ArticlePDF Available

Digestibility and metabolizable energy of maize gluten feed for dogs as measured by two different techniques

Authors:
  • Grandfood Industria e Comercio Ltda

Abstract and Figures

Maize gluten feed (MGF) is a co-product of wet milling of maize, and is composed of structures that remain after most starch, gluten and germ has been extracted from the grain. Although currently used in dog foods, its digestibility and energy values have not been documented. Two techniques were used to determine nutrient digestibility of MGF for dog foods. Both techniques used extruded diets fed to Beagle dogs, with six replicates per diet. The first study used a difference method in which 300g/kg of a reference diet was replaced by MGF. Based on the difference method, the coefficient of total tract apparent digestibility (CTTAD) of MGF was 0.53 for dry matter (DM), 0.69 for crude protein (CP), 0.74 for fat, 0.99 for starch, and 0.55 for gross energy (GE). The calculated metabolizable energy (ME) of MGF was 7.99MJ/kg (as-fed). The second study used a regression method and included a basal diet and a basal diet with 70, 140 and 210g MGF/kg of diet (as a substitute for maize starch). Maize gluten feed inclusion resulted in a linear reduction of CTTAD of DM (R2=0.99; P
Content may be subject to copyright.
Animal
Feed
Science
and
Technology
169 (2011) 96–
103
Contents
lists
available
at
ScienceDirect
Animal
Feed
Science
and
Technology
journal
homepage:
www.elsevier.com/locate/anifeedsci
Digestibility
and
metabolizable
energy
of
maize
gluten
feed
for
dogs
as
measured
by
two
different
techniques
I.M.
Kawauchia,
N.K.
Sakomurab,
R.S.
Vasconcellosa,
L.D.
de-Oliveiraa,
M.O.S.
Gomesa,
B.A.
Loureiroa,
A.C.
Carciofic,
aSao
Paulo
State
University
– UNESP,
Via
de
Acesso
Prof.
Paulo
Donato
Castellane,
s/n,
Jaboticabal,
SP
14884-900,
Brazil
bAnimal
Science
Department,
Sao
Paulo
State
University
UNESP,
Via
de
Acesso
Prof.
Paulo
Donato
Castellane,
s/n,
Jaboticabal,
SP
14884-900,
Brazil
cVeterinary
Medicine
and
Surgery
Department,
Sao
Paulo
State
University
UNESP,
Via
de
Acesso
Prof.
Paulo
Donato
Castellane,
s/n,
Jaboticabal,
SP
14884-900,
Brazil
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
14
April
2010
Received
in
revised
form
12
May
2011
Accepted
21
May
2011
Keywords:
Fiber
Maize
Short
chain
fatty
acids
Starch
a
b
s
t
r
a
c
t
Maize
gluten
feed
(MGF)
is
a
co-product
of
wet
milling
of
maize,
and
is
composed
of
struc-
tures
that
remain
after
most
starch,
gluten
and
germ
has
been
extracted
from
the
grain.
Although
currently
used
in
dog
foods,
its
digestibility
and
energy
values
have
not
been
documented.
Two
techniques
were
used
to
determine
nutrient
digestibility
of
MGF
for
dog
foods.
Both
techniques
used
extruded
diets
fed
to
Beagle
dogs,
with
six
replicates
per
diet.
The
first
study
used
a
difference
method
in
which
300
g/kg
of
a
reference
diet
was
replaced
by
MGF.
Based
on
the
difference
method,
the
coefficient
of
total
tract
appar-
ent
digestibility
(CTTAD)
of
MGF
was
0.53
for
dry
matter
(DM),
0.69
for
crude
protein
(CP),
0.74
for
fat,
0.99
for
starch,
and
0.55
for
gross
energy
(GE).
The
calculated
metabo-
lizable
energy
(ME)
of
MGF
was
7.99
MJ/kg
(as-fed).
The
second
study
used
a
regression
method
and
included
a
basal
diet
and
a
basal
diet
with
70,
140
and
210
g
MGF/kg
of
diet
(as
a
substitute
for
maize
starch).
Maize
gluten
feed
inclusion
resulted
in
a
linear
reduc-
tion
of
CTTAD
of
DM
(R2=
0.99;
P<0.001),
CP
(R2=
0.95;
P=0.002),
fat
(R2=
0.87;
P=0.009),
starch
(R2=
0.81;
P<0.001),
and
GE
(R2=
0.99;
P<0.001).
Faecal
production
increased
lin-
early
from
56
g
to
107
g/dog/d
(R2=
0.99;
P<0.001),
with
a
linear
reduction
of
faecal
DM
(R2=
0.99;
P<0.001)
and
a
linear
increase
in
faecal
lactic
acid
concentration
(P<0.02).
Both
urine
(R2=
0.77;
P=0.029)
and
faeces
(R2=
0.92;
P=0.019)
showed
a
linear
reduction
in
pH.
Results
of
ingredient
CTTAD
obtained
by
the
regression
and
difference
methods
were
close
(6%
or
less
of
variation)
for
CP,
fat,
and
starch,
and
also
for
ME
content
(1.4%
higher
for
the
difference
method),
but
the
two
methods
disagreed
on
calculated
CTTAD
of
DM
and
organic
matter.
The
high
dietary
fiber
content
of
MGF
(382
g/kg)
may
explain
the
low
digestibility
of
this
ingredient.
Maize
gluten
feed
could
be
a
useful
ingredient
for
formulations
designed
to
have
low
energy
or
reduce
the
urine
pH
of
dogs.
© 2011 Elsevier B.V. All rights reserved.
1.
Introduction
Several
co-products
of
maize
are
available
for
use
in
human
and
animal
nutrition.
Maize
gluten
feed
(MGF)
is
a
co-product
of
wet
milling
of
maize,
and
is
composed
of
structures
that
remain
after
most
starch,
gluten
and
germ
has
been
extracted
Abbreviations:
CTTAD,
coefficient
of
total
tract
apparent
digestibility;
DM,
dry
matter;
ME,
metabolizable
energy;
SAS,
Statistical
Analysis
Systems.
