Does macauba cake alter nutrient digestibility and microorganism population in the rumen of sheep?

Twenty Santa Inês male sheep were used in a randomized block design to evaluate the effects of different levels of macauba cake (MC) on nutrient digestibility and the population of microorganism in the rumen. The animals were divided into four groups according to MC levels (0, 10, 20, and 30% of DM) and initial body weight ranging from 32.75 to 52.17 kg. Diets were isonitrogenous and formulated to attend metabolizable energy requirements, and feed intake was regulated with 10% allowance for leftovers. Each experimental period lasted 20 days, with the final 5 days reserved for sample collection. Macauba cake inclusion did not affect the dry matter, organic matter, or crude protein intake but increased the ether extract, neutral detergent fiber, and acid detergent fiber intake, mainly because of changes in the concentrations of these components in diets with a higher level of MC. With MC inclusion, a linear decreasing effect was observed for the dry matter and organic matter digestibility, and a quadratic effect with a maximum point of 21.5% was observed for the acid detergent fiber digestibility. A relative reduction of 73% in anaerobic fungal populations was observed with the lowest level of MC inclusion, and a relative increase of 162% in methanogenic populations was observed with the highest level of MC inclusion. The increasing level of macauba cake up to 30% of the diet of lambs reduced the dry matter digestibility and reduced anaerobic fungi but increased methanogenic population.


Introduction
The use of agro-industrial residues in ruminant feed has increased due to the greater availability of these byproducts (Santos et al. 2015), the possibility for reducing feeding costs by replacing traditional feedstuffs (Oliveira et al. 2012), and the reduced environmental effects because most of the byproducts are environmental waste management problems.
The macauba (Acrocomia aculeata (Jacq.) Lodd. Ex Mart.) is a highly productive oil seed palm, which is native to South America. It has attracted interest as an oil source, with potential for the production of biodiesel, cosmetics, and food. The oil from the fruit pulp has high levels of palmitic (C16:0), oleic (C18:1), and linoleic (C18:2) acids, which account for approximately 90% of total fatty acids (Coimbra and Jorge 2012).
During the process of oil extraction using a hydraulic press, large quantities of biomass residue are generated, which represent 50% of the total quantity of the processed fruit. This residual biomass, called macauba cake (MC), contains low levels of protein (7.41 ± 0.75) and high levels of fiber (67.03 ± 4.85) and ether extracts (17.34 ± 2.30) (Azevedo et al. 2013;Rufino et al. 2011;Santos et al. 2017). Therefore, MC could be considered an alternative ruminant feed. However, a high lipid level in the diet of ruminants (more than 5% of fatty acids in the dietary DM) can inhibit rumen fermentation, Gram-positive bacteria, Archaea, and protozoa (Palmquist and Jenkins 1980).
A recent study demonstrated that 100 g kg −1 of MC in diet reduced ruminal protozoa concentrations in dairy cows (Santos, Magalhães, Azevedo, Vieira, França, Geraseev and Duarte 2015) and Santa Ines lambs (Santos, Azevedo, Júnior, Rodriguez, Duarte and Geraseev 2017). This alteration in ruminal protozoa suggests that other ruminal populations, such as cellulolytic bacteria, could also be affected, which would change the bioavailability of nutrients or modulate rumen fermentation. The high lipid and fiber contents in MC may decrease dry matter intake and affect digestibility and performance due to an interference with fiber digestion. In this study, the aims were to evaluate the digestibility of nutrients and the changes in some ruminal microorganism populations important to fiber digestion in sheep fed diets containing different amounts of macauba cake.

