Acidifiers blend on digestibility, blood metabolites and performance of nursery piglets

The aim of this study was to evaluate the effect of inclusion of two acidifiers blend (with different protection levels) on the acidification potential in vitro, apparent total tract digestibility of diet nutrients, blood metabolites, and average daily gain of nursery piglets. Two commercial acidifiers blend additives (AC1 and AC2) were evaluated. The acidification potential of the two acidifiers blend added to the diet were tested by buffering capacity and initial pH in vitro. For the evaluation of apparent digestibility of diet nutrients, blood metabolites and average daily gain, two experiments were carried out. The AC2 acidifier had a greater effect on the initial reduction in pH and a lower linear buffering rate of the diet. The use of acidifiers did not improve the nutrient digestibility of the diets and had no effect on the blood metabolite concentrations of urea, total protein, albumin, globulin, and lactate. Although the inclusion of acidifiers in the diets did not affect the apparent total tract digestibility of the diet nutrients the inclusion of the AC2 acidifier at different levels showed a negative quadratic effect on the average daily gain, with the best inclusion level being 0.26%.

As an alternative, studies have been carried out using acidifiers blend as feed additive to reduce gastric pH, altering the action of digestive enzymes and reducing the proliferation of pathogenic microorganisms, with the aim of improving the apparent total tract digestibility, feed conversion and performance of the animals as a whole (Diao et al., 2014). The main action mechanisms of acidifiers are the reduction of gastric pH, the antimicrobial effect and pathogen inhibition and the stimulation of pancreatic enzyme secretion (Papatsiros et al., 2012). Additionally, acidifiers serve as an energy source during the intermediate metabolism of the gastrointestinal tract and can improve the mineral utilization by the process of chelation (Suryanarayana et al., 2012). Therefore, the use of acidifiers in combination or as a substitute of antibiotics could be a potential alternative. The use of organic acids blends as antibiotic substitute improved performance, serum immunity, intestinal morphology, microbiota (Long et al., 2018) and reduced the diarrhea in weaned piglets (Lei et al., 2017). However, the action potential of acidifiers in diets depends on its composition, pKa and level of inclusion in the piglets diets.
Results from studies using acidifiers blend as feed additive in piglets' diets are still quite controversial, where inclusion levels depend on both the chemical composition and the physicochemical properties of the diet as well as on the acidifying and buffering power of the acidifier used . Additionally, few studies have evaluated the impact of acidifiers on pH and buffer capacity in order to determine the appropriate inclusion level in diets prior to in vivo experimentation (Bockor et al., 2017).
Due to the properties of acidifiers, positive effects are expected on the feed palatability and digestibility, reducing buffering capacity of feed, improving the digestive system, health, and animal performance (Nowak et al., 2021). However, this expected effect depends on the quantity and quality of the acidifier, also considering the composition of the diet. Furthermore, unprotected acidifiers can be easily buffered by the effect of certain ingredients or absorbed prior to the effect on the intestinal environment (Canibe et al., 2001).
Through advances in technology, the incorporation of acidifiers has improved the palatability aspects and potential of dissociation through the process of micro-encapsulation (Grilli et al., 2010). This process involves molecules of organic acid or a blend, with a thin layer of polymers or lipids, which when ingested by the animal slowly dissociate the acids in the small intestine, showing better results compared to unprotected acidifiers (Piva et al., 2007;Grilli et al., 2010). Furthermore, encapsulated compound acidifiers improve the intestinal morphology and function by reducing the gastrointestinal pH, so as enhance the intestinal adaptation and immunity, the growth performance improvement in weaning piglets occurs (Jia et al., 2010).
Thus, the aim of this study was to evaluate i) the effect of inclusion of two acidifiers blend (with different protection levels) on the acidification potential in vitro, apparent total tract digestibility of diet nutrients and blood metabolites of nursery piglets; ii) the effect of different inclusion levels of the acidifier blend with a higher protection level on the average daily gain of the nursery piglets.

MATERIAL AND METHODS
Animal care procedures throughout the study followed the protocols approved by the Ethics Committee for Animal Use (ECAU) at the Federal Catarinense Institute (IFC), numbers 105/2015 (experiment 1) and 152/2016 (experiment 2).
