J. Anim. Physiol. a. Anim. Nutr.

86 (2002), 246256
2002 Blackwell Verlag, Berlin
ISSN 09312439

College of Animal Science and Technology, China Agricultural University, Beijing, China
and 2Institute of Animal Nutrition, Federal Agricultural Research Centre Braunschweig,
Braunschweig, Germany

The estimation of utilizable amino acids (uAA) of feeds for

ruminants using an in vitro incubation technique
By G. Y. Zhao1 and P. Lebzien2

Summary
The in vitro incubation technique of Zhao and Lebzien (2000; Arch. Anim. Nutr. 53, 293
302) was used for the estimation of utilizable amino acids (uAA) (sum of amino acids from
undegraded feed protein and microbial protein, when N is not limiting) of feeds for
ruminants. The rumen fluid from a cow fed only with hay (Expt. 1) and that from a sheep
fed with a mixed ration (Expt. 2) was compared with respect to estimation of uAA. In
Expt. 1, 30 feeds and feed mixtures were tested and in Expt. 2, 33 feeds and feed mixtures
were tested. A close linear relationship was found between the utilizable crude protein
(uCP undegraded feed protein + microbial protein) (X, g/kg) calculated from in vivo
experiments and the uAA (Y, g/kg) estimated from in vitro incubations both in Expt. 1: y
0:95 x  1:39; r2 0:85; p < 0:001; n 30; and in Expt. 2: Y 0:85X  6:67; r2
0:85; p < 0:001; n 33. Statistical analysis indicates that there was a significant regressive
relationship between uAA determined with the rumen fluid of a sheep (X, g/kg) and uAA
determined with the rumen fluid of a cow (Y, g/kg): Y 1:06X 12:4; r2
0:80; p < 0:001; n 27. The results indicate that the in vitro incubation technique of
Zhao and Lebzien (2000) can be used for the estimation of uAA of feeds for ruminants. As
a rumen-fistulated cow is more expensive than a rumen-fistulated sheep, it is suggested to
use a sheep fed a mixed ration as the donor of rumen fluid for the estimation of uAA of
feeds with in vitro incubation. Further experiments should be performed to standardize the
method and to test the most valid length of the incubation period.

Introduction
Lebzien et al. (1996) summarized the results of 532 individual cow experiments on the
measurement of crude protein (CP) flow at the duodenum, conducted in vivo with cows
fistulated at the rumen and duodenum. They concluded that the prediction of the total
crude protein supply at the duodenum is more accurate and simpler than the separate
prediction of the rumen microbial crude protein and the rumen undegraded dietary crude
protein. They proposed to predict the utilizable crude protein (uCP) at the duodenum by a
significant regression between intake of digestible organic matter (DOM) and CP and
rumen undegradable crude protein (UDP), which was later adopted in the new German
protein evaluation system (GfE, 1997).
With reference to the first stage of Tilley and Terrys (1963) in vitro digestion
technique, Zhao and Lebzien (2000) developed an in vitro incubation technique for the
estimation of the uCP of feeds for cattle. They used rumen fluid from a ruminally fistulated
dairy cow fed on hay and found a significant regressive relationship between the calculated
uCP based on in vivo data sets (Lebzien et al., 1996) and the determined uCP using the
U. S. Copyright Clearance Center Code Statement: 09312439/2002/86070246 $ 15.00/0

www.blackwell.de/synergy

The in vitro incubation technique for the estimation of uAA

247

in vitro incubation technique. They suggested that the developed in vitro incubation
technique can be used for quick and accurate estimation of uCP content of feeds. As the
uCP contains non-protein nitrogen, e.g. in the form of nucleic acid-N it will be more useful
to modify the in vitro incubation technique to estimate the utilizable amino acids (uAA) at
the duodenum for ruminants. As ruminally fistulated sheep are more commonly used than
ruminally fistulated cows in many laboratories, in particular as the donor of rumen fluid
for the Hohenheim gas-test (Naumann and Bassler, 2001), the objective of the present
experiments was to study the feasibility of estimating uAA of different feeds with an
in vitro incubation technique, and to compare the uAA estimated with rumen fluid from a
cow fed with hay with that from a sheep fed with a ration like that used when rumen fluid
for the gas-test is needed.

