Influence of mimosa (Acacia mearnsii) tannin encapsulated with sunflower oil on rumen fermentation, methane and in vitro organic matter digestibility

Abstract

The utilization of tannin as an enteric-methane reducing additive has received wide attention in ruminant nutrition research but its astringency and the binding to digestible nutrients remain a major limitation. Encapsulation of tannin could mask the bitter taste while also delivering a controlled release of the bioactive into the rumen with ameliorated impact on feed intake and nutrient digestibility. This research aimed at evaluating the influence of encapsulating mimosa tannin with sunflower oil, and different inclusion levels into a substrate, on in vitro gas, methane, and organic matter digestibility, as well as rumen ammonia and volatile fatty acid concentration. The encapsulation process was done using solid-in-oil-in-water method. A dose comparison of the unencapsulated at 20 g/kg DM was compared with the encapsulated at 10–30 g/kg DM (EMTS¹⁰, EMTS²⁰ and EMTS³⁰) equivalent using two substrates a roughage (low digestibility) and a total mixed ration (high digestibility). The inclusion of UMT and EMTS decreased (p < 0.05) gas, methane, methane/in vitro organic matter digestibility (IVOMD), gas/IVOMD, methane/total volatile fatty acid (TVFA) and gas/TVFA as well as gas yield from insoluble but slowly fermentable fraction, b and effective gas production (EGP). However, IVOMD was reduced (p < 0.05) only by EMTS³⁰. While substrate type affected ammonia concentration, neither substrate nor tannin inclusion affected TVFA concentration (p > 0.05). In conclusion, non-encapsulated and encapsulated mimosa tannin in sunflower oil decreased methane without any reduction in IVOMD, ammonia and total VFA at 20 g/kg DM inclusion and at 30 g/kg DM, encapsulated tannin decreased IVOMD further. The effectiveness of this encapsulation is recommended for other ruminant nutrition applications where targeted delivery of the active ingredient is desirable.

Introduction

Tannins have gained significant attention for their potential to mitigate enteric methane emissions, as well as their ability to enhance dietary protein bypass and improve amino acid utilization^1,2. Minimizing enteric methane (CH4) emissions reduces the environmental footprint and improve feed conversion efficiency as part of the sustainable livestock production goals^2,3. Various strategies have been proposed to reduce methane emissions from ruminants, with tannin supplementation emerging as a promising option due to the abundance of tannins and their relatively easier incorporation into existing livestock systems⁴.

Among tannin sources, the black wattle (Acacia mearnsii), also known as mimosa tannin, is particularly noteworthy. It is recognized as an invasive species in South Africa and is rich in tannins⁵. Numerous studies have demonstrated the potential of mimosa extract to modulate rumen fermentation, reduce enteric methane production, and promote dietary protein bypass, reduced excretion of urinary nitrogen while some studies have equally observed improved feed efficiency^2,4,6.

However, the oral administration of moderate to high doses of tannin additives has been found to decrease feed intake and delay dry matter digestibility, largely due to the astringent nature of tannins and their solubility in the gastrointestinal tract^7,8. To counter these adverse effects, tannin encapsulation technologies have been proposed as a means to neutralize the bitterness of tannins and regulate their dissolution in the digestive tract^9,10. This approach could reduce enteric methane emissions and dietary protein degradation by rumen microbes without negatively impacting dry matter intake or digestibility.

Encapsulation is commonly used to achieve specific outcomes, such as masking unpleasant tastes, extending shelf life, protecting active ingredients from oxidative damage, and enabling controlled or targeted release^11,12. Recent advances in encapsulation technology have yielded promising results across various livestock applications to deliver active ingredients to target location, bypass rumen or optimise rumen release rate¹³. Encapsulation of tannin reduced the solubility to 24%, 21%, and 19% in simulated rumen, abomasum, and small intestine conditions, respectively, compared to the much higher solubility levels of 94%, 92%, and 96% of the unencapsulated tannin extract over a 24-hour period¹⁰.

A double emulsion encapsulated tannin extract is reported to exhibit high encapsulation efficiency and controlled release⁹. These controlled-release systems can minimise the negative side effects such as the astringency of tannins by masking the bitter taste in the mouth as well as modulate the significant reduction in protein and fibre digestibility through bind-release approach as the feed interact with the rumen environment^14,15. Previous studies employing the double emulsion have shown good product stability and because of the multicompartmentalized structure, allows for a controlled release of the bioactive core material from the internal to the external aqueous phase^12,15,16.

