms, as well as the taxa from which they
are generated.
Fibre digestion by the rumen microbiome is of current research interest. Improved understanding
of the microbial communities and the strategies they employ to degrade recalcitrant feed has
many applications for ruminant production. This includes improving feed utilisation, the
sustainability of production, animal health and performance. Currently, research into using
exogenous enzymes to improve fibre digestibility in the rumen have been largely unsuccessful.
An improved understanding of the dynamics of the rumen microbial population and their
synergistic interactions for effective fibre degradation is essential for developing strategies to
enhance ruminal feed conversion. Enzyme discovery with meta-omic approaches targeted at
23
Page 24 of 42
detecting high Vmax or rate limiting enzyme activities within the rumen could lead to the
development of effective enzyme formulations with the potential to enhance the digestion of low
enzyme activities that act synergistically with the hydrolases of rumen microbial communities
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
could also potentially improve fibre utilisation within the rumen. Future advancement in various
omics approaches can extensively improve our understanding about the mode of nutrient
utilisation by rumen microbes from low quality forages. Due to posttranscriptional modification
and regulations, the correlation between transcriptomic and proteomics are often low and
variable (Greenbaum et al. 2003; Csardi et al. 2015). Development of direct methods to quantify
and characterise translational processes like metaRibo-Seq (Ribosome profiling) hold promise.
metabolomics analysis into models that describe microbial activity at all levels, could improve
Conflict of interest
Acknowledgements
Authors wish to acknowledge the support of Alberta Agriculture and Forestry, Genome Canada,
Genome Alberta, and Genome Quebec for their generous support. Stephanie A Terry
acknowledges the financial support of Meat and Livestock Australia (MLA) during her PhD
candidacy. We also kindly acknowledge Mohammad Marami Miliani for his assistance with
figures.
24
Page 25 of 42
References
Badhan, A., and McAllister, T.A. 2016. Designer Plants for Biofuels: A Review. Curr.
Metabolomics. 4: 49-55.
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Badhan, A., Jin, L., Wang, Y., Han, S., Kowalczys, K., Brown, D.C., Ayala, C.J., Latoszek-
Green, M., Miki, B., Tsang, A., and McAllister, T. 2014a. Expression of a fungal ferulic
acid esterase in alfalfa modifies cell wall digestibility. Biotechnol. Biofuels. 7: 39.
Badhan, A., Wang, Y., Gruninger, R., Patton, D., Powlowski, J., Tsang, A., and McAllister, T.
pretreated alfalfa hay and barley straw by rumen enzymes and commercial cellulases.
Badhan, A., Wang, Y., Gruninger, R., Patton, D., Powlowski, J., Tsang, A., and McAllister, T.A.
of Recalcitrant Cell Wall Constituents Using Enzyme Fingerprinting. BioMed Res. Int.
2015: 13.
Badhan, A.K., Chadha, B.S., Saini, H.S., 2007a. Purification of the alkaliphilic xylanases from
Myceliophthora sp. IMI 387099 using cellulose-binding domain as an affinity tag. World J
Badhan, A.K., Chadha, B.S., Kaur, J., Sonia, K.G., Saini, H.S., and Bhat, M.K. 2007b.Role of
Beauchemin, K.A. 1992. Effects of ingestive and ruminative mastication on digestion of forage
25
Page 26 of 42
Bensousssan, L., Moais, S., Dassa, B., Friedman, N., Henrissat, B., Lombard, V., Bayer, E.A.,
Brulc, J.M., Antonopoulos, D.A., Berg-Miller, M.E., Wilson, M.K., Yannarell, A.C., Dinsdale
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
,E.A., Edwards, R.E., Frank, E.D., Emerson, J.B., Wacklin, P., Coutinho, P.M., Henrissat,
B., Nelson, K.E., and White, B.A. 2009. Gene-centric metagenomics of the fibre-adherent
bovine rumen microbiome reveals forage specific glycoside hydrolases. Proc. Natl. Acad.
