Effect of phytase on intestinal phytate breakdown, plasma inositol concentrations, and glucose transporter type 4 abundance in muscle membranes of weanling pigs

Abstract

The objective of this present study was to determine the effects of phytase dosing on growth performance, mineral digestibility, phytate breakdown, and the level of glucose transporter type 4 (GLUT4) in muscle plasma membranes of weanling pigs. A total of 160 barrows were used in a randomized completely block design and assigned to 4 treatments for a 7-wk study. Depending on the feeding phase, diets differed in dietary calcium (Ca) and phosphorus (P) levels (positive control [PC]: 8 to 6.8g/kg Ca; 7.3 to 6.3 g/kg P; negative control [NC]: 5.5 to 5.2 g/kg Ca; 5.4 to 4.7 g/kg P). NC diets were supplemented with phytase at 0 (NC); 500 (NC + 500 FTU); or 2,000 FTU/kg (NC + 2,000 FTU) phytase units/kg. Blood was collected after fasting (day 48) or feeding (day 49) for measurement of plasma inositol concentrations. On day 49, 2 pigs per pen were euthanized, and duodenal and ileal digesta samples were collected to determine inositol phosphates (InsP_6-2) concentrations. High phytase supplementation increased BW on days 21, 35, and 49 (P < 0.05). Over the entire feeding period, ADG, ADFI, and feed efficiency were increased by NC + 2,000 FTU compared with the other treatments (P < 0.05). Postprandial plasma inositol concentration was increased in NC + 2,000 (P < 0.01), but there was only a tendency (P = 0.06) of a higher fasting plasma inositol concentration in this group. Inositol concentrations in the portal vein plasma (day 49) were not different among treatments. Duodenal digesta InsP₅ and InsP₆ concentrations were similar in PC and NC, but higher in these 2 treatments (P < 0.05) than those supplemented with phytase. Phytase supplementation decreased InsP_6-4, resulting in increased InsP_3-2 and myo-inositol concentrations. Similar effects were found in ileal contents. Compared with NC, phytase supplementation resulted in greater cumulative InsP_6-2 disappearance (93.6% vs. 72.8% vs. 25.0%, for NC + 2,000 FTU, NC + 500 FTU and NC, respectively, P < 0.01) till the distal ileum. Longissimus dorsi muscle plasma membrane GLUT4 concentration was increased by NC + 2,000 FTU (P < 0.01) compared with NC. In summary, high phytase supplementation increased growth performance of nursery pigs. The higher myo-inositol release from phytate could contribute to the increased expression of GLUT4 in muscle plasma membranes. Further investigation is needed to determine whether this is associated with enhanced cellular glucose uptake and utilization.

INTRODUCTION

Plant seeds store phosphorus mainly as phytate (Kumar et al., 2012). The phytate molecule is essentially a salt of myo-inositol (a.k.a. inositol) phosphate because it is able to bind minerals within the molecule. Phytate is also a major antinutrient in monogastric animals as it cannot be efficiently degraded due to low phytase activity in the gastrointestinal tract. Inositol phosphates (InsPs) can bind minerals and this reduces digestibility of bound cations, including calcium (Ca), phosphorus (P), and Zinc (Woyengo et al., 2009; Woyengo and Nyachoti, 2013) as well as iron (Laird et al., 2018). Phytate also binds proteins and enzymes such as trypsin and α-amylase and inhibits their activities, consequently reducing protein and carbohydrate digestibility (Singh and Krikorian, 1982; Deshpande and Cheryan, 1984). This effect ultimately reduces growth performance of animals. Recently, feeding high level of phytase in the diet has become more common, and effects attributable to the released myo-inositol from phytate degradation have been recently discussed (Laird et al., 2018; Sommerfeld et al., 2018). Apart from these positive effects of phytate degradation and the enhanced P utilization by the animal, phytase use can have beneficial effects on the environment by limiting the release of undigested phosphorus into soil and ground water.

