Endo-1,3-β-D-glucanase activity influences in vitro ruminal fermentation of diets varying in forage:concentrate

Abstract

With growing concerns to reduce global methane emissions, there is a pressing need to explore alternative strategies to mitigate methane production in ruminant livestock species. The objectives of this study were to evaluate the effects of increasing dosage of endo-1,3-β-D-glucanase on in vitro ruminal fermentation from diets varying in forage:concentrate. Ruminal contents were obtained from 2 ruminally cannulated Angus × Holstein steers. The experimental design was a randomized complete block design with a 3 × 6 factorial arrangement of treatments. The basal substrates (DM basis) were high-forage (90% tall fescue hay, 10% corn; 90F:10C), equal forage:concentrate (50% tall fescue hay, 50% corn; 50F:50C), or high-concentrate (90% corn, 10% tall fescue hay; 10F:90C). Six doses of endo-1,3-β-D-glucanase were tested: 0, 5, 10, 25, 50, and 100 units/100 mL of ruminal inoculum. Recombinant endo-1,3-β-D-glucanase was diluted to working concentrations with 100 mM sodium acetate buffer, and then working solutions (1 mL) were added to fermentation vessels immediately before inoculum addition. Substrates were added to fermentation vessels and incubated for 48 h. There were 2 technical replicates per treatment in each run, and 3 runs to constitute the experiment. Results were analyzed with the GLM procedure of SAS. Increasing dosage of endo-1,3-β-D-glucanase did not influence (P ≥ 0.10) the rate of gas production or cumulative methane production. Cumulative gas production (mL, mL/g substrate, and mL/g substrate digested) increased linearly (P ≤ 0.03) with increasing endo-1,3-β-D-glucanase activity. Microbial biomass, true dry matter (DM) digestibility, apparent DM digestibility, and neutral detergent fiber digestibility were not influenced (P ≥ 0.21) by increasing endo-1,3-β-D-glucanase dose. The molar acetate proportion and acetate:propionate decreased linearly (P < 0.001) with increasing endo-1,3-β-D-glucanase activity. The molar propionate, butyrate, and valerate proportions in the fermentation media linearly increased (P < 0.001) with increasing endo-1,3-β-D-glucanase activity. Decreasing forage:concentrate decreased (P < 0.05) gas and methane production per gram of substrate digested and increased (P < 0.001) digestibility and total short-chain fatty acid (SCFA) concentration. These data demonstrate that endo-1,3-β-D-glucanase modulates in vitro ruminal fermentation of diets varying in forage:concentrate by altering molar SCFA proportions.

Endo-1,3-β-D-glucanase activity and decreasing forage:concentrate altered the short-chain fatty acid profile in vitro by decreasing the 2-carbon acetate proportion and increasing 3-, 4-, and 5-carbon short-chain fatty acid proportions.

Introduction

Enzymatic inhibition of methanogenesis, via 3-nitrooxypropanol or halogenated compounds, has been shown to be an effective solution for mitigating methane emissions in ruminants (Romero-Perez et al., 2014; Hristov et al., 2015; Roque et al., 2019; Gyeltshen et al., 2025). 3-Nitrooxypropanol inhibits methyl-coenzyme M reductase by binding to the active site, resulting in oxidation of the nickel ion required for activation (Duin et al., 2016). Halogenated compounds, such as bromoform, are thought to inhibit coenzyme M methyltransferase and methyl-coenzyme M reductase during ruminal methanogenesis (Romero et al., 2023; Li et al., 2025). In contrast to enzymatic inhibitors, lysozyme enzymes or anti-bacterial compounds such as monensin modify ruminal microbial populations to decrease substrate supply for methanogenesis (Biswas et al., 2016; Islam and Lee, 2019). However, these technologies target microorganisms, such as bacteria and protozoa, that are not directly responsible for producing methane in the rumen (Biswas et al., 2016; Islam and Lee, 2019; Park et al., 2019).

Although the supplementation of enzymatic inhibitors or rumen fermentation modifiers has resulted in well-documented reductions in methane emissions (Beauchemin et al., 2020), few studies have attempted to supplement enzymes that can target methanogenic archaea specifically. Some studies have shown positive responses with lytic endo-isopeptidases with pure cultures of ruminal methanogens; however, the same enzymes were not effective when evaluated in batch culture containing mixed ruminal microbial populations (Leahy et al., 2010; Altermann et al., 2018). A primary difference between bacteria and methanogenic archaea in the rumen is the composition of their cell wall: where ruminal bacteria contain peptidoglycan that is linked by β-1,4-glycosidic bonds, and some species of methanogenic archaea in the rumen contain pseudomurein, also known as pseudopeptidoglycan, which is linked by β-1,3-glycosidic bonds (Kandler and Konig, 1993). Pseudomurein is inert to the hydrolytic activities produced by bacterial cell wall hydrolases and lysozyme (Kandler and Konig, 1993). Endo-1,3-β-D-glucanase hydrolyzes β-1,3-glycosidic bonds in the internal β-1,3-glucan chain, and the activity exists in many organisms, including bacteria, yeast, fungi, and plants (Erfle and Teather, 1991). Some cellulolytic microbes in the rumen have β-1,3-glucanase activity (Erfle and Teather, 1991), but no studies have evaluated the effectiveness of endo-1,3-β-D-glucanase supplementation on the inhibition of ruminal methanogenesis.

