Published December 8, 2014
Increased puroindoline levels slow ruminal digestion of wheat
(Triticum aestivum L.) starch by cattle
C. G. Swan,* J. G. P. Bowman,† J. M. Martin,* and M. J. Giroux*1
*Department of Plant Sciences and Plant Pathology; and †Department of Animal and Range Sciences,
Montana State University, Bozeman 59717
creased starch digestibility by 30.8% (P < 0.05). Experiment 2 separated the effects of particle size and total
PIN content on digestion by milling the hardest and
softest lines such that the mean particle size was nearly
identical. Increased PIN decreased DMD by 21.7% (P
< 0.05) and starch digestibility by 19.9% (P < 0.05)
across particle sizes smaller than whole kernel. Experiment 3 addressed the time course of PIN effects in the
rumen by observing ground samples of the hardest and
softest lines over a 12-h in situ period. Increased PIN
decreased DMD by 10.4% (P < 0.05) and starch digestibility by 11.0% (P < 0.05) across all time points. Dry
matter and starch digestibility results demonstrated
that increased expression of PIN was associated with
a decreased rate of ruminal digestion independent of
particle size. Puroindolines seem to aid in the protection
of starch molecules from microbial digestion in the rumen, potentially increasing the amount of starch entering the small intestine.
ABSTRACT: Starch is the primary nutrient in ruminant diets used to promote high levels of performance.
The site of starch digestion alters the nature of digestive
end products (VFA in the rumen vs. glucose in the small
intestine) and the efficiency of use. Cereal grain endosperm texture plays a major role in the rate and extent
of starch degradation in ruminants. Wheat grain texture is regulated by the starch surface protein complex
friabilin that consists primarily of puroindoline (PIN)
A and B. Soft kernel texture in wheat is a result of both
PIN genes being in the wild type active form and bound
to starch. The objective of this study was to investigate
the effect of varying PIN content in wheat on the rate
of starch digestion in the rumen of beef cattle. In Exp.
1, 6 transgenic soft pin a/b isolines created in a hard
wheat background, and 2 hard wheat controls were
milled to yield a wide range of mean particle sizes across
all lines. Milled samples were incubated in situ for 3 h.
Increased expression of both PINA and PINB decreased
DM digestibility (DMD) by 29.2% (P < 0.05) and de-
Key words: digestibility, grain hardness, particle size, puroindoline, wheat
2006 American Society of Animal Science. All rights reserved.
INTRODUCTION
J. Anim. Sci. 2006. 84:641–650
in starch granule structure among species of cereal
grains may account for distinct rates of digestion patterns. Protein and structural carbohydrates within the
cereal kernel may be more important in determining
the extent of ruminal starch digestion than the starch
type (McAllister et al., 1993).
Wheat (Triticum aestivum L.) grain hardness is determined by the degree of adhesion between starch
granules and the protein matrix, regulated by the protein complex friabilin. Friabilin, isolated from the surface of starch granules, contains 2 major proteins, puroindolines (PIN) A and B. Puroindolines contain a
unique tryptophan-rich domain believed to be involved
in their binding to the phospholipids of starch granules
(Gautier et al., 1994). Soft wheat results from both pin
genes being in the wild type form, whereas hard wheats
have an absence or alteration in either pin gene (Morris
et al., 2001). Hard wheat transformed with added wild
type pina, pinb, or pina and b resulted in isolines with
a wide range of grain textures (Hogg et al., 2004).
Cereal grains are typically fed to ruminants to increase starch intake. The site of starch absorption along
the gastrointestinal tract affects cattle performance
and feed efficiency. Slower rates of digestion increase
the amount of starch bypassing the rumen. Starch digested in the small intestine can produce up to 42%
more energy than fermentation (Owens et al., 1986)
because of a more efficient use of digestive end products
(glucose vs. VFA).
Grain texture plays a major role in the rate and location of starch digestion in ruminants (Philippeau et al.,
1999). Small grains (wheat, barley, or oats) are more
rapidly fermented than corn and sorghum. Variations
1
Corresponding author: [email protected]
Received March 10, 2005.
Accepted September 8, 2005.
641
642
Swan et al.
Table 1. Puroindoline (PIN) expression, kernel characteristics, and chemical composition
(DM basis) of wheat lines varying in PIN expression
ID1
161
Hi-Line
HWC avg
HGA1
HGA3
HGA avg
HGB6
HGB12
HGB avg
HGAB12
HGAB18
HGAB avg
LSD4 (0.05)
LSD5 (0.05)
PINA
level2
PINB
level2
SKCS grain
hardness3
Kernel
diameter,3
mm
Kernel wt,3
mg
ADF, %
Protein, %
Starch,
%
1.00
1.00
1.00
7.00
8.00
7.50
1.00
1.00
1.00
5.75
7.25
6.50
0.41
0.29
1.00
1.00
1.00
1.00
1.00
1.00
3.75
4.00
3.88
3.75
5.00
4.38
0.41
0.29
75.97
78.86
77.41
55.36
41.48
48.42
26.88
19.10
22.99
21.43
15.93
18.68
3.27
2.31
2.19
2.05
2.12
2.25
2.00
2.12
1.90
2.08
1.99
1.98
1.94
1.96
0.10
0.07
26.55
25.10
25.82
27.93
24.85
26.39
23.99
24.88
24.44
25.52
24.71
25.12
2.69
1.91
4.16
4.79
4.48
4.27
4.78
4.52
4.84
4.73
4.78
4.94
4.61
4.78
0.13
0.09
18.34
18.14
18.24
17.11
18.52
17.82
18.50
18.00
18.25
19.66
19.85
19.75
0.02
0.01
61.00
60.62
60.80
58.52
60.31
59.42
60.24
61.96
61.10
59.27
61.05
60.16
0.36
0.26
1
Lines grown in 2003 under nonirrigated conditions in Bozeman, MT. All lines are isolines created using
the hard red spring variety Hi-Line. HWC = hard wheat controls. All lines contain the Bar marker gene,
with 161 containing only Bar, and all other HG designated lines containing added PINA (HGA), PINB
(HGB), or both PINA and PINB (HGAB). Two independently created lines make up each PIN expression
group.
