Effects of bacterial direct-fed microbial combinations on beef cattle growth performance, feeding behavior, nutrient digestibility, ruminal morphology, and carcass characteristics

Abstract

The effects of the dietary inclusion of a mixture of bacterial direct-fed microbial (DFM) on feedlot beef cattle growth performance, carcass characteristics, nutrient digestibility, feeding behavior, and ruminal papillae morphology were evaluated. Crossbred-Angus steers (n = 192; initial body weight (BW) = 409 kg ± 8 kg) were blocked by BW and randomly assigned into 48 pens (4 steers/pen and 16 pens/treatment) following a randomized complete block design. A steam-flaked corn-based fishing diet was offered to ad libitum intake once daily for 153 d containing the following treatments: (1) Control (no DFM, lactose carrier only); (2) treat-A (Lactobacillus animalis, Propionibacterium freudenreichii, Bacillus subtilis, and Bacillus licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg)/animal-daily minimum; and (3) treat-B, the same DFM combination, but with doses at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected. Data were analyzed using the GLIMMIX procedure of SAS, with pen as the experimental unit, the fixed effect of treatment, and the random effect of BW-block, while preplanned contrasts comparing Control × treat-A or treat-B were used. Steers offered treat-A had increased carcass-adjusted average daily gain (P = 0.03) by 6.7%, gain efficiency (P < 0.01) by 6%, tended (P = 0.07) to have increased carcass-adjusted final BW by 15 kg, and hot carcass weight (P = 0.07) by 10 kg, while treat-B did not differ (P ≥ 0.17) from control. Overall dry matter (DM) intake (P = 0.36) and other carcass traits (P ≥ 0.13) were not affected by treatments. Steers offered treat-A tended to have increased digestibility of DM (P = 0.07) by 3%, neutral detergent fiber (P = 0.10), and hemicellulose (P = 0.08) by 9% compared with control, while treat-B did not differ (P ≥ 0.10) from control. No treatment × period interactions (P ≥ 0.21) or main effects of treatment (P ≥ 0.12) were observed during 24-h feeding behavior. Steers ruminated, ate, chewed, and were more active (P ≤ 0.01) during the second behavioral assessment (day 113), while drinking behavior was not affected (P ≥ 0.88). Ruminal papillae morphology and ruminal ammonia concentration (ruminal fluid collected at slaughter facility) were not affected by treatment (P ≥ 0.39). Steers offered the DFM treat-A had improved growth performance and it positively affected carcass weight and nutrient digestion. The DFM combinations did not seem to affect feedlot cattle feeding behavior or ruminal papillae morphology.

Lactobacillus animalis, Propionibacterium freudenreichii, Bacillus subtilis, and Bacillus blicheniformis offered to feedlot finishing yearlings improved average daily gain and feed efficiency, positively impacting hot carcass weight and nutrient digestibility, without affecting feeding behavior and ruminal papillae.

Introduction

Continuous regulatory scrutiny is limiting beef cattle dietary antibiotic additives that may restrict beef production efficiency. The growing concern about antibiotic use in food production has challenged ruminant nutritionists to develop alternative feeding strategies targeting specific changes in cattle’s digestive physiology in order to achieve the efficacy that traditional feed additives have shown to provide. In 1989, the Food and Drug Administration defined probiotics used in livestock production as “direct-fed microbial” (DFM), which were deemed as naturally occurring microorganisms that altered ruminal fermentation and intestinal function. The use of several DFM combinations has shown a potential to positively affect ruminal function, as it is the first site of a meaningful microbial activity within the gastrointestinal tract of ruminants (Yoon and Stern 1995).

The most common bacterial strains typically incorporated into cattle diets are those that produce lactate and those that use lactate, as they work synergistically to positively influence ruminal fermentation (Ban & Guan, 2021). Lactate-producing bacteria, like Lactobacillus animalis, have demonstrated resilience in acidic conditions (Nagaraja and Titgemeyer, 2007). This adaptability enables the ruminal microbiome to acclimate to harsh environments while continuing to generate organic acids, such as lactate, which in turn fosters the growth of lactate-utilizing bacteria like Propionibacterium freudenreichii (Krehbiel et al., 2003; Wilson and Krehbiel, 2012; Yoon and Stern, 1995). The combination of both bacteria allows lactic acid to be converted into energy-efficient volatile fatty acids, like propionate, and hence maintain a safer ruminal environment for the microbial ecosystem (Retta, 2016). Many studies have observed that the combination of both types of bacteria has improved feedlot growth performance and at times carcass traits (Elam et al., 2003; Raeth-Knight et al., 2007; Vasconcelos et al., 2008).