Diet composition influences the ruminal fermentation of carbohydrates, with fermentation of high-fiber diets producing relatively more acetate and greater methane per unit of fermentable organic matter and fermentation of high-starch diets producing relatively more propionate and less methane per unit of fermentable organic matter (Johnson and Johnson, 1995). A recent meta-analysis determined that decreasing forage:concentrate resulted in a 9% reduction in methane emissions per unit of gain on average (Arndt et al., 2022). Because of changes in ruminal fermentation end-products (acetate, propionate, methane) with changes in dietary composition, it is additionally unknown how endo-1,3-β-D-glucanase influences the in vitro fermentation of diets varying in forage:concentrate. Therefore, the objectives of this experiment were to evaluate the effects of increasing endo-1,3-β-D-glucanase activity on in vitro ruminal fermentation from high-forage and high-concentrate substrates. The hypothesis of this experiment was that increasing endo-1,3-β-D-glucanase activity would result in decreased in vitro methane production across high-forage and high-concentrate substrates.

Materials and Methods

All animal procedures were approved by the University of Kentucky Animal Care and Use Committee (Protocol 2020-3546).

Experimental design and in vitro ruminal fermentation

The experimental design was a randomized complete block design with a 3 × 6 factorial arrangement of treatments. For each treatment, there were 2 technical replicate fermentation vessels each day that were averaged. The experiment was replicated on three separate days. Bulk samples of tall fescue hay and ground corn were obtained from the University of Kentucky C. Oran Little Research Center, partially dried in a forced-air oven (55 °C; 24 h), and ground to pass a 1-mm screen using a Wiley mill. The basal substrates were high-forage (90% tall fescue hay, 10% corn; 90F:10C), equal forage:concentrate (50% tall fescue hay, 50% corn; 50F:50C), or high-concentrate (90% corn, 10% tall fescue hay; 10F:90C). Substrates were combined in the proportions described on a dry matter (DM) basis. Six doses of endo-1,3-β-D-glucanase were tested: 0, 5, 10, 25, 50, and 100 units (U)/100 mL of ruminal inoculum. According to the manufacturer, recombinant endo-1,3-β-D-glucanase (E-LAMHV; Neogen, Lansing, MI) was synthesized from barley (Hordeum vulgare), and the enzymatic activity was stable from pH 3.0 to 9.0 (optimal pH = 5.0). One unit of endo-1,3-β-D-glucanase activity was equal to one micromole of glucose-reducing-sugar equivalents produced per minute in the presence of laminarin (10 mg/mL) in 100 mM sodium acetate buffer (pH 5.0; 40 °C).

Conditions of the in vitro gas production protocol were similar to those described previously (Goering and Van Soest, 1970; Marten and Barnes, 1979; Cone, 1998; Pell et al., 1998; Theodorou et al., 1998). Before sampling of ruminal contents, the buffer, macromineral, micromineral, and reducing solutions were prepared as described by Goering and Van Soest (1970). Fermentation vessels used in the current study were 250-mL coated glass bottles (Cat. #7056; ANKOM Technology, Macedon, NY). Substrates (500 mg) were pre-weighed into acetone-rinsed fiber filter bags (Cat. #F57; ANKOM Technology) and the bags were heat-sealed. Filter bags were wrapped with a custom-designed clip composed of vinyl 14 Ga wire to ensure submersion at the bottom of the fermentation vessel. Fermentation vessels were fitted with rubber stoppers (24 mm ID × 3 mm; Seal Innovations Inc., Huntington Beach, CA) containing valved in-line coupling bodies (PMCD1304; CPC, Arden Hills, MN) and capped with open-top screw caps (240746; WHEATON Lab 45 Accessory, Replacement Cap; DWK Life Sciences, Vineland, NJ).

Ruminal contents were collected from two ruminally cannulated Angus × Holstein steers (body weight = 654 ± 17.3 kg) that were housed in a vented gable roof barn with outdoor dry lot pens (2.4 m × 14.6 m) at the University of Kentucky C. Oran Little Research Center in Versailles, KY, USA. Steers were fed a corn silage-based diet (Table 1) once daily to supply 2.0 times the net energy required for maintenance. The diet was formulated to exceed requirements for ruminally degradable protein, metabolizable protein, vitamins, and minerals (NASEM, 2016). Ruminal contents (1 kg) were collected from the medial rumen (Storm and Kristensen, 2010) from each steer 5 h after feeding and pooled into an insulated container (YETI Rambler One Gallon Water Jug; Yeti Holdings, Inc., Austin, TX) for transport to the University of Kentucky Ruminant Nutrition Laboratory in Lexington, Kentucky.

Table 1.

Composition of basal diet fed to steers

Item
Ingredient composition, DM basis
Corn silage, %	72.5
Dried corn distillers’ grains with solubles, %	17.5
Finely ground corn, %	7.18
Limestone, %	1.76
Trace mineral premix, %1	0.46
Urea, %	0.33
Tallow, %	0.23
Vitamin A, D, & E premix, %2	0.02
Rumensin 90, %3	0.01
Tylan 40, %4	0.01
Chemical composition
Dry matter, %	42.6
Total starch, % of DM	38.9
Neutral detergent fiber, % of DM	25.4
Acid detergent fiber, % of DM	16.1
Crude protein, % of DM	10.0
Crude fat, % of DM	4.51
Ca, % of DM	0.84
P, % of DM	0.30
Net energy for maintenance, Mcal/kg5	1.77
Net energy for gain, Mcal/kg5	1.14

¹Contained: 56.34% Cl, 36.53% Na, 1.2% S, 0.06% Ca, 9.29 g Fe/kg, 5.52 g Zn/kg, 4.79 g Mn/kg, 1.84 g Cu/kg, 120 mg I/kg, 68.9 mg Co/kg, and 18.5 mg Se/kg on a DM basis.