Materials and methods
Twenty castrated male Santa Ines lambs were used. Initially, the animals were individually identified, treated for ecto-and endo-parasites, and immunized for clostridiosis. The sheep were housed in metabolic cages (1.10 × 1.10 m), which had individual feeders, waterers, and collectors of feces and urine.
The study used a randomized block design to separate the animals into four groups, each group having a different level of MC in their diet (0%, 10%, 20%, and 30% of DM). The blocks were determined by the initial body weight of the animals. In the first block, the average live weight of animals was 52.17 ± 4.8 kg, in the second block 48.17 ± 1.25 kg, in the third block 45.62 ± 2.1 kg, in the fourth block 37.65 ± 6.1 kg, and in the fifth block 32.75 ± 6.3 kg. The diets were prepared according to the recommendations of the (NRC, 2007) for metabolizable energy requirements, and consumption was regulated to maintain 10% of leftovers (Table 1). The diet was provided twice a day at 07:00 and 15:00. For each block, the experimental period consisted of 15 days of adaptation and 5 days of sample collection. The experimental collection period included the measurements of the total amount of leftovers and daily production of feces per animal. Representative samples of the leftovers, feces, and urine were collected daily. To avoid the loss of nitrogenous compounds by volatilization, 100 mL of 10% sulfuric acid (H 2 SO 4 ) was added to the urine collectors. The samples were frozen at − 20 ℃ for subsequent laboratory analyses.
The DM, organic matter (OM), CP, EE, NDF, and ADF intakes were calculated by the difference in the daily weight offered and leftover by lambs. The in vivo digestibility of each chemical component (DM, OM, CP, EE, NDF, ADF) was calculated for each animal using the average individual nutrient intake and fecal output.
Four hours after the feed on the last day of the collection period, the rumen fluid was extracted from each animal using an esophageal probe coupled to a vacuum pump. A single sample per animal was taken. Rumen fluid samples were stored in a freezer at − 20 ℃ for subsequent analysis.
Before DNA extraction, the sample was well homogenized with a vortex. DNA extraction was performed according to the method described by Makkar (2005) and was carried out in duplicate. The DNA concentration and purity of the sample (260 nm/280 nm) were determined using a Nanodrop 1000 spectrophotometer at absorbances of 260 and 280 nm. The extracted DNA was used as a template for the quantitative PCR assay using primer pairs for rumen bacteria, rumen fungi, Ruminococcus flavefaciens, Fibrobacter succinogenes, and rumen methanogens, as described by Denman and McSweeney (2006) and Denman et al. (2007). The DNA samples were diluted to a final concentration of 10 ng/μL. The primers were used at a concentration of 10 mM along with SYBR Green. The final volume of the reaction was 25 µL. DNA amplification was performed on an Applied Biosystems 7500 thermal cycler using the following program: 1 cycle of 95 ℃ for 5 min; 40 cycles of 95 ℃ for 10 s and 60 ℃ for 30 s; and one cycle of 95 ℃ for 2 min, 60 ℃ for 15 s, and 95 ℃ for 15 s. The microbial population levels were expressed relative to the total bacterial population (ΔCt). These ΔCt values were calculated by the difference between the cycle threshold (Ct) values of the target and reference genes (16S rRNA of the bacteria). The ΔΔCts were determined by the difference between the ΔΔCts of the target groups in the experimental and control diets. The percentages of rumen fungi, Ruminococcus flavefaciens, Fibrobacter succinogenes, and rumen methanogens relative to the total bacterial population were calculated from the ΔCt values as 100 × (2 ∆Ct ) −1 (Denman and McSweeney 2006) and the expression of the target groups relative to the control treatment as 2 −Δ∆Ct .
Nutrient intake and digestibility were analyzed using PROC REG, part of the statistical software SAS version 9.4 (SAS Institute Inc). For microorganisms, the variance analyses were performed using the AOV function and Duncan's test in the ExpDes package. In the multivariate analysis, the original data were standardized, and the principal components were obtained using the princomp function of R software.

Results
Including MC in the diet of the sheep did not change the intake of DM, OM, or CP (Table 2). However, MC inclusion increased the intake of EE, NDF, and ADF (P < 0.01), possibly because of changes in the concentrations of these components in diets with elevated MC levels ( Table 1). Inclusion of MC reduced the in vivo digestibility of DM and OM but increased the digestibility of ADF (P < 0.01), with a peak at the 21.5% MC inclusion level (Table 2). However, no effect was observed in CPD, EED, and NDFD.
Principal component analysis (Fig. 1) revealed that the first two components collectively accounted for 90.93% of the variation. In corroboration to our findings, it showed that without MC inclusion in the diet, the F. succinogenes, R. flavefaciens, and anaerobic fungal populations, as well as the DM and CP digestibility, tended to increase. In addition, at the 30% MC inclusion level, the methanogenic Archaea population, NDF digestibility, ADF digestibility, and EE digestibility showed a tendency to increase.

