| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL NUTRITION |
* Department of Animal and Veterinary Science, University of Idaho, Moscow 83844;
and
Beef Northwest Feeders, Nyssa, OR 97913;
and
Pioneer Hi-Bred International, Johnston, IA 50131; and and
Koers-Turgeon Consulting Services Inc., Amarillo, TX 79188
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Key Words: in situ • intestine • rumen • starch • steer
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
). Moreover, starch is more digestible in the rumen and the total digestive tract from high-moisture than from dry-rolled corn (Owens et al., 1986). Although corn hybrids differ minimally in starch concentration (B. Mahanna, Pioneer Hi-Bred International, personal communication), they differ markedly in starch vitreousness (Majee et al., 2003). Furthermore, greater kernel vitreousness has been associated with less rapid ruminal (in situ) starch disappearance (Philippeau and Michalet-Doreau, 1997). Based on a survey of 14 widely diverse corn types, Philippeau et al. (1999a) proposed that 88.5% of the variation in ruminal (in situ) starch disappearance could be attributed to kernel vitreousness.
Maturity at harvest can also affect vitreousness and in situ fermentation of starch from corn. Comparing 2 corn hybrids at 2 maturities, Philippeau and Michalet-Doreau (1997) noted that vitreousness differed between dent and flint genotypes (26.5 vs. 38.3%), but vitreousness differed even more between immature and mature grains (32 vs. 60.2%). Correa et al. (2002) reported that with advancing maturity of a dent genotype, kernel vitreousness increased, and ruminal in situ disappearance decreased. Macken et al. (2003) indicated that vitreousness might interact with moisture content of corn, noting that feed efficiency was poorer from a flinty hybrid than from a floury hybrid when corn was dry-rolled; however, when fed as high-moisture corn, feed conversion for steers fed these 2 grains were not statistically different. The extent to which endosperm vitreousness interacts with kernel moisture at harvest to affect starch digestibility is not known.
Therefore, the objective of our study was to characterize the influence of endosperm vitreousness and kernel moisture on digestion of high-moisture corn by feedlot steers.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
) and the ratio of hard to soft endosperm (Li et al., 1996). Hybrids were harvested on 3 dates, yielding grain at 3 moisture levels (DRY, MID, and WET). Grain was harvested with a John Deere (Moline, IL) model 9610 (8-row) combine. At harvest, the FLO corn contained 27.7, 32.5, and 36.5% moisture, whereas the VIT corn harvested on the same dates contained 28.5, 30.0, and 35.5% moisture for DRY, MID, and WET harvest dates, respectively. Visual assessment of harvested grain revealed that kernels on all harvest dates had reached physiological maturity (e.g., presence of a black abscission layer at the tip of the kernel). The MID moisture level was intended to represent an ideal moisture (e.g., 30 to 32%) for harvest and storage of high-moisture corn grain, whereas the DRY and WET harvest times were an attempt to bracket the kernel moisture range typically harvested at dairies or feedlots. In retrospect, including a lower moisture level or even dry-rolled corn from these 2 hybrids would have been useful for comparative purposes to confirm that digestibility is lower for dry corn that is more vitreous, as has been reported in trials by others. Of special interest for comparison with corn silage, grain moisture levels tested in this study all were drier than for grain harvested with corn silage; with whole-plant corn silage harvested at approximately 46% DM, corn grain moisture typically is approximately 36%. Within 3 h after harvest, the grain was rolled through a mill (Farm King model Y 100; Buhler Industries Inc., Winnipeg, Manitoba, Canada) with a roller gap maintained to fracture every kernel. Rolled corn was packed in 1.0 x 1.25 x 1.5 m cardboard boxes equipped with plastic liners and sealed. High-moisture corn treatments were allowed to ferment for a minimum of 45 d before being fed; feeding studies lasted 126 d. Subsamples of high-moisture corn obtained during each experimental period (described later) were lyophilized for particle size analysis (described later).
Metabolism Study
Animals, Feeding, and Experimental Timeline.
