By-Products of the Babassu Agribusiness for Ruminant Diets

The use of babassu agro-industrial residues in animal feed, in addition to being an economic option of great importance in reducing the environmental impact in regions of the Brazilian Cerrado, provides the production of good quality foods of animal origin due to its nutritional characteristics. However, information related to the nutritional components of babassu by-industrial residues has not yet been sufficiently explored. The aim of this study was to evaluate the nutritional potential of some by-products from the babassu production chain through chemical composition and in situ degradability analyses. The experiment was conducted in a completely randomized experimental design, with 4 by-products from babassu processing (cake, greasy, fine flour, and 95 µm flour) and 5 replications. The by-products differ in terms of chemical composition, except for the hemicellulose content. For the degradability of fraction “a” of dry matter, it presented a higher percentage for 95 µm flour. Fine flour and 95 µm flour presented the highest fractions “b” and “c,” potential, and effective degradability of dry matter. For the degradation of crude protein, the highest percentages of potential and effective degradability were observed for greasy and 95 µm flour. The highest standardized potentially degradable fraction and the highest passage rate were obtained by 95 µm flour, which also showed greater degradation for dry matter, crude protein, and neutral detergent fiber. Among the by-products studied, the babassu cake has superior chemical composition; however, the 95 µm flour presented nutritional value and satisfactory rumen degradation to be used as an additive or in partial replacement of traditional concentrates.


Introduction
Te continuous search for alternative feeds that can replace the traditionally used feeds in formulating diets for ruminants, aiming to reduce costs while maintaining or even enhancing animal performance, is a major concern for animal nutritionists [1,2].Terefore, various agribusiness by-products have shown potential for use in ruminant feeding due to their nutritional components such as carbohydrates, proteins, minerals, fbers, vitamins, antioxidants, and bioactive compounds, which are essential for animal development, as well as their ability to improve the quality of fnal products (meat and milk) [3][4][5][6].
Technological innovations aimed at enhancing the efciency of animal production systems are necessary to sustainably address global food security demands [7].As reported by Goenag et al. [8], for every ton of fruits and vegetables produced within the food chain worldwide, approximately 10-60% of losses occur during processing, resulting in 6 million tons of solid waste and by-products annually.Terefore, the introduction of these by-products into ruminant diets represents a signifcant opportunity for reducing feeding costs, considering the seasonality of forage production [9,10], and consequently achieving greater proftability for the production chain.Additionally, it provides a viable solution for the disposal of by-products, as the vast majority of them lack established applications, are considered surplus in the production chain, and are likely contributors to environmental issues [4,8].
Babassu palm (Attalea speciosa) is cultivated throughout the Americas, Africa, Asia, and the Middle East [11].In Brazil, babassu is mainly found among the states in the North-Northeast regions, with the highest occurrence in the babassu palm forests, a transitional region between the Amazon rainforest and the semiarid region of Brazil [12].In the state of Maranhão, approximately 10 million hectares of babassu are found, accounting for about 94% of the national production [13].
Babassu is considered the world's largest source of wild seed oil, accounting for approximately 60 to 72% of the weight of the almond.However, industrial extraction of babassu oil generates 337 kg of by-products per ton of almonds used [12].Te generated by-products (cake, bran, and four from babassu mesocarp) have been increasingly used in animal nutrition as alternative feed or in replacement of other traditional feeds [11,12,[14][15][16][17].
Terefore, determining the chemical and bromatological composition, as well as the kinetic parameters of ruminal degradation of these by-products, is of utmost importance in the nutritional context, given that they have not yet been fully elucidated.Tis information serves as the starting point for formulating and balancing diets for ruminants, maximizing nutrient synchronization, minimizing energy and nitrogen losses due to ruminal fermentation, and promoting greater efciency of microbial synthesis [18].
We hypothesize that there are nutritional diferences among the babaçu by-products, which will directly afect ruminal degradation.Tus, the aim of this study was to evaluate the chemical composition and in situ degradability of babassu by-products.

