The Rumen Metatranscriptome Landscape Reflects Dietary
1
Adaptation and Methanogenesis in Lactating Dairy Cows
2
3
Bastian Hornung
1#,*,=
,
Bartholomeus van den Bogert
2,3#,%
, Mark Davids
1,$
, Vitor A.P. Martins
4
dos Santos
1
, Caroline M. Plugge
3
, Peter J. Schaap
1,2
, Hauke Smidt
3
5
6
1
Laboratory of Systems and Synthetic Biology, Wageningen University & Research,
7
Wageningen, The Netherlands
8
2
Top Institute Food and Nutrition (TIFN), Wageningen, The Netherlands
9
3
Laboratory of Microbiology, Wageningen University & Research, Wageningen, The
10
Netherlands
11
=
Present address: Leiden Universitair Medisch Centrum LUMC, Leiden, The Netherlands
12
%
Present address: BaseClear B.V., Leiden, The Netherlands
13
$
Present address: Academisch Medisch Centrum (AMC), Amsterdam, The Netherlands
14
15
# These authors contributed equally to this work
16
Running title: Methanogenesis in the rumen metatranscriptome
17
*
Corresponding author: bastian.hornung at gmx dot (Germany)
18
Other addresses:
19
20
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted March 3, 2018. ; https://doi.org/10.1101/275883doi: bioRxiv preprint
Abstract
21
Methane eructed by ruminant animals is a main contributor to greenhouse gas emissions and
22
is solely produced by members of the phylum Euryarchaeota within the domain Archaea.
23
Methanogenesis depends on the availability of hydrogen, carbon dioxide, methanol and
24
acetate produced, which are metabolic products of anaerobic microbial degradation of feed-
25
derived fibers. Changing the feed composition of the ruminants has been proposed as a
26
strategy to mitigate methanogenesis of the rumen microbiota.
27
We investigated the impact of corn silage enhanced diets on the rumen microbiota of rumen-
28
fistulated dairy cows, with a special focus on carbohydrate breakdown and methanogenesis.
29
Metatranscriptome analysis of rumen samples taken from animals fed corn silage enhanced
30
diets revealed that genes involved in starch metabolism were significantly more expressed
31
while archaeal genes involved in methanogenesis showed lower expression values. The
32
nutritional intervention also influenced the cross-feeding between Archaea and Bacteria.
33
The results indicate that the ruminant diet is important in methanogenesis. The diet-induced
34
changes resulted in a reduced methane emission. The metatranscriptomic analysis provided
35
insights into key underlying mechanisms and opens the way for new rational methods to
36
further reduce methane output of ruminant animals.
37
38
Keywords: rumen, cow, microbiota, methane, greenhouse effect, metatranscriptome,
39
RNAseq, microbial ecology
40
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted March 3, 2018. ; https://doi.org/10.1101/275883doi: bioRxiv preprint
Introduction
41
Reduction of global greenhouse gas (GHG) output is necessary to prevent a further increase in
42
global warming, which is predicted to result in multiple detrimental effects for the
43
environment and human affairs (Schleussner et al., 2016). The necessary measures are
44
focused on the industrial and agricultural sectors in developed countries, with the aim to
45
reduce carbon dioxide, methane and other GHG emissions. One of the predominant sources of
46
methane emission, estimated to be as high as ~35% of the total anthropogenic methane
47
emissions worldwide (IPCC, 2007;McMichael et al., 2007), is the agricultural sector, and
48
especially the eructation by ruminant animals (Murray et al., 2007).
49
Ruminal microbes play a pivotal role in the breakdown of animal feed and contribute between
50
35 to 50% of the animal’s energy intake (Bergman, 1990). The ruminal microbial
51
composition is complex, with diverse populations including bacteria, archaea, fungi, and
52
protozoa. Their functional capacity is vast and has not yet been fully elucidated (Hess et al.,
53
2011;Li et al., 2012).
54
Notwithstanding the ruminal microbial complexity, methane is solely produced by a few
55
members of the phylum Euryarcheota belonging to the Archaea (Hook et al., 2010). It has
56
been shown that a change in diet can have a significant effect on the methane
emissions of
57
ruminants (Liu et al., 2011;van Gastelen et al., 2015), but the mechanisms that drive this
58
change are not fully understood. The methanogenic archaea are not directly involved in the
59
breakdown of the feed, but rely on their relationships with other community members that
60
provide the necessary substrates for methanogenesis like hydrogen, formate and methanol.