2
Level of PINA and PINB relative to the levels in the hard wheat control variety Hi-Line. Data on PINA
and PINB levels taken from Hogg et al. (2004).
3
Single Kernel Characterization System 4100 (Perten Instruments, Springfield, IL) grain hardness. Hard
wheats have values greater than 50 and soft wheats less than 50 units.
4
Compares individual line means.
5
Compares group means.
Transgenic PIN experiments have demonstrated that
wheat with high levels of one PIN is intermediate in
grain texture, and if both PIN are active, then soft
wheat texture results (Hogg et al., 2004). The objective
of these studies was to investigate the effect of varying
PIN content and particle size in wheat on the rate of
digestion in the rumen by using PIN isolines that varied
only in PIN content.
MATERIALS AND METHODS
Genetic Material
The hard red spring wheat cultivar Hi-Line (Lanning
et al., 1992) was used for transformation. Native HiLine contains the soft type pina sequence (pina-D/a)
and the variant pinb sequence (pinB-D1b), which contains a glycine to serine substitution at the 46th residue
of pinb (Giroux et al., 2000). Hi-Line was transformed
with vectors containing wild type pina, pinb or both
pina and pinb (pin). Pin were under the control of the
wheat glutenin promoter (Hogg et al., 2004). Isolines
with a wide range of grain textures resulted, and a
subset of the lines presented in Hogg et al. (2004) were
chosen for this study. Two lines with added pina (HGA1,
HGA3) formed the intermediate grain texture HGA
group; 2 lines with added pinb (HGB6, HGB12) formed
the soft HGB group; 2 lines with added pina and pinb
(HGAB12, HGAB18) formed the very soft HGAB group;
and 2 hard wheats [native Hi-line and Hi-Line transformed with only bar (line 161 of Hogg et al., 2004)]
formed the hard wheat controls (HWC). Seeds used in
the study were obtained from a single 4-row plot grown
during the 2003 season at the Montana State University-Bozeman Arthur H. Post Field Research Farm under dry-land conditions.
Isoline Characterization
Isolines used were analyzed for common feed and
grain characteristics (Table 1) to demonstrate that lines
were nearly identical except for the presence of the pin
transgene(s) dictating PIN content and grain hardness.
Three independent 100-seed replicate samples per line
were analyzed for kernel hardness, kernel weight, and
kernel size using the Single Kernel Characterization
System 4100 (Perten Instruments, Springfield, IL). Dry
matter content of each line × treatment was determined
using AOAC method 930.15 (2000) for oven drying and
replicated twice. Acid detergent fiber for each line was
determined using the protocol described by Van Soest
et al. (1991) and replicated twice. Crude protein for
each line was measured by AACC method 46-30 (2000)
using a LECO FP-328 nitrogen analyzer (LECO Corporation, St. Joseph, MI) and replicated 4 times. Starch
content for each line was determined using a modified
protocol of the Megazyme total starch assay kit (Megazyme International, Brey, Ireland) and was replicated
4 times.
Line × Milling Treatment Experiment
Each isoline was milled using 4 treatments. Treatments were selected to give a wide range of mean parti-
Puroindolines slow wheat starch digestion
Table 2. Geometric mean particle size, in ␮m, of 4 wheat
groups varying in puroindoline expression and milled to
4 particle sizes (line × milling treatment experiment and
time course experiment)
Milling treatment1
ID2
Fine
Medium
Coarse
Cracked
161
Hi-Line
HWC avg
HGA1
HGA3
HGA avg
HGB6
HGB12
HGB avg
HGAB12
HGAB18
HGAB avg
LSD3 (0.05)
LSD4 (0.05)
161
169
165
144
141
143
127
120
124
113
101
107
64
45
426
442
434
366
358
362
318
315
312
304
276
290
64
45
829
839
834
772
756
764
613
610
607
596
574
584
64
45
1,717
1,729
1,721
1,688
1,674
1,681
1,586
1,581
1,583
1,567
1,535
1,551
64
45
1
Fine = Perten Laboratory Mill 3303 on setting #0, Medium =
Perten Laboratory Mill 3303 on setting #3, Coarse = Perten Laboratory Mill 3303 on setting #6, and Cracked = Bühler mill on setting
11.5. Particles <90 microns were discarded.
2
Lines grown in 2003 under nonirrigated conditions in Bozeman,
MT. All lines are isolines created using the hard red spring variety,
Hi-Line. HWC = hard wheat controls. All lines contain the Bar marker
gene, with 161 containing only Bar, and all other HG-designated
lines containing added PINA (HGA), PINB (HGB), or both PINA and
PINB (HGAB). Two independently created lines make up each PIN
expression group.