Corresponding
author.
Tel.:
+55
16
3209
2631;
fax:
+55
16
3203
1226.
E-mail
address:
aulus.carciofi@gmail.com
(A.C.
Carciofi).
0377-8401/$
see
front
matter ©
2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.anifeedsci.2011.05.005
I.M.
Kawauchi
et
al.
/
Animal
Feed
Science
and
Technology
169 (2011) 96–
103 97
from
the
grain.
Approximately
two
thirds
of
MGF
is
composed
of
fibrous
residue
and
the
remaining
one
third
of
dried
steep
liquor
(Blasi
et
al.,
2001).
This
ingredient
has
been
used
as
a
source
of
protein
and
energy
for
ruminants
(Fleck
et
al.,
1988)
but
low
energy
value,
high
fiber
content
(Yen
et
al.,
1974;
Castanon
et
al.,
1990),
and
amino
acid
imbalances
are
considered
limitations
on
the
use
of
MGF
for
single
stomach
animals
(Yen
et
al.,
1971).
Although
used
in
dog
foods,
no
information
about
MGF
digestibility
could
be
found
in
a
review
of
the
current
literature.
The
method
used
for
evaluating
ingredient
digestibility
is
important.
The
difference
(Sakomura
and
Rostagno,
2007)
and
regression
(Shen
et
al.,
2002;
Fang
et
al.,
2007)
methods
have
been
largely
used
in
poultry
and
swine.
Different
methods
can
produce
similar
results,
as
verified
with
swine
by
Fan
and
Sauer
(1995),
but
these
techniques
have
not
been
extensively
studied
in
dogs
(Harmon,
2007),
and
need
further
evaluation
for
extruded
dog
diets
(Fortes
et
al.,
2010).
Considering
the
high
fiber
content
of
MGF,
it
is
assumed
that
its
inclusion
in
extruded
diets
for
dogs
will
reduce
the
food
digestibility
and
energy
density.
However,
the
comprehension
of
the
MGF
effects
in
the
diets
will
enable
a
better
use
of
the
ingredient.
This
study
used
two
techniques,
the
difference
and
the
regression
methods,
for
estimating
the
coefficient
of
total
tract
apparent
digestibility
(CTTAD)
and
the
metabolizable
energy
(ME)
content
of
MGF
in
extruded
diets
for
dogs.
2.
Materials
and
methods
2.1.
Animals
and
experimental
design
All
experimental
procedures
were
approved
by
the
Ethics
and
Animal
Welfare
Committee
of
the
College
of
Agricultural
and
Veterinarian
Sciences
(São
Paulo
State
University,
Jaboticabal,
Brazil).
Thirty-six
healthy
adult
Beagle
dogs,
3.4
±
0.4
years
old
and
10.2
±
0.3
kg
body
weight
were
used
in
two
experiments.
The
dogs
were
kept
in
individual
stainless
steel
metabolic
cages
(100
cm
×
100
cm
×
100
cm)
equipped
with
a
system
to
separate
faeces
and
urine.
Throughout
both
studies,
mean
ambient
temperature
was
23.75
±
0.8 C
and
a
12
h
dark:12
h
light
cycle
was
provided.
In
the
first
experiment
CTTAD
and
ME
of
MGF
was
calculated
by
the
difference
method.
Twelve
dogs
were
allotted
to
one
of
two
diets,
the
reference
diet
and
the
test
diet.
The
second
experiment
evaluated
the
digestibility
of
MGF
by
the
regression
method.
This
experiment
followed
a
randomized
block
design
with
four
diets
and
two
blocks
of
12
animals
each,
for
a
total
of
24
dogs
and
six
replicates
per
diet.
These
two
different
methods
were
used
in
the
determination
of
MGF
digestibility
in
order
to
compare
both
methodologies
for
a
high
fiber
ingredient
in
extruded
diets.
In
both
experiments
dogs
were
fed
the
experimental
diets
for
10
days
following
the
guidelines
of
the
Association
of
American
Feed
Control
Officials
(2004).
The
dogs
were
fed
once
per
day
(0800).
The
ME
of
each
diet
was
estimated
from
the
chemical
composition
and
the
amount
supplied
calculated
as
397
kJ
ME
per
kg0.75 (National
Research
Council,
2006).
Bowls
were
removed
after
30
min
and
any
food
remaining
was
weighed
and
recorded.
Dogs
had
free
access
to
water.
On
the
first
day
of
collection
(day
6),
all
faeces
and
urine
were
removed
from
the
cages
and
discarded
before
0730
and
total
faecal
and
urine
output
for
each
dog
was
collected
from
this
point
for
the
next
five
days.
Faeces
and
urine
were
collected
twice
per
day
and
pooled
per
dog
and
replicate.
Faecal
samples
were
weighed
and
frozen
(15C).
Urine
samples
were
collected
in
plastic
bottles
with
1
mL
of
sulfuric
acid
solution
(1
Eq/L)
and
frozen.
In
the
second
experiment,
as
the
diets
were
formulated
to
be
isometric,
some
additional
measurements
were
made.
Faecal
samples
were
scored
according
to
the
following
system:
1
=
watery—liquid
that
can
be
poured;
2
=
soft,
unformed—stool
assumes
shape
of
container;
3
=
soft,
formed,
moist—softer
stool
that
retains
shape;
4
=
hard,
formed,
dry
stool—remains
firm
and
soft;
5
=
hard,
dry
pellets—small,
hard
mass.
Dogs
were
kept
in
the
cages
an
additional
three
days,
and
during
this
time
urine
was
collected
in
plastic
bottles
stored
in
ice
under
the
cage
funnel.
Each
24-h
of
urine
production
was
pooled
by
dog
and
the
volume,
density
(by
refractometer,
T2-NE,
Atago
Co.,
Tokyo,
Japan)
and
pH
(DM20,
Digicrom
Analítica,
São
Paulo,
Brazil)
were
quantified.
Fresh
faecal
samples
(approximately
10
g)
were
also
collected
daily
to
measure
short
chain
fatty
acids
(SCFA).