In addition to lactate-producing and lactate-utilizing bacteria, the addition of spore-forming bacteria like Bacillus could be beneficial to cattle in the feedlot in order to further improve digestion. Bacillus strains are able to germinate, grow, and resporulate in the rumen, therefore, could provide long-lasting effects to the entire gastrointestinal tract as opposed to just the ruminal compartment (Green et al., 1999). Some species, for instance, Bacillus subtilis have been observed to promote the growth of proteolytic bacteria in the rumen, as well as greater apparent total tract digestibility of nitrogen and greater ammonia ruminal concentrations (Sun et al., 2010; Ciao et al., 2015; Deng et al., 2018). Other Bacillus species have shown the ability to produce beneficial products, such as starch and fiber-degrading enzymes (Ferrari et al., 1993). In particular, Bacillus licheniformis has been observed to improve fiber digestibility by stimulating cellulolytic bacteria in the rumen (Qiao et al., 2010), while also producing an α-amylase enzyme that specifically hydrolyzes starch (Deng et al., 2018). Together, both Bacilli species have been shown to improve the digestibility of dry matter (DM) and neutral detergent fiber (NDF) of forages, as well as improving starch digestion of concentrate sources (Pan et al., 2022).

Considering the individual benefits that each species provides to the gastrointestinal tract of ruminants, it was hypothesized that a unique combination of L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis would alter ruminal fermentation patterns and therefore provide improvements of feedlot beef cattle growth performance, feeding behavior, apparent total tract nutrient digestibility (ATTND), carcass traits, and ruminal papillae morphology.

Materials and Methods

All experimental procedures involving the use of animals were performed in accordance with the Texas Tech University Animal Care and Use Committee, ACUC Protocol # 20077-10.

Cattle receiving and experimental design

A total of 230 crossbred Angus steers single sourced from a ranch located in Plainview-Texas were received at the Texas Tech University Beef Center during the spring season. Steers were placed into pens and offered a receiving diet to ad libitum intake for ~6 wk for adaptation to the new environment, identification of animals with unwilling temperament, and to allow them to reach an appropriate body weight (BW) for the beginning of the study. Steers were tagged using a low-frequency electronic identification tag on the left ear and a numbered identification tag on the right ear. Steers were also processed using the following protocol: a Mycoplasma bovis bacterin (Myco-B one dose, American Animal Health, Inc. Grand Prairie, TX); 5-way viral respiratory vaccine (BoviShield Gold Zoetis, Florham Park, NJ), Clostridial vaccine (UltraChoice 7, Zoetis, Florham Park, NJ), Doramectin (0.5%) at 0.5 mg/kg (Dectomax Pour-on, Zoetis, Florham Park, NJ), and oral Fenbendazol (10%) at 5 mg/kg (Safe-guard, Merck Animal Health, Madison, NJ). One week prior to study initiation, steers were individually weighed and sorted into 16 (BW) blocks (three pens of four animals for each block) following a randomized complete block design, allowing them to adapt to the new grouping and small pen environment. Within cattle BW blocks, steers were randomly allocated to treatments and placed into 48 pens (15.5 m × 3 m), totaling 192 steers under experimentation.

Diets, treatments, feeding strategy, and cross-contamination precautions

The experiment was initiated during late spring after the week of adaptation to the new grouping and small pens allocation. Steers were kept on a receiving diet for an additional 5 d after the initiation of the study and then stepped up every 5 d (four step-up diets shown in Table 1) until steers reached the finisher diet within day 21. Dietary treatments consisted of the following: (1) Control—no direct-fed microbial (lactose carrier only at 2 g as-fed/steer/d;8 g/pen); (2) treat-A (L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg/steer/d minimum; provided by Chr. Hansen Incorporated., Milwaukee, WI [Research Award # 80711]); and (3) treat-B, same DFM combination, and dose as treat-A, but at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected. The DFM diluted mixtures and carrier (Control) were preweighed and stored at −20 ºC until incorporated into the diet 10 min before feeding (projected 2 g as-fed/steer/d- within 128 g as-fed preweighed aluminum bags).

Table 1.

Dietary ingredients and analyzed nutritional composition of step-up and finisher diets offered to yearling steers with or without bacterial direct-fed microbial mixture

Item	Basal diets
	Step 1	Step 2	Step 3	Step 4	Finisher
Diet inclusion, % DM
Steam-flaked corn, 335 g/L	22	32.75	43.50	54.25	65
Wet corn gluten feed, quality: sweet bran	54.1	45.67	37.25	28.62	20
Low-quality Alfalfa Hay	20	17	14	11	8
Yellow grease	—	0.75	1.5	2.25	3
Mineral vitamin supplement1	2.5	2.38	2.25	2.13	2
Limestone	1.4	1.45	1.5	1.55	1.6
Urea	—	—	—	0.2	0.4
Analyzed nutrient composition
NEm, Mcal/kg3	1.75	1.86	1.95	2.02	2.12
NEg, Mcal/kg3	1.13	1.22	1.30	1.37	1.45
Crude protein, %	18.68	17.00	15.31	14.14	12.97
Acid detergent fiber, %	10.85	10.84	8.90	7.85	7.85
Neutral detergent fiber, %	27.10	27.06	22.96	21.22	21.22
Crude fat, %	2.54	3.23	4.87	5.1	5.54
Calcium, %	0.95	0.87	0.87	0.89	0.86
Phosphorus, %	0.66	0.61	0.58	0.60	0.46
Magnesium, %	0.32	0.31	0.30	0.31	0.24
Potassium, %	1.32	1.24	1.22	1.24	0.92
Sulfur, %	0.28	0.27	0.26	0.25	0.19

¹Supplement contained (DM basis) 68.2599% carrier (cottonseed meal), 0.5% antioxidant (Endox; Kemin Industries, Inc., Des Moines, IA), 3.76% urea, 10% potassium chloride, 15% sodium chloride, 0.0022% cobalt carbonate, 0.1965% copper sulfate, 0.0833% iron sulfate, 0.0031% ethylenediamine dihydroiodide, 0.167% manganous oxide, 0.125% selenium premix (0.2%), 0.9859% zinc sulfate, 0.0099% vitamin A (1,000,000 IU/g), 0.157% vitamin E (500 IU/g), and provided (dietary) 30 mg/kg of monensin (0.75% Rumensin-90 in supplement; Elanco Animal Health, Indianapolis, IN).

