Effects of Three Yeast Strains on In Vitro Rumen Fermentation of Corn Stover and a Total Mixed Ration

1. Introduction

With the growth of the human population, the demand for agricultural products in developing countries is expected to double by 2030 [1]. This increased demand necessitates the intensification of livestock production, prompting the use of new strategies to enhance the production of animal-derived products [2]. In ruminants, this challenge has led nutritionists to manipulate rumen fermentation to improve animal performance and health [3]. One such strategy involves using yeast additives as probiotics to enhance the fermentation of fibrous foods and nutrient absorption [4]. These additives can stimulate the growth of beneficial microbes in the rumen, particularly cellulolytic bacteria and fungi [5,6], stabilize rumen pH, reduce methane production [7,8,9], improve fermentation patterns, reduce pathogen concentrations, and increase meat and milk production [10]. Despite extensive research on the efficacy of yeast additives in ruminant diets, there are still gaps and inconsistencies in the findings. These variations can be attributed to differences in animal species, yeast genus and species, dosage, and application methods, as some strains have specific actions while others are multifunctional [11,12,13].

Institutions worldwide are conducting preliminary research on yeast strains other than Saccharomyces spp., such as Pichia guilliermondii, Issatchenkia orientalis, Candida tropicalis, and Candida norvegensis. In vitro results suggest these strains improve fermentation patterns [11,14,15]. In some cases, these benefits surpass those of commercial Saccharomyces cerevisiae, despite it being the only yeast currently authorized for commercial probiotic use in ruminants. Therefore, it is crucial to explore the advantages of different yeast genera for their potential use as probiotic feed additives in livestock systems. The primary objective of this study was to evaluate the effects of Saccharomyces cerevisiae, Pichia guilliermondii, and Candida norvegensis on rumen fermentation in vitro using corn stover and a total mixed ration as substrates.

2. Materials and Methods

2.1. Location of the Study

The study was conducted in Chihuahua, Chihuahua, Mexico, located at latitude 28°35'10.9" N, longitude 106°6'26.6" W, and an altitude of 1,440 meters above sea level. The part of the research that involved the use of animals was carried out in accordance with the regulations of the Institutional Bioethics Committee (case number: CFTZyE-ACTA-101/2015: AGREEMENT 4.2).

2.2. Experimental Design

A completely randomized design with a 4 x 2 factorial arrangement and four replications was used to study the effect of three yeast strains and a non-yeast control treatment on in vitro rumen fermentation of corn stover and a totally mixed ration (TMR) as substrates. The ingredients of the TMR are shown in Table 1.

The yeast strains used in the experiment were: Pichia guilliermondii (Levica 27), Candida norvengensis (Levazoot 15), and Saccharomyces cerevisiae (Levucell® SC 10; Lallemand Animal Nutrition). The experimental treatments are presented in Table 2.

2.3. Experimental Procedure

The test was performed in vitro in 18-mL tubes, maintaining an effective volume of 15 mL. Corn stover and TMR used as substrates were pre-dried at room temperature (27°C) under the sun and ground through a 1.0-mm mesh prior to use. The chemical composition of the substrates was determined according to [16] and is shown in Table 3. Subsequently, 0.5 g of each substrate was weighed into the 18-mL test tubes for incubation.

Rumen fluid was extracted from two fistulated Pelibuey male sheep (32 kg and approximately 12-month-old), which were kept in individual pens and fed for seven days a diet containing 40% corn stover and 60% concentrate, with free access to water. Rumen fluid extraction was performed on day eight and prior to the first feed offering (9:00 a.m.). The rumen fluid was filtered through muslin and was used to prepare the fermentation medium, which contained only filtered rumen fluid (LRF). The LRF (15 mL each) was distributed to each of four vials containing the substrate, and four tubes without substrate were used as blank controls. The tubes were prepared under a constant stream of nitrogen gas to maintain an anaerobic atmosphere then capped for incubation as indicated below.