Immediately
after
collection,
faecal
samples
were
mixed
in
30
mL
of
a
16%
(v/v)
formic
acid
solution,
precipitated
at
4C
for
72
h,
and
the
supernatant
centrifuged
(5804R,
Eppendorf,
Hamburgo,
Brazil)
three
times
at
5000
rpm
and
15 C
for
15
min.
Faecal
pH
was
determined
by
mixing
10
mL
of
distilled
water
with
5
g
of
fresh
faeces
and
measuring
the
result
with
a
pH
meter
(DM20,
Digicrom
Analítica
Ltda,
São
Paulo,
Brazil).
2.2.
Foods
The
same
sample
of
MGF
was
used
for
both
experiments
(Table
1).
In
the
first
experiment
the
reference
diet
(Table
2)
was
formulated
according
to
AAFCO
(2004)
recommendations
for
adult
dogs.
The
test
diet
was
prepared
using
700
g
of
the
reference
diet
and
300
g
of
MGF,
on
an
as-fed
basis.
This
substitution
level
was
tested
in
other
experiments
and
was
considered
adequate
for
ingredients
with
high
fiber
content
(Sakomura
and
Rostagno,
2007).
In
the
second
experiment
the
same
reference
diet
from
experiment
one
was
used
as
a
basal
food
(food
without
MGF).
Three
other
diets
were
formulated
with
70,
140
and
210
g/kg
of
MGF
(as-fed
basis).
The
MGF
was
added
to
these
foods
in
substitution
of
maize
starch,
and
the
other
ingredients
were
not
altered
(Table
2).
The
adopted
range
of
MGF
inclusion
from
0
to
210
g/kg
was
established
to
evaluate
the
ingredient
effects
on
nutrient
digestibility,
food
ME,
dog
faecal
characteristics,
and
some
gut
fermentation
end
products
formation.
The
MGF
inclusion
was
limited
to
210
g/kg
to
avoid
the
formulation
of
very
high
fiber
diets
(more
than
170
g/kg
of
dietary
fiber),
which
is
not
routinely
used
in
dog
foods.
98 I.M.
Kawauchi
et
al.
/
Animal
Feed
Science
and
Technology
169 (2011) 96–
103
Table
1
Chemical
composition
of
maize
gluten
feeda,b(g/kg,
as-fed
basis).
Item
Dry
matter 894.3
Organic
matter
834.6
Crude
protein
222.8
Acid-hydrolyzed
fat
40.2
Crude
fiber 70.5
Insoluble
dietary
fiber 379.1
Soluble
dietary
fiber 3.0
Total
dietary
fiber
382.1
Starch 112.1
Ash
59.7
Calcium
0.1
Phosphorus
10.0
Potassium
15.4
Magnesium
3.6
Sodium 0.14
Chloride
0.01
Sulfur 2.5
Lactic
acid
55.8
Food
base
excess
(mEq/kg)c100.68
Gross
energy
(MJ/kg) 18.5
aAnalysed
in
duplicate.
bRefinazil
(Corn
Products
Brazil,
Mogi-Guac¸
u,
Brazil).
cFood
base
excess
(mEq/kg)
=
(49.9
×Ca)
+
(82.3
×
Mg)
+
(43.5
×
Na)
+
(25.6
×
K)
(64.6
×
P)
(62.4
×
S)
(28.2
×
Cl)
(Kienzle
and
Wilms-Eilers,
1994).
All
diets
were
mixed
and
ground
in
a
hammer
mill
(Model
4,
D’Andrea,
Limeira,
Brazil)
fitted
with
a
0.8
mm
screen
before
being
extruded
and
kibbled
under
identical
processing
conditions
in
a
single-screw
extruder
(Mab
400S,
Extrucenter,
Monte
Alto,
Brazil).
The
manufacturing
process
was
controlled
by
adjusting
kibble
density
between
360
and
400
g/L
(as-is
basis)
every
20
min
to
ensure
consistent
cooking
and
kibble
quality
(size
and
expansion).
Extruder
pre-conditioning
temperature
was
kept
above
90 C.
Water,
steam,
screw
speed
and
ration
flux
were
adjusted
according
to
diet
formulation,
and
the
extrusion
temperature
varied
between
120
and
135 C.
The
chemical
composition
of
the
foods
is
presented
in
Table
2.
Table
2
Ingredient
and
chemical
compositions
of
the
experimental
diets
(g/kg,
as-fed
basis).
Item
Maize
gluten
feed
(MGF)
addition
Diets,
regression
method
(g
MGF/kg)
Diets,
difference
method
(g
MGF/kg)
0
(basal)
70
g
140
g
210
g
0
g
(reference
diet)
300
g
(test-diet)
Ingredient
composition
Poultry
by
product
meal
296.9
296.9
296.9
296.9
296.9
207.8
Maize
222.3
222.3
222.3
222.3
222.3
155.6
Maize
starch
210.0
142.6
75.2
7.7
210.0
147.0
Maize
gluten
feed
0.0
70.0
140.0
210.0
0.0
300.0
Broken
rice
170.6
170.6
170.6
170.6
170.6
119.4
Poultry
fat
58.1
55.5
52.9
50.4
58.1
40.7
Minor
ingredientsa42.1
42.1
42.1
42.1
42.1
29.5
Total
1000
1000
1000
1000
1000
1000
Chemical
compositionb
Dry
matter
934.3
933.3
937.8
938.0
934.3
943.7
Crude
protein
202.8
217.5
240.1
250.5
202.8
231.8
Acid-hydrolyzed
fat 102.5
103.0
106.3
106.7
102.5
97.6
Starch
469.7
433.0
381.7
334.2
469.7
380.6
Dietary
fiber
76.5
96.0
127.5
165.1
76.5
107.1
Food
density
(g/L)
362
363
394
384
362
368
Starch
gelatinization
index
(%)
95.6
93.9
96.1
90.3
95.6
91.3
Food
base
excess
(mEq/kg)c416
379
342
305
aSupplied
per
kilogram
of
diet:
vitamin
A,
15,000
IU;
vitamin
D,
1300
IU;
vitamin
E,
200
IU;
thiamin,
10
mg;
riboflavin,
14
mg;
pantothenic
acid,
60
mg;
niacin,
90
mg;
pyridoxine,
9
mg;
folic
acid,
0.50
mg;
vitamin
B12,
0.2
mg
iron,
130
mg;
copper,
13
mg;
magnesium,
13
mg;
zinc,
180
mg;
iodine,
2
mg;
selenium,
0.3
mg;
choline
chloride,
2
g;
potassium
chloride,
5.8
g;
sodium
chloride,
3.2
g;
dicalcium
phosphate,
1
g;
palatant,
20
g;
mold
inhibitor
(Mold
Zap:
ammonium
dipropionate,
acetic
acid,
sorbic
acid
and
benzoic
acid
Alltech
do
Brasil
Agroindustrial
Ltda,
Curitiba,
Brazil),
1
g;
antioxidant
(Banox:
butylated
hydroxyanisole,
butylated
hydroxytoluene,
propyl
gallate
and
calcium
carbonate
– Alltech
do
Brasil
Agroindustrial
Ltda,
Curitiba,
Brazil),
0.4
g.