Despite these advances, there is still gaps in knowledge of the optimal inclusion level of oil-encapsulated mimosa tannin for methane reduction without penalizing digestibility under forage vs. TMR fermentation. The hypothesis of the study is that higher doses of encapsulated mimosa can be included in the diet of ruminants compared to the crude extract. The objective of this study is to explore the effects of different doses of sunflower oil-encapsulated mimosa tannin on in vitro gas production and methane production as well as fermentation kinetics, volatile fatty acids and ammonia nitrogen concentrations of Eragrostis hay and a total-mixed-ration as substrate.

Results

In vitro gas, methane and fermentation of Eragrostis curvula hay and total mixed ration as affected by unencapsulated and encapsulated acacia tannin

Table 1 presents the summary of gas, methane, in vitro organic matter digestibility and fermentation parameters from Eragrostis curvula hay (ECH) and total mixed ration (TMR) incubated with unencapsulated mimosa tannin at 20 g/kg DM of feed (UMT) and encapsulated mimosa tannin at 10, 20 and 30 g/kg DM (EMT^S10, EMT^S20 & EMT^S30, respectively.

Table 1.

Effects of encapsulating mimosa tannin as additive on in vitro gas production, methane, and rumen fermentation characteristics of Eragrostis curvula hay (ECH) and total mixed ration (TMR) diets.

1Parameters	Gas (mL/g)	Methane (mL/g)	IVOMD (g/kg)	Ammonia (mM)	TVFA (mmol/L)
ECH
Control	184	5.32	571	17.0	81.3
UMT	160	4.47	564	13.8	84.7
EMTS10	178	4.66	565	13.8	84.1
EMTS20	164	4.34	563	16.1	85.8
EMTS30	158	4.30	550	15.2	95.9
TMR
Control	199	5.54	728	25.0	98.6
UMT	171	4.81	715	21.4	89.0
EMTS10	198	5.27	718	21.9	87.8
EMTS20	170	4.76	712	22.7	91.5
EMTS30	163	4.45	692	23.7	89.9
SEM	5.69	0.20	0.75	2.21	6.37
P-values	S	0.003	0.011	<0.001	<0.001	0.224
	T	<0.001	<0.001	0.012	0.658	0.828
	S*T	0.632	0.817	0.890	0.996	0.508
Contrast	C vs “all tannin”	<0.001	<0.001	0.021	0.234	0.785
	UMT vs “EMT”	0.202	0.953	0.346	0.514	0.658
	UMT vs EMTS20	0.837	0.663	0.813	0.466	0.776
	UMT vs EMTS30	0.363	0.200	0.021	0.452	0.350

UMT = unencapsulated mimosa tannin @ 20 g/kg DM; EMT^S10, EMT^S20 & EMT^S30 = encapsulated mimosa tannin at 10, 20 & 30 g/kg DM, respectively.

¹Gas and methane in mL/g DM substrate; IVOMD = in vitro organic matter digestibility (g/kg); Ammonia in mg/100 mL; TVFA = total volatile fatty acids (mmol/L).

P- values: S = effect of substrate; T = effect of mimosa tannin; S*T = interaction effect of substrate and tannin. Contrasts: C vs. all tannin, control vs. the average of all tannin-included treatments; UMT vs. EMT, Unencapsulated tannin vs. average of all encapsulated-tannin treatments; UMT vs. EMT^{S20 =} Unencapsulated tannin vs. encapsulated tannin at 20 g/kg inclusion; UMT vs. EMT^S30 = Unencapsulated tannin vs. encapsulated tannin at 30 g/kg inclusion.

There was no significant interaction effect between substrate type and mimosa tannin extracts on the gas, methane (CH₄), in vitro organic matter digestibility (IVOMD), ammonia nitrogen and total volatile fatty acid (TVFA) productions. However, the simple effects of substrate type and tannin inclusions were significant on total gas and methane production, as well as IVOMD. Total gas production as well as methane were generally higher in the TMR substrate. Irrespective of the addition of mimosa tannin extracts, the proportion of total gas, methane, IVOMD, ammonia and TVFA were higher (p < 0.05) in the TMR diet compared with the ECH substrate. Although substrate type affected ammonia concentration, the effect on TVFA was not significant. While tannin inclusion generally reduced gas, methane and IVOMD, encapsulation did not affect gas and methane production when compared to the unencapsulated tannin treatment. Encapsulated tannin at 30 g/kg resulted in a lower substrate IVOMD compared to the crude extract at 20 g/kg (p = 0.021).