Christopherson, M.R., Dawson, J.A., Stevenson, D.M., Cunningham, A.C., Bramhachariya, S.,
Weinmer, P.J., Kendziorski, C., and Suen, G. 2014. Unique aspects of fiber degradation
Ciric, M., Moon, C.D., Leahy, S.C., Creevey, C.J., Altermann, E., Attwood G.T., Rakonjac, J.,
and Gagic, D. 2014. Metasecretome-selective phage display approach for mining the
Comtet-Marre, S., Parisot, N., Lepercq, P., Chaucheyras-Durand, F., Mosoni, P., Peyretaillade,
E., Bayat, A. R., Shingfield, K. J., Peyret, P., and Forano, E. 2017. Metatranscriptomics
Reveals the Active Bacterial and Eukaryotic Fibrolytic Communities in the Rumen of
Creevey, C. J., Kelly, W. J., Henderson, G., and Leahy, S. C. 2014. Determining the culturability
Csárdi, G., Franks, A., Choi, D.S., Airoldi, E.M. and Drummond, D.A. 2015. Accounting for
26
Page 27 of 42
Processes, Largely Determine Steady-State Protein Levels in Yeast. PLOS Genetics. 11:
e1005206
Dai, X., Zhu, Y., Luo, Y., Song, L., Liu, D., Liu, L., Chen, F., Wang, M., Li, J., Zeng, X., Dong,
Z., Hu, S., Li, L., Xu, J., Huang, L., and Dong, X. 2012.Metagenomic Insights into the
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Dai, X., Tian, Y., Li, J., Su, X., Wang, X., Zhao, S., Liu, L., Luo, Y., Liu, D., Zheng, H., Wang,
J., Dong, Z., Hu, S., and Huang, L. 2015. Metatranscriptomic Analyses of Plant Cell Wall
Dassa, B., Borovok, I., Ruimy-Israeli, V., Lamed, R., Flint, H.J., Duncan, S.H., Henrissat, B.,
Coutinho, P., Morrison, M., Mosoni.P., Yeoman, C.J., White, B.A., and Bayer. E.A. 2014.
Can. J. Anim. Sci.
Dodd, D., Mackie, R.I., and Cann, I.K.O. 2011. Xylan degradation, a metabolic property shared
Edwards, J.E., Forster, R.J., Callaghan, T.M., Dollhofer, V., Dagar, S.S., Cheng, Y., Chang, J.,
Kittelmann, S., Fliegerova, K., Puniya, A.K, Henske, J. K., Gilmore, S. P., O'Malley, M.
A., Griffith, G. W., Smidt, H.. 2017. PCR and Omics Based Techniques to Study the
Ezer, A., Matalon, E., Jindou, S., Borovok, I., Atamna, N., Yu, Z., Morrison, M., Bayer, E.A.,
27
Page 28 of 42
Findley, S. D., Mormile, M. R., Sommer-Hurley, A., Zhang, X. C., Tipton, P., Arnett, K., Porter,
J. H., Kerley, M., and Stacey, G. 2011. Activity-based metagenomic screening and
Gharechahi, J., and G. H. Salekdeh. 2018. A metagenomic analysis of the camel rumen’s
microbiome identifies the major microbes responsible for lignocellulose degradation and
Greenbaum, D., Colangelo, C., Williams, K., and Gerstein, M. 2003. Comparing protein
abundance and mRNA expression levels on a genomic scale. Genome Biol, 4: 117.
Gruninger, R.J., Puniya, A.K., Callaghan, T.M., Edwards, J.E., Youssef, N., Dagar, S.S.,
Fliegerova, K., Griffith, G.W., Forster, R., Tsang, A., McAllister, T., and Elshahed, M.S.
Can. J. Anim. Sci.
taxonomy, life cycle, ecology, role and biotechnological potential. FEMS Microbiol. Ecol.
90: 1-17.
Gruninger, R. J., Nguyen, T. T. M., Reid, I., Yanke, L. J., Wang, P., Abbott, D. W., Tsang, A.
active enzymes expressed by diverse genera of anaerobic fungi isolated from the rumen.
Haitjema, C.H., Gilmore, S.P., Henske, J.K., Solomon, K.V., de Groot, R., Kuo, A., Mondo, S.
