Effects of Dietary Fat Type and Xylanase Supplementation to Rye-Based

Broiler Diets on Selected Bacterial Groups Adhering to the Intestinal

Epithelium, on Transit Time of Feed, and on Nutrient Digestibility

S. DÄNICKE,*,1 W. VAHJEN,† O. SIMON,† and H. JEROCH*

*Institut für Tierernährung und Vorratshaltung, Landwirtschaftliche Fakultät, Martin-Luther-Universität Halle-Wittenberg,
06108 Halle (Saale) Emil-Abderhalden-Straße 26, Germany and †Institut für Tierernährung,
Freie Universität Berlin, 14195 Berlin, BruAAaummerstraße 34, Germany

ABSTRACT Two experiments were conducted to ex-
amine the effects of different fat types, i.e., soybean oil
(S) and beef tallow (T), in rye-based broiler diets, either
unsupplemented (−) or supplemented (+) with xylanase
(Avizyme 1300 at 1 g/kg diet), on selected bacterial
groups adhering to the epithelium of crop, duodenum,
jejunum, and ileum (Experiment 1, 16 d of age), on mean
retention time (MRT) of digesta, and on digestibility of
N and dry matter in successive segments of the digestive
tract (Experiment 2, 24 d of age). Live weight of enzyme-
treated and S-fed chickens was significantly higher than
that for unsupplemented or T-fed birds, respectively, in
both experiments.
In Experiment 1, a reduction in bacterial colonization
from crop to duodenum was followed by a continuous
(Key words: broiler, dietary fat type, xylanase, intestinal bacteria, feed transit)
increase as far as the ileum. Xylanase supplementation
significantly reduced enterobacteria and total anaerobe
microbes with a similar trend for Gram-positive cocci
and enterococci. The latter two groups were significantly
enhanced in birds fed T. In Experiment 2, xylanase sup-
plementation resulted in a decrease in MRT in several
segments of the digestive tract. This effect was most pro-
nounced in the small intestine, where MRT of 268, 217,
241, and 209 min in groups S−, S+, T−, and T+, respec-
tively, were measured. Apparent digestibility of N and
dry matter was slightly improved by xylanase supple-
mentation in the jejunum and ileum. Nitrogen digestibil-
ity by the terminal ileum was 80.3, 83.7, 79.4, and 82.2%
for the S−, S+, T−, and T+ groups, respectively, and dry
matter digestibility amounted to 61.2, 65.5, 62.1, and
64.0%, respectively.
1999 Poultry Science 78:1292–1299

INTRODUCTION 1971; Misir and Marquardt, 1978a,b; Day and Thomas,

Increased intestinal viscosity was shown to be the most
important physiological impact when rye diets are fed to
broiler chickens (Bedford and Classen, 1992). It has been
proved that this increased viscosity is closely related to
soluble pentosans in rye and that their in vivo gel-forming
properties can be greatly diminished by supplementation
of the diet with endoxylanases (Bedford and Classen,
1992). Furthermore, it has been reported that enzyme
effects on crude fat and fatty acid digestibility were more
pronounced when the source of fat was beef tallow com-
pared with soybean oil (Dänicke et al., 1997b). On the
other hand, it has clearly been demonstrated that antibi-
otic supplementation of rye broiler diets improved perfor-
mance of the birds markedly (MacAuliffe and McGinnis,
Received for publication November 10, 1998.
Accepted for publication April 15, 1999.
1
To whom correspondence should be addressed: Sven Dänicke, Insti-
tute of Animal Nutrition, Federal Agricultural Research Centre,
Braunschweig (FAL), Bundesallee 50, D-28 116 Braunschweig, Germany;
sven.daenicke@fal.de
1980) and changed the pattern of both luminal and adher-
ing flora as well as microbial activity (Wagner and
Thomas, 1978; Untawale and McGinnis, 1979). It was
therefore suggested that the antinutritive effects of pento-
sans might be mediated through the gut microflora, and
that a longer transit time of digesta caused by increased
viscosity could increase microbial activity in the chicken
(Choct and Annison, 1992; Choct et al., 1992). Hence, the
objective of the present study was to examine selected
groups of microbes adhering to epithelium and mucus
of the small intestine (Experiment 1) and to determine
the mean retention time (MRT) and the apparent digest-
ibility of dry matter and N in different segments of the
digestive tract of the chicken (Experiment 2).
MATERIALS AND METHODS
Experimental Design
Diets tested in these experiments were the most ex-
treme diets of an experimental design described by Dän-
Abbreviation Key: MRT = mean retention time; S = soybean oil; T
= beef tallow; − = without enzyme addition; + = with enzyme addition.