bAnalysed
in
duplicate.
cFood
base
excess
(mEq/kg)
=
(49.9
×
Ca)
+
(82.3
×
Mg)
+
(43.5
×
Na)
+
(25.6
×
K)
(64.6
×
P)
(62.4
×
S)
(28.2
×
Cl)
(Kienzle
and
Wilms-Eilers,
1994).
I.M.
Kawauchi
et
al.
/
Animal
Feed
Science
and
Technology
169 (2011) 96–
103 99
2.3.
Laboratory
analyses
Faecal
samples
were
freeze
dried
for
48
h
(Savant
ModulyoD
Freeze
Dryer,
Thermo
Electron
Corporation,
Milford,
EUA).
Freeze
dried
faecal
samples
and
diets
were
ground
in
a
cutting
mill
(Mod
MA-350,
Marconi,
Piracicaba,
Brazil)
fitted
with
a
1
mm
screen.
Urine
samples
(approximately
90
g)
were
weighed
and
placed
in
three
Petri
dish
(30
g
each),
dried
in
a
forced
air
oven
at
55 C
for
24
h
(320-SE,
Fanem,
São
Paulo,
Brazil),
and
the
dry
residue
placed
in
silicon
capsules
for
combustion.
Maize
gluten
feed,
diets
and
faecal
samples
were
tested
for
dry
matter
(DM)
by
oven-drying
the
sample
(method
934.01),
ash
by
muffle
furnace
incineration
(method
942.05),
crude
protein
(CP)
by
the
Kjeldahl
method
(method
954.01),
and
acid-
hydrolyzed
fat
using
a
Soxhlet
apparatus
(method
954.02)
according
to
the
Association
of
Official
Analytical
Chemists
(1995).
Organic
matter
(OM)
was
calculated
as
1000-ash.
Minerals
were
analysed
after
nitro
perchloric
digestion;
phosphorus
was
measured
by
visible
spectrophotometery
(Labquest
Bio
2000.
Labtest
Diagnóstica
S.A.,
Lagoa
Santa,
Brazil),
calcium,
potas-
sium,
magnesium,
chloride
and
sodium
by
flame
atomic
absorption
spectrophotometry
(GBC-932
AA,
Scientific
Equipment
PTY
LTD,
Melbourne,
Australia),
and
sulfur
by
the
turbidimetric
method
according
to
the
Association
of
Official
Analytical
Chemists
(1995).
Dietary
fiber
was
measured
by
using
a
combination
of
enzymatic
and
gravimetric
procedures
according
to
Prosky
et
al.
(1992).
Total
amount
of
starch
was
determined
according
to
the
method
described
by
ICC
(1995)
and
Karkalas
(1985).
Gross
energy
(GE)
was
determined
in
a
bomb
calorimeter
(model
1261,
Parr
Instrument
Company,
Moline,
IL,
USA).
Faecal
SCFA
were
analysed
by
gas
chromatography
(model
9001,
Finnigan,
San
Jose,
USA)
according
to
Erwin
et
al.
(1961)
and
using
a
glass
column
2
m
in
length
and
3.17
mm
in
width,
covered
with
80/120
Carbopack
B-DA/4%
Carbowax
20M.
Nitrogen
was
the
carrier
gas
with
a
flow
rate
of
25
mL/min.
Working
temperatures
were
220 C
at
injection,
210 C
in
the
column,
and
250 C
in
the
flame
ionization
detector.
Lactic
acid
was
measured
in
MGF
and
faecal
samples
according
to
Pryce
(1969),
using
a
colorimetric
method
(Spectrophotometer
Quick
Lab,
Drake,
São
José
do
Rio
Preto,
Brazil).
All
analyses
were
carried
out
in
duplicate
and
the
coefficient
of
variation
was
below
5%
in
all
cases.
2.4.
Coefficient
of
total
tract
apparent
digestibility
and
metabolizable
energy
Coefficient
of
total
tract
apparent
digestibility
and
ME
of
the
experimental
diets
were
calculated
according
to
the
quanti-
tative
collection
of
faeces
and
urine
protocol
and
calculation
procedures
described
by
Association
of
American
Feed
Control
Officials
(2004).
From
the
results
of
diets
CTTAD,
in
each
experiment
the
CTTAD
and
ME
of
MGF
were
calculated
using
a
different
method.
In
the
first
experiment
MGF
CTTAD
was
calculated
based
on
the
CTTAD
values
of
the
reference
diet,
CTTAD
values
of
the
test
diet,
and
the
inclusion
level
of
MGF
corrected
to
DM-basis,
using
the
equation
proposed
by
Matterson
et
al.