Materials and methods

Expt. 1: Incubation with rumen fluid from a dairy cow
Animal and feeding
A non-lactating German Friesian cow (live weight 680 kg) fitted with a large rumen fistula
(inner diameter 100 mm) was used as the donor of rumen fluid. The animal was fed
with 4 kg grass hay and 50 g mineral mixture twice daily [total dry matter intake (DMI): 7.1
kg/day] and had free access to drinking water.
Feed samples
Twenty-four air-dry feeds (Table 3) and six feed mixtures (Table 1) were ground to pass a
3-mm sieve for incubation and determination of dry matter. All feedstuffs had been
Table 1. Composition of feed mixtures (%)
Feed mixture
Ingredients

10

11

12

Barley
Beet pulp (dried)
Citrus pulp
Coconut cake
Corn
Corn gluten feed
Corn silage B
Corn starch
Faba beans
Manioc
Palm kernel cake
Protected rapeseed
meal1
Rapeseed meal 1
Rapeseed meal 2
Rapeseed meal 3
Soya bean hulls
Soya bean meal 1
Soya bean meal 2
Mineral mixture
Wheat grain

20

10

30

30

10

20
40

40

30

20

40

10

50

50

50

30

16.5

18.4

50
4.1

18.4

50
4.1

21.6

51
4.9

21.6

51
4.9

80

18

40

10

30

20

50

30

20

3.5

27

0.5

27

0.5

22

0.5

22
0.5

Formaldehyde treated

248

G. Y. Zhao and P. Lebzien

Table 2. Amino acid (AA) composition of feeds
Total
Crude
AA
protein (16 AA)
a
a

Feeds
Barley
Beet pulp
Citrus pulp
Coconut meal
Corn (n 2)
Corn gluten feed
Corn silage (n 2)
Faba bean
Grass meal
Grass silage
Hay (n 2)
Manioc
Oats
Palm kernel cake
Protected rapeseed
meal1
Rapeseed meal 2
Rapeseed meal 3
Soya bean hulls
Soya bean meal 1
Soya bean meal 2
Wheat grain
Wheat straw
Feed mixtures
Mix 1
Mix 2
Mix 3
Mix 4
Mix 5
Mix 6
Mix 7
Mix 8
Mix 9
Mix 10
Mix 11
Mix 12

Methionine

Lysine

Leucine

107
67
59
191
107
201
85
258
109
127
135
41
107
171
353

105
62
46
166
110
193
59
245
100
101
100
34
99
150
326

1.9
1.4
0.7
2.7
2.0
3.9
1.2
2.2
1.6
2.0
1.7
0.6
1.9
3.6
7.5

1.8
2.3
1.5
1.6
1.8
2.0
2.0
0.9
1.6
2.0
1.7
1.8
1.9
2.4
2.3

4.2
4.5
1.9
5.1
3.7
8.5
1.8
16.9
4.2
3.7
4.7
1.9
4.7
6.2
17.9

4.0
7.3
4.2
3.1
3.4
4.4
3.0
6.9
4.2
3.7
4.7
5.7
4.7
4.1
5.5

8.4
5.2
3.8
13.3
14.1
20.7
7.4
20.8
8.7
9.4
8.8
2.9
8.3
11.4
26.7

8.0
8.4
8.3
8.0
12.8
10.7
12.5
8.5
8.7
9.3
8.8
8.4
8.4
7.6
8.2

365
343
251
466
427
137
53

353
332
217
494
437
130
37

8.1
7.3
3.3
6.9
6.1
2.3
0.6

2.3
2.2
1.5
1.4
1.4
1.8
1.6

21.9
21.6
15.8
32.6
28.8
4.8
1.9

6.2
6.5
7.3
6.6
6.6
3.7
5.2

28.6
26.9
17.8
40.5
36.7
10.3
3.1

8.1
8.1
8.2
8.2
8.4
7.9
8.5

243
103
148
173
79
214
179
158
149
165
157
65

213
91
135
157
82
183
183
139
133
154
141
62

4.3
2.0
2.4
3.5
1.5
3.7
3.7
3.1
2.8
2.5
2.3
1.1

2.0
2.2
1.8
2.2
1.8
2.0
2.0
2.2
2.1
1.6
1.6
1.8

10.7
3.7
6.6
9.4
3.4
11.3
11.2
7.4
7.2
8.6
7.8
2.4

5.0
4.1
4.9
6.0
4.1
6.2
6.1
5.3
5.4
5.6
5.5
3.9

18.3
7.6
12.0
12.7
6.6
14.8
11.1
12.6
12.1
14.0
13.1
5.5

8.6
8.4
8.9
8.1
8.0
8.1
6.1
9.1
9.1
9.1
9.3
8.8

a: g/kg dry matter, b: percentage of 16 analysed amino acids (TAA).