Discussion
In this study, the NDF content increased with MC inclusion; however, this was not a limiting factor for DM consumption (Table 2). This could be due to the physical processing of the MC, which was ground to a granulometry similar to that of the other foods used in the concentrate, which could reduce the physical effectiveness of the fibrous fraction. Dantas Filho et al. (2007) evaluated the performance of Santa Ines sheep fed diets with different inclusion levels of dehydrated cashew pulp, with NDF levels similar to those in the present study. These authors did not observe differences in the DM intake, although the inclusion of the byproduct increased fiber levels, according to the authors, this fact can be attributed, in part, to the processing of the byproduct, which was finely ground, resulting in a faster rate of passage of the food through the rumen.
In our study, there was a decrease in DM and OM digestibility with MC inclusion probably due to the increase in the fibrous fraction, which is common when there is a substitution of conventional for alternative ingredients. Diets composed of high levels of fatty acids, mainly unsaturated, may inhibit the digestion of fiber in the rumen because of their  This analysis was performed with the chemical composition of the diet, digestibility of the experimental diets, and its components: 0, control treatment; 10, inclusion of 10% of macauba cake; 20, inclusion of 20% of macauba cake; 30, inclusion of 30% of macauba cake; DM, dry matter digestibility; NDF, neutral detergent fiber digestibility; ADF, acid detergent fiber digestibility; CP, crude protein digestibility; EE, ethereal extract digestibility cytotoxicity for microorganisms (Allen 2000). In this study, the increase in the EE level in the diets with MC inclusion did not compromise fiber digestion since, and no difference was observed in the NDF digestibility, while ADF digestibility was increased. A maximum point of 21.5% was observed for the acid detergent fiber digestibility; this was unexpected, as it has been reported in the literature (Santos, Azevedo, Júnior, Rodriguez, Duarte and Geraseev 2017) that MC is substantially composed of unsaturated fatty acids. The percentage of F. succinogenes and R. flavefaciens relative to the total bacterial population was not influenced by the inclusion of MC (Table 3). A number of studies have shown that an increase in the EE concentration in the diet reduces the populations of cellulolytic bacteria (Potu et al. 2011), F. succinogenes (Yang et al. 2009), and R. flavefaciens (Wang et al. 2018). Although there was an increase in the dietary EE with MC inclusion, there was also an increase in the NDF content of the diet, which may have contributed to the maintenance of these fibrolytic bacteria. It is important to note that, for all treatments, F. succinogenes comprised a greater percentage of the total bacterial population than R. flavefaciens. This was unsurprising as F. succinogenes has been described as one of the major cellulolytic bacterial species present in the rumen (Forsberg et al. 1997).
The increase in Archaea (Table 3) occurred due to a reduction in the nonfibrous carbohydrate content and an increase in the fibrous fraction in the diet with MC inclusion. Martin et al. (2010) reported a decrease in the ruminal production of CH 4 with diets containing more soluble sugars because of a shift in the VFA production from acetate to propionate. Ruminal dissolved hydrogen positively correlates with total VFA and molar proportions of propionate and negatively correlates with the molar proportion of acetate and gaseous methane emissions (Wang et al. 2016). In this study, the NDF intake increased with MC inclusion, which may have contributed to the reduction in the ruminal dissolved hydrogen concentration and the increase in methanogenic Archaea.
In this study, we observed a significant reduction in the anaerobic fungal population with MC inclusion (Table 3). The high lipid content potentially reduced the fiber adhesion of zoosporangia and mycelia structures, resulting in the reduced growth of these important fibrolytic microorganisms. Other studies have also reported a reduction in anaerobic fungi with diets containing high levels of lipids. Adult Merino wethers were offered diets containing 15% unprotected and protected cottonseed kernel supplements. Rumen anaerobic fungi were undetectable with 5.9% free lipids but were present at normal levels with 4 and 2.1% free lipids. Concomitant with the absence of anaerobic fungi, the numbers of protozoa tended to be lower, while the bacterial counts increased (Faichney et al. 2002). Veneman et al. (2015) also reported a reduction in anaerobic fungi with the inclusion of linseed oil in the diet of dairy cows.
In a previous study, we evaluated the rumen protozoa profile of lambs fed diets with MC. The concentrations of small, medium, and large protozoa, as well as the total, were maximal at an inclusion of 10% in the diet. Above this level, theses ciliate populations substantially decreased along with a reduction in the genus diversity. The fatty acid composition of the MC (per kg fatty acids) was 15.3 g of C12:0, 14.7 g of C14:0, 178.0 g of C16:0, 40.6 g of C16:1, 18.0 g of C18:0, 550.3 g of C18:1, 150.0 g of C18:2, 17.6 g of C18:3, and 15.5 g of other acids. A ratio of 1.0/3.4 saturated/unsaturated fatty acids was associated with antiprotozoal effects (Santos, Azevedo, Júnior, Rodriguez, Duarte and Geraseev 2017). The mix of fatty acids might also promote the antifungal effect observed in the present study, which requires elucidation in future studies.
Agricultural residues, like MC, are lower cost alternatives to be used in ruminant diets. Our study revealed that increasing level of this residue (up to 30% of total DM) reduced dry matter digestibility, promoted an antifungal effect, and increased the methanogenic population. This could be explained by the increase of NDF and EE contents of those diets; however, further studies are necessary to fully understand how microbial colonization occurs and its effect on animal performance.
Funding This project was financed by the National Council for Scientific and Technological Development -CNPq (475368/2012-6), Coordination of Superior Level Staff Improvement -CAPES (code 001), and the Pro-Reitoria de Pesquisa of Federal University of Minas Gerais (PRPq/ UFMG).

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval
The animal experiments were conducted in accordance with the ethical principles of animal experimentation, approved in pro-