Use of microbial (fungal or bacterial derived) phytase has led to a reduction in the level of inorganic phosphorus used in animal diets. Currently, phytase is mostly added at 500 FTU/kg to pig diets as the standard inclusion rate (Wilcock and Walk, 2016). It has been reported that 500 FTU/kg phytase is equal to 0.3 to 1.7 g/kg inorganic P depending on the phytase source (Augspurger et al., 2003). It has been shown that 2,000 to 2,500 FTU phytase/kg feed could lead to an additional increase in growth performance of pigs beyond the expected growth increase from just releasing adequate P to support growth (Santos et al., 2014; Laird et al., 2018). Zeng et al. (2016) reported that phytase up to 20,000 FTU/kg feed could increase digestibility of CP and AA. However, the mechanism of the extra phosphoric effect of higher phytase inclusion level than recommended (super-dosed phytase) is still poorly defined. One possible reason is that the more complete phytate degradation by super-dosed phytase releases higher levels of myo-inositol, which is known to have insulin-like effects (Huber, 2016; Lee and Bedford, 2016). Myo-inositol may increase insulin sensitivity by enhancing the concentration of PIP3 in the cell (Jiang et al., 1998). In addition, myo-inositol may also promote insulin secretion from pancreatic β cells. In mammalian cells, pyrophosphates (InsP₇) are synthesized from myo-inositol through a series of chemical reactions involving many enzymes such as inositol hexakisphosphate (InsP₆) kinases, diphosphoinositol pentakisphosphate (PP-IP₅) kinases (Wilson et al., 2013). It has been reported that overexpression of InsP₆ kinases in pancreatic β cells increased production of InsP₇, and this stimulated exocytosis of insulin in a dose-dependent manner (Illies et al., 2007). Insulin is known to stimulate glucose uptake by increasing the translocation of intracellular glucose transporter type 4 (GLUT4) vesicles to the plasma membrane of myocytes (Kahh, 1996), and it is suggested that glucose transport is the rate limiting step in muscle glycogen synthesis (Richter and Hargreaves, 2013). Thus, by promoting glucose utilization in muscle, and perhaps other tissues, phytase could be contributing to increased animal growth, through a mechanism that is different from its effect in increasing P availability. Therefore, the objective of the present study was to determine the effect of high phytase level on growth performance, mineral digestibility, and inositol phosphate disappearance, and to determine effects on plasma metabolites and GLUT4 muscle plasma membrane concentration.

MATERIALS AND METHODS

Animals

All animal procedures were approved by the Purdue Animal Care and Use Committee. A total of 160 weanling barrows (initial BW 5.6 ± 0.5 kg) were used in a randomized complete block design. The experiment was performed in 2 replicate runs using 80 pigs per run. In each run, treatments were replicated in 5 pens with pigs of similar BW (block) in the same pen. There were 4 pigs per replicate pen. Therefore, for the 2 runs of the experiment, there was a total of 10 replicate pens per treatment. Pigs were fed according to a 3-phase feeding program post-weaning; phase 1 (days 0 to 21), phase 2 (days 21 to 35), and phase 3 (days 35 to 49).

Dietary Treatments

All diets were fed as mash diets. During phase 1, pigs in the positive control (PC) and negative control (NC) treatments were fed a common control diet which met their nutrient requirements to ensure sufficient phosphorus build up during this period. The 2 enzyme groups were supplemented with 500 and 2,000 FTU/kg phytase on top of the PC diet. At the end of phase 1 (days 0 to 21), half of the pigs fed with control diet during phase 1 were assigned to the PC treatment which had sufficient dietary phosphorus concentration that met the nutrient requirement of weanling pigs (National Research Council [NRC], 2012), and the other half were given the NC diet which contained reduced standardized total tract digestible (STTD) Ca (−1.6 g/kg) and P (−1.4 g/kg) compared with PC diets. Two levels of phytase (500 and 2,000 FTU/kg feed, Quantum Blue, AB Vista, Marlborough, UK) were added to the NC diet for treatments NC + 500 FTU and NC + 2,000 FTU, respectively. Titanium dioxide (TiO₂) was added at 0.5% to phase 3 diets to determine apparent ileal digestibility (AID) of calcium, phosphorus, and disappearance of InsP₆ and cumulated InsP_6-2. The diet formulations are presented in Table 1.

Table 1.

Ingredient composition and analyzed nutrient composition of experimental diets on an as-fed basis of PC and NC diets

	Days 0 to 21	Days 21 to 35		Days 35 to 49
	PC	PC	NC	PC	NC
Ingredient, g/kg
Corn	380.25	458.9	472.8	441.9	456.3
Soybean meal	250	300	300	300	300
Soy protein concentrate	88	58.5	58.5	58.5	58.5
Whey, dried	180	100	100	100	100
Lactose	30	51	51	51	51
Fish meal	40	0	0	0	0
Soybean oil	12	9	3.5	9	3.4
Limestone	7.5	8.75	7.6	9	8.2
Monocalcium phosphate	6.5	11.25	4	9	1
l-lysine–HCl	0	1	1	1	1
dl Met	0.6	0.6	0.6	0.6	0.6
TiO2 premix1	0	0	0	15	15
Salt	1	1	1	1	1
ZnO	0.15	0	0	0	0
Vitamin premix2	2.5	2.5	2.5	2.5	2.5
Mineral premix3	1.5	1.5	1.5	1.5	1.5
Total	1,000	1,000	1,000	1,000	1,000
Calculated composition
ME (kcal/kg)	3,415	3,354	3,354	3,310	3,311
CP, g/kg	254	231	232	230	231
Ca, g/kg	8.0	7.1	5.5	6.8	5.2
Total P, g/kg	7.3	6.8	5.4	6.3	4.7
STTD P, g/kg	4.8	4.4	3.0	3.9	2.5
Analyzed composition
CP, g/kg	248	225	228	231	229
Ca, g/kg	8.2	7.3	5.7	7.5	5.1
Total P, g/kg	7.1	6.4	5.2	5.7	4.1
InsP6-2, µmol/kg	—	—	—	17,224	14,859
myo-inositol, µmol/kg	—	—	—	422	340

¹TiO₂ premix resulting in 3g Ti/kg feed.