1292
 EFFECTS OF FAT TYPE AND XYLANASE ADDITION IN BROILERS 1293
TABLE 1. Composition of the experimental diets

Group
Ingredients S− S+ T− T+
(%)
Rye1 56.0 56.0 56.0 56.0
Soybean protein isolate 18.8 18.8 18.8 18.8
DL-methionine 0.43 0.43 0.43 0.43
L-lysine-HCl 0.13 0.13 0.13 0.13
DL-threonine 0.25 0.25 0.25 0.25
Soybean oil 10.0 10.0 ... ...
Cellulose 5.0 5.0 0.89 0.89
Tallow ... ... 10.0 10.0
Maize starch 3.5 3.5 8.7 8.7
Sand 1.09 1.09 ... ...
Limestone 0.5 0.5 0.5 0.5
Dicalcium-phosphate 2.95 2.95 2.95 2.95
Na-chloride 0.35 0.35 0.35 0.35
Enzyme preparation2 ... 0.1 ... 0.1
Premix3 1.0 1.0 1.0 1.0
Calculated analysis
Crude protein 23.0
Crude fat 11.0
ME, kcal/kg 3,180
Lysine 1.25
Met 0.7
Met + Cys 0.94
Threonine 0.99
Tryptophan 0.27
Ca 0.9
P 0.7
Na 0.14
1
Variety “Marder”, insoluble pentosans, 66.2 g/kg dry matter; soluble pentosans, 35.5 g/kg dry matter.
2
Avizyme 1300 (Finnfeeds International Ltd., Marlborough, Wiltshire, UK SN8 1XN): 2,700 IU xylanase (pH
5.3) from Trichoderma longibrachiatum, β-glucanase (traces), 800 IU protease (pH 7.0) from Bacillus subtilis.
3
Vitamin-mineral premix provided per kilogram of diet: Fe, 60 mg; Cu, 5 mg; Zn, 51.4 mg; Mn, 60.8 mg; Se,
0.2 mg; I, 0.6 mg; vitamin A, 12,000 IU; cholecalciferol, 3,000 IU; vitamin E, 49 IU; vitamin B1, 2.1 mg; vitamin
B2, 6.6 mg; vitamin B6, 4.1 mg; vitamin B12, 20.7 µg; pantothenic acid, 15 mg; nicotinic acid, 36 mg; folic acid,
1 mg; biotin, 102 mg; choline chloride, 700 mg; ethoxyquin, 120 mg; Zn-bacitracin, 50 mg.

icke et al. (1999) in which a dose response study with
respect to dietary soluble pentosan level was conducted.
Wheat and rye were used as the sole plant fiber and
pentosan source. These cereals were blended in such a
way as to give four basal diets with increasing pentosan
contents. To achieve this aim, wheat was replaced by rye
as a percentage of total dietary cereals, at a rate of 0, 33, 66,
and 100%. This substitution corresponded to calculated
dietary soluble pentosans of 7.7, 11.0, 14.3, and 17.6 g/
kg, respectively. Each of these diets contained either soy-
bean oil (S) or beef tallow (T) as the dietary fat type.
Finally, each diet was tested without (−) or with (+) xyla-
nase supplementation. The xylanase preparation was ob-
tained from Trichoderma longibrachiatum and contained
3,000 IU/g, measured at pH 6.0 on the basis of reducing
substance formation. β-Glucanase activity and cellulase
activity measured under the same conditions were very
low (40 and 14 IU/g, respectively). The inclusion rate of
the enzyme preparation was 1 g/kg. The four rye-diets
used in the present experiments represented the highest
concentrations of dietary soluble pentosans (Table 1).
Animal and Laboratory Procedures
Birds used for the present investigations were selected
from the mentioned experiment (Dänicke et al., 1999) at
7 d of age (Experiment 1) and at 14 d of age (Experiment
2). The average body weight of their respective groups
were 125, 135, 123, and 133 g for S−, S+, T−, and T+ groups,
respectively, at 7 d of age and 286, 340, 243, and 303
g, respectively, at 14 d of age. A total of 12 chicks and 16
chicks were used in Experiment 1 and 2, respectively
(three and four replications per group, respectively). Birds
were placed into single metabolic cages and the pelleted
experimental diets were offered for ad libitum con-
sumption.
Experiment 1
The procedure was performed when the chicks were 16
d of age. Preparation of tissues for determining adhering
microbes is described by Vahjen et al. (1998) in detail.
Individual broilers were consecutively processed after
feed withdrawal of 2 h such that no more than 4 h passed
between extraction of tissue samples and incubation of
samples on the media. Birds were killed and the digestive
tract was quickly dissected. Crop, duodenum (pylorus to
entrance of bile ducts), jejunum (bile duct entrance to
Meckel’s diverticulum), and ileum (Meckel’s diverticu-
lum to the ileo-cecal junction) were ligated and subse-
quently separated. Dilutions from extracted and washed
tissue samples were applied onto selective media agar
1294 DÄNICKE ET AL.