²Composed of vitamin A acetate (1,814 kIU/kg), D-activated animal sterol (source of vitamin D₃; 363 kIU/kg), vitamin E supplement (227 IU/kg), roughage products, calcium carbonate, and mineral oil.

³Contained 199 mg monensin per gram of premix.

⁴Contained 18.1 g tylosin phosphate per kilogram of premix.

⁵Calculated from tabular values (NASEM, 2016).

Mixed ruminal contents were blended under CO₂ headspace for 30 s, squeezed through four layers of cheesecloth, and added to the inoculum. Fermentation vessels were gassed with CO₂ for 20 s to facilitate an anaerobic headspace. Recombinant endo-1,3-β-D-glucanase was diluted to desired working concentrations with 100 mM sodium acetate buffer (pH = 5.0) with 1 mg/mL bovine serum albumin. One milliliter of endo-1,3-β-D-glucanase working solutions were added to the fermentation vessel immediately prior to inoculum addition. Fermentation vessels were dispensed with buffered inoculum (99.1 ± 2.43 g) using a peristaltic pump (FlexiPump Pro; Interscience International, Woburn, MA) and secured into a 39 °C recirculating (IC-400 Standard Immersion Circulator; Antylia Scientific, Vernon Hills, IL) water bath with platform clamps (EPM1190-9002; MilliporeSigma, Burlington, MA). After the addition of all fermentation vessels to the water bath, the valves of the fermentation vessel caps were opened simultaneously to release any accumulated pressure. Then, valves of the fermentation vessels were connected to pressure transducers (Cerabar PMC21; Endress + Hauser, Greenwood, IN) with tubing (MFLX06404-14; Masterflex L/S Precision Pump Tubing) and cumulative gas pressure was measured in 5-min intervals over a 48-h incubation period with Memograph M RSG45 Data Manager (Endress + Hauser, Greenwood, IN).

Sample collection and analysis

Samples of the basal diet (Table 1) and basal substrates (Table 2) were collected for nutrient analysis, including DM, starch, crude protein and fat, neutral detergent fiber (NDF), acid detergent fiber, and minerals by the Dairy One Forage Laboratory (Ithaca, NY). Diet samples were partially dried at 60 °C for 4 h in a forced-air oven (NFTA 2.2.1.1.) and then ground to pass a 1-mm screen using a Wiley mill. DM concentration was determined by oven-drying for 3 h at 105 °C (NFTA 2.1.4.). Diet samples were prepared for total starch analysis (method 2014.10) by gelatinizing the sample in water, followed by a 2-step enzymatic hydrolysis with α-amylase and glucoamylase in acetate buffer (AOAC, 2023). Free glucose concentration was measured using the glucose oxidase electrode of a YSI Series 2950 D-1 Biochemistry Analyzer (YSI Inc., Yellow Springs, OH) and then multiplied by 0.9 to convert to anhydroglucose, as it occurs in starch (McCleary et al., 1994). Nitrogen concentration was analyzed by combustion (AOAC, 1999; method 990.03) using a CN628 Carbon/Nitrogen Determinator (Leco Corporation, St. Joseph, MI). Crude protein concentration was calculated by multiplying N concentration × 6.25. Crude fat concentration was determined with ether extraction using an ANKOM X15 Extractor (ANKOM Technology Method 2, Macedon, NY). NDF and acid detergent fiber concentration were determined sequentially using the filter bag technique (ANKOM Technology Methods 14 and 15, respectively) with an automated fiber analyzer (ANKOM DELTA; ANKOM Technology, Macedon, NY). Samples were digested in 50 mL MARSXPress vessels (CEM Corporation, Matthews, NC) using a MARS 6 Microwave Digestion System (CEM Corporation, Matthews, NC), and mineral concentrations (Ca and P) were determined using inductively coupled plasma-optical emission spectroscopy (iCAP PRO XP ICP-OES; Thermo Fisher Scientific Inc., Beverly, MA).

Table 2.

Composition of the high-forage (90F:10C; 90% tall fescue hay, 10% corn), equal (50F:50C; 50% tall fescue hay, 50% corn), and high-concentrate (10F:90C; 10% tall fescue hay, 90% corn) basal substrates

	Forage:concentrate
Item	90F:10C	50F:50C	10F:90C
Dry matter, %	98.1	98.0	97.8
Total starch, % of DM	7.80	36.0	63.2
Neutral detergent fiber, % of DM	56.5	33.0	12.2
Acid detergent fiber, % of DM	30.0	16.7	4.80
Crude protein, % of DM	8.80	8.10	7.90
Crude fat, % of DM	1.71	2.88	3.79
Ca, % of DM	0.36	0.21	0.06
P, % of DM	0.34	0.34	0.34
Net energy for maintenance, Mcal/kg1	1.09	1.45	1.80
Net energy for gain, Mcal/kg1	0.54	0.87	1.18

¹Calculated from tabular values (NASEM, 2016).

At the completion of the 48-h incubation period, the cumulative measurement of gas pressure via the Memograph M RSG45 Data Manager was stopped by disconnecting the coupling valves. Valves of the fermentation vessel caps were connected to a 3-way stopcock consisting of the male coupling valve, a 10 mL syringe, and a needle (21 Ga × 38.1 mm). Headspace gas was collected into the syringe and then injected into 10-mL serum tubes (BD Vacutainer; Beckton, Dickinson and Company, Franklin Lakes, NJ). Serum tubes filled with headspace gas were stored at room temperature until analysis for methane concentration. Vessels were removed from the water bath and transferred to an ice bath to stop the fermentation. Fermentation vessel caps were unscrewed, and pH was measured with a combination electrode (SevenCompact S220; Mettler-Toledo, Columbus, OH). The filter bags containing residual substrates were removed from fermentation vessels, rinsed with cold water, and were dried in a forced-air oven (Model 338F; Fisher IsoTemp Oven 300 Series; FisherScientific) at 105 °C for 24 h. One milliliter of the fermentation media was collected and prepared for ammonia and short-chain fatty acid (SCFA) concentration analysis (Trotta et al., 2023) by combining with 100 µL of 500 g/L metaphosphoric acid (Erwin et al., 1961) and 100 µL of 85 mM 2-ethylbutyrate as an internal standard.