Two commercial acidifiers blend additives (AC1 and AC2, Sanex Comércio e Indústria Veterinária Ltda, Curitiba, Paraná, BR) were evaluated in vitro. The AC1 acidifier blend was a compound of five organic acids including benzoic, lactic, formic, fumaric and citric acids, with a concentration of 7286 mEq H + /kg, 23.5% in a protected form. The AC2 acidifier blend was composed of benzoic, sorbic, formic, lactic and phosphoric acids, with a concentration of 7302mEq of H + /kg, 53.8% in a protected form. Both products were subjected to the same lipid protection process, patented by the company.
To evaluate the buffering capacity, nine inclusion levels (0.25, 0.50, 0.75, 1, 2, 5, 9, 12 and 25%) of the AC1 and AC2 acidifiers blend were added to the diet and evaluated for buffering capacity by the linear buffering rate according to Bockor et al. (2017). To evaluate the initial pH of the diet, five inclusion levels (0.25, 0.50, 0.75, 1 and 2%) of the AC1 and AC2 acidifiers blend were added to the diet.
The pH was measured with the pH meter (pH 330, SET, WTW) using a 0.5 g sample in a solution with distilled and deionized water under constant agitation. Buffering capacity was measured using titration curves. The initial pH was measured, then NaOH (0.05 mol/l) was used to raise the pH to 8.0, and HCl (0.05 mol/l and 0.1 mol/l, respectively) was used to lower the pH to 2.0, with constant reductions around 0.200 pH points. From the titration curves pH values were transformed by exponential function (1/ pH) and the linear buffer rate (LBR) was obtained as the inverse of the angular coefficient between the transformed pH measurement as the dependent variable and the amount of acid added as independent variable.
The diet used for both the initial pH and the LBR evaluations was an isoenergetic and isoproteic diet formulated to attend to the nutritional requirements of piglets during the nursery phase, according to Rostagno et al. (2011) (Table 1). Statistical analysis was performed using SAS version 9.3 (Statistical Analysis Institute, Cary, NC, USA). The results obtained from the initial pH and LBR (dependent variables) were adjusted for linear regressions and the level of inclusion of the acidifiers blend was considered as an independent variable. The regression lines were compared with a significance level of 5%. Differences between the intercept and the slope of the data adjustment for the linear regressions of the different acidifiers blend were evaluated.
To evaluate the apparent total tract digestibility of diet nutrients, blood metabolites, and average daily gain of nursery piglets two experiments were carried out in at a Pig Experimental Farm located in Santa Catarina State in southern Brazil (26°22'12"S and 48°43'20"W), with an altitude of 9 m. The climate is Cfa (moist mesothermal with hot summer), according to the Koppen classification system.
Experiment 1 -To evaluate the apparent total tract digestibility three treatments (control, AC1 and AC2) were used, with six animals per replication, allocated in 18 cages. The experiment was repeated twice in time, totalizing 12 animals per treatment. After weaning at 28 days of age, the animals (Large White x Landrace; 50% males and 50% females) were identified with ear tags, individually weighed (6.78 ± 1.01 kg), and placed into adjustable individual cages with feces and urine collection systems and pacifiers installed. The animals underwent a five-day diet adaptation period followed by a five-day collection period, totaling an experimental period of 10 days. The room temperature was 28ºC.
All animals received an isoenergetic and isoproteic diet formulated to attend to the nutritional requirements of piglets during the nursery phase, according to Rostagno et al. (2011) ( Table 1) used also in the in vitro test. However, the treatments differed in the acidifier blend used as feed additive (AC1 and AC2). Thus, the three experimental treatments were: control group, which received the basal diet and 0.3% addition of refined sugar cane (CONT); AC1 group, which received the basal diet and the addition of the AC1 acidifier blend (AC1); and the AC2 group, which received the basal diet and the addition of the AC2 acidifier blend (AC2), both acidifier blend with a 3% inclusion level in the basal diet. Refined sugar cane was added in the control group to equalize the energy provided by the acidifier blend. The animals received diets in two daily meals (6:00 am and 6:00 pm), with free access to water. After the adaptation period, the amount of feed provided daily for each animal was calculated based on the metabolic weight (LW 0.60 ) and adjusted to the level of metabolizable energy of diet according to Noblet and Perez (1993).
