Quality of sorghum hybrid silages at different storage times

.


Introduction
Livestock production has an impact on the global economy, being essential for the economic survival of several regions around the world.In the Pampa Biome in Brazil, Argentina, and Uruguay, livestock production is the main economic activity, with cattle and sheep being the dominant species (Marchi et al., 2018).However, due to subtropical and temperate climates with four, well-characterized seasons (Roesch et al., 2009) and extremely sandy texture and fragility of soil, the natural grasslands have seasonal forage production and quality (Rubert et al., 2018).
To balance the forage seasonality caused by droughts and low temperatures, and to ensure the sustainability of livestock production in the Pampa region, forage planning must be adopted (Malaguez et al., 2017).The production of silages is one of the options in a forage planning system, and for less fertile soils, the sorghum plant is the most suitable.
Silage making is a method of feed preservation in anaerobic conditions, whereby biochemical processes in ensiled forage ensure a reduction in pH and preservation of nutritional value.However, this can be altered by failures in the ensiling process or by the fermentative dynamics during silage storage.The chemical composition of silages can be changed depending on the region of production (Bernardes et al., 2018).Information on silage sorghum fodder obtained in the Pampa region is scarce, as is analyses of the nutritional content over time.Thus, the objective was to study was to study the nutritional changes in the silage of four sorghum hybrids after different storage times.

Material and Methods
The work was conducted at the Federal University of Pampa Experimental farm and Animal Nutrition Laboratory of Unipampa -Uruguaiana Campus, located at the Rio Grande do Sul, Brazil (29°45'17" S; 57°05'18" W; 66 m above sea level).
The design was completely randomized plots; subdivided in a 4 × 9 arrangement, with four replications.The plots were allocated to four sorghum hybrids, Qualysilo, Chopper, Dominator, and Maxisilo.In the subplots, sampling times were considered in fresh material (time zero) and after 1, 3, 7, 14, 28, 56, 112, and 224 days of storage.
Cultures were implanted using a continuous flow seeder system with a spacing of 0.34 m.At the time of sowing, the seeds were treated with CRUISER® (thiamethoxam) insecticide.A base fertilizer of 120 kg/ha of formulated 8:20:15 (N:P:K) was used.A cover fertilizer of 50 kg/ha of nitrogen was applied as urea 45 d after sowing.
The harvest was carried out when it was identified that the panicle of the plants presented grains with 70% pasty consistency and 30% milky consistency, obtaining a forage mass with an ideal dry matter content for silage of ~35%.With the aid of a tractor, the green material was cut 15 cm from the ground and the equipment knives were adjusted for particle chopping of 2-5 cm.
The harvested material was stored in experimental silos of polyvinyl chloride (PVC), equipped with a Bunsen-type valve to allow gases to escape during fermentation.At the bottom, 0.5 kg of clean sand was used for the purpose of draining the effluents during storage.For adequate compaction, a density of 600 kg/m 3 was used.
At the time of the opening of the silos, hydrogen potential (pH) of silages was measured according to the method of Cherney & Cherney (2003).Ammonia nitrogen (NH3-N) was determined according to the method of Bolsen et al. (1992).
The organic matter (OM) content was calculated according to the formula: The concentrations of neutral detergent fibre (NDF) and acid detergent fibre (ADF) were determined according to the method of Van Soest et al. (1991).The hemicellulose (HEM) content was calculated as the difference between NDF and ADF.Neutral detergent insoluble nitrogen (NDIN) and acid detergent insoluble nitrogen (ADIN) levels were determined according to the method of Licitra et al. (1996).Concentrations of total carbohydrates (TC) and non-fibre carbohydrates (NFC) were obtained using the formulae (Sniffen et al. (1992): Total digestible nutrients (TDN) were calculated according to the method of Bolsen (1996) and dry matter consumption as a percentage of live weight (DMCLW) was calculated by estimate, according to the method of Mertens (1997).Digestible dry matter (DIGDM) was determined according to the method of Rohweder et al. (1978).