The ratios of total gas to methane, methane to IVOMD, gas to IVOMD, methane to TVFA and gas to TVFA is presented in Table 2. The interaction effects between the substrates and mimosa tannin inclusion were not significant for all the ratios. The simple effect of substrate type and tannin inclusion showed significant effect on methane/IVOMD and gas/IVOMD. However, methane/gas, methane/TVFA and gas/TVFA did not show significant effect associated with substrate or tannin inclusion. Generally, tannin inclusion reduced methane/IVOMD, gas/IVOMD, methane/TVFA and gas/TVFA compared to the control substrates. However, the inclusion of encapsulated tannin at 20–30 g/kg did not show any difference in methane/IVOMD, gas/IVOMD, methane/TVFA and gas/TVFA when compared to the unencapsulated mimosa at 20 g/kg.

Table 2.

The ratios of gas, methane, in vitro organic matter digestibility (IVOMD) and total volatile fatty acids (TVFA) as influenced by the inclusion of encapsulated mimosa tannin in substrate fermentation.

1Parameters	Methane/ Gas	Methane/ IVOMD (mL/g)	Gas/IVOMD (mL/Kg)	Methane/TVFA	Gas/TVFA
ECH
Control	0.029	0.097	3.49	0.057	2.27
UMT	0.028	0.085	3.04	0.055	1.89
EMTS10	0.026	0.093	3.50	0.062	2.13
EMTS20	0.027	0.085	3.01	0.053	1.92
EMTS30	0.027	0.081	2.96	0.050	1.70
TMR
Control	0.028	0.073	2.53	0.066	2.03
UMT	0.028	0.063	2.24	0.053	1.94
EMTS10	0.027	0.065	2.48	0.056	2.31
EMTS20	0.028	0.061	2.30	0.051	1.87
EMTS30	0.028	0.062	2.29	0.047	1.81
SEM	0.001	0.004	0.10	0.004	0.10
p - value	S	0.760	<0.001	<0.001	0.784	0.873
	T	0.494	0.007	<0.001	0.031	0.001
	S*T	0.909	0.807	0.344	0.529	0.334
Contrasts	control vs “all tannin”	0.263	0.001	0.001	0.021	0.017
	UMT vs “all EMT”	0.258	0.863	0.169	0.698	0.649
	UMT vs EMT-2%	0.563	0.770	0.862	0.546	0.809
	UMT vs EMT-3%	0.442	0.559	0.851	0.173	0.132

UMT = unencapsulated mimosa tannin @ 20 g/kg DM; EMT^S10, EMT^S20 & EMT^S30 = encapsulated mimosa tannin at 10, 20 & 30 g/kg DM, respectively.

IVOMD = in vitro organic matter digestibility (g/kg); TVFA = total volatile fatty acids.

P- values: S = effect of substrate; T = effect of mimosa tannin; S*T = interaction effect of substrate and tannin. Contrast analysis across treatments: C vs. all tannin, control vs. the average of all tannin-included treatments; UMT vs. EMT, Unencapsulated tannin vs. average of all encapsulated-tannin treatments; UMT vs. EMT^{S20 =} Unencapsulated tannin vs. encapsulated tannin at 20 g/kg inclusion; UMT vs. EMT^S30 = Unencapsulated tannin vs. encapsulated tannin at 30 g/kg inclusion.

Gas production characteristics of Eragrostis curvula hay and total mixed ration as influenced by unencapsulated and encapsulated acacia tannin

The summary of gas production kinetics (Table 3) showed that there was no significant interaction effect between substrate type and mimosa tannin extracts on gas volume from slowly fermentable fractions (b), the rate of fermentation (c) and effective gas production (EGP). However, the main effect of substrate indicated that ‘b’ was not significantly greater (p = 0.051) in ECH diet, while ‘c’ and EGP were significantly higher in TMR substrate. The simple effect of tannin revealed that there was no significant difference in gas volume from slowly fermentable fractions (b), (p = 0.066) and rate of fermentation (c) (p = 0.323). Nevertheless, EGP was affected by substrate with incubations containing TMR substrates having higher EGP while tannin generally reduced EGP (p < 0.01) but the effect of encapsulation was not significant on EGP.

Table 3.

Influence of encapsulated mimosa tannin on in vitro fermentation kinetics of Eragrostis hay and a total mixed ration substrate.

1Parameters	2b (mL/g)	3c (mL/h)	4EGP (mL/g)
ECH
Control	242	0.032	90.0
UMT20	237	0.025	77.4
EMTS10	227	0.033	89.8
EMTS20	230	0.025	81.0
EMTS30	207	0.028	70.3
TMR
Control	233	0.039	104
UMT20	204	0.038	87.6
EMTS10	226	0.040	104
EMTS20	208	0.036	84.7
EMTS30	195	0.037	82.9
SEM	12.0	0.003	3.48
5p - values	S	0.051	<0.001	<0.001
	T	0.066	0.323	<0.001
	S*T	0.723	0.803	0.539
Contrasts	control vs “all tannin”	0.037	0.330	<0.001
	UMT vs “all EMT”	0.607	0.558	0.300
	UMT vs EMT-2%	0.915	0.762	0.923
	UMT vs EMT-3%	0.109	0.820	0.101

UMT = unencapsulated mimosa tannin @ 20 g/kg DM; EMT^S10, EMT^S20 & EMT^S30 = encapsulated mimosa tannin at 10, 20 & 30 g/kg DM, respectively.