The feces (total fecal collection method) were collected twice a day for a five-days period, weighed, packed in plastic bags individually identified, pre-dried at 65°C, milled (with a 1 mm mesh) and homogenized. The pre-dried feces were analyzed for dry matter (DM), organic matter (OM), crude protein (CP) and ash (A) according to AOAC (Official…, 1990). These results were used to calculate the apparent total tract digestibility of the diet nutrients (ATTDnutrient), as follow: (((Nutrient diet-Nutrient feces) / Nutrient diet) x100).
To evaluate blood metabolites, 12 nursery piglets (Large White x Landrace; 50% males and 50% females) were identified with ear tags, individually weighed (6.85 ± 0.85 kg), standardized according to sex and live weight and allocated into three pens with a total area of 2.68 m 2 with a fenestrated plastic floor suspended 0.90 m from the ground and with pacifiers installed, and received one of the experimental diets (CONT, AC1 or AC2) ad libitum. The initial room temperature was 28ºC, being reduced by 2 degrees per week. The number of animals per treatment (4) was determined using the minimum number of replications formula (Kaps and Lamberson, 2009) to detect differences between treatments at a level of 5% and 80% power of the test being in accordance with ECAU regulations.
Animal blood collections for blood metabolites analyzes were performed by venipuncture of the jugular using 10 ml vacutainer tubes at weaning (beginning of the experiment) and again 12 days later. To obtain the serum, the samples were centrifuged at 7,000G for 5 min. The serum was then stored in Eppendorf tubes and frozen at -20°C for further analysis of urea (1,013-4/50, Labtest, Lagoa Santa, MG, Brazil; sensitivity 0.94mg/dL), albumin (1,007-250, Labtest, Lagoa Santa, MG, Brazil; sensitivity 0.015g/dL), lactate (138-1/50, Labtest, Lagoa Santa, MG, Brazil; sensitivity 0.4 mg/dL), and total protein (99-1/250, Labtest, Lagoa Santa, MG, Brazil; sensitivity 0.0168g/dL), that were colorimetrically measured in duplicate using commercial kits developed for veterinary or human serum validated for multi-species (all coefficients of variation below 10%). All analyzes were performed in a commercial laboratory. Globulin was calculated as follows: globulin=total protein-albumin.
Statistical analysis was performed using SAS version 9.3 (Statistical Analysis Institute, Cary, NC, USA). For the digestibility analyses a completely randomized design with three treatments and twelve replications (repeated twice in time) was used. The GLM procedure was used to test the effects of treatment (CONT, AC1 and AC2) on apparent total tract digestibility of diet nutrients. Sex was included in the model as a random effect. The effect of the experiment repetition over time was included in the model. The animal was considered the experimental unit. The model used was Yijl= μ+ αi+ βj + γl + εijl, where Yijl is the observation given acidifiers i, time j and sex l; μ is the overall mean; αi is the effect of acidifier i; βj is the effect of time j; γl is the effect of sex l and εijl is the random error associated with each observation, being μ= 0 e variance= 1. Means were compared using the Tukey test with a significance level of 5%. There were no significant interaction effects, therefore they were removed of the final model.
For the blood metabolites analyses a completely randomized design with three treatments and four replications was used. The GLM procedure was used to test the effects of treatment (CONT, AC1 and AC2) on blood metabolites. Sex was included in the model as a random effect. Blood metabolite data was analyzed as a repeated measure over time. The animal was considered the experimental unit. The model was the same used for digestibility analyses, except that instead of considering the effect of the experiment repetition over time, the effect of the blood sample collection was considered. Means were compared using the Tukey test with a significance level of 5%.