²Vitamin premix supplied per kilogram of diet: 3,635 IU vitamin A, 363 IU vitamin D3, 26.4 IU vitamin E, 3.6 mg vitamin K, 1,206 µg menadione, 21.2 µg vitamin B12, 4.2 mg riboflavin, 13.5 mg d-pantothenic acid, and 19.5 mg niacin.

³Mineral premix supplied per kilogram diet: 9 mg Cu (as copper sulfate),0.34 mg I (as Ca iodate), 97 mg Fe (as ferrous sulfate), Fe (as ferrous sulfate), 12 mg Fe (as ferrous sulfate), and 97 mg Zn (as zinc oxide).

Experiment Procedure and Sample Collection

Pigs were weighed every week. Feed intakes were recorded and feed efficiency was calculated based on the ratio of ADG and ADFI (G:F). On day 47, feeders were closed for about 12 h in the evening, and blood samples were collected from the jugular vein from 2 pigs per pen on day 48 (fasting blood). On day 49, the same 2 pigs from each pen were injected with 0.5 mL of Telazol [Zoetis Inc., MI, reconstituted with 5 mL of xylazine (RXV Inc., CA)] to provide 0.2-mg tiletamine base, 0.2-mg zolazepam base, and 0.2-mg mannitol per kg BW of pig to induce anesthesia and euthanized with CO₂ asphyxiation. Blood samples from the portal vein (portal blood) and jugular vein (fed blood) were collected in a vacutainer tube containing lithium heparin (BD Inc., NJ). Tubes were shaken slightly to mix thoroughly with the anticoagulant and immediately put on ice. Digesta from proximal duodenum (about 50 cm length from duodenal bulb) and distal ileum (about 50 cm length from ileal–cecal junction) were collected and pooled from 2 pigs per pen and immediately frozen at −20 °C. Longissimus dorsi muscle tissue was collected from each pig and immediately frozen in liquid nitrogen.

Phosphorus, Calcium, and Inositol Phosphate Determination

Duodenal and ileal digesta were freeze dried and ground to pass through a 0.5-mm screen before analysis. Subsamples of ileal digesta and diets were sent to the University of Missouri (MO) for determination of titanium concentration. Briefly, samples were digested in H₂SO₄. Then 30% H₂O₂ was added and the absorbance of samples was measured at 406-nm wavelength (Myers et al., 2004). Phosphorus concentration was determined by digesting the samples in concentrated nitric acid and 70% perchloric acid. Absorbance was measured at 620 nm using a spectrophotometer (SpectraCount, model AS1000; Packard, Meriden, CT; AOAC, 2006) as described by Zhai and Adeola (2013). Calcium concentration was determined using an atomic absorption spectrometer (AAnalyst 300; PerkinElmer, Norwalk, CT). Inositol phosphates (InsP₆, InsP₅, InsP₄, InsP₃, and InsP₂) were determined by the postcolumn UV detection method of Phillippy and Bland (1988) after Blaabjerg et al. (2010). Milled, dry feed, or digesta (100 mg) were extracted with 5 mL, 100 mm NaF, 20 mM disodium EDTA, pH 10, for 30 min with shaking, followed by 30 min in a bath sonicator at approximately 10 °C and a further 2 h standing at 4 °C. The extract was centrifuged at 9,000 × g for 15 min at 4 °C and an aliquot of the supernatant filtered through a 13-mm, 0.45-µm pore-size PTFE syringe filter (Kinesis, UK). Samples (20 µL) were injected onto a 3 mm × 200 mm Dionex CarboPac PA-200 column with 3 × 50 mm guard column of the same material. The column was eluted at a flow rate of 0.4 mL/min with water (A) and 0.6 M methanesulfonic acid (B), mixed according to the following schedule: time (min), % B; 0, 0; 25, 60; 28, 100; 38, 100. The postcolumn reagent was added at a flow rate of 0.2 mL/min and the whole routed through a 192 µL volume knitted reaction coil to a UV detector set at 290 nm. For inositol measurement of feed and digesta, the acid extract was diluted 50-fold with 18.2 MΩ.cm water and aliquots (20 µL) analyzed by HPLC with pulsed amperometric detection according to Lee et al. (2018). A 7-point calibration curve (0 to 8 µM) gave a linear correlation coefficient r² > 0.996. AID of phosphorus, calcium, InsP₆, and cumulated InsP_6-2 were calculated using following equation:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{I}\mathrm{D},\mathrm{\%}=[1-(\mathrm{T}{\mathrm{i}}_{\mathrm{i}}/\mathrm{T}{\mathrm{i}}_{\mathrm{o}})\times ({\mathrm{Y}}_{\mathrm{o}}/{\mathrm{Y}}_{\mathrm{i}})]\times 100,}\end{array}}$$

where Ti_i and Ti_o are the titanium concentrations of the diet and ileal digesta, respectively (mg/kg of DM), and Y_o and Y_i are the concentrations of phosphorus, calcium, InsP₆, or cumulated InsP_6-2 in the ileal digesta and diet, respectively (nmol/g DM).

