Directory UMM :Data Elmu:jurnal:A:Animal Feed Science and Technology:Vol87.Issue3-4.Oct2000:
Animal Feed Science and Technology
87 (2000) 231±239
The in¯uence of diet of the donor animal on the
initial bacterial concentration of ruminal ¯uid and
in vitro gas production degradability parameters
S. Nagadi, M. Herrero, N.S. Jessop*
Institute of Ecology and Resource Management, The University of Edinburgh,
West Mains Road, Edinburgh EH9 3JG, UK
Received 3 December 1999; received in revised form 10 July 2000; accepted 24 July 2000
Abstract
Six sheep were fed twice a day on a different ratio of sheep pellets and hay (20:80 (diet 1), 40:60
(diet 2) and 80:20 (diet 3)) in a replicated Latin square design to study the effect of the host diet on
the bacterial concentration of ruminal liquor and in vitro gas production degradability parameters of
cellulose, glucose and hay. Bacterial DM, bacterial absorbance and the volume of gas produced in
the absence of substrate increased as the ratio of sheep pellet to hay increased. The gas production
degradability parameters obtained from ®tting data to the model Gas B 1 ÿ expÿC tÿLag ) were
also affected by changing the ratio of sheep pellets to hay in the diet of donor animals. For each
substrate, incubation with ruminal ¯uid taken from sheep fed on diet 2 or 3 gave higher (P < 0:05)
asymptotic values `B' (except for hay), rates `C' of gas production and lower Lag times (cellulose
and hay only) than when incubated in the ruminal ¯uid taken from sheep fed on diet 1. The
digestibility of NDF from cellulose and hay was not affected by diet. Bacterial DM was strongly
related to the absorbance of ruminal ¯uid and the volume of gas produced in the absence of
substrate (R2 0:99, P < 0:001). Results suggest that changing the ratio of concentrate to hay
reduces the initial bacterial concentration and affects the gas production degradability parameters
but the estimation of bacterial DM either from bacterial absorbance or volume of gas produced
without substrate was not affected by changing the diet of donor animal. # 2000 Elsevier Science
B.V. All rights reserved.
Keywords: Gas production; Degradability; Forage quality; Rumen microbes; Techniques
*
Corresponding author. Tel.: 44-131-535-4141; fax: 44-131-667-2601.
E-mail address: [email protected] (N.S. Jessop).
0377-8401/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 7 - 8 4 0 1 ( 0 0 ) 0 0 1 9 7 - 8
232
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
1. Introduction
The in vitro gas production technique has the ability to characterise feeds not only by
the quantity of digestible carbohydrate they provide but also by the rate at which these
nutrients are released. Such characteristics are of central importance to understanding the
dynamics of rumen fermentation. For this system to be useful for routine evaluation of
forages, it must produce results with high precision and repeatability. One factor which
can in¯uence the in vitro gas production pro®les, is the microbial activity of the inoculum
(Jessop and Herrero, 1998). This could be affected by the frequency of sampling of
ruminal liquor (Nagadi et al., 1999), the time of collection (Menke and Steingass, 1988;
Pell and Scho®eld, 1993; Cone et al., 1996) and the extent of dilution with buffer (Pell
and Scho®eld, 1993; Rymer et al., 1999).
The diet of the host animal in¯uences the chemical environment within the rumen and
subsequently the microbial population of ruminal ¯uid (Weiss, 1994). Several studies
have indicated that the diet of donor animal in¯uences both the total gas production (Trei
et al., 1970; Menke and Steingass, 1988) and the gas production pro®les (Bonsi et al.,
1995; Cone et al., 1996; Huntington et al., 1998; Das and Singh, 1998).
The diet of donor animal differs between research groups that use the in vitro gas
production technique. Pell and Scho®eld (1993); Theodorou et al., 1994 and BluÈmmel
and Becker (1997) use ruminal ¯uid taken from donor animals fed on hay only, others use
ruminal ¯uid taken from host animals fed on a particular ratio of hay to concentrate
(Menke and Steingass, 1988; Beuvink et al., 1992; Cone et al., 1996) although the ratio
varies between groups and the composition of the concentrate is rarely speci®ed. These
differences in the diet of donor animals might be one of the factors that affects microbial
activity or concentration and subsequently the gas production pro®les and causes
differences in gas production data between laboratories (Moss et al., 1998). Preliminary
results have shown that the relationship between bacterial DM and bacterial absorbance
can be used to predict the microbial activity of the inoculant (Nagadi et al., 1999).
However, it is not known if this relationship is affected by the diet of the host animal.
The aims of this study were to examine the effect of varying the ratio of hay to
concentrate in the diet of the host animal on bacterial concentration and gas production
degradability parameters, and, whether the relationship between bacterial absorbance at
600 nm and bacterial DM varies across diets.
2. Materials and methods
2.1. Animals and diet
The experiment was conducted as two (replicated) 3 3 Latin squares using six
ruminally ®stulated Suffolk sheep in three periods. In each period sheep were fed twice a
day (at 8.00 and 17.00 h) on one of each of diet 1 (200 g sheep pellets and 800 g hay, diet
2 (400 g sheep pellets and 600 g hay) or diet 3 (800 g sheep pellets and 200 g hay). The
hay contained 631 g NDF and 77 g CP (N 6:25) per kg DM. The sheep nuts contained
251 g NDF, 54 g sugar, 235 g starch and 170 g CP per kg DM.
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
233
The sheep were given three weeks to adapt to each diet before and between successive
periods. Water was available to the sheep ad libitum.
2.2. Inoculum preparation
A sample of rumen contents containing both solid and liquid material was taken from
each sheep before morning feeding and collected in pre-warmed vacuum ¯asks. In the
laboratory ruminal ¯uid was ®ltered through two layers of muslin cloth and the solid
material was squeezed lightly. The strained ruminal ¯uid was mixed (1:2 v/v) with
anaerobic medium as described by Menke and Steingass (1988). Once prepared, the
suspension of micro-organisms was kept at 398C with CO2 bubbling through for approx.
