Effects of high-amylase corn on performance and carcass quality of finishing beef heifers

Abstract

Developed initially for use in fuel ethanol production, Enogen Feed Corn (EFC; Syngenta Crop Protection) is genetically modified to express high concentrations of α-amylase in the corn kernel. Experiments were conducted to evaluate processing characteristics of EFC, in vitro digestion, and effects on feedlot performance, carcass characteristics, and liver abscess incidence. Experiment 1 used a randomized complete block design (3 × 3 × 5 factorial) to evaluate starch availability, in situ dry matter disappearance (ISDMD), in vitro gas production (IVGP), and volatile fatty acid (VFA) profiles of in vitro cultures. Grains (EFC or mill-run control [CON]) were flaked to a density of 360 g/L, and mixtures with 0%, 25%, 50%, 75%, or 100% EFC were prepared. Grains were tempered with added moisture (0%, 3%, or 6%) prior to steam conditioning for 15, 30, or 45 min. No two- or three-way interactions were observed. Adding moisture improved starch availability (linear; P < 0.01), and tended to improve ISDMD (linear, P = 0.06). Steam conditioning for 30 min improved starch availability, IVGP, and production of acetate, propionate, butyrate, valerate, and total VFA (P < 0.01) compared with conditioning for 15 or 45 min. Starch availability, ISDMD, IVGP, acetate, propionate, valerate, and total VFA production increased with an increasing proportion of EFC (linear, P < 0.01). Experiment 2 used 700 beef heifers (394 ± 8.5 kg initial body weight [BW]) fed finishing diets with steam-flaked corn as CON or EFC for 136 d. Targeting similar starch availabilities, grains were processed to 360 g/L (CON) and 390 g/L for CON and EFC, respectively. Heifers were blocked by BW, stratified, and then randomly assigned to 28 dirt-surfaced pens (25 animals per pen). Dry matter intakes were similar between treatments (P = 0.78), but cattle fed EFC had greater average daily gain (P < 0.01), improving feed efficiency by 5% (P < 0.01). Hot carcass weight was 6 kg greater for EFC cattle (P <0.01) than CON. No differences were observed for longissimus muscle area (P = 0.89), 12th-rib fat thickness (P = 0.21), or USDA yield grade (P = 0.13). Cattle fed CON had greater marbling scores than EFC (P = 0.04), but this did not affect the USDA quality grade (P > 0.33). Cattle fed EFC had 23% fewer abscessed livers than CON (P = 0.03). High-amylase corn may be used to improve microbial digestion, mill-throughput, and cattle performance, and it may mitigate liver abscesses.

Introduction

Enogen Feed Corn (EFC; Syngenta Seeds, LLC) is characterized by high-amylase expression in the kernel endosperm. It was originally developed and has been used extensively for the production of grain-derived fuel ethanol. Corn is well established as the principal ingredient fed to U.S. finishing cattle, and the starch therein provides a major proportion of dietary energy. Ruminants have limited capacity for pancreatic-amylase secretion and consequently are limited in their capacity for post-ruminal digestion of starch (Harmon et al., 2004). It has also been noted that intestinal digestion of starch is energetically more favorable when compared with ruminal digestion (McLeod et al., 2007). It is plausible that any ruminally undigested starch could be further degraded in the small intestine with the addition of α-amylase produced by the grain, potentially providing an energetic advantage.

Steam-flaking corn improves energy utilization from corn, and improvements by this processing technique are extensively documented (Owens et al., 1997; Zinn et al., 2002). Limited literature is available on finishing cattle fed EFC, and in the case of steam-flaking EFC, this is the first such research to be performed. Actions of heat-stable α-amylase within EFC could potentially be enhanced through the flaking process. Moreover, preliminary studies from our laboratory indicated that targeted starch availabilities for flaked grain could be achieved with lesser degrees of processing in EFC compared with control (CON) grain, thereby increasing throughput and decreasing energy expenditures associated with steam flaking. Objectives of studies reported herein were to examine processing parameters, in situ dry matter disappearance (ISDMD), in vitro gas production (IVGP), and volatile fatty acid (VFA) profiles of in vitro cultures, and the effects of EFC on feedlot performance, carcass characteristics, and liver abscess incidence and severity.

Materials and Methods

All studies were conducted in accordance with the procedures approved by the Kansas State University Institutional Animal Care and Use Committee.

Experiment 1

Experimental design and grain treatment

Grain mixtures were prepared using whole shelled EFC and mill-run corn (CON) as control. Grain types were blended in EFC:CON proportions of: 0:100, 25:75, 50:50, 75:25, and 100:0. Samples (2 kg) were placed into glass jars and water was applied at 0%, 3%, or 6% (w/w). Jars were put on a rotary device at 30 revolutions/min for 1 h to disperse moisture throughout the grains. After mixing, grains were placed into perforated steel baskets and inserted into a custom-fabricated steam table with 12 individual steam chambers. Randomized samples were then conditioned with steam for 15, 30, or 45 min. Treatments were arranged as a 5 × 3 × 3 factorial, with all 45 treatments prepared in duplicate.

Immediately after steam conditioning, samples were flaked using a dual-drive roller mill (R & R Machine Works; Dalhart, TX) with 46 × 91 cm corrugated rolls. Samples were placed into the flaker through a conveyance system above the rolls, collecting flaked product simultaneously underneath. Grains were flaked to a bulk density of 360 g/L. Enzymatic starch availability was determined soon after by steeping 25 g corn flakes in 100 mL 2.5%-amyloglucosidase solution heated to 55 C for 15 min (Sindt et al., 2006b). The liquid fraction was then filtered, and percent solubles were determined using a handheld refractometer. Percent solubles and dry matter (DM) were then converted to starch availability using regression equations. Another portion of grain was placed into a 105 °C forced-air oven for 24 h for the determination of moisture content. All remaining grain was frozen and retained for future in vitro and in situ analyses. This process took place in September 2016 at the Kansas State University Beef Cattle Research Center in Manhattan KS.