Gas samples were analyzed for methane concentration using gas chromatography with flame ionization detection (Trotta et al., 2023) (Agilent 7890A GC; Agilent Technologies, Santa Clara, CA, USA). Acidified samples of the fermentation media were frozen at −20 °C to facilitate protein precipitation, thawed, centrifuged (20,000 × g; 15 min; 4 °C), and the supernatants were transferred to autosampler crimp-top vials. Samples of the fermentation media were analyzed for SCFA concentrations using gas chromatography with flame ionization detection (Trotta et al., 2023; 8890 GC System; Agilent Technologies Inc., Santa Clara, CA). Concentrations of individual SCFA in the fermentation media were determined in reference to an analytical standard (50 mM acetate, 12 mM propionate, 4.5 mM isobutyrate, 10 mM butyrate, 4 mM isovalerate, 4.5 mM valerate) containing metaphosphoric acid and 2-ethylbutyrate. Total SCFA concentration was considered as the sum of acetate, propionate, isobutyrate, butyrate, isovalerate, and valerate concentrations. Molar proportions of SCFA were calculated as the individual SCFA concentration divided by the total SCFA concentration and multiplied by 100. Ammonia concentration in the fermentation media was analyzed using the glutamate dehydrogenase procedure (Kun and Kearney, 1974) adapted to a multi-mode plate reader (BioTek Synergy HTX; Agilent Technologies Inc.) as described by Trotta et al. (2024).

Apparent DM digestibility, true DM digestibility, NDF digestibility, and microbial biomass concentration of residual substrates in filter bags were analyzed using methods previously described (Ajayi et al., 2025). Dried filter bags were removed from the forced-air oven, placed into a desiccator to cool, and weighed to determine the DM content of the residue remaining for apparent DM digestibility. Filter bags were then added to an ANKOM 200 fiber analyzer to determine true DM digestibility and NDF digestibility. Microbial biomass was considered as the difference between true DM digested and apparent DM digested divided by the initial substrate DM mass.

In vitro gas production

Cumulative gas pressure in the vessel was corrected for atmospheric pressure and then converted into standard atmospheric pressure using the ideal gas law. Cumulative gas pressure was converted to moles of gas produced and then to milliliters of gas produced using Avogadro’s law. Cumulative gas production was corrected for the gas volume in the headspace of each fermentation vessel. The gas volume in the headspace of each fermentation vessel was calculated as the total volume of the vessel with the screw cap (301 ± 5.32 mL) minus the sum of the volume of the buffered inoculum, filter bag with substrate, and clip (106 ± 2.43 mL). Cumulative gas production was fitted to the one-phase association exponential model using GraphPad Prism 10.4 (Dotmatics, Boston, MA):

$${\displaystyle \begin{array}{l}{\displaystyle Y=Y_0+(P-Y_0)\times 1{\mathrm{e}}^{-r(t-\lambda )}}\end{array}}$$

where Y is cumulative gas production, Y₀ is the average Y value up to time λ, P is the Y value at infinite time, r is the rate of gas production, t is the time in hours, and λ is the lag time in hours. Y₀ was constrained to equal 0. Methane production was determined by multiplying the methane concentration of gas samples by cumulative gas production after 48-h of fermentation.

Statistical analysis

All variables were checked for normality using the Shapiro–Wilk test of the UNIVARIATE procedure of SAS (version 9.4; SAS Institute Inc., Cary, NC). The fermentation vessel was considered the experimental unit (n = 3). The data were analyzed as a randomized complete block design using the GLM procedure of SAS for fixed effects of replicate, enzyme dose, forage:concentrate, and the enzyme dose × forage:concentrate interaction. Blocks (replicates) were included in the model statement as fixed effects, as recommended by Dixon (2016). Least squares means and their standard errors were computed for each fixed effect. If the interaction was significant, then least squares means were separated using the Tukey-Kramer adjustment. If the interaction was not significant, then polynomial contrast coefficients (linear, quadratic) were used to describe least squares means for effects of enzyme dose and forage:concentrate after adjustment for unequal spacing using the IML procedure. Results were considered significant if P ≤ 0.05.

Results

For all parameters evaluated, the enzyme dose × forage:concentrate interactions were not significant (data not shown) and therefore, least squares means for the effects of enzyme dose and forage:concentrate were presented and evaluated independently using polynomial contrasts.

The rate of gas production was not influenced (P ≥ 0.60) by increasing concentration of endo-1,3-β-D-glucanase (Table 3). Cumulative gas production (mL, mL/g substrate, and mL/g substrate digested) increased linearly (P ≤ 0.03) with increasing endo-1,3-β-D-glucanase activity. Cumulative methane production (mL, mL/g substrate, and mL/g substrate digested) was not influenced (P ≥ 0.12) by increasing endo-1,3-β-D-glucanase concentration. Microbial biomass, true DM digestibility, apparent DM digestibility, and NDF digestibility were not influenced (P ≥ 0.21) by increasing endo-1,3-β-D-glucanase dose.

Table 3.