²b = gas volume from slowly fermentable fraction; ³c = rate of fermentation of fraction b (ml/h); ⁴EGP = effective gas production.

P- values: S = effect of substrate; T = effect of mimosa tannin; S*T = interaction effect of substrate and tannin. Contrast analysis across treatments: C vs. all tannin, control vs. the average of all tannin-included treatments; UMT vs. EMT, Unencapsulated tannin vs. average of all encapsulated-tannin treatments; UMT vs. EMT^{S20 =} Unencapsulated tannin (20 g/kg) vs. encapsulated tannin at 20 g/kg inclusion; UMT vs. EMT^S30 = Unencapsulated tannin (20 g/kg) vs. encapsulated tannin at 30 g/kg inclusion.

Volatile fatty acids concentration of Eragrostis curvula hay and total mixed ration as affected by unencapsulated and encapsulated acacia tannin

The influence of unencapsulated and encapsulated mimosa tannin extracts on molar proportion of volatile fatty acids (VFAs) (Table 4) showed that there were no interaction effects (p > 0.05) between substrate type and tannin for all VFA molar proportions as well as the ratio of acetate: propionate. Furthermore, the simple main effect of substrate type showed significant difference for acetate, Isobutyrate, butyrate, Isovaleate, valerate and the acetate-propionate ratio. Acetate proportion was higher in the ECH (72.2 mol/100mol.) than TMR (68.99 mol/100 mol.) while butyrate and isobutyrate proportion was higher in the TMR (7.86 mol/100mol and 1.71 mol/100mol) compared to the ECH substrate (6.06 mol/100mol and 1.58 mol/100mol), respectively. Substrate type did not influence propionate proportion, but A/P ratio was higher in the ECH compared to TMR substrate.

Table 4.

The in vitro rumen volatile fatty acid proportions (mol/100 mol) from Eragrostis hay and total mixed ration fermentation as affected by the inclusion of encapsulated mimosa tannin as additive.

Treatments	Acetate	Propionate	Isobutyrate	Butyrate	Isovalerate	Valerate	Acetate: Propionate
ECH
Control	73.0	15.8	1.48	5.70	2.28	1.74	4.6
UMT20	72.1	16.1	1.50	5.94	2.47	1.89	4.5
EMTS10	71.4	16.1	1.60	6.35	2.60	1.85	4.4
EMTS20	72.7	15.5	1.51	6.03	2.49	1.77	4.7
EMTS30	71.5	15.9	1.81	6.30	2.64	1.85	4.5
SEM
TMR
Control	70.1	15.3	1.7	7.8	3.0	2.10	4.7
UMT20	68.3	16.7	1.8	8.0	3.0	2.20	4.1
EMTS10	68.3	16.3	1.7	7.9	3.3	2.50	4.2
EMTS20	69.8	16.3	1.6	7.6	2.8	1.90	4.3
EMTS30	68.5	16.7	1.7	8.1	2.9	2.10	4.1
SEM	0.44	0.40	0.07	0.28	0.15	0.12	0.13
p-values	S	<0.001	0.261	0.018	<0.001	<0.001	<0.001	0.006
	T	0.004	0.364	0.103	0.751	0.467	0.170	0.084
	S*T	0.890	0.565	0.086	0.914	0.645	0.274	0.476
Contrast	control vs “all tannin”	0.004	0.086	0.265	0.366	0.308	0.996	0.018
	UMT vs “all EMT”	0.757	0.551	0.959	0.841	0.829	0.897	0.516
	UMT vs EMT-2%	0.047	0.295	0.220	0.682	0.596	0.311	0.161
	UMT vs EMT-3%	0.663	0.845	0.172	0.599	0.979	0.847	0.871

UMT = unencapsulated mimosa tannin @ 20 g/kg DM; EMT^S10, EMT^S20 & EMT^S30 = encapsulated mimosa tannin at 10, 20 & 30 g/kg DM, respectively.

P- values: S = effect of substrate; T = effect of mimosa tannin; S*T = interaction effect of substrate and tannin. Contrast analysis across treatments: C vs. all tannin, control vs. the average of all tannin-included treatments; UMT vs. EMT, Unencapsulated tannin vs. average of all encapsulated-tannin treatments; UMT vs. EMT^{S20 =} Unencapsulated tannin (20 g/kg) vs. encapsulated tannin at 20 g/kg inclusion; UMT vs. EMT^S30 = Unencapsulated tannin (20 g/kg) vs. encapsulated tannin at 30 g/kg inclusion.