Formaldehyde treated

included in the in vivo trials used for the formulation of equations for prediction of uCP
(Lebzien et al., 1996) before.
Incubation
The incubation procedures were based on the in vitro incubation technique of Zhao and
Lebzien (2000). The rumen fluid was taken just before morning feeding and was filtered
through one layer of cheese-cloth. Centrifugation tubes that were about 2 cm in diameter,
6 cm in length and 80 ml in volume were used as the incubation vessels. About 0.5 g of feed
sample was weighed into each centrifugation tube. The tubes were pre-warmed to 38  C
before incubation. The buffers for the incubation solution were made as following:

The in vitro incubation technique for the estimation of uAA

249

Table 3. The utilizable (u) crude protein (CP)-, uAmino acids (AA)-, uMethionine-, uLysine- and
uLeucine-content (g/kg incubated dry matter) of feeds and feed mixtures estimated using in vitro
incubation technique with rumen fluid of a cow (Expt. 1)
uMethionine

Feeds
Barley
Beet pulp
Corn 1
Citrus pulp
Coconut cake
Corn gluten feed
Corn silage 1
Faba bean
Grass meal
Grass silage
Hay 1
Manioc
Oat
Palm kernel cake
Protected rape
seed meal3
Rapeseed meal 1
Rapeseed meal 2
Rapeseed meal 3
Soya bean hulls
Soya bean meal 1
Soya bean meal 2
Soya bean meal 3
Wheat grain
Wheat straw
Feed mixtures
Mix 1
Mix 2
Mix 3
Mix 4
Mix 5
Mix 7

uLysine

uLeucine

Regressive
uCP
calculated
(g/kg) uCP1 (g/kg)

g/kg

(%)2

g/kg

(%)2

g/kg

(%)2

uAA
(g/kg)

3.9
3.2
3.6
2.2
3.5
3.7
1.7
2.6
2.5
2.9
3.0
1.7
2.7
3.7
6.4

2.2
2.1
2.2
2.1
1.8
2.1
2.0
1.6
2.2
2.1
2.2
2.2
2.1
2.1
2.1

14.5
12.7
10.6
9.4
7.4
12.3
5.5
12.4
7.8
9.2
10.4
7.4
9.2
7.6
18.3

8.1
8.2
6.5
8.9
3.9
7.1
6.4
7.5
6.8
6.8
7.6
9.4
7.0
4.4
6.0

14.3
12.9
18.6
8.5
15.2
17.0
8.9
14.4
10.2
12.3
11.3
6.2
10.7
13.2
25.8

8.0
8.3
11.3
8.0
8.0
9.8
10.3
8.7
8.9
9.1
8.2
7.8
8.2
7.6
8.5

179
155
164
106
191
174
86
166
114
135
137
79
131
173
305

164
157
154
136
236
210
105
187
178
165
139
101
164
163
343

162
141
157
142
201
182
130
194
137
143
145
131
140
175
341

5.6
5.1
4.5
3.4
4.5
5.0
5.0
4.0
1.1

2.1
2.2
2.2
1.8
1.8
1.6
1.9
2.3
2.3

18.2
15.4
15.2
14.8
17.9
21.0
19.0
15.2
3.7

6.9
6.7
7.5
7.6
7.1
6.6
7.3
8.6
7.7

23.8
20.6
18.3
16.6
23.0
28.4
23.6
13.8
4.0

9.0
8.9
9.0
8.6
9.2
9.0
9.1
7.8
8.3

263
231
203
194
251
316
259
177
48

230
256
233
207
289
309
270
173
63

264
219
214
184
319
303
308
172
76

5.3
3.2
4.8
3.2
2.2
3.8

2.2
2.2
2.2
2.2
1.9
2.2

16.5
9.0
14.9
13.1
9.0
13.8

6.8
6.1
6.7
9.0
7.6
7.9

20.5
11.9
18.8
12.1
9.5
14.0

8.4
8.1
8.4
8.3
8.1
8.0

244
147
223
146
118
175

258
165
242
161
127
208

227
154
172
183
149
177

Calculated uCP according to: uCP (187:7  115:4 UDP/CP) DOM + 1.03 UDP/DM
Percentage of 16 analysed amino acids
3
Formaldehyde treated
2