J., Salamov, A.A., LaButti, K., Zhao, Z., Chiniquy, J., Barry, K., Brewer, H.M, Purvine,
S.O., Wright, A.T,, Hainaut, M., Boxma, B., van Alen, T., Hackstein, J.H.P., Henrissat, B.,
Baker, S.E., Grigoriev, I.V. and O'Malley, M.A. 2017. A parts list for fungal cellulosomes
28
Page 29 of 42
Haltrich, D., Nidetzky, B., Kulbe, K.D., Steiner, W., and Župančič, S., 1996. Production of
Henderson, G., Cox, F., Ganesh, S., Jonker, A., Young, W., and Janssen, P.H. 2015. Rumen
microbial community composition varies with diet and host, but a core microbiome is
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Hess, M., Sczyrba, A., Egan, R., Kim, T-W, Chokhawala, H., Schroth, G., Luo, S., Clark, D.S.,
Chen, F., Zhang, T., Mackie, R.I., Pennacchio, L.A., Tringe, S.G., Visel, A., Woyke, T.,
Wang, Z., and Rubin, E.M. 2011. Metagenomic Discovery of Biomass-Degrading Genes
Huws S.A., Creevey, C.J., Oyama, L.B., Mizrahi, I., Denman, S.E., Popova, M., Muñoz-
Tamayo, R., Forano, E., Waters, S.M., Hess, M., Tapio, I., Smidt, H., Krizsan, S.J., Yáñez-
Can. J. Anim. Sci.
Ruiz, D.R., Belanche, A., Guan, L., Gruninger, R.J., McAllister, T.A., Newbold, C.J.,
Roehe, R., Dewhurst, R.J., Snelling, T.J., Watson, M., Suen, G., Hart, E.H., Kingston-
Smith, A.H., Scollan, N.D., do Prado, R.M., Pilau, E.J., Mantovani, H.C., Attwood, G.T.,
Edwards, J.E., McEwan, N.R., Morrisson, S., Mayorga, O.L., Elliott, C. and Morgavi, D.P.
Jose, V. L., Appoothy, T., More, R. P., & Arun, A. S. 2017. Metagenomic insights into the
rumen microbial fibrolytic enzymes in Indian crossbred cattle fed finger millet straw. AMB
Joseleau, J.P., Comtat, J., and Ruel, K. 1992. Chemical structure of xylans and their interactions
in the plant cell walls, in: Visser, J., Beldman, G., VanSomeren, M.A.K., Voragen, A.G.J.,
29
Page 30 of 42
Kaoutari, A.E., Armougom, F., Gordon, J.I., Raoult, D., and Henrissat, B. 2013.The abundance
and variety of carbohydrate-active enzymes in the human gut microbiota. Nat. Rev. Micro.
11: 497-504.
Kim, S., and Dale, B.E. 2004.Global potential bioethanol production from wasted crops and crop
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Koike, S., Handa, Y., Goto, H., Sakai, K., Miyagawa, E., Matsui, H., Ito,S., and Kobayashi, Y.
2010. Molecular Monitoring and Isolation of Previously Uncultured Bacterial Strains from
Lam, K., Tse, D., LaButti, K. and Khalak, A. 2015. FinisherSC: a repeat-aware tool for
Li, R.W., Connor, E.E., Li, C., Baldwin, V.I.R.L., and Sparks, M.E. 2012. Characterization of
Can. J. Anim. Sci.
the rumen microbiota of pre-ruminant calves using metagenomic tools. Environ. Microbiol.
14: 129-139.
Li, R.W. 2015.Rumen metagenomics, in: Puniya, A.K., Singh, R., Kamra, D.N., (Eds.), Rumen
Li, F., and Guan, L.L. 2017. Metatranscriptomic profiling reveals linkages between the active
rumen microbiome and feed efficiency in beef cattle. Appl Environ Microbiol. 83:e00061-
17
Li, F., Andre, L.A.N., Ghoshal, B., and Guan, L. L. 2018. Symposium review: Mining
metagenomic and metatranscriptomic data for clues about microbial metabolic functions in
30
Page 31 of 42
McCartney, D.H., Block, H. C., Dubeski, P. L., and Ohama, A. J. 2006. Review: The
composition and availability of straw and chaff from small grain cereals for beef cattle in
Miyazaki, K., Martin, J. C., Marinsek-Logar, R., and Flint, H. J. 1997. Degradation and
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
utilization of xylans by the rumen anaerobe Prevotella bryantii B 14. Anaerobe. 3: 373–
381.
Miyazaki, K., Hirase, T., Kojima, Y., and Flint, H.J. 2005. Medium- to large-sized xylo-
Montella, S., Ventorino, V., Lombard, V., Henrissat, B., Pepe, O., and Faraco, V. 2017.
