Evaluation of different inclusion levels of dry live yeast impacts on various rumen parameters and in situ digestibilities of dry matter and neutral detergent fiber in growing and finishing beef cattle

Abstract

This study evaluated the effects of supplementing dry live yeast (LY; Saccharomyces cerevisiae) on in vitro gas production (IVGP) fermentation dynamics, pH, and CH₄ concentration at 48 h, and in situ rumen parameters and digestibility of DM (DMD) and NDF (NDFD) of growing cattle during 3 feeding phases: grower (GRW) for 17 d (38% steamed-flaked corn; SFC), transition (TRANS) for 15 d (55.5% SFC: 1.2 Mcal/kg NE_g), and finisher (FIN) for 13 d (73% SFC: 1.23 Mcal/kg NE_g). Twenty British-crossbred, ruminally cannulated steers (183 kg ± 44 kg) 6 mo of age were blocked by weight into 5 pens containing Calan gate feeders and received a control (CON) diet (17.2% CP, 35.8% NDF, 86.7% DM) without LY on days −12 to 0. After that, animals were randomly assigned to treatments (TRT), 5 animals per TRT: CON or LY at inclusion rates of 5 g/d (LY1), 10 g/d (LY2), or 15 g/d (LY3) top dressed every morning at 0800 for 45 d. The DMD and NDFD were assessed during 7 separate collection days using in situ nylon bags containing 5 g of GRW, TRANS, or FIN diets, incubated at 1200 for 48 h. Protozoa counts (PC) were determined during 5 collection periods. Data were analyzed as a repeated measure within a randomized complete block design, assuming a random effect of the pen. For GRW, TRT altered the total gas production of the nonfiber carbohydrate (NFC; P = 0.045) and the fractional rate of degradation (kd) of the fiber carbohydrate (FC) pool (P = 0.001) in a cubic pattern (P ≤ 0.05): LY2 had the most gas production and fastest kd. TRT also influenced DMD (P = 0.035) and NDFD (P = 0.012) with LY2 providing the greatest digestibility. For TRANS, TRT tended to affect the NFC kd (P = 0.078) and influenced pH (P = 0.04) and DMD (P < 0.001) in which LY2 yielded the fastest kd, highest pH, and greatest DMD. For FIN, there was an effect of TRT on total gas production (P < 0.001) and kd (P = 0.004) of the NFC pool, FC kd (P = 0.012), in vitro CH₄ concentration (P < 0.001), PC (P < 0.001), DMD (P = 0.039), and NDFD (P = 0.008). LY1 had the highest PC and provided the greatest DMD and NDFD. LY2 had the fastest kd of both the NFC and FC pools and had the least CH₄ concentration. LY3 had the greatest NFC gas production. No specific dose–response pattern was observed, but 10 g/d provided the most beneficial result for all diets. We concluded that supplementation with LY affected IVGP as well as ruminal parameters and digestibilities.

Introduction

The continued escalation of livestock production will probably continue due to the increasing worldwide demand for livestock products. Researchers are continually investigating ways that beef cattle production can become a more efficient and economical process. Meat yields have been improved by supplementing livestock with different sources of feedstuffs and feed additives that provide not only appropriate levels of protein, vitamins, minerals, and energy but also adequate animal health. Due to current trends in consumer preferences and government regulation through directives such as the Veterinary Feed Directive, interest has been sparked in finding alternative means where we can still receive similar results as the currently medicated feed additives. This directive has increased the use of alternative additives in animal nutrition. One of these alternative sources, direct-fed microbials have been consistently investigated because they have been shown to improve animal performance due to their ability to modify the rumen environment and overall function (Tedeschi et al., 2011). Of these direct-fed microbials, live yeast (LY) is one of the most studied, specifically Saccharomyces cerevisiae. The effect of LY has been extensively studied in dairy cattle (Desnoyers et al., 2009). The role of LY in ruminants is not clearly defined, but it is suggested to improve DM digestibility (DMD) and stabilize ruminal pH, thus increasing growth performance and health in intensive feeding systems (McAllister et al., 2011; Crossland et al., 2018, 2019). Saccharomyces cerevisiae live cells have been shown to stimulate fermentation of mixed ruminal microorganisms (Lila et al., 2004). These findings are thought to occur because of the alteration of fermentative pathways from lactate to propionate by increasing the lactate-utilizing and cellulolytic bacterial populations (Chaucheyras et al., 1996; Lila et al., 2004). It has been found that yeast assists with digestion and metabolism of feedstuffs in ruminants in multiple aspects such as increasing nutrient digestibility, optimizing proportions of volatile fatty acids, decreasing ruminal ammonia nitrogen, palliation of pH fluctuation, and stimulation of microbial populations (Chaucheyras-Durand et al., 2008).

In addition, it has been proven to provide various growth factors, pro-vitamins, and other stimulants to rumen microorganisms (Jouany, 2006). Moreover, S. cerevisiae is said to have the ability to decrease the redox potential of the rumen (Marden et al., 2008) and promotes a more favorable environment for the development of microorganisms, mainly cellulose consumers, which maximize the degradation rate of fiber (McAllister et al., 2011).

The effects of such direct-fed microbials on beef cattle under feedlot conditions are not as well investigated as is the inclusion of these additives in the diets of dairy cattle. Therefore, the objective of this study was to determine the benefits of supplementing LY to growing beef cattle receiving 3 consecutive feedlot diets, grower (GRW), transition (TRANS), and finisher (FIN) phases, when examining multiple rumen parameters and in situ DMD and NDF digestibility (NDFD).

Materials and Methods

Animals were cared for according to the guidelines of the Texas A&M University Institutional Care and Use Committee (IACUC protocol #2018-0039). Given the results from a preliminary study (Cagle et al., 2018), the present work was designed to study the effects of the selected levels of inclusion of LY (S. cerevisiae Sc47 CNCM I-4407, Actisaf HR+, Phileo Lesaffre Animal Care, Milwaukee, WI, 1.10¹⁰ CFU/g) in diets of 20 growing beef cattle. Rumen pH, protozoa counts, DMD, and NDFD were determined from a 45-d in situ trial. Concurrently, rumen fluid was collected and used on 48-h in vitro fermentation assays to determine the total gas production as well as the fractional rate of degradation (kd) and methane concentration (CH₄) using the in vitro gas production (IVGP) technique.