In situ dry matter disappearance

Approximately, 2.5 g (DM basis) of each flaked grain sample (three conditioning times, three moisture additions, five grain blends, and prepared in duplicate for a total of 90 samples) were weighed and sealed into Dacron bags (5 × 10 cm with 50 ± 10 µm porosity; ANKOM Technologies, Macedon, NY), with three bags representing each sample. In situ measurements were taken using 14-h ruminal incubation periods repeated on three separate days. Randomized blocks were animal within day, using six fistulated Jersey steers, such that each of the 45 treatment combinations was evaluated in each of the six fistulated steers. Two blank bags (sealed empty Dacron bags) were included as an adjustment factor in each animal during each incubation period to account for ruminal material that inherently enters bags and remains after rinsing. Weight gained in blanks was assumed to be equal to grain-containing samples and was removed during final calculations. Bags (including blanks) were suspended ruminally for 14 h, after which they were removed and rinsed. Bags were then dried at 105 °C in a forced-air oven for 24 h and weighed to calculate percent DM disappearance (using blank banks for adjustment).

IVGP and VFA profiles

In vitro studies were completed in duplicate for each flaked sample (45 treatment combinations originally flaked in duplicate = 90 samples) over four separate runs (45 samples per run). Each run consisted of a single replicate of each treatment combination and was used as a random effect in the statistical model. Two additional blank bottles were used in each run to adjust for background VFA concentrations contributed by ruminal fluid inoculum. Randomized samples (3 g whole corn flakes, DM basis) were placed into 250-mL fermentation bottles equipped with ANKOM pressure sensing modules (ANKOM Technologies). Ruminal fluid was collected from two cannulated steers (blocks) fed a 50:50 concentrate:roughage diet. Individual ruminal fluid samples were strained through eight layers of cheesecloth, transferred into separatory flasks, degassed using nitrogen, capped, and placed into a 39 °C incubation chamber. After approximately 1 h, strained ruminal fluid samples were separated into three distinct layers, and the center, microbe-rich fraction was used as an inoculant. Ten milliliters of inoculum were combined with 140 mL McDougall’s artificial saliva and flaked corn, degassed with nitrogen, and capped with an ANKOM module. ANKOM modules measured cumulative IVGP at 15-min intervals. Fermentation vessels were placed into an orbital shaker and incubated at 39 °C for 24 h with light agitation. Initial and final pH values were recorded for each culture. Following incubation, 4 mL of the liquid fraction from each bottle was combined with 1 mL of 25% metaphosphoric acid and frozen. Frozen samples were then thawed, vortexed, and centrifuged, and the top layer of supernatant was transferred into gas chromatography vials.

VFA concentrations of cultures were determined using an Agilent 7890 gas chromatograph (Agilent Technologies, Santa Clara, CA) equipped with a Nukol capillary column (15 m × 0.35 mm, d_f 0.50 µm). Concentrations of acetate, propionate, isobutyrate, butyrate, isovalerate, valerate, isocaproate, caproate, and heptanoate were determined for each sample. No internal standards were used.

Statistical analyses

Data were analyzed using the MIXED models procedure of the Statistical Analysis System (SAS version 9.4; SAS Inst. Inc., Cary, NC). Fixed effects included percent EFC, percent added moisture, steam conditioning time, and all two- and three-way interactions. The experimental unit for ISDMD was the Dacron bag (n = 270), and the experimental unit for in vitro studies was fermentation bottle (n = 180). Random effect was block (animal within day was block for ISDMD, and run was block for IVGP and VFA analyses). Orthogonal contrasts were used to identify linear and quadratic effects. Modeling of IVGP also included time as a repeated measure, and the Slice option was used to characterize treatment differences at different points in time. Statistical significance was declared at P < 0.05, and tendency for significance at P < 0.10.

Experiment 2

The trial was initiated in December 2016 and ended April 2017, taking place at the Kansas State University Beef Cattle Research Center, Manhattan, KS.

Experimental design

A randomized complete block design with two treatments was carried out using 700 crossbred beef heifers (394 kg ± 8.5 initial body weight [BW]). Two lots of cattle, kept separate during blocking, were utilized in the trial; 350 heifers received in June 2016 were used previously in a receiving trial examining trace mineral supplementation. The second lot of cattle was received in November 2016, targeting similar initial BW between lots at study initiation. Heifers within lot were stratified by BW, randomized within strata, blocked by BW, and then randomly assigned within the block to treatments. Cattle were assigned 28 dirt-surfaced pens (25 animals per pen; 14 pens per treatment) providing approximately 18.5 m² of surface area for each animal. Treatments randomly assigned within the block consisted of CON steam flaked to 360 g/L and EFC steam flaked to 390 g/L. Grain treatments were designed to target similar daily starch availability, based on the observations from preliminary work. The CON was flaked at approximately 6 t/h, while throughput for EFC was approximately 9 t/h, thereby decreasing steam-conditioning time for EFC compared with CON.

Animal processing, housing, and handling

Upon arrival at the Kansas State University Beef Cattle Research Center, heifers were given ad libitum access to alfalfa hay and water. Cattle were of sale barn origin and received as two separate lots, each processed 24 to 48 h after arrival. Processing included vaccination using a 5-way viral vaccine (Bovishield Gold-5; Zoetis, Parsippany, NJ), a 7-way clostridial (UltraBac 7 Somubac; Zoetis), prophylactic administration of tilmicosin (Micotil; Elanco Animal Health, Greenfield, IN), and application of a topical parasiticide (Dectomax; Zoetis). Animals were identified with a uniquely numbered ear tag, and BW was recorded. On day 1 of trial initiation, starting BW was recorded as animals were sorted into pens, and heifers were implanted with a trenbolone/estradiol implant (Component TE-IH with Tylan; Elanco Animal Health). On day 84, heifers were reimplanted (Component TE-200 with Tylan; Elanco Animal Health) and treated with a pour-on insecticide (Standguard; Elanco Animal Health).