20 min before addition to syringes.
2.3. Measurement of bacterial DM
Bacterial DM was measured directly by centrifuging 30 ml of the buffered rumen ¯uid
at 1000g (to remove protozoa and fungi and particulate matter) for 5 min at 308C. The
pellet was discarded and the supernatant was re-centrifuged at 26000g for 15 min at
48C. The supernatant was discarded whilst the bacterial pellet was resuspended in NaCl
solution (9 g/l) and centrifuged again. The supernatant was discarded and the bacterial
pellet was dried at 808C to constant weight for determination of bacterial DM (Henning
et al., 1991). Bacterial DM was also measured indirectly by reading the absorbance at
600 nm (Wells and Russell, 1996) of ruminal ¯uid that had been diluted ®fty fold with
buffer.
2.4. Gas production measurements and analytical procedures
The in vitro gas production was performed as described by Jessop and Herrero (1996).
Gas production was measured in triplicate without added substrate (BGV) and from the
incubation of 50 mg DM D-glucose, 200 mg DM a-cellulose, and hay (as fed to the
sheep). The cumulative gas production for each syringe was recorded at 1, 2, 3, 4, 5, 6,
8 h; thereafter every 4 h until 60 h, and then at 72, 84, 96 and 120 h and corrected for
BGV. A further correction was made for the hay sample only by subtracting the gas
produced up to 4 h which is assumed to represent the gas produced from fermentation of
soluble carbohydrate (Jessop and Herrero, 1996). Data were then ®tted to the model
Gas B 1 ÿ expÿC tÿLag ) (Krishnamoorthy et al., 1991) using the Marquart algorithm
as implemented by GraFit (Leatherbarrow, 1992) where B is the asymptotic gas
production from the fermentation of carbohydrate (ml), C the fractional rate of gas
production (/h), t time (h) and `Lag' the lag phase before the fermentation of
carbohydrate begins (h). At the end of the fermentation (120 h), residual matter was
collected from each syringe and analysed for NDF using a modi®ed micro technique
(Pell and Scho®eld, 1993). This enabled NDF digestibility to be estimated. The contents
of each syringe was emptied into 100 ml medical ¯at bottles. Each syringe was washed
three times with 5 ml of NDF solution, the washings being added to the appropriate
bottle.
234
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
2.5. Statistical analysis
Analysis of variance (Genstat, 1997) was used to compare bacterial DM, bacterial
absorbance, BGV and estimated gas production degradability parameters between
different diets. Least signi®cant differences were calculated from the standard error of the
differences between means. Regression was used to study the relationship between
bacterial DM, absorbance and BGV.
3. Results
The effect of sheep diet on the bacterial DM content of ruminal ¯uid, bacterial
absorbance and BGV are summarised in Table 1. Ruminal ¯uid from sheep fed on diet 3
had the highest (P < 0:05) mean values of bacterial DM, bacterial absorbance and BGV
followed by ruminal ¯uid obtained from sheep fed on diet 2 and 1, respectively.
The effect of sheep diet on gas production degradability parameters, NDF digestibility
and ®nal pH of glucose, cellulose and hay are presented in Table 2. It can be seen that
ruminal ¯uid taken from sheep fed on diet 2 or 3 resulted in an increased (P < 0:05)
asymptotic gas production for cellulose and glucose (B), total gas production from the fermentation of hay (A B) and fractional rate of gas production (C) from the fermentation
of glucose, cellulose and hay when compared with the ruminal ¯uid taken from sheep fed
on diet 1. The lag phase before the fermentation of carbohydrate begins (Lag) for cellulose
and hay was signi®cantly shorter when incubated in ruminal ¯uid taken from sheep fed on
diet 2 and 3 than that incubated in diet 1. The gas produced from the fermentation of
soluble material (A) for hay was signi®cantly higher when incubated in ruminal ¯uid
obtained from sheep fed on diet 2 and 3 than that incubated in diet 1 whereas the asymptotic
gas production from the fermentation of NDF (B) was not effected by the diet of donor
animal. The NDF digestibility (NDFD) of cellulose and hay was not signi®cantly
in¯uenced by source of ruminal ¯uid whether taken from sheep fed on diet 1, 2 or 3.
Cellulose, glucose and hay incubated in ruminal ¯uid taken from sheep fed on diet 3 had the
highest (P < 0:05) ®nal culture pH followed by those incubated in diet 2 and 1, respectively.
The relationship between bacterial DM and bacterial absorbance is shown in Fig. 1.
Bacterial DM was linearly related (R2 0:99, P < 0:001) to bacterial absorbance such that
Bacterial DM mg=10 ml of strained rumen fluid
863 SE 4:5 Bacterial Absorbance
Table 1
The effect of donor diet on bacterial DM, bacterial absorbance and the blanks gas volumea
Donor diet
Bacterial DM (mg/10 ml)
Bacterial absorbance (A)
Blanks gas volume (ml)
1
2
3
SED
73.9a
119.0b
155.0c
5.35
0.085a
0.137b
0.179c
0.007
11.5a
18.7b
24.5c
0.98
a
The letters a, b, c in the same column denote a signi®cant difference (P < 0:05) between means.
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
235
Table 2
The effect of donor diet on gas production parameters, NDF digestibility and ®nal pHa
Substrate
Gas production
parametersb
Glucose
B (ml)
C (/h)
Lag (h)
Final pH
Cellulose
Hay
Donor dietc
1
2
3
SED
14.6a
0.143a
0
6.56a
17.4b
0.204b
0
6.59b
17.7b
0.206b
0
6.61c
0.51
0.0038
±
0.002
B (ml)
C (/h)
Lag (h)
NDFD (g/kg DM)
Final pH
84.7a
0.073a
5.2a
999
6.53a
90.7b
0.101b
4.5b
1000
6.57b
91.1b
0.102b
4.4b
998
6.59c
0.49
0.0029
0.10
1.1
0.003
A (ml)
B (ml)
C (/h)
Lag (h)
A B (ml)
NDFD (g/kg DM)
Final pH
4.7a
30.0
0.028a
5.9a
34.7a
531
6.61a
7.1b
31.5
0.037b
5.3b
38.6b
538
6.64b
7.0b
31.9
0.040b
5.2b
39.0b
541
6.70c
0.20
1.7
0.0017
0.08
0.89
6.9
0.006
a
The letters a, b, c in the same row denote a signi®cant difference (P < 0:05) between means.