Animals, Equipment, and Feeding Regimen

Twenty British-crossbred, ruminally cannulated steers (183 kg ± 44 kg) 6 mo of age from Texas A&M AgriLife Research Center in McGregor, TX, were used in this experiment. Cattle were blocked by weight resulting in 2 pens of heavyweight steers, 1 pen of medium weight steers, and 2 pens of lightweight steers that were housed in 5 separate pens of 4 animals each. Each pen contained Calan gate feeders (American Calan, Northwood, NH), and water was always accessible. On day −12, cattle were fitted with the Calan sensor. The Large Ruminant Nutrition System (http://www.nutritionmodels.com/lrns.html; accessed on June 20, 2018; Tedeschi and Fox, 2018) was used to formulate all diets using the following ingredients: medium chopped alfalfa hay, bermudagrass hay, dried distiller’s grain, steamed-flaked corn, and a mineral supplement (Table 1). From days −12 to −1, animals were adapted to the grower diet (17.21% CP, 35.8% NDF, 86.7% DM) without LY supplement in the Calan gate feeders so they could become acclimated to their individual feed bunks in the Calan gate system and adjusted to a total mixed ration. Beginning on day 0, each of the diets was fed sequentially as follows: grower (GRW) for 17 d fed during weeks 1 and 2, transition (TRANS) for 14 d fed during weeks 3 and 4, and finishing (FIN) for 14 d during weeks 5 and 6 of the study (Table 1).

Table 1.

Ingredient and chemical composition of diets fed to steers during each growing period

Items	Diets1
	Grower	Transition	Finisher
Ingredients, % of the diet AF
Alfalfa hay, medium chopped	25	16.15	7.3
Bermudagrass hay	8.0	7.6	7.2
Dried distiller’s grains	22	15.5	9.0
Steam-flaked corn	38	55.5	73
Minerals	1.0	1.0	1.0
Limestone	1.0	1.0	1.0
Urea	0.5	1.0	1.5
Molasses	4.5	2.25	0.0
Chemical composition2, % DM
DM	86.7	90.7	87.3
CP	17.2	18.0	16.1
SP, % CP	31.7	31.0	35.4
ADIN	1.9	1.7	1.2
NDIN	2.4	2.1	1.8
ADF	24.8	13.9	10.7
NDF	35.8	23.8	21.9
Lignin	6.0	3.7	2.7
Sugar	6.5	3.7	2.1
Starch	24.7	37.3	55.2
Fat	3.7	4.3	2.9
Ash	6.5	6.1	4.5
Ca	1.2	1.0	0.70
P	0.44	0.46	0.38
Mg	0.29	0.23	0.16
K	1.1	0.95	0.71
S	0.36	0.29	0.19
Na	0.23	0.18	0.13
Fe	221.0	329.7	256.0
Mn	65.0	63.3	47.0
Zn	63.0	58.7	40.0
Cu	26.0	18.0	13.0
Cl	0.44	0.32	0.30
TDN	68.8	76.5	78.4
NEm (Mcal/kg)	1.6	1.8	1.9
NEg (Mcal/kg)	1.0	1.18	1.23

¹Two hundred milligrams of monensin was supplemented to every animal during the grower and transition feeding phases, and 360 mg was supplemented to every animal during the finisher feeding phase.

²Chemical analyses and the calculation of TDN, NE_m, and NE_g were determined and estimated by Cumberland Valley Analytical Services (Waynesboro, PA).

Treatments and Experimental Design

In addition, on day 0, treatments were randomly assigned to animals using a randomized complete block design (5 pens; 4 treatments; 4 animals per pen). Each animal within a pen was randomly assigned to a treatment (TRT), allowing 5 animals per TRT, which were as follows: control (CON), LY1 (5 g/d), LY2 (10 g/d), and LY3 (15 g/d). Each animal was offered its weighted amount of ration twice daily at 0800 and 1700 in its corresponding bunk. Treatments were top dressed and thoroughly handed mixed during the morning feeding. Baseline ruminal contents were collected on day −1 and subsequent collections were taken every 7 d following for a total of 8 collection days throughout the trial. There were two collections during the GRW and FIN phase and 3 collections during the TRANS phase.

Rumen Sampling and Analyses

During the collection process for each TRT, whole rumen contents were obtained from the cranial, middle, and caudal compartments of the rumen through the rumen cannula, and split into 2 portions: one portion was frozen for future chemical assays and the second portion was used in the IVGP assay as described below. Through the rumen cannula, a combined rumen content (approximately 1 L) was suctioned with a rumen fluid extractor that contained a plastic tube with a strainer cap at the end to prevent a large mass of rumen particles being retrieved. Rumen fluid samples were strained through 8 layers of cheesecloth and immediately placed into individual stainless-steel thermoses. Headspace was minimizing to maintain both temperature and an anaerobic environment. Rumen fluid was immediately transported to the Ruminant Nutrition Laboratory in Kleberg Center at Texas A&M University for the IVGP technique and subsamples to be taken and stored for VFA analyses and protozoa counts to be executed at a later time.

Meanwhile, in situ pre-prepared nylon bags were placed into each animal. Rumen pH measurements of individual animals were taken from 3 separate locations: the reticulum, the dorsal portion of the rumen, and the caudal portion of the rumen. The sampling locations were approximately 16 inches from the outside of the rumen cannula opening. The pH of each rumen fluid sample was immediately recorded using a Thermo Scientific Orion A221 portable pH meter (Thermo Fisher Scientific, Waltham, MA).

Lactate and Volatile Fatty Acids

Approximately 8 mL of rumen fluid from each sample was transferred into individual falcon tubes containing 2 mL of metaphosphoric acid (2 Falcon tubes per animal) for lactate and VFA analyses, and then frozen at −20 °C. Lactate was determined following the methodology described by Barker and Summerson (1941), using a UNICO UV-2000 spectrophotometer (UNICO Instruments Co., Ltd, Shangai, China). The VFA concentration was measured by gas chromatography following the methodology described by Hinton et al. (1990), using an Agilent 6890N (Agilent Technologies, Santa Clara, CA) equipped with a capillary column (HP-FFAP, 10 m by 530 µm by 1.00 µm film thickness (Agilent Technologies). For VFA analysis, helium was used as the carrier gas at a flow rate of 27 cm/s, and the oven temperature was programmed as 100 °C for 1 min, increasing from 100 °C to 190 °C at 20 °C/min and holding 190 °C for 3 min.