plates in 20 µL droplets. All media were incubated for
48 h at 37 C.
McKonkey media, esculin/bile acid media, and total
anaerobe media2 were used to isolate enterobacteria,
Gram-positive cocci, and total anaerobes, respectively.
Total colony-forming units on the respective media were
counted using a stereo microscope. Evaluation of colony-
forming units were carried out according to Vahjen et
al. (1998).
where yijk = tested parameter of a broiler k fed a diet
containing fat type i and xylanase level j; ai (fixed effect)
= fat type (S, soybean oil, T, tallow); bj (fixed effect) =
xylanase supplementation (−, without, +, with xylanase);
ai × bj = interactions between fat type and xylanase supple-
mentation; eijk = error term.
Digestibility of nitrogen and dry matter and MRT were
analyzed by a three-way ANOVA with repeated mea-
surements:

Experiment 2 yijkl = µ + ai + bj + ck + (a × b)ij + (a × c)ik +

(b × c)jk + (a × b × c)ijk + dl(a × b) + eijkl
Diets contained TiO2 as an inert marker. It was assumed
that the birds were in a steady state at 23 d of age with where yijkl = tested parameter of a broiler l fed a diet
respect to intake and excretion of TiO2 after consuming containing fat type i and xylanase level j in intestinal
the marker containing diets for 5 d. The quantity of feed
segment ki (fixed effect) = fat type (S, soybean oil, T,
eaten within 24 h was recorded from 23 to 24 d of age.
tallow); bj (fixed effect) = xylanase supplementation (−,
Birds that consumed feed ad libitum were weighed exactly
without, +, with xylanase); ck (fixed effect) = intestinal
24 h after starting to record their feed intake and the
segment (C, crop, G, proventriculus and gizzard, D, duo-
quantity of consumed feed was determined. After live
denum, J, jejunum, I, ileum, R, rectum, Cc, ceca); (a × b)ij
weight was recorded, chicks were killed and the digestive
= interactions between fat type and xylanase supplemen-
tract was quickly dissected and segments described above
tation; (a × c)ik = interaction between fat type and intesti-
were separated. Digesta of the respective intestinal seg-
nal segment; (b × c)jk = interaction between xylanase
ments was squeezed into preweighed test tubes and
supplementation and intestinal segment; (a × b × c)ijk =
weight was recorded, frozen, and stored for further analy-
interaction between fat type, xylanase supplementation
sis at −20 C. Digesta samples were freeze-dried and finely
and intestinal segment; dl(a × b) = effect of repeated mea-
ground to pass through a 0.2-mm sieve using a freezer
surements (different segments) within the same bird l;
mill (Model 6700-230).3 Dry matter and N were analyzed
eijkl = error term.
by standard methods of the Verband Deutscher Landw-
Significant differences between the means were de-
irtschaftlicher Untersuchungsund Forschungsanstalten
tected by the Tukey significant difference (HSD) test. All
(VDLUFA) as described by Naumann and Bassler (1993);
statistics were carried out using the Statistica for the Win-
TiO2 was determined according to Brandt and Allam
dows operating system (Statsoft, 1994).
(1987).
Digestibility of dry matter and N was calculated as
follows (Krawielitzki et al., 1987): RESULTS

Digestibility (%) = Experiment 1

(
1–
TiO2 in diet
TiO2 in digesta
ⴢ
Nutrient in digesta
Nutrient in diet
ⴢ100.) Live weight of broilers at Day 16 is shown in Table 2.
Birds fed S diets were significantly heavier than birds
fed T diets. Live weight was significantly increased by
Mean retention time (MRT) can be calculated from the
xylanase supplementation to the T diet. No interactions
ratio between TiO2 present in the respective segments of
were observed between fat type and enzyme supplemen-
the digestive tract and the daily TiO2 intake. Calculation
tation.
can be formulated as follows (Persson and Svensson, 1960,
Van Der Klis et al., 1990):
TABLE 2. Live weight of the chickens, 16 d of age, Experiment 1
TiO2 in segment (mg)
MRT (min) = 1,440 min/dⴢ
TiO2 intake (mg/d) Fat Xylanase Liver weight
(g)
Statistical Analysis Soybean oil Without 259.0ab
3,000 IU/kg diet 295.0a
Two-way factorial ANOVA were applied for live
Tallow Without 197.7c
weight and feed intake: 3,000 IU/kg diet 241.0b
Probability
Yijk = µ + ai + bj + (a × b)ij + eijk Fat 0.0001
Enzyme 0.0014
Fat × enzyme 0.6701
Pooled SEM 8.3
2
Merck, Darmstadt, Germany 64271. a–c
Values within columns with no common superscript differ (P <
3
SPEX CertiPrep, Inc., Metuchen, NJ 08840. 0.05).
 EFFECTS OF FAT TYPE AND XYLANASE ADDITION IN BROILERS 1295
TABLE 3. Tissue associated bacterial counts in different segments of digestive tract of chickens (16 d of
age) in dependence on dietary fat type and xylanase supplementation, Experiment 1

