Rice is the second most cultivated cereal worldwide (FAO, 2016). Therefore, for each ton of rice grain harvested, one ton of straw remains in the field (Doyle et al., 1986). Despite the low nutritional value of rice straw due to its high silica content, low ruminal degradation of carbohydrates, and low nitrogen content, when stored in bales, it presents significant potential for strategic use in critical periods of food availability or in ruminant production systems with low nutrient requirements.

Technologies to increase the use of low-quality feeds, such as rice straw, consist in optimizing nutrient availability for the ruminal fermentation, ensuring no deficiency of nutrients to the microorganisms. Increased bacterial growth may result in increased extraction, through fermentation, of the roughage carbohydrate energy and, as a result, microbial cells synthesized in the rumen are available for amino acid digestion and absorption in the intestine (Leng, 1990).

This nutrient availability optimization in cattle fed rice straw may be achieved with mineral and protein-energy supplementation aiming to improve forage digestibility to maximize its intake (Barbosa et al., 2007), meeting animal requirements for maintenance and moderate weight gain (Lima, 2002). Furthermore, this supplementation increases the ruminal ammonia nitrogen concentration and meets the requirements of the ruminal microorganisms, allowing maximum fermentation rates (Fike et al., 1995).

Additionally, the increase in the metabolizable protein availability and the increase in the absorbed protein:energy ratio reduced the metabolic heat production, promoting greater intake and raising the gain rates (Leng, 1990; Poppi and McLennan, 1995).

On the other hand, supplementation with feed of low nutritional value, with exogenous fibrolytic enzymes, aims to increase nutrient use and animal production efficiency (Nsereko et al., 2000; Beauchemin et al., 2003) and reduce the fecal output. These enzymes potentiate the degradation of fibrous polysaccharides togheter with the enzymes produced by the rumen microorganisms, stimulating total digestion and degradation rate, thus improving the digestibility of fibrous feeds (Newbold, 1997).

The hypothesis that the improvement of the rice straw in vitro fermentation process may be achieved by using additives was investigated. Therefore, the objective of this study was to evaluate, through in vitro techniques, the effects of mineral and protein-energy and exogenous fibrolytic enzyme supplements on digestibility, fermentation parameters and kinetics, maximum gas production, methane (CH₄) and carbon dioxide (CO₂) production, and volatile fatty acid concentration in rice straw.

Animal care procedures throughout the study followed protocols approved by the Ethics Committee for Animal Use (ECAU) of the Universidade Federal do Rio Grande do Sul, number 18442/2010.

Two straws (low and high nutritional value) were incubated in vitro without supplementation (WS), with mineral and protein-energy supplement (MPES), with exogenous fibrolytic enzymes supplement (ES), and with the combination of the two supplements (MPES + ES) (Table 1). Four experimental periods were used, totaling four treatment replications. Duplicate bottles were also included in each run as blanks.

Table 1

Chemical composition (g kg–1 of dry matter)	High-nutritional value straw	Low-nutritional value straw	MPES1	ES
Dry matter2	873	915	927	915
Organic matter2	831	828	400	955
Crude protein2	44.7	37.4	398	240
Neutral detergent fiber3	732	781	196	317
Acid detergent fiber4	424	469	69.8	130
Cellulose5	393	435	45.3	90.7
Hemicellulose5	307	302	126	186
Lignin5	28.0	34.4	24.5	39.6
Silica5	95.9	126	-	-
IVOMD	529	424	653	765

¹ Each kg contains: 60 g calcium; 30 g phosphorus; 14 g sulfur; 74 g sodium; 0.21 g manganese; 1.3 g zinc; 0.06 g cobalt; 0.12 g copper; 0.02 g iodine; 0.009 g selenium; 0.28 g fluorine; 330 g non-protein nitrogen; and 220 g total digestible nutrients.

The mineral and energy-protein supplement used was commercially available and prepared in compliance with the nutritional standards of the NRC (1996) for beef cattle, whose daily intake recommendation is 50 g 100 kg^–1 body weight, composed of non-protein nitrogen source, macro and micro minerals, cottonseed and soybean meal, and wheat bran. The enzyme supplement was a commercially available source of xylanase, whose daily intake recommendation is 15 g head^–1, consisting of corn distillers’ dried grains with solubles, plant protein products, Yucca schidigera plant extract, and dried Trichoderma longibrachiatum fermentation extract (Alltech Inc.).

