Materials and Methods

The study was conducted in the Forage Production Sector of the Department of Animal Science of the Federal University of Paraíba (UFPB) in Areia – Paraíba, Brazil. Laboratory analyses were performed in the forage production and animal nutrition laboratories of the same institution.

The cochineal nopal cactus (Nopalea cochenillifera Salm Dyck), commonly known in Brazil as palma Miúda, was used. The cultivated cactus had a regrowth age of two years and was obtained from a crop that was previously established crop on the experimental farm of the Empresa Estadual de Pesquisa Agropecuária da Paraíba (Paraiba's Enterprise for Agricultural Research, EMEPA), located in the Agreste of Paraíba mesoregion, Curimataú microregion, municipality of Soledade, at coordinates 7° 8′ 18″ S and 36° 27′ 2″ W.Gr. at an altitude of 534 m. Based on the Köppen classification, the climatic type of the region is Bsh (hot semiarid) with rainfall from January to April, an average annual temperature of 24°C, average relative humidity of ~68%, and average annual rainfall of ~400 mm; drought conditions persist during almost the entire year (Oliveira Jr. et al., Reference Oliveira Junior, Barreiro Neto, Ramos, Leite, Brito and Nascimento2009).

The cochineal nopal cactus was harvested manually, preserving the primary cladodes. Later, the cladodes were processed in a stationary forage machine (MC1001N Laboremus®), obtaining particles of ~2 cm. Additives were mixed with the cochineal nopal cactus as follows: Control: no additive; urea: addition of urea to 2% of the DM weight; LB: addition of L. buchneri CNCM I – 4323 (1.0 × 10¹¹ cfu/g); this was applied as a commercial inoculant (Lalsil® AS, Lallemand) that was developed to increase the aerobic stability of silage, and it was applied as recommended by the manufacturer; and U + LB: addition of urea and L. buchneri in the above-mentioned proportions.

Two experiments were performed in this study. The first experiment evaluated the silages at various storage times, and the second experiment evaluated the properties of the silages during the aerobic stability test.

First experiment: The silages were prepared in 48 experimental PVC (polyvinyl polychloride) silos, 10 cm in diameter and 30 cm in height, to achieve a final packing density of 600 ± 20 kg of fresh matter/m³. After addition of the silage material, the silos were closed and stored at room temperature in a covered, dry and ventilated place until being opened at 7, 15, 60 and 120 days after ensiling. At each opening time, the microbial populations, the fermentation profile, the fermentation losses and the chemical composition of the silage were determined.

Second experiment: For the aerobic stability test, the silages were produced using the same compaction density mentioned above in 12 PVC silos (15 cm diameter and 30 cm height) fitted with Bunsen valves for release of the gases resulting from fermentation. At the bottom of each silo, 1 kg of dry sand was added and covered with a TNT screen (nonwoven fabric) to collect the effluents. At the end of this process, the silos were closed, weighed and stored at room temperature in a covered, dry and ventilated place until they were opened 120 days after ensiling.

The aerobic stability test lasted 96 h; the microbial populations, fermentation profile, chemical composition and recovery of DM from the silage were determined after 0, 48 and 96 h of air exposure. At each evaluation time in both experiments, the pH of the silage was determined according to the methodology described by Bolsen et al. (Reference Bolsen, Lin, Brent, Feyerherm, Urban and Aimutis1992), and samples (~400 g) of the silage were collected for subsequent evaluations.

The microbial populations were quantified before ensiling and in the silages using selective culture media for each microbial group. MRS (de Man, Rogosa and Sharpe) agar containing 0.4% nystatin was used for lactic acid bacteria, Violet Red Bile agar was used for enterobacteria and potato dextrose agar containing 1% of a 10% tartaric acid solution was used for moulds and yeasts.

The quantification of the microbial groups was performed using 10 g of a sample obtained from the replicates of each treatment. To each sample, 90 ml of sterile distilled water was added to obtain a 10⁻¹ dilution, and the sample was homogenized for 1 min. Serial dilutions ranging from 10⁻¹ to 10⁻⁹ were then prepared, and the culture was performed in sterile disposable Petri dishes. The Petri dishes were incubated at specific incubation temperatures for each microbial group (Ávila et al., Reference Ávila, Carvalho, Pinto, Duarte and Schwan2014; Santos et al., Reference Santos, Pereira, Garcia, Ferreira, Oliveira and Silva2014). For lactic acid bacteria, the dishes were incubated at 37°C for 48 h; for enterobacteria, they were incubated at 30°C for 24 h, and for moulds and yeasts, they were incubated at 28°C for 72 h. Petri dishes containing 30–300 colony-forming units (cfu) were considered appropriate for counting.