Gram
Intestinal Entero- postive Entero- Total
segment Fat Xylanase bacteria cocci cocci anaerobes
(cfu/log10)
Crop Soybean oil Without 6.58a 5.04b 4.18 5.89abab
3,000 IU/kg diet 5.38b 5.59ab 4.46 6.02a
Tallow Without 5.53b 6.22a 4.00 6.76a
3,000 IU/kg diet 5.94ab 5.14b 4.28 4.82b
Duodenum Soybean oil Without 6.26b 3.96 3.16b 5.74
3,000 IU/kg diet 3.54c 4.26 3.88a 4.29
Tallow Without 4.96a 5.01 4.09a 5.48
3,000 IU/kg diet 5.19a 4.72 3.08b 4.66
Jejunum Soybean oil Without 6.35a 4.62b 3.43 6.20
3,000 IU/kg diet 4.42b 4.06c 2.95 4.72
Tallow Without 5.73a 5.26a 4.01 5.35
3,000 IU/kg diet 5.71a 4.82b 3.88 5.30
Ileum Soybean oil Without 6.13a 4.25c 3.80b 6.50
3,000 IU/kg diet 6.13a 4.85ab 4.29ab 5.97
Tallow Without 6.49a 6.35a 5.90a 6.49
3,000 IU/kg diet 5.03b 5.06b 3.58b 5.52
Probability
Fat 0.7967 0.0000 0.0555 0.3749
Enzyme 0.0000 0.0147 0.1083 0.0001
Intestinal segment 0.0000 0.0000 0.0002 0.0169
Fat × enzyme 0.0003 0.0005 0.0080 0.6742
Fat × intestinal segment 0.1243 0.0849 0.0603 0.9765
Enzyme × intestinal segment 0.0869 0.3808 0.0410 0.9328
Fat × enzyme × intestinal segment 0.0000 0.0067 0.0012 0.0805
Pooled SEM 0.29 0.25 0.38 0.52
a–c
Values within columns with no common superscript within intestinal segment differ (P < 0.05).

Tissue-associated colony-forming units in crop, duode-
num, jejunum, and ileum dependence on fat type and
xylanase supplementation are summarized in Table 3.
Significant xylanase effects were observed for the genera
of Enterobacteriaceae (P < 0.001), for total anaerobes (P
< 0.001), and for Gram-positive cocci. Feeding T influ-
enced the counts of Gram-positive cocci significantly (P
< 0.001). Similar changes in the counts of all examined
bacterial groups were observed when the direction from
crop to ileum was considered (significant effect of intesti-
nal segment). There were high concentrations of colony-
forming units adhering to the crop epithelium, a marked
decrease from crop to duodenum, which was followed
by a continuous increase up to the ileum. Significant inter-
actions between fat type and enzyme addition were ob-
served for Enterobacteriaceae, Gram-positive cocci, and
enterococci (P < 0.001, P < 0.001, and P < 0.01, respec-
tively). Significant three-way interactions between fat
type, xylanase supplementation, and intestinal segment
were detected for enterobacteriaceae, Gram-positive
cocci, and enterococci (P < 0.01). These interactions indi-
cate that fat type and xylanase addition interacted in a
different way in different intestinal segments to some
extent and should be considered in interpreting of the
main effects.
Experiment 2
Feed intake during the 24-h experimental period as
well as live weight of the birds at the end of this period
are shown in Table 4. Birds fed the T-diet consumed
slightly less feed. The respective live weights reflected
the tendencies observed for feed intake, but in this case,
the fat and enzyme effects were significant (P < 0.05).
Again, T-birds had the lowest live weight, which resulted
in the significant interaction between fat type and enzyme
supplementation (P < 0.001). The MRT in the different
segments, as shown in Table 5, were summed and the
resulting MRT in the whole digestive tract is also shown
in Table 4. No significant effects were observed. However,
MRT for enzyme-supplemented groups were more than
1 h shorter than in the unsupplemented groups. Observ-
ing the mean values of MRT over fat type and intestinal
segments revealed a significant reduction after xylanase
supplementation. Indeed, enzyme addition resulted in a
reduction of MRT in several segments of the digestive
tract irrespective of fat type (Table 5, Figure 1), especially
in the small intestine (jejunum and ileum), whereas in
other segments no consistent tendencies were observed.
Reduction in MRT in jejunum and ileum was paralleled
by small increases in dry matter and N digestibility. The
pattern of MRT and digestibility descending from crop
to rectum was the same in all groups, but was modified
by dietary treatments (Figures 1 to 3). Digestibility values
declined from crop to duodenum and were found to be
negative in proventriculus plus gizzard and duodenum,
which indicated high rates of secretion into these seg-
ments. This negative digestibility was followed by contin-
uous positive absorption in the jejunum and ileum.
Changes in digestibilities from ileum to rectum were only
1296 DÄNICKE ET AL.
TABLE 4. Feed intake and live weight at day of experiment and mean retention time of digesta in the
entire digestive tract of chickens (24 d of age, Experiment 2)

Mean
Feed Live retention
Fat Xylanase intake weight time
(g/d) (g) (h)
Soybean oil Without 91.3 728a 8.3
3,000 IU/kg diet 90.1 686a 6.3
Tallow Without 78.0 577b 6.9
3,000 IU/kg diet 90.8 714a 6.4
Probability
Fat 0.1004 0.0108 0.3540
Enzyme 0.1272 0.0373 0.0732
Fat × enzyme 0.0713 0.0009 0.2655
Pooled SEM 3.5 20 0.7
a,b
Values within columns with no common superscript differ (P < 0.05).