Plasma Sample Preparation Metabolites, and Inositol Concentration

Heparinized blood samples were centrifuged at 2,000 × g for 15 min at 4 °C and the supernatant was transferred to a new tube and stored at −80 °C before analysis. Plasma glucose concentrations were determined using the Autokit glucose kit (Wako Pure Chemical Industries Ltd., Chuo-Ku Osaka, Japan) following the manufacturer’s protocol. Nonesterified fatty acid concentrations were determined using a NEFA kit from the same manufacturer. Plasma insulin concentration was determined using the porcine insulin ELISA kit following the manufacturer’s protocol (inter- and intra-assay CV: 2.7% and 3.5%; Mercodia, Uppsala, Sweden). Plasma triglyceride was determined with the triglyceride determination kit (Sigma–Aldrich, St Louis, MO). For analysis of plasma inositol, frozen plasma samples were first thawed, then 1 N perchloric acid was added (2 volume of perchloric acid to 1 volume of plasma). Samples were incubated at 4 °C for 30 min before centrifugation at 17,500 × g for 10 min. The supernatant was transferred to a new tube and sent to University of East Anglia (Norwich, UK) for myo-inositol analysis by HPLC-pulsed amperometry by the method of Lee et al. (2018). The acidified plasma samples were diluted 20-fold with 18.2 MΩ.cm water and 20-µL samples injected. A 5-point calibration curve of inositol standards 0 to 1 µM gave a linear correlation coefficient r² > 0.996.

Determination of Muscle Plasma Membrane GLUT4 Concentration and Western Blot Analysis

Frozen muscle samples were thawed on ice, and approximately 1 g of sample was mixed with 2-mL mannitol-HEPES buffer (100 mM mannitol, 10 mM Tris–HEPES, pH 7.4) and homogenized. Homogenized samples were centrifuged at 1,800 × g for 10 min at 10 °C, and the supernatant filtered through a 70-nm nylon mesh. The infiltrate was then centrifuged at 45,000 × g for 45 min at 4 °C to obtain a crude membrane pellet. The membrane pellet was resuspended in 1 mL of mannitol-HEPES buffer to obtain the crude membrane protein fraction. Protein concentration in the crude membrane fraction was determined with the bicinchoninic acid (BCA) protein assay kit (Sigma–Aldrich). Membrane GLUT4 concentrations were measured using the porcine glucose transporter 4 ELISA kit (inter- and intra-assay CV: <15%; MyBioSource, Inc., San Diego, CA) following the manufacturer’s protocol. Data obtained were standardized with the total protein concentration in each sample.

Protein expression of Akt and phosphorylated-Akt was determined by western blot using SDS–PAGE (Yan and Ajuwon, 2017). Muscle samples were homogenized in 1× RIPA buffer [50 mmol/L Trizma-HCl (pH 7.4), 15 mmol/L NaCl, 0.25% deoxycholic acid, 0.1% Triton X, 10 mmol/L EDTA, 1 mmol/L Na₂VO₃, and protease inhibitor cocktail] (Sigma–Aldrich). Tissue homogenates were centrifuged at 10,000 × g for 10 min at 4 °C. Protein concentration in clear homogenates was determined with the BCA protein assay kit (Sigma–Aldrich). Protein samples were resolved by 10% acrylamide gel. Proteins were transferred to a 0.2-µm pore-size nitrocellulose membrane by the semidry method (Bio-Rad, Hercules, CA). Membranes were blocked in 5% bovine serum albumin (BSA) in tris-buffered saline (TBS, 50 mmol/L Tris–HCl, pH 7.4, 150 mmol/L NaCl). Immunoblotting was performed using a polyclonal antibody against mouse protein kinase B (Akt) and phosphorylated-Akt (Cell Signaling Technology, Beverly, MA) at a dilution of 1:1,000. Membranes were stripped and reblotted with rabbit anti-β-actin antibody (Cell Signaling Technology) at a dilution of 1:1,000. Blots were then incubated with IRDye 800CW secondary antibodies (LI-COR, Lincoln, NE) at a dilution of 1:2,000. Signal was detected and quantified with the imaging software on the Odyssey CLx machine (Samkoe et al., 2019; LI-COR).

Statistical Analysis

Data were analyzed using the Proc GLM procedure of SAS (SAS Inst. Inc., Cary, NC) for a randomized complete block design with diet as the main effect. Pen was the experimental unit. The model included period (2 runs), replicate (block) within period and diet. There was no difference between the first and second runs of the experiment. Therefore, data from both runs were pooled. Results are reported as least square means and SEM. Means were different at P ≤ 0.05. When diet effect was significant at P < 0.05, differences between means were compared using the Tukey’s test. Superscript designations were used to indicate significant mean differences.