3
Compares line means within a milling treatment.
4
Compares group means within a milling treatment.
cle sizes across all lines to simulate as-fed grain and
grain after mastication and rumination. Treatments
used were: cracked [Bühler mill (Bühler AG, Uzwil,
Switzerland) on setting 11.5] coarse, medium, and fine
[Perten Laboratory Mill 3303 (Perten Instruments,
Springfield, IL) settings #6, #3, and #0 with standard
grinding wheels, respectively]. Geometric mean particle size analysis was conducted according to the method
described by Baker and Herrman (2002). Forty grams
of each line was milled per treatment. After milling,
samples were placed on a series of 5 International Standards Organization sieves. Sieves used were 2,360,
1,700, 850, 425, and 90 ␮m in screen opening diameter.
The sieve stack was shaken for 5 min using a RoTap
shaker (Tyler Co., Mentor, OH). Because of the pore
size of the bags used for in situ analysis, particles <90
␮m were removed from all samples. Geometric mean
particle size (dgw) of each line × treatment was calculated on a weight basis of the geometric mean of the
diameter openings in 2 adjacent sieves in a stack using
the equation (Pfost and Headley, 1976) (dgw) = log−1 [Σ
(Wi log di) / Σ Wi] in which Wi = weight of material in
sieve i and di = diameter of the sieve i. The geometric
mean particle diameter of each line × treatment is given
in Table 2.
In situ DM digestibility (DMD) was determined using
the Vanzant et al. (1998) method. Duplicate 5-g samples
643
of each line × treatment combination were placed into
preweighed and numbered 10- × 20-cm, 50 ␮m-poresize polyester bags (Ankom Technology, Macedon, NY).
Thirty experimental bags plus 1 standard and 1 blank
were placed in the rumen of each of 2 ruminally cannulated cows fed a grain-based diet at the same time. The
2 ruminally cannulated beef cows, each consuming low
quality grass hay ad libitum and 3.6 kg/d of dry-rolled
barley, were fed for 14 d before being used for the in
situ work. Samples were incubated in the rumen. After
incubation, bags were hand washed under cold water
to stop microbial digestion. Bags were dried for 48 h at
60°C in a forced air oven. The equation DMD, g/kg =
[sample weight in * (mean DM value/100)] − (dried
sample weight out − dried blank) * 1,000/sample weight
in * (mean DM value/100), was used (Bowman et al.,
2001). Starch digestibility was determined using the
residues of samples incubated in the rumen. For each
line × treatment combination, starting starch content
was determined after milling with all particles <90 ␮m
removed by sifting over a 90-␮m screen. Final starch
content was determined after the completion of the incubation period. The equation starch digestibility, % =
[(initial starch weight − ending starch weight)/initial
starch weight] * 100, was used with all samples adjusted for DM recovery after rumen incubation.
Similar Particle Size Experiment
The hardest (Hi-Line) and the softest (HGAB18) lines
were milled such that the geometric mean particle diameter of treatments across lines was nearly identical.
This was accomplished by collecting fractions of the
milled line from each International Standards Organization sieve in the stack. Sieves used were: 3,350, 2,360,
1,700, 850, 425, and 90 ␮m in screen opening diameter.
Particle size ranges were: 2,360 to 3,349 ␮m, 1,700 to
2,359 ␮m, 850 to 1,699 ␮m, 425 to 849 ␮m, and 90 to
424 ␮m. The mills used were: fine = Perten Laboratory
Mill 3303 (Perten Instruments, Springfield, IL) on setting #0, coarse = Perten Laboratory Mill 3303 on setting
#6, and cracked = Bühler mill on setting 11.5. Dry matter digestibility and starch digestibility for each line
× particle size range combination were determined as
described in the line × milling treatment experiment.
Time Course Experiment
Two hundred grams of Hi-Line and HGAB18 were
cracked (Bühler mill on setting 11.5). Hi-Line had a
mean particle size of 1,729 ␮m, whereas HGAB18 had
a mean particle size of 1,535 ␮m. Duplicate samples
for each line × time period combination were placed in
the rumen of each of 2 ruminally cannulated cows at
the same time. Duplicate samples for each line were
removed from the rumen of each cow after 0.5, 1, 1.5,
2, 3, 4, 6, 9, and 12 h. In situ DMD and starch digestibility were determined as described for the line × milling
treatment experiment.
644
Swan et al.
Starch Granule Visualization Using Scanning
Electron Microscope
Wheat meal (100 mg) ground through a UDY mill
(0.5-mm mesh; Seedburo Equipment Co., Chicago, IL)
was placed on top of 1 mL of chloroform in a 2.0-mL
tube at 22°C. Samples were allowed to sit for 1 h with
occasional stirring of the meal with a small spatula.
After 1 h, the supernatant and suspended wheat meal
were aspirated off, leaving settled starch on the bottom
of the tube. The remaining starch granules were
washed in acetone and allowed to dry completely. A
thin layer of dried starch was attached to aluminum
electron microscope pucks with double-sided tape. The
puck was coated with gold. Images were generated with
a JEOL Model 6100 Scanning Electron Microscope
(JEOL U.S.A. Inc., Peabody, MA) at 1,000× magnification (20 kV).