(1965):
CTTADing
=
CTTADrd
+CTTADtd
CTTADrd
inclusion
level
of
test
in
ingredient
in
basal
(g/kg)/100
where
CTTADing
is
the
coefficient
of
total
tract
apparent
digestibility
of
the
MGF,
CTTADrd
is
the
coefficient
of
total
tract
apparent
digestibility
of
the
reference
diet
and
CTTADtd
is
the
coefficient
of
total
tract
apparent
digestibility
of
the
test
diet.
In
the
second
experiment
the
results
of
food
CTTAD
and
ME
were
fit
into
a
linear
regression
model
and
the
CTTAD
and
ME
of
the
MGF
was
estimated
by
the
intercept
when
the
MGF
content
was
extrapolated
to
1000
g/kg
(Harmon,
2007;
Fan
and
Sauer,
1995).
2.5.
Statistical
analysis
Experiment
one
was
analysed
as
a
completely
randomized
design,
using
the
general
linear
models
procedure
of
SAS
(1997).
Model
sums
of
squares
were
separated
in
treatment
(food)
and
animal
effects.
Data
from
the
second
experiment
were
analysed
as
a
completely
randomized
block
design
using
the
general
linear
models
procedure
of
SAS
(1997).
Model
sums
of
squares
were
separated
into
treatment
(food),
block,
and
animal
effects.
When
treatment
differences
were
detected
on
variance
analysis,
polynomial
regressions
were
done
to
describe
the
relationship
between
MGF
inclusion
level
and
evaluated
criteria.
Values
of
P<0.05
was
considered
significant.
All
data
were
found
to
comply
with
the
assumptions
of
ANOVA
models.
Results
were
presented
as
means
±
standard
error.
3.
Results
3.1.
Maize
gluten
feed
chemical
composition
The
chemical
composition
of
MGF
(Table
1)
corresponded
with
previous
estimates
(National
Research
Council,
2006;
Rostagno
et
al.,
2005),
except
for
a
low
starch
content.
3.2.
Digestibility,
faecal
and
urine
production
Diets
were
readily
consumed
without
diarrhea,
vomiting
or
refusal.
There
were
no
differences
in
DM
intake
(P>0.05,
Table
3)
but
protein,
starch,
fiber
and
lactic
acid
ingestion
differed
(P<0.01)
as
the
dietary
concentration
of
these
nutri-
100 I.M.
Kawauchi
et
al.
/
Animal
Feed
Science
and
Technology
169 (2011) 96–
103
Table
3
Food
consumption
(g/dog/day
as-fed
basis),
coefficient
of
total
tract
apparent
digestibility
(CTTAD),
and
metabolizable
energy
content
(ME,
MJ/kg
as-fed
basis)
of
the
experimental
diets.
Item
Maize
gluten
feed
(MGF)
addition
Diets,
regression
method
(g
MGF/kg)
ContrastscDiets,
difference
method
(g
MGF/kg)
0
(basal) 70
g
140
g
210
g
S.E.M.aLinear
Quadratic
0
g
(reference)
300
g
(test-diet)
S.E.M.b
Mean
body
weight,
kg
10.3
10.6
10.0
10.0
0.2
0.416
0.786
10.3
11.8 d0.3
Nutrient
intake
Dry
matter
162.2
167.2
174.9
166.2
3.4
0.480
0.350
162.2
179.4d3.9
Crude
protein 35.2
38.9
44.8
44.4
0.9
0.002
0.240
35.2
41.6d1.2
Acid-hydrolyzed
fat 17.8
18.5
19.8
18.9
0.4
0.180
0.290
17.8
17.5
0.3
Total
dietary
fiber 12.4
16.1
22.3
27.4
0.4
<0.001
0.400
12.4
36.9d3.7
Starch 81.5
77.6
71.2
59.2
1.4
<0.001
0.160
81.5
68.3d2.3
Lactic
acid
0
0.69
1.4
2.1
0.2
<0.001
0.540
0
3.1d0.5
CTTAD
Dry
matter
0.828
0.794
0.752
0.728
0.002
<0.001
0.194
0.828
0.741d0.014
Organic
matter
0.910
0.874
0.831
0.802
0.002
<0.001
0.257
0.910
0.798d0.017
Crude
protein
0.838
0.826
0.820
0.800
0.004
0.002
0.593
0.838
0.794d0.008
Acid-hydrolyzed
fat
0.894
0.882
0.883
0.874
0.002
0.009
0.714
0.894
0.851d0.002
Total
dietary
fiber 0.352
0.173
0.190
0.258
0.012
0.017
<0.001
0.352
0.361
0.009
Starch
0.998
0.998
0.998
0.996
0.001
<0.001
0.052
0.998
0.998
0.001
Gross
energy 0.908
0.876
0.841
0.813
0.002
<0.001
0.514
0.908
0.802d0.016
ME
14.8
14.5
14.3
13.4
0.054
<0.001
0.011
14.8
13.0d0.427
aS.E.M.
=
standard
error
of
the
mean
(n
=
24).
bS.E.M.
=
standard
error
of
the
mean
(n
=
12).
cLinear
and
quadratic
effect
of
MGF
additions.
dDifference
between
reference
and
test-diet
of
the
difference
method
(P<0.05).
ents
changed
with
the
addition
of
MGF.
Maize
gluten
feed
addition
resulted
in
a
linear
reduction
of
dietary
DM,
OM,
CP,
fat,
starch,
and
GE
CTTAD
(P<0.01),
and
also
ME
content
of
the
food
(P<0.001).
Dietary
fiber
digestibility
(P<0.001)
showed
a
quadratic
reduction
with
the
addition
of
MGF.
Dietary
fiber
consumption
(g/dog/day)
was
negatively
correlated
with
CTTAD
of
DM
(CTTAD
of
DM
=
0.902
0.00636
×
dietary
fiber
ingestion;
P<0.001;
R2=
0.88),
CTTAD
of
OM
(CTTAD
of
OM
=
0.991
0.00694
×
dietary
fiber
ingestion;
P<0.001;
R2=
0.93),
and
CTTAD
of
GE
(CTTAD
of
GE
=
0.979
0.00606
×
dietary
fiber
ingestion;
P<0.001;
R2=
0.92).
