The E. coli 6-phytase used in this study was produced through controlled fermentation (Advanced Enzyme Technologies). The enzyme is presented as cream to off-white colored granules with an activity level of 10,000,000 FTU/kg and has been found to remain heat-stable at temperatures up to 85°C.

The experiment was conducted in two distinct phases (Phase 1, d 1 to 21; phase 2, d 22 to 42). This trial was conducted at the commercial swine farm in Haman, South Korea, from June until July 2023. Four hundred cross-bred (Duroc× Landrace×Yorkshire) newly weanling 21-day-old weanling piglet’s initial body weight (BW) was 6.42±0.01 kg. Those were allotted to one of five dietary treatments: control (CNT), a corn-soybean meal basal diet that met or surpassed the requirements of nutrients referred to the NRC [17] guidelines; CNT with 500 FTU/kg; CNT with 750 FTU/kg; CNT with 1,000 FTU/kg; CNT with 1,500 FTU/kg. The levels of phytase supplementation (0, 500, 750, 1,000, and 1,500 FTU/kg) were selected to represent a stepwise increase from standard commercial doses to higher-than-conventional levels. This range was designed to evaluate the dose-response relationship within a physiologically relevant window, while avoiding levels typically considered as super-dosing. The incremental spacing also allowed for the use of orthogonal polynomial contrasts to assess linear and non-linear trends in growth and nutrient utilization parameters. Each treatment had ten replicated pens which assigned eight weanling piglets per pen for group housed. The sex ratio of weanling piglets in each pen is 4:4 (4 castrated male and 4 female) and applied to every pen. The experiment space consists of 50 pens (2.28×1.85 = 4.2 m²) with a partially slatted concrete floor. The environment was controlled by forced ventilation, automated fan-assisted central heating, and artificial lighting to provide a least of 12 hours light in a day. Each pen was equipped with two feeding points-type nipple drinker and one hopper-type feeder. Through the whole experimental period, both water and feed were offered ad libitum to weanling piglets and the pens were cleaned daily. The formula and chemical composition of the diets on an as-fed basis are presented in Table 1.

The BW and cumulative feed intake were measured on the final day of each phase. Through the whole experimental period, growth performance was evaluated at the initial and final days of both phases by calculating data related to average daily weight gain (ADG), average daily feed intake (ADFI), BW, and gain-to-feed ratio (G:F). The ADG was determined by calculation of dividing the weight gain of each weanling piglet through the whole phase of the experimental period by 21 days (d 1 to 21, d 22 to 42). The ADFI was determined by calculation of dividing the cumulative feed intake of each experiment phase by 21 days (d 1 to 21, d 22 to 42). The G:F was calculated by dividing the ADG by the ADFI. Throughout the whole experimental period, mortality was monitored, and no mortality was observed.

The weanling piglets were fed a 2.5 g/kg chromium oxide (Cr₂O₃) contained feed for seven days (d 15 to 21, d 36 to 42) of each experimental period to analyze the nutrient apparent total tract digestibility (ATTD). On days 20 to 21 and 41 to 42, the fecal samples of the weanling piglets were collected. After collecting all fecal samples from each pen, the samples were frozen at −20°C for subsequent assessment. On days 21 and 42, the blood samples for inositol assay were collected by 2 weanling piglets from the anterior vena cava (castrated male and female) of each pen. Randomly selected two weanling piglets (one castrated male and one female) from each pen were anesthetized using pentobarbital sodium at a dose of 200 mg/kg BW and subsequently euthanized through exsanguination at the end of the experiment. Digesta samples collected from the terminal of the ileum were obtained and reserved after being flushed with distilled water and stored at −20°C until AA ileal digestibility was calculated.

The ATTD of dry matter (DM), crude protein (CP), gross energy (GE), ether extract (EE), ash, Ca, and P was calculated using Cr₂O₃ concentration as a marker. The frozen sample was defrosted and desiccated in an air-forced oven (model FC-610, Advantage, Toyo Seisakusho) operating at 75°C for 72 hours. At the beginning of the assay, the dried fecal samples were finely ground through a one-millimeter screen (Christy and Norris Hammer Mill). The GE content was measured by a bomb calorimeter (Parr 1261 bomb calorimeter; Parr Instruments). The CP content assessment was performed using the Kjeldahl method AOAC International 990.03 [18]. The EE content was estimated with AOAC International method 2003.6 [18] and a Soxhlet device (Soxtec 2050; FOSS North America) was used for EE extraction. The ash content was analyzed with AOAC International method 942.05 [18]. The Ca and P contents were assessed with AOAC International method 985.01 [18] after the samples were digested by concentrated nitric acid and 70% perchloric acids. The concentration of Ca was measured by a flame atomic absorption spectrometer (Varian AA FS240; Varian). The concentration of P was measured by a spectrophotometer (Jasco V-550; Jasco) as followed by a previous study [19]. Briefly, samples were digested using concentrated nitric acid and 70% perchloric acid. After digestion, a blue chromogenic reaction was induced by adding acid molybdate and Fiske and Subbarow Reducer (Sigma-Aldrich). The absorbance of the resulting solution was then measured at 620 nm using a spectrophotometer. The concentration of chromium was determined by a spectrometer (Jasco V-650; Jasco).

