Bacterial colonisation of fresh and dried perennial ryegrass
in the rumen
J.E. Edwards (jae@aber.ac.uk),  S. Jones, R. Sanderson and A.H. Kingston-Smith
Introduction
The first step of degradation of plant material within the rumen involves
rapid colonisation of the material by a complex bacterial community1.
Previously, colonisation of conserved hay stems by celluloytic bacteria
(Fibrobacter succinogenes (Fs), Ruminococcus albus (Ra) and R.
flavefaciens (Rf)) was shown to occur at equal rates2. However, how this
compares to fresh grass is unclear.
Aim
To characterise early (<2 h) populations of rumen eubacteria and
cellulolytic bacteria colonising fresh and dried perennial ryegrass (PRG),
and determine the corresponding dry matter (DM) loss.
Method
? Fresh or dried PRG (mechanically processed to mimic mastication) was
incubated in sacco in the rumen of three rumen fistulated, non-lactating
dairy cows grazing a ryegrass sward.
? For each cow, duplicate polyester bags of each forage type were
incubated per time point (5, 10, 15, 30, 60 and 120 min) with 0 min bags
processed directly. Bag residues were hand washed and snap-frozen in
liquid N. Rumen contents were also sampled (0, 60 and 120 min) and
snap-frozen.
? DNA was extracted from the residual DM (RDM), and the colonising
bacteria analysed by eubacterial 16S ribosomal DNA based denaturing
gradient gel electrophoresis (DGGE)1 and quantitative PCR (eubacteria1,
Fs2, Ra2 and Rf2). Rumen contents were analysed similarly.
Results
? PRG preparation and incubation time did not affect the composition of
the colonising rumen eubacterial populations consistently (Fig 1).
? Colonising eubacteria, Fs, Ra and Rf increased over time (P <0.01), and
were greater with fresh PRG than dried (P <0.001) (Fig 2).
? Relative abundance of the celluloytic bacteria in rumen content was
Rf>Fs>Ra for all cows, but for colonising cellulolytic bacteria the relative
species abundance differed by cow (data not shown).
? Initial DM loss (0 min) was greater with dried PRG than with fresh (18.4
v 5.5 %; P <0.01).
? A linear interaction (P <0.001) between forage and time in terms of DM
loss (relative to 0 min) was observed (Fig 3), with dried PRG showing
greater apparent DM loss after 2 h than fresh PRG (25.0 v 7.3 %; P
<0.05).
Conclusion
? Colonising rumen bacterial populations were larger with fresh rather
than dried PRG, but this was not reflected in DM loss.
? Animal differences in relative abundances of colonising cellulolytic
bacteria were more apparent than any forage associated effect on total
population composition.
? Clarification as to whether the observed differences in initial and
ruminal DM loss may have resulted from differing responses to the
mechanical processing (to mimic mastication) is required.
Fig 1. Cluster analysis (% similarity) of DGGE profiles of the eubacteria
colonising fresh (F) and dried (D) PRG incubated in the rumen of three 
different cows (a-c). Branch labels denote incubation time (min) and PRG 
preparation (e.g. 120 F).
Fig 2. Eubacterial DNA (a) and cellulolytic bacterial DNA (b-d) on fresh 
(red) and dried (black) PRG incubated in sacco in the rumen.  Data points 
represent the mean of duplicate bags for each cow.
Fig 3. DM loss from fresh (red) and dried (black) PRG incubated in sacco in 
the rumen. Data points represent the mean of duplicate bags for each cow.
Acknowledgements
The financial support of the Biotechnology and Biological Sciences
Research Council is gratefully acknowledged.
References
1. Edwards JE, Huws SA, Kim EJ & Kingston-Smith AH (2007) FEMS 
Microbiol. Ecol. 62, 323-335.
2. Koike S, Pan J, Kobayashi Y & Tanaka K (2003) J. Dairy Sci. 86, 1429-1435.
100908070605040
60 D
30 D
120 F
60 F
30 F
120 D
10 F
5 D
15 D
10 D
15 F
5 F
0 D
0 F
100908070605040
5 D
5 F
0 D
0 F
60 D
60 F
30 D
30 F
15 F
10 F
10 D
120 D
120 F
15 D
1009080706050
60 F
30 F
120 F
30 D
15 D
60 D
15 F
10 D
10 F
120 D
5 D
5 F
0 D
0 F
(a) (b) (c)
Incubation time (min)
Log
10
(ng
DN
A/
g RDM)
(a) Eubacteria (b) Fs
(c) Ra (d) Rf
Incubation time (min)
% DM
 Loss 
(rela
tiv
e t
o 0 
mi
n)