Equally, the main effect of tannins indicated that the molar concentrations of acetate were significantly influenced while other VFA components were not significantly affected. Contrast comparisons show that compared to the control, tannin inclusion significantly reduced acetate proportion (p = 0.004) but did not increase propionate proportion (p = 0.086). Nevertheless, compared to substrate only, tannin inclusion shifted reduced A/P ratio. While encapsulated tannin at 20 g/kg showed ameliorative effect on acetate reduction compared to the unenapsulated tannin (p = 0.047), this was not observable when encapsulated tannin concentration increased to 30 g/kg. For all other VFA proportions, the effect of encapsulation was not seen compared to the crude extract at same concentration. Equally, increasing the encapsulated tannin dose to 30 g/kg did not affect VFA proportions when compared to the unencapsulated tannin at 20 g/kg.

Discussion

In this study, inclusion of both unencapsulated and encapsulated mimosa tannin at 20 g/kg dosage significantly reduced the quantity of in vitro gas and methane. Previous studies have evidenced the effect of tannins on rumen fermentation to be via direct inhibition of methanogenesis as well as overall reduction in organic matter fermentation¹⁷ as seen in the current study. Phenolic compounds like tannins are thought to exhibit differential antimicrobial effects by disrupting the cytoplasmic membrane, which hampers proton-electron transport, active transport, membrane potential, and cellular content stability. This disruption reduces ATP synthesis, ultimately leading to cell death^17,18. The present finding compares favourably with a previous report¹⁹ where approximately 11%, 13% and 16% decrease in methane was recorded after fermentation substrate was supplemented with unpurified, ethyl acetate purified, and pentanol purified mimosa tannin extracts at 20 g/kg substrate DM, respectively. Furthermore, Adejoro et al..⁹ observed that the inclusion of lipid encapsulated mimosa tannins reduced in vitro gas and methane by 7.0%, and 16%, respectively.

The influence of tannin supplementation on gas and rumen fermentation is widely reported to depend largely on the dosage of extracts used aside the proportion of hydrolysable vs. condensed tannin fraction^2,18,20,21. While, several studies have shown that lower inclusion levels less than 50 g/kg of feed DM) of mimosa tannin did not reduce gas, methane, ammonia nitrogen and total volatile fatty acid concentrations^9,19,22, encapsulated mimosa at 30 g/kg resulted in methane reduction comparable to the 20 g/kg unencapsulated mimosa treatment. Although this was accompanied by decline in IVOMD and gas production, encapsulation of tannin has potential to modulate the release of tannin in the rumen to achieve methane reduction with other co-benefits.

In vitro gas, methane, IVOMD, ammonia and TVFA were greater in TMR substrate compared to Eragrostris forage substrate. This could be attributed to increase in fermentable organic matter in the substrate. In concurrence with our finding, previous studies also reported higher methane and rumen fermentation metabolites-VFA and ammonia in high protein-based substrate than fibre-based diet¹⁹.

Moreover, addition of encapsulated mimosa tannin at 20 g/kg of substrate DM reduced slightly more methane compared to equivalent inclusion level of unencapsulated mimosa tannin. This may be linked to the pattern of release of the bioactive tannin from the encapsulation matrix or the role of the lipid wall material which might complement the tannin towards suppressing methanogenesis. Previous studies have observed the effect of oil as a supplement or wall as part of other additives in reducing in methane and ammonia nitrogen in ruminants due to potential alteration in microbiome dynamics such as protozoa defaunation^9,23.

The current results showed that the volume of gas from slowly fermentable fraction was higher in Eragrostis hay while the rate of fermentation and effective gas production were greater in the total mixed ration. This is due to the higher fibre fraction of the Eragrostis hay regardless of tannin inclusion. Tannins are well known to reduce potential gas production through prevention of rumen microbial growth and activities as well as reduction in feed digestibility^1,4,7. This also agrees with the previous report¹⁹ (Ibrahim and Hassen 2022) involving different mimosa tannin extracts. In contrast, Adejoro et al..⁹ reported an increase in the b-fraction when encapsulated mimosa tannins in palm oil and lard (as wall materials) were separately added to either Eragrostis or total mixed ration substrate. The rise in gas volume from slowly fermentable fraction observed in that study might be attributed to the lower tannin release rate recorded (19%) compared to the tannin concentration released by sunflower oil microcapsule (22%) used in the present study.