Experiment 2 -To evaluate the average daily gain (ADG) of nursery piglets an experiment was carried out with 24 nursery piglets (Large White x Landrace; 50% males and 50% females) at 28 days of age, allocated in four treatments with six animals per replication. After weaning, the animals were identified with ear tags, individually weighed (8.59 ± 1.19kg), standardized according to sex and live weight and allocated into groups of three animals in pens with a total area of 2.68 m 2 , with a fenestrated plastic floor suspended 0.90 m from the ground and with pacifiers installed. The initial room temperature was 28ºC, being reduced by 2 degrees per week.
All animals received the same basal diet used in Experiment 1, however, the treatments differed in the inclusion level (0, 0.2, 0.4 and 0.6%) of the AC2 acidifier blend. The concentration of the acidifier blend in each diet was calculated according to the predicted buffering capacity. The number of animals per treatment (6) was determined using the minimum number of replications formula (Kaps and Lamberson, 2009) to detect differences between treatments at a level of 5% and 80% power of the test being in accordance with ECAU regulations. Also, the animal was considered the experimental unit.
During the 21-day experimental period, the animals were weighed individually every 7 days to evaluate the average daily gain (ADG) of each animal.
Statistical analysis was performed using SAS version 9.3 (Statistical Analysis Institute, Cary, NC, USA). The experimental design was a completely randomized with four treatments and six replications. To evaluate the effect of the treatment (0, 0.2, 0.4 and 0.6% of AC2 in the diet) on the piglets ADG the REG procedure was used to determine the optimal level of acidifier blend in the diet to obtain the higher ADG.

RESULTS
The initial drop in pH differed according to the acidifier blend used (Figure 1, Table 2), with the addition of AC2 showing the greatest decrease in pH (P < 0.001), despite the concentrations of AC1 and AC2 being similar (7289 and 7302 mEq H + /kg, respectively).
Linear regression analysis demonstrated that the addition of the acidifier blend to the diet presented different linear buffering rates (LBR) (Figure 2, P < 0.001). The diet using AC2 as feed additive had a lower LBR than the diet with AC1 (Table 2, P < 0.001).
There was no effect of treatment on dry matter, organic matter, ash, and crude protein digestibility (P > 0.05), with average values of 86.93, 86.93, 68.26 and 85.46%, respectively (Table 3).
Regardless of the collection time (at weaning and 12 days later), there was no effect of treatment (P > 0.05) on the concentrations of urea, total protein, albumin, globulin, and lactate, with mean values of 17.36 mg/ dL, 4.88 g/dL, 3.32 g/dL, 1.56 mg/dL and 9.23 mmol/L, respectively (Table 4).    CONT: group which received only basal diet; AC1: group which received basal diet and addition of acidifier AC1; AC2: group which received basal diet and addition of acidifier AC2, both with inclusion level of 0.3% in the basal diet; Pr>F: probability; means followed by standard error of the mean. CONT: group which received only basal diet; AC1: group which received basal diet and addition of acidifier blend AC1; AC2: group which received basal diet and addition of acidifier blend AC2, both with inclusion level of 0.3% in the basal diet; Pr>F: probability; means followed by standard error of the mean; col 1: blood sample collected at weaning; col 2: blood sample collected 12 days after weaning. 1 Variance analysis between collection time; 2 Variance analysis between treatments of blood sample collected 12 days after weaning.
A negative quadratic effect on the ADG of piglets was observed with the addition of the AC2 acidifier blend to the diet (P < 0.05), indicating 0.26% as the best level of AC2 inclusion, in terms of the higher ADG (Figure 3).

DISCUSSION
The efficiency of acidifiers additives in weaned piglet diets are controversial due to several factors, such as the growth stage, the diet complexity, the type of acid used and level of inclusion (Suryanarayana et al., 2012), as well as the palatability of the diet, which can stimulate or inhibit consumption (Kil et al., 2011). These inconsistencies are also related to the nonevaluation of the diet buffer capacity with the inclusion of acidifiers at the formulation time (Corassa et al., 2006).