A is the gas produced at 4 h and B, C and Lag are parameters obtained by ®tting the gas production data,
corrected for A, to the equation:
b
gas B 1 ÿ eÿC tÿLag
where B is the asymptotic gas production from fermentation of NDF, C is the fractional rate of gas production
and Lag is the time taken for gas production from B to begin.
c
1 Ð sheep pellets:hay 20:80; 2 Ð sheep pellets:hay 40:60; 3 Ð sheep pellets:hay 80:20.
Fig. 1. The relationship between bacterial DM (mg/10 ml of strained rumen ¯uid) and bacterial absorbance
(n 18).
236
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
Fig. 2. The relationship between bacterial DM (mg/10 ml of strained rumen ¯uid) and the volume of gas
produced in the absence of substrate (n 18).
Fig. 2 shows the relationship between bacterial DM and BGV. There was a linear
relationship (R2 0:99, P < 0:001) between bacterial DM and BGV such that
Bacterial DM mg=10 ml of strained rumen fluid 6:34 SE 0:085 BGV ml
4. Discussion
The primary objective of this work was to study the in¯uence of donor diet on initial
microbial concentration and gas production degradability. Results clearly showed that
bacterial DM, bacterial absorbance and BGV increased as the ratio of concentrate (sheep
pellet) to hay increased in the donor animals' diet and that gas production degradability
parameters of cellulose, glucose and hay were affected when incubated in ruminal ¯uid
taken from sheep fed on diet 1. These differences might be attributed to the high ratio of
low quality hay to low ratio of concentrate in diet 1 of the donor animal in¯uencing the
chemical environment within the rumen resulting in lower microbial concentrations
(Weiss, 1994). The correspondingly low initial microbial concentration in strained
ruminal liquor (Table 1) was below that recommended by Jessop and Herrero (1998) and
Nagadi et al. (1999). These results are in agreement with Menke and Steingass (1988) who
recommended the diet of the host animal to be 50±60% hay and 40±50% concentrates since
they observed a 25% reduction in the total gas production when rumen ¯uid was taken
from sheep fed on only straw and no concentrate. They referred to the low level of microbial activity in the ruminal ¯uid of the donor animals fed on straw since, in their study,
BGV (regarded as a measure of microbial activity) was around 3 ml compared to 12±16 ml
from ruminal ¯uid taken from animals receiving a diet of hay and concentrates.
The results of this study also indicate that host diet affects the ®nal pH although
differences were small. Dietary effects on gas production degradability (B, C and Lag)
and ®nal pH observed in this study are in agreement with some earlier works. Trei et al.
(1970) studied the in¯uence of inoculum source on in vitro gas production and found that
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
237
total gas production was greater when the inoculum was obtained from grain fed steers as
compared to that from hay fed steers. Bonsi et al. (1995) studied the effect of rumen
ecology on in vitro gas production of four sun-dried fodder trees incubated in ruminal
¯uid taken from sheep fed on teff straw alone or supplemented with either Sesbania
sesban or Leucaena leucocephala. They reported that the total and the rate of gas
production from fodder trees were lower when incubated in ruminal ¯uid taken from
sheep fed on teff straw as compared with that taken from sheep fed on either S. sesban or
L. leucocephala. Cone et al. (1996) tested the in¯uence of source of ruminal ¯uid on the
rate of gas production and found that the rate of gas produced from the incubation of
maize cob mix in ruminal ¯uid taken from sheep fed on 800 g hay and 200 g concentrate
was higher than that incubated from sheep fed on hay only.
Huntington et al. (1998) studied the effect of host diet on the gas production pro®les of
grass hay and high temperature dried grass. They found that grass hay and high
temperature dried grass incubated in ruminal ¯uid taken from a silage-based diet increased
the asymptotic value, time dependent rate and ®nal pH but that Lag time, total VFA and
OMD were not affected by the diet of donor animal. Recently, Das and Singh (1998) studied
the effect of varying the level of berseem supplementation in the diet of donor animals on in
vitro gas production of wheat straw. They noted that both the total and the rate of gas
production were increased by increasing the level of berseem in donor diet up to 30%.
Jessop and Herrero (1998) proposed that if microbial activity was low, this would
become a limiting factor and a signi®cant proportion of degraded carbohydrate would be
incorporated into new microbial matter rather than being fermented to products that gave
rise to gas production. The reduced asymptotic gas production (B) but similar NDFD
observed in this study when the concentration of bacterial DM and BGV of ruminal ¯uid
taken from sheep fed on diet 1 were below the minimum values of microbial activity
recommended in earlier work (90 mg/ml and 15 ml, respectively, Nagadi et al., 1999;
Jessop and Herrero, 1998) would support this assumption.
The coef®cients relating bacterial DM to both bacterial absorbance (863 s.e. 4.5 mg
bacterial DM in 10 ml strained ruminal ¯uid per unit of absorbance) and BGV (6.34 s.e.
0.085 mg bacterial DM in 10 ml strained ruminal ¯uid per ml BGV) obtained in this
study were very close to those (bacterial absorbance (868 s.e. 39 mg bacterial DM in
10 ml strained ruminal ¯uid per unit of absorbance) and BGV (589 s.e. 0.043 mg
bacterial DM in 10 ml strained ruminal ¯uid per millilitre BGV) observed in earlier work
(Nagadi et al., 1999). This indicates that the estimation of bacterial DM by reading the
absorbance at 600 nm of ruminal ¯uid diluted 50-fold or from BGV is not affected by the
diet of donor animal. Moreover, the above relationship between bacterial DM and
bacterial absorbance could be used for monitoring the effect of donor diet on microbial
concentration of ruminal liquor and whether or not such liquor provides suf®cient
microbial matter for the in vitro gas production technique.