²Microbial treatments (A and B) or carrier only (control) were kept under refrigeration (−20 °C) and added (calculated 2 g as-is/animal daily) to each mixer batch 10 min prior to feeding. Mixer disinfection (187 ppm of available chlorine) and flush-diet were performed between A and B treatments.

³Calculated using 2016, BCNRM, beef cattle nutrient requirement model (NASEM, 2016).

Bunk calls were made at 1400 hours every day to adjust for feed offerings if needed, targeting clean bunk management. Feed refusals (if present) were collected prior to feeding and weighed to adjust for daily intake. Feed refusals were dried in a forced-air oven (Sheldon Manufacturing, Cornelius, OR) at 100 ºC for 24 h to calculate daily DM and subtracted from the daily amount offered of diet DM, aiming to calculate the DM intake (DMI). Additional diet and ingredient samples were collected once a week and dried (55 ºC for 48 h) for further diet nutrient analyses.

Three batches of feed were made daily using a horizontal Roto-Mix mixer (Roto-Mix, Dodge City, KS) and offered to pens at 1500 hours within treatment in the following order: (1) Control, (2) treat-A, and (3) treat-B. Prior to the preparation of the feed batch for treat-B, the mixer was disinfected with a solution (3.78 L) containing 187 mg/kg of available chlorine (0.26% solution of a 7.5% commercial bleach [sodium hypochlorite]) and allowed to air-dry. In addition, 38 spare steers were strategically used within the feeding management allowing for the use of a mixer-flush-diet (right after the sanitation protocol aforementioned) to minimize the potential for cross-contamination between bacterial treatments. Lastly, all utensils (e.g., shovels, bags, and buckets) were individualized to each treatment and kept on three 4-wheel-garden carts all identified by distinct treatment color. These procedures were performed in an attempt to decrease the risk of cross-contamination during bunk cleaning or any other bunk management.

BW and carcass measurements

Unshrunk, two consecutive days, individual BW were collected on days 0, 30, 60, 90, 121, 130, and 153 before feeding. The two consecutive days of BW measurements were averaged for calculation purposes. Weights were taken using a “Cow Power 1050” Squeeze Chute (Arrow CattleQuip, Manitoba, Canada) placed on Tru-Test HD5T load cells (Tru-test Ltd, Auckland, New Zealand). On day 29 of the experiment, steers were implanted with a slow-release 200 mg of trenbolone acetate and 40 mg estradiol implant (Revalor-XS, Merck Animal Health, Madison, NJ). On day 130 (pens 25 to 48) and day 153 (pens 1 to 24) when ~65% or more steers had sufficient visual finish to grade USDA Choice, cattle were shipped to a federally inspected slaughter facility located in Friona, TX. Trained personnel from Texas Tech University coordinated with the slaughter facility to collect hot carcass weight (HCW), and the USDA-inspected packing plant camera system was used to determine yield grade, quality grade, marbling score, longissimus muscle area, and 12th rib fat thickness, while liver scores were recorded by Texas Tech University trained personnel. The dressing percentage was calculated by dividing HCW by the non-shrunk final BW. Carcass-adjusted BW was calculated from HCW divided by the average dressing percent across treatments for each harvesting group (60.93% and 60.43% for pens 25 to 48 and 1 to 24, respectively), and adjusted by a 4% shrink. Carcass-adjusted average daily gain (ADG) was calculated from carcass-adjusted final shrunk BW, initial BW, and days on feed, and carcass-adjusted G:F (feed efficiency) was calculated as carcass-adjusted ADG divided by the average DMI for the experimental period.

Apparent total tract nutrient digestibility

A digestibility assessment was conducted from days 68 to 72. During this period, refusals were collected prior to feeding, diet samples were collected from all pens during feeding, and fecal samples were collected at 0700 and 1700 hours from at least three steers within each pen, and frozen (−20 ºC). Diet and refusal samples were dried in a force-air oven (100 ºC for 24 h) to calculate the DMI of pens. A subsample of diets and refusals was dried at 55 ºC for 72 h and ground to pass a 1-mm screen using a Wiley Mill (Thomas Scientific, Swedesboro, NJ) for further laboratorial nutrient analyses. Fecal samples were composited by pen (10 samples per pen) by using ~30 g (as-fed) from each homogenized sample, dried in a forced-air oven at 55 ºC for 120 h, and ground to 1 mm for laboratorial nutrient analyses.