Laboratory Analyses.
Laboratory analyses were conducted at the Laboratory of Animal Products (LAPOA) and the Laboratory of Bromatology and Animal Nutrition, both belonging to the Center for Agricultural and Environmental Sciences at the Universidade Federal do Maranhão, Chapadinha, Maranhão, Brazil (03 °44′33″S latitude and 43 °21′21″W longitude).

Samples and Experimental Design.
Te experiment was conducted in a completely randomized experimental design, with 4 by-products from babassu processing (cake, greasy, fne four, and 95 µm four) (treatments) and 5 replications.Te evaluated by-products were obtained from babassu processing by the company Florestas Brasileiras S.A., Itapecuru Mirim, Maranhão, Brazil.
To obtain the babassu cake, the babassu almond is washed, weighed, and ground to facilitate cooking and pressing.Cooking releases the oil particles contained in the cells; in addition, it eliminates toxins that may be present in the almond.After cooking, the almond is pressed to extract the oil, leaving the babassu cake (Figure 1).
Babassu four was obtained through the process of classifying and grinding the mesocarp as part of the full use of the coconut, in which to obtain this, it is necessary to remove impurities from the mesocarp, which is done with the aid of sieves, which they have holes with diferent diameters, which give them diferent particle sizes for commercialization.After sieving, the mesocarp is moistened and roasted.After this process, fours are generated, which difer in texture and particle size (Figure 2).At a commercial level, the babassu four used in these studies was classifed according to the following granulometry: premium four (less than 150 μ-premium type) and fne four (150 μ-type I).
Greasy babassu by-product (Figure 3) (also known as borra de babaçu) is obtained after the frst step of the refning process (degumming) that produces oil for human consumption.

2.4.
In Situ Degradability.Samples (4 g) were placed in bags (13 × 4 cm) made of nonwoven fabric with a weight of 100 g/m 2 [26].For in situ incubation, a rumen-fstulated bovine with an average live weight of 400 kg was used.Te bags were inserted into the rumen simultaneously and removed at 0, 3, 6, 12, 24, and 72 hours.Immediately after removal from the rumen, the bags were immersed in cold water and then manually washed in running water at room temperature.Te bags from time zero were used to determine the soluble fraction (fraction a); they were placed in a water bath at 39 °C for one hour and then washed together with bags from other incubation times until the water became clear.
Te bags were dried in an oven (60 ± 5 °C; 72 hours) and weighed.Te residues remaining in the bags were analyzed for dry matter, crude protein [19], and neutral detergent fber [27] contents.Te estimation of potential (PD) and efective (ED) in situ degradability parameters of dry matter and crude protein was determined based on the method proposed by Ørskov and McDonald [28]: where a � water-soluble fraction; b � water-insoluble fraction, but potentially degradable; c � degradation rate of fraction b in the nylon bag after time zero, t � incubation time (h), and k � passage rate (passage rates of 2, 5, and 8% per hour were considered [29]).Te degradability of NDF was estimated using the model proposed by Waldo et al. [30]: where Rt � residue at time t (hours); b � fraction of potentially degradable NDF in the rumen; c � degradation rate of fraction b; I � indigestible fraction.

Statistical Analysis.
Te data were subjected to analysis of variance using PROC MEANS.Te parameters a, b, and c and the in situ degradability curves of the nutritional principles were analyzed using the procedure for nonlinear models (PROC NLIN).Means were compared using the Tukey test, considering α � 0.05, using the PROC MIXED procedure.All analyses were performed using the statistical software SAS [31], version 9.0.