In Vitro Gas Production Measurements

About 6 mL of rumen fluid subsamples from each treated animal was homogenized to make a representative sample of each TRT (30 mL). Using a portion of the treatment-specific homogenous samples as the inoculum, an in vitro anaerobic fermentation and gas production analysis (i.e., IVGP) was performed on a total of 384 samples (48 samples from each time point collection performed in duplicates using 2 separate fermentation chambers). The IVGP technique has been previously described in detail (Tedeschi et al., 2011; Tedeschi and Fox, 2018), but briefly, it utilizes an incubation chamber to mimic rumen temperature (39 °C) with a multiplate stirrer housing 24 or 36 Wheaton bottles in each chamber. Approximately 200 mg of each of the phase-specific diets (GRW, TRANS, and FIN depending on the feeding phase), ground to 2 mm, was weighed and transferred into 125-mL Wheaton bottles containing equal-sized magnetic stir bars. Samples were dampened with 2.0 mL of distilled H₂O to prevent particle scattering during subsequent CO₂ flushing to maintain an oxygen-reduced atmosphere. Meanwhile, an anoxic media (Goering and Van Soest, 1970) was continuously flushed with O₂-free CO₂, and then 14 mL was added to each bottle under CO₂. Each bottle was then sealed with lightly greased butyl rubber stoppers and closed with aluminum crimps (Bellco Industries, Vineland, NJ). Bottles were instantly placed in the 39 ºC incubator and connected to their respective pressure sensors via needle insertion. Representative rumen fluid samples from treated cattle were then again filtered through 4 layers of cheesecloth and glass wool, into a flask continually flushed with CO₂, and 4 mL of previously prepared rumen inoculum was injected anoxically into each Wheaton fermentation bottle via a needle and syringe which contained either a blank, 200 mg of alfalfa as the laboratory standard, or phase-specific diet in quadruplicates, respectively. Internal pressure was equilibrated to atmospheric pressure at time 0 by piercing rubber stoppers with a needle for approximately 5 s, before initiating recording. Once the pressure was equalized in all bottles, software recording was initialized, and atmospheric pressure was recorded at 5-min intervals for 48 h. Real-time plotting of the fermentation profile over time for each bottle was monitored for abnormalities. After 48 h of fermentation, software recording was terminated, and bottles were placed in a refrigerator (−8 °C) to cease fermentation. The headspace gas was sampled (1 mL) from each bottle and analyzed for methane concentration using gas chromatography (Allison et al., 1992). The final incubation pH was measured on the remaining rumen fluid. Then, 40 mL of neutral detergent solution (Van Soest et al., 1991) was added to each bottle, resealed, and autoclaved for 15 min at 120 °C. The undegraded fiber was then filtered gravimetrically using Whatman 54 filter paper, oven-dried at 60 °C for 48 h, and the residue was weighed to calculate IVGP DMD. All steps of the IVGP process were completed for 2 separate chambers (i.e., duplicates), which allowed for 48 samples per collection (24 in each chamber). The logarithmic 2-pool nonlinear function (Schofield et al., 1994; Tedeschi and Fox, 2018) was used to characterize the cumulative gas production. Specific variables within these 2 pools were IVGP-a (asymptote; cumulative gas production of the nonfiber carbohydrate pool; Fig. 1A), IVGP-b (factional rate of degradation of the nonfiber carbohydrate pool; Fig. 2A), IVGP-d (asymptote; cumulative gas production of the fiber carbohydrate pool; Fig. 1B), and IVGP-e (factional rate of degradation of the fiber carbohydrate pool; Fig. 2B). Adjustments to the parameters of the IVGP were conducted as proposed by Tedeschi and Fox (2018).

Figure 1.

Effects of dry live yeast on the in vitro (A) total gas production of the nonfiber concentrate pool and (B) total gas production of the fiber concentrate pool (closed diamonds and solid line = CON; closed squares and long dashes = LY1 (5 g/d); closed triangles and short dashes = LY2 (10 g/d); and crosses and dotted line = LY3 (15 g/d)). GRW diet was fed during weeks 1 and 2, TRANS was fed during weeks 3 and 5, and FIN diet was fed during weeks 6 and 7.

Figure 2.

Effects of dry live yeast on the in vitro (A) fractional rate of degradation of the nonfiber concentrate pool and (B) fractional rate of degradation of the fiber concentrate pool (closed diamonds and solid line = CON; closed squares and long dashes = LY1 (5 g/d); closed triangles and short dashes = LY2 (10 g/d); and crosses and dotted line = LY3 (15 g/d)). GRW diet was fed during weeks 1 and 2, TRANS was fed during weeks 3 and 5, and FIN diet was fed during weeks 6 and 7.

Protozoa Counts

In accordance with methods described by Dehority (1984), protozoa counts were performed without staining. The counting technique was an adaptation of the procedure described by Purser and Moir (1959). In summary, a 1-mL subsample of the original rumen fluid from each animal was added to 10-mL formalin to achieve a 1:10 dilution of the original rumen contents. A 1-mL aliquot of the formalinized sample was pipetted into a Sedgewick Rafter counting chamber using a 1-mL pipet with a 3-mm-wide orifice. Protozoa were counted at a 100× magnification with a counting grid 0.5 mm square in the eyepiece. Twenty-five evenly spaced grids from the entire chamber surface were counted for each rumen fluid sample (120 total samples). Protozoa per mL of rumen fluid were then calculated as follows: the sum of protozoa counted in all 25 grids was multiplied by the dilution factor, which was then multiplied by the multiple of the volume of a square and the total number of squares counted (25).

In Situ Ruminal Incubations

In our study, small nylon bags, 5 × 10 cm, 50-µm porosity (Ankom Technologies, Macedon, NY) were weighed, filled with 5 g of ground sample (to pass through a 2-mm screen), and sealed (Vanzant et al., 1998). Two sealed blank bags and 5 bags filled with the phase-specific diet (GRW, TRANS, FIN) were incubated in the rumen of each animal for a 48-h period each week. The empty bags were used as blanks to correct for feed particles and microorganisms that may have adhered to the nylon bags after incubation. The nylon bags were held together in a 32 × 42 cm polyester bag with a nylon zipper during the 48-h incubation. After removal from the rumen, nylon bags were rinsed with distilled water to remove large particles of rumen contents and washed through a series of ten 3-min washes cycles in a washing machine consisting of a 2-min wash and a 1-min spin (Vanzant et al., 1998). On completion, the bags were placed in a forced-air oven and dried at 55 °C for 48 h or until steady weight.

