Study area

The study was carried out at the Faculty of Veterinary Medicine and Animal Husbandry of the University of Juárez, Durango State. Forage nopal variety AV6 was randomly harvested from a cultivated nopal plantation, located next to the Faculty and within Durango municipality, Mexico.

Solid state fermentation (SSF) and preparation of microsilos

The nopal cactus stalks were cut into pieces of approximately 1 cm² using a stainless steel knife and placed in 19 l plastic containers, where they were inoculated with Sacharomyces cerevisiae (1% m/m). The fermentation process was carried out for 48 h at 25 ºC. Treatments consisted of including cactus and fermented cactus to corn fodder, as shown in Table 1. Experimental microsilos were prepared with chopped corn forage without grain, and in mature state with a particle size of 2 to 4 cm (Variety: Hybrid 21/20) (T1, n=7), corn stover with fresh nopal (T2, n=7). Besides prepared with corn stubble with fermented nopal (T3, n=7) in plastic containers (30 cm diameter × 50 cm high), hermetically sealed for 30 d. After this time, the microsilos were opened for later analysis.

Fermentative variables

Once the microsilos were opened, the following variables were evaluated: pH (Hanna instruments, model HI 83142); lactic acid according to ^{Borshchevskaya et al. (2016)}; as well as volatile fatty acid and ammonia nitrogen (NH3-N) contents, using the procedures proposed by ^{Galyean (2010)}.

Chemical analysis

Samples from each experimental microsilos were dried in a forced air oven at 55 °C for 72 h; subsequently, the particle size was reduced to 1 mm in a Wileymil mill (Arthur H Thomas, Philadelphia, PA, USA), to determine the dry matter (DM) contents (method 934.01). Crude protein (CP) concentrations were determined by the micro-Kjeldhal technique (method 920.87), using the conversion factor (6.25) (^{AOAC, 2010}). The concentration of NDF, ADF were obtained according to the procedures proposed by ^{Van Soest (1991)} and the gas production parameters according to the technique described by ^{Menke and Steingass (1988)}.

Gas in vitro production

Approximately 1 g of sample from each experimental microsilo was placed in glass modules in a pressure transducer equipment, ANKOM brand. In addition, it was incubated in triplicate with a 2:1 solution of ruminal buffer-liquid solution, according to the procedure described by ^{Murillo-Ortiz et al. (2018)}. Incubations were carried out from 0 to 96 h and recording pressure values at the same time. The gas production kinetics was estimated using the Gompertz function (^{Murillo-Ortiz et al., 2018}), according to the following equation:

GP = Gmax ∗ exp[−A ∗ exp(−k ∗ t)]

Where GP= gas production at time t (ml), Gmax= maximum gas production (ml), k= constant gas production rate (h^-1) and A= lag phase (h). From the 24 h incubation, the valve was opened to release gas for 2 s from each module. The gas released from each module was connected to a portable CH₄ and CO₂ analyzer through a tube to measure the concentration of these gases according to the procedures established by the manufacturer (GEM^TM5000, LANDTEC, USA).

In vitro fermentation parameters

To evaluate fermentation parameters, 1 g of sample was placed in nylon bags (ANKOM, F500 nylon bags; ^{ANKOM, 2018}), pre-weighed and placed inside ANKOM modules and incubated in triplicate with buffer solution: ruminal fluid, in a 2:1 ratio according to ^{Murillo- Ortiz et al. (2018)}). After 24 h of continuous fermentation, the modules were opened and immediately pH was measured (Hanna instruments, model HI 83142). The bags were removed from the modules and were washed with distilled water and dried at 65°C for 48 h. In vitro digestibility of dry matter (IVDDM) was calculated based on the difference in dry matter content of the substrate before and after incubation. Additionally and approximately 1.0 ml of the filtrate was centrifuged at 3,000×g for 5 min; then, 500 μl of the supernatant liquid was acidified with 150 μl of 25 % metaphosphoric acid to evaluate volatile fatty acids. Also approximately 1.0 ml of the filtrate was placed in tubes and was acidified with 30 μl of 50 % v/v sulfuric acid to determine N-NH₃ (^{Galyean, 2010}).