Statistical Analysis
Data characterizing the initial grain samples for each
genotype were analyzed using 1-way analysis of variance. Duplicate samples for DMD and starch digestibility were averaged before analysis. The model for the
line × milling experiment was a factorial treatment
structure with factors for group, lines within group,
milling treatment and their interactions, and cows as
blocks. Data obtained for the similar particle size experiment were analyzed using a 2-factor factorial treatment structure with genotypes and particle size category as factors and cows as blocks. Analyses were accomplished using PROC GLM of SAS (SAS Inst., Inc.,
Cary, NC).
The time course experiment was analyzed using a
repeated measures model with time as the repeated
factor and a first order autoregressive covariance structure using PROC MIXED of SAS following methods
outlined by Littell et al. (1998). The DM and starch
disappearance, and rate constant (Kd) and lag time of
DM and starch disappearance were calculated as described by Bowman and Firkins (1993) using PROC
NLIN of SAS. The basic model used was from Mertens
and Loften (1980). Dry matter and starch were partitioned into 3 fractions defined as immediately soluble
(Fraction A), disappearing at a measurable rate (Fraction B), and undegradable (Fraction C). The rate constant, lag time, and fractions B and C were determined
from the nonlinear model, while Fraction A was calculated as (100 − B − C).
RESULTS AND DISCUSSION
Scanning Electron Microscope Analysis
of Starch Granules
Barlow et al. (1973) first reported that starch granules from soft and hard wheat varieties differed in the
amount of material adhering to their surface after milling with greater adherence seen in hard textured
wheats. Beecher et al. (2002) determined that the
amount of material adhered to starch was reduced by
complementing the pinb-D1b hardness mutant allele
with the wild type pinb-D1a, which also restored soft
texture. However, the appearance of super-soft wheat
starch granules containing overexpressed wild type
pina and pinb had never been examined. Figure 1 (Panels A and B) contains scanning electron microscope photos of starch granules prepared from the hard wheat HiLine and the super-soft transgenic HGAB18 (samples
described in Table 1). Line HGAB18 is a transgenic
version of Hi-Line having high expression levels of
PINA and PINB (Hogg et al., 2004). Both samples display some type A (large, oblong) and B (smaller, round)
starch granules. However, the amount of material adhering to the surface of type A granules is dramatically
different between the samples. Hi-Line granules (Figure 1A) are clumped together along with protein bodies,
oblong, and rough in texture with cracks on the surface
of the large granules. Line HGAB18 granules (Figure
1B) are single and discrete, smooth on the surface, and
have virtually no adhering B granules or protein bodies
associated with the type A granules.
Using scanning electron microscope analysis, the protein matrix of corn was observed to limit access of ruminal bacteria to starch granules (McAllister et al., 1991).
Comparing corn and barley ruminal starch digestion,
McAllister et al. (1993) concluded that structural components associated with or within the endosperm were
responsible for the differences in starch digestion. To
determine if physical differences in starch granules
during digestion could be seen between hard and soft
wheats, Hi-Line and HGAB18 were incubated for 4 h
in the rumen and then prepared for scanning electron
microscope analysis. Hi-Line granules (Figure 1C) after
incubation were large and clumped together with the
protein matrix, had deep type A pitting, and showed
signs of digestion in all type A granules. Line HGAB18
granules (Figure 1D) after incubation were individual,
had shallow type A pitting, and showed signs of digestion in only about 1/4 of type A granules. Starch granules from the soft seeds of HGAB18 appeared more
resistant to microbial digestion than the starch granules from the hard textured Hi-Line control variety.
The differences seen after digestion may be a result of
greater damage to the surface of starch granules in HiLine vs. HGAB18. Soft wheats fracture easily, resulting
in less starch damage after milling than hard wheats
(Symes, 1965). The findings here support current thinking that the starch granule surface is likely the site of
functional differences between soft and hard wheats.
The lack of interaction of the protein matrix with starch
granules in HGAB18 indicates that PIN directly control
grain softness by reducing the interaction between
starch granules and their surrounding protein matrix.
Line × Milling Treatment Experiment
To examine the effects of added PIN and grain hardness upon DMD and starch digestibility, we chose a
Puroindolines slow wheat starch digestion
645
Figure 1. Scanning electron microscope analysis of purified starch granules taken at 1000× magnification. A) HiLine, hard wheat control. Granules are oblong and rough in texture with cracks on the surface of the large granules.
B) HGAB18, super-soft wheat. Granules are single, discrete, and smooth on the surface. C) Hi-Line after incubating
in the rumen for 4 h. Granules are clumped together and show widespread signs of pitting from microbial digestion.
D) HGAB18 after incubation in the rumen for 4 h. Granules show less severe and less abundant pitting than Hi-line.
Arrows in C) and D) mark pitting of starch granules.