Animals were housed in dirt-surfaced pens that provided approximately 18 m² of surface area/animal; fences and gates were constructed of steel pipe. Automatic waterers provided ad libitum access to water were shared between adjacent pens, and water was from a municipal supply.

Diet preparation

Heifers were transitioned to finishing diets at the start of the trial over 21 d using three step-up diets with concentrate:roughage ratios of: 60:40, 71:29, and 92:18 (7 d per step) for gradual adaptation to the finishing diet. Treatments were applied during dietary adaptation (CON or EFC were fed as a concentrate during the transition). Both grain types were steam flaked daily, using the same flaking system as described for exp. 1. Water was applied to grains prior to steam conditioning with a moisture applicator (SarTec; Anoka, MN) to obtain final moisture content of 21% to 22% in the final flaked grain products. The CON grain was processed to a final density of 360 g/L at a rate of 6 t/h, and EFC was processed to a density of 390 g/L at a rate of 9 t/h, with the goal of attaining similar starch availabilities. Experimental diets are shown in Table 1. Ractopamine hydrochloride (Optaflexx 45; Elanco Animal Health) was included in total mixed rations for the final 39 d on feed at the rate of 300 mg/animal daily. Rations were mixed and delivered once daily, beginning at approximately 0800 hours, and fed ad libitum. Feed intakes were visually monitored and adjusted daily as needed so that only trace amounts of residual feed remained in the bunks before feeding each morning. Orts were collected as needed to account for unconsumed feed, dried at 55 °C for 48 h, and subtracted from feed deliveries to estimate actual dry matter intake (DMI) for each pen. Subsamples of each feed ingredient were collected weekly, dried at 55 °C for 48 h, and composited into monthly samples, which were later analyzed for nutrient composition (SDK Labs; Hutchinson, KS).

Table 1.

Composition of finishing diets fed to beef heifers fed either steam-flaked EFC or CON¹

Item	CON	EFC
Ingredient, % of DM
Mill-run corn (steam-flaked)	85.40	0.00
High-amylase corn (steam-flaked)	0.00	85.40
Ground alfalfa hay	7.00	7.00
Soybean meal	1.64	1.64
Feed additive premix2	2.20	2.20
Supplement3	3.76	3.76
Analyzed composition, % of diet DM4
Crude protein	15.44	15.37
ADF	6.51	7.15
Ether extract	3.58	4.16
Calcium	0.60	0.59
Phosphorus	0.23	0.24
Potassium	0.64	0.68

¹Optaflexx (Elanco Animal Health, Greenfield, IN) was fed at 300 mg ractopamine–HCl per heifer daily for the final 39 d on feed.

²Consisted of ground corn and 1% tallow and provided 0.05 mg/kg melengestrol acetate (MGA; Zoetis, Parsippany, NJ) in total diet DM.

³Consisted of urea, salt, limestone, trace mineral premix, vitamin premix, and potassium chloride to provide (on total diet DM basis) 0.15 mg/kg cobalt, 10 mg/kg copper, 0.50 mg/kg iodine, 20 mg/kg manganese, 0.10 mg/kg selenium, 30 mg/kg zinc, 2,205 IU/kg vitamin A, 22 IU/kg vitamin E, and 36.4 mg/kg monensin (Rumensin; Elanco Animal Health, Greenfield, IN).

⁴Analyzed nutrient composition of ingredients in total diet (SDK Labs; Hutchinson, KS). Each ingredient was analyzed by month as composites from weekly samples (five composites per ingredient).

Harvest

On day 136, pens of animals were weighed on a platform scale immediately before loading onto trucks for transport. Final BW was calculated by multiplying the mean BW for each pen by 0.96 to account for 4% shrink during transport. Heifers were loaded onto trucks and transported approximately 440 km to a commercial abattoir in Lexington, NE. Records collected on the day of slaughter by trained Kansas State University personnel included animal identification, harvest order, hot carcass weight (HCW), and liver abscess prevalence and severity using the system described by Brown et al. (1975). Following a refrigeration period of approximately 24 h, longissimus muscle area (LMA), 12th-rib subcutaneous fat thickness, marbling score, and USDA yield and quality grades were determined. The dressing percentage was calculated by averaging HCW within the feedlot pen and dividing that value by final shrunk BW.

Grain characteristics

Starch availability, DM content, and particle size distribution were measured for processed grains daily. Grain DM was determined by drying in a forced-air oven at 105 °C for 24 h. Enzymatic starch availability was determined using the methodology described in exp. 1. Particle size was determined by weighing approximately 200 g of flaked corn placed onto a set of sieves, with decreasing screen sizes in the order: 4,750, 3,350, 2,360, 1,700, 1,180, 850, 600 µm, and a solid pan. The stack was placed into a Ro-Tap orbital shaker (RX-30; W. S. Tyler, Mentor, OH) for 5 min. Particles were removed from each sieve and weighed. Mean geometric particle size was calculated as described by Baker and Herrman (2002) using equations developed by Pfost and Headley (1976).

Statistical analyses

Analyses of BW, DMI, average daily gain (ADG), and feed efficiency employed the MIXED procedure of the SAS (version 9.4), with pen as the experimental unit, grain type as the fixed effect, and block as the random effect. Carcass characteristics were analyzed similarly, with the exception that categorical carcass traits (USDA quality grade, USDA yield grade, and liver abscess prevalence and severity) were analyzed with the GLIMMIX procedure of SAS, reflecting the proportion of animals within each pen that appeared within each category. For the relationship between HCW and liver abscess, data were analyzed with GLIMMIX using liver abscess category, treatment, and the interaction as fixed effects, and block as a random effect.

