Amylolytic Enzyme in Rehydrated Corn and Sorghum Grain Silage in the Diet of Ears

Aimed of this work was to evaluate rehydrated corn and sorghum grains silages, with and without α-AMYLASE, on fermentation pro�le, nutritional value, digestion and metabolism on diets for sheep. Two experiments were conducted. In the �rst experiment 28 silos were divided into: 1-RSGS (rehydrated sorghum grain silage); 2-RSGS+A (rehydrated sorghum grain silage + amylase,); 3-RCGS (rehydrated corn grain silage); 4 RCGS+A (rehydrated corn kernel silage + amylase). In experiment II, 12 lambs were distributed in: RSGS; RSGS+A; RCGS; RCGS+A. In experiment I, there was an effect of grain x enzyme interaction for GL MN. The enzyme reduced the NFC content. In experiment II intake of DM kg/day was not affected by the starch content in the diet, with an average of 1.2 kg/day meaning that the type of grain and the enzyme had no in�uence on the intake of animals. There was an interaction for the intake of starch (kg/day), blood urea and N-NH 3 . Lambs fed with RCGS+A had a higher concentration of ruminal ammonia. The use of enzymes improved the fermentation pro�le of the silages. RSGS can replace RCGS in sheep feed without modifying nutrient intake and digestibility.


Introduction
The high starch content in corn and sorghum grains encourages the use of these foods in ruminant diets, as they are a source of energy for the growth of ruminal microorganisms, which are responsible for producing volatile fatty acids (VFAs), responsible for the supply animal energy (Faustino et al., 2018).
However, the digestibility of starch granules can be affected by hydrophobic matrix, which prevents the xation of microorganisms in rumen, reducing starch digestibility (Ferraretto et al., 2015).
The hydrophobic effect of prolamins is reduced when the grains undergo mechanical, chemical and enzymatic processing that is capable of breaking the hydrogen bonds, releasing starch granules (Silva et al., 2020).
The reconstitution and silage process of the grain makes starch more available to be used in the rumen fermentation process (Ferraretto et al., 2015).The addition of exogenous enzymes in the silage of rehydrated grains helps the starch digestive process (Oliveira et al., 2019).
Considering the effect of rehydration and the use of α-AMYLASE, the hypotheses raised in this work were that rehydrated sorghum silage without and with enzyme has a nutritional value similar to rehydrated corn silage, digestion and metabolism and that will not be in uenced by the type of grain and/or enzyme utilization, rehydrated sorghum grain silage can replace rehydrated corn grain silage without loss of animal performance.
The aimed of this research was to evaluate the rehydrated corn and sorghum grains silages, without and with α-AMYLASE, on fermentation pro le, nutritional value, digestion and metabolism on diets for sheep.

Experiment I
The corn and sorghum grains used in this experiment were harvested during the 2018/2019 harvest.After harvesting, the corn and sorghum grains were ground in knife mills (4 mm), hydrated until reaching a dry matter content between 50 and 55% and homogenized.The experimental silos were made in polyethylene tubes 40 cm in height and 30 cm in diameter.At the bottom of the silos, dry sand (2 kg) was placed, separated from the forage by a nylon cloth (50 mesh -porosity) to quantify the e uent produced.The material was compacted with a density of 930kg/m 3 .The experimental silos were sealed with adhesive tape, weighed and stored.The chemical composition of the material before ensiled is shown in Table 1.Twenty-eight experimental silos were used, distributed in a 2 X 2 factorial scheme, with two levels of enzyme and two cereals, where the treatments were: Treatment 1: rehydrated corn silage without the addition of α-AMYLASE enzyme; Treatment 2: rehydrated corn silage with α-AMYLASE (amylase, Kerazyme 3035, enzymatic activity 300 U mL −1 ); Treatment 3: rehydrated sorghum silage without the addition of α-AMYLASE enzyme; Treatment 4: rehydrated sorghum silage with α-AMYLASE (amylase, Kerazyme 3035, enzymatic activity 300 U mL −1 ).
At 45 days of fermentation, the silos were weighed again to determine gas losses and then opened, samples were collected and taken to pre-drying in a forced ventilation oven at 55º C for 72h and after this period they were ground.in a Willey type mill at 1 mm, and dried for 16h in an oven at 105º C to determine dry matter (DM, method 950.15), ash (method 942.05), organic matter (MO, 1000-ash), crude protein (CP, N × 6.25, Kjeldahl method 984.13) ether extract (EE, method 920.39), acid detergent ber (ADF) and lignin (method 973.18), (AOAC, 2000).(NDF) (Van Soest et al., 1991) The starch content was determined according to the methodology described by Hendrix (1993).The concentrations of non-brous carbohydrates (NFC) were obtained from the equation: NFC = 100 -(%CP + %EE + %ashes + %NFC).
Microbiological analyzes were performed according to Silva et al. (1997).
For ruminal kinetics, three animals cannulated in the rumen were used, distributed in a randomized block design.The bags were then deposited in the ventral sac region of the rumen by incubation time in reverse order (0, 3, 6, 9, 12, 24, 36 and 48 hours) to be removed all at the same time, at the end of the period, and in this way, promote uniform washing of the material when the rumen is removed.The remaining residues were quanti ed for DM contents (950.15)(AOAC, 2000).

