Introduction

The sustainability of animal diets is crucial in the development of livestock production systems, and feed efficiency can be improved by reusing food waste (i.e. bakery by-products, residues of the sugar industry, pineapple and citrus by-products) in ruminant diets, thus diminishing the use of food grains (Makkar and Ankers 2014).

The use of excretes from livestock is an alternative method of supplying non-protein nitrogen (NPN) in the feed of ruminants (Ortiz et al. 2007; Nasiru et al. 2014); however, they usually have low energy content for optimal rumen bacteria growth. The use of by-products from the baking and sugar industries provides an attractive energy source for ruminants. Increased cellulolytic activity of microorganisms in the rumen can increase the digestible energy due to better utilization of fibrous feeds, and otherwise improve the supply of microbial protein (Chandrasekharaiah et al. 2012). Several studies (Bórquez et al. 2009; Trujillo et al. 2014) have proved that the inclusion of up to 50 % silage manure as dry matter does not affect the intake and metabolic response in lambs, thus presenting an option as a sustainable resource in ruminant feeding.

The aim of this study was to evaluate the effects of swine manure (SM), poultry waste (PW) and urea (U) as nitrogen sources, with the inclusion of molasses (M) or bakery by-product (BB) as carbohydrate sources, on the chemical composition of silages and their intake and digestibility in lambs.

Materials and methods

Experimental silages and chemical analysis

Four silages were prepared using nitrogen (SM, PW and U) and energy sources (M and BB), and mixed with corn stover in different proportions as a fiber source (Table 1). The silages were prepared with PW combined with BB (PWBB), SM with BB (SMBB), SM with M (SMM) and U with M (UM). Each combination was ensiled with a bacterial additive (Sill-All 4 × 4 Alltech®, 10 mg/kg DM; Streptococcus faecium, Lactobacillus plantarum, Pediococcus acidilactici and Lactobacillus salivarius and enzymes cellulase, hemicellulase, pentosanase and amylase). The mixing process was performed by adding water (480 ml/kg fresh matter) to SM, PW and U, followed by M or BB. Once diluted, these combinations were mixed with corn stover in different proportions in plastic bags with a capacity of 50 kg, in three replications, compacted and sealed to prevent the ingress of air; for details see Mejía-Uribe et al. (2013).

Table 1 Proportion of the ingredients and chemical composition (g/kg DM) of the experimental silages
Full size table

After 60 days, the silage bags were opened (Serrano-García et al. 2008) and samples of 500 g were taken from each bag; pH was determined with a pH meter (Conductronic pH 130), samples were dried in a forced air oven (60 °C, 48 h) and grounded in a Willey mill (2 mm diameter). Silage samples were analyzed for dry matter (DM, #934.01), ash (#942.05) and N (#954.01) according to AOAC (1997). Neutral and Acid detergent fiber (NDF and ADF; Van Soest et al. 1991) were analyzed using an ANKOM200 Fiber Analyzer Unit (ANKOM Technology Corporation, Macedon, NY, USA), and lignin (AOAC 1997; #973.18). NDF was assayed with alpha amylase and sodium sulfite in the NDF. Both NDF and ADF are expressed without residual ash. Moisture content of the silages was determined through distillation with toluene (Haigh and Hopkins 1977).

Animals and diets

Four Hampshire lambs with live weight (LW) of 30 ± 3.0 kg and <1 year old, provided with ruminal cannulas, were fed with the four experimental silages using a 4 × 4 Latin square design. The animals were placed in metabolic cages. The diet consisted of the inclusion of silages (Table 1) and concentrate supplement (Table 2) in order to meet growth requirements (NRC 2007). At the beginning of the experiment, animals were dewormed (IVOMEC®; Ivermectin 1 ml 50 kg LW), supplemented with ADE complex (1 ml/head IM) and vaccinated (Bobact 8®, 2.5 ml/animal). The experimental diets were formulated to contain approximately 140 g/kg CP and 10.25 MJ ME/kg DM on average; treatments were administered ad libitum twice a day at 08.00 and 16.00 h. Each experimental period lasted 21 days, allowing 14 days for adaptation to the diet and 7 days for sample collection. Feed and ort samples were collected on days 14 to 21, weighed and composited daily, both for each individual sheep and across days. Total fecal matter and urine were weighed and sub-sampled (10 % of wet weight) for each lamb and period and stored at −20 °C for laboratory analysis.

Table 2 Proportion of the ingredients used and chemical composition (g/kg DM) of the diets for growing lambs, with the inclusion of silages of poultry waste (PW), swine manure (SM) or urea (U) with the inclusion of bakery by-product (BB) or molasses (M)
Full size table

On day 21, ruminal fluid samples (250 ml) were drawn using a suction strainer, obtained via ruminal cannula at 0 (previous ingestion), 3, 6, 9 and 12 h after feeding. Samples were filtered through a double layer of cheesecloth gauze, and the pH was recorded (Conductronic pH 130).

Statistical analysis

Chemical composition data were processed as a complete randomized design (Steel et al. 1997), using the following model:

$$Y_{ij} = \mu + T_{i} \, + \varepsilon_{ij} ,$$
(1)

where Y ij is the response variable, μ is the general mean, T i is the effect due to diet and ε ij is the random experimental error.

The in vivo experiment data were analyzed according to a 4 × 4 Latin square design, following the model:

$$Y_{ijk} \, = \mu \, + T_{i} \, + A_{j} \, + P_{k} + \varepsilon_{ijk}$$
(2)

where Y ij is the response variable, μ is the general mean, T i is the effect due to diet, A j is the animal effect, P k is the effect due to experimental period, and ε ijk is the random experimental error. The GLM procedure of SAS (2002) was used. The means of treatments were compared by Tukey’s test (Steel et al. 1997) where the effect was significant (P ≤ 0.05).