TABLE 5. Mean retention time of digesta and apparent digestibility of nitrogen and dry matter in
consecutive segments of the digestive tract of chickens (24 d of age, Experiment 2)

Apparent digestibility
Intestinal Mean retention
segment Fat Xylanase time N Dry matter
(min) (%)
Crop Soybean oil Without 60 −2.2 −9.0
3,000 IU/kg diet 56 −8.2 −10.3
Tallow Without 41 −10.3 −13.4
3,000 IU/kg diet 49 −2.7 −11.6
Proventriculus
+ gizzard Soybean oil Without 58 −24.9 −37.0
3,000 IU/kg diet 18 −54.6 −68.9
Tallow Without 49 −13.6 −68.9
3,000 IU/kg diet 52 −7.1 −27.9
Duodenum Soybean oil Without 18 −112.9 −21.8
3,000 IU/kg diet 11 −92.2 −25.1
Tallow Without 13 −131.8 −31.4
3,000 IU/kg diet 8 −134.2 −22.9
Jejunum Soybean oil Without 110 47.2 45.3
3,000 IU/kg diet 102 57.5 50.8
Tallow Without 99 42.5 42.1
3,000 IU/kg diet 84 44.2 44.7
Ileum Soybean oil Without 140 80.3 61.2
3,000 IU/kg diet 104 83.7 65.5
Tallow Without 129 79.4 62.1
3,000 IU/kg diet 117 82.2 64.0
Rectum Soybean oil Without 36 77.9 68.6
3,000 IU/kg diet 27 80.0 65.1
Tallow Without 34 80.2 68.2
3,000 IU/kg diet 23 80.8 67.5
Ceca Soybean oil Without 77 68.5 67.8
3,000 IU/kg diet 59 54.7 54.9
Tallow Without 50 63.2 66.9
3,000 IU/kg diet 63 72.9 70.2
Probability
Fat 0.4245 0.8350 0.9397
Enzyme 0.0914 0.7539 0.6111
Intestinal segment 0.0000 0.0000 0.0000
Fat × enzyme 0.2035 0.3653 0.0055
Fat × intestinal segment 0.6192 0.0075 0.8706
Enzyme × intestinal segment 0.6844 0.8481 0.9561
Fat × enzyme × intestinal segment 0.6439 0.4195 0.0026
Pooled SEM 12 9.8 6.8
 EFFECTS OF FAT TYPE AND XYLANASE ADDITION IN BROILERS
FIGURE 1. Effect of fat type and xylanase supplementation on mean
retention time in successive segments of the gastrointestinal tract of
male chickens (24 d of age, Experiment 2, average of four birds per
treatment ± standard error). Bars from left to right in each group: crop,
proventriculus plus gizzard, duodenum, jejunum, ileum, and rectum.
slight, which suggests that digestion and absorption were
terminated in the ileum.
DISCUSSION
Birds fed the S or + diets were significantly heavier
than animals fed the T or − diets in both experiments.
Xylanase effects were more pronounced in birds fed Tin
Experiment 2, which is in accordance with earlier observa-
tions (Dänicke et al., 1997b). Detection of microbes adher-
ing to epithelia and mucus of different segments of the
digestive tract of chicken is in general agreement with
literature findings (Fuller, 1973, 1984; Vahjen et al., 1998).
Segment-dependent changes in composition of intestinal
microflora have been reported (Huhtanen and Pensack,
1965; Watkins and Kratzer, 1983; Vahjen et al., 1998).
Enzyme supplementation significantly depressed total
anaerobe colony-forming units of tissue-associated bacte-
FIGURE 2. Effect of fat type and xylanase supplementation on appar-
ent nitrogen digestibility in successive segments of the gastrointestinal
tract of male chickens (24 d of age, Experiment 2, average of four birds
per treatment ± standard error). Bars from left to right in each group:
crop, proventriculus plus gizzard, duodenum, jejunum, ileum, and
rectum.
1297
ria, with especially large differences in duodenal and jeju-
nal samples from birds fed S diets. The same effect could
be observed for Enterobacteria. Similar observations were
made by Vahjen et al. (1998) who fed a wheat-based diet,
either unsupplemented or xylanase-supplemented, to
broiler chickens from 1 to 28 d of age and recorded age
and xylanase-dependent changes in luminal and tissue-
associated bacteria.
Growth of tissue associated bacteria is more or less
independent of digesta flow rates; thus, surface or sub-
strate factors determine the amount of bacteria in the
mucosal layer of the intestine.
Higher digesta viscosity changes the morphology of
host intestinal tissues (Johnson and Gee 1986), which re-
sults in prolonged epithelial villi and higher turnover of
epithelial cells in rats (Gee et al. 1996). Digesta viscosities
from animals receiving the same diets as in this experi-
ment were measured to be 99, 30, 220, and 96 mPas in
birds fed S−, S+, T−, and T+, respectively, in jejunal digesta
supernatant and 464, 140, 810, and 264 mPas in ileal di-
gesta supernatant, respectively (Dänicke et al., unpub-
lished data). Thus, viscosity reduction in birds fed
enzyme-supplemented diets may have induced modifi-
cations in intestinal tissue morphology leading to reduced