The straw rice intake of 100 g kg^–1 of body weight was considered.

The in vitro digestibility was determined by the two-stage digestion technique proposed by Tilley and Terry (1963). Ruminal inoculum was collected from two fasting Texel sheep with an average weight of 60 kg adapted for 10 days to a diet based on alfalfa hay. Two hours after morning feed, rumen fluid and part of the rumen solid material were obtained to collect microorganisms adhered to the substrate. All collected material was homogenized in a blender at a ratio of 1:1 (solid:liquid portion) and filtered through four layers of gauze adding CO₂.

In vitro organic matter digestibility (IVOMD) was calculated by the difference between the incubated and undigested organic matter (OM) present in crucibles (Goering and Van Soest, 1970).

In vitro cumulative gas production was obtained through the Theodorou et al. (1994) methodology modified by Mauricio et al. (1999), using a pressure transducer data logger (PDL 200 LANA/SCENE USP, Piracicaba/SP, Brazil) connected to a three-output valve. The first output was connected to the pressure transducer, the second to the needle (no. 22) to be inserted into the bottle stopper, and the third to a plastic syringe to measure the volume. Ruminal inoculum was obtained as described previously. The bottles were sealed with rubber stoppers and aluminum rings. In each experimental period, two bottles per treatment and per time were incubated, totaling 224 bottles plus the blank bottles (two blanks per incubation time, totaling 28 bottles). Four experimental periods were used, totaling four replications per treatment over time.

Pressure and volume of gas were measured at 0, 1, 3, 6, 9, 12, 18, 24, 30, 36, 48, 60, 72, and 96 h post-incubation. Gas production was expressed in mL of gas produced per gram of organic matter incubated.

For degradation rate adjustment, gas production data were fitted using a bicompartimental model (Schofield et al., 1994): V (t) = A / (1 + exp × (2 – 4 × B × (T – C))) – 1 + D / (1 + exp (2 – 4 × E × (T – F))) – 1, in which V (t) = cumulative gas production at time t (mL g^–1 OM); A = maximal gas production of the rapid fermentation fraction (mL); B = fermentation rate of A (h); C = lag time of the fraction A (h); D = maximal gas production of the slow fermentation fraction (mL); E = fermentation rate of D (h); F = lag time of the fraction D (h); and T = incubation time (h). The model parameters were estimated by interactive Marquardt method inserted into the NLIN procedure of SAS (Statistical Analysis System, version 9.3).

The partition factor (PF) was determined according to Makkar (2004): FP = mg OM truly degraded/mL gases. For this calculation, we considered 36 h of incubation (time in which half of the maximal gas production was produced by treatments).

At 6, 12, 24, 48, 72, and 96 h, the fermentation was stopped and the pH measurements were performed immediately. Subsequently, the contents of the bottles were filtered through a sintered-glass crucible of coarse porosity (100 to 160 µm). Crucible containing residue from the filtration was oven-heated at 105 °C for 12 h, weighed, resulting in a moisture-free residue, and subsequently heated at 450 °C for 5 h. In vitro organic matter degradability was calculated by the difference between the incubated and undigested organic matter present in crucibles (Goering and Van Soest, 1970).

To study the ruminal degradability kinetics, the degradability obtained at different times was adjusted using the McDonald (1981) model: Yt = a + b (1 – exp – c (t – to)), in which Yt = losses for degradation after t hours; a = immediately solubilized substrate; b = insoluble material, but potentially degradable; a + b = potential degradability; c = degradation rate of b; t = incubation time (h); and to = lag time. The effective degradability (ED) was calculated using the equation proposed by Ørskov and McDonald (1979): ED = a + [(b × c) / (c + k)] exp (–(c + k) t), in which a, b, c, and t followed previous definitions and k = ruminal outflow rates of 0.02 or 0.05 h^–1.