5. Conclusions
Results show that changes in the microbial concentration in ruminal ¯uid caused by
variation in diet alter the gas production pro®le obtained from this in vitro method. Care
238
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
should be taken to ensure that the inoculum used contains suf®cient microbial activity. As
a guide, a ratio of 60% hay and 40% concentrates in the donor animal's diet is
recommended although the quality of the hay and composition of the concentrate will
in¯uence this. The relationship between bacterial DM and bacterial absorbance can be
used to determine whether the diet of the host animal allows the attainment of a suf®cient
microbial concentration in ruminal ¯uid or not since results suggest that the diet of donor
animal does not affect the relationships between bacterial DM, bacterial absorbance,
bacterial DM and BGV.
Acknowledgements
We are grateful for the skilled technical assistance of G.F. Allan and the statistical
advice of Mr. E.A. Hunter, Biomathematics and Statistics, Scotland.
References
Beuvink, J.M.W., Spoelstra, S.F., Hogendorp, R.J., 1992. An automated method for measuring time-course of
gas production of feedstuffs incubated with buffered rumen ¯uid. Netherlands J. Agric. Sci. 40, 401±407.
Blummel, M., Becker, K., 1997. The degradation characteristics of 54 roughages and roughage NDF as
described by in vitro gas production and their relationship to voluntary feed intake. Br. J. Nutr. 77, 757±768.
Bonsi, M.L.K., Osuji, P.O., Tuah, A.K., 1995. Effect of supplementing teff straw with different levels of
leucaena or sesbania leaves on the degradabilities of teff straw, sesbania, leucaena, tagasate and veronia and
on certain rumen and blood metabolites in Ethiopian Menz sheep. Anim. Feed Sci. Technol. 52, 101±129.
Cone, J.W., van Gelder, A.H., Visscher, G.J.W., Oudshoorn, L., 1996. In¯uence of rumen ¯uid and substrate
concentration on fermentation kinetics measured with a fully automated time related gas production
apparatus. Anim. Feed Sci. Technol. 61, 113±128.
Das, A., Singh, G.P., 1998. Effect of different levels of berseem supplementation in donor animal diet on in vitro
dry matter digestibility and gas production of wheat straw. Indian J. Anim. Sci. 68, 1267±1269.
Genstat 5, 1997. Statistical Software, Release 4.1. Statistics Department, IACR-Rothamsted, Harpenden, Herts,
UK.
Henning, P.H., Steyn, D.G., Meissner, H.H., 1991. The effect of energy and nitrogen supply pattern on rumen
bacterial growth in vitro. Anim. Prod. 53, 165±175.
Huntington, J.A., Rymer, C., Givens, D.I., 1998. The effect of host diet on the gas production pro®le of hay and
high-temperature dried grass. Anim. Sci. 67, 59±64.
Jessop, N.S., Herrero, M., 1998. Modelling fermentation in an in vitro gas production system: effect of microbial
activity. In: Deaville, E.R., Owen, E., Adesogan, A.T., Rymer, C., Huntington, J.A., Lawrence, T.L.J. (Eds.),
In Vitro Techniques for Measuring Nutrient Supply to Ruminants, Occasional Publication no. 22, British
Society of Animal Science, pp. 81±84.
Jessop, N.S., Herrero, M., 1996. In¯uence of soluble components on parameter estimation using the in vitro gas
production technique. Anim. Sci. 62, 626±627.
Krishnamoorthy, U., Soller, H., Steingass, H., Menke, K.H., 1991. A comparative study on rumen fermentation
of energy supplements in vitro. J. Anim. Physiol. Anim. Nutr. 65, 28±35.
Leatherbarrow, R.J., 1992. GraFit version 3, Erithacus Software Ltd., Staines, UK.
Menke, K.H., Steingass, H., 1988. Estimation of the energetic feed value obtained from chemical analysis and in
vitro gas production using rumen ¯uid. Anim. Res. Dev. 28, 7±55.
Moss, A.R., Pell, A.N., Givens, D.I., 1998. Use of the Cornell University gas production technique to measure
kinetics of carbohydrate degradation in ®ve foods: a comparison between laboratories. In: Deaville, E.R.,
Owen, E., Adesogan, A.T., Rymer, C., Huntington, J.A., Lawrence, T.L.J. (Eds.), In Vitro Techniques for
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
239
Measuring Nutrient Supply to Ruminants, Occasional Publication no. 22, British Society of Animal Science,
pp. 81±84.
Nagadi, S., Herrero, M., Jessop, N.S., 1999. Effect of frequency of ovine ruminal sampling on microbial activity
and substrate fermentation. Proc. Br. Soc. Anim. Sci., 154 (abstract).
Pell, A.N., Scho®eld, P., 1993. Computerised monitoring of gas production to measure forage digestion in vitro.
J. Dairy Sci. 76, 1063±1073.
Rymer, C., Huntington, J.A., Givens, D.I., 1999. Effect of inoculum preparation method and concentration
method of incubation and pre-soaking the substrate on gas production pro®le of high temperature dried grass.
Anim. Feed Sci. Technol. 78, 199±213.
Theodorou, M.K., Williams, B.A., Dhanoa, M.S., Mcallan, A.B., France, J., 1994. A simple gas production
method using a pressure transducer to determine the fermentation kinetics of ruminant feeds. Anim. Feed
Sci. Technol. 48, 185±197.
Trei, J., Hale, W., Theurer, B., 1970. In vitro gas production from processed grain. J. Anim. Sci. 30, 825±831.
Weiss, W.P., 1994. Estimation of digestibility of forages by laboratory methods. In: George, C., Fahey Jr. (Eds.),
Forage Quality, Evaluation and Utilization. Madison, USA, pp. 644±681.
Wells, J.E., Russell, J.B., 1996. Why do many ruminal bacteria die and lyse so quickly? J. Dairy Sci. 79, 1487±
1495.