RESULTS

Growth Performance and Mineral Digestibility

Phosphorus concentration in the PC and NC diets were 7.1, 6.4 and 5.2, 5.7 and 4.1 g/kg feed for phases 1, 2, and 3, respectively (Table 1). Body weights were not different between PC and NC treatments on days 35 and 49. The application of phytase at 500 FTU/kg feed had no significant effect on BW compared with NC or PC treatments. However, phytase supplementation at 2,000 FTU/kg feed significantly increased BW (P < 0.01; Tables 2 and 3). Phytase at 500 FTU/kg feed had no significant effect on ADG, ADFI, and G:F in phases 2, 3, and overall, whereas phytase at 2,000 FTU/kg feed significantly increased these parameters (P < 0.05; Table 3). Both phytase application rates increased AID of P compared with the NC. Pigs that received the PC diet had higher daily Ca and P absorption than NC pigs (P < 0.01; Table 4), and these were not different from the NC + 2,000 FTU pigs.

Table 2.

Growth performance of pigs fed experimental diets during days 0 to 2¹

	PC	PC + 500 FTU	PC + 2,000 FTU	SEM	P-value
BW, kg
Day 0	5.6	5.6	5.6	0.08	0.99
Day 21	8.2b	8.8ab	9.5a	0.81	<0.01
Days 0 to 21
ADG, g/d	120b	148ab	186a	39.8	<0.01
ADFI, g/d	234b	244ab	281a	40.3	0.02
G:F, g/kg	504b	600ab	650a	100.9	<0.01

¹Values are means of each treatment (n = 20 for PC, a combination of each of the 10 replicates in the PC and NC pig treatments because both groups were on the same diet during this period; n = 10 for each of the PC + 500 FTU and PC + 2,000 FTU treatments); means with different superscripts differ significantly (P < 0.05). PC = positive control, PC + 500 FTU and PC + 2,000 FTU = PC added with phytase by 500 or 2,000 FTU/kg feed, respectively.

Table 3.

Growth performance of pigs fed experimental diets different in dietary P, Ca, and phytase levels¹

Item	PC	NC	NC + 500 FTU	NC + 2,000 FTU	SEM	P-value
BW, kg
Day 21	8.2b	8.2b	8.8ab	9.5a	0.26	<0.01
Day 35	14.1b	13.7b	14.4ab	16.1a	0.45	<0.01
Day 49	23.1ab	21.4b	23.4ab	26.0a	0.69	<0.01
Phase 2
ADG, g/d	417ab	388b	404ab	468a	18.89	0.03
ADFI, g/d	696b	711b	716b	811a	22.52	<0.01
G:F, g/kg	594	542	561	578	16.87	0.18
Phase 3
ADG, g/d	636a	556b	637a	708a	21.01	<0.01
ADFI, g/d	1,163ab	1,070b	1,143ab	1,235a	29.96	<0.01
G:F, g/kg	551ab	521b	558ab	574a	11.83	0.02
Overall
ADG, g/d	352b	321b	361b	416a	14.09	<0.01
ADFI, g/d	628b	612b	636b	705a	17.43	<0.01
G:F, g/kg	559ab	522b	565ab	588a	12.36	<0.01

¹Values are means of 10 replicate pens per treatment (5 replicate pens per run); means with different superscripts differ significantly (P < 0.05). PC = positive control, NC = negative control, NC + 500 FTU and NC + 2,000 FTU = NC added with phytase added by 500 or 2,000 FTU/kg feed, respectively.

Table 4.

Apparent ileal Ca and P digestibility and calculated absorption thereof in piglets fed diets different in P, Ca, and phytase levels at the end of the overall feeding period of 49 days¹

Item	PC	NC	NC + 500 FTU	NC + 2,000 FTU	SEM	P-value
Ca digestibility, %	76.6ab	70.3b	69.6b	81.6a	2.70	<0.01
P digestibility, %	55.3ab	45.4c	54.9ab	61.4a	2.54	<0.01
Ca absorption, g/d	6.4a	4.0b	4.0b	6.3a	0.23	<0.01
P absorption, g/d	3.5a	2.1c	2.6bc	3.1ab	0.16	<0.01

¹Values are means of 10 replicate pens per treatment (5 replicate pens per run), means with different superscripts are different (P < 0.05). PC = positive control, NC = negative control, NC + 500 FTU and NC + 2,000 FTU = NC added with phytase added by 500 or 2,000 FTU/kg feed, respectively. The absorption of Ca and P was calculated by multiplying the analyzed digestibility of Ca and P with average daily mineral feed intake.

Plasma Metabolites and Inositol Concentration

Fasting blood plasma glucose, triglyceride, and insulin concentrations were not different among treatments (Table 5). However, the NEFA concentration was higher in phytase-fed piglets with differences being significant only for the NC + 500 FTU treatment compared with NC and PC (P < 0.01; Table 5). In general, fasting blood plasma inositol concentration was 4- to 6-fold higher than levels analyzed in fed or portal blood plasma (Table 6). Fed blood plasma inositol concentration was greatest in the NC + 2,000 FTU treatment (P < 0.01; Table 6), and for this treatment, a tendency of a higher fasting blood plasma inositol concentration was observed also (P = 0.06; Table 6). However, inositol concentration in the portal vein blood was not different among treatments (Table 6).

Table 5.