Results and Discussion
Te diferences in the composition of babassu by-products are primarily due to the type of processing used to obtain each by-product, combined with variations in climate and region of the samples used, as these variables can alter the chemical composition of this ingredient.Te diferent babassu by-products showed diferences in chemical composition (P < 0.05), except for hemicellulose (P � 0.0677) (Table 1).It was possible to observe that the greasy presented a higher content of DM (P < 0.05) (Table 1), which is related to its processing, obtained from residues of the epicarp (outer fbrous part; 12% of the total fruit weight) and mesocarp (grainy starch-rich layer located between the epicarp and the endocarp (23% of the total fruit weight)) [32], where these components have in their constitution a lignifed, hard, impermeable, and resistant tissue, which would limit feed intake by small ruminants.To the best of our knowledge, there are no studies evaluating the inclusion of greasy in ruminant diets, making it necessary and pertinent for future studies to elucidate the current literary limitations regarding this ingredient.
Te babassu cake showed higher contents of ash and CP (P < 0.05) compared to the other by-products (Table 1).Te values obtained are below those reported by Castro et al. [33] with 205.8 g/kg CP and 44.9 g/kg ash when using babassu cake in sheep diets and by Castro et al. [34] with 288.6 g/kg CP and 51.3 g/kg ash when incorporating babassu cake into cattle diets.Te above-mentioned authors observed an increase in the intake of these nutrients by animals with the increase in the levels of babassu cake in the diets ofered.
Te CP content of babassu cake is above the range of 6 to 8% recommended by Van Soest [25] for efective ruminal microbial fermentation, with good nitrogen availability, which can enhance fber digestion and enable increased intake by ruminants.Babassu cake is the main by-product derived from babassu, being widely used for protein enrichment of silages [17,35] and diets ofered to ruminants as a source of roughage or as a substitute for soybean meal [33,34,36].
Te fne four showed higher contents of EE and LIG (P < 0.05) compared to the other by-products (Table 1), while the 95 µm four showed higher levels of TC and TDN (P < 0.05) in its composition compared to the other studied by-products.Te results obtained difer from those found by Sá et al. [37] when evaluating the use of babassu endocarp four in the formulation of diets for sheep.Te authors found 86.6 g/kg of EE, 206.7 g/kg of lignin, 843.1 g/kg of TC, and 474 g/kg of TDN in the endocarp four I evaluated.According to the authors, endocarp four I has a fne and powdery particle size, as it is separated by a suction system, being composed only of the endocarp, whereas endocarp four II has a coarse particle size rich in fbrous bundles and small almond pieces.Tus, we can infer, based on the characteristics of the fours studied, that the fours evaluated in this study can be called endocarp four I (95 µm four) and endocarp four II (fne four).
Te diference in EE content between fne four and 95 µm four may be due to the type of processing used to obtain them, as they difer in texture and particle size, with 95 µm four classifed as premium four primarily intended for human consumption.Te EE content of 95 µm four is below the maximum level (70 g/kg of EE) recommended by Van Soest [25] as limiting to ruminal fermentation.However, the EE levels observed for fne four, greasy, and cake are above this threshold, deserving careful attention when formulating feed with these by-products.
Higher values of NDF and ADF were presented by babassu cake and fne four compared to the other babassu by-products (P < 0.05) (Table 1).Te results obtained are higher than those obtained by Dutra Santos et al. [38] for fne four (644.0 g/kg of NDF and 390.2 g/kg of ADF) and by Castro et al. [34] for babassu cake (595.0 g/kg of NDF and 303.9 g/kg of ADF).Te observed results for NDF and ADF contents may be due to contaminations during the extraction of mesocarp by epicarp and endocarp, which are fbrous [39].Te high content of fbrous components classifes babassu cake and fne four as fbrous feedstufs, whose characteristics may limit the inclusion of these ingredients in diets ofered to ruminants, mainly limiting their energy content [37].
Te observed NDF and ADF results are above the 60% NDF recommended by Van Soest [25] and the range presented by Almeida et al. [40] for ADF (20 to 30%) for use in ruminant diets.Tus, we can infer that the high proportions of NDF and ADF in babassu cake and fne four indicate the presence of lignocellulosic constituents, which are less utilized by the animals and negatively correlated with dry matter intake and digestibility [41].Babassu cake and fne four also showed higher levels of cellulose (P < 0.05), which is correlated with the high levels of NDF and ADF present in these ingredients.