Both fecal and diet samples were analyzed for a dietary internal marker (indigestible NDF [iNDF]) according to Gregorini et al. (2008), Cole et al. (2011), and Krizsan and Huhtanen (2013), aiming to estimate total fecal output and calculate for ATTND. Briefly, Ankom F57 bags containing 0.5 g of sample were incubated for 288 h inside a ruminally cannulated steer that was offered a hay-based diet, followed by a tap water rinse (until clear water) and NDF analysis that included α-amylase and sodium sulfite (Van Soest et al., 1991). Bags were then given a final rinse with acetone and dried to calculate for iNDF without discounting the final residual ash. ATTND was determined as follows:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{T}\mathrm{T}\mathrm{N}\mathrm{D},\mathrm{\%}=100\times [\frac{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}.\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{e}\mathrm{d}}{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}.\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}}\times \frac{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}.\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}}{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}.\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{e}\mathrm{d}}].}\end{array}}$$

Feeding behavior

During days 106 and 113 of the experiment, a feeding behavior assessment was performed for a 24-h period, according to Ovinge et al. (2018). Personnel were trained to observe and record behavior every 5 min of each steer in every pen. Behavior recorded consisted of whether cattle were active, resting, eating, drinking, or ruminating. Chewing time was later calculated by adding time spent eating and ruminating. Refusals were collected prior to feeding to adjust for DM intake in order to calculate time spent ruminating, eating, chewing, and drinking per kilogram of DM, OM, NDF, ADF, and hemicellulose (HEM) as well as the digestible nutrient intake.

Laboratorial analyses

A subsample of the diets collected weekly were composited within period and dried at 55 ºC using a forced-air oven for 48 to 72 h. These composites were then ground to 1 mm using a Wiley Mill for nutrient analyses. Method 950.01 (AOAC, 1990) was used for laboratorial DM (100 ºC for 4 h), then organic matter was calculated by subtracting the ash residue (600 ºC for 4 h; 942.05; AOAC, 2005). Neutral and acid detergent fiber were analyzed (Ankom 200, Macedon, NY) in which the NDF procedure contained thermo-stable α-amylase and sodium sulfite then followed by an acetone rinse, and accounting for the residual ash (Van Soest et al., 1991). Other dietary nutrient analyses used for dietary description purposes only were analyzed by a commercial laboratory (Servitech, Amarillo, TX).

Rumen morphology collection and analysis

Upon harvest, ruminal tissue was collected for papillae morphology according to Daniel et al. (2006). A fragment (4 cm²) from the cranial sac of the rumen was collected within the slaughter facility at harvest, rinsed with deionized water, submerged in 70% alcohol within a whirl-pack bag, and kept under refrigeration (2 to 4 ºC) until analysis. Briefly, ruminal specimens were shaped (1 cm²), while papillae were completely counted by three trained personnel (average used for reporting), followed by the removal of 12 representative papillae of each specimen. The average area of the papillae and the area of the base of the fragment were measured using electronic scanning software (ImageJ v 1.8.0_172). Measurements and calculations were performed as described by Rush et al. (2023).

Statistical analysis

Data were analyzed using the GLIMMIX procedure of SAS (SAS Inst., Inc., Cary, NC) considering pen as the experimental unit in a randomized complete block design. Treatment was considered a fixed effect with a random effect of BW-block to evaluate intake, ATTND, growth performance, carcass characteristics, and ruminal morphology. Feeding behavior was analyzed using repeated measures with the fixed effects of treatment, phase, and their interaction considering BW block as a random effect. Covariance structures for repeated measures were chosen based on the smallest Akaike information criterion. Kenward Rogers was used to adjust for the degree of freedom bias. The same model with a non-Gaussian distribution (binomial) was used for carcass data (USDA quality grade), so the inverse link function was used to convert values back to averages reported on result tables. F-test-protected preplanned orthogonal contrasts were used to compare control treatment vs. treat-A and control vs. treat-B. Differences were considered significant at P ≤ 0.05 and tendencies were discussed when 0.05 < P ≤ 0.10.

Results

By design, initial BW was not different (P = 0.27) among treatments, in which steers weighed ~409 kg at the beginning of the experiment (Table 2). Steers offered treat-A tended (P = 0.07) to have a 15-kg greater carcass-adjusted final BW compared to control, while steers offered treat-B did not differ from control (P = 0.51). Steers offered treat-A had increased (P = 0.03) overall (day 0 to end) ADG by 6.7% and feed efficiency (G:F) by 6% (P = 0.01) compared to control. Neither ADG nor G:F were affected (P ≥ 0.17) by treat-B compared to control. A decrease of ~1.2% in DMI for steers offered treat-A and treat-B on day 30 (P < 0.01), and for treat-B on day 60 (P = 0.05) compared to control was observed, while overall (day 0 to end) DMI was not affected (P = 0.36) by dietary treatments.

Table 2.