At these same times, two aliquots of 5 mL of the filtrate were collected, one for volatile fatty acid and another for ammonia nitrogen (NH₃-N) determination. In the aliquots for NH₃-N determinations, 1 mL of 0.18 molar sulfuric acid was added to avoid nitrogen losses. Aliquots were frozen until analysis.

Volatile fatty acid concentrations - acetic (C2), propionic (C3), and butyric acids (C4) - were determined by high-performance liquid chromatography, in a chromatograph (Shimadzu model 14-B) equipped with UV detector, pre-column and column (Aminex HPX-87H, BioRad^®). Sulfuric acid was used as eluent at 0.01 molar concentration, at a 0.6 mL min^–1 flow rate and 50 °C operating temperature. The detection wavelength was set at 210 nm. Volatile fatty acid concentrations were calculated from the calibration curves using standards (Sigma^®, analytical grade) at 0.1 to 2.5 g L^–1 concentrations.

Ammonia nitrogen concentrations were determined by magnesium oxide distillation according to AOAC (1995).

The volume of gas produced during the intervals of 12, 24, 36, 48, 72, and 96 h of incubation was collected, measured, and stored in 20-mL vacutainer tubes without additive for the gas analysis.

Methane and CO₂ gases were analyzed by gas chromatography (Shimadzu^® “greenhouse” model) equipped with three packed columns operating at 70 °C. Nitrogen as a carrier gas (25 mL min^–1), injector (250 °C) with direct sampling of 1 mL, and electron capture detector with Ni₆₃ at 325 °C were used.

Peak gas areas were determined automatically by integration. Methane volume produced at time x was calculated in accordance with Tavendale et al. (2005): CH₄ production (mL g^–1 dry matter (DM)) at time x = (% CH₄ (x) – % CH₄ (x – 1)) × 40/100 + CH₄% (x) × GP / 100, in which x time = 12, 24, 36, 48, 72, and 96 fermentation h; x – 1 = previous time; 40 = head space in the fermentation bottle in mL; GP = volume of gas produced in mL. This calculation resulted in the volume of CH₄ gas produced between each time interval. The sum of these volumes resulted in the accumulated volume of CH₄ for 96 h. The same formula was used to calculate CO₂ production.

Digestibility, degradation parameters, effective degradation of organic matter, maximum gas production from the rapidly and slowly degradable fractions and their respective degradation rates, time of colonization, and partition factor data were analyzed using the PROC MIXED of SAS. The following statistical model was used:

Yijkl = µ + αi + βj + αβij + γk + eijkl,

in which Yijkl = dependent variables; µ = overall mean of the observations; αi = fixed effect of the straw (i = 1, 2); βj = fixed effect of the supplement (j = 1, 2, 3, 4); αβij = straw × supplement interaction effect (i = 1, 2, and j = 1, 2, 3, 4); γk = ramdom effect of the period (k = 1, 2, 3, 4); and eijkl = random residual experimental error.

Gas production, pH, NH₃, CH₄, CO₂, and VFA data were analyzed as repeated measures over time using the same procedure. The following statistical model was used:

Yijklm,1 = µ + αi + βj + αβij + εij+ τk + ατik + βτjk + αβτijk + γl + eijklm,

in which Yijklm,1 = dependent variables; µ = overall mean of the observations; αi = fixed effect of straw (i = 1, 2); βj = fixed effect of the supplement (j = 1, 2, 3, 4); αβij = straw × supplement interaction effect (i = 1, 2, and j = 1, 2, 3, 4); εij = random residual experimental error; τk = fixed effect of the time ((k = 6, 12...96), or k = 12, 24…96)); ατik = straw × time interaction effect; βτjk = supplement × time interaction effect; αβτijk = straw × supplement × time interaction effect; γl = ramdom effect of the period (k = 1, 2, 3, 4); and eijklm = experimental error associated with the observation Yijklm,l level.

Using Akaike information criterion, the CS structure (symetry compound) was regarded as the best model for the residual covariance structure.

There was a significant interaction between the supplement and rice straw nutritional value for in vitro dry matter and organic matter digestibilities (Table 2). The mineral and protein-energy supplement improved the in vitro dry matter and organic matter digestibility of the rice straw with high nutritional value, while the rice straw with low nutritional value without supplementation showed lower values for in vitro dry matter and organic matter digestibility.