The interaction between substrate type and tannin inclusion did not result in significant substrate × tannin effects on IVOMD, indicating that the effect of tannin on digestibility was largely independent of whether the substrate was a fibrous forage (Eragrostis curvula hay) or a more protein- and starch-rich TMR. Furthermore, the observed patterns reflect the expected differences in fibre–protein interaction with tannins and the effect on digestibility¹⁸. In both substrates, tannin addition reduced IVOMD, but the reduction was greater at higher tannin levels, particularly in the ECH substrate. This supports the idea that in high-fibre, lower-protein substrates, the reduced availability of protein for tannin complexation may allow tannins to exert a more direct inhibitory effect on cell wall fermentation, contributing to lower gas production, reduced EGP, and slightly decreased digestibility^18,20. In contrast, the TMR’s higher protein content may have buffered the tannin effect through protein–tannin binding, which likely reduced the extent to which tannins interacted with fibre-degrading microbes, helping maintain IVOMD values closer to the control treatment.

Acetic acid and acetate: propionate ratios were higher in ECH diet, while, iso-butyric, butyric, iso-valeric and valeric acids were greater in TMR substrate. The multiplication and activities of rumen microbes is well known to depend largely on the nature of the substrate^9,23. On the other hand, several studies revealed that lower (50 g/kg DM) inclusion levels of tannins had little or no influence on rumen fermentation parameters². This assertion is corroborated by the findings in the present investigation. Furthermore, moderate to high concentration of tannin were reported to depress the activities of acetic acid producing bacteria more than those of propionate producers leading to a decrease in acetic: propionic acid ratio⁴ and this is confirmed in the current study.

Encapsulation influenced tannin effects primarily in the fermentation end-products, particularly VFA profiles, rather than in overall digestibility responses. While encapsulation did not significantly alter total VFA concentrations or their ratios relative to methane or gas, there was evidence of a moderating effect on acetate proportion at the 20 g/kg inclusion rate. This suggests that encapsulation may partially delay tannin release, reducing the immediate inhibitory impact on fibrolytic microbes that drive acetate production, an effect detectable in both the high-fibre ECH and the TMR. However, this amelioration was not sustained at a higher inclusion dose of 30 g/kg, implying that encapsulation does not prevent tannin overwhelming the microbial tolerance threshold.

Conclusion

The inclusion of both non-encapsulated and encapsulated mimosa tannin significantly decreased methane but also in vitro gas, and organic matter digestibility. In vitro gas, methane IVOMD, ammonia, TVFA, as well as effective gas production were higher in the total mixed ration diet compared to Eragrostis hay substrate. Encapsulation of mimosa tannin with sunflower oil at 20 g/kg DM achieved a methane-reducing effect similar to that of the unencapsulated form while maintaining digestibility, whereas inclusion at 30 g/kg DM led to a further decline in IVOMD. Further developments on the encapsulation of tannin is warranted to enhance its application in ruminant systems.

Materials and methods

The study was carried out in compliance with the recommendations of the South African National Health Research Ethics Council and in accordance with ARRIVE guidelines. The experimental procedure was reviewed and approved by the Animal Ethics Committee of the University of Pretoria, South Africa, with references number, EC075-17.

Materials

The Acacia mearnsii (also known as mimosa or black wattle) extract used in this study was gifted by the UCL Company Pty Ltd, Kwa-Zulu Natal, South Africa. The sunflower oil was obtained from a local grocery store in Pretoria while Span80, Tween80 and dichloromethane were purchased from Sigma-Aldrich, St. Louis, MO, USA. All other reagents utilized were of analytical grade and procured from Sigma-Aldrich, Johannesburg, South Africa.

Tannin encapsulation with sunflower oil

The tannin product upon the analysis of total phenol, non-tannin phenol and total tannin, condensed tannin and hydrolysable tannin^24,25 yielded 699 g/kg DM tannin phenol, 32.3 g/kg DM non-tannin phenol, 678 g/kg DM of total tannin, 222 g/kg DM condensed tannin and 464 g/kg DM hydrolysable tannin¹⁹. The encapsulation process was done using a double emulsion (solid-in-oil-in-water) technique as described by in previous study⁹. Briefly, 8.5 g of mimosa tannin powder was added into a beaker containing 30 mL sunflower oil solution in DCM (50 mg/mL) mixed with 0.5% (w/v) Span80 and stirred for 120 s with the aid of magnetic stirrer at 400 revolutions per minute, rpm. Another solution was prepared by mixing 300 mL water and 1% (w/v) of Tween80 using an iron rod homogenizer (PRO400DS, Pro Scientific Inc., Oxford, CT 06478, USA) set at 20,000 rpm for 180 s. The two solutions were mixed together and stirred for three hours using a magnetic stirring plate set at 800 rpm to evaporate the DCM. The product was squeezed using cheese cloth, rinsed with water and freeze-dried over a period of 5 days. The dried sunflower oil microparticle was ground to powder and kept in fridge for analysis. The yield and encapsulation efficiency was 58% and 68%, respectively¹⁰.