colonization surface, altered adhesion structures, and a
increased luminal release of epithelial cell (and adhering
bacteria) due to a stimulated proliferation rate of mucosal
cells. The latter was reported for chickens (Rolls et al.,
1978) and rats (Southon et al., 1985) when diets contained
elevated amounts of nonstarch polysaccharides, which
potentially increase digesta viscosity. The observed ef-
fects on microbial colonization were most pronounced in
duodenal and jejunal samples with short digesta retention
times and, thus, low steady state concentrations of soluble
arabinoxylans. An overall lower steady state nutrient con-
centration due to higher resorption rates, lower endoge-
nous loss, or composition change in mucus and shed
epithelial cells, would lead to growth depression for bac-
FIGURE 3. Effect of fat type and xylanase supplementation on appar-
ent dry matter digestibility in successive segments of the gastrointestinal
tract of male chickens (24 d of age, Experiment 2, average of four birds
per treatment ± standard error). Bars from left to right in each group:
crop, proventriculus plus gizzard, duodenum, jejunum, ileum, and
rectum.
1298 DÄNICKE ET AL.
teria, which depend on more complex nutrient composi-
tions as provided by the host.
The response of cocci on enzyme supplementation
seemed less drastic, which indicates that this group of
bacteria is not influenced by morphological changes or
nutrient composition as mediated by enzyme treatment.
On the other hand, T feed resulted in an increased number
of cocci in the jejunum and ileum, but seemed to depress
growth of total anaerobic bacteria and enterobacteria. It
is well known from ruminant nutrition that higher con-
centrations of unsaturated fatty acids (mainly oleic acid,
linoleic acid, and linolenic acid) in the diet change the
microbial composition and activity in the rumen mark-
edly (for reviews see Jilg et al., 1988; Demeyer and Van
Nevel, 1995). Significant higher pH values were measured
in most intestinal segments when T was used as dietary
fat instead of S (Dänicke et al., 1997a). Thus, fat type clearly
influences the microbial composition in the intestine, and
in this experiment, growth of cocci indicated changed
microflora due to different fats. It is not known whether
cocci growth was actively stimulated by T or whether it
was merely a result of an inhibition of other bacteria,
possibly due to increased secretion of bile acids by the
host. Bile acids have a bacteriostatic effect, with most
members of the intestinal microflora being resistant, but
with differences in toleration of bile acid concentrations.
Cocci can be distinguished by their ability to grow in the
presence of different bile salt concentrations and, thus,
more bile acid resistant cocci, such as enterococci, may
become dominant under higher bile acid concentrations.
Active deconjugation of conjugated bile acids, as carried
out by some lactobacilli, also occurs in the intestinal tract,
but it is unknown whether the generated bile salt hy-
drolase activity is sufficient to quantitatively interfere
with the bile salt metabolism.
The antibiotic Zn-bacitracin was included in all diets.
The polypeptide bacitracin has a narrow spectrum of ac-
tivity against Gram-positive bacteria such as clostridia,
certain staphylococci, and streptococci. Thus, total anaer-
obic and cocci colony-forming units are biased by selec-
tion of bacitracin-resistant bacteria in the intestinal tract.
Nevertheless, as all diets contained the antibiotic, relative
changes between trial groups are still valid, although
influences on antibiotic abundance in different viscosity
milieus or proteolytic breakdown remain unknown.
The MRT was shortened by xylanase supplementation.
This effect was more pronounced in jejunum and ileum
than in other segments of the digestive tract. The small
intestine is the location where the digesta stays for a
relatively long time and where the soluble pentosan-
caused increase in intestinal viscosity is supposed to im-
pair nutrient digestibility, which makes both the observed
MRT and digestibilities more reliable than those observed
for other segments. Reduction in MRT over the whole
digestive tract after enzyme addition (Table 4) to diets
rich in gel-forming antinutritive substances is in
agreement with other findings (Jeroch et al., 1990; Almirall
and Esteve-Garcia, 1994; Dänicke et al., 1997a). The MRT
for the entire small intestine were 268, 217, 241, and 209