Growth performance of steers offered a steam-flaked corn-based finishing diet with or without bacterial direct-fed microbial mixtures

	Treatment1					Contrasts
Item	Control	A	B	SEM2	F-test	CTL vs. A	CTL vs. B
Body weight, kg
Carc. adjusted3 final body weight	638	653	642	7	0.07	0.03	0.51
Average daily gain (ADG), kg
Days 0 to 30	1.47	1.5	1.47	0.064	0.93	0.73	0.98
Days 0 to 60	1.97	2.02	1.92	0.046	0.32	0.49	0.41
Days 0 to 90	1.91	2.01	1.86	0.038	0.02	0.05	0.35
Days 0 to 121	1.83	1.91	1.81	0.032	0.06	0.08	0.59
Days 0 to end (carc. adjusted)	1.62	1.73	1.64	0.033	0.03	0.01	0.66
Dry matter intake (DMI), kg/d
Days 0 to 30	8.6	8.5	8.5	0.15	<0.01	<0.01	<0.01
Days 0 to 60	9.6	9.6	9.5	0.15	0.09	0.94	0.05
Days 0 to 90	10.2	10.2	10	0.16	0.24	0.9	0.13
Days 0 to 121	10.6	10.6	10.4	0.17	0.41	0.85	0.30
Days 0 to end	10.8	10.9	10.7	0.15	0.36	0.75	0.30
Gain:feed
Days 0 to 30	0.171	0.177	0.172	0.0074	0.84	0.59	0.93
Days 0 to 60	0.205	0.211	0.202	0.0052	0.35	0.38	0.57
Days 0 to 90	0.187	0.198	0.185	0.0037	0.01	0.01	0.66
Days 0 to 121	0.174	0.181	0.174	0.0029	0.03	0.02	0.99
Days 0 to end (carc. adjusted)	0.15	0.159	0.154	0.0029	0.01	<0.01	0.17

¹(1) Control (no DFM, lactose carrier only); (2) treat-A (L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg)/animal-daily minimum; and (3) treat-B, the same DFM combination, but with doses at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected.

²SEM. standard error of the mean (n = 16 pens/treatment).

³Carc. adjusted, final body weight calculated from hot carcass weight, common harvesting group dressing percentage, and a 4% pencil shrink ([HCW/DP]*0.96).

HCW tended (P = 0.07) to be 10 kg greater for steers offered treat-A compared to control (Table 3). All other carcass variables were not affected (P ≥ 0.13) by treatment.

Table 3.

Carcass characteristics of steers offered a steam-flaked corn-based finishing diet with or without direct-fed microbial mixtures

	Treatment1					Contrasts
Item	Control	A	B	SEM2	F-test	CTL vs. A	CTL vs. B
Hot carcass weight (HCW), kg	403	413	406	4.5	0.07	0.03	0.51
Dressing percent3	60.36	60.82	60.86	0.194	0.13	0.1	0.07
Twelveth rib fat, mm	15.13	15.2	14.66	0.654	0.76	0.93	0.56
LM area, cm	101	103	100	1.2	0.14	0.12	0.76
Marbling score4	468	457	468	8.8	0.53	0.33	0.98
Calculated yield grade	2.88	2.85	2.88	0.111	0.96	0.8	0.99
Quality grade, %
Choice	90.97	84.97	84.55	8.315	0.52	0.32	0.29
Select	9.03	15.03	15.45	8.528	0.52	0.32	0.29
Liver scores, %
A+, A, A−, and condemned	16.11	6.45	13.33	4.744	0.27	0.11	0.67

¹(1) Control (no DFM, lactose carrier only); (2) treat-A (L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg)/animal-daily minimum; and (3) treat-B, the same DFM combination, but with doses at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected.

²SEM, standard error of the mean (n = 16 pens/treatment).

³Dressing percent calculated as HCW divided by the unshrunk final BW.

⁴Marbling score: 400 = small; 500 = modest; 600 = moderate.

Steers offered treat-A tended (P = 0.07) to have an increase in digestibility of DM by 2.2% (P = 0.07), NDF by 5.3% (P = 0.10), and HEM by 5.3% (P = 0.08), compared to control, while ADF digestibility was not affected (P = 0.24) by dietary treatments (Table 4). During the digestibility assessment, intakes of DM, OM, NDF, and ADF were not affected (P ≥ 0.13) by treatment.

Table 4.

Apparent total tract nutrient digestibility of beef steers offered a steam-flaked corn-based diet with or without bacterial direct-fed microbial mixtures

	Treatment1					Contrasts
Item	Control	A	B	SEM2	F-test	CTL vs. A	CTL vs. B
Total apparent tract digestibility, %
Dry matter	77.1	79.34	78.71	0.68	0.07	0.03	0.10
Organic matter	81.53	82.98	82.53	0.68	0.32	0.14	0.31
NDF	51.59	56.88	51.54	1.99	0.10	0.07	0.98
ADF	47.16	52.22	48.77	2.15	0.25	0.10	0.60
Hemicellulose	53.88	59.36	53.05	2.1	0.08	0.07	0.78

¹(1) Control (no DFM, lactose carrier only); (2) treat-A (L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg)/animal-daily minimum; and (3) treat-B, the same DFM combination, but with doses at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected.

²SEM, Standard error of the mean (n = 16 pens/treatment).

No treatment × period interactions (P ≥ 0.21) were observed for feeding behavior variables (Table 5). Overall, the time spent ruminating, chewing, eating, and drinking were not affected (P ≥ 0.17) by treatment; however, in general, steers ruminated, chewed, ate, and were more active (P ≤ 0.01) during the second behavioral assessment compared to the first. Regardless of treatment, steers spent more time (min/kg; P ≤ 0.01) ruminating, eating, and chewing during the second assessment (day 113) compared to the first (day 106). Intake (min/kg) was not affected (P ≥ 0.12) by treatment. Time spent drinking (min/d and min/kg) was not affected (P ≥ 0.81) by period or treatment.