For statistical analysis of the data, the SISVAR statistical program was used.The data were submitted to variability analysis and, when the significance was established, the parameters were compared using Tukey's test (5%).All silage storage times were compared using regression analysis.Linear and quadratic models were tested, and the coefficients of the equations for each model were tested at the 5% level of significance using Student's t-test.To choose the regression model that best explained the behaviour of the data, the determination coefficient (R²) was considered in addition to the significance.Only the equations of the contents that obtained R 2 ≥0.6 were discussed; the equation generated by the analysis explained 60% of the variation in the content in relation to time, whereas 40% of the variation was random.

Results and Discussion
There was an effect of the interaction of factors on the variables studied, except for CP, FC, NDF, TDN and DMCLW, which were affected only by the isolated factors (Table 1).
The dry matter contents did not fit the tested regression models, and in all silage times, the lowest DM content was observed in the silages obtained with Maxisilo (Table 3).As forage silage with a low DM content causes quality losses (Borreani et al., 2018) and promotes the development of Clostridium (Driehuis et al., 2018), this material must be carefully monitored in respect of the DM content of the silage.For obtaining quality silages, the indicated DM required is ~30% (McDonald et al., 1991) for a rapid reduction of pH due to the action of lactic acid bacteria (Muck et al., 2018) and the inhibition of the development of undesirable microorganisms (Driehuis et al., 2018).
Qualysilo, Chopper, and Dominator silages increased NH3-N up to 153, 50, and 168 days, respectively, with subsequent reduction, whereas in Maxisilo, NH3-N increased 0.0237% of total N every day of the fermentation period (Table 2).NH3-N is the result of the plant and microbial proteolysis that occurs inside the silo and is higher in humid silages due to the action of microorganisms of the genus, Clostridium (Kung et al., 2018).
Proteolysis starts right after cutting the forage and extends during ensiling until the time when acidic conditions are established inside the silo (Woolford, 1984).The NH3-N production in silos occurs through biochemical means called deamination, decarboxylation, oxidation, and reduction (Pires et al., 2013), and, together with other non-nitrogen compounds, they interfere with ruminal metabolism and the regulation of consumption in ruminants (Grant & Ferraretto, 2018).Although increases in NH3-N were observed in all silages, quality was preserved, because no silage time produced NH3-N higher than the threshold values of 10 to 15% of total N (Kung et al., 2018).
Rapid pH reduction was observed in all hybrids (Table 2), indicating adequate preservation of silages (Van Soest, 1994).On the third day of fermentation, Chopper, Dominator, and Maxisilo silages had a pH of <4.2; the range indicated for well-preserved silages extends to 3.6 (Moura et al., 2016).Although the Qualysilo silage took 6 d to reduce the pH, it was also fast (Table 2).Together with NH3-N, pH is a safe indicator of the fermentative quality of silages, since low values of both indicate a rapid stabilization of ensiled material (Moura et al., 2016).
Mineral matter was higher in silages obtained using hybrid, Maxisilo, after 14 d of fermentation, with the opposite result for OM (Table 3), indicating a lower nutritional value for this silage.Since feed OM contains nutrients that are potential energy providers such as carbohydrates, lipids, and proteins, its reduction can also reduce the energy potential of foods (Malaguez et al., 2017).Not a nutritionally significant source of minerals, MM contains a large amount of silica, which is responsible for the thickening of the cell wall together with lignin (Ruppenthal et al., 2016).In fodder, this mineral is characterized as a potent reducer of the digestibility of the constituents of the cell wall, with a consequent reduction in DM consumption, since its solubility in the rumen environment inhibits the action of cellulolytic ruminal microorganisms (Van Soest & Jones, 1968).Similarly, increasing the amount of silica in its dry mass increases the ash content (Tolentino et al., 2016).
The silage CP was markedly altered by the silage storage time only until the third day of fermentation, with a reduction of 1.21 g/kg per day (ŷ = -1.2106x+ 64.204; R² = 0.99).After this period, CP changes were not statistically significant, indicating that proteolysis, the process responsible for these reductions, occurred more intensely in the first days of the fermentation process.Differences were observed between the studied hybrids.The silage obtained with Maxisilo had a lower CP content (51.33 g/kg) than the others.Silage with a CP content of 57.49, 60.51, and 67.74 g/kg, respectively, were obtained from the hybrids, Qualysilo, Chopper, and Dominator, which was lower than that of Stella et al. (2016), who determined a CP content of 73 g / kg in sorghum silages.