min in groups S−, S+, T−, and T+, respectively. These
values are slightly longer than those reported in the litera-
ture, 166 to 202 min (Hurwitz and Bar, 1966; Shires et al.,
1987; van Der Klis et al., 1990).
An apparent relationship between MRT and bacterial
counts is mediated by changes in intestinal viscosity. Dif-
ferences in MRT as such seems to be less important for
counts of adhering bacteria.
Apparent digestibilities of N and dry matter in seg-
ments of the digestive tract (Table 5) show negative values
in crop, gizzard, and duodenum. The negative digestibili-
ties reflect the intensive secretion processes in the upper
part of the digestive tract, which, for N, is especially
pronounced in the duodenum due to stomach, bile, and
pancreatic secretions. In these upper segments, the data
are not related to the factors of fat type or xylanase ad-
dition.
The slope of increase in digestibility decreases from
duodenum to ileum. Although positive, xylanase effects
were not significant for N and dry matter digestibility in
jejunum and ileum, which are the main nutrient absorp-
tive sites. Similar trends were reported by Imondi and
Bird (1965) and Bielorai et al. (1973) for intestinal N digest-
ibility and by van der Klis et al. (1990) for dry matter
digestibility. Because the differences in apparent digest-
ibility between unsupplemented and xylanase-supple-
mented animals are more pronounced in duodenum and
ileum than in the rectum, a shift of nutrient absorption
towards the upper region in enzyme-supplemented ani-
mals seems indicated. This result would mean an im-
proved availability of the nutrients for the animal due to
reduced microbial metabolism of the substrates.
The MRT and digestibilities observed for the ceca
should only be taken as a rough approximation, as the
ceca empty at greater time intervals than the remaining
parts of the digestive tract (McNab, 1973). Also, parame-
ters as calculated for gizzard might be biased by the pres-
ence of grit or pseudo-grit and its behavior compared to
the movement of digesta.
In conclusion, there seems to be some evidence that
dietary fat and xylanase interact with the intestinal micro-
ecological system to affect feed transit and nutrient diges-
tion in different segments of the digestive tract which, in
turn, cause differences in live weight.
ACKNOWLEDGMENT
The authors are grateful to Finnfeeds International,
Marlborough, UK, for financial support of the exper-
iments.
REFERENCES
Almirall, M., and E. Esteve-Garcia, 1994. Rate of passage of
barley diets with chromium oxide: Influence of age and poul-
try strain and effect of β-glucanase supplementation. Poultry
Sci. 73:1433–1440.
Bedford, M. R., and H. L. Classen, 1992. Reduction of intestinal
viscosity through manipulation of dietary rye and pentosa-
nase concentration is effected through changes in the carbo-
 EFFECTS OF FAT TYPE AND XYLANASE ADDITION IN BROILERS
hydrate composition of the intestinal aqueous phase and
results in improved growth rate and food conversion effi-
ciency of broiler chicks. J. Nutr. 122:560–569.
Bielorai, R., M. Tamir, E. Alumot, A. Bar, and S. Hurwitz, 1973.
Digestion and absorption of protein along the intestinal tract
of chicks fed raw and heated soyabean meal. J. Nutr.
103:1291–1298.
Brandt, M., and S. M. Allam, 1987. Analytik von TiO2 im Darmin-
halt und Kot nach Kjeldahlaufschluß. Arch. Anim. Nutr.
37:453–454.
Choct, M., and G. Annison, 1992. Anti-nutritive effect of wheat
pentosans in broiler chickens: Roles of viscosity and gut
microflora. Br. Poult. Sci. 33:821–834.
Choct, M., G. Annison, and R. P. Trimble, 1992. Soluble wheat
pentosans exhibit different anti-nutritive activities in intact
and cecectomized broiler chickens. J. Nutr. 122:2457–2465.
Dänicke, S., O. Simon, H. Jeroch, and M. Bedford, 1997a. Interac-
tions between dietary fat type and xylanase supplementation
when rye-based diets are fed to broiler chickens. 1. Physico-
chemical chyme features. Br. Poult. Sci. 38:537–545.
Dänicke, S., O. Simon, H. Jeroch, and M. Bedford, 1997b. Interac-
tions between dietary fat type and xylanase supplementation
when rye-based diets are fed to broiler chickens. 2. Perfor-
mance, nutrient digestibility and the fat-soluble vitamin sta-
tus of livers. Br. Poult. Sci. 38:546–556.
Day, R. M., and O. P. Thomas, 1980. Growth depression of
chicks fed a crude rye extract containing pectic substances.
Poultry Sci. 59:2754–2759.