Experimental procedures involving animals were approved by the University of Idaho Institutional Animal Care and Use Committee. Six ruminally and duodenally cannulated Angus-Jersey crossbred steers (450 kg of BW), which had been used in a previous study (Szasz et al., 2005), were fed the 6 test grains to examine the effects of kernel vitreousness and moisture at harvest on intake and digestibility of high-moisture corn. Steers were housed in individual pens (approximately 25 m2) and had free choice access to fresh water. Pens had concrete floors covered with rubber mats, with half of each pen having an overhead shelter. Solid waste was removed from each pen daily. Animals were fed a concentrate diet with 88.25% of diet DM being high-moisture corn. Diets also contained (DM basis) 6.0% chopped alfalfa hay, 2.0% corn gluten meal, 0.75% urea, and 3.0% supplement. Diets were formulated to contain a minimum of 12.5% CP, 0.6% Ca, 0.6% K, 0.2% S, 33 mg/kg of monensin (Elanco Animal Health, Indianapolis, IN), and 11 mg/kg of tylosin (Elanco Animal Health). To maintain the correct proportions of ingredients in the diet, DM content of each of the 6 corns was assayed daily; other dietary ingredients were assayed weekly. Diets were prepared daily as a total mixed ration at 0600 and were offered in 3 equal portions at 0700, 1100, and 1500 to simulate commercial feedyard management practices. Chromic oxide (15 g/d) was mixed thoroughly into each diet daily as an external marker for calculating digestibility. Feed offered, diet DM, and feed refusals were recorded daily. Periods within the experiment lasted 21 d, with steers being adapted to their diets for 10 d (d 1 to 10) and voluntary DMI being monitored for 7 d (d 11 to 17). The final 4 d of each period were intended for collection of ruminal fluid, duodenal and fecal samples, and in situ degradation measurements.
For particle size analysis, samples of high-moisture corn were obtained daily during each collection period and were stored at – 20 ° C. Later, samples were composited within period according to hybrid-kernel moisture subclass, thereby yielding a total of 36 samples for analysis (i.e., 6 samples/treatment). Samples were lyophilized and dry-sieved using an analytical throw-action sieve shaker (Retsch model AS200; Rheinische, Germany) equipped with USA standard testing sieves (4,750, 3,350, 2,000, and 850-µm holes) arranged in descending order above a collection pan. Sieving continued for 10 min. Weight over size generated during the dry-sieving procedure was used to compute the geometric mean diameter (dgw) and SD (Sgw) of the particles according to the equations described by Baker and Herrman (2002):
 |
where di = the geometric mean diameter and Wi = the weight fraction on the ith sieve.
Surface area (SA) of the particles was estimated using the equation:
 |
where ßs = the shape coefficient for calculating surface area of particles (fixed at 6); ßv = the shape coefficient for calculating volume of particles (fixed at 1); and 
= the specific weight of material (fixed at 1.32; Pfost and Headley, 1976).
In Situ Disappearance.
Samples of each of the 6 high-moisture corn treatments obtained at the beginning of each period and stored at – 20 ° C were used to determine the in situ disappearance. Approximately 10 g of DM from fresh (nondried, unground) high-moisture corn was placed in 5 x 10 cm Dacron bags (pore size 53 µm; Ankom Corp., Fairport, NY). Duplicate bags were inserted into the rumen at specific time intervals on d 20 of each experimental period for simultaneous removal, such that ruminal incubations lasted 2, 4, 6, 8, 12, 18, 24, and 48 h. Zero-hour bags were soaked in cold water for 15 min and were rinsed until the rinse water ran clear to determine the quantity of high-moisture corn that was present as soluble material and as small particles. After ruminal incubation, bags were washed and dried as described above. Incubation bag residues, composited within period, steer, and time, were ground through a mixer mill (mixer mill MM 200, Retsch, Haan, Germany) and analyzed for starch using methods described later. In situ starch and DM disappearance were calculated as the amount of starch or DM that disappeared during washing plus ruminal incubation.
Digestibility.