Chemical analysis of the substrates

The substrates (Eragrostis hay and the total mixed ration) were obtained from the feed store of experimental farm, University of Pretoria. The feeds were milled through 2 mm screen and the chemical composition analysed according to the Association of Official Analytical Chemists (AOAC)²⁶ procedures for dry matter (DM; ID 934.01), ash (ID 942.05), and crude protein (CP) using Nitrogen analyzer (FP-2000, Leco Instrumente GmbH, Kirchheim, Germany; ID 968.06). The proportion of neutral detergent fibre (NDF), acid detergent fibre (ADF) and acid detergent lignin (ADL) of the substrates were analysed using Fibre Analyzer (ANKOM 200/220, ANKOM Technology, New York, USA) in sequence²⁷. The ingredient and chemical composition of the substrates is presented in Table 5. The ingredient composition of the TMR include a proprietary concentrate mix, lucerne and Eragrostis hay.

Table 5.

Chemical composition of substrates (Eragrostis curvula hay and the total mixed ration) used for in vitro incubation.

Chemical composition g/kg DM	Total mixed ration	Eragrostis
DM	943	925
Crude protein	167	79.9
Neutral detergent fibre	369	726
Acid detergent fibre	241	397
Acid detergent lignin	78.2	77.7
Ash	72.4	45.9

In vitro gas, methane, ammonia and volatile fatty acids analyses

Rumen fluid was obtained from three cannulated Pinzyl steers (average age of 30 ± 3 months and weighing 472 ± 17 kg). The steers had been fitted with rumen cannulas at 21 ± 2 months of age, maintained within the cow herd of the University of Pretoria Teaching and Research Farm and routinely monitored by the duty veterinarian. These steers consumed Eragrostis hay lucerne hay ad libitum. A total of 1000 mL of rumen fluid was obtained from each steer and the rumen fluid was sieved with the aid of four folds of clean cheese cloth into a pre-heated thermos flask and immediately transported to the laboratory. Buffer media was prepared²⁸ with modifications proposed by previous research reports²⁹. The rumen fluid and the buffer were mixed under a continuous stream of CO₂ gas at the ratio of 3:5 and rumen inoculum was constantly flushed with CO₂ under inside a water bath set at 39 °C. The inoculum was dispensed (40 mL) into serum bottles containing 400 mg each of either substrate. The serum bottles were instantly locked with stoppers fitted with needles attached to three-way stopcocks and revolved in a shaker at 120 revolutions per minute (rpm), and heated at 39 °C. For each substrate, treatments comprised of (i) feed only (ii) substrate + unencapsulated mimosa tannin 20 g/kg DM of feed, UMT), (iii) feed + encapsulated mimosa equivalent to 10 g UMT/kg DM of feed, EMT^S10 (iv) feed + encapsulated mimosa equivalent to 20 g UMT/kg DM of feed, EMT^S20 (v) feed + encapsulated mimosa equivalent to 30 g UMT/kg DM of feed, EMT^S30. All treatments were incubated in four replicates while four blank bottles (containing only rumen inoculum) were included to correct for gas associated with baseline rumen inoculum, and four experimental runs were carried out in a completely randomized design with factorial treatment.

Gas pressure built up inside the serum bottles was measured as previously described³⁰ at 3, 6, 12, 24 and 48 h post incubation, using a semi-automated pressure Transducer (PX4200–015GI; Omega Engineering Inc., Laval, QC, Canada) linked with a digital data tracker (220 series indicators; Omega Engineering Inc.). The bottle headspace was calculated by subtracting the serum bottle contents (inoculum plus substrate) from the bottle volume obtained from the weight of bottle filled with water minus empty bottle weight. Gas pressure in psi was converted to gas volume and the cumulative gas volume were estimated in mL using the equation, shown in (1) below.

1

where, V_x = volume of gas in mL, V_j = headspace of serum bottle, and P_psi = pressure recorded in psi by the data logger³⁸.

Gas production characteristics were estimated using the Ørskov & McDonald non-linear model³¹ as presented in Eq. (2).

2

where, y = amount of gas produced at t = time; b = quantity of gas from the insoluble but slowly fermentable fraction; and c = rate of fermentation of ‘b’ fraction.