Morgavi, D.P., Kelly, W.J., Janssen, P.H., and Attwood, G.T. 2013. Rumen microbial
Naas, A.E., Mackenzie, A.K., Mravec, J., Schückel, J., Willats, W.G., Eijsink, V.G., and Pope,
Parks, D.H., Rinke, C., Chuvochina, M., Chaumeil, P.A., Woodcroft, B.J., Evans, P.N.,
Hugenholtz, P., and Tyson, G.W. 2017. Recovey of nearly 8000 metagenome assembled
Pope, P.B., Denman, S.E., Jones, M., Tringe, S.G., Barry, K., Malfatti, S.A., McHardy, A.C.,
Cheng, J.F., Hugenholtz, P., McSweeney, C,S., and Morrison, M. 2010. Adaptation to
31
Page 32 of 42
herbivory by the Tammar wallaby includes bacterial and glycoside hydrolase profiles
different from other herbivores. Proc. Natl. Acad. Sci. 107: 14793-14798.
Qi, M., Wang, P., O'Toole, N., Barboza, P.S., Ungerfeld, E., Leigh, M.B., Selinger, L.B., Butler,
G., Tsang, A., McAllister, T.A., and Forster, R.J. 2011. Snapshot of the Eukaryotic Gene
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Ribeiro, G.O., Gruninger, R., Badhan, A., and McAllister, T.A. 2016. Mining the rumen for
Rosewarne, C.P., Pope, P.B., Cheung, J.L., and Morrison, M. 2014. Analysis of the bovine
rumen microbiome reveals a diversity of Sus-like polysaccharide utilization loci from the
Ross, M.G., Russ, C., Costello, M., Hollinger, A., Lennon, N.J., Hegarty, R., Nusbaum, C. and
Can. J. Anim. Sci.
Jaffe, D.B. 2013. Characterizing and measuring bias in sequence data. Genome Biol. 14:
R51.
Ross, E.M., Moate, P.J., Marett, L.C., Cocks, B.G., and Hayes, B.J. 2013. Metagenomic
Seshadri, R., Leahy, S.C., Attwood, G.T., Teh, K.H., Lambie, S.C., Cookson, A.L., Eloe-
Fadrosh, E.A., Pavlopoulos, G.A., Hadjithomas, M., Varghese, N.J., Paez-Espino, D.,
collaborators, H. project, Perry, R., Henderson, G., Creevey, C.J., Terrapon, N., Lapebie,
P., Drula, E., Lombard, V., Rubin, E., Kyrpides, N.C., Henrissat, B., Woyke, T., Ivanova,
N.N., and Kelly, W.J. 2018. Cultivation and sequencing of rumen microbiome members
from the Hungate1000 Collection. Nat. Biotechnol. 36: 359. Singh, N. K. 2015.
Proteomics in studying the molecular mechanisms of fibre digestion. Pages 100 – 100 in
32
Page 33 of 42
Shabat, S., Sasson, K.G., Doron-Faigenboim, A., Durman, T., Yaa-coby, S., Berg Miller, M.E.,
mechanisms underlie the en-ergy harvest efficiency of ruminants. ISME J. 10: 2958–2972.
Shi, W., Moon, C.D., Leahy, S.C., Kang, D., Froula, J., Kittelmann, S., Fan, C., Deutsch, S.,
Gagic, D., Seedorf, H., Kelly, W.J., Atua, R., Sang, C., Soni, P., Li, D., Pinares-Patiño,
C.S., McEwan, J.C., Janssen, P.H., Chen, F., Visel, A., Wang, Z., Attwood, G.T., and
Rubin,E.M. 2014. Methane yield phenotypes linked to differential gene expression in the
Shinkai, T., Mitsumori, M., Sofyan, A., Kanamori, H., Sasaki, H., Katayose, Y., and Takenaka,
Can. J. Anim. Sci.
Singh, N. K. 2015. Proteomics in studying the molecular mechanisms of fibre digestion. Pages
Söllinger, A., Tveit, A. T., Poulsen, M., Noel, S. J., Bengtsson, M., Bernhardt, J., Frydendahl
Hellwing, A. L., Lund, P., Riedel, K., Schleper, C., Højberg, O., and Urich, T. 2018.
Stewart, R. D., Auffret, M.D., Warr, A., Wiser, A.H., Press, M,O., Langford, K.W., Liachko, I.,
Snelling, T,J., Dewhurts, R,J., Walker, A.W., Roehe, Rainer., and Watson, M. 2017.