Buffer I: 23.5 g Na2 HPO4  12H2 O (per analysis p.a.), 12.5 g NaHCO3 (p.a.) and 11.5 g
NH4 HCO3 (p.a.) were dissolved in 400 ml distilled water. NH4 HCO3 was used as the
nitrogen source for the microbes to replace part of the NaHCO3 in the original buffer
(Tilley and Terry, 1963).
Buffer II: 23.5 g NaCl (p.a.), 28.5 g KCl (p.a.), 6.0 g MgCl2  6H2 O (p.a.) and 2.63 g
CaCl2  2H2 O (p.a.), were dissolved in 1000 ml distilled water.
In a beaker, 400 ml of buffer I, 50 ml of buffer II and an adequate volume of distilled
water were mixed to yield a final volume of 500 ml. Two hundred and fifty ml of mixed
buffer and 1000 ml of distilled water were mixed and pre-warmed to 38  C in a flask and
the mixture was continuously gassed with CO2 and was stirred automatically by a
magnetic stirrer. Then 312.5 ml of rumen fluid was added into the flask. After the buffer
and rumen fluid were mixed well, 50 ml of buffer-rumen fluid mixture was given into each

250

G. Y. Zhao and P. Lebzien

incubation tube and sealed with a rubber stopper fitted with a small glass tube with a
rubber balloon at the top. Four sets of incubation with two duplicates for each feed sample
and two blanks in each set were used. The incubation tubes were kept at 38  C in a water
bath and were shaken occasionally during incubation. After 24 h of incubation, all tubes
were taken out of the water bath. The pH value was immediately measured. All the
incubation residues were freeze-dried for further analysis.
Expt. 2: Incubation with rumen fluid from a sheep
Animal and feeding
An adult sheep (live weight 51 kg), fitted with a large rumen fistula (inner diameter 40 mm),
fed with a mixed ration twice daily in two equal meals, was used as the donor of rumen
fluid. The mixed ration which had a dry matter content of 85.5% contained 700 g grass hay
and 600 g mixed concentrates (65% wheat grain, 29% soya bean meal, 1% soya bean oil
and 5% mineral and vitamin mixture). Crude protein content in the ration DM was about
16.5%. The sheep had free access to drinking water.
Feed samples
Twenty-two air-dry feeds (Table 4) and 11 feed mixtures (Table 1), which were used in
different in vivo experiments before, were ground to pass a 3-mm sieve for incubation and
determination of dry matter.
Incubation
The rumen fluid was taken 7 h after morning feeding. The amount of NH4 HCO3 in buffer
I was reduced to avoid too high levels of ammonia in incubation, and the amount of
NaHCO3 was increased to keep the same level of HCO3 in buffer I as in Expt. 1. The
composition of modified buffer I was: 23.5 g Na2 HPO4  12H2 O (p.a.), 15.0 g NaHCO3
(p.a.) and 9.5 g NH4 HCO3 (p.a.), dissolved in 400 ml distilled water. The other incubation
procedures and buffer II were the same as in Expt. 1.
Chemical analysis
The feed samples were dried for 6 h at 105  C for the determination of dry matter content.
Ammonia N in the incubation residues was analysed by steam distillation and titration
prior to freeze-drying of the residues. Amino acids in the feed samples and the incubation
residues were determined by Degussa (Germany) using an amino acid analyser with cation
exchange resin column and ninhydrin detection according to the EC Directive 98/64/EC of
3 September 1998. Cyst(e)ine and methionine were oxidized to cysteic acid and methionine
sulphone respectively prior to hydrolysis. Tryptophan was not determined. In the
following only values for the most important amino acids methionine, lysine and leucine
will be listed, to reduce the number of tables and figures.
Calculation and statistical analysis
The uAA and uCP were calculated as following:
uAA AAresidue  AAblank =Incubated sample DM

where uAA (g/kg DM) refers to uAA in feed, AAresidue (g) refers to AA content of
incubation residue, AAblank (g) to AA content of incubation residue of the blank and DM
(kg) to dry matter,

The in vitro incubation technique for the estimation of uAA

251

uCP CPresidue  CPammonia  CPblankresidue  CPblankammonia =

Incubated sample DM

where uCP (g/kg DM) refers to uCP in feed, CPresidue (g) refers to crude protein in
incubation residue, CPammonia (g) to crude protein from ammonia in incubation residue,
CPblankresidue (g) refers to crude protein content of residue in blank, CPblankammonia (g)
to crude protein from ammonia in incubation residues of blank and DM (kg) to dry
matter.
The uCP and uAA determined with in vitro incubation technique were compared
statistically with the uCP values calculated by the following multiple regression equation
based on 335 in vivo experiments on fistulated cows (Lebzien et al., 1996):
uCP 187:7  115:4UDP=CPDOM 1:03UDP=DM

r2 0:91, CV (%) 8.9, where uCP (g) refers to utilizable crude protein in feed, UDP (g)
refers to rumen undegradable crude protein and DOM (kg) to digestible organic matter.