Statistical analysis

The data obtained were analyzed with a completely randomized design, using the GLM procedures of ^{SAS (2010)}. Means were compared with Tukey's multiple range test and declared significant differences when P≤0.05.

Chemical composition.

Dry matter (DM) content decreased with the inclusion of corn stubble (P<0.05, Table 2). Silages containing stubble and nopal (t2 and t3) presented adequate DM contents, since according to ^{Pineda-Cordero (2016}) silages containing between 30 and 35 % DM are considered of good quality. In addition, the moisture present in silages determines the type of fermentation that took place during the ensiling process. Other factors that determine fermentation quality are the soluble carbohydrates present and the buffering capacity of the forage used (^{Bernal et al., 2002}).

Crude protein (CP) concentration was different among treatments (P<0.05, Table 2). The addition of cactus and fermented cactus increased 43.54 % and 79 % the CP concentration in silages, respectively, compared to t1. The increases at t2 and t3 are due to the initial protein content in the corn stubble before ensiling (mature green forage contained 4.9 %, while the stubble contained 5.2 % CP; results not shown) and to the proliferation of cellular protein from the Saccharomyces cereviseae yeast used to ferment the nopal. However, ^{Alhanafi et al. (2019)}) recorded a lower CP content in silages of Opuntia ficus, indica added with Atriplex (6.41 %); compared to t2 of this study.

The concentration of neutral detergent fiber (NDF) was lower in T1 (P<0.05, Table 2); an increase of 18% was observed in T2. This increase is due to the presence of corn stubble, which has a high amount of fiber; however, part of the hemicellulose is hydrolyzed during the ensiling process. At this stage, pentoses are released and they are fermented into lactic acid and acetic acid (^{McDonald et al., 2002}).

Likewise, acid detergent fiber (ADF) content showed the same behavior; ADF increased 59 and 49 % at T2 and T3, respectively (P<0.05, Table 2). In the same way, the increases in ADF were attributed to the presence of corn stubble at T2 and T3. Nevertheless, NDF and ADF contents were within the range of good quality forages.

Fermentation parameters of silage process

The pH values were higher in t2 and t3, with respect to t1 (P<0.05, Table 3); however, all were within the ideal values, which is indicative of a good fermentation and conservation process. It is worth mentioning that the speed with which a silage achieves a pH <4.0 guarantees silage stability and reduces nutrient loss by secondary fermentations, or by bacterial and fungal contamination (^{Ha Vu et al., 2019}).

The concentration of ammonia nitrogen in the silages increased with the cactus and fermented cactus inclusion (P<0.05, Table 3). This increase recorded in t2 and t3 silages could have been caused by an increase in protein degrading microorganisms (^{Berumen et al., 2015}; Ruangyote and Metha, 2018), or also the use of corn stover reduces the amount of soluble carbohydrates and therefore increases protein degradation (^{Herremans et al., 2019}). However, to classify a silage as good quality, the maximum ammonia nitrogen concentration should be 7-20 % of total nitrogen (^{Sánchez and García, 2017}); therefore, the experimental silages obtained values that are within this range and therefore indicate that an adequate fermentation process was carried out.

The concentration of acetic acid was different between treatments (P<0.05, Table 3). Volatile fatty acids are the product of fermentations induced by the presence of coliform bacteria that transform lactic acid into acetic and butyric acid, and are present in manure and soil. According to ^{Kung et al. (2018)}, acetic acid concentrations are between 0.5 and 2.0%, when DM content is 45 to 55 %; therefore, the values obtained in this work are within the range for good quality silages (0.5-1.1 %). Propionic acid values were different among treatments (P<0.05, Table 4). The results obtained in this study are equal to those reported by ^{González et al., (2019)} in corn silage with cactus and fermented cactus (4.0%). Butyric acid concentrations were higher in silages containing stubble (P<0.05, Table 3); however, these values indicate that there was adequate fermentation. In addition, ^{Da Silva et al. (2020)} report lower propionic acid values in cactus silages with gliricidia. According to the lactic and butyric acid values obtained in the silages of this study, it can be inferred that the inclusion of cactus and fermented cactus in the stubble promotes an increase in lactic acid and a decrease in butyric acid, which results in silages with good fermentative and nutritional quality.