Although
statistically
significant,
CTTAD
of
CP
(CTTAD
of
CP
=
0.865
0.00227
×
dietary
fiber
ingestion;
P=0.002;
R2=
0.37),
CTTAD
of
fat
(CTTAD
of
fat
=
0.905
0.00112
×
dietary
fiber
ingestion;
P=0.01;
R2=
0.27),
and
the
CTTAD
of
starch
(CTTAD
of
starch
=
0.100–0.000123
×
dietary
fiber
ingestion;
P<0.001;
R2=
0.53)
were
less
influenced
by
dietary
fiber
ingestion,
as
can
be
seen
by
the
smaller
slope
of
the
equations.
Faecal
production
increased
linearly
(R2=
99;
P<0.001),
and
faecal
pH
(R2=
0.92;
P<0.01)
and
DM
decreased
linearly
(R2=
0.99;
P<0.01)
with
increased
amounts
of
MGF
(Table
4).
Faecal
total
SCFA
concentration
increased
quadratically
(P<0.02),
Table
4
Faecal
and
urine
evaluation,
and
short
chain
fatty
acid
(SCFA,
mmol/g
of
faecal
DM)
concentration
in
dog
faeces.
Item
Maize
gluten
feed
(MGF)
addition
Diets,
regression
method
(g
MGF/kg)
ContrastscDiets,
difference
method
(g
MGF/kg)
0
(basal)
70
g
140
g
210
g
S.E.M.aLinear
Quadratic
0g
(reference)
300g
(test-diet)
S.E.M.b
Faecal
evaluation
g/dog/day
(fresh)
56
73
94d107d3
<0.001
0.680
56.09
115.53 d9.859
Faeces
dry
matter
(g/kg)
497
475
448
424
4
<0.001
0.880
496.9
407.6d14.836
Score
4.7
3.8
3.6
3.5
0.0
<0.001
0.420
4.7
3.6d0.044
pH
7.45
7.43d7.32d7.20
0.04
0.019
0.520
SCFA
Acetic
212
169
171
198
6
0.420
0.006
Propionic
73
67
83
97
4
0.012
0.180
Butyric 35
27
30
30
1
0.320
0.110
Total
SCFA
320
263
283
325
10
0.690
0.019
Lactic
acid
8.5
9.1
12.7
12.8
0.7
0.018
0.870
Urine
evaluation
mL/dog/day
231
288
309
200
20
0.680
0.055
Specific
gravidity 1.033
1.026
1.028
1.038
0.001
0.023
0.064
pH
7.42
7.16
7.23
6.99
0.06
0.029
0.943
aS.E.M.
=
standard
error
of
the
mean
(n
=
24).
bS.E.M.
=
standard
error
of
the
mean
(n
=
12).
cLinear
and
quadratic
effect
of
MGF
additions.
dDifference
between
reference
and
test-diet
of
the
difference
method
(P<0.05).
I.M.
Kawauchi
et
al.
/
Animal
Feed
Science
and
Technology
169 (2011) 96–
103 101
Table
5
Extruded
maize
gluten
feed
(MGF)
coefficient
of
total
tract
apparent
nutrient
digestibility
(CTTAD)
and
metabolizable
energy
(ME;
MJ/kg
as-fed
basis)
for
dogs
calculated
by
the
regression
and
difference
methods.
Item
Equation
R2P>F
Calculated
CTTAD
of
MGF
RegressionaDifference b
CTTAD
Dry
matter
y
=
0.00046803x
+
0.8261286
0.99
<0.001
0.36
0.53
±
0.31
Organic
matter
y
=
0.00051212x
+
0.9086437
0.99
<0.001
0.40
0.52
±
0.22
Crude
protein y
=
0.000117175x
+
0.8364403
0.95
0.001
0.72
0.69 ±
0.31
Acid-hydrolyzed
fat y
=
0.00009992x
+
0.8920960
0.87
0.006
0.79
0.74
±
0.13
Total
dietary
fiber y
=
0.00046872x
+
0.2915582
0.79
<0.001
0.18
0.36
±
0.65
Starch y
=
0.00000794x
+
0.9983967
0.81
0.009
0.99
0.99
±
0.01
Gross
energy
y
=
0.00044755x
+
0.9067011
0.99
<0.001
0.46
0.55
±
0.25
ME
y
=
0.00702355x
+
14.90357
0.82
<0.001
7.88
7.99
±
0.92
aValues
calculated
by
the
intercept
when
the
MGF
content
was
extrapolated
to
1000
g/kg.
bMean
±
standard
error.
Values
calculated
by
difference
using
the
method
of
Matterson
et
al.
(1965).
mainly
due
to
a
linear
increase
in
propionic
acid
concentration
(P<0.01).
Faecal
lactic
acid
also
increased
linearly
with
MGF
addition
(P<0.02),
and
there
was
a
linear
decrease
in
urinary
pH
(P<0.03).
The
CTTAD
of
MGF
estimated
by
the
difference
and
regression
methods
is
shown
in
Table
5.
Results
were
similar
for
CP,
fat
and
starch
digestibility,
and
for
MGF
ME
content
whereas
estimates
of
DM
and
OM
digestibility
obtained
with
the
difference
method
were,
respectively,
47%
and
30%
greater
than
values
estimated
by
the
regression
method.
None
of
the
methods
could
estimate
fiber
digestibility
with
both
methods
providing
negative
values.
4.
Discussion
The
reduced
starch
content
verified
in
the
MGF
sample
used
in
this
study
could
be
attributed
to
an
improved
efficiency
of
starch
extraction
from
maize
during
it
process,
which
can
reduce
the
energy
value
of
the
ingredient.
The
lactic
acid
verified
in
MGF
was
likely
produced
during
the
formation
of
maize
steep
liquor,
which
was
later
dried
and
incorporated
as
an
ingredient.
The
co-product
had
also
high
amounts
of
insoluble
fiber.
The
linear
reduction
in
CTTAD
of
nutrients
from
the
addition
of
MGF
to
the
diets
could
be
explained
by
the
higher
fiber
content
of
the
ingredient.