Yeast cultures were added to the tubes just before the addition of the LRF. For the preparation of the Levica 27 and Levazoot 15 inocula, the strains were activated by two aerobic subcultures in malt extract broth (DIBICO^®, Cuautitlán Izcallí, Mexico) at 110 rpm (Orbital Shaker incubator; New Brunswick Model Innova 4000, Nijmegen, Netherlands), 30°C, and 24 hours of incubation. From the activated cultures, 10% (v/v) was used as inocula for 50 mL of malt extract broth (DIBICO^®) in 100-mL flasks then inoculated again at 30°C and 110 rpm (Orbital Shaker incubator) for 24 hours. From these cultures, 0.5 mL (equivalent to 10 g of yeast/adult animal daily) was added to the corresponding tubes for in vitro rumen fermentation. The cultures of Levica 27 and Levazoot 15 had final concentrations of 5.23 x 10⁹ and 1.32 x 10¹⁰ cfu/mL, respectively, in each tube. In the case of Levucell® SC 10, 1 mg of the commercial product equivalent to 10 g recommended for an adult ruminant was added to respective tubes. Finally, the tubes were immediately capped and incubated at 39°C and 110 rpm (Orbital Shaker incubator). After 24 hours, the tubes were placed on ice to stop fermentation and prepared for the analysis of the samples.

2.4. Variables Evaluated

pH was measured using a Hanna Instruments Model HI 9017 (Arvore-Vila do Conde, Portugal) and NH₃-N concentration was measured by visible UV spectrophotometry (Varioskan Flash Thermo Scientific v4.00.53, Vantaa, Finland) following the method of Broderick and Kang [17]. Total gas production was measured using a FESTO® pressure transducer (Siemens, Munich, Germany). A 1-mL gas sample was taken from each tube to determine its composition by gas chromatography using a GOW-MAC Series 580 chromatograph equipped with a Carbosphere® 80/100 packed column (5682PC) (GOW-MAC Instrument Company). Carbon dioxide was used as the carrier gas at a flow rate of 20 mL/min to determine the individual production of hydrogen, methane, and carbon dioxide after 24 hours of sample incubation. The molar concentration of volatile fatty acids (VFA) was determined by gas chromatography with flame ionization detection (Claurus 400® gas chromatograph; Perkin Elmer, Shelton, CT, USA) with a Varian capillary CP-wax58 (FFAP) CB column (15 m x 0.53 mm, 0.5 μm). The injected sample volume was 0.6 μL, which was previously conditioned with meta-phosphoric acid according to [18].

2.5. Statistical Analyses

Analysis of variance was performed using the SAS GLM procedure (Statistical Analysis System, version 9.3; Cary, NC, USA). The model equation that was fitted is as follows (Equation (1)):

y_ijk = µ + α_i + β_j + αβ_ij + ε_ijk

(1)

where $y_{ijk}$ is the measured response variable, $\mu$ is the general mean, ${\alpha }_i$ is the fixed effect of the treatment (i = 1, 2, 3, 4), ​ ${\beta }_j$is the fixed effect of diet (j = 1, 2), ${\alpha \beta }_{ij}$is the interaction effect between treatment and diet, and $e_{ijk}$ is the random error term.

3. Results

Table 4 shows the effect of yeast on gas production and pH. There was no interaction between diet and yeast strain (P > 0.05) for total gas production (TGP) or its composition (H₂, CO₂, and CH₄). However, these variables were different (P < 0.05) for each type of diet.

The molar concentration of hydrogen in the corn stover diet exceeded (P < 0.05) that obtained with the TMR. Conversely, the molar concentrations of CO₂ and CH₄ were higher (P < 0.05) with the TMR than with corn stover, demonstrating the advantage of using a balanced diet for the appropriate growth and activity of ruminal microorganisms.

There was no interaction (P > 0.05) between diets and yeasts for pH, but differences (P < 0.05) were obtained among the studied strains when the TMR was used. The yeast strain S. cerevisiae resulted in a higher pH value compared to the strains of the genera Pichia and Candida. However, no difference (P > 0.05) was found between any of these strains and the control treatment. With corn stover, no yeast effect (P > 0.05) was found.

Table 5 shows results for VFA and NH₃-N concentration, showing interaction (P < 0.05) between yeast strains and diets evaluated. The inclusion of the three yeast strains increased the molar concentration (P < 0.05) of total VFA and acetic, propionic, and butyric acids in ruminal fluid. The C₂:C₃ ratio remained unchanged (P > 0.05) when corn stover was used as the substrate and decreased (P < 0.05) with the inclusion of the Levica 27 and Levazoot 15 strains in the TMR treatment. In the case of the commercial product Levucell® SC 10, this ratio remained similar (P > 0.05) to the control treatment.