33
Page 34 of 42
Assembly of 913 microbial genomes from metagenomic sequencing of the cow rumen.
Nat.Commun. 9: 870.
Svartstrom, O., Alneberg, J., Terrapon, N., Lombard, V., Bruijn, I., Malmsten, J., Dalin, A.,
Muller, E., Shah, P., Wilmes, P., Henrissat, B., Aspeborg, H., and Andersson, A. 2017.
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Ninety-nine de novo assembled genomes from the moose (Alces alces) rumen microbiome
provide new insights into microbial plant biomass degradation. ISME J. 11: 2538-2551.
Wang, L., Hatem, A., Catalyurek, U.V., Morrison, M., and Yu, Z. 2013. Metagenomic Insights
Wu, S., Baldwin, R.L. VI, Li, W., Li, C., Connor, E.E., and Li, R.W. 2012. The bacterial
Youssef, N.H., Couger, M.B., Struchtemeyer, C.G., Liggenstoffer, A.S., Prade, R.A., Najar, F.
Z., Atiyeh, H.K., Wilkins, M.R. and Elshahed, M.S. 2013. The genome of the anaerobic
fungus Orpinomyces sp. strain C1A reveals the unique evolutionary history of a
Zhang, X.Z., Zhang, Z., Zhu, Z., Sathitsuksanoh, N., Yang, Y., and Zhang, Y.-H.P. 2010. The
86: 525–53
34
Page 35 of 42
Figure Captions
Figure 1. Plant cell wall structural complexity and the enzymes involved in its deconstruction.
Structural components are a representation only and do not reflect precise anatomical locations.
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
cellulase (A) hemicellulase (B) oligosaccharide degrading enzyme (C) read in total non-rRNA.
35
Page 36 of 42
Table 1. Advantages and disadvantages of the use of metagenomics and metatranscriptomics in rumen
microbial profiling
Advantages Disadvantages
Metagenomics Composition of microbial Does not reflect gene
DNA based approach community expression
Functional potential of Does not account for post-
microbial community transcriptional and post-
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
translational modifications
Can sequence whole genome Low sequence similarity
or segmented regions
DNA stability Under representation of
eukaryotic DNA and CBMs
Favours numerically abundant
genes
expression
Accounts for post-
transcriptional changes
Page 37 of 42
Table 2. Summary of the most abundant genera from various metagenome (MG) and metatranscriptome (MT) analysis
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Analysis Sample Most abundant phylum in order Most abundant genera in order
MG Yak, bovine (Dai et al. Bacteroidetes, Firmicutes, Fibrobacteres Information not available
2012)
Reindeer, cervid (Pope et Bacteroidetes, Firmicutes Information not available
al. 2012)
Dairy cows (Wu et al. Bacteroidetes, Firmicutes, Proteobacteria , Information not available
2012) Fibrobacteres, Spirochaetes
Jersey, bovine (Wang et al. Information not available Clostridium, Ruminococcus, Butyrivibrio, unclassified
2013) Lachnospiraceae
Bovines, caprids, cervids, Information not available Prevotella, Butyrivibrio, Ruminococcus,
camelids (Henderson et al. Lachnospiraceae, Ruminococcaceae, Bacteroidales,
2015) Clostridiales
Cattle, goats, sheep (Li et Bacteroidetes, Firmicutes, Proteobacteria, Spirochaetes, Prevotellaceae, Lachnospiraceae, Ruminococcaceae,
al. 2015) Fibrobacteres, Verrucomicrobia, Synergistetes,, Veillonellaceae, Acidaminococcaceae
Can. J. Anim. Sci.