The fresh samples were kept in a glass tube with vacuum condition and hydrolyzed by 6 N HCl (including phenol) for 24 hours at 110±2°C. The classification of AA was confirmed by the Waters Ion Exchange High-Performance Liquid Chromatography system and software (ver. 3.05.01, Millennium, Waters) was used to link the chromatograms.

The concentration of myo-inositol was measured in accordance with a previous study [20]. Proteins in plasma samples were precipitated through the addition of acetonitrile, and the samples were subsequently lyophilized prior to derivatization. A two-step derivatization process, involving oxidation followed by salinization, was employed. Deuterated myo-inositol was used as an internal standard, and myo-inositol levels were quantified using an Agilent 5977A gas chromatograph coupled with a mass spectrometer (Agilent Technologies).

The statistical analysis of this experiment was computed by Analysis of Variance through the General Linear model (GLM) contained in SAS software ver. 9.2 (SAS Institute). Orthogonal polynomial contrasts were utilized to calculate linear and quadratic effects of the supplementation of phytase. The experiment was designed as a randomized complete block design with phytase treatment as a fixed effect and block as a random effect. Pen was an experimental unit, blocked by the initial weight of weanling piglets. The probability values of p≤0.05 were treated as the significant difference between the mean values. When phytase showed a significant difference in parameters at p<0.05 and p<0.01, the Tukey test was used to compare the difference between means.

The effect of phytase supplementation on growth performance is shown in Table 2. In phase 1, weanling piglets exhibited a linear effect in ADG and G:F (p<0.05) with increasing dietary phytase levels, whereas ADFI remained unchanged. In phase 2, ADG and G:F showed a linear effect (p<0.01) with higher phytase levels whereas ADFI showed a tendency of quadratic effect (p = 0.084). Over the entire study, there was a linear effect (p<0.01) in ADG and G:F whereas no significant differences were noted in ADFI.

The effect of phytase supplementation on ATTD is shown in Table 3. In phase 1, the ATTD of DM and EE showed a tendency toward linear effect (p = 0.054 and 0.082, respectively). Piglets fed diets with increasing phytase concentrations demonstrated a linear effect (p<0.05) in CP and GE ATTD. No difference was shown in ash ATTD. Additionally, weanling piglets fed diets with increasing phytase levels showed a linear effect (p<0.05) in the digestibility of Ca, while P ATTD showed a tendency toward linear effect (p = 0.056). In phase 2, a quadratic effect (p<0.05) on DM ATTD. The ATTD of CP showed a linear effect (p<0.05), while GE ATTD did not show any linear effect but tended towards quadratic effects (p = 0.059). Additionally, no differences were found in EE and ash ATTD, whereas the ATTD of Ca and P showed linear and quadratic effects (p<0.01).

The effects of varying phytase supplementation on AA digestibility are detailed in Table 4. For indispensable AA, there was a linear effect (p<0.05) in the digestibility of lysine (Lys), phenylalanine (Phe), and leucine (Leu) with higher phytase levels. The digestibility of arginine (Arg), and tryptophan (Trp) were linearly affected (p<0.05) by supplemental phytase, while no differences were shown for histidine, isoleucine, methionine, and threonine. The digestibility of valine showed a tendency towards linear effect (p = 0.06). For dispensable AA, cysteine (Cys) and alanine (Ala) digestibility was linearly affected (p< 0.05) by supplemental phytase. No significant differences were found in the digestibility of aspartate, glutamine, glycine, proline, serine, and tyrosine.

In phase 1, there were no significant differences in myo-inositol levels among the dietary treatments (Figure 1). However, in phase 2, a linear increase (p<0.01) in myo-inositol levels was shown with dietary phytase supplementation levels.