For adequate performance, ruminants depend on a balanced supply of nutrients (Martineau et al., 2011), of which nitrogen is the most critical (Hristov et al., 2019).Nitrogen compounds are a source of amino acids, are incorporated in nucleic acids and are essential for the synthesis of microbial protein (Schuba et al., 2017).However, not all nitrogenous fractions that make up the CP of foods are used by ruminants, so their separate study becomes relevant to estimate the nutritional value of feeds (Martineau et al., 2011).
The content of NDIN and ADIN of the studied silages (Table 2) requires attention because they exceeded the upper limit of 311 g/kg of total N for conserved fodder (Machacek & Kononoff, 2009).These parameters indicate the nitrogen fractions linked to the plant cell wall and insoluble in neutral and acid detergent, respectively.As they are expressed in relation to total nitrogen, the higher the values, the lower the availability of nitrogen for use by animals, especially in the case of ADIN.Although ADIN can be partially digestible (Machacek & Kononoff, 2009), this fraction is accepted as the nitrogen that is nutritionally unavailable to ruminants (Sniffen et al., 1992) and it is negatively correlated with the digestibility of N in the diet (Van Soest, 1994).These values can be influenced by the physiological maturity stage of the plant during harvest and its lignification (Moura et al., 2016) and heat damage (Machacek & Kononoff, 2009) during prolonged exposure to temperatures above 45-50 °C (Kung et al., 2018).This condition favours the occurrence of the Maillard reaction (Gayer et al., 2019), which is the non-enzymatic chemical polymerization of soluble sugars and hemicellulose with amino acids of food (McDonald et al., 2010).The content of EE was higher in Chopper silage in practically all fermentation times (Table 2) due to the greater participation of grains in the ensiled mass of this hybrid.In the study of TC, no statistically significant differences were observed, with an average value of 846.21 g/kg.The NFC content was higher in the Maxisilo silage, whereas the highest concentration of NFC from the seventh day of fermentation was observed in the Chopper silage (Table 3).As NFC are rapidly fermentable in the rumen, they provided a higher energy content in the Chopper silage (TDN: 596 g/kg) compared to the other hybrids (Qualysilo: 519 g/kg; Maxisilo: 503 g/kg, and Dominator: 517 g/kg).These, since they are synchronized with the availability of nitrogenous compounds, are the most efficient in increasing the production of milk and meat as they are substrates for the synthesis of microbial protein.For this reason, its maintenance throughout the silage storage period, as observed in this study (Table 2) and also by Hristov et al. (2019) in corn silages, is desirable.
As NDF is directly related to the consumption of DM (Hristov et al., 2019), these parameters must be studied together.When studying the cell wall components of the silages, it was found that the NDF did not change over the storage period, in agreement with the results of Hristov et al. (2019), who studied long-term corn silage storage.Of the studied silages, Maxisilo had a higher NDF content (734 g/kg) than the others (average: 547 g/kg).For DM consumption expressed in relation to percentage of live weight, the Chopper hybrid silage could be added in greater quantity to the diets, reaching up to 2.08% of live weight.With a lower inclusion limit and requiring greater caution in its use in diets, especially in high production animals, Maxisilo silage was obtained with its inclusion limited to 1.74% of live weight.For both silages of the Qualysilo and Dominator hybrids, the consumption potential obtained was intermediate and identical at 1.89% of live weight.
NDF is the main source of energy in ruminants (Krämer-Schmid et al., 2016), and together with CP, reflects the nutritional value of bulky feed (Zhang et al., 2018).Thus, as an NDF content in the diet above 485 g/kg and 353 g/kg limits consumption and performance of dairy cows and beef cattle in feedlots, respectively (Arelovich et al., 2008), due to the rumen-filling effect (Krämer-Schmid et al., 2016), the limits of inclusion of these silages in diets should be considered.