Results and Discussion

Experiment 1

There were no two- or three-way interactions between factors for any experimental analyses (P > 0.10), thus only the main effects of each treatment are shown. No caproate, isocaproate, or heptanoate was detected when analyzing VFA profiles of in vitro cultures.

Moisture addition to grains

The overall impact of adding moisture prior to steam conditioning corn appeared minimal. Starch availability, ISDMD, and IVGP effects can be seen in Table 2. There was a linear increase (P < 0.01) in enzymatic starch availability when the amount of tempered moisture increased from 0% to 6%. A tendency for linear improvement (P = 0.06) of ISDMD occurred with increasing amounts of tempered moisture. Tempering with moisture did not affect IVGP (linear; P = 0.81; quadratic; P = 0.42). Adding 3% water to corn resulted in lower acetate production compared with grains with 0% or 6% added moisture (quadratic effect; P = 0.04), and tendencies for reduced propionate (quadratic; P = 0.07) and total VFA (quadratic; P = 0.08) when compared with 0% or 6% treatments (Table 3). Moisture addition prior to steam conditioning corn had no other effects on VFA profiles.

Table 2.

Effects of moisture addition to grain (tempering) prior to steam conditioning on starch availability, ISDMD, and production of fermentative gasses by in vitro cultures of mixed ruminal microbes (IVGP)

	Tempering moisture, %1				P-value
Item	0	3	6	SEM	Linear	Quadratic
Starch availability, %2	47.2	49.5	51.2	0.60	<0.01	0.76
ISDMD, %3	40.3	40.7	42.0	1.62	0.06	0.51
IVGP, mL4	110.1	107.4	111.2	7.33	0.81	0.42

¹Performed by adding water (w/w) with whole grains into glass jars. Jars were placed on a roller platform and mixed 60 min prior to steam conditioning.

²Measured using the refractive index method (Sindt et al., 2006b) once per sample (90 observations, 30 per treatment) shortly after flaking.

³Measured over 3 d using six ruminally fistulated steers by incubating 2.5 g (DM) corn flakes in Dacron bags ruminally for 14 h (270 observations, 90 per treatment).

⁴Mean gas production during the 24-h incubation period. Measured by incubating 3 g (DM) flaked corn as substrate, with 10 mL ruminal fluid inoculum and 140 mL McDougall’s Buffer at 39 °C. Repeated with two replicates per sample (90 samples, 180 observations, 60 per treatment).

Table 3.

Effects of moisture addition to grain (tempering) prior to steam conditioning on the production of VFAs by in vitro cultures of mixed ruminal microbes fed steam-flaked corn as substrate

	Tempering moisture, %1				P-value
VFA, mmoles/g substrate DM2	0	3	6	SEM	Linear	Quadratic
Acetate	1.60	1.54	1.60	0.063	0.91	0.04
Propionate	1.85	1.78	1.84	0.063	0.71	0.07
Acetate:propionate	0.87	0.87	0.88	0.048	0.31	0.50
Butyrate	0.52	0.51	0.52	0.067	0.72	0.60
Isobutyrate	0.003	0.002	0.002	0.0021	0.59	0.97
Valerate	0.063	0.063	0.064	0.0116	0.68	0.80
Isovalerate	0.006	0.006	0.005	0.0026	0.29	0.71
Total	4.05	3.90	4.02	0.082	0.79	0.08

¹Performed by adding water (w/w) with whole grains into glass jars. Jars were placed on rollers and mixed 60 min prior to steam conditioning.

²VFAs were measured by gas chromatography following 24-h incubation of cultures containing 3 g (DM) flaked corn as substrate, 10 mL ruminal fluid inoculum, and 140 mL McDougall’s Buffer at 39 °C. Initial incubation was repeated with two replicates per sample (90 samples, 180 observations, 60 per treatment).

These results are in agreement with those of Sindt et al. (2006a), who observed no effect on IVGP when tempering with 6%, 10%, or 14% added moisture. Sindt et al. (2006b) observed a tendency for moisture addition (0%, 6%, or 12%) to increase enzymatic starch availability, but no effect on in vivo VFA profile. Moisture addition to corn prior to flaking has shown little to no impact on cattle performance (Zinn et al., 2002; Gutierrez et al., 2018) but may be used to improve flake durability and potentially reduce the amount of steam needed to condition corn (Sindt et al., 2006a).

Steam conditioning time

The impact of steam conditioning time is displayed in Tables 4 and 5. In most cases, treating corn with 30 min of steam was optimal when compared with 15 or 45 min, with this amount yielding the greatest numerical value for starch availability, IVGP, acetate, propionate, butyrate, valerate, and total VFA (quadratic; P < 0.01). Compared with 15 or 45 min, conditioning with 30 min of steam also reduced acetate:propionate (quadratic; P < 0.01). Conditioning grain for 30 min resulted in greater concentrations of isobutyrate compared with 15 or 45 min of steam conditioning (quadratic; P = 0.03), while 15 min of steam conditioning yielded the greatest production of isovalerate (linear; P < 0.01). Steam conditioning time had no discernible impact on ISDMD (P > 0.10).

Table 4.

Effects of steam conditioning time for flaked corn on starch availability, ISDMD, and production of fermentative gasses by in vitro cultures of mixed ruminal microbes (IVGP)

	Conditioning time, min				P-value
Item	15	30	45	SEM	Linear	Quadratic
Starch availability, %1	48.5	51.5	47.9	0.60	0.52	<0.01
ISDMD, %2	36.6	44.4	42.0	2.66	0.15	0.12
IVGP, mL3	102.3	121.3	104.9	7.33	0.57	<0.01

¹Measured using the refractive index method (Sindt et al., 2006b) once per sample (90 observations, 30 per treatment) shortly after flaking.