The
negative
correlation
between
dietary
fiber
ingestion
and
CTTAD
of
DM
and
OM
detected
in
the
present
study
was
previously
reported
(Fahey
Jr.
et
al.,
1992;
Cole
et
al.,
1999;
Earle
et
al.,
1998).
This
effect
may
not
only
be
related
to
a
very
low
fiber
digestibility
in
dogs
but
also
to
a
shortened
transit
time
in
the
small
intestine
caused
by
insoluble
fiber,
as
verified
in
pigs
(Le
Goff
et
al.,
2002;
Wilfart
et
al.,
2007)
and
dogs
(National
Research
Council,
2006).
Maize
gluten
feed
addition
to
the
diets
had
a
weaker
effect
on
CP,
fat
and
starch
CTTAD
than
on
DM
CTTAD.
This
trend
could
be
explained
by
the
reduced
interference
of
low
fermentable
insoluble
fiber
on
nutrient
digestibility
in
dogs
(Sunvold
et
al.,
1995b)
in
comparison
to
soluble
and
fermentable
fibers
(Silvio
et
al.,
2000).
Starch
CTTAD
was
high
for
all
evaluated
diets,
as
previously
demonstrated
in
properly
extruded
dog
foods
(Carciofi
et
al.,
2008).
To
reduce
dietary
energy
use,
high
fiber
diets
have
been
used
for
dogs
during
weight
loss
programs
(Jewell
et
al.,
2000;
Carciofi
et
al.,
2005).
It
is
believed
that
fiber
could
help
weight
loss
and
promote
a
satiety
stimulus
in
dogs
(Laflamme,
2006).
Considering
the
low
ME
and
high
dietary
fiber
content
of
MGF,
this
ingredient
could
be
considered
for
diets
with
reduced
energy
content,
including
foods
for
old
dogs,
obese
animals,
dogs
with
low
physical
activity,
and
also
for
dogs
with
diabetes
mellitus.
Increased
faecal
excretion
and
reduced
faecal
score
with
MGF
addition
reflects
the
increase
in
dietary
insoluble
fiber.
Insoluble
fiber
causes
water
retention
in
the
faecal
bolus,
increasing
the
bulk
(Meyer
and
Tungland,
2001).
However,
this
effect
was
not
very
pronounced;
the
faeces
remained
at
an
adequate
score
without
any
loose
stools
detected
during
the
experiment.
The
linear
reduction
in
faecal
pH
could
be
explained
by
the
increased
total
SCFA
and
lactic
acid
concentrations
in
the
faeces.
Two
sources
of
lactic
acid
can
be
considered:
an
increased
production
of
lactic
acid
during
fiber
fermentation
in
the
colon,
or
increased
consumption
of
lactic
acid
as
a
compound
of
MGF.
However,
the
exact
role
of
ingested
lactic
acid
on
faecal
pH
remains
controversial,
as
this
organic
acid
is
absorbed
by
the
host
and
metabolized
by
gut
microorganisms,
and
does
not
tend
to
accumulate
in
the
colon
(Bernalier
et
al.,
1999).
Short
chain
fatty
acids
are
also
readily
absorbed
by
dog
colon
mucosa
(Herschel
et
al.,
1981),
and
faecal
measurements,
as
in
the
present
study,
may
not
show
quantitative
alterations
of
the
fermentative
activity
in
the
large
intestine.
Another
consideration
is
the
limited
capacity
of
intestinal
fermentation
in
dogs
due
to
the
short
and
non-sacculated
large
intestine
(Sunvold
et
al.,
1995a).
So,
a
high
rate
of
SCFA
absorption
and
limited
intestinal
fermentation
could
explain,
at
least
partially,
the
reduced
influence
of
MGF
addition
on
faecal
SCFA
concentrations,
as
observed
previously
by
Guevara
et
al.
(2008).
The
reduction
in
urine
pH
caused
by
an
increased
ingestion
of
MGF
could
be
explained
by
the
macro
element
composition
of
the
ingredient.
The
negative
food
base
excess
of
MGF
reflects
a
large
increased
concentration
of
anions
relative
to
cations
(Kienzle
and
Wilms-Eilers,
1994).
The
sulfur
content
of
MGF
is
relatively
high,
probably
due
to
sulfate
inclusion
during
steep
liquor
processing
of
maize,
thereby
explaining
its
negative
food
base
excess.
Most
dry
dog
diets
have
a
positive
food
102 I.M.
Kawauchi
et
al.
/
Animal
Feed
Science
and
Technology
169 (2011) 96–
103
base
excess,
resulting
in
alkaline
urine
(Carciofi,
2007)
and
possibly
predisposing
animals
to
struvite
uroliths
(Gevaert
et
al.,
1991).
Even
considering
that
most
struvite
uroliths
in
dogs
are
related
to
bacterial
cystitis,
it
may
be
interesting
to
examine
the
effect
of
the
addition
of
ingredients
with
negative
food
base
excess,
such
as
MGF,
on
struvite
urolith
formation.
Ingredient
digestibility
studies
are
scarce
in
dogs,
and
food
ME
content
is
currently
estimated
by
chemical
composition
(NRC,
2006),
which
does
not
take
into
account
the
energy
content
of
the
specific
ingredients.
Data
on
digestibility
and
ME
content
of
individual
ingredients
could
be
of
value
for
diet
formulation
by
allowing
a
technical
ingredient
selection
based
on
nutrient
digestibility
rather
than
on
crude
or
gross
nutrients
(Fortes
et
al.,
2010).
Several
factors
might
affect
ingredient
digestibility.
Nutrient
concentration
is
an
important
factor;
a
study
of
19
ingredients
used
in
dog
foods
(Kendall
and
Holme,
1982)
found
a
positive
correlation
between
fat
and
protein
content
and
ingredient
digestibility.
However,
for
fiber
and
ash
contents
this
correlation
was
negative.
The
selection
of
a
method
to
study
ingredient
digestibility
must
consider
the
possible
limitations
of
each
one.