Te greasy and 95 µm four show higher NFC content compared to the other babassu by-products (P < 0.05) (Table 1).Tis is likely related to the lower levels of fbrous constituents present in these by-products compared to babassu cake and fne four.Te NFC values obtained for the 95 µm four are higher than the 351.0 g/kg found by Santos et al. [14].Tere are no studies presenting the chemical composition of greasy babaçu by-product.
Te highest percentage of the water-soluble fraction (fraction a) of DM was presented by the 95 µm four (37.34%) compared to the other by-products (Table 2).Tis diference can be explained by the lower fber content observed in the 95 µm four and its particle size, which provided greater solubility of the DM compared to the other byproducts.Zeoula et al. [42] mention that the particle size of incubated feed may explain the high values of the soluble fraction due to losses in the washing process.
On the other hand, fne four showed the highest average percentage (56.93%)for the water-insoluble but potentially degradable fraction (fraction b) (Table 2).Tis percentage is related to the higher levels of NDF, ADF, CEL, and lignin that this ingredient presented in its composition.Te highest degradation rate of fraction b per hour (fraction c) observed in the greasy (8.84%/hour) (Table 2) possibly occurred due to the lower levels of NDF and ADF.
Te 95 µm four and fne four showed higher values of PD and ED of DM at the three passage rates (2%, 5%, and 8%/h) (Table 2).Tese results were a consequence of the  characteristics of fractions "a" and "b" presented by the 95 µm four and fne four and the high starch content they possess in their composition, as described by Cinelli et al. [43], where the authors observed 60.05% of starch for babassu four, which according to Van Soest [25], being a nonstructural carbohydrate, can be extensively degraded by ruminal microorganisms.
It is observed that the ED for all by-products decreased as the passage rate increased, corroborating Costa et al. [44], who emphasize that this result is due to the shorter time the feed remains in the rumen, thus reducing the time ruminal microorganisms can act.Te ED values of the 95 µm four and fne four are above those described by Sousa et al. [39] in in vitro degradability tests, who observed values for ED of DM between 27.76 and 34.01%for type I mesocarp four, and between 32.32 and 39.48%/hour for type II mesocarp four.Tis diference may be attributed to the higher content of fbrous constituents found by the authors, compared to the results obtained in this study.On the other hand, the ED values of babassu cake, at diferent passage rates, are higher than those described by Farias et al. [45], 25.75 and 33.64%/hour.
Regarding the parameters of protein degradation (Table 2), the 95 µm four presents a higher value of fraction "a," which may be related to its particle size, leading to greater losses during bag washing.Nocek [26] emphasized that particle losses during bag washing can lead to large variations (physical disappearance), which seems to have occurred with the 95 µm four.
For fractions "b" and "c" of protein, the greasy and the 95 µm four showed higher values compared to the fne four and cake.Te higher values of this fraction presented by the  Scientifca greasy and the 95 µm four may result from the lower levels of fbrous constituents (Table 1) presented by these byproducts and also from the smaller particle size of the 95 µm four.However, despite having the highest protein content (155.16g/kg) among the analyzed by-products, the cake was the by-product that obtained the lowest value for fraction "b," indicating that a large amount of protein in this by-product may be associated with NDF and ADF.Santos et al. [14], when including mesocarp four of babassu in diets for sheep, observed that as the levels of babassu four in the diets increased, the fber content also increased, which led to a reduction in protein degradation, since fbrous components such as ADF and lignin are responsible for low digestion of the cell wall in the rumen.Te greasy and the 95 µm four showed much higher potential and efective degradability of crude protein compared to the fne four and cake (Table 2).Although the groats and the 95 µm four has low CP levels (Table 1), the higher PD values observed for these by-products were due to the larger fractions "b" and "c," and because of the lower levels of NDF, ADF, LIG, and higher NFC (Table 1) compared to the other by-products.Regarding ED, these results were mainly due to the values obtained in fractions "a," "b," and "c," which are used in the equation to calculate ED; thus, the results follow the efects observed for these fractions.