Results

Chemical composition of silages

Silage pH was higher for UM (P < 0.001) compared with the rest of the treatments (Table 2); the OM content was higher (P < 0.001) for SMBB and UM, followed by PWBB compared with SMM. The CP concentration was higher (P < 0.001) for UM than SMBB and SMM. There were no differences (P > 0.05) for ADF and lignin among silages. The NDF content was higher (P < 0.03) for PWBB and SMBB than UM. The inclusion of experimental silages (Table 2) in the diets ranged from 360 to 406 g/kg DM, and the CP of the diets varied from 130 to 141 g/kg, being lower for SMBB and SMM. The NDF and ADF content in the diets were lower in PWBB and SMM compared with SMBB and UM diets.

Nutrient intake and digestibility

Table 3 shows the pH, intake, digestibility and nitrogen balance in growing lambs fed manure silage. By far the most effective source of nitrogen and minerals (ash) was provided by the livestock manure (swine and poultry), urea as we know, is only a source of nitrogen. Ruminal pH values in lambs were similar (P > 0.05). The DM, OM and ADF intakes were higher (P < 0.05) for silage based on UM compared with the rest of the treatments. The NDF intake was higher (P < 0.05) for UM, followed by PWBB and SMM, and lower for SMBB. Digestibility of DM, OM and NDF was similar among treatments (P > 0.05). The ADF digestibility was lower (P < 0.05) for SMM compared with UM treatment. Nitrogen intake (g N/day) was higher (P < 0.05) for UM compared with the rest of the treatments. Nitrogen excretion (feces and urine) and retention were similar among treatments (P > 0.05).

Table 3 Intake (g/kg LW0.75), digestibility (g/kg) and N balance (g/day) in lambs fed with silages of poultry waste (PW), swine manure (SM) or urea (U) with the inclusion of bakery by-product (BB) or molasses (M)
Full size table

Discussion

Chemical composition of silages

All the silages had acceptable quality (Frenkel 1984), except UM, which showed higher pH values (Table 1). This might be due to the fact that BBP and M were a better carbohydrate sources to be fermented into desirable organic acids such as lactic, acetic and propionic acids, diminishing the pH. The low pH in silages in our study suggests that undesirable fecal microorganisms (Coliforms, Salmonella, Shigella, Proteus), yeasts and molds may had been eliminated (Pagán et al. 2014; Serrano-García et al. 2008), which have been reported for cattle manure 1 week after ensiling (Cornman et al. 1981; McCaskey and Wang 1983). Optimum moisture recommended for manure-blended silage is at least 600 g/kg (McCaskey and Wang 1983), whereas in our study moisture of silages ranged from 524 to 685 g/kg. Bórquez et al. (2009) using cattle manure silage with BB, the amount of DM was lower with respect to which was added M. The amount of water varies among the different ingredients added in the silages, thus varying among silages. The CP and NDF content can vary depending on the type and amount of bedding floor used by the livestock species (Tobia and Vargas 2000). Evans and Smith (1986) reported that the use of U leads to changes in cell wall components of forages treated, destroying the linkages of phenolic groups between hemicellulose and lignin, which solubilizes the hemicellulose and make it available to the cell wall unless UM silage includes more corn stover than the rest of the silages. Tobia and Vargas (2000) found a similar protein content in PW and NDF, but higher for ADF. Mthiyane et al. (2001) showed a lower content of CP and OM compared with the present study, but higher in NDF for PW. These variations in nutrient content depend on the type of food, floor and management of excreta that is offered to poultry and pigs (Tobia and Vargas 2000; Morales et al. 2002; Teixeira et al. 2015), effecting a wide variation in the chemical composition of PW and SM.

Nutrient intake and digestibility

The inclusion level of silage in the diet had no effect on ruminal pH in the lambs (Table 3), which could be due to silage inclusion which ranged around 60 % as fresh matter and there was a sufficient amount of NDF in the entire diet (ranging 417–471 g/kg DM). The intake of DM, OM, NDF and ADF was similar to Trujillo et al. (2014). The digestibility (g/kg) of DM and OM based on PW silage was lower than Morales and Egaña (1997), but similar for NDF digestibility; these variations depend on the chemical composition of the different poultry waste sources. In the present study, there were no differences (P > 0.05) among DM, OM and NDF digestibility, which was similar to Obeidat et al. (2011) and Trujillo et al. (2014), but ADF digestibility was higher (P < 0.05) for UM compared with SMM. This effect could be related with the lower ADF intake and higher ash content in the SMM diet, as was found by Jakhmola et al. (1988) and Iñiguez-Covarrubias et al. (1990).

The lower N intake in the silage diets compared with the UM diet is a direct response to the higher DM intake in the UM diet compared with the rest of the silages; the N retained in lambs fed diets with the inclusion of PW and SM silage was numerically lower (P = 0.34) than UM, which provided more efficient N retention in the animals. The inclusion level of silages in the diets of up to 37 % neither enhanced nor adversely affected animal performance as compared with UM, which are in agreement with the results of Zia-ul-Hassan et al. (2011) and Sarwar et al. (2011) in feeding traits with different levels of cattle manure in lactating Nili–Ravi buffaloes and growing cattle calves, respectively; and with Trujillo et al. (2014) who studied different levels of SM or PW silages in growing lambs.

Conclusion

UM silage showed higher pH and DM, OM, CP content, but lower NDF. Treatment with these silage also showed higher DM, OM, NDF and ADF intake. Nitrogen intake was higher with la inclusion of UM silage, but there was no difference in N retention. It is concluded that UM silage inclusion in complete diets for growing lambs can be as good as the use of livestock manures silages.