Six fecal and duodenal digesta samples were obtained on d 19 through 21 of each experimental period, such that every 4 h of a 24-h feeding cycle was represented. Fecal and duodenal digesta samples, composited by animal within period, were frozen at – 20 ° C. Fecal composites were later transferred to a forced-air oven and dried at 55 ° C for 72 h; dried samples were ground through a 2-mm screen of a Udy cyclone mill (Udy Corp., Fort Collins, CO). Composited duodenal samples were thawed, mixed thoroughly, and a sub-sample was obtained. Subsamples were lyophilized and subsequently homogenized using a mortar and pestle. Ground fecal and duodenal digesta samples were analyzed for DM and ash (AOAC, 1990), total starch by an assay kit (Megazyme International Ireland Ltd., Wicklow, Ireland), and Cr (Williams et al., 1962) by inductively coupled plasma. Representative samples of feed ingredients were obtained during each experimental period. Samples were dried in a forced-air oven at 55 ° C for 72 h; dried samples were ground through a 2-mm screen of a Udy cyclone mill and analyzed for DM, ash, and starch, as previously described. Nutrient digestibilities were calculated using ratios of diet and fecal analyzed components to Cr (Merchen, 1988).
Ruminal Fermentation.
Ruminal fluid samples (200 mL) were collected at 0300, 0700, 0900, 1100, 1300, 1500, 1900, 2100, and 2300 on d 18 of each experimental period. Ruminal fluid samples were strained through 2 layers of cheesecloth, and their pH was measured immediately with a pH meter equipped with a glass electrode (Orion Research, Boston, MA). Ruminal fluid pH was obtained by submersing the pH combination electrode (model SA720, Orion Research) directly into strained ruminal liquor. A numerical integration method was used to estimate the number of hours during which the ruminal fluid pH fell below a specified value; computations were made using a DATA step macro in SAS (SAS Inst. Inc., Cary, NC; Szasz, 2006). After the pH measurement, aliquots were obtained, acidified (1 mL of 9 M H2SO4 per 100 mL of ruminal fluid) and frozen at – 20 ° C. Subsequently, the sample was thawed, and an aliquot was centrifuged (20,000 x g for 15 min) and analyzed for NH3 (Broderick and Kang, 1980) with a spectrophotometer (Lachat AE Quikchem, Milwaukee, WI). Another fresh aliquot (10 mL) was added to a 15-mL sample cup containing 2 mL of 25% (wt/vol) meta-phosphoric acid, and the mixture was stirred vigorously and frozen. Later, it was centrifuged (20,000 x g for 15 min) and assayed for VFA by GLC (Supelco, 1998; GS model 6890, Hewlett Packard, Palo Alto, CA).
Computations and Statistical Analyses.
All data were analyzed using SAS. Using the NLIN procedure, in situ DM and starch disappearance values were fitted to the equation of Orskov and McDonald (1979):
 |
where p = DM or starch that had disappeared (%) at time t; a = the rapidly soluble fraction (%); b = the potentially degraded fraction (%); c = the rate at which b is degraded (%/h); and t = the incubation time (h).
Effective ruminal DM or starch degradability (ED) was calculated using the equation:
 |
where a, b, and c = constants from the nonlinear model described previously, whereas k = fractional ruminal outflow rate that was assumed to be 5.0%/h (Orskov and McDonald, 1979). A 5.0%/h outflow rate has been used to estimate effective degradability of feedstuffs in previous trials (Hristov and McAllister, 2002; Szasz et al., 2005).
Response variables were analyzed as a 6 x 6 Latin square with a 2 x 3 factorial arrangement using the MIXED procedure. The statistical model was:
 |
where µ = the overall mean; 
i = the random effect of steer (i = 1 to 6); ßj = the fixed effect of period (j = 1 to 6); 
k = the fixed effect of hybrid; 
l = the fixed effect of kernel moisture at harvest; kl = the fixed effect interaction term; and 
ij(kl) = the residual error term that was assumed to be distributed normally.
To examine the response surface for kernel moisture at harvest, the 2 df for kernel moisture, as well as the 2 df for the interaction of kernel moisture with vitreousness, were partitioned into linear and quadratic effects. Lamba coefficients were computed for unequally spaced levels of kernel moisture using the IML procedure. A mixed effects model (using the MIXED procedure) was also used to analyze repeated ruminal fluid measures obtained over time. Effects contained within the previous model were used together with a sampling time effect and the resulting interactions with treatment factors. Multiple covariance matrix structures were assessed for the data using an iterative method; the best-fitting structure was selected based on the Bayesian information criterion.