subset of the samples examined by Hogg et al. (2004;
Table 1). All transgenic lines were made using explant
material from the hard red spring variety Hi-Line. The
first group (HWC) consisted of the control variety HiLine and a Bar control gene only line termed 161. The
HGA group consisted of 2 PINA overexpressing lines,
and the HGB group contained 2 PINB overexpressing
lines. The final group was HGAB, which consisted of 2
transgenic lines, each with high levels of PINA and
PINB. The 4 groups with 2 lines per group and with
varying expression of the PIN proteins were milled using 4 milling treatments. The 4 milling treatments resulted in average particle sizes after milling that were
smaller (P < 0.05) for the medium and coarse milling
treatments (Table 2) for the HGA, HGB, and HGAB
groups relative to the HWC group. Subsamples of each
milling treatment sample were incubated in the rumen
for 3 h. Lines within group source of variation was not
significant for DMD (P = 0.378) or starch digestibility
(P = 0.167), indicating the 2 lines within a group were
similar. Milling treatments did interact with groups for
DMD (P = 0.002) and starch digestibility (P = 0.002),
but milling treatments did not show interactions with
lines within group for either DMD (P = 0.879) or starch
digestibility (P = 0.977). Ruminal DMD and starch digestibility milling treatment × group combination
means are presented in Table 3. The HWC and added
PINA groups had the highest DMD values across all
milling treatments and were not different from each
other except for the coarse milling treatment (P =
0.002). The HGB and HGAB groups were lower (P =
0.001) in DMD than HWC and HGA groups across all
milling treatments. The HGAB group tended to be lower
than the HGB group, but that difference reached statistical significance (P = 0.002) only in the medium milling
treatment. Results for starch digestibility generally
mirrored those for DMD with starch digestibility declining as particle size increased. The HWC group had the
646
Swan et al.
Table 3. In situ DM disappearance (DMD) and starch disappearance of 4 wheat groups
varying in puroindoline expression and milled to 4 particle sizes (line × milling treatment experiment)
Item
No. of observations1
Fine2
Medium
Coarse
Cracked
HWC3
HGA
HGB
HGAB
HWC
HGA
HGB
HGAB
HWC
HGA
HGB
HGAB
HWC
HGA
HGB
HGAB
SE
Particle size P-value
PIN expression P-value
Line (PIN expression) P-value
Particle size × PIN P-value
Particle size × line (PIN) P-value
DMD,
%
Starch
disappearance,
%
128
70.42z
67.06y
59.75x
58.15wx
67.73yz
66.29y
57.67wx
49.43v
58.90x
54.94w
40.45u
39.46u
33.78t
31.59t
23.78s
20.42r
1.369
0.001
0.001
0.378
0.002
0.879
64
96.17y
94.48y
82.03x
74.87w
94.70y
92.83y
77.67wx
63.75v
84.81x
76.69w
58.35u
55.02tu
51.16t
45.76s
37.02r
34.10r
2.012
0.001
0.001
0.167
0.002
0.977
Within a column, means without a common superscript letter differ (P < 0.05).
DMD values are the mean of 8 replications and starch digestibility values are the mean of 4 replications
per particle size × PIN treatment.
2
Particle sizes: Fine = Perten Laboratory Mill 3303 on setting #0, Medium = Perten Laboratory Mill 3303
on setting #3, Coarse = Perten Laboratory Mill 3303 on setting #6, and Cracked = Bühler mill on setting
11.5.
3
Lines grown in 2003 under nonirrigated conditions in Bozeman, MT. All lines are isolines created using
the hard red spring variety Hi-Line. HWC = hard wheat controls. Lines designated HG contain added PINA
(HGA), PINB (HGB), or both PINA and PINB (HGAB). Two independently created lines make up each PIN
expression group.
r–z
1
greatest (P = 0.001) starch digestibility followed by the
HGA, HGB, and HGAB groups.
The HGAB group showed increased expression of
both PINA and PINB, the HGB group had increased
expression of PINB, and the HGA group had increased
expression of PINA. As a result, the HGAB group had
a super-soft texture, the HGB group had a soft texture,
the HGA group had an intermediate texture, and the
HWC had a hard texture (Table 1). The HGA group
had a 7.5× increase in PINA and a 28-unit decrease in
hardness when compared with the HWC group. The
increase in PINA decreased DMD an average of 5.0%
and starch digestibility an average of 5.9% across all
treatment levels. The HGB group had a 3.9× increase
in PINB and a 54.4 unit decrease in hardness when
compared with the HWC group. The increased expression of PINB decreased DMD an average of 22.7% and
starch digestibility an average of 22.9% across all treatment levels. The HGAB group had a 6.5× increase in
PINA, a 4.4× increase in PINB, and a 58.7 unit decrease
in grain hardness when compared with the HWC group.
The increased expression of PIN decreased DMD an
average of 29.2% and decreased starch digestibility an
average of 30.8% across all treatment levels.
The line × milling treatment experiment indicated
that PIN content affected the rate of wheat starch digestion in the rumen. Increased expression of PINB and
both PINA and PINB led to a significant reduction in
DMD and starch digestibility across all milling treatments. The largest reduction was achieved by the addition of PINA and PINB. A potential complicating factor
in this study is the effect of grain texture on particle
size. Soft wheats, having a softer endosperm, fracture
easily, requiring less energy to mill than hard wheats
(Symes, 1965, 1969). As a result, soft wheats yield
smaller particles on the same mill setting, suggesting
that the effect of PIN upon digestibility may reflect both
particle size variation and PIN expression variation.
Similar Particle Size Experiment
To separate the effect of particle size and PIN content
on DMD and starch digestibility, Hi-Line (hardest) and
HGAB18 (softest) were milled such that the mean parti-
647
Puroindolines slow wheat starch digestion
Table 4. In situ DM disappearance (DMD) and starch disappearance of 2 wheat lines
varying in puroindoline expression and milled to specific particle size ranges
Item
No. of observations1
Hi-Line2
HGAB18
0.09 to 0.424 mm3
0.425 to 0.849 mm
0.85 to 1.69 mm
1.7 to 2.35 mm
>2.36 mm
0.09 to 0.424 mm
0.425 to 0.849 mm
0.85 to 1.69 mm
1.7 to 2.35 mm
>2.36 mm
Line P -value
Particle size P-value
Line × particle size P-value
SE
DMD,
%
Starch
disappearance,
%
40
68.59z
65.81z
59.44y
33.42v
2.78t
62.28y
52.23x
42.49w
23.93u
2.74t
0.001
0.001
0.001
1.159
20
94.69z
94.42z
85.31y
54.66v
24.27t
88.90y
73.04x
62.48w
41.54u
24.47t
0.001
0.001
0.001
1.552
Within a column, means without a common superscript letter differ (P < 0.05).