Table 5.

Feeding behavior of beef steers offered a steam-flaked corn-based finishing diet with or without bacterial direct-fed microbial mixtures

	Treatment1				Phase			F-test			Contrasts
Item	Control	A	B	SEM1	Day 106	Day 113	SEM1	Trt	Phase	Trt × phase	CTL vs. A	CTL vs. B
Activity, min/d
Ruminating	161	168	153	6.30	151	171	4.40	0.23	<0.01	0.57	0.44	0.35
Eating	80	86	83	4	78	87	2.75	0.4	<0.01	0.37	0.18	0.5
Chewing2	241	254	235	7	229	258	5.23	0.17	<0.01	0.33	0.19	0.59
Drinking	16	17	16	2	16	16	1.40	0.74	0.9	0.44	0.85	0.58
Active	136	146	136	10	133	146	9.28	0.35	<0.01	0.21	0.2	0.97
Resting	1,047	1,024	1,053	16	1062	1020	13.72	0.11	<0.01	0.74	0.11	0.67
Rumination min/kg of intake
DM	15	15	14	0.7	14	16	0.5	0.45	<0.01	0.68	0.42	0.66
OM	16	16	15	0.7	15	17	0.5	0.43	<0.01	0.68	0.41	0.65
NDF	82	85	82	3.8	78	88	2.8	0.74	<0.01	0.70	0.56	0.88
ADF	243	246	231	11.1	225	255	7.9	0.58	<0.01	0.67	0.83	0.44
HEM	125	131	126	5.9	119	135	4.2	0.72	<0.01	0.72	0.44	0.86
Rumination min/kg of digestible intake
DM	19	19	18	0.9	18	20	0.7	0.53	<0.01	0.70	0.71	0.46
OM	19	20	18	0.9	18	20	0.7	0.49	<0.01	0.70	0.58	0.53
NDF	166	151	163	9.9	150	170	7.9	0.39	<0.01	0.79	0.20	0.81
ADF	541	478	487	32.7	471	532	24.4	0.25	<0.01	0.67	0.13	0.18
HEM	239	223	246	14.8	221	251	11.9	0.37	<0.01	0.85	0.33	0.68
Eating min/kg of intake
DM	7	8	8	0.4	7	8	0.3	0.34	<0.01	0.39	0.15	0.32
OM	8	8	8	0.4	8	9	0.3	0.33	<0.01	0.39	0.15	0.33
NDF	40	44	44	2.1	40	45	1.6	0.34	<0.01	0.39	0.22	0.19
ADF	119	126	125	6.1	117	130	4.6	0.65	<0.01	0.36	0.39	0.49
HEM	61	67	68	3.2	62	69	2.4	0.21	<0.01	0.40	0.16	0.11
Eating min/kg of digestible intake
DM	9	10	10	0.5	9	10	0.4	0.62	<0.01	0.36	0.35	0.49
OM	9	10	10	0.5	9	10	0.4	0.52	<0.01	0.37	0.27	0.44
NDF	83	77	88	5.9	78	87	4.8	0.26	<0.01	0.28	0.42	0.40
ADF	270	247	263	20.2	245	275	15.1	0.63	<0.01	0.32	0.35	0.78
HEM	119	114	133	8.8	116	128	7.2	0.12	<0.01	0.28	0.57	0.15
Chewing min/kg of intake 3
DM	22	23	22	0.9	21	24	0.6	0.35	<0.01	0.43	0.20	0.94
OM	23	25	23	0.9	22	25	0.7	0.34	<0.01	0.43	0.20	0.96
NDF	123	129	126	4.8	118	133	3.5	0.59	<0.01	0.45	0.31	0.66
ADF	362	372	356	14.0	342	385	10.1	0.66	<0.01	0.39	0.58	0.74
HEM	186	198	194	7.4	181	204	5.4	0.44	<0.01	0.48	0.21	0.39
Chewing min/kg of digestible intake 2
DM	28	29	28	1.1	27	30	0.9	0.57	<0.01	0.40	0.47	0.76
OM	29	30	28	1.2	27	31	0.9	0.48	<0.01	0.42	0.34	0.87
NDF	248	228	251	14.4	228	257	11.7	0.28	<0.01	0.35	0.20	0.85
ADF	811	725	750	49.0	717	808	36.4	0.33	<0.01	0.25	0.15	0.30
HEM	358	337	379	21.6	336	379	17.7	0.18	<0.01	0.41	0.34	0.36
Drinking min/kg of intake
DM	1	2	1	0.1	1	1	0.1	0.73	0.88	0.48	0.55	0.88
OM	2	2	2	0.2	2	2	0.1	0.72	0.88	0.48	0.55	0.87
NDF	8	9	8	0.8	8	8	0.7	0.87	0.87	0.47	0.64	0.98
ADF	24	25	23	2.4	24	24	2.1	0.84	0.88	0.48	0.80	0.73
HEM	12	13	13	1.2	13	13	1.1	0.84	0.86	0.47	0.57	0.83
Drinking min/kg of digestible intake
DM	2	2	2	0.2	2	2	0.2	0.81	0.94	0.45	0.80	0.70
OM	2	2	2	0.2	2	2	0.2	0.79	0.94	0.45	0.71	0.75
NDF	17	15	16	1.7	16	16	1.5	0.74	0.88	0.36	0.45	0.82
ADF	54	48	49	5.7	51	50	4.7	0.59	0.81	0.34	0.35	0.42
HEM	24	22	24	2.6	24	23	2.2	0.76	0.92	0.36	0.55	0.93

¹(1) Control (no DFM, lactose carrier only); (2) treat-A (L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg)/animal-daily minimum; and (3) treat-B, the same DFM combination, but with doses at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected.