|
RESULTS AND DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
). In addition, GMD increased (P = 0.04) linearly with kernel moisture content. Geometric SD was also greater (P = 0.10) for VIT than FLO, reflecting greater variability in size of VIT corn particles. Calculated surface area averaged 15.8% more (P = 0.03) for VIT than for FLO corn particles. Philippeau and Michalet-Doreau (1998) measured GMD of fresh kernels obtained from whole-plant corn harvested at 30% moisture and chopped in a blender before ensiling. Geometric mean diameter of their processed grain (4,133 and 3,941 µm for dent and flint corn, respectively) was within the range of GMD of rolled corn grain used in the current study (3,456 to 4,440 µm) and indicates that, in contrast with dry corn, more vitreous grain, when processed as high-moisture corn, may generate processed grain with a smaller mean particle. Using dry grain, Philippeau et al. (1999b) processed flint and dent dry corn grain through a hammer mill equipped with a 3-mm screen. After being processed, flint hybrids that contained a greater proportion of vitreous endosperm had 12.4% more coarse particles (>4,000 µm) than dent hybrids; particles from processed flint corn had 46.2% less surface area than particles from dent corn. This indicates that when processed dry, more vitreous grain tended to be more resistant to particle size reduction when being rolled, contrary to the results with high-moisture grain that we observed and Philippeau and Michalet-Doreau (1998) also noted with high-moisture corn grain. Although effects of endosperm type on particle size of processed, high-moisture corn have not been studied extensively, these results suggest that, compared with more vitreous kernels, floury corn kernels, when moist, are more pliable and less damaged when rolled. In contrast, vitreous corn kernels in the form of high-moisture corn appeared brittle and shattered into finer particles when rolled. Whether these particle size differences can be ascribed fully to vitreousness alone, to the method of comminution, or to other genetic or environmental differences among these grain samples that differed in vitreousness is not certain.
|
). The amount of rapidly degraded fraction (a), which includes small particles that wash through the pores of Dacron bags plus solubles, for both DM and starch increased linearly (P < 0.001) with harvest kernel moisture. This commensurately decreased the amounts of potentially degraded (b) DM and starch. Extent of DM and starch degradation in the rumen at a calculated hourly ruminal passage rate of 5% increased linearly (P < 0.001) with kernel moisture. Rapidly degradable DM (fraction a) was greater (P = 0.10) for FLO than VIT. An interaction of vitreousness with the quadratic effect of moisture was noted (P < 0.001), such that fraction a and effective degradation for starch tended to be greater for the vitreous hybrid at the lowest and greatest moisture content but lower for the vitreous hybrid at the intermediate moisture content. Calculated extent of ruminal degradation tended to parallel surface area of these grain samples. Remond et al. (2004) determined that with semi-flint corn, rapidly and potentially degraded starch increased linearly with decreasing particle size. Galyean et al. (1981) also noted that in situ starch disappearance of high-moisture corn increased by 95% when particle size was reduced by half (3,000 to 1,500 µm). Lykos and Varga (1995) found that particle size of dry corn was inversely related to ruminal digestion. Based on these data, particle size alone has a profound effect on starch disappearance in situ. Unfortunately, in situ loss may not adequately predict in vivo measurements due to the difference in particle size exposed to ruminal microbes and uncertainty about extent of ruminal degradation of the small particles in fraction a that are considered to be fully digested by in situ calculations.
|
). Mean DMI (6.5 kg) across dietary treatments appeared low. Relatively low DMI may be explained, at least in part, by the dairy breed influence of the cattle used. Lehmkuhler et al. (2003) noted DMI of Jersey steers during the finishing phase was 7.2 kg/d. Additionally, we employed a feedbunk management system intended to minimize feed refusals. Collectively, these factors may account for relatively low DMI reported in our study. Ruminal starch digestion for these samples of high-moisture corn averaged 90.1%. Although high when compared to literature estimates of ruminal digestion of starch for dry corn grain (Owens and Zinn, 2005), these values are consistent with results of recent studies with high-moisture corn. For example, Cooper et al. (2002) reported ruminal starch digestion was 91.7% for feedlot finishing diets containing 81.8% rolled high-moisture corn (29% kernel moisture) on a DM basis. In our study, intestinal starch disappearance measured either as a percentage of starch intake or as grams digested daily was not different among dietary treatments. However, when calculated as a percentage of starch entering the duodenum, intestinal starch digestibility was 2.9 to 4.1 percentage units greater (P = 0.02) for VIT than FLO corn. At the DRY kernel moisture, intestinal digestibility of non-starch OM also was greater (P < 0.05) for VIT than FLO. At each moisture level, grain with a smaller mean particle size was more extensively digested in the intestines. Large particle size has been shown to limit post-ruminal starch digestion (Rust, 1983). Remond et al. (2004) noted that as a percentage of duodenal passage, starch apparently digested in the small intestine increased linearly with decreasing particle size. Turgeon et al. (1983) investigated the effect of corn particle size on site and extent of starch digestion in beef steers; they noted that total tract starch digestibility was greater for cracked corn (GMD = 2,384 µm) than for whole corn (GMD = 5,977 µm). Hence, greater intestinal and total tract starch digestibility for more vitreous grain likely was a result of smaller mean particle size. This would support the concept that particle size may limit extent of intestinal digestion of DM and starch from processed corn grain. For total tract DM and OM digestion, endosperm vitreousness interacted (P < 0.06) with kernel moisture. Similarly, endosperm vitreousness tended to interact (P < 0.09) with kernel moisture for total tract starch digestibility. Total tract digestion of starch, DM, and OM was slightly but consistently greater for VIT than FLO, with the greatest difference existing at the driest kernel moisture, in which the difference in calculated surface area of the 2 grain types also was greatest.