61
Microbial ecology in cows and other ruminants has been investigated using 16S ribosomal
62
RNA (rRNA) genes as molecular markers (Fernando et al., 2010;Pitta et al., 2014), the sheep
63
rumen microbial metatranscriptome has been investigated (Shi et al., 2014), and in cows
64
specialized and general microbial functions have been examined (Brulc et al., 2009;Hess et
65
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted March 3, 2018. ; https://doi.org/10.1101/275883doi: bioRxiv preprint
al., 2011;Poulsen et al., 2013;Dai et al., 2014;Dassa et al., 2014;Roehe et al., 2016;Jose et al.,
66
2017). Understanding the mechanisms that influence cow rumen methanogenesis requires
67
community-level analysis of active metabolic functions, however, a comprehensive analysis
68
of diet-dependent effects on the functional landscape of the rumen microbiota is lacking. Here
69
we investigated the effect of feed composition on bovine rumen activity patterns with a
70
special focus on methane
metabolism. By analysis of the rumen metatranscriptome landscapes
71
in animals fed mixed grass silage (GS) and corn silage (CS) diets, we were able to elucidate
72
the impact of the diet on the expression of methanogenic pathways and on the relationships of
73
methanogens with other community members.
74
75
Materials and Methods
76
77
Study design and sampling
78
The study design has been described in detail by Van Gastelen et al. (van Gastelen et al.,
79
2015). Briefly, the experiment was performed in a complete randomized block design with
80
four dietary treatments and 32 multiparous lactating Holstein-Friesian cows. Cows were
81
blocked according to lactation stage, parity, milk production, and presence of a rumen fistula
82
(12 cows). Within each block cows were randomly assigned to 1 of 4 dietary treatments. All
83
dietary treatments had a roughage-to-concentrate ratio of 80:20 based on dry matter. In the
84
four diets, the roughage consisted of either 100% GS (GS100), 67% GS and 33% CS (GS67),
85
33% GS and 67% CS (GS33), or 100% CS (GS0; all dry matter basis). This study, including
86
the rumen fluid sampling, was conducted in accordance with Dutch law and approved by the
87
Animal Care and Use Committee of Wageningen University.
88
89
Sample collection and processing
90
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted March 3, 2018. ; https://doi.org/10.1101/275883doi: bioRxiv preprint
In total, samples from 12 rumen fistulated cows, three per dietary treatment, were used for
91
metatranscriptome analysis. Rumen fluid was collected 3 hours after morning feeding on day
92
17 of the experimental period (for further details regarding the whole experimental period, see
93
(van Gastelen et al., 2015)). The samples were obtained as described previously (van
94
Zijderveld et al., 2011), and collected from the middle of the ventral sac. The rumen fluid
95
samples were immediately frozen on dry ice and subsequently transported to the laboratory
96
where the samples were stored at -80C until further analysis.
97
For RNA extraction, 1 ml rumen fluid was centrifuged for 5 min at 9000 g, after which the
98
pellet was re-suspended in 500 µl TE buffer (Tris-HCl pH 7.6, EDTA, pH 8.0). Total RNA
99
was extracted from the resuspended pellet according to the Macaloid-based RNA isolation
100
protocol (Zoetendal et al., 2006) with the use of Phase Lock Gel heavy (5 Prime GmbH,
101
Hamburg) (Murphy and Hellwig, 1996) during phase separation. The aqueous phase was
102
purified using the RNAeasy mini kit (Qiagen, USA), including an on-column DNAseI
103
(Roche, Germany) treatment as described previously (Zoetendal et al., 2006). Total RNA was
104
eluted in 30 µl TE buffer. RNA quantity and quality were assessed using NanoDrop ND-1000
105
spectrophotometer (Nanodrop Technologies, Wilmington, USA) and Experion RNA Stdsens
106
(Biorad Laboratories Inc., USA).
107
rRNA was removed from the total RNA samples using the Ribo-Zero
TM
rRNA removal Kit
108
(Meta-Bacteria; Epicentre, Madison, WI, USA) using 5 μg total RNA as input. Subsequently,
109
barcoded cDNA libraries were constructed for each of the rRNA depleted samples using the
110
ScriptSeq™ Complete Kit (Bacteria; Epicentre) according to manufacturer’s instructions in
111
combination with Epicentre’s ScriptSeq Index PCR Primers.
112
The barcoded cDNA libraries were pooled and sent to GATC Biotech (Konstanz, Germany)
113
for 150 bp single end sequencing on one single lane using the Illumina HiSeq2500 platform in
114
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted March 3, 2018. ; https://doi.org/10.1101/275883doi: bioRxiv preprint