87 (2000) 231±239
The in¯uence of diet of the donor animal on the
initial bacterial concentration of ruminal ¯uid and
in vitro gas production degradability parameters
S. Nagadi, M. Herrero, N.S. Jessop*
Institute of Ecology and Resource Management, The University of Edinburgh,
West Mains Road, Edinburgh EH9 3JG, UK
Received 3 December 1999; received in revised form 10 July 2000; accepted 24 July 2000
Abstract
Six sheep were fed twice a day on a different ratio of sheep pellets and hay (20:80 (diet 1), 40:60
(diet 2) and 80:20 (diet 3)) in a replicated Latin square design to study the effect of the host diet on
the bacterial concentration of ruminal liquor and in vitro gas production degradability parameters of
cellulose, glucose and hay. Bacterial DM, bacterial absorbance and the volume of gas produced in
the absence of substrate increased as the ratio of sheep pellet to hay increased. The gas production
degradability parameters obtained from ®tting data to the model Gas B 1 ÿ expÿC tÿLag ) were
also affected by changing the ratio of sheep pellets to hay in the diet of donor animals. For each
substrate, incubation with ruminal ¯uid taken from sheep fed on diet 2 or 3 gave higher (P < 0:05)
asymptotic values `B' (except for hay), rates `C' of gas production and lower Lag times (cellulose
and hay only) than when incubated in the ruminal ¯uid taken from sheep fed on diet 1. The
digestibility of NDF from cellulose and hay was not affected by diet. Bacterial DM was strongly
related to the absorbance of ruminal ¯uid and the volume of gas produced in the absence of
substrate (R2 0:99, P < 0:001). Results suggest that changing the ratio of concentrate to hay
reduces the initial bacterial concentration and affects the gas production degradability parameters
but the estimation of bacterial DM either from bacterial absorbance or volume of gas produced
without substrate was not affected by changing the diet of donor animal. # 2000 Elsevier Science
B.V. All rights reserved.
Keywords: Gas production; Degradability; Forage quality; Rumen microbes; Techniques
*
Corresponding author. Tel.: 44-131-535-4141; fax: 44-131-667-2601.
E-mail address: [email protected] (N.S. Jessop).
0377-8401/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 7 - 8 4 0 1 ( 0 0 ) 0 0 1 9 7 - 8
232
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
1. Introduction
The in vitro gas production technique has the ability to characterise feeds not only by
the quantity of digestible carbohydrate they provide but also by the rate at which these
nutrients are released. Such characteristics are of central importance to understanding the
dynamics of rumen fermentation. For this system to be useful for routine evaluation of
forages, it must produce results with high precision and repeatability. One factor which
can in¯uence the in vitro gas production pro®les, is the microbial activity of the inoculum
(Jessop and Herrero, 1998). This could be affected by the frequency of sampling of
ruminal liquor (Nagadi et al., 1999), the time of collection (Menke and Steingass, 1988;
Pell and Scho®eld, 1993; Cone et al., 1996) and the extent of dilution with buffer (Pell
and Scho®eld, 1993; Rymer et al., 1999).
The diet of the host animal in¯uences the chemical environment within the rumen and
subsequently the microbial population of ruminal ¯uid (Weiss, 1994). Several studies
have indicated that the diet of donor animal in¯uences both the total gas production (Trei
et al., 1970; Menke and Steingass, 1988) and the gas production pro®les (Bonsi et al.,
1995; Cone et al., 1996; Huntington et al., 1998; Das and Singh, 1998).
The diet of donor animal differs between research groups that use the in vitro gas
production technique. Pell and Scho®eld (1993); Theodorou et al., 1994 and BluÈmmel
and Becker (1997) use ruminal ¯uid taken from donor animals fed on hay only, others use
ruminal ¯uid taken from host animals fed on a particular ratio of hay to concentrate
(Menke and Steingass, 1988; Beuvink et al., 1992; Cone et al., 1996) although the ratio
varies between groups and the composition of the concentrate is rarely speci®ed. These
differences in the diet of donor animals might be one of the factors that affects microbial
activity or concentration and subsequently the gas production pro®les and causes
differences in gas production data between laboratories (Moss et al., 1998). Preliminary
results have shown that the relationship between bacterial DM and bacterial absorbance
can be used to predict the microbial activity of the inoculant (Nagadi et al., 1999).
However, it is not known if this relationship is affected by the diet of the host animal.
The aims of this study were to examine the effect of varying the ratio of hay to
concentrate in the diet of the host animal on bacterial concentration and gas production
degradability parameters, and, whether the relationship between bacterial absorbance at
600 nm and bacterial DM varies across diets.
2. Materials and methods
2.1. Animals and diet
The experiment was conducted as two (replicated) 3 3 Latin squares using six
ruminally ®stulated Suffolk sheep in three periods. In each period sheep were fed twice a
day (at 8.00 and 17.00 h) on one of each of diet 1 (200 g sheep pellets and 800 g hay, diet
2 (400 g sheep pellets and 600 g hay) or diet 3 (800 g sheep pellets and 200 g hay). The
hay contained 631 g NDF and 77 g CP (N 6:25) per kg DM. The sheep nuts contained
251 g NDF, 54 g sugar, 235 g starch and 170 g CP per kg DM.
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
233
The sheep were given three weeks to adapt to each diet before and between successive
periods. Water was available to the sheep ad libitum.
2.2. Inoculum preparation
A sample of rumen contents containing both solid and liquid material was taken from
each sheep before morning feeding and collected in pre-warmed vacuum ¯asks. In the
laboratory ruminal ¯uid was ®ltered through two layers of muslin cloth and the solid
material was squeezed lightly. The strained ruminal ¯uid was mixed (1:2 v/v) with
anaerobic medium as described by Menke and Steingass (1988). Once prepared, the
suspension of micro-organisms was kept at 398C with CO2 bubbling through for approx.
20 min before addition to syringes.