|
). Callison et al. (2001) observed that total tract OM digestion decreased linearly as corn particle size increased. Although we did not determine digestibility of N, others have reported increased ruminal N degradation (Cerneau and Michalet-Doreau, 1991; Lykos and Varga, 1995) and increased N apparently digested in the total tract (Remond et al., 2004) associated with decreasing particle size of grain.
Ruminal Fermentation.
For ruminal fermentation measurements, sampling time did not interact with endosperm type or kernel moisture (P > 0.10). Endosperm vitreousness interacted (P = 0.01) with the linear effect of kernel moisture for mean ruminal fluid pH, such that pH tended to be lower for FLO compared with VIT at the driest kernel moisture but tended to be greater for FLO compared with VIT at wettest kernel moisture (Table 4). An interaction between endosperm vitreousness and the linear effect of kernel moisture was noted (P < 0.06), such that maximum ruminal fluid pH was lower and NH3 concentration was greater for the vitreous hybrid at the greatest moisture content. Variation (measured as SD) in ruminal fluid pH was greater (P = 0.03) for grain from the more floury hybrid. Time that ruminal pH remained below pH of 5.5 or 6.0 was not affected by high-moisture corn treatment. Considering that diets were offered 3 times each day using a feed delivery method designed to minimize feed refusals, it is not surprising that steers were able to manage their ruminal acid load. Therefore, any potential acidosis conditions that might be induced by dietary treatments would have gone undetected. Nevertheless, sampling protocol may have also contributed to a lack of statistical difference in ruminal fluid pH among dietary treatments. Ruminal fluid NH3, averaging 14.1 mg/100 mL, was not affected by high-moisture corn treatment. With 16.2% CP diets containing 81.75% rolled high-moisture corn (29% moisture) and 2.0% urea, Cooper et al. (2002) reported that ruminal fluid NH3 concentrations were approximately 14 mg/100 mL. Diets in the current study should have provided adequate ruminal fluid NH3 concentrations to maximize bacterial CP production (Satter and Slyter, 1974).
|
in ruminal fluid obtained from steers fed a diet containing 75% high-moisture corn. If propionate concentration is considered to reflect ruminal starch digestion, the slightly greater propionate percentage with FLO than VIT corn treatments is inconsistent with the in situ results in which VIT corn had greater ruminal starch disappearance. Nevertheless, numerical values for total VFA were consistent with the extent of ruminal starch digestion for VIT high-moisture corn. In conclusion, high-moisture corn with more vitreous endosperm yielded particles with a smaller size and greater surface area than a hybrid with more floury endosperm. As noted in previous studies with dry grain, smaller particle size was associated with more rapid in situ digestion and slightly greater intestinal and total tract starch digestibility. Accordingly, the potential negative effects on starch digestion associated with vitreous endosperm may be circumvented through ensiling and processing of high-moisture corn. Future research with high-moisture corn should identify particle sizes or kernel processing methods that strike a balance between agronomics and animal performance and health.
1 Corresponding author: jszasz{at}vetmed.wsu.edu
Received for publication May 3, 2006. Accepted for publication May 11, 2007.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