NH₃-N levels (Table 5), there was an interaction between yeast strains and diets used in this study. The lowest (P < 0.05) molar concentration was obtained with corn stover (6.1 mM) and was similar among yeast strains with this substrate. However, with the TMR, NH₃-N concentration increased to values between 47 mM and 66 mM. Additionally, differences (P < 0.05) for NH₃-N were observed among yeast strains. With Levica 27 and Levazoot 15, NH₃-N concentration was higher than the control treatment, but with the commercial product Levucell® SC 10, this concentration was lower (P < 0.05) than the control treatment.

4. Discussion

4.1. Gas Production and Composition

Yeasts have the ability to alter the fermentation process in the rumen, resulting in variations in gas production and composition. These variables are commonly used to evaluate the efficiency of ruminal fermentation [19]. Generally, the use of yeasts in ruminant feeding is associated with improvements in dry matter (DM) and structural carbohydrate digestibility, which leads to higher gas production, an expected outcome in this study. However, no variation in gas production was found when yeasts were included. In agreement with our results, González et al. (2023) [15] found no effects on in vitro dry matter digestibility or in vitro neutral detergent fiber digestibility when adding a strain of yeast P. guilliermondii (Levica 27) to the in vitro fermentation of corn stover. The effectiveness of yeasts as additives for dry matter and structural carbohydrate digestibility has given variable results and can be influenced by several factors, including the characteristics of the diet used [20,21], the yeast strain employed [15], and the dosage used [22]. Results are also variable regarding the influence of diet on the effect of yeasts in rumen fermentation. Roa et al. (1997) [23] reported that the best results were obtained when yeasts were used with high-quality forages. In contrast to our results, previous studies by Marrero et al. (2014) [24] and Castillo et al. (2016) [14] reported significant increases (P<0.05) in gas production when using Candida norvegensis yeast strains Levica 25 and Levazoot 15, respectively, in the in vitro fermentation of fibrous substrates. Those authors attributed the results to yeasts in the rumen improving the degradation rates of structural carbohydrates and therefore gas production.

Total gas production was higher when TMR was used. This variable increased from 47 mL of gas with corn stover to 94.8 mL with TMR, likely due to the better nutritive value of the latter. Corn stover is a low-quality fibrous substrate that might not meet the requirements of ruminal microorganisms, generating low fermentative capacity. Supporting this statement, Ikhimioya (2008) [25] mentioned that for optimal microbial growth in the rumen and consequent high fermentative activity, a minimum of 80 g/kg of DM of crude protein is required, whereas the protein content in the stover used in this study was 59 g/kg of DM.

The inclusion of yeasts can not only increase total gas production but also cause changes in the composition of gases produced by improving rumen efficiency [26]. Studies evaluating the effect of yeasts on methane production vary significantly, with some reporting reductions [27,28], others increases [22,29], and still others no changes [30]. In the current study, yeast inclusion had no effect on methane production in either of the two diets used. The increase in CH₄ production with TMR compared to the corn stover diet was possibly due to greater substrate degradation, a result that aligns with results presented by Benchaar et al. (2024) [22].

4.2. pH

The pH in the stover diet with yeast inclusion showed values close to neutrality (6.4), consistent with the pH observed when low-quality fibrous substrates are supplied. These observations are in line with the results reported by Marrero et al. (2006) [31], working with S. cerevisiae and Levica 25, and Ruiz et al. (2016) [32], using C. norvegensis, when studying the effects of yeast on the in vitro ruminal fermentation of fibrous substrates. The lack of difference observed with this type of diet when yeasts were included may be due to the establishment of methanogenic bacteria populations in the rumen when high-fiber diets are used. These bacteria utilize H₂ and CO₂ for methane formation, and their utilization of these gases, especially H₂, contributes to ruminal pH stabilization [33]. Similar results, with no differences in pH when yeasts were added to fibrous substrate fermentation, have been reported by Díaz et al. (2017) [21] and Anjum et al. (2018) [34].

With TMR as substrate, pH values ranged from 5.6 to 5.8, with the highest value obtained with the S. cerevisiae strain. The effects of yeast cultures on ruminal pH reported in the literature are variable and mainly depend on experimental conditions [35], with the primary factors being the composition of the diets evaluated [8,21,34,36] and the yeast species used [32]. Authors like Chaucheyras-Durand and Fonty (2008) [37] demonstrated that yeasts, when high-concentrate diets are used, can mainly increase ruminal pH or reduce its variability due to a decrease in lactate concentration. These outcomes are attributed to the interactions between yeast cells and lactate-metabolizing bacteria [20].