Digestibility Analyses

Dry matter digestibility

After nylon bags were removed from the forced-air oven, they were placed in a desiccator, and individual dry weights were obtained from all samples. The residual weight of each sample was determined after drying to calculate in situ DMD by dividing the residue weight by the original sample weight before incubation.

Neutral detergent fiber digestibility

After dry weights were retrieved on each in situ bag sample, they went through additional wash cycles in the ANKOM machine to determine the NDF residue. The NDFD was determined by methods described originally by Van Soest and Robertson (1980) using an Ankom 200 Fiber Analyzer (Ankom Technologies). Once dry, they were immediately placed into a desiccator until final weights were able to be taken. The NDFD was calculated as follows:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{\%}\mathrm{N}\mathrm{D}\mathrm{F}\mathrm{D}=(1-\frac{W_3-(W_1\times C_1)}{W_2})\times 100}\end{array}}$$ (1)

where W₁ is the bag tare weight, W₂ is the sample weight, W₃ is the dried weight of bag post-incubation, and C₁ is the blank bag correction or the running average of the final oven-dried weight divided by the original bag weight.

Statistical Analyses

The effect of TRT was tested using the least-square means and orthogonal contrasts, assuming an evenly spaced daily amount of LY fed to animals. The PROC IML was used to obtain the orthogonal coefficients for linear, quadratic, and cubic contrasts. The interaction between TRT, week, DMI, or the covariate was removed from the statistical model if not significant at P < 0.05. Data were considered significant at alpha level P ≤ 0.05, and tendencies were discussed at P ≤ 0.10.

In vitro analysis

The PROC MIXED of SAS (SAS Inst. Inc., Cary, NC) was used to analyze the data from representative rumen fluid from treated animals (IVGP-a, IVGP-b, IVGP-d, IVGP-e, and methane) as complete randomized block design. The incubation box was the random effect, TRT was the fixed effect, and the incubation bottle within the box was the subject.

In situ analysis

The PROC MIXED of SAS (SAS Inst. Inc., Cary, NC) was also used to analyze the data collected from animals (pH, total VFA concentration, acetate-to-propionate (A:P) ratio, protozoa, DMD, and NDFD) as complete randomized block design. The protozoa count was logarithmized before analysis. The pens were the random effect, TRT was the fixed effect, and the animal within pen was the experimental unit. The average DMI of each animal was used as a covariate for all animal variables but removed if not significant at P < 0.15. For the total VFA concentration and acetate-to-propionate analyses, the boxplot analysis was used to determine extreme outliers, which were removed from the analysis. Similarly, the initial pH and initial protozoa concentration of the animals were also used as covariates. The weeks, or time, of rumen fluid collections were analyzed as repeated-measures design using the REPEATED statement.

Results and Discussion

In Vitro Fermentation

The IVGP technique allowed for insight into the fermentative capacity of each of the adapted, representative, treated rumen fluid samples. The total gas production is shown in Table 2 and Fig. 1, and the fractional rate of degradation is shown in Table 2 and Fig. 2.

Table 2.

Effect of dry live yeast on in vitro gas production (IVGP) parameters and methane production of representative rumen fluid samples from like-treated growing steers fed 3 types of diets

Items1	Dietary treatment2, g/hd/d				SEM	P-values			Contrasts
	CON (0 g)	LY1 (5 g)	LY2 (10 g)	LY3 (15 g)		TRT	Time (T)3	TRT × T	L	Q	C
Grower
IVGP-a4, mL	5.98	7.2	6.42	7.18	0.528	0.195	0.053	0.041	0.191	0.625	0.081
IVGP-d4, mL	7.75ab	7.19b	8.27a	7.93ab	0.309	0.045	0.021	0.247	0.246	0.710	0.014
IVGP-b4, 1/h	14.3	12.1	14	13.7	1.16	0.479	0.679	0.010	0.978	0.391	0.199
IVGP-e4, 1/h	3.10a	2.78b	3.35b	3.21b	0.173	0.001	<0.001	0.095	0.063	0.371	0.001
Methane, mL	33.3	31.4	34.2	34.2	2.61	0.508	<0.001	0.615	0.409	0.540	0.264
Transition
IVGP-a4, mL	13.29	13.76	12.966	14.35	0.528	0.278	<0.001	0.127	0.313	0.387	0.151
IVGP-d4, mL	10.2a	10.5a	9.71a	9.44b	0.333	0.124	0.757	0.507	0.045	0.389	0.284
IVGP-b4, 1/h	15.3ab	15.4ab	16.2a	14.1b	0.560	0.078	<0.001	0.002	0.263	0.052	0.164
IVGP-e4, 1/h	3.31	3.51	3.5	3.55	0.121	0.509	0.020	0.703	0.197	0.542	0.637
Methane, mL	13.1	12.7	13.2	12.6	0.493	0.487	<0.001	<0.001	0.486	0.645	0.193
Finisher
IVGP-a4, mL	10.5b	9.68bc	8.39c	12.6a	0.701	<0.001	<0.001	0.005	0.113	0.001	0.051
IVGP-d4, mL	10.5	9.94	9.21	9.98	0.573	0.470	<0.001	0.003	0.381	0.252	0.509
IVGP-b4, 1/h	19.2ab	16.6b	21.2a	14.5b	1.32	0.004	<0.001	0.003	0.123	0.125	0.002
IVGP-e4, 1/h	4.20a	4.08ab	4.29a	3.81ab	0.104	0.012	0.961	<0.001	0.045	0.090	0.024
Methane, mL	9.78a	8.70b	7.14c	7.57c	0.248	<0.001	0.220	<0.001	<0.001	0.004	0.027

¹Items are variables analyzed during each feeding phase using representative rumen fluid sampled from like-treated animals.

²Dietary treatment values are given as least-squares means.

³All animals were fed for 8 wk (12-d adaptation and 7 observation periods consisting of 17 d GRW, 15 d TRANS, 13 d FIN).

⁴a = the asymptote measurement of the nonfiber carbohydrate pool (total gas production); d = the asymptote measurement of the fiber concentrate pool (total gas production); b = the fractional rate of degradation of the nonfiber carbohydrate pool; e = the fractional rate of degradation of the fiber concentrate pool.

^a,b,cMeans with different superscripts differ by P ≤ 0.05.