Least squares means for the effect of endo-1,3-β-D-glucanase on in vitro gas production, methane production, and dry matter (DM) digestibility

	endo-1,3-β-D-glucanase, U/100 mL							P-value
Item	0	5	10	25	50	100	SEM1	Linear	Quadratic
Rate of gas production, %/h	7.53	7.37	8.39	7.45	7.52	7.54	0.290	0.60	0.91
Cumulative gas production
mL	107	109	104	110	110	112	1.89	0.03	0.61
mL/g substrate	218	221	212	224	224	227	3.83	0.03	0.46
mL/g substrate digested	410	407	398	411	412	432	9.49	0.02	0.50
Cumulative methane production
mL	2.63	2.75	2.67	2.52	2.80	2.87	0.122	0.12	0.65
mL/g substrate	5.34	5.60	5.42	5.12	5.70	5.82	0.248	0.14	0.68
mL/g substrate digested	10.0	10.3	10.1	9.5	10.4	11.0	0.499	0.12	0.44
%	2.46	2.53	2.55	2.30	2.55	2.57	0.0882	0.41	0.48
True DM digestibility, %	66.5	67.7	66.8	67.7	67.4	66.4	0.711	0.55	0.23
Apparent DM digestibility, %	53.7	55.1	53.9	55.2	54.5	53.3	0.794	0.35	0.21
Microbial biomass, mg/g DM	128	126	129	125	129	131	2.65	0.23	0.61
NDF digestibility, %	46.3	48.3	46.6	48.1	47.6	45.9	1.15	0.46	0.26

¹Standard error of the mean (n = 3).

The rate of gas production responded quadratically (P = 0.04), with the peak rate observed with the 90F:10C diet (Table 4). Cumulative gas production (mL and mL/g substrate) peaked (quadratic: P < 0.01) with the 10F:90C diet. When gas production was expressed as mL/g substrate digested, there was a quadratic response (P = 0.04); however, the peak was observed with the low concentrate diet, and decreased as the concentrate amount increased. Methane (mL and mL/g substrate) responded linearly (P < 0.001), with methane increasing as the concentrate ratio of the diet increased. However, decreasing forage:concentrate decreased (quadratic: P = 0.04) methane production when expressed per gram of substrate digested. Methane proportion in gas samples was not influenced (P ≥ 0.52) by forage:concentrate. Decreasing forage:concentrate increased (P ≤ 0.03) apparent DM, true DM, and NDF digestibility. Microbial biomass concentration was greatest (quadratic: P = 0.01) in the high-forage diet.

Table 4.

Least squares means for the effect of forage:concentrate on in vitro gas production, methane production, and dry matter (DM) digestibility

	Forage:concentrate				P-value
Item	90F:10C	50F:50C	10F:90C	SEM1	Linear	Quadratic
Rate of gas production, %/h	9.34	7.99	5.57	0.205	<0.001	0.04
Gas production at 48 h
mL	83.0	106	137	1.34	<0.001	0.008
mL/g substrate	168	214	280	2.71	<0.001	0.05
mL/g substrate digested	438	400	396	6.71	<0.001	0.04
Methane production at 48 h
mL	2.07	2.67	3.37	0.0860	<0.001	0.64
mL/g substrate	4.20	5.42	6.88	0.175	<0.001	0.59
mL/g substrate digested	10.8	10.1	9.72	0.352	0.03	0.63
%	2.48	2.52	2.46	0.0623	0.85	0.50
True DM digestibility, %	52.1	66.1	83.1	0.503	<0.001	0.03
Apparent DM digestibility, %	38.4	53.7	70.7	0.562	<0.001	0.23
Microbial biomass, mg/g DM	136	124	124	1.87	<0.001	0.01
NDF digestibility, %	26.1	46.0	69.3	0.813	<0.001	0.09

¹Standard error of the mean (n = 3).

Final pH, ammonia concentration, and total SCFA concentration were not influenced (P = 0.01) by increasing endo-1,3-β-D-glucanase activity (Table 5). The molar acetate proportion and acetate:propionate decreased linearly (P < 0.001) with increasing endo-1,3-β-D-glucanase activity. The molar propionate, butyrate, and valerate proportions in the fermentation media increased linearly (P < 0.001) with increasing endo-1,3-β-D-glucanase activity. Isovalerate and isobutyrate molar proportions in the fermentation media decreased linearly (P < 0.01) with increasing endo-1,3-β-D-glucanase activity.

Table 5.

Least squares means for the effect of endo-1,3-β-D-glucanase on pH, ammonia, and short-chain fatty acid (SCFA) concentration in the fermentation media

	endo-1,3-β-D-glucanase, U/100 mL							P-value
Item	0	5	10	25	50	100	SEM1	Linear	Quadratic
pH	6.97	6.97	6.99	6.97	6.97	6.90	0.00974	0.19	0.82
Ammonia, mM	20.3	19.9	20.2	21.1	21.3	21.5	0.780	0.12	0.49
Total SCFA, mM	58.0	58.5	58.6	59.6	59.6	58.0	0.993	0.94	0.12
Acetate, mol/100 mol	52.8	52.7	52.6	52.1	51.3	50.7	0.183	<0.001	0.06
Propionate, mol/100 mol	21.7	22.0	22.0	22.5	22.5	23.0	0.175	<0.001	0.14
Butyrate, mol/100 mol	14.6	14.6	14.5	14.7	15.2	15.4	0.133	<0.001	0.43
Valerate, mol/100 mol	5.26	5.14	5.18	5.20	5.40	5.50	0.0732	<0.001	0.96
Isovalerate, mol/100 mol	3.83	3.77	3.80	3.75	3.75	3.78	0.0260	0.004	0.99
Isobutyrate, mol/100 mol	1.85	1.81	1.83	1.80	1.81	1.77	0.0184	0.01	0.71
Acetate:propionate	2.45	2.42	2.40	2.33	2.29	2.22	0.0258	<0.001	0.06

¹Standard error of the mean (n = 3).