Table 2

Suplement	Rice straw nutritional value		Mean	SEM
	High	Low
IVDMD (g kg–1 fresh material)
Without supplement	473Ba	392Bb	432	0.2190
Mineral and protein-energy supplement (MPES)	519Aa	423Ab	472	0.2190
Exogenous fibrolytic enzyme supplement (ES)	467Ba	435Ab	451	0.2190
MPES + ES	475Ba	422Ab	448	0.2190
Mean	484	417
SEM	0.1548	0.1548
Significance (P =)
Nutritional value × suplement	<0.0001
IVOMD (g kg–1 DM)
Without supplement	568Ba	495Bb	532	0.2633
MPES	604Aa	515Ab	559	0.2633
ES	561Ba	529Ab	545	0.2633
MPES + ES	551Ba	519Ab	535	0.2633
Mean	571	515
SEM	0.1862	0.1862
Significance (P =)
Nutritional value × suplement	<0.0001

DM - dry matter; SEM - standard error of the mean.

Organic matter degradation parameters were influenced only by the rice straw nutritional value (Table 3). Straw with high nutritional value had greater content of readily soluble fraction of OM (a) (P<0.05) compared with the straw with low nutritional value, 188 and 162 g kg^–1 OM, respectively.

Table 3

Parameter	a (g kg−1 OM)	b (g kg−1 OM)	c (h)	lag time (h)	ED 0.02	ED 0.05
Straw nutritional value
High	188A	656	0.017	9.41	442A	297A
Low	168B	645	0.017	9.71	408B	266B
SEM	0.53	2.32	0.0012	0.68	0.74	0.65
Suplement
Without supplement	183	661	0.016	10.25	424	277
Mineral and protein-energy supplement (MPES)	193	675	0.016	9.83	425	283
Exogenous fibrolytic enzyme supplement (ES)	187	652	0.017	9.30	420	276
MPES + ES	192	617	0.019	8.86	432	289
SEM	0.74	3.28	0.0017	0.96	1.05	0.92
Significance (P =)
Nutrition value	<0.0001	0.5401	0.7170	0.4988	<0.0001	<0.0001
Suplement	0.1572	0.1300	0.1445	0.1536	0.4108	0.1849
Nutritional value × suplement	0.3115	0.6349	0.2517	0.5617	0.4301	0.3154

OM - organic matter; ED - effective degradability; SEM - standard error of the mean.

Averages of insoluble fraction, but potentially degradable (b), degradation rate of the insoluble fraction, but potentially degradable (c), and lag time of OM were 651 g kg^–1 OM, 0.0017 h^–1, and 9.56 h, respectively (Table 3), without influence of the straw nutritional value.

Effective degradability obtained for the solid fraction passage rates (k = 0.02 and 0.05 h^–1) were influenced by the straw nutritional value (P<0.05) (Table 3).

There was an interaction effect between the straw nutritional value and the incubation time on the in vitro cumulative gas production (Table 4). From 18 h of incubation, the cumulative gas production of high-nutritional value straw was greater than for the low-nutritional value straw and at the end of the 96 h of incubation, this production was 200.22 and 186.13 mL g^–1 OM, respectively, showing better fermentation for the high-nutritional value straw.

Table 4

Incubation time (hours)	Rice straw nutritional value		Mean	SEM
	High	Low
1	0.03K	0I	0.01	1.30
3	0.92K	0.84I	0.87	1.30
6	5.45JK	2.12I	3.78	1.30
9	10.57IJ	4.58I	7.66	1.30
12	15.62I	7.12I	11.37	1.30
18	30.82Ha	18.63Hb	24.72	1.30
24	50.82Ga	36.26Gb	43.54	1.30
30	78.74Fa	60.36Fb	69.55	1.30
36	103.69Ea	83.38Eb	93.53	1.30
48	134.97Da	114.84Db	124.90	1.30
60	161.71Ca	144.86Cb	153.28	1.30
72	179.93Ba	165.40Bb	172.66	1.30
96	200.22Aa	186.13Ab	193.18	1.30
Mean	69.54	58.89
SEM	0.49	0.49
Significance (P =)
Nutritional value × incubation time	<0.0001

OM - organic matter; SEM - standard error of the mean.