²SEM, standard error of the mean (n = 16 pens/treatment).

³Chewing activities are calculated by adding time spent eating and time spent ruminating.

Ruminal tissue morphology was not affected (P > 0.39) by treatment (Table 6).

Table 6.

Ruminal papillae morphology of beef steers offered a steam-flaked corn-based finishing diet with or without bacterial direct-fed microbial mixtures

	Treatment1					Contrasts
Item	Control	A	B	SEM2	F-test	CTL vs. A	CTL vs. B
Average papillae area, cm2	0.38	0.40	0.37	0.02	0.39	0.51	0.47
Papillae number, no./cm2 of fragment	47.44	51.98	49.49	3.95	0.70	0.40	0.70
Absorptive surface area, cm2 per fragment cm2	37.38	39.30	49.49	3.21	0.87	0.62	0.95
Ruminal absorptive Surface area, %	96.99	97.04	96.76	0.29	0.74	0.91	0.55

¹(1) Control (no DFM, lactose carrier only); (2) treat-A (L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis), at 1:1:1:3 ratio, respectively; totaling 6 × 10⁹ CFU (50 mg)/animal-daily minimum; and (3) Treat-B, the same DFM combination, but with doses at 1:1:3:1 ratio. Bacterial counts were ~30% greater than the minimum expected.

²SEM, standard error of the mean (n = 16 pens/treatment).

Discussion

The objective of this study was to evaluate the effects of a combination of bacterial DFM including L. animalis, P. freudenreichii, B. subtilis, and B. licheniformis on growth performance, carcass characteristics, ATTND, feeding behavior, and ruminal morphology of feedlot beef yearling steers offered a steam-flaked corn-based finishing diet.

Growth performance. An increased final BW was also observed by Galyean et al. (2000) when steers were offered a combination of L. acidophilus and P. freudenreichii at various concentrations by an average of 2% across all treatments (treat-G: 1 × 10⁶ CFU L. acidophilus strain 45, 1 × 10⁹ CFU P. freudenreichii; treat-Y: 1 × 10⁴ CFU L. acidophilus strain 45, 1 × 10⁴ CFU L. acidophilus strain 51, 1 × 10⁹ CFU P. freudenreichii; treat-B: 1 × 10⁶ CFU L. acidophilus strain 45, 1 × 10⁶ CFU L. acidophilus strain 51, 1 × 10⁹ CFU P. freudenreichii). Greater overall ADG of ~6% for steers offered treat-A corroborates with the findings reported by Galyean et al. (2000) and Swinney-Floyd et al. (1991), which supplemented L. acidophilus (1 × 10⁸ CFU/animal-daily) and Propionibacterium strain P63 (1 × 10⁹ CFU/animal-daily) to cattle fed a 90% concentrate diet. Supplementation of the current DFM does not appear to impact DM intake. Several experiments utilizing similar combinations of DFM have also observed no discernible difference in this variable (Galyean et al., 2000; Brashears et al., 2003; Elam et al., 2003; Peterson et al., 2007; Raeth-Knight et al., 2007; Vasconcelos et al., 2008). While there was a slight decrease of ~0.1 kg/d observed in both treatments during the initial 30 d and in treat-B alone during the first 60 d in the current experiment, this change may not be deemed a substantial effect. This decrease coincided with the period when animals were undergoing a step-up process and acclimating to a high-concentrate finisher diet. With the increase in ADG and consistently unaffected DMI, a greater G:F of ~5% was observed for animals offered treat-A. An improved G:F in cattle with the supplementation of L. acidophilus was observed by Vasconcelos et al., (2008), and also, similar responses were observed by Galyean et al. (2000), which reported a 2% increase in G:F across treatment groups that were offered varying concentrations of L. acidophilus and Propionibacterium freudenreichii. Moreover, Rush et al., (2023)) reported a 7.3% improvement in G:F for cattle also offered similar treatment DFM mixtures. When cattle were offered a 90% concentrate diet (65% steam-flaked corn-based diet) with DFM containing Lactobacillus (L. acidophilus; 1 × 10⁹ CFU/animal-daily), the authors observed that DMI was not affected; however, a tendency for a 5% improvement in G:F was observed for cattle offered the DFM (Brashears et al., 2002). Although literature results are not directly comparable due to differences in bacterial species, strains, concentrations, and diet types, it seems reasonable to assume that the supplementation of current bacterial DFM mixtures to cattle does not seem to cause negative effects, while potential improvements in cattle growth performance have been more often reported. The detailed mechanism of action of DFM is still unknown; however, Krehbiel et al. (2003) suggested a combination of lactate-producing bacteria and lactate-utilizing bacteria to be used in order to see improved ADG (2.5% to 5%), while other performance variables such as DMI and G:F are less consistent across DFM studies. Such statement highlights the importance of experiments attempting to measure additional variables other than growth performance, in which potential mechanisms of action can be elicited, such as nutrient digestion and ruminal morphology.