Decreasing forage:concentrate linearly decreased (P < 0.001) the final pH of the fermentation media (Table 6). Ammonia concentration in the fermentation media was not influenced (P ≥ 0.09) by forage:concentrate. Total SCFA concentration linearly increased (P < 0.001) with decreasing forage:concentrate. As forage:concentrate decreased from 90F:10C to 50F:50C, there were quadratic responses (P < 0.001) observed for the molar acetate proportion and acetate:propionate because decreasing forage:concentrate decreased these variables and plateaued with further decreases in forage:concentrate. Decreasing forage:concentrate increased (P < 0.001) the molar propionate proportion as forage inclusion decreased from 90% to 50%, without further decreases as forage inclusion decreased to 10%. The molar butyrate proportion linearly increased (P < 0.001) with decreasing forage:concentrate. Decreasing forage:concentrate from 90F:10C to 50F:50C increased the molar valerate proportion by 20.3% and decreasing forage:concentrate from 50F:50C to 10F:90C increased the molar valerate proportion by a further 24.3%. Increasing forage:concentrate linearly increased (P < 0.001) the molar proportions of isovalerate and isobutyrate in the fermentation media.

Table 6.

Least squares means for the effect of forage:concentrate on pH, ammonia, and short-chain fatty acid (SCFA) concentration in the fermentation media

	Forage:concentrate				P-value
Item	90F:10C	50F:50C	10F:90C	SEM1	Linear	Quadratic
pH	7.06	6.98	6.88	0.00699	<0.001	0.45
Ammonia, mM	20.0	20.7	21.4	0.551	0.09	0.96
Total SCFA, mM	54.6	58.6	63.0	0.702	<0.001	0.76
Acetate, mol/100 mol	56.3	51.6	48.2	0.129	<0.001	<0.001
Propionate, mol/100 mol	21.1	22.8	22.9	0.124	<0.001	<0.001
Butyrate, mol/100 mol	12.2	14.9	17.4	0.0938	<0.001	0.42
Valerate, mol/100 mol	4.28	5.15	6.40	0.0517	<0.001	0.006
Isovalerate, mol/100 mol	4.10	3.74	3.47	0.0184	<0.001	0.07
Isobutyrate, mol/100 mol	2.00	1.79	1.64	0.0130	<0.001	0.13
Acetate:propionate	2.68	2.26	2.11	0.0182	<0.001	<0.001

¹Standard error of the mean (n = 3).

Discussion

In the current study, we investigated the effects of increasing dosage of exogenous endo-1,3-β-D-glucanase (0, 5, 10, 25, 50, and 100 U/100 mL) in the in vitro ruminal inoculum preparation. Pseudomurein, or pseudopeptidoglycan, is a principal component of methanogenic archaeal cell walls, which is distinguished from bacterial cell walls because of linkages consisting of β-1,3-glycosidic bonds (Kandler and Konig, 1993). Endo-1,3-β-D-glucanase acts on substrates with linear glucose sequences with β-1,3-bonds (Fontaine et al., 1997) and hydrolyzes internal bonds in glucan (glucose and oligosaccharides as end-products) and functions in transglycosylation reactions (glycosides form as products; Usoltseva et al., 2020). Although animal and feed management, diet formulation, and rumen manipulation have well-documented effects resulting in decreased methane emissions of ruminants (Arndt et al., 2022), we investigated an alternative strategy using endo-1,3-β-D-glucanase because of increased global demand to mitigate methane emissions of ruminants without decreasing animal productivity. The current study evaluates a product-based mitigation strategy and an absolute mitigation strategy (Arndt et al., 2022), with the factors of interest being decreasing forage:concentrate and adding endo-1,3-β-D-glucanase, respectively.

The predominant β-glucanases that are commercially available act on either β-1,3- or β-1,4-glucans with exo- or endo-targeted hydrolysis along polymers. β-1,3-glucanases are classified as exo-β-1,3-glucanase (EC 3.2.1.58) and endo-β-1,3-glucanase (EC. 3.2.1.39) based on their mechanism of action (Barras and Stone, 1969). The exo-β-1,3-glucanase hydrolyzes the β-1,3-glycosidic bond from the non-reducing end, releasing glucosyl residues (Wang et al., 2023a), and endo-β-1,3-glucanase (laminarinase) hydrolyzes the β-1,3-glucan glycosidic bond in the internal β-1,3-glucan chain to generate oligosaccharides with a low degree of polymerization (Wang et al., 2023a). Endo-β-1,4-glucanase (cellulase) hydrolyzes glycosidic bonds within β-1,4-glucans such as cellulose (Linton, 2020). Although previous studies have evaluated the supplementation of β-glucans in cattle (Ma et al., 2015; Cherdthong et al., 2018; Reis et al., 2022; Wang et al., 2025), less information is available from studies that have evaluated the supplementation of β-glucanase enzymes. Of those studies that have evaluated β-glucanase supplementation, most were focused on β-1,4-glucan hydrolysis (Beauchemin et al., 2000; Holtshausen et al., 2011; Peters et al., 2015). To our knowledge, this is the first study that has evaluated the effects of endo-1,3-β-D-glucanase activity on in vitro ruminal fermentation.