²Measured over 3 d using six ruminally fistulated steers by incubating 2.5 g (DM) corn flakes in Dacron bags ruminally for 14 h (270 observations, 90 per treatment).

³Mean gas production during 24-h incubation period. Measured by incubating 3 g (DM) flaked corn as substrate, with 10 mL ruminal fluid inoculum and 140 mL McDougall’s Buffer at 39 °C. Repeated with two replicates per sample (90 samples, 180 observations, 60 per treatment).

Table 5.

Effects of steam conditioning time on the production of VFAs from in vitro cultures of mixed ruminal microorganisms fed steam-flaked grains as substrate

	Conditioning time, min				P-value
VFA, mmoles/g substrate DM1	15	30	45	SEM	Linear	Quadratic
Acetate	1.56	1.67	1.51	0.063	0.13	<0.01
Propionate	1.75	1.99	1.73	0.063	0.56	<0.01
Acetate:propionate	0.89	0.85	0.88	0.048	<0.01	<0.01
Butyrate	0.50	0.58	0.48	0.067	0.16	<0.01
Isobutyrate	0.002	0.004	0.001	0.0021	0.11	0.03
Valerate	0.058	0.073	0.058	0.0117	0.95	<0.01
Isovalerate	0.007	0.005	0.005	0.0026	0.02	0.22
Total VFA	3.88	4.32	3.78	0.082	0.25	<0.01

¹VFAs were measured by gas chromatography following 24-h incubation of cultures containing 3 g (DM) flaked corn as substrate, 10 mL ruminal fluid inoculum, and 140 mL McDougall’s Buffer at 39 °C. Initial incubation was repeated with two replicates per sample (90 samples, 180 observations, 60 per treatment).

These results reflect the digestive properties of starch, as steam conditioning for a short time likely does not yield enough opportunity for starch gelatinization (Zinn et al., 2002; Huntington et al., 2006). Svihus et al. (2005) described the relationship between the degree of gelatinization and digestibility as being linear. Kurakake et al. (1997) observed that starch granule swelling power and solubility decreased with higher temperature, implying more rigidity, which they attributed to restructuring between amylose and amylopectin. We speculate that extended periods of steam conditioning may result in similar changes, or possibly induce early-stage Maillard reactions between proteins and reducing sugars that adversely impact starch availability. It should be noted that the device used for steam conditioning of grains in this experiment utilized relatively small volumes of grain (approximately 4 liters) that reached maximum temperature within 1 or 2 min. Commercial-scale steam conditioning chambers are far larger, requiring 15 to 20 min or longer for grains to reach peak temperature. We recognize that the absolute times for steam conditioning evaluated in this experiment may not be directly applicable to commercial-scale processing systems; nevertheless, the results do suggest that there is an optimal time for steam conditioning, and excessive conditioning times may actually be counterproductive.

High-amylase corn concentration

High-amylase corn improved starch availability and all measures of microbial digestion compared with the CON corn. Enzymatic starch availability improved linearly (P < 0.01) with an increasing percentage of EFC in flaked grain mixtures (Table 6). Also in Table 6, EFC increased ISDMD (linear; P < 0.01), with 100% EFC resulting in an 11% improvement over CON. This is a 32% increase in ISDMD. A visual graphic of IVGP improvement using EFC over time is in Figure 1. Using greater proportions of EFC linearly improved IVGP (P < 0.01) over CON. Differences between treatments (evaluated using the Slice option) manifested as early as 6 h into the study and were maintained for the balance of the incubation period. Table 7 displays the effects of EFC on in vitro VFA profile. As may be expected based off IVGP, increasing EFC concentration in grain blends linearly increased production of acetate, propionate, valerate, and total VFA (P < 0.01). Acetate-to-propionate ratio decreased (linear; P < 0.01) with greater proportions of EFC. The only negative effect of EFC was a decrease in isovalerate (linear; P < 0.01) by increasing EFC content.

Table 6.

Effects of increasing proportion of steam-flaked EFC in grain mixtures on starch availability and ISDMD

	Proportion of grain mixture as EFC, %						P-value
Item	0	25	50	75	100	SEM	Linear	Quadratic
Starch availability, %1	45.0	47.7	50.5	50.7	52.7	0.77	<0.01	0.16
ISDMD, %2	34.8	39.5	40.8	43.8	46.0	1.70	<0.01	0.27

¹Measured using the refractive index method (Sindt et al., 2006b) shortly after flaking, with a total of 18 samples per treatment.

²Measured using six ruminally fistulated steers by incubating 2.5 g (DM) corn flakes in Dacron bags ruminally for 14 h (270 observations, 54 per treatment).

Figure 1.

Effect of the proportion of EFC in grain mixtures on the production of fermentative gasses by in vitro cultures of mixed ruminal microbes with steam-flaked grains as substrate. Cumulative gas production was recorded at 15-min intervals throughout a 24-h incubation period at 39 °C using culture bottles equipped with pressure sensing modules (ANKOM Technologies, Macedon, NY). Cultures contained 3 g (DM) of flaked corn as substrate, 10 mL ruminal fluid inoculum, and 140 mL McDougall’s buffer. Repeated with two replicates per sample (90 samples, 180 observations, and 36 per treatment). Linear effect of EFC proportion, P < 0.01; Quadratic effect of EFC proportion, P = 0.21; SEM = 7.81.

Table 7.