Results and discussion

The crude protein content and amino acid composition of feeds and feed mixtures are given
in Table 2.
The uCP- and uAA-content of feeds and feed mixtures
The in vitro determined uCP and uAA as well as utilizable (u) Methionine, uLysine and
uLeucine in the incubation residues per kg incubated DM of Expt. 1 are shown in Table 3.
On an average the in vitro determined uCP values were only 5 g or 2.6% higher than the
values calculated by regression. But seven out of the 30 values differ by more than 30 g or
15%. That is quite high, but the coefficient of variation for the prediction equation (3) used
for the calculation of the uCP content, amounted also to 8.9% (Lebzien et al., 1996) and
the differences can be caused as well by the inherent error of the prediction equation as
independent reference as by errors of the in vitro method.
It was found that there was a close relationship between regressive calculated uCP (X,
g/kg) and in vitro determined uCP (Y, g/kg): Y 0:93X 17:50; r2 0:84; p < 0:001;
n 30, which is quite similar to the regressive relationship reported by Zhao and Lebzien
(2000).
In Fig. 1 it is shown that there was also a close relationship between the regressive
calculated uCP and the in vitro determined uAA. The results point out that the in vitro
incubation technique can be used for the estimation of uAA of feeds.
The results (Table 3) also show that there was a close relationship between the
in vitro determined uCP and uAA r2 0:90; n 30; p < 0:001), between uAA and

Fig. 1. The relationship between regressive calculated utilizable crude protein

(uCP) and in vitro determined utilizable
amino acids (uAA) with rumen fluid of a
cow (Expt. 1).

252

G. Y. Zhao and P. Lebzien

Table 4. The utilizable (u) crude protein (CP)-, uAmino acids (AA)-, uMethionine-, uLysine- and
uLeucine-content (g/kg incubated dry matter) of feed and feed mixtures estimated using in vitro
incubation technique with rumen fluid of a sheep (Expt. 2)
uMethionine

Feeds
Barley
Beet pulp
Citrus pulp
Coconut meal
Corn 2
Corn gluten feed
Corn silage 2
Faba bean
Grass meal
Grass silage
Hay 2
Manioc
Oats
Palm kernel cake
Protected rapeseed meal3
Rapeseed meal 2
Rapeseed meal 3
Soya bean hulls
Soya bean meal 1
Soya bean meal 2
Wheat grain
Wheat straw
Feed mixtures
Mix 1
Mix 2
Mix 3
Mix 4
Mix 5
Mix 6
Mix 8
Mix 9
Mix 10
Mix 11
Mix 12
13

uLysine

uLeucine

Regressive
uCP calculated
(g/kg) uCP1 (g/kg)

g/kg

(%)2

g/kg

(%)2

g/kg

(%)2

uAA
(g/kg)