In the current study, we found that increasing endo-1,3-β-D-glucanase activity did not influence in vitro methane production. Although reasons are unclear, we speculate that there could be multiple explanations for the lack of response to in vitro methane production. It is possible that endo-1,3-β-D-glucanase did not penetrate the outer glycoprotein-rich layer of the archaeal cell wall, preventing the hydrolysis of pseudomurein in the middle and inner layers of the archaeal cell wall (Leahy et al., 2010). Others have speculated that the presence of specific endo-isopeptidases may be required to hydrolyze the glycoprotein-rich outer layer of the cell wall first (Leahy et al., 2010; Altermann et al., 2018). The endo-1,3-β-D-glucanase used in the current study may not have performed best under the conditions of the current experiment because of differences in the chemical properties of the in vitro ruminal inoculum compared with optimal conditions for endo-1,3-β-D-glucanase activity (pH, temperature, substrate concentration). The optimal pH of endo-1,3-β-D-glucanase was 5 (active between pH 3 to 9) and there was an optimal temperature range between 40 and 50 °C, respectively, indicating that the enzyme may act suboptimally in the mixed in vitro ruminal inoculum. The degree of effectiveness of endo-1,3-β-D-glucanase could have been due to the form of glucanase that we utilized in the experiment, as many 1,3-β-D-glucanases have modular structures (Usoltseva et al., 2020). Two types of 1,3-β-D-glucanases derive from plant origin and contain varying modular structures, larger proteins with a molecular weight range of 42 to 50 kDa and 2 domains (N-terminal domain with catalytic activity and carbohydrate-binding C-terminal domain) and smaller proteins with a molecular weight range of 33 to 41 kDa that do not contain a carbohydrate-binding domain (Usoltseva et al., 2020). The enzyme we utilized in the current study had a molecular weight of 34.1 kDa, indicating that it lacked a carbohydrate-binding domain. In plant 1,3-β-D-glucanases, the carbohydrate-binding domains increase the affinity of the enzyme’s binding surface for 1,3-glucans (Zamora-Carreras et al., 2015; Usoltseva et al., 2020), possibly indicating that the plant-based endo-1,3-β-D-glucanase (belonging to glycoside hydrolase family 17) utilized in this study had a decreased electron binding affinity for the 1,3-β-D-glucans found in the inner cell wall layers of methanogens and therefore, did not interact.

Despite the lack of response in methane production in the current study, endo-1,3-β-D-glucanase altered proportions of SCFA that typically correspond with changes in ruminal H₂ utilization that would result in decreased methane production (Janssen, 2010; Wang et al., 2017, 2018). Results from the current study with endo-1,3-β-D-glucanase were similar to others evaluating the effects of monensin on ruminal fermentation because of decreased acetate and increased propionate molar proportions (Richardson et al., 1976; Thornton and Owens, 1981; Trotta et al., 2023). With greater inclusion of endo-1,3-β-D-glucanase, there were decreases in the molar acetate proportion and increases in the molar proportions of propionate. The shift from acetate to propionate with endo-1,3-β-D-glucanase could suggest theoretical increases in the efficiency of converting feed energy to absorbable energy from acidic end products (Richardson et al., 1976). Propionate production from microbial fermentation of carbohydrates can be used as a substrate for hepatic gluconeogenesis, where propionate provides 60% to 74% of the total carbon pool for glucose synthesis (Aschenbach et al., 2010). The shift in acetate:propionate with endo-1,3-β-D-glucanase suggests that greater carbon was retained in the ruminal aqueous pool as proportions shifted from 2-carbon acetate to 3-carbon propionate. This data might suggest that supplemental endo-1,3-β-D-glucanase activity could aid in the production of precursors for hepatic gluconeogenesis and, therefore, energy and improved feed efficiency. This suggests that endo-1,3-β-D-glucanase could potentially be included in diets as a rumen fermentation modifier to shift acetate:propionate across high-forage and high-concentrate diets with the marketing advantage of being classified as a non-antibiotic alternative.

The shift in SCFA proportions upon the application of endo-1,3-β-D-glucanase from 2-carbon acetate to 3-carbon propionate may suggest changes in the production of carbon- and hydrogen-containing intermediates and end-products of rumen fermentation. Additionally, the combined increased molar proportions of propionate, butyrate, and valerate with endo-1,3-β-D-glucanase suggest a change in ruminal hydrogen utilization. This could occur by increasing dissolved hydrogen in the aqueous pool or by shifting metabolic hydrogen available for methane, H₂ gas, or other metabolic hydrogen sinks (Ungerfeld, 2020). Propionate (C₃H₅O₂), butyrate (C₄H₇O₂), and valerate (C₅H₉O₂) contain longer carbon chains than acetate (C₂H₃O₂) and can therefore bind more hydrogen atoms, serving as a hydrogen sink, which might suggest that less hydrogen atoms would be available for ruminal methanogenesis. However, in vitro methane production was not influenced by endo-1,3-β-D-glucanase in the current study. Despite increasing proportions of SCFA with greater carbon-chain lengths, there was not a concomitant reduction in methane production with endo-1,3-β-D-glucanase. Interestingly, others have reported significant changes in ruminal SCFA profiles without measurable changes in methane in vivo (Beauchemin and McGinn, 2006; Beauchemin et al., 2007). Reasons for these observations are unclear but may suggest that hydrogen sinks other than methane are affected with decreasing acetate:propionate (Ungerfeld, 2018, 2020; Beauchemin et al., 2020).