Experiment II
Twelve whole lambs (27.3 ± 7.5 kg of body weight and 6.4 ± 0.3 months), were distributed in three 4 × 3 Latin squares, consisting of periods of 14 days.Diets were formulated with an average daily gain of 200 g, using the Small Ruminants Nutritional System (Table 2).The animals were housed in metabolic cages, fed twice a day.
To assess intake, leftovers were weighed daily and the supply adjusted for ad libitum intake with leftovers calculated at 10 to 15%.
To estimate the total apparent digestibility of dry matter and nutrients, total feces were collected from the 15th to the 20th day of each experimental period.The samples obtained were homogenized to compose a sample composite of each animal in each period.The stool samples collected were pre-dried in an oven with forced ventilation (60°C/72 hours) and processed in a knife mill with 1 mm porosity sieves.Subsequently, these samples were analyzed for DM, MO, PB, EE, NDF and starch according to the methodology previously described for food analysis.At 12th and 13th day total urine collection was performed to quantify the urinary volume.The samples were stored for analysis of allantoin, uric acid, xanthine and hypoxanthine (Chen and Gomes 1992).
From the 15th to the 19th of each experimental period, total urine collections were performed to quantify the urinary volume.Spot samples were collected during spontaneous urination, at 11:00 am, that is, four hours after the provision of the meal at 7:00 am.A 10 mL aliquot of urine was diluted in 40 mL of 0.036 N sulfuric acid.
In this process, the pH was adjusted, if necessary, to values below 3, with droplets of concentrated sulfuric acid, in order to prevent bacterial destruction of the purine derivatives and precipitation of uric acid.The samples were stored at -18 ºC for further analysis of purine derivatives allantoin, uric acid, xanthine and hypoxanthine (CHEN and GOMES 1992).
For the nitrogen balance, the quanti cation of nitrogen content in urine and feces was performed according to (AOAC, 2000).The calculation was performed according to the following formulas: On the 21st day, 4 hours after feeding, ruminal uid was collected through an esophageal tube (Ortolani et al., 1981).After collection, pH was measured.Ammonia was determined by the methodology of Broderick and Kang (1980), and SCFAs by Erwin et al. (1961).

Experiment I and II
The data obtained were submitted to SAS, verifying the normality of the residues and homogeneity of the variances.The data were analyzed by PROC MIXED adopting a signi cance of 5%, according to the model: Y i = dependent variable, µ = overall mean, S i = silage xed effect (i = 1 to 2); E j = enzyme effect and S i * E j = interaction effect and e ij = error.The degrees of freedom were corrected by DDFM=kr.
Y ijkl = µ + A i + P j + C k + D l + e ijkl Y ijyk = dependent variable, µ = overall mean, A i = animal effect (j = 1 to 12), P j = period effect (y = 1 to 3), C k = squared effect (k =1 to3), D l = diet effect (l = 1 to 3) and e ijkl = error.The random effect of the model was characterized by: A i and P j.

Results
Experiment I The type of grain, enzyme and interaction of these factors (P =0.452 and P =0.317) did not affect e uent losses in kg/ton or %DM.The enzyme increased losses by gas losses in natural (GL MN; P=0.026), with the effect of grain x enzyme interaction.
RCGS+A had the highest LAB count, in uencing a reduction of 0.78% in pH and 23% in the production of N-NH 3 (%NT).The enzyme reduced the fungus count by 2.52% (Table 4).There was an effect of the enzyme on LAB (P=0.022),fungi (P=0.004),pH (P=0.021) and (N-NH 3 (%NT)) (P=0.001).The enzyme reduced the NFC (P=0.032),increased (P=0.017) the NDF content, reducing the starch content (P=0.047).RCGS had higher levels of OM, EE, starch, TND and NEg compared to RSGS (Table 5).Starch intake (P=0.047)showed a grain x enzyme interaction effect.RCGS+A had 8% higher than the other treatments (Table 7).Lambs fed with RCGS+A had a higher concentration of ruminal ammonia compared to the other treatments.There was interaction of the type of grain with the enzyme on blood urea and N-NH 3. RCGS+A had the highest production of N-NH 3 .There was an effect of the interaction of grain x enzymes on the fermentative pro le for N-NH 3. The animals fed with silage-containing enzymes showed higher (P=0.021,P=0.012 and P=0.044) production of ammonia, propionate and methane compared to animals that did not receive enzyme silage.Lambs fed with RCGS had higher production of butyrate and total acids of 30.65% and 3.8% compared to those fed RSGS (Table 9).