Effects of increasing proportion of steam-flaked EFC in grain mixtures on the production of VFAs by in vitro cultures of mixed ruminal microorganisms fed steam-flaked corn as substrate

	Proportion of grain mixture as EFC, %						P-value
VFA, mmoles/g substrate DM1	0	25	50	75	100	SEM	Linear	Quadratic
Acetate	1.49	1.51	1.61	1.63	1.66	0.067	<0.01	0.64
Propionate	1.69	1.73	1.85	1.90	1.95	0.067	<0.01	0.83
Acetate:propionate	0.89	0.88	0.87	0.87	0.86	0.048	<0.01	0.76
Butyrate	0.53	0.48	0.52	0.51	0.54	0.067	0.21	0.06
Isobutyrate	0.004	0.000	0.003	0.003	0.002	0.0021	0.98	0.33
Valerate	0.053	0.056	0.067	0.063	0.076	0.0119	<0.01	0.61
Isovalerate	0.008	0.005	0.007	0.005	0.004	0.0027	<0.01	0.78
Total	3.77	3.78	4.05	4.12	4.24	0.097	<0.01	0.93

¹VFAs were measured by gas chromatography following 24-h incubation of cultures containing 3 g (DM) flaked corn as substrate, 10 mL ruminal fluid inoculum, and 140 mL McDougall’s buffer at 39 °C (36 observations per treatment).

Only limited literature are available on EFC digestibility, none of which pertain to steam-flaked corn. Jolly-Breithaupt et al. (2016b) observed numerical improvements in the ruminal digestibility of organic matter (OM) and starch when feeding a finishing diet using dry-rolled EFC to fistulated cattle. They saw no effects on VFA concentrations. It is likely that the processing method has a substantial impact on EFC digestibility, as the application of moisture and heat from steam enhances EFC α-amylase activity. It is important to note that we observed no quadratic effects of EFC for any analysis; this means that the actions of EFC are likely confined to that portion of the grain mixture and do not affect the CON portion of the blends.

Experiment 2

Four animals were removed from the CON group for nontreatment-related reasons: three due to calving and one was found deceased due to respiratory disease. Four animals were also removed from the EFC group for nontreatment reasons: one due to a bacterial infection, one due to respiratory disease, one due to a displaced abomasum, and one due to a hip-injury, all of which caused severe weight loss.

Grain characteristics

Laboratory analysis (SDK labs) of nutrient composition between CON and EFC are shown in Table 8. High-amylase corn had greater acid detergent fiber (ADF; P < 0.01), and potassium (P = 0.03) components compared with CON. High-amylase corn also had a tendency (P = 0.06) for greater fat content (ether extract as measurement) as opposed to CON. Observed differences are not likely due to the α-amylase component of EFC. No differences were evident between flaked grains for protein, calcium, or phosphorus.

Table 8.

Nutrient analyses¹ of steam-flaked CON and EFC

Item, % of DM	CON	EFC	SEM	P-value
Crude protein	8.77	8.69	0.09	0.55
ADF	3.69	4.45	0.14	< 0.01
Ether extract	3.58	4.27	0.18	0.06
Calcium	0.016	0.014	0.002	0.59
Phosphorous	0.226	0.246	0.008	0.13
Potassium	0.334	0.374	0.008	0.03

¹Analyzed by SDK Labs, Hutchinson, KS. Weekly samples (~1 kg) were composited into monthly samples, which were then analyzed (five composites per grain type).

The characteristics of grains are presented in Table 9. By design, moisture content of both corn types had no difference after steam flaking (P = 0.55), and starch availability was similar, although there was a tendency (P = 0.08) for EFC to yield a greater starch availability value than CON. Even though EFC was flaked to a greater bulk density (390 vs. 360 g/L), it still resulted in a smaller mean particle size (P < 0.01) when compared with CON; Figure 2 displays particle size distribution across sieve sizes. This can likely be explained observationally, as EFC flakes appeared more fragile and did not hold together as well as CON. There was no difference in the standard deviation of grain particle size (P = 0.99). Sindt et al. (2006a) observed no effects on finishing performance or carcass characteristics when feeding different particle sizes of steam-flaked corn, this being 4,667 or 3,330 µm (smaller particle size achieved by 15 min of mixing, not by reducing flake density).

Table 9.

Characteristics of CON and EFC after steam flaking to a bulk density of 360 g/L (at 6 t/h) or 390 g/L (at 9 t/h), respectively, from daily grain samples

Item	CON	EFC	SEM	P-value
DM, %	78.8	78.9	0.70	0.55
Starch availability, %1	51.3	52.1	0.30	0.08
dgw, µm2	4,413	4,292	24	<0.01
Sgw3	1.50	1.50	0.017	0.99

¹Measured using the refractive index method (Sindt et al., 2006b) immediately after flaking.

²Mean geometric particle size: measured by placing flaked grains (200 g) in a set of sieves with descending screen sizes in a Ro-Tap orbital shaker (RX-30; W. S. Tyler, Mentor, OH), and weighing each fraction following a 5-min rotary tapping cycle. Calculated using equations described by Pfost and Headley (1976).

³Standard deviation of particle size: calculated using equations described by Pfost and Headley (1976) following the procedure described above².

Figure 2.

The particle size distribution of CON or EFC after steam flaking to a bulk density of 360 g/L (at 6 t/h) or 390 g/L (at 9 t/h), respectively. Particle size distributions were measured by placing daily samples of flaked grains (200 g) in a set of sieves with descending screen sizes in a Ro-Tap orbital shaker (RX-30; W. S. Tyler, Mentor, OH) and weighing each fraction following a 5-min rotary tapping cycle. Superscript letters (^a,b) indicate differences (P < 0.05) within sieve size.