Maximum gas production of the rapidly degradable fraction of organic matter (A) was influenced by the straw nutritional value. High-nutritional value straw produced 113.98 mL, while the low value produced 75.61 mL of gases related to fraction A. However, the maximum gas production of the slowly degradable fraction of organic matter (D) was influenced by straw nutritional value and by the supplement. Low- and high-nutritional value straws produced 113.43 and 89.27 mL, respectively (P<0.05). Supplements contributed to the reduction in the maximum gas production related to fraction D compared with the treatment without supplement (Table 5).

Table 5

Parameter	A (mL)	B (h)	C (h)	D (mL)	E (h)	F (h)	PF (mg mL−1)
Straw nutritional value
High	113.98A	0.044	19.74A	89.27B	0.022B	36.34A	2.68
Low	75.61B	0.038	13.04B	113.43A	0.026A	30.60B	2.61
SEM	5.08	0.33	1.60	5.56	0.15	2.12	0.12
Suplement
Without supplement	94.05	0.043	14.92	110.90A	0.022	34.91	2.37
Mineral and protein-energy supplement (MPES)	100.01	0.041	14.66	88.08B	0.022	32.17	2.88
Exogenous fibrolytic enzyme supplement (ES)	93.17	0.041	18.39	99.26AB	0.024	33.40	2.73
MPES + ES	91.96	0.041	17.58	107.15AB	0.024	33.41	2.60
SEM	7.19	0.47	2.73	7.86	0.22	3.00	0.17
Significance (P =)
Nutrition value	<0.0001	0.0987	0.0004	0.0003	0.0242	0.0134	0.6844
Suplement	0.6895	0.9635	0.2803	0.0408	0.6238	0.8402	0.2867
Nutritional value × suplement	0.4071	0.8900	0.5968	0.6769	0.2910	0.9407	0.2298

PF - partition factor; OM - organic matter; SEM - standard error of the mean.

The degradation rate of slowly degradable fraction (E) was greater for the low-nutritional value straw than the high nutritional value, 0.026 and 0.022 h^–1, respectively (P<0.05). The lag times of rapidly (C, h) and slowly degradable (F, h) fractions were greater for the high-nutritional value straw than for the lower value (P>0.05). The partition factor, was not affected by treatments (P>0.05), indicating that the fermentation efficiency was not affected by supplementation (Table 5).

The pH decreased, while the NH₃-N concentration increased with the increase in incubation time (Table 6). The NH₃-N concentrations were also influenced by the straw nutritional value (P<0.05) and the supplement (P<0.05). Low- and high-nutritional value straw showed NH₃-N concentrations of 14.05 and 14.38 mg dL^–1, respectively. The MPES + ES and MPES supplements showed the highest concentrations of NH₃-N, 14.93 and 14.79 mg dL^–1, respectively, differing from the supplements ES and WS, whose NH₃-N concentrations were 13.78 and 13.35 mg dL^–1, respectively (Table 6).

Table 6

Parameter	pH	NH3-N (mg dL–1)	Incubation time (h)	CO2 (mL g–1 DM)
Straw nutritional value
High	7.10	14.05B		19.03A
Low	7.08	14.38A		16.85B
SEM	0.028	0.11		0.1896
Suplement
Without supplement	7.00	13.35B		18.74A
Mineral and protein-energy supplement (MPES)	7.12	14.79A		17.40B
Exogenous fibrolytic enzyme supplement (ES)	7.11	13.78B		18.07AB
MPES + ES	7.12	14.93A		17.57B
SEM	0.040	0.15		0.2670
Incubation time (h)
6	7.19A	11.87E	12	19.63A
12	7.09AB	12.54DE	24	19.80A
24	7.12AB	13.22D	36	19.27A
48	7.10AB	14.30C	48	17.08B
72	7.05AB	15.84B	72	16.00B
96	6.94B	17.51A	96	15.88B
SEM	0.050	0.19		0.4549
Significance (P =)
Nutritional value	0.6949	0.0399		<0.0001
Suplement	0.1095	<0.0001		0.0020
Incubation time	0.0042	<0.0001		<0.0001
Nutritional value × suplement	0.8219	0.9383		0.4105
Nitritional value × incubation time	0.3318	0.1139		0.5586
Suplement × incubation time	0.7613	0.5552		1.000
Nutritional value × supplement × incubation time	0.4054	0.8061		1.000

DM - dry matter; SEM - standard error of the mean.