Actinobacteria
Moose, cervid (Svartström Bacteroidetes, Firmicutes, Spirochaetes, Proteobacteria, Information not available
et al. 2017) Actinobacteria
Camel, camelid Bacteroidetes, Firmicutes, Spirochaetes, Proteobacteria, Information not available
(Gharechahi et al. 2018) Verrucomicrobia
All ruminants (Seshadri et Firmicutes, Bacteroidetes, Lachnospiraceae Information not available
al. 2018)
MT Holstein, bovine (Dai et al. Firmicutes, Bacteroidetes, Spirochaetes, Proteobacteria, Prevotella, Ruminococcus, Clostridium, Butyrivibrio,
2015) Actinobacteria, Fibrobacteres Eubacterium, Bacteroides, Treponema, Blautia,
Roseburia
Holstein, bovine (Shinkai Bacteroidetes, Firmicutes, Fibrobacteres Prevotellaceae, Fibrobacteraceae, Bacteroidacae,
et al. 2016) Lachnospiraceae, Clostridiacae
Dairy, bovine (Comtet- Firmicutes, Bacteroidetes, Fibrobacteres, Information not available
Marre et al. 2017) Proteobacteria, Spirochaetae and Lentisphaerae
Beef steers, bovine (Li et Proteobacteria, Firmicutes, Bacteroidetes, Spirochaetes, Succinivibrionaceae, Prevotellaceae, Ruminococcaceae,
al. 2017) Cyanobacteria, Synergistetes Lachnospiraceae, Veillonellaceae, Spirochaetaceae,
Dethiosulfovibrionaceae, and Mogibacteriaceae
Page 38 of 42
Table 3. Comparison of CAZy genes identified in various rumen metagenome (MG) and metatranscriptome (MT) analysis
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only.
Moose, cervid
** ND ND * * * * 4.82
(Svartström et al. 2017)
Camel, camelid
** ND ND ** * * * 6.41
(Gharechahi et al. 2018)
All ruminants
** * ND ** * * * 3.16
(Seshadri et al. 2018)
MT Muskoxen, bovine (Qi et
**** *** NDa *** ND *** *** 48.2
al. 2011)
Holstein, bovine (Dai et
*** * ND **** * ** *** 35.4
al. 2015)
Holstein, bovine (Shinkai
** * ND *** * * * 10.34
et al. 2016)
Dairy, bovine (Comtet-
** * ND ** * ** ** 13.70
Marre et al. 2017)
Crossbred steers, bovine
**** * ND **** * ** ** 29.28
(Li et al. 2017)
Hemicellulase and debranching enzymes
% of
GH8 GH10 GH11 GH12 GH26 GH28 GH51 GH53 GH54 GH62 GH67 GH78
total GH
MG Angus, bovine (Brulc et
* * * ND * * ND ** * ND ND *** 12.11
al. 2009)
Page 39 of 42
al. 2011)
Yak, bovine (Dai et al.
* **** * ND ** * ND *** * ND *** ** 33.57
2012)
Reindeer, cervid (Pope et
* ** * ND ** ** *** ** * ND ** *** 29.63
al. 2012)
Jersey, bovine (Wang et
ND **** ND * * ND ND *** ND ND ND *** 29.96
al. 2013)
Friesian, bovine (Ciric et
* ** * * ** ** ** ** * * ** ** 13.68
al. 2014)
Indian Holstein Friesian,
* ** * * * * * *** * ND * ** 12.86
bovine (Jose et al. 2017)
Moose, cervid
* ** * * ** ** ** ** ** ND * ** 13.39
(Svartström et al. 2017)
Camel, camelid
* ** * * ** ** ** ** * ND * ** 14.48
(Gharechahi et al. 2018)
All ruminants
* ** * * * * ** * * * * * 6.12
(Seshadri et al. 2018)
Can. J. Anim. Sci.
al. 2013)
Friesian, bovine (Ciric et
* *** *** ** * * * * *** * ** 29.66
al. 2014)
Indian Holstein Friesian,
** *** *** * * ** ** * ** ND * 28.94
bovine (Jose et al. 2017)
Moose, cervid
** *** *** ** ** * * * *** ND ** 28.46
(Svartström et al. 2017)
Camel, camelid
* *** *** ** ** * * * *** ND ** 28.06
(Gharechahi et al. 2018)
All ruminants
** ** ** * * * * * ** * * 16.70
(Seshadri et al. 2018)
MT Muskoxen, bovine (Qi et
** * *** ND ND * ND ND **** ND * 27.7
al. 2011)
Holstein, bovine (Dai et
*** *** **** * * * * * *** ND ** 40.19
al. 2015)
Holstein, bovine (Shinkai
* ***** **** ** * * * * **** ND ND 44.26
et al. 2016)
Can. J. Anim. Sci.
bND: not detected, *: 0<n<1, **: 1<n<5, ***: 5<n<10, ****:10<n<15, *****: n>15 where n is the relative percentage of the GH domain to the total number of
GH domain.
Can. J. Anim. Sci.
Downloaded from www.nrcresearchpress.com by RUTGERS UNIVERSITY on 08/09/19. For personal use only. Page 41 of 42