Ruminal fermentation parameters

N-NH₃ concentration was lower in t1, compared to t2 and t3 (P<0.05, Table 4). The inclusion of cactus and fermented cactus in the stubble silages promoted an increase of 48 % and 27.7 % in N-NH₃, respectively. Changes in this variable indicate that proteolysis is taking place and it increased due to crude protein increase in the silage by the addition of stubble and fermented nopal. However, there are studies that affirm that increases in N-NH₃ concentration are because the protein is not incorporated into microbial protein synthesis, which would reflect an energy loss in the ruminant (^{Rodríguez et al., 2007}). On the contrary, other authors also found increases for N-NH₃ when crude protein content was increased through the addition of urea in pineapple silages (^{López-Herrera et al., 2014}). Additionally, these results agree with those obtained by ^{Pinho et al. (2017)} in nopal silages (17 mg/dL) at 9 h of incubation. Thus, in this study, the N-NH₃ concentrations recorded in all treatments presented adequate levels higher than 5 mg/dL, which allows guaranteeing microbial protein synthesis (^{Rodríguez et al., 2007}).

On the other hand, acetic acid concentration decreased and propionate concentration increased in corn stubble and fermented cactus silage (P<0.05; Table 4), compared to t1 and t2. In this sense, a decrease in acetic acid concentration at t2 and t3 is closely related to a decrease in the fermentation of structural carbohydrates (NDF and ADF). Thus, since structural carbohydrates have the highest concentration in the nutrients of t2 and t3, it is suggested that adequate fermentation is not taking place (^{Van Soest, 1994}). However, ^{Sánchez et al. (2014)} also detected a decrease in acetate concentration and an increase in ruminal propionate. Regularly, the production rate of propionate and other AGV is directly related to the consumption of fermentable substrates from the diet, which favors propionate synthesis from microbial fermentation by amylolytic bacteria (^{Van Soest, 1994}).

Maximum gas production (Gmax) was higher at t1 (P<0.05; Table 5). The inclusion of cactus and fermented cactus in silages with corn stubble decreased maximum gas production 31.9 % and 48.7 %, respectively. The values obtained in this study are lower than those reported in leucaena and star grass (234 and 154 ml/g, respectively) by ^{Naranjo et al. (2016).} Similarly, ^{González et al. (2019)} recorded higher Gmax values in corn silages with nopal (176 ml). The decrease in gas production observed in this study is attributed to the decrease in dry matter digestibility at t2 and t3, as a result of the addition of corn stubble that NDF and ADF values are higher than those recorded at t1. As previously mentioned, the presence of structural carbohydrates prevents adequate fermentation, which is reflected in gas production.

On the other hand, the dormancy period "A" decreased 55 % at t2 (stubble + nopal silage). To explain this result, variables that were not considered in this study, such as soluble carbohydrate content or lignin, must be taken into account, since the faster onset of fermentation depends on these. In this sense, ^{López-Inzunza et al. (2017)} reported longer dormancy times for silages with higher ADF content, with respect to NDF content in silages with pineapple stubble; these authors showed structural carbohydrate concentrations and dormancy times similar to those reported in this study (72 % NDF and an A of 3.8 h). Moreover, the A values obtained in this study are also similar to those recorded by ^{González et al. (2019)} in corn forage silages with nopal cactus. Additionally, methane production decreased 32 and 49 % in silages that included nopal and fermented nopal, respectively (P<0.05; Table 5). Similarly, the methane: CO2 ratio also decreased with the addition of fermented nopal and nopal without fermented, as well as corn stubble. This variable is closely related to ruminal methane synthesis. Higher values in this ratio suggest an increase in ruminal methane synthesis through the CO₂ reduction pathway (^{Murillo et al., 2018}). Thus, although a reduction in methane and CO₂ production is closely related to a decrease in the fermentative quality of t2 and t3 silages, this reduction also suggests changes or even an inhibition in methanogenic populations (^{Tavendale et al., 2005}. In addition, the decrease in methane is directly related to an increase in ruminal.