The volume of CO₂ produced (mL g^–1 DM) was related to the straw nutritional value, the supplement, and the incubation time (Table 6). Low-nutritional value straw produced 16.85 mL g^–1 DM, while the high value produced 19.03 mL g^–1 DM of CO₂. The supplements MPES and MPES + ES produced lower volumes of CO₂ (17.40 and 17.57 mL g^–1 DM) compared with supplements WS and ES (18.74; 18.07 mL g^–1 DM), being important for mitigating CO₂. As the incubation time increased from 12 to 96 h, the CO₂ production decreased from 19.63 to 15.88 mL g^–1 DM, respectively.

There was an interaction between the straw nutritional value and the incubation time on in vitro CH₄ production (Table 7). In the first 12 h of fermentation, CH₄ production was similar between straws. Starting from 12 h to the end of the incubation period, there was a linear increase in the CH₄ volumes for both straws; however, greater CH₄ volume was produced by high-nutritional value straw compared with the low-nutritional value straw. Nevertheless, at the end of the 96-h incubation, the CH₄ production rate in the total gas volume was 0.15 for both straws.

Table 7

Incubation time (hours)	Rice straw nutritional value		Mean	SEM
	High	Low
12	1.16E	0.75F	0.96	0.53
24	7.97Da	5.64Eb	6.80	0.53
36	14.97Ca	11.66Db	13.32	0.53
48	22.32Ba	18.46Cb	20.39	0.53
72	27.43Aa	23.18Bb	25.30	0.53
96	29.39Aa	27.03Ab	28.31	0.53
Mean	17.21	14.45
SEM	0.45	0.45
Significance (P =)
Nutritional value × incubation time	0.0017

DM - dry matter; SEM - standard error of the mean.

Volatile fatty acid concentration and the acetate:propionate ratio were influenced by the interaction between the straw nutritional value and the incubation time (Table 8). The greater acetic acid concentration and the lowest propionic acid concentration were observed for the low-nutritional value straw with 6 h of incubation (71.80 and 21.03 mol 100 mol^–1 total volatile fatty acid, respectively). Thus, the greater acetate:propionate ratio was also observed for this straw (3.60 mol 100 mol⁻¹ total volatile fatty acid) (Table 8). However, the greater concentration of butyric acid was observed for the high-nutritional value straw with 6 h of incubation (7.61 mol 100 mol^–1 total volatile fatty acid).

Table 8

Incubation time (hours)	Rice straw nutritional value		Mean	SEM
	High	Low
Acetate
6	67.58Ab	71.80Aa	69.69	0.54
12	68.55A	69.86AB	69.21	0.54
24	66.94A	67.85B	67.40	0.54
48	68.19A	67.73B	67.96	0.54
72	68.34A	66.84B	67.59	0.54
96	69.02A	68.71AB	68.86	0.54
Mean	68.10	68.80
SEM	0.31	0.31
Significance (P =)
Nutritional value × incubation time	0.0057
Propionate
6	24.79Ba	21.03Db	22.91	0.44
12	25.73AB	23.20CD	24.46	0.44
24	27.15AB	25.96AB	26.55	0.44
48	27.81A	28.89A	28.35	0.44
72	26.51AB	28.17AB	27.34	0.44
96	25.79AB	25.97AB	25.88	0.44
Mean	26.30	25.53
SEM	0.25	0.25
Significance (P =)
Nutritional value × incubation time	0.0057
Butirate
6	7.61Aa	7.16Ab	7.39	0.47
12	5.71AB	6.93A	6.32	0.47
24	5.90AB	6.18A	6.04	0.47
48	3.98B	3.37B	3.68	0.47
72	5.13B	4.98AB	5.05	0.47
96	5.18B	5.31AB	5.24	0.47
Mean	5.58	5.56
SEM	0.19	0.19
Significance (P =)
Nutritional value × incubation time	<0.0001
Acetate:propionate ratio
6	3.12Ab	3.60Aa	3.36	0.31
12	2.71AB	3.08B	2.90	0.31
24	2.48B	2.67BC	2.58	0.31
48	2.47B	2.37C	2.41	0.31
72	2.60B	2.40C	2.50	0.31
96	2.73AB	2.68BC	2.70	0.31
Mean	2.68	2.80
SEM	0.30	0.30
Significance (P =)
Nutritional value × incubation time	0.0103