Carcass characteristics. HCW tended to be 10 kg (2.5%) heavier for steers offered treat-A, which also experienced improved ADG and G:F. Galyean et al. (2000) offered L. acidophilus and P. freudenreichii at various concentrations, and HCW increased on average by 2.2%, similar to what was observed in the current experiment. However, Vasconcelos et al. (2008) used similar DFM treatments as those reported by Galyean et al. (2000) and did not observe differences in HCW or any other carcass characteristics. Dressing percentage, longissimus muscle area, liver scores, and USDA quality grades were not affected by treatment, in accordance with Galyean et al. (2000), Brashears et al. (2002), and Vasconcelos et al. (2008), which had offered DFM treatments to beef cattle finishing diets. An increase in ADG, while KPH, back fat, and marbling score not being affected corroborates with a greater HCW and final BW, which indicates that current bacterial DFM mixtures did not induce local (gastrointestinal tract) or any other type of energy partitioning in the carcass.

Apparent total tract nutrient digestibility. The digestibility of DM tended to be greater for both treatments A and B by 3% and 2%, respectively. In addition, NDF digestibility tended to increase for treat-A by 9%, which was also corroborated by a tendency of greater digestion of HEM by 9%. The greater digestibility of fiber could be attributed to the inclusion of Bacillus species within the microbial combinations offered. Ferrari et al. (1993) stated that Bacillus species may produce a wide variety of extracellular enzymes, like cellulase and lipases that could be beneficial to the digestion of nutrients. Qiao et al. (2010) observed that the inclusion of B. licheniformis improved fiber digestion by stimulating cellulolytic bacteria growth in the rumen of Chinese Holstein steers offered a 40:60 concentrate to forage diet which supports the tendencies for improved NDF and hemicellulose digestion in the current experiment. In an attempt to better understand post-ruminal digestion, researchers have studied the microbiota of the large intestine, which represents the second fermentation chamber in ruminants (Callaway et al., 2010; Durso et al., 2012; Mao et al., 2015; Oliveira et al., 2013; Liu et al., 2016). These studies reported that the predominant bacteria present in the hindgut are Firmicutes and Bacteroidetes which degrade carbohydrates. Fuerniss et al. (2022) evaluated the effects of a combination of four Bacilli strains (2 billion CFU combination of B. amyloliquefaciens, B. subtilis, B. pumilus, and B. licheniformis) on hindgut microbiota, and reported that the relative abundance of fibrolytic bacteria was increased for animal offered Bacilli DFM, which allows for greater capacity of fiber degradation. Improved fiber degradation has also been observed on monogastric animals when offered a combination of Bacillus spp. (Payling et al., 2017; Cai et al., 2015). In addition, B. licheniformis has been shown to specifically hydrolyze starch which, when combined with a high-concentrate diet with increased availability of starch, could also allow for quicker starch fermentation of the diet (Ferrari et al., 1993; de Boer et al., 2004; Deng et al., 2018; Pan et al., 2022).

Feeding behavior and ruminal papillae morphology. In the current study, feeding behavior was not affected by the inclusion of DFM in the diet. It was important to evaluate whether the high-concentrate diet with or without the inclusion of the DFM was causing any digestive disturbances, which could potentially affect cattle feeding patterns. As Gonzales et al. (2012) highlighted in their review, feeding behavior data can be indicative of whether an animal is undergoing ruminal acidosis through a matter of decreased eating time and decreased meal size. Ruminal papillae morphology was not affected with the inclusion of current DFM mixtures, which indicates that improved digestion could potentially occur postruminally with hindgut fermentation, as other experiments have suggested.

Conclusion

The bacterial DFM mixture of treat-A improved feedlot steer yearlings' growth performance, which positively affected carcass weight. An improvement in nutrient digestibility seems to be related to such effect, while mechanisms of action involving changes in animal feeding behavior or ruminal papillae morphology do not seem to be related to the use of these current DFM combinations in beef cattle consuming steam-flaked corn-based finishing diets.

Glossary

Abbreviations

ADF

acid detergent fiber

ADG

average daily gain

APN

average papillae number

APA

average papillae area

ASA

absorptive surface area

ATTND

apparent total tract nutrient digestibility

BW

body weight

DFM

direct-fed microbials

DM

dry matter

DMI

dry matter intake

DP

dressing percentage

EID

electronic identification

G:F

feed efficiency

HCW

hot carcass weight

HEM

hemicellulose

iNDF

indigestible nutrient detergent fiber

LM

longissimus muscle

NDF

neutral detergent fiber

OM

organic matter

PAA

papillae absorptive area

SARA

subacute ruminal acidosis

SAS

Statistical Analysis Software

Conflict of interest statement

The authors report no conflicts of interest.