Total gas production

A TRT by time interaction was observed (P < 0.05) during the fermentation of the nonfiber carbohydrate (IVGP-a) when cattle were fed the GRW and FIN diet, and there was an effect of time (P < 0.001) during the TRANS diet. Overall, TRT did not affect IVGP-a during the feeding of the GRW or TRANS diet; however, TRT tended to respond in a cubic pattern during the GRW (P = 0.081) with LY1 producing the most total gas. When looking at the weeks when cattle were fed the FIN diet, there was an effect of TRT (P < 0.001) and time (P < 0.001) in addition to the interaction of the two, as mentioned above. The TRT responded in a quadratic (P < 0.001) and cubic (P < 0.051) pattern where LY3 produced the greatest amount of total gas overall. When comparing individual TRT within the same run, differences were detected in weeks 1, 2, 4, 5, and 6 (Fig. 1A). This illustrates the TRT by time interaction when comparing TRT effects across all weeks. This interaction is most likely due to the different diets being fed during each of the feeding phases (GRW: weeks 1, 2, TRANS: weeks 3, 4, 5, and FIN: weeks 6, 7). These interactions suggest that TRT responses may have different outcomes depending on the diet being fed and how long animals have been fed the specific diet (Table 2 and Fig. 1A).

During the fermentation of the fiber carbohydrate pool (IVGP-d; Fig. 1B), TRT (P = 0.045) and time (P = 0.021) affected total gas production in a cubic fashion (P = 0.014) when the cattle received the GRW diet in which LY2 had the greatest total gas production. No differences were observed overall when cattle went through TRANS (P ≥ 0.05), but TRT responded in a linear fashion (P = 0.045) in which LY1 produced the greatest amount of gas averaged over the whole feeding periods. When comparing TRT within each time period of the TRANS phase, TRT influenced total gas production in time period 4 in which LY3 had the least total gas production (P < 0.05). There was a TRT by time interaction detected (P = 0.003) as well as an effect of time (P < 0.001) during the FIN phase, but not TRT effect was observed (P = 0.47).

The total gas production is assumed to represent the digestibility of the substrate being fermented (Tedeschi and Fox, 2018). The TRT responses varied from diet to diet, but the inclusion of LY in the diets increased total gas production, suggesting greater digestibility of the substrate incubated. These findings are in accordance with Tang et al. (2008) and Elghandour et al. (2014) who reported that supplementation of a yeast culture increased the total gas production when incubating different types of diets. Although they supplemented with a yeast culture product which does not contain a high count of LY cells like a complete LY product does, they observed similar results possibly indicating yeast alone, no matter the product (culture or live), may interact with the rumen environment in some manner. In their case, the yeast culture could have also behaved as a substrate for the in vitro incubation. The intensity of this interaction may vary depending on the particular yeast product. The results we found in vitro are likely due to increased production of propionic acid caused by an improvement in rumen fermentation, which in turn decarboxylate succinate to propionate and CO₂ (Wolin et al., 1997). Unfortunately, we failed to confirm this conjecture in our in situ A:P ratio data (Table 3). The inclusion of direct-fed microbial like yeast-based products might not only improve total gas production but can also make qualitative changes in the gases produced through increasing animals and rumen efficiency to help contribute fewer negative effects on the environment (Hristov et al., 2013).

Table 3.

Effects of dry live yeast on rumen parameters and in situ digestibility of DM and NDF in growing steers fed 3 types of diets

Items1	Dietary treatment2, g/d				SEM	P-values3			Contrast P-values			Covariate P-values1
	CON (0)	LY1 (5)	LY2 (10)	LY3 (15)		TRT1	T1,4	TRT × T	Linear	Quad	Cubic	DMI1	T × DMI	TRT × DMI	TRT × T × DMI	IpH	IPC	IPC × TRT5
Grower
Protozoa6	10.4a	104a	10.1b	10.1b	0.061	0.175	0.164	0.884	0.300	0.047	0.167	0.002	—	<0.001	—	—	<0.001	0.369
pH	6.31a	6.33a	6.45a	6.26b	0.086	0.145	0.525	0.751	0.042	0.753	0.229	0.182	0.883	0.157	0.823	0.398	—	—
Total VFA, mM	88.9	76.7	78.9	86.4	7.38	0.603	0.424	0.239	0.872	0.192	0.781	—	—	—	—	—	—	—
A:P ratio	2.87	2.99	3.07	2.70	0.249	0.742	0.171	0.318	0.687	0.335	0.721	—	—	—	—	—	—	—
DMD, %	75.6	75.2	77.6	73.8	1.76	0.035	0.062	<0.001	0.058	0.556	0.035	0.797	0.005	0.042	<0.001	—	—	—
NDFD, %	81.1	80.7	82.2	79.7	1.23	0.012	0.177	<0.001	0.018	0.520	0.029	0.796	0.021	0.014	<0.001	—	—	—
Transition
Protozoa6	10.4	10.5	10.2	10.4	0.085	0.118	—	—	0.232	0.063	0.842	0.335	—	0.314	—	—	0.233	0.108
pH	6.26b	6.31a	6.44a	6.37a	0.079	0.041	0.014	0.091	0.178	0.069	0.027	0.943	0.184	0.044	0.104	0.022	—	—
Total VFA, mM	18.3	23.7	24.4	25.7	2.82	0.304	—	—	0.091	0.481	0.676	—	—	—	—	—	—	—
A:P ratio	4.14	3.94	4.12	3.59	0.291	0.528	—	—	0.273	0.581	0.417	—	—	—	—	—	—	—
DMD, %	78.4	78.5	81.3	77.0	1.89	<0.001	0.364	<0.001	0.001	0.003	0.011	<0.001	0.685	<0.001	<0.001	—	—	—
NDFD, %	84.4	84.9	86.6	84.2	1.06	<0.001	0.438	<0.001	0.002	0.032	0.031	<0.001	0.242	0.001	<0.001	—	—	—
Finisher
Protozoa6	10.2c	10.6a	10.3b	10.5a	0.048	<0.001	0.100	0.134	0.020	<0.001	0.014	0.017	—	<0.001	—	—	0.001	<0.001
pH	5.68	5.66	5.75	5.47	0.108	0.956	0.036	0.046	0.719	0.769	0.781	0.512	0.034	0.924	0.035	0.178	—	—
Total VFA, mM	63.0b	80.8ab	68.8b	91.4a	7.54	0.047	0.186	0.160	0.033	0.717	0.039	—	—	—	—	—	—	—
A:P ratio	1.96	1.82	1.52	1.53	0.347	0.667	0.542	0.542	0.268	0.804	0.728	—	—	—	—	—	—	—
DMD, %	78.0	81.5	79.9	80.0	1.76	0.039	0.135	0.009	0.145	0.022	0.620	0.277	0.045	0.053	0.030	—	—	—
NDFD, %	85.3	86.6	86.3	86.2	0.672	0.008	0.096	0.000	0.551	0.001	0.539	0.454	0.021	0.012	0.001	—	—	—

¹Items are variables collected during each feeding phase on individually treated animals. DMD = DM digestibility, IPC = initial protozoa count (log₁₀/ml), IpH = initial pH, NDFD = NDF digestibility, T = time, and TRT = treatment.