In the current study, there were observed increases in butyrate and valerate molar proportions with increasing doses of endo-1,3-β-D-glucanase and decreasing forage:concentrate. Butyrate and valerate are extensively metabolized by the ruminal epithelium first-pass (Kristensen and Harmon, 2004). Butyrate promotes the growth of ruminal papillae by increasing cellular proliferation and decreasing apoptosis (Sakata and Tamate, 1978; Mentschel et al., 2001). Butyrate also influences the metabolic activity of the ruminal epithelium, mRNA expression of transcripts encoding proteins that mediate SCFA transport, ruminal epithelial blood flow, and ruminal motility (Górka et al., 2018). Inflammatory responses of the ruminal epithelium are decreased with butyrate by decreasing nuclear factor kappa B activity and thus, suppresses the transcription of pro-inflammatory cytokines (Zhang et al., 2018). Increased molar proportions of valerate could possibly stimulate the growth of cellulolytic bacteria, as valerate fulfills the necessary carbon-chain requirement for cellulolytic rumen bacteria (Nagaraja et al., 1997; Hackmann, 2024). Collectively, the increased molar proportions of butyrate and valerate with endo-1,3-β-D-glucanase or decreasing forage:concentrate demonstrate an increase in substrates available for use by ruminal epithelial cells and ruminal microbes, respectively. These findings may suggest that dietary inclusion of endo-1,3-β-D-glucanase or decreasing forage:concentrate may have positive implications for providing energy substrates used by cells and microbes of the ruminant gastrointestinal tract.

In the current study, endo-1,3-β-D-glucanase did not influence in vitro ruminal digestibility (microbial biomass, true DM digestibility, apparent DM digestibility, and NDF digestibility). Forages contain greater proportions of structural carbohydrates like cellulose and hemicellulose that are characterized by β-1,4-glycosidic linkages of resident glucose molecules. The enzyme used in the study catalyzes the hydrolysis of glucans containing continuous β-1,3-linked glucosyl residues, suggesting that endo-1,3-β-D-glucanase does not hydrolyze cellulose and therefore, would not enhance fiber digestibility (Moore and Stone, 1972). Grains contain greater proportions of nonstructural carbohydrates like starch that contain α-1,4- and α-1,6-glycosidic bonds, suggesting likewise that endo-1,3-β-D-glucanase does not enhance starch digestibility. Therefore, the lack of responses in digestibility parameters with endo-1,3-β-D-glucanase occurred as expected, based on the hydrolytic properties of the enzyme and substrates available with varying forage:concentrate.

In the current study, we utilized diets containing increasing proportions of ground corn as a concentrate source (10%, 50%, and 90% of DM). We observed well-established effects in regard to altering forage:concentrate ratio in the diet. We observed an increase in vitro methane production (mL and mL/g substrate) when forage:concentrate decreased, but decreased in vitro methane production (mL/g substrate digested) when forage:concentrate decreased. These results align with the findings of others demonstrating that increasing the dietary starch:NDF decreases methane production per gram of substrate digested in cattle (Trotta et al., 2018; Galyean and Hales, 2024). Additionally, the 10% reduction in in vitro methane production per gram of substrate digested with decreasing forage:concentrate was similar to the 9% reduction in methane emissions per unit of milk produced or gain reported in a meta-analysis conducted by Arndt et al. (2022). In the current study, decreasing forage:concentrate by replacing tall fescue hay with ground corn increased in vitro apparent DM, true DM, and NDF digestibility, which agrees with other in vitro experiments evaluating forage:concentrate (Trotta et al., 2018; Wang et al., 2023b). Like other studies have found, diets with greater content of starch (decreased forage:concentrate) favored propionate and butyrate formation and decreased acetate:propionate, while diets containing greater proportions of forage (increased forage:concentrate) favored acetate formation (Moss et al., 2000; Aguerre et al., 2011; Ramos et al., 2021). Total SCFA concentration increased as concentrate proportions of the diet increased, which is in accordance with the results of other in vitro fermentation studies (Xu et al., 2010; Weimer et al., 2011). Decreased final pH of the in vitro fermentation media was associated with increased concentration of total SCFA in the current study when forage:concentrate decreased, in accordance with the findings of others (Burrin and Britton, 1986; Lee et al., 2018).

Conclusions

Endo-1,3-β-D-glucanase activity modified in vitro ruminal fermentation similarly across substrates varying in forage:concentrate. Increasing endo-1,3-β-D-glucanase activity from 0 to 100 U/100 mL in the in vitro ruminal inoculum influenced in vitro ruminal fermentation by altering molar SCFA proportions. Increasing endo-1,3-β-D-glucanase activity from 0 to 100 U/100 mL or decreasing forage:concentrate resulted in decreased acetate proportion and increased molar proportions of propionate, butyrate, and valerate. These data suggest increased carbon and metabolic hydrogen retention as SCFA in the in vitro ruminal fermentation media. Increasing endo-1,3-β-D-glucanase activity from 0 to 100 U/100 mL did not influence in vitro methane production from high-forage, mixed, or high-concentrate substrates. The lack of quadratic responses with increasing endo-1,3-β-D-glucanase activity may indicate that the optimal level of inclusion was not reached. Decreasing forage:concentrate influenced in vitro ruminal fermentation by decreasing gas and methane production per gram of substrate digested and increasing digestibility and total SCFA concentration. More research is needed to understand how changes in molar SCFA proportions with endo-1,3-β-D-glucanase activity influence the fate of ruminal carbon- and hydrogen-containing end-products and the effects of endo-1,3-β-D-glucanase activity in vivo.

Glossary

List of Abbreviations

DM

dry matter

NDF

neutral detergent fiber

SCFA

short-chain fatty acid

U

unit

Conflict of Interest Statement

The authors declare no real or perceived conflict of interest.

Author Contributions

Daniel P. Seeforth (Formal analysis, Funding acquisition, Investigation, Writing - original draft, Writing - review & editing), and Ronald Trotta (Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing - original draft, Writing - review & editing)