2.3. Measurement of bacterial DM
Bacterial DM was measured directly by centrifuging 30 ml of the buffered rumen ¯uid
at 1000g (to remove protozoa and fungi and particulate matter) for 5 min at 308C. The
pellet was discarded and the supernatant was re-centrifuged at 26000g for 15 min at
48C. The supernatant was discarded whilst the bacterial pellet was resuspended in NaCl
solution (9 g/l) and centrifuged again. The supernatant was discarded and the bacterial
pellet was dried at 808C to constant weight for determination of bacterial DM (Henning
et al., 1991). Bacterial DM was also measured indirectly by reading the absorbance at
600 nm (Wells and Russell, 1996) of ruminal ¯uid that had been diluted ®fty fold with
buffer.
2.4. Gas production measurements and analytical procedures
The in vitro gas production was performed as described by Jessop and Herrero (1996).
Gas production was measured in triplicate without added substrate (BGV) and from the
incubation of 50 mg DM D-glucose, 200 mg DM a-cellulose, and hay (as fed to the
sheep). The cumulative gas production for each syringe was recorded at 1, 2, 3, 4, 5, 6,
8 h; thereafter every 4 h until 60 h, and then at 72, 84, 96 and 120 h and corrected for
BGV. A further correction was made for the hay sample only by subtracting the gas
produced up to 4 h which is assumed to represent the gas produced from fermentation of
soluble carbohydrate (Jessop and Herrero, 1996). Data were then ®tted to the model
Gas B 1 ÿ expÿC tÿLag ) (Krishnamoorthy et al., 1991) using the Marquart algorithm
as implemented by GraFit (Leatherbarrow, 1992) where B is the asymptotic gas
production from the fermentation of carbohydrate (ml), C the fractional rate of gas
production (/h), t time (h) and `Lag' the lag phase before the fermentation of
carbohydrate begins (h). At the end of the fermentation (120 h), residual matter was
collected from each syringe and analysed for NDF using a modi®ed micro technique
(Pell and Scho®eld, 1993). This enabled NDF digestibility to be estimated. The contents
of each syringe was emptied into 100 ml medical ¯at bottles. Each syringe was washed
three times with 5 ml of NDF solution, the washings being added to the appropriate
bottle.
234
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
2.5. Statistical analysis
Analysis of variance (Genstat, 1997) was used to compare bacterial DM, bacterial
absorbance, BGV and estimated gas production degradability parameters between
different diets. Least signi®cant differences were calculated from the standard error of the
differences between means. Regression was used to study the relationship between
bacterial DM, absorbance and BGV.
3. Results
The effect of sheep diet on the bacterial DM content of ruminal ¯uid, bacterial
absorbance and BGV are summarised in Table 1. Ruminal ¯uid from sheep fed on diet 3
had the highest (P < 0:05) mean values of bacterial DM, bacterial absorbance and BGV
followed by ruminal ¯uid obtained from sheep fed on diet 2 and 1, respectively.
The effect of sheep diet on gas production degradability parameters, NDF digestibility
and ®nal pH of glucose, cellulose and hay are presented in Table 2. It can be seen that
ruminal ¯uid taken from sheep fed on diet 2 or 3 resulted in an increased (P < 0:05)
asymptotic gas production for cellulose and glucose (B), total gas production from the fermentation of hay (A B) and fractional rate of gas production (C) from the fermentation
of glucose, cellulose and hay when compared with the ruminal ¯uid taken from sheep fed
on diet 1. The lag phase before the fermentation of carbohydrate begins (Lag) for cellulose
and hay was signi®cantly shorter when incubated in ruminal ¯uid taken from sheep fed on
diet 2 and 3 than that incubated in diet 1. The gas produced from the fermentation of
soluble material (A) for hay was signi®cantly higher when incubated in ruminal ¯uid
obtained from sheep fed on diet 2 and 3 than that incubated in diet 1 whereas the asymptotic
gas production from the fermentation of NDF (B) was not effected by the diet of donor
animal. The NDF digestibility (NDFD) of cellulose and hay was not signi®cantly
in¯uenced by source of ruminal ¯uid whether taken from sheep fed on diet 1, 2 or 3.
Cellulose, glucose and hay incubated in ruminal ¯uid taken from sheep fed on diet 3 had the
highest (P < 0:05) ®nal culture pH followed by those incubated in diet 2 and 1, respectively.
The relationship between bacterial DM and bacterial absorbance is shown in Fig. 1.
Bacterial DM was linearly related (R2 0:99, P < 0:001) to bacterial absorbance such that
Bacterial DM mg=10 ml of strained rumen fluid
863 SE 4:5 Bacterial Absorbance
Table 1
The effect of donor diet on bacterial DM, bacterial absorbance and the blanks gas volumea
Donor diet
Bacterial DM (mg/10 ml)
Bacterial absorbance (A)
Blanks gas volume (ml)
1
2
3
SED
73.9a
119.0b
155.0c
5.35
0.085a
0.137b
0.179c
0.007
11.5a
18.7b
24.5c
0.98
a
The letters a, b, c in the same column denote a signi®cant difference (P < 0:05) between means.
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
235
Table 2
The effect of donor diet on gas production parameters, NDF digestibility and ®nal pHa
Substrate
Gas production
parametersb
Glucose
B (ml)
C (/h)
Lag (h)
Final pH
Cellulose
Hay
Donor dietc
1
2
3
SED
14.6a
0.143a
0
6.56a
17.4b
0.204b
0
6.59b
17.7b
0.206b
0
6.61c
0.51
0.0038
±
0.002
B (ml)
C (/h)
Lag (h)
NDFD (g/kg DM)
Final pH
84.7a
0.073a
5.2a
999
6.53a
90.7b
0.101b
4.5b
1000
6.57b
91.1b
0.102b
4.4b
998
6.59c
0.49
0.0029
0.10
1.1
0.003
A (ml)
B (ml)
C (/h)
Lag (h)
A B (ml)
NDFD (g/kg DM)
Final pH
4.7a
30.0
0.028a
5.9a
34.7a
531
6.61a
7.1b
31.5
0.037b
5.3b
38.6b
538
6.64b
7.0b
31.9
0.040b
5.2b
39.0b
541
6.70c
0.20
1.7
0.0017
0.08
0.89
6.9
0.006
a
The letters a, b, c in the same row denote a signi®cant difference (P < 0:05) between means.