Feedlot performance

The effects of EFC on gain and efficiency of feedlot heifers are found in Table 10. There was no difference in BW at trial initiation (P = 0.52), but cattle fed EFC were heavier than CON on the final day (P < 0.01). Thus, over the 136-d period, EFC cattle had improved ADG (P < 0.01). There was no difference in DMI between treatments (P = 0.78), which results in 5% greater feed efficiency for cattle fed EFC (P < 0.01) compared with CON. When comparing the effects of EFC when dry-rolled (DRC) or as high-moisture corn (HMC), and combined with one of two different byproducts fed to finishing steers, Jolly-Breithaupt et al. (2016a) observed a numerical 3.9% increase in gain:feed (G:F) when comparing EFC–DRC with modified wet distillers grains plus solubles (MDGS) opposed to their control. They also observed a 2.1% numerical improvement in G:F using EFC fed as HMC (with MDGS). When sweet bran was included instead of MDGS, EFC–HMC had the opposite effect, as feed efficiency decreased by 2.1% (numerically). Jolly-Breithaupt et al. (2016a) also presented a corn type × corn processing interaction, in which final BW (P = 0.02) and ADG (P = 0.04) were greatest in those fed EFC as DRC, and lowest when fed as HMC. These results suggest at least some similarities to what we observed when steam-flaking EFC. It also suggests that the feed effects of EFC may vary by corn processing method.

Table 10.

Finishing performance of heifers fed diets containing steam-flaked CON or steam-flaked EFC¹

Item	CON	EFC	SEM	P-value
Initial BW, kg	395	394	8.6	0.52
Final BW, kg2	588	599	10.7	<0.01
DMI, kg/d	10.00	10.07	0.196	0.78
ADG, kg	1.60	1.69	0.028	<0.01
G:F	0.160	0.168	0.002	<0.01

¹Trial utilized 700 beef heifers in a randomized complete block design, with 25 animals per pen, 14 pens per treatment, and fed 136 d.

²Gross BW (measured immediately prior to loading cattle onto trucks for transport to abattoir) multiplied by 0.96 to account for 4% shrink.

Other work has been performed to evaluate the effectiveness of adding exogenous α-amylase to feedlot diets. In two experiments from Tricarico et al. (2007), α-amylase (Amaize, Alltech Inc., Nicholasville, KY) fed at a rate of 950 DU/kg DM (DU defined as one dextrinizing unit is the quantity needed to dextrinize soluble starch at 30 °C, pH 4.8, at 1 g/h) was first added to steam-flaked corn diets using alfalfa or cottonseed hulls as roughage. The second experiment identified amylase quantity effects when feeding cracked corn (CC) or HMC finishing diets. Experiment 1 showed no effects of added amylase or amylase × roughage source interactions for any feedlot performance data over the entire feeding period on beef steers. There were no interactive effects between amylase and corn processing methods in exp. 2, where finishing beef heifers had optimal ADG and DMI (quadratic effect, P = 0.04 and P = 0.07, respectively) when amylase was added to diets at a rate of 580 DU/kg DM. They saw no effects on feed efficiency. While the increase in ADG is comparable to our study, this was with CC and HMC; no effects were observed when amylase was added to the steam-flaked corn diet. The amylase content of EFC grain varies with hybrid and environmental conditions during the growing period. Furthermore, processing temperatures needed for optimal amylase activities may be different for exogenous enzymes compared with those contained within EFC.

Carcass characteristics

The effects of EFC on carcass merit are shown in Table 11. The improved daily gain in cattle fed EFC resulted in greater carcass weight, as heifers produced approximately 6 kg heavier carcasses (P < 0.01) when compared with CON. No differences in LM area or 12th-rib fat were evident. The CON diet yielded carcasses with a numerically greater marbling score (P = 0.04) over EFC, but this did not impact the USDA quality grades (P > 0.33). There also was a tendency (P = 0.09) for an increased percentage of USDA yield grade 3 carcasses in heifers fed EFC compared with CON. Jolly-Breithaupt et al. (2016a) evaluated the impact of grain processing method with EFC, and observed tendencies for interaction between corn type and processing method on HCW, 12th-rib fat, and marbling score. When consuming EFC–DRC, HCW and marbling score were greater than the control counterpart. When steers consumed EFC–HMC, HCW and marbling scores decreased compared with the HMC control. Twelfth-rib fat thickness was least for cattle fed EFC–DRC and greatest for cattle fed EFC–HMC. The authors did not speculate regarding the possible cause for lower marbling scores and increased 12th-rib fat thickness when EFC was fed as HMC. It is difficult to compare these results as differences in enzyme activation likely occur when comparing their processing methods to steam-flaked EFC. The increase in HCW when EFC was fed as DRC agrees with our findings, while the decrease in marbling score feeding EFC as HMC is an additional caveat. Jolly-Breithaupt et al. (2016b) do offer insight into the potential mechanism of action of high-amylase corn, citing increased total tract digestion of OM and starch for diets based on DRC. For finishing diets based on steam-flaked corn, total tract starch digestion in flaked corn diets is typically 99% or more (Sindt et al., 2006b), suggesting that opportunities to increase total tract starch digestibility are very limited. Altering the site of digestion to favor ruminal or small intestinal digestion may, therefore, be a more plausible mechanism of action.

Table 11.

Carcass characteristics of heifers fed diets containing steam-flaked CON or steam-flaked EFC¹

Item	CON	EFC	SEM	P-value
HCW, kg	366	372	6.41	<0.01
Longissimus muscle area, cm2	94.7	94.6	1.02	0.89
12th-rib fat thickness, cm	1.16	1.19	0.045	0.21
Marbling score2	605	589	10	0.04
USDA Prime, %	6.6	4.9	1.68	0.33
USDA Choice, %	68.7	70.4	4.44	0.62
USDA Select, %	10.7	11.4	2.58	0.79
USDA sub-Select, %3	9.0	9.3	2.61	0.68
USDA yield grade
Yield grade 1, %	23.2	19.5	3.77	0.22
Yield grade 2, %	49.4	47.7	3.01	0.64
Yield grade 3, %	25.1	30.9	3.07	0.09
Yield grade 4, %	2.0	2.0	0.76	1.00
Yield grade 5, %	0.3	0.0	0.20	0.32

¹Trial utilized 700 beef heifers in a randomized complete block design, with 25 animals per pen, 14 pens per treatment, and fed 136 d prior to transport to a commercial abattoir, wherein carcass data were collected.