In agreement with the results obtained with the NFC, the ADF was lower in the silages of the Chopper hybrid, whose result was mainly influenced by the content of lignin, a component of the ADF, also lower in the silages of this hybrid (Table 2).The NDF content does not affect the digestibility of the ADF, however, the reverse occurs (Kendall et al., 2009).For this reason, it is essential to know the ADF content of bulky feed for ruminants.The ADF is used as a predictor of digestibility due to the presence of lignin (Machacek & Kononoff, 2009).Agreeing with results observed by Hristov et al. (2019) in corn silages, this fraction increased in all silages over the storage period (Table 2).The increase is due to the concentration of lignin in the ensiled mass resulting from the consumption of carbohydrates in the fermentation phase of the silages.
High levels of lignin in silages are not desirable because it is practically indigestible and chemically phenolic; it causes a physical and chemical barrier to fibrolytic microorganisms, compromising the digestibility of the fibrous fractions of the diet.Although NDF is the most widely-used consumption predictor, feed with a high ADF also has potential for consumption depression due to the filling effect (Poczynek et al., 2020).Thus, the higher the lignin content, the greater the filling effect (Poczynek et al., 2020) and the lower the digestibility of the diet (Marcos et al., 2018), both accentuated by cellulose content (Table 2).
The B2 fraction was higher in Maxisilo silages (447 g/kg) and lower in Qualysilo silage (329 g/kg) while Chopper (409 g/kg) and Dominator (370 g/kg) silages were similar (data presented only in the text).This fraction provides energy slowly in the rumen and can affect the efficiency of microbial synthesis and animal performance (Perim et al., 2014) when present in high proportions in the diets.The main components of fraction B2 are cell wall degradable carbohydrates, which, despite a slow rate of digestion (Du et al., 2020), are considered as potentially digestible fibre fractions in the rumen (Brandstetter et al., 2019).
Fraction C was lower in the silages of the Chopper hybrid (Table 2), indicating a silage of better nutritional quality.Fraction C corresponds to unavailable fibre fractions and lignin (Brandstetter et al., 2019), representing the cell wall carbohydrates unavailable for ruminal use (Du et al., 2020).Increases in this fraction in feed are related to lower NDF digestibility and, consequently, there is a greater potential for this to have a filling effect (Brandstetter et al., 2019).
The digestibility of DM was higher in the Chopper silage and lower in the Maxisilo silage (Table 2).As digestibility is directly affected by CP, NDF, and lignin, the less digestible silage was obtained with the Maxisilo hybrid, precisely due to the lower CP and higher lignin than the other hybrids (Table 2).As NFC and A+B1 fractions also have a direct relationship with digestibility (Du et al., 2020), due to the higher content of these fractions in the Chopper silage, greater digestibility was also observed in this silage (Table 2).Digestibility declined linearly over time in Chopper silage and a quadratic response was observed in Qualysilo and Dominator silages (Table 2).These changes are due to the consumption of NFC, especially fraction A, which due to its content of water-soluble sugars, organic acids, and short chain oligosaccharides (Du et al., 2020), is used as a substrate for fermentation inside the silo.
RVF is a quality indicator used when referring to the concentration of constituents of the plant cell wall (Gayer et al., 2019).Chopper provided silage with the best RVF (96.15%) and Maxisilo with the smallest (71.11%).Qualysilo (81.52%) and Dominator (83.94%) produced silages with intermediate and statistically similar RVF (data not shown in tables).The results found are explained by the proportion of carbohydrates in the silages, since the higher the cellulose, hemicellulose and lignin content, the lower the feed RVF, indicating lower or higher quality materials (Gayer et al., 2019).These fractions correspond to the fibrous carbohydrates in the food, which contain fractions B2 and C, which, due to their slow digestion (Du et al., 2020) and practically non-existent content (Brandstetter et al., 2019), respectively, give the forage a low nutritional value, as evidenced by the low RVF.
Digestible energy (DE) was lower in Maxisilo silage due to its higher content of fibrous constituents and lower content of rapidly fermentable carbohydrates (fractions A+B1) (Table 2).Since these carbohydrates are an important source of digestible energy in food, even in silages with desirable low levels of lignin, without the presence of fermentable carbohydrates, digestibility and DE will be low (Hristov et al., 2019).