A is the gas produced at 4 h and B, C and Lag are parameters obtained by ®tting the gas production data,
corrected for A, to the equation:
b
gas B 1 ÿ eÿC tÿLag
where B is the asymptotic gas production from fermentation of NDF, C is the fractional rate of gas production
and Lag is the time taken for gas production from B to begin.
c
1 Ð sheep pellets:hay 20:80; 2 Ð sheep pellets:hay 40:60; 3 Ð sheep pellets:hay 80:20.
Fig. 1. The relationship between bacterial DM (mg/10 ml of strained rumen ¯uid) and bacterial absorbance
(n 18).
236
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
Fig. 2. The relationship between bacterial DM (mg/10 ml of strained rumen ¯uid) and the volume of gas
produced in the absence of substrate (n 18).
Fig. 2 shows the relationship between bacterial DM and BGV. There was a linear
relationship (R2 0:99, P < 0:001) between bacterial DM and BGV such that
Bacterial DM mg=10 ml of strained rumen fluid 6:34 SE 0:085 BGV ml
4. Discussion
The primary objective of this work was to study the in¯uence of donor diet on initial
microbial concentration and gas production degradability. Results clearly showed that
bacterial DM, bacterial absorbance and BGV increased as the ratio of concentrate (sheep
pellet) to hay increased in the donor animals' diet and that gas production degradability
parameters of cellulose, glucose and hay were affected when incubated in ruminal ¯uid
taken from sheep fed on diet 1. These differences might be attributed to the high ratio of
low quality hay to low ratio of concentrate in diet 1 of the donor animal in¯uencing the
chemical environment within the rumen resulting in lower microbial concentrations
(Weiss, 1994). The correspondingly low initial microbial concentration in strained
ruminal liquor (Table 1) was below that recommended by Jessop and Herrero (1998) and
Nagadi et al. (1999). These results are in agreement with Menke and Steingass (1988) who
recommended the diet of the host animal to be 50±60% hay and 40±50% concentrates since
they observed a 25% reduction in the total gas production when rumen ¯uid was taken
from sheep fed on only straw and no concentrate. They referred to the low level of microbial activity in the ruminal ¯uid of the donor animals fed on straw since, in their study,
BGV (regarded as a measure of microbial activity) was around 3 ml compared to 12±16 ml
from ruminal ¯uid taken from animals receiving a diet of hay and concentrates.
The results of this study also indicate that host diet affects the ®nal pH although
differences were small. Dietary effects on gas production degradability (B, C and Lag)
and ®nal pH observed in this study are in agreement with some earlier works. Trei et al.
(1970) studied the in¯uence of inoculum source on in vitro gas production and found that
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
237
total gas production was greater when the inoculum was obtained from grain fed steers as
compared to that from hay fed steers. Bonsi et al. (1995) studied the effect of rumen
ecology on in vitro gas production of four sun-dried fodder trees incubated in ruminal
¯uid taken from sheep fed on teff straw alone or supplemented with either Sesbania
sesban or Leucaena leucocephala. They reported that the total and the rate of gas
production from fodder trees were lower when incubated in ruminal ¯uid taken from
sheep fed on teff straw as compared with that taken from sheep fed on either S. sesban or
L. leucocephala. Cone et al. (1996) tested the in¯uence of source of ruminal ¯uid on the
rate of gas production and found that the rate of gas produced from the incubation of
maize cob mix in ruminal ¯uid taken from sheep fed on 800 g hay and 200 g concentrate
was higher than that incubated from sheep fed on hay only.
Huntington et al. (1998) studied the effect of host diet on the gas production pro®les of
grass hay and high temperature dried grass. They found that grass hay and high
temperature dried grass incubated in ruminal ¯uid taken from a silage-based diet increased
the asymptotic value, time dependent rate and ®nal pH but that Lag time, total VFA and
OMD were not affected by the diet of donor animal. Recently, Das and Singh (1998) studied
the effect of varying the level of berseem supplementation in the diet of donor animals on in
vitro gas production of wheat straw. They noted that both the total and the rate of gas
production were increased by increasing the level of berseem in donor diet up to 30%.
Jessop and Herrero (1998) proposed that if microbial activity was low, this would
become a limiting factor and a signi®cant proportion of degraded carbohydrate would be
incorporated into new microbial matter rather than being fermented to products that gave
rise to gas production. The reduced asymptotic gas production (B) but similar NDFD
observed in this study when the concentration of bacterial DM and BGV of ruminal ¯uid
taken from sheep fed on diet 1 were below the minimum values of microbial activity
recommended in earlier work (90 mg/ml and 15 ml, respectively, Nagadi et al., 1999;
Jessop and Herrero, 1998) would support this assumption.
The coef®cients relating bacterial DM to both bacterial absorbance (863 s.e. 4.5 mg
bacterial DM in 10 ml strained ruminal ¯uid per unit of absorbance) and BGV (6.34 s.e.
0.085 mg bacterial DM in 10 ml strained ruminal ¯uid per ml BGV) obtained in this
study were very close to those (bacterial absorbance (868 s.e. 39 mg bacterial DM in
10 ml strained ruminal ¯uid per unit of absorbance) and BGV (589 s.e. 0.043 mg
bacterial DM in 10 ml strained ruminal ¯uid per millilitre BGV) observed in earlier work
(Nagadi et al., 1999). This indicates that the estimation of bacterial DM by reading the
absorbance at 600 nm of ruminal ¯uid diluted 50-fold or from BGV is not affected by the
diet of donor animal. Moreover, the above relationship between bacterial DM and
bacterial absorbance could be used for monitoring the effect of donor diet on microbial
concentration of ruminal liquor and whether or not such liquor provides suf®cient
microbial matter for the in vitro gas production technique.