²500 to 599 = small degree of marbling; 600 to 699 = modest degree of marbling.

³Carcasses graded as USDA Standard, Commercial, Utility, or Cutter.

Tricarico et al. (2007) observed no interactions between exogenous amylase addition and roughage source in finishing diets containing steam-flaked corn. LMA was greater for steers consuming the amylase diet compared with those fed the control diet. In a second study, these authors observed that HCW was optimal when cattle were fed 580 DU/kg DM amylase (quadratic effect). Like exp. 1, they observed a quadratic increase in the LM area (quadratic effect) with the 580 DU/kg DM concentration being optimal. A tendency for fat thickness to increase in response to increasing amylase concentration was also observed, and 580 DU/kg DM amylase resulted in the lowest numerical yield grade. The fact that HCW increased with added amylase would agree with our findings. However, differences in backfat, LM area, and overall yield grade do not agree with our observations. The actions of exogenous amylase enzyme added to a finishing diet may be different than those of EFC, for which amylase is an inherent component of the grain. Meaningful comparisons at this stage again may be difficult with limited literature on finishing cattle with ECF or adding amylase enzyme.

Liver abscess prevalence and severity are shown in Table 12. Note that tylosin was not included in experimental diets (Table 1). Finished beef heifers fed EFC had fewer total abscessed livers at slaughter than their CON counterparts (P = 0.03). This difference occurs due to fewer moderate (P = 0.03) and severe (P = 0.11) abscessed livers in the EFC group. No previous peer-reviewed literature has identified any liver abscess effects when feeding EFC (through any processing method) or by feeding with added amylase enzyme. Detrimental effects of liver abscesses on cattle gain (Potter et al., 1985) resulting in lighter, poorer quality carcasses (Brown and Lawrence, 2010) have been well established. The relationship between liver abscess severity and HCW for each treatment is shown in Figure 3. There were effects by diet (P < 0.01), liver abscess severity (P < 0.01), and tendency for a treatment × liver abscess interaction (P = 0.09). The EFC group maintained heavier carcasses and did not display the same decrease in HCW when abscess severity increased that was observed for CON cattle. A biological explanation for this effect of EFC on liver abscesses in not readily apparent. It is conceivable that the use of EFC shifted the site of starch digestion, resulting in less starch appearing in the hindgut. If the effects of EFC on the incidence of abscessed livers are real, the implications could be substantial, as alternative methods for the prevention of liver abscesses are clearly warranted.

Table 12.

Liver abscess incidence and severity¹ in heifers fed diets containing steam-flaked CON or steam-flaked EFC²

Item	CON	EFC	SEM	P-value
Total abscessed livers, %	34.4	26.6	2.47	0.03
A−, %	11.9	12.7	1.80	0.73
A0, %	14.7	9.2	1.74	0.03
A+, %	7.5	4.6	1.40	0.11

¹Severity classified using the system described by Brown et al. (1975) wherein A⁻ denotes livers with one or two small abscesses or inactive scars, A⁰ denotes livers with 1 to 2 large abscesses or multiple small abscesses, and A⁺ denotes livers with multiple large abscesses.

²Trial utilized 700 beef heifers in a randomized complete block design, with 25 animals per pen, 14 pens per treatment, and fed 136 d prior to transport to a commercial abattoir.

Figure 3.

Relationship between liver abscess severity and HCW of finishing heifers fed diets containing steam-flaked CON or EFC. Trial utilized 700 beef heifers in a randomized complete block design, with 25 animals per pen, 14 pens per treatment, and fed 136 d prior to transport to a commercial abattoir. Abscess severity classified using the system described by Brown et al. (1975), wherein A⁻ denotes livers with one or two small abscesses or inactive scars, A⁰ denotes livers with one to two large abscesses or multiple small abscesses, and A⁺ denotes livers with multiple large abscesses. Effect of treatment, P < 0.01; Effect of liver abscess, P < 0.01; treatment × liver abscess interaction, P = 0.09.

The research aimed at determining the site and extent of digestion of high-amylase hybrids that are steam flaked may help to explain production responses observed in this experiment. Processing characteristics of EFC are likely attributable to the presence of amylase enzyme, as we have not previously encountered corn grains that respond in this manner to flaking.

Conclusions

High-amylase corn could be used advantageously by producers to reduce production costs associated with steam flaking. Implications including reduced steam use, reduced grain processing (bulk density), and 50% increased mill throughput (reduced labor) are all potential benefits that occur prior to feed being delivered to bunks. Improvements in microbial digestion, ADG, feed efficiency, HCW, and liver abscess mitigation are potential benefits associated with feeding high-amylase corn.

Glossary

Abbreviations

ADF

acid detergent fiber

ADG

average daily gain

BW

body weight

CC

cracked corn

CON

control grain or mill-run, which is grain of unknown genetic characterization

DM

dry matter

DMI

dry matter intake per day

DRC

dry-rolled corn

EFC

Enogen Feed Corn, a genetic modification event that results in amplified expression of α-amylase in the corn kernel

G:F

average daily gain/average daily dry matter intake

HCW

hot carcass weight

HMC

high-moisture corn

ISDMD

in situ dry matter disappearance

IVGP

in vitro gas production

LMA

longissimus muscle area

MDGS

modified wet distillers grains plus solubles

OM

organic matter

VFA

volatile fatty acid

Conflict of interest statement

The authors declare that there are no conflicts of interest associated with this research.