DMD values are the mean of 8 replications, and starch digestibility values are the mean of 4 replications
per line × particle size treatment.
2
Genotypes were the hard red spring variety Hi-Line and HGAB18, which is Hi-Line transformed to have
high levels of PINA and PINB expression.
3
Particle size ranges were adjusted to the stated size ranges by sifting after milling using the fine (for
0.09 to 0.424 mm), coarse (for 0.425 to 0.849 and 0.85 to 1.69 mm), and cracked (for 1.7 to 2.35 mm) mill
settings. The mills used were: fine = Perten Laboratory Mill 3303 (Perten Instruments, Springfield, IL) on
setting #0, coarse = Perten Laboratory Mill 3303 on setting #6, and cracked = Bühler mill on setting 11.5.
t–z
1
cle size per treatment was nearly identical, and incubated in the rumen for 3 h. Dry matter digestibility
and starch digestibility results are presented in Table 4.
A line × particle size treatment interaction was detected
(P = 0.001) indicating that lines did not react similarly
across particle size treatments. Line HGAB18 had
lower DMD than Hi-Line (P = 0.001) among particles
ranging in size from 0.09 ␮m to 2.35 ␮m. No differences
(P = 0.94) in DMD were seen in particles above 2.36
␮m in size. Starch digestibility declined with increasing
particle size, but HGAB18 also had lower starch digestibility than Hi-Line (P = 0.001) at particles sizes ranging from 0.09 ␮m to 2.35 ␮m. No differences (P = 0.85)
in starch digestibility were seen in particles above 2.36
␮m in size. For wheat, particles above 2.36 ␮m are
generally whole kernels. Seeing no differences between
whole kernels of Hi-Line and HGAB18 indicated that
PIN proteins had no affect on the seed coat or aleurone
layers of the kernel and that differences in DMD and
starch digestibility for small particles was a function
of PIN interaction with starch. The increased PINA
and PINB expression in HGAB18 decreased DMD an
average of 21.7% and starch digestibility an average of
19.9% across particle sizes smaller than whole kernel
(<2.36 mm). This experiment indicated that increased
PIN expression decreased DMD and starch digestibility, and was largely independent of particle size.
Time Course Experiment
To investigate the effects of time in the rumen on
PIN proteins, Hi-Line and HGAB18 were cracked, and
disappearance was observed at various time points over
a 12-h period in the rumen. Dry matter and starch
digestibility results are presented in Figure 2 and Table
5. Dry matter digestibility increased over time for both
lines. Line HGAB18 was consistently lower in DMD
than Hi-Line across all time points. However, the difference between lines became less with time after 4 h
leading to a line × time interaction (P = 0.001). Similarly, starch digestibility increased over time with
HGAB18 being lower than Hi-Line. Again, the difference between HGAB18 and Hi-Line became less with
time giving rise to a line × time interaction (P = 0.01).
Increased expression of PINA and PINB in HGAB18
decreased DMD by 10.4% and starch digestibility by
11.0% across all time points.
Line HGAB18 had a 27.9% lower (P = 0.001) immediately soluble fraction of DM than did Hi-Line, and a
24.5% slower (P = 0.003) DM disappearance rate (Table
5). The immediately soluble fraction of starch tended
(P = 0.118) to be lower for HGAB18 than for Hi-Line,
and HGAB18 had a 30.2% slower (P = 0.003) starch
disappearance rate. In addition, HGAB18 had a lower
(P = 0.050) undegraded starch fraction compared with
Hi-Line, indicating that slowing starch digestibility via
increasing PIN expression will not necessarily reduce
extent of starch digestion.
The rate of passage of feedstuffs out of the rumen is
affected by specific gravity and particle size. Grains,
although rapidly fermented, are small and dense, and
thus pass rapidly compared with roughages (Firkins et
al., 2001). The time course experiment demonstrated
that a difference in starch digestibility could still be
648
Swan et al.
Figure 2. In situ dry matter (DM) and starch disappearance (0 to 12 h) of Hi-Line (䊉) and HGAB18 (䊏), 2 wheat
lines varying in puroindoline expression. A) DM disappearance (%), and B) starch disappearance (%). Wheat line ×
time interactions were significant for DM (P = 0.002) and starch disappearance (P = 0.006). Within a time period,
means without a common superscript letter differ (P < 0.05).