Organic acids (lactic, acetic, propionic and butyric acids) and ethanol were determined using the methodology described by Siegfried et al. (Reference Siegfried, Ruckemann and Stumpf1984) in a Shimadzu high-performance liquid chromatography system that included a model SPD-10ª VP detector coupled to an ultraviolet detector, at a wavelength of 210 nm. For the evaluation of ammoniacal nitrogen and the buffering capacity (BC) of the silages, the methods described by Bolsen et al. (Reference Bolsen, Lin, Brent, Feyerherm, Urban and Aimutis1992) and Playne and McDonald (Reference Playne and Mcdonald1966), respectively, were used.

Approximately 300 g of silage samples were pre-dried in a forced air oven (60°C), processed in a Willey mill with a 1 mm sieve and analysed according to the AOAC protocols (1990) for the determination of DM (ID 930.15) and crude protein (CP) (ID 954.01) content. The WSC content of the samples was determined using the concentrated sulphuric acid method described by Dubois et al. (Reference Dubois, Gilles, Hamilton, Rebers and Smitch1956) as modified by Corsato et al. (Reference Corsato, Scarpare Filho and Sales2008). The methodologies described by Van Soest et al. (Reference Van Soest, Robertson and Lewis1991) were used to determine the level of neutral detergent fibre (NDF), not analysed with heat-stable amylase and sodium sulphite (NDFom) and not expressed exclusive of residual ash.

The chemical composition of cochineal nopal cactus before ensiling was 159.9 g/kg of DM, 21.6 g/kg DM of CP, 226.4 g/kg DM of NDF not analysed with heat-stable amylase and sodium sulphite and 173.5 g/kg DM of WSC. The DM losses in the silages in the form of gas and effluent were quantified by weight difference using the equations described by Zanine et al. (Reference Zanine, Santos, Dórea, Dantas, Silva and Pereira2010):

$$G = \lpar {{\rm WFc}-{\rm WFo}} \rpar /\lpar {{\rm FMc\ } \times {\rm DMc}} \rpar \times 10000$$

where G is the gas losses (% of DM), WFc is the weight of the filled silo at closing (kg), WFo is the weight of filled silo at opening (kg), FMc is the forage mass at silo closing (kg) and DMc is the forage DM concentration at silo closing (%).

$$E = \lsqb {\lpar {{\rm WEf}-{\rm Tb}} \rpar -\lpar {{\rm WEi\ }-{\rm Tb}} \rpar } \rsqb /{\rm FMi} \times 100$$

where E is the effluent losses (kg/t of fresh matter), WEi is the weight of the empty silo + sand at closing (kg), WEf is the weight of the empty silo + sand at opening (kg), Tb is the weight of the empty silo (kg) and FMi is the forage mass at silo closing (kg).

The DM recovery was estimated based on the difference in feed weight and DM concentration before and after ensiling using the equation described by Zanine et al. (Reference Zanine, Santos, Dórea, Dantas, Silva and Pereira2010):

$${\rm DMR} = \lpar {{\rm FMo} \times {\rm DMo}} \rpar /\lpar {{\rm FMc} \times {\rm DMc}} \rpar \times 100$$

where DMR is the DM recovery rate (%), FMo is the forage mass at silo opening (kg), DMo is the forage DM concentration at silo opening (%), FMc is the forage mass at silo closing (kg) and DMc is the forage DM concentration at silo closing (%).

After 120 days of ensiling, the aerobic stability of the silages (expressed in hours) was evaluated by monitoring the surface and internal temperatures of the silages air exposure. The silage samples were placed without compaction in unlidded PVC experimental silos and maintained in a controlled environment (25°C). Temperatures were recorded hourly using digital laser and digital immersion thermometers positioned at the centre of the silage mass. The beginning of deterioration was defined as the point at which the internal temperature of the silage reached 2°C above room temperature (Kung Jr. et al., Reference Kung, Robinson, Ranjit, Chen, Golt and Pesek2000).

For both experiments, the experimental design was completely randomized in a factorial design. First experiment adopted a 4 × 4 (four additives and four opening times) design with three replicates per treatment. For the variables associated with aerobic stability (second experiment), a 4 × 3 factorial design (four additives and three air exposure times) with three replicates per treatment was used. The data were statistically analysed by analysis of variance, and when significant, the means were compared using Tukey's test at a significance level of 5% using SISVAR software (Ferreira, Reference Ferreira2008).

The statistical model used was as follows:

$$Y_{ijk} = \mu + \alpha _i + \beta _j + \gamma _{ij} + \varepsilon _{ijk}$$

where μ is the population mean, α_i is the treatment effect, β_j is the time effect, γ_ij is the treatment × time interaction effect and ε_ijk is the residual error.

Data regarding the temperatures in the aerobic stability test and the quantification of the microbial groups (in logarithmic units, log₁₀) were analysed in a descriptive manner.