Discussion
The lower REC DM for silages with α-AMYLASE is effect of the higher LG in this material, an effect caused by the enzyme that provided a greater amount of substrates, causing a greater amount of aerobic bacteria to occur, competing with LAB, for sugars and consuming the organic acids, in uencing for an extension of the fermentation, producing CO 2 , generating greater losses of DM (Oliveira et al., 2019).
The higher LG in silages containing enzymes is linked to secondary fermentation, carried out by yeasts, enterobacteria, and aerobic bacteria that grow at a pH close to 4, using fermentable carbohydrates, found in greater quantities in silages containing enzyme, due to the action of α -AMYLASE that degrades starch into smaller molecules, making them available for use by microorganisms, resulting in an increase in LG and reducing REC DM, (Gizotto et al., 2020).
The pH of silages containing α-AMYLASE was not negatively in uenced by the lower amount of LAB, with a pH close to 3.8 which reduced the amount of fungi in the silage, in uencing the nutritional value of the silage, as it was not found a signi cant difference in chemical composition compared to other treatments.
Losses of DM during the fermentation process result in increased concentrations of more than 11% NDF, due to the effect of concentration, as it is an insoluble fraction, which is not used by microorganisms as a substrate during the period that the material was ensiled (Oliveira et al., 2019).
The 2.67% reduction in NFC is due to the action of α-AMYLASE, which solubilized part of the starch in sugars that can be fermented by LAB, reducing their concentrations in the ensiled material.The reduction in NFC concentrations was expected due to the rehydration of grains that potentiate the activity of endogenous and exogenous amylolytic enzymes, increasing starch degradation, increasing the concentration of propionate in the rumen, in uencing a 16.5% reduction in the production of methane (Mombach et al., 2019).
With the exception of RCGS, all silages had acceptable pH values (3.6 to 4.2) (Oliveira et al., 2019).This nding may be related to the greater amount of ammonia that neutralized the desirable acids resulting in a nal pH increasing even with a greater amount of LAB (SILVA et al., 2020), corroborating the high presence of fungi, total and aerobic bacteria.
RCGS was the treatment with the lowest fraction "a" (37.61%), this is due to the greater microbial activity observed, which consumed highly fermentable carbohydrates, reducing the amount of carbohydrates in the "a" fraction (Oliveira et al., 2019).
The treatments containing α-AMYLASE had a higher soluble fraction, and RSGS+E had a higher fraction "a" (54.09), due to the effect of α-AMYLASE that hydrolyzed starch into smaller molecules, intensifying the amount of fermentable carbohydrates, facilitating the action of bacteria in the rumen environment (Oliveira et al., 2019), however, was the treatment that presented the highest fraction C (0.09%) and FI (23.9%).
The intake of DM kg/day was not affected by the starch content in the diet, with an average Intake of 1.2 kg/day meaning that the type of grain and the enzyme had no in uence on the intake of animals, being in uenced by the concentration of propionate and lactate in the rumen.Due to the hypophagic effect of propionate, potentiated by increasing its concentration in the rumen, reducing Intake by increasing the amount of glucose in the bloodstream, stimulating the hypothalamic neurons in the satiety center (Silva et al., 2020).
The increase in the concentration of propionate in the rumen reduced the production of methane enteric, reducing the concentration of acetate being the main precursor for the production of methane, however the formation of propionate does not produce carbon dioxide, conserving more energy and decreasing the production of methane (Chandrakar et al., 2021).
All treatments had ruminal pH close to 5.8, being in uenced by the production of lactic and propane, which is considered a low value, which may be indicative of ruminal acidosis, caused by the high amount of starch present in the diets and its high digestibility (< 90%), increasing rumen fermentation and consequently lowering the pH, in addition to in uencing DM Intake (Silva et al., 2020).
The increase in propionate is related to the high passage rate found in treatments containing enzyme with an average of 50% readily available starch, which in uenced the increase in starch intake, being of rapid fermentation, in uencing the production of lactic acid being one of the main drivers for the production of propionate (Giubert et al., 2013).
The lower production of ammonia in the rumen of animals fed with RCGS (23.59%) is associated with lower hydrolysis of sorghum grain protein, contributing to lower fermentation and consequently lower production of ammonia in the rumen (Chandrakar et al., 2021).

Conclusion
The use of amylolytic enzymes improved the fermentation pro le of the silages, reducing the pH values and the concentration of N-NH3 (%NT).The enzyme-containing silages showed a signi cant increase in effective degradability (DF).
Rehydrated sorghum grain silage can replace rehydrated corn grain silage in sheep feed since no signi cant difference was found in nutrient intake and digestibility. Declarations Ruminal KineticsData were tted to a nonlinear regression using the SAS (Sas Institute, NC, Cary) (Orskov andMcdonald,  1979):Y=a+b (1-e -ct), Y = accumulated degradation, after time t; a = intercept of the degradation curve when t = 0; b = potential for degradation of the water-insoluble fraction; a+b = potential degradation when time is not a limiting factor; c = rate of degradation by the fermentative action of b; t = incubation time.

Table 1
Chemical composition of grains before ensiling (percentage based on DM)

Table 3 .
Fermentation losses according to experimental treatments

Table 9 .
Ruminal fermentation pro le