2.6
2.5
2.7
3.2
2.2
3.0
1.9
3.1
1.9
2.3
2.3
2.3
2.1
3.2
5.8

2.1
2.0
3.1
1.6
1.6
1.9
2.3
1.9
1.9
2.0
1.8
2.2
1.7
2.4
1.9

12.4
11.4
7.5
8.8
7.7
10.5
7.2
14.5
7.4
7.1
8.7
11.0
9.5
6.3
19.0

10.2
9.0
8.5
4.5
5.7
6.6
8.6
8.8
7.3
6.2
6.9
10.7
7.8
4.6
6.4

10.1
11.0
7.6
15.9
16.9
16.5
8.2
14.6
9.2
10.9
11.8
8.6
10.4
10.8
26.0

8.3
8.7
8.6
8.2
12.5
10.4
9.8
8.8
9.0
9.5
9.4
8.3
8.5
7.9
8.7

122
127
88
194
135
159
84
165
102
115
126
103
122
136
299

142
145
117
213
144
172
104
187
127
156
167
135
139
154
307

162
141
142
201
156
182
128
194
137
143
147
131
140
175
341

6.7
4.9
2.7
4.4
4.2
2.9
0.9

2.7
2.8
1.8
1.9
1.8
2.4
1.8

16.5
12.2
11.9
17.7
16.5
12.5
4.4

6.7
7.0
7.8
7.8
7.1
10.2
8.6

21.9
16.3
13.0
21.5
21.0
10.1
4.3

8.9
9.3
8.5
9.5
9.1
8.2
8.4

247
175
153
227
232
123
51

252
183
165
235
242
136
67

219
214
184
319
303
172
76

3.0
2.5
2.9
3.3
2.9
2.4
3.5
4.0
3.2
3.4
2.0

1.7
1.9
2.0
2.3
2.0
1.8
2.9
3.5
2.9
2.4
2.0

13.0
8.5
9.9
12.5
11.3
12.6
9.1
8.7
8.8
9.2
9.6

7.6
6.5
6.7
8.6
7.7
9.3
7.5
7.6
7.9
6.6
9.4

15.2
11.0
13.1
12.5
14.8
11.4
11.4
10.7
10.1
11.0
9.2

8.8
8.5
8.9
8.6
10.1
8.4
9.4
9.3
9.1
7.9
9.0

172
130
148
145
146
135
121
115
111
140
102

199
144
161
158
154
152
127
114
117
130
119

227
154
172
183
149
183
156
155
175
171
144

See Table 3

uMethionine (r2 0:90; p < 0:001, n 30), between uAA and uLysine (r2 0:82;
p < 0:001, n 30) and between uAA and uLeucine (r2 0:96; p < 0:001, n 30). These
results indicate that the proportion of uAA and the proportions of uMethionine,
uLysine and uLeucine in uAA are relatively constant irrespective of the feedstuff
incubated. It may be possible to calculate uAA from uCP and uMethionine, uLysine or
uLeucin content of feeds from uAA.
The results from Expt. 2 are given in Table 4. On an average the differences between the
in vitro determined values and the values calculated by equation (3) are larger (18 g or
11.6%) than in Expt. 1. Possible reasons are discussed later on. Nevertheless there was a
close relationship between the regressive calculated uCP (X, g/kg) and the in vitro
determined uCP (Y, g/kg): Y 0:79X 19:35; r2 0:81; p < 0:001, n 33, which was

The in vitro incubation technique for the estimation of uAA

253

similar to that of Zhao and Lebzien(2000). But although r2 is nearly the same as in Expt. 1,
the relationship differs more from Y X than in Expt. 1 and points to a systematic effect.
The same was true for the relationship between the regressive calculated uCP and the
in vitro determined uAA (Fig. 2).
It was also observed that there is a close relationship between the in vitro determined
uCP and uAA r2 0:96; p < 0:001; n 33), between uAA and uMethionine (r2 0.70,
n 33, p < 0:001), between uAA and uLysine r2 0:70; n 33; p < 0:001) and between
uAA and uLeucine r2 0:93; n 33; p < 0:001.
Comparison of the amino acid pattern of feeds and incubation residues
Statistical analysis indicates that there was a significant regressive relationship between
total amino acids (TAA) of feeds (X, g/kg DM) and TAA of incubation residues (Y, g/kg
incubated DM) in Expt. 2: (Y 0:38X 81:6; r2 0:73; p < 0:001; n 33), but the slope
was only 0.38. This means, that TAA content of the feedstuff is only one factor influencing
uAA content of feedstuffs. This is in accordance with results from in vivo investigations
(Lebzien et al., 1996) showing that uCP content of feedstuff mainly depends on energy
content of the feedstuff if N is not limiting. There were also significant regressive
relationships (p < 0:001) between the methionine, lysine and leucine contents of feeds
(g/kg DM) and the uMethionine, uLysine and uLeucine contents of incubation residues (g/
kg incubated DM; r2 0:76; r2 0:65; r2 0:75).
The mean TAA in incubation residues (mean value 144 g/kg incubated DM) were lower
than the TAA in feeds (mean value 163 g/kg DM). The mean TAA-content in feed, of
course, depends on what kind of feed is used. The variation of TAA in feeds in Table 2
(SD 112 g/kg DM), was about twice that of uAA (Tables 3 and 4; SD 59 g/kg incubated
DM), indicating that the TAA of incubation residues of different feeds are more constant
than the TAA of feeds. The reason for the similarity of TAA of incubation residues is, that
the AA content and the AA pattern of incubation residues mainly depend on the AA
content and the AA pattern of microbial protein as the percentage of microbial protein in
the incubation residues is higher than the percentage of the undegraded feed protein.
Comparison of the amino acid pattern of incubation residues, duodenal protein
and rumen microbial protein
Lebzien (1997) summarized the data on the amino acid pattern of the protein at the
duodenum of ruminants and reported that the ranges of methionine, lysine and leucine