The
regres-
sion
method
is
more
expensive
than
the
difference
method
since
it
requires
the
use
of
several
experimental
diets.
However,
important
information
about
the
influence
of
different
levels
of
the
ingredient
on
diet
digestibility
and
faecal
production
can
be
obtained,
allowing
establishment
of
the
adequate
inclusion
amount.
The
main
limitation
of
the
regression
method
is
the
fact
that
the
generated
equation
is
valid
only
for
the
range
of
ingredient
inclusion
used.
The
extrapolation
of
the
inclusion
level
to
1000
g/kg
would
increase
the
experimental
error,
and
could
provide
results
that
are
not
consistent
with
the
physiology
of
the
animal.
This
extrapolation
also
depends
on
finding
a
linear
regression
between
ingredient
inclusion
and
food
digestibility.
Even
though
this
equation
was
significant
in
the
range
of
the
evaluated
inclusion
rates
used
in
the
test
diets,
higher
ingredient
additions
could
result
in
quadratic
or
cubic
responses,
making
the
linear
extrapolation
not
always
valid.
This
could
be
the
case
of
estimated
MGF
CTTAD
of
DM
and
OM
in
the
present
study
because
the
linear
extrapolation
resulted
in
very
low
values
that
might
not
be
correct.
The
difference
method,
however,
assumes
no
associative
effects
between
the
evaluated
ingredient
and
the
basal
diet,
which
does
not
always
occur.
Depending
on
the
chemical
composition
of
the
ingredient
being
tested,
inclusion
can
result
in
increased
or
decreased
basal
diet
digestibility,
leading
to
estimation
errors.
However,
since
diet
digestibility
was
determined
and
not
extrapolated,
the
results
can
be
considered
to
have
good
reliability
(Sakomura
and
Rostagno,
2007).
In
the
present
study
not
all
nutrient
digestibilities
could
be
determined
by
the
regression
and
difference
methods.
Both
methods
agreed
for
the
estimates
of
MGF
CTTAD
of
CP,
fat,
and
starch
and
for
ME
content
but
dietary
fiber
CTTAD
could
not
be
estimated
by
either
method.
Despite
the
fact
that
the
diets
in
this
study
had
higher
CTTAD
of
dietary
fiber
than
estimates
for
soy
hulls
(Cole
et
al.,
1999)
or
by-products
of
maize
ethanol
extraction
(Guevara
et
al.,
2008),
a
negative
fiber
CTTAD
for
MGF
was
estimated
by
both
calculation
methods.
Indeed,
the
CTTAD
of
dietary
fiber
indicates
a
moderate
fermentation
of
the
fiber
fraction
of
the
MGF.
This
suggests
a
possible
associative
effect
of
MGF
and
the
basal
diet
in
the
difference
method,
and
a
lack
of
linearity
of
fiber
CTTAD
in
the
regression
method
due
to
a
reduced
capacity
of
dogs
to
use
the
fiber
fraction
of
the
diet.
Thus,
the
choice
of
the
most
adequate
method
to
estimate
CTTAD
will
depend
on
the
research
objectives
and,
especially,
the
composition
of
the
ingredient
to
be
studied.
5.
Conclusions
The
use
of
MGF
as
a
diet
ingredient
must
reflect
its
high
fiber
level
and
low
DM
CTTAD
and
ME
content.
Maize
gluten
feed
could
be
an
effective
ingredient
for
formulations
designed
to
have
low
energy
or
reduced
food
base
excess.
Both
the
difference
and
regression
methods
provided
estimates
of
protein,
fat
and
starch
digestibilities
and
for
metabolizable
energy
content
of
the
ingredient,
but
differences
between
the
methodologies
and
the
impact
of
ingredient
chemical
composition
must
be
considered
for
interpretation
of
the
results.
Acknowledgements
The
authors
acknowledge
the
financial
support
of
Corn
Products
S.A.,
and
the
financial
and
technical
support
of
Mogiana
Alimentos
S.A.
(Guabi)
to
the
Laboratory
of
Research
in
Nutrition
and
Nutritional
Diseases
of
Dogs
and
Cats
“Prof.
Dr.
Flávio
Prada”.
References
Association
of
American
Feed
Control
Officials,
2004.
Dog
and
Cat
Nutrient
Profiles.
Official
Publication
of
the
Association
of
American
Feed
Control
Officials
Incorporated.
AAFCO,
Oxford,
IN,
USA.
Association
of
Official
Analytical
Chemists,
1995.
Official
Methods
of
Analysis,
16th
ed.
AOAC,
Washington,
DC,
USA.
Bernalier,
A.,
Dore,
J.,
Durand,
M.,
1999.
Biochemistry
of
fermentation.
In:
Gibson,
G.R.,
Roberfroid,
M.B.
(Eds.),
Colonic
Microbiota,
Nutrition
and
Healthy.
Kluwer
Academic
Publishers,
Dordrecht.
Blasi,
D.A.,
Brouk,
M.J.,
Douillard,
J.S.,
Montgomery,
S.P.,
2001.
Corn
Gluten
Feed:
Composition
and
Feeding
Value
for
Beef
and
Dairy
Cattle.
Kansas
State
University,
Agricultural
Experimental
Station
and
Cooperative
Extension
Service,
Manhattan,
14
p.
(Bulletin,
MF-2488).
Carciofi,
A.C.,
Gonc¸
alves,
K.N.V.,
Vasconcellos,
R.S.,
Bazolli,
R.S.,
Brunetto,
M.A.,
Prada,
F.,
2005.
A
weight
loss
protocol
and
owners
participation
in
the
treatment
of
canine
obesity.
Cienc.
Rural
35,
1331–1338.
Carciofi,
A.C.,
2007.
Métodos
para
estudo
das
respostas
metabólicas
de
cães
e
gatos
a
diferentes
alimentos.
Rev.
Bras.
Zootecn.
36,
235–249.
Carciofi,
A.C.,
Takakura,
F.S.,
de-Oliveira,
L.D.,
Teshima,
E.,
Jeremias,
J.T.,
Brunetto,
M.A.,
Prada,