²Dietary treatment values are given as least-squares means adjusted for significant covariates.

³If there was a significant (P ≤ 0.05) interactions between dietary treatment, T, and DMI, the dietary treatment means are reported for the average of DMI and average overall runs in that feeding phase.

⁴All animals were fed for 8 wk (12-d adaptation and 7 observation periods consisting of 17 d GRW, 15 d TRANS, 13 d FIN). Protozoa, total VFA, and A:P ratio were only collected during 5 of the periods, and all other variables were collected for 7 periods.

⁵If there were significant (P ≤ 0.05) interactions between dietary treatment and DMI and dietary treatment and initial protozoa, the dietary treatment means are reported for the average of DMI and the average concentration of initial protozoa count.

⁶log₁₀/ml.

^a,b,cMeans with different superscripts differ by P ≤ 0.05.

Fractional rate of degradation

There was an interaction of TRT by time (P = 0.010) on the fractional rate of degradation (kd) of the nonfiber carbohydrate pool (IVGP-b; Table 2 and Fig. 2A) when cattle were fed the GRW diet. There was no overall effect of TRT or time independently (P = 0.477, P = 0.679, respectively), but as shown in Fig. 2A, there were some differences in TRT between week 1 and the interaction can clearly be identified by TRT responding invertedly between week 1 and 2. This finding indicates that the effects of TRT may be dependent on how long animals receive LY and how adapted they are to a particular diet. During the TRANS and FIN phase, there was an interaction of TRT and time (P = 0.002, P = 0.003, respectively) as well as an independent effect of and time (P < 0.001) in both feeding phases. The TRT also tended to affect the kd (P = 0.078) in the TRANS phase in a quadratic pattern (P = 0.052) and had a significant effect (P = 0.004) in the FIN phase with a cubic response (P = 0.002), where LY2 numerically had the fastest kd overall during both feeding phases. Figure 2A reflects the differences of the TRTs within each run and displays a clear image of how TRT responses may be dependent on how long animals are on the diet in each feeding phase.

Similar interactions were observed for the kd of the fiber carbohydrate pool (IVGP-e; Table 2 and Fig. 2B) throughout the feeding phases. An effect of TRT (P = 0.001), time (P < 0.001), and a tendency for the interaction of the two (P = 0.095), was observed during the feeding of the GRW diet as well as a cubic pattern for TRT response (P = 0.006). The LY2 had the quickest kd averaged over the entire GRW phase. When cattle went through the TRANS phase an influence of time was detected (P = 0.020), but there was no effect of TRT, contrast pattern, or a TRT by time interaction detected (P = 0.509, P ≥ 0.05, P = 0.197, respectively). The length of the transition phase in our study may also explain this lack of response. However, numerically, CON had the slowest kd throughout the TRANS phase. A TRT by time (P < 0.001) interaction, effect of TRT (P = 0.012), as well as significant cubic and linear pattern responses (P = 0.024, P = 0.045, respectively) occurred within the FIN diet. Coinciding with results in the GRW phase, LY2 resulted in the numerically fastest kd in the FIN phase as well.

The fractional rate of degradation indicates the proportion of the substrate that disappears per unit of time. Our findings may support Dawson et al. (1990) conclusion that active dry LY remains able to stimulate rumen microbes. Ando et al. (2004) experienced the same results of increase kd with the inclusion of dried brewers’ yeast, which are common cultures of S. cerevisiae species. While brewer’s yeast does not contain LY organisms like that of the LY product used in the current study, it is still adding credibility to yeast’s ability to alter the rumen environment.

Rumen Fluid Measurements

Methane

Decreasing methane emissions from ruminants without sacrificing animal production is a constant objective of ruminant nutritionists. Using LY as a means of mitigation have been studied, but most results are inconclusive or no response has been reported (Martin et al., 2010). Some studies performed have posed the idea that various yeast products might stimulate the acetogens to compete or to cometabolize hydrogen with methanogens, thereby reducing methane emissions (Mwenya et al., 2004; Elghandour et al., 2014). Overall, there was quite a variability in TRT response on methane production (Table 2 and Fig. 3), as expected by a variable that is affected by many factors simultaneously (Van Soest, 1994; Tedeschi and Fox, 2018). During the GRW phase, there was only an effect of time observed (P < 0.001) in which more methane was produced during week 2 than in week 1 (26.3 vs. 40.3 mL, respectively; P < 0.001). This was likely initiated by an increase in DMD experienced from week 1 to week 2 in the GRW phase. There was an interaction detected between TRT and time (P < 0.001) as well as an effect of time (P < 0.001) in the TRANS phase. Methane increased significantly from weeks 3 to 4 to 5 (10.6 to 12 to 12.6 mL, respectively; P < 0.001). This could be expected since there was an increase in the percentage of starch in the diet from one run to another. With an increasing percentage in starch, the rumen microbes could have adapted, and an increase in protozoa populations could have been present to aid in the digestion of the starch (as discussed next). Protozoa assist methanogens, so they could have contributed to the increase in methane from period to period. There was no recorded effect of TRT overall, but as illustrated in Fig. 3, the response varied depending on what week was observed. There was a TRT by time interaction (P < 0.001) during the FIN phase as well. There was also an effect of TRT (P < 0.001) on methane production in linear, quadratic, and cubic patterns (P < 0.001, P = 0.004, P = 0.027, respectively) in which LY had the least methane production.

Figure 3.

Effects of dry live yeast on the methane production of growing cattle (closed diamonds and solid line = CON; closed squares and long dashes = LY1 (5 g/d); closed triangles and short dashes = LY2 (10 g/d); and crosses and dotted line = LY3 (15 g/d)). GRW diet was fed during week 1 and 2, TRANS was fed during week 3 and 5, and FIN diet was fed during week 6 and 7.