Fig. 2. The relationship between

regressive calculated utilizable
crude protein (uCP) and in vitro
determined utilizable amino acids
(uAA) with rumen fluid of a sheep
(Expt. 2)

254

G. Y. Zhao and P. Lebzien

percentages of duodenal protein (consisting of undegraded dietary protein and rumen

microbial protein) are 0.63.0, 4.79.5 and 6.811.9%, respectively. Clark et al. (1992)
reported that the ranges of methionine, lysine and leucine percentages of rumen microbial
protein are 1.14.9, 4.99.5 and 5.39.7% respectively. The means and ranges for
percentages of methionine, lysine and leucine in TAA of incubation residues (Tables 3 and
4), microbial protein and duodenal digesta protein are compared in Table 5. The results
indicate that the contents of methionine, lysine and leucine in TAA of incubation residues
are within or quite close to the ranges of methionine, lysine and leucine of duodenal
protein from in vivo experiments and the ranges of rumen microbial protein. Although
TAA and methionine, lysine and leucine of feeds had significant influence on uAA,
uMethionine, uLysine and uLeucine contents, in accordance with the conclusions of
Lebzien (1997), it was impossible to calculate the AA pattern of uAA from the AA pattern
of feeds as the degradation and synthesis of AA in rumen fermentation and in in vitro
incubation are influenced by so many factors.
The results imply that the in vitro incubation conditions in the present experiments are
quite similar to the average rumen fermentation with respect to the amino acid composition
of duodenal protein and the rumen microbial protein, indicating that the in vitro
incubation technique could be a valid method for the estimation of uAA, utilizable
Methionine, uLysine and uLeucine of feeds for ruminants.
Comparison of amino acid composition of incubation residues from Expts. 1 and 2
For 27 identical feeds uAA- and uCP-content were determined using as well rumen fluid
from a cow as rumen fluid from a sheep. Statistical analysis indicates that there was a
significant regression relationship between uAA content of these feeds determined with
rumen fluid of a sheep and that with rumen fluid of a cow (Fig. 3).
There was also a significant regression relationship between uCP of feeds determined
with rumen fluid of a sheep and that with rumen fluid of a cow (Fig. 4). This was also valid
for the estimated amounts of uMethionine, uLysine or uLeucine per kg incubated feed DM
determined with rumen fluid of a sheep and those with rumen fluid of a cow respectively
(uMethionine: r2 0:59; p < 0:001; uLysine: r2 0:74; p < 0:001 and uLeucine: r2
0:79; p < 0:001). These results indicate that the in vitro incubation with rumen fluid of a
cow fed only with hay and with rumen fluid of a sheep fed with the mixed ration were
Table 5. Comparison of means and ranges for percentages (g/100 g amino acids1 ) of methionine,
lysine and leucine in rumen micobial protein (CLARK et al., 1992), duodenal digesta protein
(LEBZIEN, 1997), in incubated feed protein and in in vitro incubation residual protein in the present
experiments
In vitro incubation
residual protein

Feed
protein

Microbial
protein

Duodenal
digesta protein

Expt. 1

Expt. 2

Methionine
Mean
Range

1.9
0.92.4

2.6
1.14.9

2.0
0.63.0

2.1
1.62.3

2.2
1.63.5

Lysine
Mean
Range

5.1
3.17.3

7.9
4.99.5

6.7
4.79.5

7.2
3.99.4

7.6
4.510.7

Leucine
Mean
Range

8.8
7.612.8

8.1
5.39.7

9.3
6.811.9

8.6
7.611.3

9.0
7.912.5

Sixteen analysed amino acids

The in vitro incubation technique for the estimation of uAA

255

Fig. 3. The relationship between

utilizable amino acids (uAA)
determined with rumen fluid of a
cow and uAA determined with
rumen fluid of a sheep