Increasing the inclusion of the acidifiers blend AC1 and AC2 as feed additive differently effected the initial dietary pH. The use of AC2 demonstrated a greater efficiency in reducing the initial pH, especially when the inclusion level was greater than 0.75%. Additionally, the buffering capacity of the diets was also lower when AC2 was added at different inclusion levels. This was expected, since in the formulation of this compound, the buffering capacity of each acid was taken into consideration, aiming to minimize the effects on the buffering capacity of the diet. The presence of phosphoric acid in the AC2 acidifier blend, which had a higher water solubility, may have promoted the greater initial pH decrease and influenced the buffering capacity, besides being the only inorganic acid present in the compounds. Another factor that can influence the buffering capacity of acidifiers is related to the level of protection used. The reason is that unprotected acidifiers can be easily buffered by components present in food ingredients or absorbed prior to their effect on the intestinal lumen. (Nowak et al., 2021).
Means followed by standard deviation of the mean; ADG: average daily gain, AIL: acidifier blend inclusion level; LW: live weight; r 2 : coefficient of determination. The main action mechanisms of acidifiers additives are the antimicrobial effect and the inhibition of pathogens, and may stimulate the secretion of pancreatic enzymes, reduce gastric pH, act as an energy source during the intermediate metabolism of the gastrointestinal tract and improve the use of minerals by the chelation process (Suryanarayana and Ramana, 2015), aiming to improve total apparent digestibility, feed conversion and performance of the animals (Diao et al., 2014). Protected acidifiers can improve intestinal morphology and function, reduce gastrointestinal pH, and improve intestinal adaptation (Jia et al., 2010), expecting positive effects on digestibility. However, in this study there was no effect of treatment on nutrient digestibility, showing that neither the addition of AC1 and AC2 acidifiers blend improved the overall nutrient digestibility of nursery piglet diets. Likewise, in a study with complex diets, the addition of acidifiers (blend and sodium butyrate) as growth promoter had no effect on increasing the nutrients digestibility of the diet (Boas et al., 2016).
Serum protein was associated with the diet provided at each stage of development and may vary according to the age and genotype of the animals (Stukelj et al., 2010). In this study, regardless of the collection time, the use of AC1 and AC2 acidifiers blend as feed additive in the diet did not influence the concentrations of urea, total protein, albumin, globulin, and lactate, with values within the reference range for the age and species evaluated, corroborating with previous studies (Stukelj et al., 2010;Mu et al., 2017).
A high level of lactate in the stomach tends to inhibit the secretion of hydrochloric acid and therefore, generates insufficient enzyme secretion (Suryanarayana and Ramana, 2015). Therefore, as presented in this study, blood lactate levels were not influenced by the addition of AC1 and AC2 acidifiers blend, regardless of the collection time. This can be considered a positive point for additive supplementation, as it did not interfere in the stomach acidity process of piglets.
Moreover, the lack of effect of AC1 and AC2 acidifiers blend on blood metabolites may be attributed to inherent piglet factors, such as greater weight at weaning and a consequently better immune response (Denck et al., 2017), which could contribute to the low response of acidifiers during the supplementation period.
In terms of the higher ADG, the best inclusion level of the AC2 acidifier blend to the diet was found to be 0.26%, agreeing with previous studies that observed that dietary supplementation with acidifiers improved growth performance in weanling piglets (Miguel et al., 2011;Lei et al., 2017;Long et al., 2018).
The improvement in the ADG of piglets supplemented with acidifier blend in the diet can be due to the modulation of pH, improving the action of digestive enzymes with a probable improvement in the nutrient utilization of the diet and in the animal performance (Diao et al., 2014).

CONCLUSION
The use of an acidifier composed of benzoic, sorbic, formic, lactic and phosphoric acids in a protected form (AC2) as feed additive had a greater effect on the initial reduction in pH and a lower linear buffering rate of the diet than those composed of five organic acids including benzoic, lactic, formic, fumaric and citric acids (AC1). The use of acidifiers blend with inclusion of 0.3% did not improve the nutrient digestibility of nursery piglets and had no effect on blood metabolites. Considering the average daily gain of the animals, the best inclusion level of the AC2 acidifier blend in the diet was 0.26%. Feeding the blend of organic acids and medium chain fatty acids reduces the diarrhea in piglets orally challenged with enterotoxigenic Escherichia coli K88. Anim. Feed Sci. Technol., v.224, p.46-51, 2017.