SEM - standard error of the mean.

In this study, the highest IVDMD and IVOMD were observed for the straw with less lignification and silicified cell wall (high nutritional value) when supplemented with mineral and protein-energy supplement. For the same straw, mineral and protein-energy supplement inhibited the effect of the exogenous fibrolitic enzyme, since the combination of these supplements resulted in a lower in vitro digestibility in relation to other supplements. These results suggest that the improvement of IVDMD and IVOMD depends on the forage chemical characteristics and the supplement used, corroborating with Morgavi et al. (2000), who stated that more detailed knowledge of the interaction between the supplement with the forage, the host, and the rumen microorganisms is necessary for the correct application of this technology. Previous research had also identified variation in rice straw digestibility (Vadiveloo, 1992; Vadiveloo, 1995) and improvement of forage in vitro digestibility with the use of mineral and protein-energy supplement (Barbosa et al., 2007) and exogenous fibrolytic enzyme (Beauchemin et al., 2003; Bassiouni et al., 2011).

With the use of supplements, an improvement in the parameters of in vitro organic matter ruminal degradability of rice straw was expected. However, the supplements did not help carbohydrate release and did not provide enough nitrogen to improve these parameters, probably due to the limiting nitrogen content in the incubated straws.

Gas production is an indirect measure of substrate degradation, mainly of carbohydrates (Menke, 1979). In the current research, there was interaction between the incubation time and the nutritional value of rice straw on the in vitro cumulative gas production. At 9 and 12 h of incubation, there was an increase in gas production due to accumulation of indirect gas products of reaction between the buffer and the propionic acid generated from the fermentation of rapidly degradable carbohydrates and the indirect gas that starts to be produced from the structural carbohydrate degradation. According to Chai et al. (2004), the gases produced in the first 3 h of incubation correspond to the soluble components of the fermentation. To the extent that the incubation time increased, the volume of gas produced was increased by the effect of the structural carbohydrate fermentation of the substrate (Theodorou et al., 1994) and at the end of 96 h of incubation, greater cumulative gas production was observed for high-nutritional value straw due to greater organic matter degradadability in this straw, corroborating with Menke et al. (1979). However, the fermentation pattern was similar between straws and gas production curves corresponded to the fermentation pattern to a substrate with forage predominance, in which, initially, sugars are fermented and later, the structural components (Getachew et al., 2005).

Supplementation with MPES and MPES + ES did not affect the in vitro cumulative gas production. Liu and Ørskov (2000) reported that treatment of rice straw with several levels of cellulase did not affect the cumulative gas production over 24 h of incubation. However, Eun et al. (2006) treated Akibali rice straw with six different enzyme products (1.25 mg g^–1 DM) and found that only two (product composed of cellulase and hemicellulase and product containing protease) increased cumulative gas production over 24 h of incubation compared with untreated rice straw. Therefore, the inconsistencies in the responses with enzyme supplements were probably related to supplement characteristics, including the enzyme activities in the conditions of rumen (temperature and pH), as well as the substrate composition (Yang et al., 2011).

The partition factor is an indicator of fermentation efficiency; thus, high value of partition factor indicates a greater incorporation of degraded organic matter in microbial mass, thereby increasing the microbial synthesis efficiency. The greater the partition factor value, the greater the forage dry matter intake (Makkar, 2004) and lower the CH₄ production in ruminants (Blümmel et al., 1999). According to Makkar (2004), partition factor may vary from 2.74 to 4.41 mg of degraded OM mL^–1 of gases produced. In the current research, the partition factor values ranged from 2.37 to 2.88 and were not influenced by the straw nutritional value or the supplement.