Fig. 4. The relationship between

utilizable crude protein (uCP)
determined with rumen fluid of a
sheep and uCP determined with
rumen fluid of a cow

correlated, but it has to be stated that the contents of uAA and uCP of the feeds determined
with the rumen fluid of the hay-fed cow were significantly higher than those determined
with rumen fluid of the sheep fed with the mixed ration (p < 0:01). The same was true for
the contents of uMethionine, uLysine and uLeucine determined with rumen fluid of the
cow and those determined with that of the sheep. But only the difference of the contents of
uMethionine or uLeucine reaches the significance level (p < 0:05). These results indicate
that the rumen microorganisms and the rumen fermentation in the cow and the sheep were
different to some extent. Additionally to differences between individuals, one reason for
the lower uAA with rumen fluid from the sheep could be that the sheep was fed with a
mixed ration causing a more complex microbial population. The lag time, therefore, could
have been shorter and the incubation time of 24 h too long. In this case, more of the
synthesized microbial protein might have been degraded and therefore the microbial netsynthesis reduced. Shorter incubation times and differences between individuals should
therefore be tested.

Conclusion
Using the in vitro incubation technique of Zhao and Lebzien(2000), it has been found
that there is a close relationship between the regressive calculated uCP and the in vitro
determined uAA both in Expts. 1 and 2. It has also been found that the pattern of
methionine, lysine and leucine of TAA in incubation residues are within or close to the
ranges of the duodenal protein and the ranges of the rumen microbial protein from
in vivo experiments, indicating that the in vitro incubations conducted in Expts. 1 and 2

256

G. Y. Zhao and P. Lebzien

were quite similar to rumen fermentation with respect to mean amino acid composition
of duodenal protein and of rumen microbial protein. These results indicate that the
in vitro incubation technique of Zhao and Lebzien (2000) can principally be used for
estimating uAA, utilizable Methionine, uLysine and uLeucine of feeds for ruminants. It
should be proved if larger differences between the regressive calculated and the in vitro
determined values for some feedstuffs are caused mainly by the inherent error of the
uCP-regression as independent reference or mainly by the error of the in vitro method.
The TAA and methionine, lysine and leucine of feeds had a significant but not decisive
influence on the uAA, uMethionine, uLysine and uLeucine contents and it was not
possible to calculate the AA pattern of incubation residues from the AA pattern of feeds
as the degradation and synthesis of AA in rumen fermentation or in in vitro incubation
are influenced by many factors. As a rumen-fistulated dairy cow is more expensive than a
rumen-fistulated sheep and most laboratories do not have fistulated cows, it is suggested
to use a rumen-fistulated sheep fed with a mixed ration as prescribed for the Hohenheim
gas-test as the donor of rumen fluid for the estimation of uAA of feeds with in vitro
incubation, but that in this case 24 h of incubation seemed to be too long and shorter
incubation periods should be tested. Further standardization of the method should be
done.

References
Clark, J. H.; Klusmeyer, T. H.; Cameron, M. R., 1992: Microbial protein synthesis and flows of
nitrogen fractions to the duodenum of dairy cows. J. Dairy Sci. 75, 23042323.
GfE (Gesellschaft fur Ernahrungsphysiologie), 1997: Zum Proteinbedarf von Milchkuhen und
Aufzuchtrindern. Proc. Soc. Nutr. Physiol. 6, 217236.
Lebzien, P., 1997: Zum Einfluss des Futterproteins auf das Aminosaurenmuster des Proteins am
bers. Tierernahrg. 25, 137153.
Duodenum von Wiederkauern. U
Lebzien, P.; Voigt, J.; Gabel, M.; Gadeken, D., 1996: Zur Schatzung der Menge an nutzbarem
Rohprotein am Duodenum von Milchkuhen. J. Anim. Physiol. a. Anim. Nutr. 76, 218223.
Naumann, C.; Bassler, R., 2001. Die chemische Untersuchung von Futtermitteln. Band III,
Methodenbuch VDLUFA, VDLUFA, Darmstadt, im Druck.
Tilley, J. M. A.; Terry, R. A., 1963: A two-stage technique for the in vitro digestion of forage crops.
J. Brit. Grassl. Soc. 40, 139155.
Zhao, G.Y.; Lebzien, P., 2000: Development of an in vitro incubation technique for the estimation of
the utilizable crude protein (uCP) in feeds for cattle. Arch. Anim. Nutr. 53, 293302.
Authors address: Dr P. Lebzien, Institute of Animal Nutrition, Federal Agricultural Research Centre
Braunschweig, Bundesallee 50, D-38116 Braunschweig, Germany. Tel.: 0531 596
3140; Fax: 0531 596 3199; E-mail: peter.lebzien@fal.de