Blood plasma metabolite concentrations of piglets fed diets different in P, Ca, and phytase levels at the end of the overall feeding period of 49 days¹

Item	PC	NC	NC + 500 FTU	NC + 2,000 FTU	SEM	P-value
Glucose, mg/dL	100.1	97.3	89.5	92.7	3.81	0.21
TG, mg/mL	0.24	0.29	0.26	0.28	0.02	0.32
NEFA, mmol/L	0.51b	0.59b	0.78a	0.68ab	0.06	0.02
Insulin, µg/L	0.023	0.021	0.025	0.023	0.002	0.55

¹Data are means of 10 replicate pens pooled from 2 pigs per pen in each treatment, means with different superscripts are different (P < 0.05). PC = positive control, NC = negative control, NC + 500 FTU and NC + 2,000 FTU = NC added with phytase added by 500 or 2,000 FTU/kg feed, respectively. TG = triglycerol.

Table 6.

Plasma inositol concentration in fasting, fed, and portal vein blood of piglets fed diets different in P, Ca, and phytase levels sampled at the end of the overall feeding period¹

Item	PC	NC	NC + 500FTU	NC + 2,000 FTU	SD	P-value
Portal plasma inositol, nmol/mL	4.17	6.38	4.78	8.53	5.24	0.14
Fasting plasma inositol, nmol/mL	25.05	23.00	21.88	37.49	16.32	0.06
Fed plasma inositol, nmol/mL	5.87b	5.97b	6.08ab	9.71a	6.67	0.01

¹Data are means of 10 replicate pens pooled from 2 pigs per pen in each treatment, means with different superscripts differ significantly (P < 0.05). PC = positive control, NC = negative control, NC + 500 FTU and NC + 2,000 FTU = NC added with phytase added by 500 or 2,000 FTU/kg feed, respectively. Values are means of each treatment with pigs were fasted overnight at day 47, and fasting blood were collected on the following day (day 48, n = 20). Fed blood and portal vein were collected on day 49 after slaughter, and (n = 20, 20, 20, 19 for PC, NC, 500, and 2,000 phytase, respectively, for fed blood), for portal vein (n = 18, 16, 20, 17 for PC, NC, 500, and 2,000 phytase, respectively).

Digesta Inositol Phosphates

Duodenal digesta InsP₅ and InsP₆ remained highest in PC and NC with no differences between these treatments. In contrast, phytase supplementation decreased duodenal InsP₆ (Table 7) and the 2,000 FTU/kg treatment significantly decreased InsP₅ level as well. Lower InsPs (InsP₄, InsP₃, and InsP₂) and myo-inositol concentrations were increased by NC + 500FTU, whereas 2,000 FTU/kg treatment increased (P < 0.01) duodenal myo-inositol level only (Table 7). Similarly, phytase supplementation decreased InsP₆ and InsP₅ concentrations in the ileal digesta. InsP₆ disappearance was greater with increasing phytase supplementation (90% and 97.9%) and cumulated InsP_6-2 disappearance was 93.6% in NC + 2,000 FTU and 72.8% in NC + 500 FTU compared with 25% to 31.5% in the control treatments (P < 0.01; Table 7). The concentration of InsP₅ in the ileal digesta was higher in the NC and PC treatments compared with treatments containing phytase (P < 0.001; Table 7). Feeding NC + 500 resulted in increased InsP₄, InsP₃, InsP₂, and myo-inositol abundance in the ileal digesta (P < 0.01; Table 7). However, in NC + 2,000 piglets, the increase of lower InsPs in the ileum was reversed compared with NC + 500 with the exception of InsP₂ and myo-inositol increased further (P < 0.001; Table 7). Although total InsP₅ and InsP₄ differed between treatments, these effects were not significant for single InsP₄ and InsP₅ isomers. Separation of InsP isomers1 of InsP₄ and InsP₅ based on existing standards and probability of peak separation in duodenal and ileal digesta of piglets fed diets different in dietary P/Ca and phytase levels is presented in Table 8. No significant differences in the concentrations of these isomers were observed.

Table 7.

Inositol phosphate concentrations (nmol/g DM) in duodenal and ileal digesta and disappearance (%) of different inositol phosphates down to the ileum of piglets fed diets different in P, Ca, and phytase levels sampled at the end of the overall feeding period of 49 days¹

Item	PC	NC	NC + 500 FTU	NC + 2,000 FTU	SEM	P-value
Duodenum, nmol/g DM
InsP6	7,615.7a	7,232.7a	2,014.3b	661.8c	657.8	<0.0001
InsP5	1,283.4a	1,160.3a	579.6a	200.6b	101.1	<0.0001
InsP4	234.7b	285.2b	970.8a	377.4b	78.9	0.0002
InsP3	313.6ab	278.9b	527.4a	317.4ab	32.1	0.02
InsP2	45.8b	97.1ab	200.9a	247.7a	21	<0.0001
myo-inositol	1,186.4b	685.9b	1,472.8ab	2,153.8a	178.2	0.0021
Ileum, nmol/g DM, nmol/g DM
InsP6	28,895.4a	29,265.0a	3,685.7b	1,012.8b	2,287.5	<0.0001
InsP5	5,451.2a	4,295.5a	1,013.7b	269.6b	380.5	<0.0001
InsP4	1,239.1b	925.1b	3,856.9a	1,708.8b	346.6	0.01
InsP3	369.9b	380.3b	1,535.6a	816.3ab	109.1	<0.0001
InsP2	210.9c	204.2c	2,118.6a	1,005.3b	162.7	<0.0001
myo-inositol	1,531.7ab	1,168.5b	6,092.3a	8,689.4a	547.9	0.001
Disappearance, %
InsP6	35.7b	25.4b	90.0a	97.9a	3.2	<0.0001
InsP6-2	31.5c	25.0c	72.8b	93.6a	4.1	<0.0001