4.3. Volatile Fatty Acids

Consistent with the results from the current study, González et al. (2023) [15] evaluated the P. guilliermondii strain Levica 27 with corn stover as a substrate and found increases in the molar concentrations of total VFA, acetic acid, and propionic acid in ruminal fluid and a decrease in the C₂:C₃ ratio within a range of 6 to 12 hours of incubation. Additionally, Ruiz et al. (2016) [32], studying the effect of a C. norvegensis on the in vitro ruminal fermentation of oat straw, reported increased molar concentrations of acetic, propionic, and butyric acids compared to the control.

However, Marrero et al. (2006) [31] did not report any effect on the molar concentration of total VFA when evaluating different yeast strains in the ruminal fermentation of cannulated Holsteins consuming fibrous diets. Similar to the latter study, Moya et al. (2009) [38] also did not observe differences with the inclusion of a S. cerevisiae in the molar concentration of VFA or the C₂:C₃ ratio in cannulated Holstein heifers consuming a transition diet.

In the present study, the three yeast strains increased the molar concentration of butyric acid. This acid is an important energy source for the enterocyte, stimulating its proliferation, health, and function, which leads to better feed utilization by the animal [39]. Increases in total and individual VFA concentrations are indicators of enhanced degradation of both the substrate and the yeast cultures added in each treatment. In the rumen environment, yeasts have a short lifespan and are believed to degrade or pass to the lower parts of the digestive tract a few hours after supplementation [40,41]. Once degraded, the basic components of yeasts can contribute to VFA production in the rumen [22]. In the current study, 1 mg of yeast was added to each tube. This additional substrate likely contributed to the increase in total and individual VFA concentrations, although this same contribution of the yeasts was not observed in the gas production variables.

The decrease in the C₂:C₃ ratio of the treatments with Levica 27 and Levazoot 15 compared to the control found in this study is explained by the higher molar concentration of propionate in the yeast treatments. Additionally, this increase resulted in changes in the molar proportion of VFA in the rumen. Similar results were reported by Lila et al. (2004) [30] and Ruiz et al. (2016) [32], who noted that one of the main effects of yeast cultures is to increase the concentration of propionate at the expense of acetate concentration. This outcome could be due to the peptide fractions of yeast cells that stimulate the growth of Megasphaera elsdenii, the main lactate-consuming microorganism in the rumen, which uses lactate to produce propionate through the acrylate pathway [36]. Additionally, yeast cultures provide vitamins (thiamine), glucans, mannanoproteins, and organic acids that stimulate the growth of fiber-digesting and lactic acid-consuming microorganisms [42].

4.4. NH₃-N Concentration

The NH₃-N concentration obtained with corn stover aligns with other studies, with values ranging from 5 to 12 mM when fibrous diets were used [32,43]. Levica 27 and Levazoot 15 increased NH₃-N levels in ruminal fluid, consistent with the results observed by Díaz et al. (2017) [21], who reported increases in NH₃-N concentration when adding a yeast hydrolysate to ruminal fluid using a 50:50 forage-to-concentrate ratio as substrate. In agreement with our results using Levica 27 and Levazoot 15 strains, Oeztuerk (2009) [44] observed that the increase in NH₃-N concentration in fermenters produced by a live yeast culture was greater than that caused by the same autoclaved cultures. He concluded that the difference was associated with the stimulation of the proteolytic activity of rumen bacteria by the live yeast culture. This author also found higher ammonia concentrations in the presence of different doses of live yeasts in two in vitro fermentation systems and attributed the result to microbial degradation of yeast cells due to their high protein content [44]. In contrast, Anjum et al. (2018) [34] stated that the addition of S. cerevisiae decreased the concentration of ammonia during buffalo ruminal fermentation. This result aligns with results reported in the current study with the S. cerevisiae strain Levucell® SC 10, where the molar concentration of NH₃-N decreased compared to the control. Some authors consider the effect might be due to increased microbial protein synthesis [42].

5. Conclusions

The results suggest that the yeast strains used in this study could improve the energy utilization of feed for production purposes due to the increased gluconeogenic potential of the diet as reflected by increased propionate production. It is well known that propionate is the only VFA that contributes to hepatic gluconeogenesis, and it is also the most energy-efficient because its production is indirectly related to methanogens through the use of metabolic hydrogen [43,45].

Among the strains evaluated, Pichia guilliermondii and Candida norvegensis gave similar or slightly better values to Saccharomyces cerevisiae for VFA production for both diets studied. Therefore, the use of Pichia guilliermondii and Candida norvegensis strains in livestock systems should be further studied and considered to achieve better and optimal production results.