Protozoa

It is very likely there was variation between samples attributable to the different DMI and initial protozoa per animal. We expected interaction between TRT and DMI and TRT and initial protozoa. In fact, each interaction was observed during all phases (Table 3). Thus, depending on the DMI and initial protozoa count, treatments may have different outcomes. On account of these outside effects and interactions, values are reported for the average of DMI and initial protozoa count. Rumen ciliate protozoa are the most numerous protozoa species in the rumen, and they readily digest starch (Michalowski, 2005; Williams, 1989). Our finding may be desirable in a high forage, GRW type diet because there is little dietary starch that needs to be slowly degraded. Because of the relationship between the number of protozoa and methane production, fewer protozoa in the rumen could be advantageous, but the reduction in methane emissions varies by diet (Hegarty, 1999). When cattle went through the TRANS period, there tended to be an interaction of TRT and initial protozoa count (P = 0.108), and TRT tended to respond again in a quadratic fashion (P = 0.063), with LY2 having the lowest count of protozoa. Animals were offered a 50% forage 50% concentrate step-up ration when these samples were collected during week 4. This is still a high-forage diet, so a lower count of protozoa could be advantageous when considering a subsequent lower production of methane. During the FIN phase, there was a significant effect of TRT (P < 0.001), DMI (P = 0.017), and initial protozoa count (P = 0.001), as well as an interaction between TRT and DMI (P < 0.001) and TRT and initial protozoa count (P < 0.001). The TRT responded most significantly in a quadratic pattern (P < 0.001) in which the CON diet had the lowest count of protozoa. These findings are in agreeance with Shen et al. (2018), who discovered that total protozoa counts were significantly greater in the rumen fluid of cattle that were supplemented with an S. cerevisiae fermentation product top dressed while receiving high starch diets. With the FIN diet of our trial being high in starch, a higher concentration of protozoa may be desirable due to the protozoa’s ability to digest more slowly than other microbes. This may aid in keeping the animal’s ruminal pH more stable and more favorable pH, reduce the likelihood of experiencing acidosis when receiving concentrate diets rich in available starch like those fed in confined feeding programs and decrease the redox potential (Slyter, 1976). Because cellulolytic bacteria are susceptible to these parameters, protozoa indirectly stimulate the bacterial cellulolytic activity and supply their activity to the rumen microbial ecosystem (Jouany and Ushida, 1999).

Ruminal pH

The digestive health of cattle and the ability for ruminants to digest feed efficiently in order to perform relies heavily on the pH of the rumen (Shabat et al., 2016), and a good understanding is a necessary prerequisite to manipulate the microbiota to optimize rumen function and productivity (Jami and Mizrahi, 2012). If a bovine’s ruminal pH drops below 5.6 for longer than 180 min, they can begin to experience subacute ruminal acidosis (Plaizier et al., 2008), which can begin to kill the rumen microflora, damage the papilla that is responsible for nutrient absorption, reduce feed efficiency, and can even become as serious as death (Owens et al., 1998). Low pH commonly occurs after an animal is fed a diet with a high percentage of starch. These high starch diets are beneficial when it comes to putting weight on cattle, and they are commonly fed in confinement feeding programs; however, it is important that cattle keep a high ruminal pH in these types of settings, so they do not experience issues as mentioned before. Due to their confinement feeding regimens, it is important that these feeding programs find sources that help maintain a healthy ruminal pH and minimize the occurrence of liver abscesses (Tedeschi and Gorocica-Buenfil, 2018). Several studies have shown that certain strains of active LY may be particularly effective at raising and stabilizing ruminal pH throughout diets that differ in their acidotic potential (Bach et al., 2007; Guedes et al., 2008; Marden et al., 2008; Crossland et al., 2019) and under thermoneutral conditions (Crossland et al., 2018). The effects of dry LY on pH are illustrated in Table 3 and Fig. 4. Although TRT only had a significant effect during the TRANS phase (P = 0.041), LY2 had a consistent higher pH throughout the feeding phases. There were no effects observed when cattle were fed the GRW diet other than TRT responding in a linear fashion (P = 0.042) with LY2 providing the highest ruminal pH. In addition to the effect of TRT, there was also an effect of time (P = 0.041), initial pH (P = 0.022), and TRT by DMI interaction (P = 0.044) as well as a tendency of an interaction between TRT and time (P = 0.091), and a 3-way interaction between TRT, time, DMI (P = 0.104) when cattle went through the TRANS phase (Table 2 and Fig. 4). There was a significant effect of time (P = 0.036), TRT by time (P = 0.046), time by DMI (P = 0.034), and a TRT by time by DMI (P = 0.035) interaction during the FIN phase.

Figure 4.

Effects of dry live yeast on the ruminal pH of growing cattle (closed diamonds and solid line = CON; closed squares and long dashes = LY1 (5 g/d); closed triangles and short dashes = LY2 (10 g/d); and crosses and dotted line = LY3 (15 g/d)). GRW diet was fed during weeks 1 and 2, TRANS was fed during weeks 3 and 5, and FIN diet was fed during weeks 6 and 7.

Volatile fatty acids

Total VFA concentration (Table 3 and Fig. 5; acetate + propionate + butyrate + isobutyrate + valeric + isovaleric, mM) was not affected by yeast TRT while cattle were fed the GRW diet (P = 0.603) or time (P = 0.424). Similarly, TRT did not significantly alter total VFA concentration during the TRANS diet (P = 0.304), but a linear trend existed (P = 0.091). However, when cattle were fed the FIN diet, yeast TRT affected (P = 0.047) total VFA concentration in a linear and cubic fashion (P ≤ 0.04) in which the LY1 and LY3 rumen fluid had the greatest concentration of total VFA concentration (80.8 and 91.4 mM, respectively; Table 3 and Fig. 5). Similar results were seen in a preliminary study (Cagle et al., 2018). Our results agree with Bakr et al. (2015) as they reported that total VFA concentrations were significantly higher in the yeast-fed animals compared with the controls throughout the study.

Figure 5.

Effects of dry live yeast on total volatile fatty acid concentration in growing cattle (closed diamonds and solid line = CON; closed squares and long dashes = LY1 (5 g/d); closed triangles and short dashes = LY2 (10 g/d); and crosses and dotted line = LY3 (15 g/d)). GRW diet was fed during weeks 1 and 2, TRANS was fed during weeks 3 and 5, and FIN diet was fed during weeks 6 and 7.