Demeyer, D. I., and C. J. van Nevel, 1995. Transformations and
effects of lipids in the rumen: Three decades of research at
Gent University. Arch. Anim. Nutr. 48:119–134.
Fuller, R., 1973. Ecological studies on the lactobacillus flora
associated with the crop epithelium of the fowl. J. Appl.
Bacteriol. 36:131–139.
Fuller, R., 1984. Microbial activity in the alimentary tract of
birds. Proc. Nutr. Soc. 43:55–61.
Gee, J. M., W. Lee-Finglas, G. W. Wortley, and I. T. Johnson,
1996. Fermentable carbohydrates elevate plasma enteroglu-
cagon but high viscosity is also necessary to stimulate small
bowel mucosal cell proliferation in rats. J. Nutr. 126:373–379.
Huhtanen, C. N., and J. M. Pensack, 1965. The role of Streptococ-
cus faecalis in the antibiotic growth effect in chickens. Poultry
Sci. 44:830–834.
Hurwitz, S., and A. Bar, 1966. Rate of passage of calcium-45
and yttrium-91 along the intestine, and calcium absorption
in the laying fowl. J. Nutr. 89:311–316.
Imondi, A. R., and F. H. Bird, 1965. The sites of nitrogen absorp-
tion from the alimentary tract of the chicken. Poultry Sci.
44:916–920.
Jeroch, H., A. Müller, and G. Gebhardt, 1990. Influence of beta-
glucanase supplementation on digestion of different barley
varieties in chickens. Pages 259–262 in: Proceedings VIII Eu-
ropean Poultry Conference, Barcelona, Spain.
Jilg, T., K. P. Aiple, and H. Steingass, 1988. Fettstoffwechsel
und Wirkungen von Futterfetten beim Wiederkäuer. Übers.
Tierernährg. 16:109–152.
1299
Johnson, I. T., and J. M. Gee, 1986. Gastrointestinal adaptation
in response to soluble non-available polysaccharides in the
rat. Br. J. Nutr. 55:497–505.
Krawielitzki, K., R. Schadereit, E. Borgmann, and B. Evers, 1987.
Prüfung von Cr2 O3 und TiO2 als Marker für die Bestimmung
von Passagerate und Proteinverdaulichkeit bei Ratten. Arch.
Anim. Nutr. 37:1085–1099.
MacAuliffe, T., and J. McGinnis, 1971. Effect of antibiotic supple-
ments to diets containing rye on chick growth. Poultry Sci.
50:1130–1134.
McNab, J. M., 1973. The avian ceaca: A review. World’s Poult.
Sci. 29:251–263.
Misir, R., and R. R. Marquardt, 1978a. Factors affecting rye
(Secale cereale L.) utilization in growing chicks. IV. The in-
fluence of autoclave treatment, pelleting, water extraction
and penicillin supplementation. Can. J. Anim. Sci. 58:731–
742.
Misir, R., and R. R. Marquardt, 1978b. Factors affecting rye
(Secale cereale L.) utilization in growing chicks. II. The influ-
ence of protein type, protein level and penicillin. Can. J.
Anim. Sci. 58:703–715.
Naumann, C., and R. Bassler, 1993. Die Chemische Untersu-
chung von Futtermitteln. VDLUFA-Verlag, Darmstadt,
Germany.
Persson, S., and S. A. Svensson, 1960. Die Durchgangszeit des
Futters durch den Verdauungstraktus von Küken. Arch.
Geflügelkd. 24:407–416.
Shires, A., J. R. Thompson, B. V. Turner, P. M. Kennedy, and
Y. K. Goh, 1987. Rate of passage of corn-canola meal and
corn-soybean meal diets through the gastrointestinal tract of
broiler and white leghorn chickens. Poultry Sci. 66:289–298.
Southon, S., G. Livesey, J. M. Gee, and I. T Johnson, 1985. Differ-
ences in intestinal protein synthesis and cellular proliferation
in well-nourished rats consuming conventional laboratory
diets. Br. J. Nutr. 53:87–95.
StatSoft, Inc., 1994. Statistica for the Windows Operating Sys-
tem. Tulsa, OK.
Rolls, B. A., A. Turvey, and M. E. Coates, 1978. The influence
of the gut microflora and of dietary fibre on epithelial cell
migration in the chick intestine. Br. J. Nutr. 39:91–98.
Untawale, G. G., and J. McGinnis, 1979. Effect of rye and levels
of raw and autoclaved beans (Phaseolous vulgaris) on adhe-
sion of microflora to the intestinal mucosa. Poultry Sci.
58:928–933.
Vahjen, W., K. Gläser, K. Schäfer, and O. Simon, 1998. Influence
of xylanase supplemented feed on the development of se-
lected bacterial groups in the intestinal tract of broiler chicks.
J. Sci. Food Agric. 130:489–450.
Van Der Klis, J. D., M.W.A. Verstegen, and W. De Wit, 1990.
Absorption of minerals and retention time of dry matter in
the gastrointestinal tract of broilers. Poultry Sci. 69:2185–
2194.
Wagner, D. D., and O. P. Thomas, 1978. Influence of diets con-
taining rye or pectin on intestinal flora of chicks. Poultry Sci.
57:971–975.
Watkins, B. A., and F. H. Kratzer, 1983. Effects of oral dosing
of Lactobacillus strains on gut colonization and liver biotin
in broiler chicks. Poultry Sci. 62:2088–2094.