5. Conclusions
Results show that changes in the microbial concentration in ruminal ¯uid caused by
variation in diet alter the gas production pro®le obtained from this in vitro method. Care
238
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
should be taken to ensure that the inoculum used contains suf®cient microbial activity. As
a guide, a ratio of 60% hay and 40% concentrates in the donor animal's diet is
recommended although the quality of the hay and composition of the concentrate will
in¯uence this. The relationship between bacterial DM and bacterial absorbance can be
used to determine whether the diet of the host animal allows the attainment of a suf®cient
microbial concentration in ruminal ¯uid or not since results suggest that the diet of donor
animal does not affect the relationships between bacterial DM, bacterial absorbance,
bacterial DM and BGV.
Acknowledgements
We are grateful for the skilled technical assistance of G.F. Allan and the statistical
advice of Mr. E.A. Hunter, Biomathematics and Statistics, Scotland.
References
Beuvink, J.M.W., Spoelstra, S.F., Hogendorp, R.J., 1992. An automated method for measuring time-course of
gas production of feedstuffs incubated with buffered rumen ¯uid. Netherlands J. Agric. Sci. 40, 401±407.
Blummel, M., Becker, K., 1997. The degradation characteristics of 54 roughages and roughage NDF as
described by in vitro gas production and their relationship to voluntary feed intake. Br. J. Nutr. 77, 757±768.
Bonsi, M.L.K., Osuji, P.O., Tuah, A.K., 1995. Effect of supplementing teff straw with different levels of
leucaena or sesbania leaves on the degradabilities of teff straw, sesbania, leucaena, tagasate and veronia and
on certain rumen and blood metabolites in Ethiopian Menz sheep. Anim. Feed Sci. Technol. 52, 101±129.
Cone, J.W., van Gelder, A.H., Visscher, G.J.W., Oudshoorn, L., 1996. In¯uence of rumen ¯uid and substrate
concentration on fermentation kinetics measured with a fully automated time related gas production
apparatus. Anim. Feed Sci. Technol. 61, 113±128.
Das, A., Singh, G.P., 1998. Effect of different levels of berseem supplementation in donor animal diet on in vitro
dry matter digestibility and gas production of wheat straw. Indian J. Anim. Sci. 68, 1267±1269.
Genstat 5, 1997. Statistical Software, Release 4.1. Statistics Department, IACR-Rothamsted, Harpenden, Herts,
UK.
Henning, P.H., Steyn, D.G., Meissner, H.H., 1991. The effect of energy and nitrogen supply pattern on rumen
bacterial growth in vitro. Anim. Prod. 53, 165±175.
Huntington, J.A., Rymer, C., Givens, D.I., 1998. The effect of host diet on the gas production pro®le of hay and
high-temperature dried grass. Anim. Sci. 67, 59±64.
Jessop, N.S., Herrero, M., 1998. Modelling fermentation in an in vitro gas production system: effect of microbial
activity. In: Deaville, E.R., Owen, E., Adesogan, A.T., Rymer, C., Huntington, J.A., Lawrence, T.L.J. (Eds.),
In Vitro Techniques for Measuring Nutrient Supply to Ruminants, Occasional Publication no. 22, British
Society of Animal Science, pp. 81±84.
Jessop, N.S., Herrero, M., 1996. In¯uence of soluble components on parameter estimation using the in vitro gas
production technique. Anim. Sci. 62, 626±627.
Krishnamoorthy, U., Soller, H., Steingass, H., Menke, K.H., 1991. A comparative study on rumen fermentation
of energy supplements in vitro. J. Anim. Physiol. Anim. Nutr. 65, 28±35.
Leatherbarrow, R.J., 1992. GraFit version 3, Erithacus Software Ltd., Staines, UK.
Menke, K.H., Steingass, H., 1988. Estimation of the energetic feed value obtained from chemical analysis and in
vitro gas production using rumen ¯uid. Anim. Res. Dev. 28, 7±55.
Moss, A.R., Pell, A.N., Givens, D.I., 1998. Use of the Cornell University gas production technique to measure
kinetics of carbohydrate degradation in ®ve foods: a comparison between laboratories. In: Deaville, E.R.,
Owen, E., Adesogan, A.T., Rymer, C., Huntington, J.A., Lawrence, T.L.J. (Eds.), In Vitro Techniques for
S. Nagadi et al. / Animal Feed Science and Technology 87 (2000) 231±239
239
Measuring Nutrient Supply to Ruminants, Occasional Publication no. 22, British Society of Animal Science,
pp. 81±84.
Nagadi, S., Herrero, M., Jessop, N.S., 1999. Effect of frequency of ovine ruminal sampling on microbial activity
and substrate fermentation. Proc. Br. Soc. Anim. Sci., 154 (abstract).
Pell, A.N., Scho®eld, P., 1993. Computerised monitoring of gas production to measure forage digestion in vitro.
J. Dairy Sci. 76, 1063±1073.
Rymer, C., Huntington, J.A., Givens, D.I., 1999. Effect of inoculum preparation method and concentration
method of incubation and pre-soaking the substrate on gas production pro®le of high temperature dried grass.
Anim. Feed Sci. Technol. 78, 199±213.
Theodorou, M.K., Williams, B.A., Dhanoa, M.S., Mcallan, A.B., France, J., 1994. A simple gas production
method using a pressure transducer to determine the fermentation kinetics of ruminant feeds. Anim. Feed
Sci. Technol. 48, 185±197.
Trei, J., Hale, W., Theurer, B., 1970. In vitro gas production from processed grain. J. Anim. Sci. 30, 825±831.
Weiss, W.P., 1994. Estimation of digestibility of forages by laboratory methods. In: George, C., Fahey Jr. (Eds.),
Forage Quality, Evaluation and Utilization. Madison, USA, pp. 644±681.
Wells, J.E., Russell, J.B., 1996. Why do many ruminal bacteria die and lyse so quickly? J. Dairy Sci. 79, 1487±
1495.