Acetate:propionate ratio

Our study found no significant changes in the A:P ratio (P > 0.10), but the average A:P ratio tended to decrease (P < 0.27) across all yeast treatments as well as the control as the cattle went from the TRANS diet to a higher concentrate diet (FIN). This follows in accordance with what Cho et al. (2014) reported when determining the effect of the energy level of the diet on the A:P ratio in the rumen of Hanwoo steers. As displayed in Table 3, the result differed as to what inclusion had the least A:P ratio throughout each diet. These are different findings that Uyeno et al. (2017) witnessed when supplementing different inclusions (0, 5, 10 g/d) of the same LY product to Holstein cows. Although there was variation, treated rumen fluid consistently had lower A:P ratios throughout all feeding phases (Table 3). Cagle et al. (2018) observed a similar trend; thus, confirming the A:P ratio patterns.

In Situ Digestibility

The numbers of rumen microbes and their activity have a direct effect on the efficiency of forage degradation (Hungate, 1966). Increased concentrations of ruminal fibrolytic bacteria have been observed to result from yeast supplementation (Wiedmeier et al., 1987; Harrison et al., 1988; Dawson and Hoppkins, 1991; Crossland et al., 2018). In the present study, the in situ digestibilities of DMD and NDFD were increased by the addition of the low to medium inclusion level of yeast throughout all feeding phases (Table 3), except that LY3 (15 g/d) had a greater oscillation than the other TRT (Fig. 6). Thus, as in agreement with the above reports, it can be concluded that the addition of yeast might stimulate specific ruminal microbes.

Figure 6.

Effects of dry live yeast on (A) dry matter digestibility (DMD) and (B) neutral detergent fiber digestibility (NDFD) in growing cattle (closed diamonds and solid line = CON; closed squares and long dashes = LY1 (5 g/d); closed triangles and short dashes = LY2 (10 g/d); and crosses and dotted line = LY3 (15 g/d)). GRW diet was fed during weeks 1 and 2, TRANS was fed during weeks 3 and 5, and FIN diet was fed during weeks 6 and 7.

Dry matter digestibility

The TRT influenced DMD in a cubic pattern (P = 0.035) in which LY2 presented the greatest DMD when cattle were fed the GRW diet. Time tended to effect DMD (P = 0.062), and there were TRT by time (P < 0.001), TRT by DMI (P = 0.005), and TRT by time by DMI (P < 0.001) interactions as well. There was more variation in TRT response during the TRANS phase than there were in the GRW or FIN (Fig. 6A), and this could be related to the way the cattle were transitioned from the GRW to the FIN diet (3 different GRW:FIN step-ups consisting of 75:25, 50:50, 25:75). An influence of TRT (P < 0.001), DMI (P < 0.001), TRT by time (P < 0.001), TRT by DMI (P < 0.001), and TRT by time by DMI (P < 0.001) on DMD was observed during the TRANS phase. The TRT responded cubically (P = 0.011) in which LY2 yielded the greatest DMD. The FIN phase resulted in effect of TRT (P = 0.039) in a quadratic response (P = 0.022) with influence of TRT by time (P = 0.009), TRT by DMI (P = 0.053), and TRT by time by DMI (P = 0.030) interactions in which LY1 had the greatest DMD. Although LY1 resulted in the greatest DMD when cattle were fed the FIN diet, LY2 and LY3 were not far behind, but as can be seen in Fig. 6A, LY2 seemed to stay very constant DMD throughout all feeding phases and had the least variation. This result could be due to the pH of the rumen being higher throughout the feeding phases, which allowed for a more favorable fermentation environment, and a more adaptive, productive microbial population (Chaucheyras-Durand et al., 2008). A less acidic and more anaerobic ruminal environment would help stimulate the growth of fiber-degrading microorganisms (Callaway and Martin, 1997) and could improve fiber degradation in the rumen (Williams et al., 1991). Although their study was done in vivo, our results are consistent with Crossland et al. (2018) in the sense that LY supplementation resulted in greater DMD. In our study, LY2 tended to yield greater ruminal pH (Fig. 4), but cautionary interpretation is needed to avoid conflict with the data presented given the considerable variation.

Neutral detergent fiber digestibility

Almost identical results to DMD were observed during each feeding phase on NDFD (Table 3 and Fig. 6B), and the same TRT response pattern with LY2 having the greatest numerical NDFD with the least variation throughout each phase minus the FIN was observed as well. This is likely to be the result of the higher pH and more productive rumen microbial population in LY2-treated cattle as previously discussed. Specifically, TRT influenced NDFD during all feeding phases (P ≤ 0.012). In addition, TRT by time, TRT by DMI, and TRT by time by DMI interactions were observed throughout all feeding phases (P < 0.05). An effect of DMI (P < 0.001) was only seen during the TRANS phase. While not reaching significance, Crossland et al. (2018) observed an increase in NDFD in LY-treated inoculum over the control. This NFDF was measured on in vitro fermentation batches, and they support that the variation between fermentation batched was the reason for it not being significant.

In summary, our results indicated that the daily supplementation of 5.0 g LY/d yielded less methane production when high-forage diets are fed, but higher protozoa counts, total VFA concentration, greater DMD, and greater NDFD might be observed when high-concentrate diets are fed. The rumen fluid from cattle supplemented with 10 g LY/d provided the greatest amount of in vitro gas production for nonfiber and fiber carbohydrate pools as well as the fastest fractional rate of fermentation for high-forage diets (i.e., GRW and TRANS phases). The supplementation of 10 g LY/d also provided the greatest DMD and NDFD during the same feeding time as well, and it provided the highest ruminal pH throughout different diets. The administration of 15 g LY/d provided no additional measurable benefits over the other inclusions during the high-forage diet (GRW phase), but it showed to have a higher total gas production and kd of the fiber carbohydrate pool as well as the least amount of methane production and highest total VFA concentration during the transition phase. In addition, the 15 g LY/d presented the greatest total gas production of the nonfiber carbohydrate pool, fastest kd of both pools, least methane production, and lowest A:P ratio during the finisher phase (high-concentrate diets). Many studies have claimed that some type of probiotic yeast supplementation can be beneficial for ruminal health and subsequent ruminal productivity, but hardly any specific conclusions are given about the optimal inclusion of LY throughout entire feeding phases like those of confinement feeding in feed yards. More titration-type studies are needed to narrow down to the optimum concentration of dietary LY supplementation. Although there is some slight variation within some variables, overall, our results indicated that daily supplementation of LY at the inclusion of 10 g LY/d may be the most optimal dosage for growing cattle being fed in confinement when considering the health and subsequent productivity of the rumen. This recommendation is based on the specific inclusion rate’s ability yield a higher pH, which commonly leads to an increase in microbial growth and an improvement in feed digestibility.