seen after 6 h in the rumen, indicating that starch
supply to the small intestine may be increased with
increased PIN levels. Differences in starch digestibility
after incubation suggest that starch granules of
HGAB18 possess a type of protection that Hi-Line does
not. Being isogenic lines, expression of PIN proteins is
the only difference between Hi-Line and HGAB18. The
time course experiment led to the conclusion that PIN
649
Puroindolines slow wheat starch digestion
Table 5. Characteristics of ruminal in situ disappearance (0 to 12 h) of DM and starch of
2 wheat lines varying in puroindoline expression (time course experiment)
Item
DM disappearance
Fraction A,2 g/kg
Fraction B,3 g/kg
Fraction C,4 g/kg
Kd,5 h−1
Lag time, h
Starch disappearance
Fraction A, g/kg
Fraction B, g/kg
Fraction C, g/kg
Kd, h−1
Lag time, h
Hi-Line1
HGAB18
SEM
P-value
11.1
68.9
20.0
0.269
0
8.0
75.3
16.6
0.203
0
0.41
1.35
1.67
0.0099
—
0.001
0.014
0.204
0.003
—
11.3
84.1
4.6
0.368
0
8.1
87.9
2.2
0.257
0
1.24
1.64
0.69
0.0165
—
0.118
0.146
0.050
0.003
—
1
Hi-Line = untransformed hard red spring control wheat; HGAB18 = Hi-Line transformed to overexpress
PINA and PINB and have very soft endosperm texture.
2
Fraction A: immediately soluble fraction.
3
Fraction B: fraction disappearing at a measurable rate.
4
Fraction C: undegraded fraction.
5
Kd: disappearance rate.
proteins aid in the protection of starch molecules from
ruminal fermentation.
Conclusions
Comparisons between cereal species have shown that
wheat starch is fermented rapidly in the rumen when
compared with barley, maize, and sorghum (HerreraSaldana et al., 1990; Owens et al., 1997). Research also
has shown that variations in starch digestion exist between cultivars of the same species (Philippeau et al.,
1999; Bowman et al., 2001). However, previous research
has relied on the use of grains within the same market
class or named varieties for comparison. Genetic factors
controlling starch digestion cannot be determined by
such comparisons due to the diversity of genetic backgrounds. Little work has been done with near-isogenic
or transformed isolines. The use of such genetic material allows the nutritional implication of specific traits
to be evaluated against a consistent genetic background. At least 2 studies in wheat have been done
with near-isogenic lines. Short et al. (2000) studied the
effects of grain hardness on AA digestion in poultry.
They indicated that hard wheat endosperm was associated with decreased AA digestibility. Chickens, being
a nonruminant animal, cannot be easily compared with
ruminants and generally are considered to be particularly sensitive to changes in quality of the diet. Garnsworthy and Wiseman (2000) used near isogenic lines to
evaluate the ruminal digestibility of wheat starch; no
differences were seen between hard and soft wheats.
However, actual grain hardness and PIN content were
not stated. Typical soft wheats have at most twice the
PIN content of typical hard wheats (Giroux and Morris,
1997; 1998). One may therefore expect that differences
due to PIN content are too small to be detected among
native wheat varieties.
Overall, our present experiment demonstrated that
PIN proteins affected DM and starch digestibility of
wheat in the rumen. Decreasing wheat grain hardness
by increasing PIN expression slowed DMD and starch
digestibility in the rumen and was largely independent
of particle size. Data indicated that PIN proteins aid
in the protection of starch molecules from fermentation
type digestion in the rumen. In barley, low in situ DMD
values are correlated with increased feed efficiency, increased ADG, and increased NE content (Bowman et
al., 2001). Slower or lower ruminally digestible starch
shifts more starch digestion from the rumen to the small
intestine. Starch digestion in the small intestine theoretically could provide up to 42% more energy than
starch fermented in the rumen (Owens et al., 1986)
because of reduction in energy loss via methane production and more efficient use of glucose as an energy
source compared with VFA. In addition, lower DMD
could reduce excessive fermentation acid production
and reduce the incidence of bloat, acidosis, and laminitis (Hunt, 1996).
IMPLICATIONS
Although wheat is not a predominant cattle feed in
the United States, it is an ideal model system to study
the effect of grain hardness and puroindoline content
on digestion in the rumen. In wheat, the puroindoline
genes control the majority of grain hardness variation.
Isolines that differed only in puroindoline content illustrated that the presence of additional puroindoline proteins slowed the digestion of starch. Using current
transformation methods, it should be possible to reduce
the rate of starch digestion in other cereals such as
barley via puroindoline overexpression. Barley is an
attractive cereal in which to conduct further research
because no true soft barleys seem to exist, nor are there
650
Swan et al.
any barley varieties that have soft wheat levels of friabilin bound to starch. Perhaps addition of puroindoline
proteins to highly fermentable feeds such as oats and
barley would decrease the rate of digestion and increase
feed efficiency and beef cattle performance.
LITERATURE CITED
AACC. 2000. Approved Methods of the American Association of Cereal
Chemists. 10th ed. Am. Assoc. Cereal Chem., St. Paul, MN.
AOAC. 2000. Official Methods of Analysis. 17th ed. Assoc. Offic. Anal.
Chem., Gaithersburg, MD.
Baker, S., and T. Herrman. 2002. Evaluating Particle Size. Kansas
State Univ. Ag. Exp. Sta. Bull. No. MF-2051, Manhattan, KS.
Barlow, K. K., M. S. Buttrose, D. H. Simmonds, and M. Vesk. 1973.
The nature of the starch-protein interface in wheat endosperm.
Cereal Chem. 50:443–454.
Beecher, B., A. Bettge, E. Smidansky, and M. J. Giroux. 2002. Expression of wild-type pinB sequence in transgenic wheat complements a hard phenotype. Theor. Appl. Genet. 105:870–877.
Bowman, J. G. P., T. K. Blake, L. M. M. Surber, D. K. Habernicht,
and H. Bockelman. 2001. Feed-quality variation in the barley
core collection of the USDA National Small Grains Collection.