The higher the DM content, the lower the NH3-N of the silages (Table 2), since adequate DM content inhibits proteolysis inside the silo with the lowest NH3-N production (Kung et al., 2018).This parameter in turn was negatively correlated with pH (-0.61**), ADIN (-0.56**),NFC (-0.65**),DMDIG (-0.17*),DE (-0.22**), and RVF (-0.31**) (Table 4).As NH3-N is the result of proteolysis inside the silo, intensified by the delay in pH reduction and this is dependent on the fermentation of NFC by lactic acid bacteria (Muck et al., 2018), the listed correlations confirm the direct or indirect relationship between NFC and the nutritional quality of silages.
The increase in NH3-N caused a reduction in ADIN in silages, confirming that proteolysis was not necessarily related to the Maillard reaction, whose main predisposing factor is temperature during fermentation (McDonald et al., 2010).ADIN showed a positive correlation (0.60**) with pH, indicating that failures in the production of silages that impair the initial fermentation and rapid reduction in pH contribute to the unavailability of N due to its complexation with ADF.The higher the content of NFC, the greater the initial fermentation of the silage and the faster its stabilization (Driehuis et al., 2018) by reducing the pH (-0.54**), providing silages of greater nutritional value, as confirmed by the positive correlation between NFC and DMDIG (0.50**), DE (0.55**), and RVF (0.45**) (Table 4).
The ADF showed a strong, positive correlation with fraction C (0.83**) and negative with TDN (-0.72),DIGDM (-0.95**), and DE (-0.95**) (Table 4).This is explained by the fact that they are inversely proportional.The smaller the insoluble fractions of the fibre, such as fraction C, a component of the ADF, the greater the availability of soluble fractions (Brandstetter et al., 2019).
The negative correlation between NDF and DMCLW (-0.89**) (Table 4) is due to the fact that silages with NDF levels close to or above 50% of DM can act as food consumption-limiting factors, characterizing the rumen-fill effect (Krämer-Schmid et al., 2016).As observed in the current study, the Maxisilo sorghum hybrid showed an NDF of approximately 73% of DM, corroborating this negative correlation.
The positive correlations between TDN and DMCLW (0.71**), DIGDM (0.72**), and DE (0.70**); and between DIGDM and TDN (0.99 **) (Table 4) are explained by the fact that the greater the amount of digestible nutrients (represented by TDN) the better the digestibility of the food (DIGDM) (Du et al., 2020), and consequently the more DE is available to the animal.High content of TDN enables a higher ruminal passage rate, thus increasing the DMCLW (McDonald et al., 2010).In the fractions of carbohydrates, the A+B1 fraction stood out with a positive correlation with DMCLW (0.75**).

Conclusion
The sorghum hybrids studied showed distinct nutritional characteristics.The production of silage from these sorghums can be carried out, as they all fit into the nutritional profiles proposed by the literature, characterizing them as good quality It is noted that the Chopper hybrid performed better, due to better digestibility levels, which may result in greater animal performance.

Table 1 .
Analysis of variance showing mean squares and sources of variation in the characteristics of the silages obtained from four sorghum materials at * significant at 1% and 5% probability by the F test, respectively.DF: degree of freedom; DM: dry matter; NH3-N: ammoniacal nitrogen (% of total N); pH: hydrogen potential; MM: mineral matter; OM: organic matter; CP: crude protein; NDIN: neutral detergent insoluble nitrogen; ADIN: acid detergent insoluble nitrogen; EE: ether extract; NFC: non-fibre carbohydrates; FC: fibre carbohydrates; ADF: acid detergent fibre; NDF: neutral detergent fibre; HEM: hemicellulose; CEL: cellulose; LIG: lignin; A+B1: A+B1 fraction of carbohydrates; Frac B2: B2 fraction of carbohydrates; Frac C: C fraction of carbohydrates; TDN: total digestible nutrients; DMCLW; dry matter consumption as a percentage of live weight; DIGDM: digestible dry matter; RVF: relative value of forage

Table 2 .
Nutritional analyses with an R² > 0.6 in regression analyses of silages obtained from four sorghum materials at different fermentation times

Table 3 .
Chemical analysis with an R² < 0.6 in regression analyses of silages obtained from four sorghum materials at different fermentation times