¹Values are means of 10 replicate pens with digesta pooled from 2 pigs per pen, means with different superscripts differ significantly (P < 0.05). Disappearance rate was calculated using same formula for AID calculation. PC = positive control; NC = negative control; NC + 500 FTU and NC + 2,000 FTU = NC added with phytase added by 500 or 2,000 FTU/kg feed, respectively; InsP = inositol phosphates.

Table 8.

Separation of inositol phosphate (InsP) isomers¹ of InsP 4 and InsP 5 based on existing standards and probability of peak separation in duodenal and ileal digesta of piglets fed diets different in dietary P/Ca and phytase levels (nmol/g DM; n = 10)

Item	PC	NC	NC + 500 FTU	NC + 2,000 FTU	SE	P-value
Duodenum
Ins(1246/2346) P4	41	0	45	0	—	—
Ins(1234/1236) P4	84	44	77	187	17	0.18
Ins(1256/2345) P4	172	241	905	378	158	0.89
Ins(12346) P5	134	119	66	31	21	0.92
Ins(12356/12345) P5	424	394	310	153	69	0.24
Ins(23456/12456) P5	639	578	192	107	105	0.25
Ins(13456) P5	86	91	63	48	13.6	0.98
Ileum
Ins(1246/2346) P4	42	36	0	0	—	—
Ins(1234/1236) P4	314	270	426	398	65	0.12
Ins(1256/2345) P4	899	651	3,664	1,090	663	0.96
Ins(12346) P5	531	497	86	67	55	0.66
Ins(12356/12345)P5	1,761	1,318	647	174	263	0.47
Ins(23456/12456)P5	2,943	2,553	355	171	377	0.58
Ins(13456) P5	215	207	94	35	19	0.12

¹Numbers in brackets give the positioning of remaining P group in the 2 possible enantiomers that cannot be separated analytically.

PC = positive control, NC = negative control, NC + 500 FTU and NC + 2,000 FTU = NC added with phytase added by 500 or 2,000 FTU/kg feed, respectively.

Analysis of Muscle Plasma Membrane GLUT4 Concentration by Western Blot

Lowest longissimus dorsi muscle plasma membrane GLUT4 concentrations were found in NC piglets, but this was not different compared with NC + 500 and PC (P > 0.05). However, the highest GLUT4 concentration was found in the NC + 2,000 treatment (P < 0.01 compared with NC; Fig. 1). There was no treatment effect on protein kinase B (Akt) and phosphorylated-Akt concentration in muscle (Fig. 2).

Figure 1.

Longissimus dorsi muscle plasma membrane GLUT4 concentration in piglets fed diets different in P, Ca, and phytase levels sampled at the end of the overall feeding period of 49 d.

Figure 2.

Longissimus dorsi muscle Akt and phosphorylated-Akt (p-Akt) protein expression in piglets fed diets different in P, Ca, and phytase levels at the end of the overall feeding period of 49 d.

DISCUSSION

Growth Performance and Mineral Digestibility

Phytase is typically added to swine diets at 500 FTU/kg feed to release 0.3 to 1.7 g/kg available phosphorus (Augspurger et al., 2003; Wilcock and Walk, 2016). However, there is growing interest in supplementing higher phytase levels (>1,500 FTU/kg feed) to diets. This is considered as super-dosing if no further reduction in dietary P and Ca is implemented with high phytase supplementation. A deficiency of nonphytate phosphorus typically results in reduction in feed intake and exogenous phytase is known to increase feed intake in weaner pigs (Kornegay and Qian, 1996). The increased BW, ADG, ADFI, and feed efficiency in pigs fed phytase at 2,000 FTU/kg diet compared with the NC (and partly the PC) also agrees with reported effects of super-dosed phytase. The work by Kies et al. (2006) and Zeng et al. (2015) showed that phytase at 15,000 or 20,000 FTU/kg diet increased ADG and feed efficiency of weanling pigs. In a study by Nyannor et al. (2007), a linear effect of phytase was found on ADG and feed efficiency when phytase was added at 16,500; 33,000; and 49,500 FTU/kg feed. In a trial with pigs fed the same phytase as used in this trial, 2,500 FTU/kg feed, but not 500 FTU/kg feed, improved ADG and G:F (Laird et al., 2018). Therefore, it became necessary to investigate possible extraphosphoric effects of phytase. A probable explanation for the performance improvements in super-dosed fed pigs could be related to an increased nutrient digestibility, especially minerals, proteins, and AA. Adedokun et al. (2015) reported that phytase increased AID of DM, nitrogen, Ca, P, and several AA in cannulated pigs using 4 levels of phytase (0, 500, 1,000, and 2,000 FYT/kg feed). Similar results were reported by Zeng et al. (2016) demonstrating that a high dose of phytase (20,000 FTU/kg feed) increased; in addition to Ca and P, the digestibility of other minerals and trace minerals (Na, K, Mg, and Zn). Nyannor et al. (2007) found that phytase added at 16,500, 33,000, and 49,500 FTU/kg feed linearly increased digestibility of DM, GE, Ca, and P. In the present study, adding 2,000 FTU/kg feed to a P-deficient corn-soybean meal-based diet increased AID of Ca and P by 16.1% and 35.2%, respectively, compared with NC. Absorption of Ca and P was also greater in pigs that received 2,000 FTU/kg feed (158% and 148% compared with NC, respectively) with no difference compared with the PC. This suggested that 2,000 FTU/kg could recover the difference in Ca and P between the NC and PC, respectively, at the very least.