The pH is an important variable to indicate the rumen status (Gunun et al., 2013), since it regulates the affinity of microorganisms with the substrate. Thus, values near neutral pH improve the bacteria adhesion to the fiber (Allen and Mertens, 1988). In the current study, the pH ranged from 6.94 to 7.19. These values are considered optimal for the normal rumen fermentation, for the synthesis of volatile fatty acid, and microbial protein (Wanapat and Pimpa, 1999; Anantasook et al., 2012), and are also within the range of 6.2 to 7.2, considered appropriate for optimal microbial activity (Van Soest, 1994), as expected for diets based on forage.

In vitro NH₃-N concentration works as an indicator of protein degradability because there is no nitrogen absorption or recycling, as in the in vivo rumen environment (Detmann et al., 2011). The average NH₃-N concentration of all treatments was 14.21 mL dL^–1 and was within the optimum ruminal NH₃-N range of 12 and 17 mL dL^–1 for optimal fermentation and rumen microbial growth (Anantasook and Wanapat, 2012; Lunsin et al., 2012). As expected, MPES and MPES + ES supplements increased the dietary levels of carbohydrate and nitrogen, resulting in an increase in NH₃-N concentration levels and decrease in CO₂ production due to the microbial mass formation.

Within the first 12 h after incubation, there was lower CH₄ production for both straws evaluated, as this period includes the lag-time phase, in which there is no methanogenesis until the sites available for microbial attachment are saturated and these synthesize its structures and enzymes (Franco et al., 2013). The linear increase in CH₄ volume for high and low straw nutritional value, from 12 h until the end of the incubation period, was associated with the slowly digestible fraction fermentation and, consequently, with acetic and butyric acid production (Getachew et al., 2005; Lee et al., 2011). The production of CH₄ at 96 h of incubation was greater for high-nutritional value straw compared with low value straw, possibly due to better digestibility of the former, corroborating with Kurihara et al. (1995), who observed that CH₄ production in cows fed forage with low digestibility was lower than in cows fed high forage digestibility. However, this disagrees with other studies that observed that CH₄ production tends to decrease with increasing protein concentration and tends to increase with increasing fiber content of the feed (Johnson and Johnson, 1995; Getachew et al., 2005). Another possibility may be related to the fiber degradability, since forage with greater content of effectively degraded fiber promotes greater CH₄ production (Demarchi et al., 2003).

Due to the lack of supplement effect on the degradation parameters, cumulative gas production, gas production kinetics, and CH₄ production, volatile fatty acid concentrations were measured to further explore any potential effect of the supplements on the rumen fermentation. However, the supplements did not affect the volatile fatty acid concentrations, but these concentrations were influenced by the interaction between the straw nutritional value and the incubation time. High levels of volatile fatty acids observed at the beginning of the fermentation can be explained by rumen fluid being obtained from animals fed a diet based on alfalfa hay. The dominance of the acetic acid concentration observed in the current study shows that when the diet had high forage content, ruminal fermentation occurred preferentially in this way and was associated with high CH₄ production, corroborating with Nussio et al. (2011).

Acetic:propionic acid ratio is an important point in rumen methanogenesis, since greater energy losses in the CH₄ form is related to the greater acetic:propionic acid ratio (Johnson and Johnson, 1995). Also, as the propionic acid is the most important fatty acid precursor of the glucose synthesis (Nagajara et al., 1997), a low acetic:propionic acid ratio reflects an improvement of the food nutritional value. In the current research, there was no influence of the supplement on the acetic:propionic acid ratio and on improvement of the straw nutritional value. The results did not differ from results of Eun et al. (2006), who observed reduction in theacetic:propionic acid ratio with EX and PROT enzymatic treatment of rice straw, suggesting that microbial interactions lead to decreased acetate and increased propionate formation from the products of cellulose and hemicellulose hydrolyses when certain types of exogenous enzymes were added to rice straw.

The use of mineral and protein-energy supplement and mineral and protein-energy + exogenous fibrolytic enzymes supplements can be used as strategy to mitigate carbon dioxide in ruminant production systems that use rice straws.