Te colonization time of the by-products was similar, averaging 5.77 hours (Table 3).Tis parameter is favored by the presence of readily fermentable substrates and by the physical and chemical characteristics of the sample's cell wall, capable of facilitating microbial colonization.Te times obtained are lower than those obtained by Freitas et al. [46], who evaluated the four and cake of babassu, obtaining times of 9.88 and 8.14 hours, respectively, for the four and cake of babassu, above the values found in this study.Te diference between the times may be related to the concentrations of NFC, which, in the study by Freitas et al. [46], are higher than those found in this study.
Te PF and IF fractions originate from the degradability of NDF [17].Tus, the highest percentages of the PF fraction were observed for the 95 µm four and fne four, consequently presenting lower IF fractions (Table 3).Te higher percentage of the PF fraction exhibited by the 95 µm four was due to its lower fber content (Table 1), while the fne four, although it has high fber content, also showed a high percentage of the PF fraction, possibly due to the high starch content it contains.However, further research is needed to assess the starch content of the babassu by-products studied here.
Te highest passage rate percentage (k) of NDF was observed for the 95 µm four (5.84%/hour).It is worth noting that passage rates above 5% are recommended for highquality forages and diets formulated with a combination of forage and concentrate components [28].Te higher passage rate exhibited by the 95 µm four was likely facilitated by its lower fber content (Table 1), allowing for faster ruminal degradation.On the other hand, the greasy showed a lower passage rate (1.93%), a value close to the 2% passage rate typical of low-quality feeds [28].Te lower passage rate of the greasy suggests that this by-product is less degradable and requires more time in the rumen to be utilized.Portela et al. [35] emphasized that the less degradable the fber, the longer the feed remains in the rumen, consequently limiting feed intake due to the sensation of fullness.
Tere was no treatment × incubation time interaction for the degradation of DM, CP, and NDF (Table 4).Isolated efect was obtained for the evaluated by-products and incubation times (P < 0.0001) in the degradation of DM.Te higher value of DM degradation presented by the 95 µm four is related to its lower fber content and higher percentages of potential and efective degradation that this byproduct presents, suggesting that this by-product undergoes greater degradation compared to others, regardless of the incubation time.It can be inferred that it is better utilized by rumen microbiota, as reported by Santos et al. [14] when including babassu mesocarp four in sheep diets.
Te by-products difered in terms of treatment in PB degradation (P � 0.0235) (Table 4).Te 95 µm four showed higher degradation compared to greasy, possibly due to particle size.Te 95 µm four also exhibited greater NDF degradation compared to the other by-products (P < 0.0001) (Table 4), which are possibly related to the higher standardized potentially degradable fraction (PF) and higher passage rate (k).Additionally, the 95 µm four may have high starch content, which, as a nonstructural carbohydrate, is extensively degradable by rumen microorganisms [47].
Te assessment of babassu by-products allows us to infer that, despite the nutritional composition and degradability exhibited by four and cake, these by-products cannot completely replace traditional concentrates.However, they enable partial inclusion in animal diets, depending on the animals' requirements and ingredient prices.By-products × incubation times � interaction efect between the by-products and the incubation times; SEM � standard error of the mean; a, b: means followed by diferent lowercase letters in the row difer the by-products; A, B: means followed by diferent uppercase letters in the row difer the incubation times; Signifcant at the 5% probability level by Tukey's test.

Conclusions
Among the studied by-products, babassu cake presents superior chemical composition.However, the 95 µm four simultaneously exhibited satisfactory nutritional value and ruminal degradation, making it suitable for use as an additive or in partial substitution of other traditional concentrates, depending on availability and afordable prices.

Table 1 :
Chemical composition of babassu by-products.in % natural matter; DM � dry matter; SEM � standard error of the mean.Means followed by the same letters in the row do not difer signifcantly from each other according to Tukey's test at a 5% probability level.a,b,c,d Means followed by the same letters in the row do not difer signifcantly from each other according to Tukey's test at a 5% probability level. *

Table 2 :
In situ degradability of dry matter and crude protein of babassu by-products.