The concentration of methane was measured following gas readings at 3, 6, 12, 24 and 48 h substrate incubation³². Using 10 mL luer-locked syringes, gas samples were drawn from each bottle and injected manually into the gas chromatography equipped with a solenoid column packed with silica gel and a flame ionization detector (8610 C BTU Gas Analyzer GC System, SRI Instruments Bad Honnef, Germany). Prior to the gas injection, the GC was calibrated using standard methane and carbon dioxide gas. The peak area by standard gas was used to generate equations to estimate concentration of the gas in the analysed samples.

After 48 h in vitro incubation, serum bottles were dipped in iced to stop further microbial activity and the bottles were subsequently centrifuged at 2500 rpm for 15 min, the supernatants filtered off using 5 mL pipette into two separate vials, one containing 1 mL 0.5 M H₂SO₄ for ammonia determination³³ and the other containing 1 mL of 25% orthophosphoric acid for volatile fatty acid concentration analysis³⁴.

In vitro organic matter digestibility evaluation

The two step digestion method³⁵ with modifications by Engels & Van der Merwe³⁶ was used to determine the digestibility of the two substrates (Eragrostis hay and a total mixed ration) after inclusion of UMT, EMT^S10, EMT^S20 and EMT^S30. All the treatments, substrates only and blanks were replicated three times in cycle, and four separate cycles were conducted. In the first step, the artificial saliva and rumen fluid (3:1 v/v) mixtures were measured into digestion tubes containing 200 mg each of either substrate mixed with either UMT, EMT^S10, EMT^S20 or EMT^S30. Thereafter, the tubes were covered with stoppers and digested in a shaker at 39 °C, 100 rpm for 48 h. At the end of the fermentation step, the tubes were removed and centrifuged at 2500 rpm for 15 min and the supernatants discarded. In the second step (digestion step), acid-pepsin solutions were added into the tubes and further incubated at 39 °C, 100 rpm for 48 h. At the end of the second step, the tubes were removed and centrifuged at 2500 rpm for 15 min, the supernatants discarded and the residue dried at 105 °C for 18 h. After drying, the tubes were weighed and then ashed at 550 °C in a muffle furnace for 3 h. The tube + ash were weighed and the digestibility of each treatment estimated from the weights of empty tubes and residues.

Statistical analysis

For the in vitro procedures, individual bottles served as analytical replicates while each incubation run served as experimental replicate. All data obtained were checked for normality using the PROC UNIVARIATE procedure for normality, then analysed using the mixed model procedure of SAS 9.4 (SAS Institute Inc., Cary, NC, USA). The model statement was stated in (3) below:

3

where, Y_ijk = mean of observation, µ = overall mean, C_h = effect of incubation run, S_i = effect of substrates (ECH and TMR), T_j = effect of treatment (UMT, EMT^S10, EMT^S20 and EMT^S30), R_k = effect of run within treatment, (ST)_ij = effect of interaction between substrate and treatment, and Ɛ_hijk = residual error. Effect of substrate, treatment and substrate-treatment interactions were set as fixed effects, while incubation run and run within treatment were set as random effects. Where interaction was significant, Tukey test was used for mean separation. Furthermore, single degree of freedom contrasts were used to compare (i) Control versus the average of all tannin treatments, “C vs. all tannin”; (ii) Unencapsulated tannin (20 g/kg) versus average of all encapsulated-tannin treatments, UMT vs. EMT; Unencapsulated tannin (20 g/kg) versus encapsulated tannin at 20 g/kg inclusion, UMT vs. EMT^S20; and iv) Unencapsulated tannin (20 g/kg) versus encapsulated tannin at 30 g/kg inclusion, UMT vs. EMT^S30.

Author contributions

Conceptualization: Shehu Lurwanu Ibrahim, Abubeker Hassen. Data curation: Shehu Lurwanu Ibrahim. Formal analysis: Shehu Lurwanu Ibrahim, Festus A Adejoro, Abubeker Hassen. Investigation: Shehu Lurwanu Ibrahim, Abubeker Hassen. Methodology: Shehu Lurwanu Ibrahim, Festus A Adejoro, Abubeker Hassen. Funding acquisition: Abubeker Hassen. Project administration: Abubeker Hassen. Resources: Abubeker Hassen. Supervision: Abubeker Hassen. Validation: Shehu Lurwanu Ibrahim, Abubeker Hassen. Visualization: Shehu Lurwanu Ibrahim, Festus A Adejoro Writing—original draft: Shehu Lurwanu Ibrahim. Writing—review and proofreading: Shehu Lurwanu Ibrahim, Festus A Adejoro; Abubeker Hassen.

Data availability

The raw data can be accessed on request directed to abubeker.hassen@up.ac.za.

Declarations

Competing interests

The authors declare no competing interests.