Crop Sci. 41:863–870.
Bowman, J. G. P., and J. L. Firkins. 1993. Effects of forage species and
particle size on bacterial cellulolytic activity and colonization in
situ. J. Anim. Sci. 71:1623–1633.
Firkins, J. L., M. L. Eastridge, N. R. St-Pierre, and S. M. Noftsger.
2001. Effects of grain variability and processing on starch utilization by lactating dairy cattle. J. Dairy Sci. 79(E. Suppl.):E218–E238.
Garnsworthy, P. C., and J. Wiseman. 2000. Rumen digestibility of
starch and nitrogen in near-isogenic lines of wheat. Anim. Feed
Sci. Technol. 85:33–40.
Gautier, M. F., M. E. Aleman, A. Guirao, D. Marion, and P. Joudrier.
1994. Triticum aestivum puroindolines, two basic cystine-rich
seed proteins: cDNA sequence analysis and developmental gene
expression. Plant Mol. Biol. 25:43–57.
Giroux, M. J., and C. F. Morris. 1997. A glycine to serine change in
puroindoline b is associated with wheat grain hardness and low
levels of starch-surface friabilin. Theor. Appl. Genet. 95:857–
864.
Giroux, M. J., and C. F. Morris. 1998. Wheat grain hardness results
from highly conserved mutations in the friabilin components
puroindoline a and b. Proc. Natl. Acad. Sci. USA 95:6262–6266.
Giroux, M. J., L. Talbert, D. K. Habernicht, S. Lanning, A. Hemphill,
and J. M. Martin. 2000. Association of puroindoline sequence
type and grain hardness in hard red spring wheat. Crop Sci.
40:370–374.
Herrera-Saldana, R. E., J. T. Huber, and M. H. Poore. 1990. Dry
matter, crude protein, and starch degradability of five cereal
grains. J. Dairy Sci. 73:2386–2393.
Hogg, A. C., T. Sripo, B. Beecher, J. M. Martin, and M. J. Giroux.
2004. Wheat puroindolines interact to form friabilin and control
wheat grain hardness. Theor. Appl. Genet. 108:1089–1097.
Hunt, C. W. 1996. Factors affecting the feeding quality of barley for
ruminants. Anim. Feed Sci. Technol. 62:37–48.
Lanning, S. P., L. E. Talbert, F. H. McNeal, W. L. Alexander, C. F.
McGuire, H. Bowman, G. Carlson, G. Jackson, J. Eckhoff, G.
Kushnak, V. Stewart, and G. Stallknecht. 1992. Registration of
‘Hi-Line’ wheat. Crop Sci. 32:283–284.
Littell, R. C., P. R. Henry, and C. B. Ammerman. 1998. Statistical
analysis of repeated measures data using SAS procedures. J.
Anim. Sci. 76:1216–1231.
McAllister, T. A., R. C. Phillippe, L. M. Rode, and K. J. Cheng. 1993.
Effect of the protein matrix on the digestion of cereal grains by
ruminal microorganisms. J. Anim. Sci. 71:205–212.
McAllister, T. A., L. M. Rode, K. J. Cheng, and C. W. Forsberg. 1991.
Selection of a sterilization method of the study of cereal grain
digestion. J. Anim. Sci. 69:3039–3043.
Mertens, D. R., and J. R. Loften. 1980. The effect of starch on forage
fiber digestion kinetics in vitro. J. Dairy Sci. 63:1437–1446.
Morris, C. F., M. Lillemo, G. M. Simeone, M. J. Giroux, S. L. Babb, and
K. K. Kidwell. 2001. Prevalence of puroindoline grain hardness
genotypes among historically significant North American spring
and winter wheats. Crop Sci. 41:218–228.
Owens, F. N., D. S. Secrist, W. J. Hill, and D. R. Gill. 1997. The effect
of grain source and grain processing on performance of feedlot
cattle: A review. J. Anim. Sci. 75:868–879.
Owens, F. N., R. A. Zinn, and Y. K. Kim. 1986. Limits to starch
digestion in the ruminant small intestine. J. Anim. Sci.
63:1634–1648.
Pfost, H., and V. Headley. 1976. Methods of determining and expressing particle size. Page 517 in Feed Manufacturing Technology
II. H. Pfost, ed. Anim. Feed Manuf. Assoc., Arlington, VA.
Philippeau, C., F. Le Deschault de Monredon, and B. Michalet-Doreau. 1999. Relationship between ruminal starch degradation
and the physical characteristics of corn grain. J. Anim. Sci.
77:238–243.
Short, F. J., J. Wiseman, and K. N. Boorman. 2000. The effect of
the 1B/1R translocation and endosperm texture on amino acid
digestibility in near-isogenic lines of wheat for broilers. J. Agric.
Sci. 134:69–76.
Symes, K. J. 1965. The inheritance of grain hardness in wheat as
measured by particle size index. Aust. J. Agric. Res. 16:113–123.
Symes, K. J. 1969. Influence of a gene causing hardness on the milling
and baking quality of two wheats. Aust. J. Agric. Res.
20:971–979.
Van Soest, P. J., J. B. Robertson, and B. A. Lewis. 1991. Methods
for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci.
74:3583–3597.
Vanzant, E. S., R. C. Cochran, and E. C. Titgemeyer. 1998. Standardization of in situ techniques for ruminal feedstuff evaluation. J.
Anim. Sci. 76:2717–2729.