There was no significant difference between the PC and NC treatments in growth performance. This could mean P was not sufficiently reduced in the NC diet compared with PC to have any effect on performance although if could be the case that even the PC diets were still not adequate in P. This is, to some extent, supported by the higher feed intake in pigs that received phytase at a level of 2,000 FTU/kg diet. However, the calculated dietary standardized total tract digestibility (STTD) of P in the NC diet was 0.3% and 0.25% for phases 2 and 3, respectively. This was about 0.14% lower compared with the respective digestible P level in PC diets. According to NRC (2012), the STTD P requirements for pigs weighing between 7 to 11 kg and 11 to 25 kg are 0.4% and 0.33%, respectively. Alexander et al. (2008) reported that there was no difference in growth performance of growing pigs with a 20% reduction in total P compared with a P adequate diet. Santos et al. (2014) also reported the lack of a difference between PC and NC with a 0.15% reduction in available P in pigs with initial BW of about 23 kg. Therefore, these results suggest that P application met the requirement and was not deficient as to reduce growth performance in PC and NC treatment groups in the present study.

The daily Ca and P absorption in the PC piglets were 60% and 35% higher compared with the absorption calculated for NC + 500 FTU animals, whereas for the NC + 2,000 FTU treatment, similar daily Ca and P absorption was noted compared with the PC (6.4 vs. 6.3 g/d and 3.5 vs. 3.1 g/d for Ca and P, respectively). However, there was no growth performance difference between the PC and NC + 500 FTU treatments, suggesting that P was limiting and thus affecting growth. Despite the similar daily Ca and P absorption between the PC and 2,000 FTU/kg diet, the latter resulted in increased ADG and G:F. Thus, it can be speculated that the positive phytase effect of the NC + 2,000 FTU treatment on growth performance was not due to increased utilization of Ca and P.

Inositol Phosphates

Phytate degradation by phytase releases P and finally, InsP₁. The final conversion of InsP₁ to inositol is thought to be effected by intestinal phosphatases. In theory, phytase could degrade phytate completely through a stepwise dephosphorylation (InsP₆ → InsP₅ → InsP₄ → InsP₃ → InsP₂ → InsP₁). However, in vivo, hydrolysis of phytate is often incomplete, leading to a mixture of inositol phosphate esters (Humer et al., 2015; Zeller et al., 2015; Kühn et al., 2016; Laird et al., 2018). The increased formation of InsP₄ to InsP₂ by 500 FTU phytase/kg feed, and a numerically higher level of myo-inositol at the 2,000 FTU/kg phytase supplementation indicated that the super-dosed level of phytase resulted in a more complete breakdown of phytase compared with supplementation at 500 FTU/kg feed. These results demonstrate that intermediate inositol phosphates such as InsP₃ and InsP₂ can be further hydrolyzed by application of high phytase concentration. These results are in line with the work done by Kemme et al. (2006) who found that InsP₆ was mainly dephosphorylated to InsP₃ and InsP₂ with 900 FTU phytase/kg feed. Similarly, Laird et al. (2018) found a dose-dependent increase in gastrointestinal InsP degradation and myo-inositol concentration in the digesta when feeding either 500 or 2,500 FTU/kg feed of the same phytase as used in this trial.

Compared with NC, the concentration of InsP₆ in the duodenum was about 72% and 91% and in the ileum 87% and 96% lower in piglets fed the 500 and 2,000 FTU phytase/kg feed, respectively. The optimal pH for Escherichia coli phytase is around 4.5 (Igbasan et al., 2000), and the activity of mucosa phytase in pigs is very low (Jongbloed et al., 1992; Schlemmer et al. 2001). This, and the fact that phytate solubility decreases with pH above 4 to 4.5 (Angel et al., 2002), indicates that the stomach is the main site for InsP₆ hydrolysis by the phytase. The InsP₆ disappearance in the ileal digesta was 97.9% with 2,000 FTU/kg which is similar to previously reported analyses in broilers in which 12,500 FTU/kg of the same phytase resulted in 92% InsP₆ hydrolysis (Zeller et al., 2015). Laird et al. (2018) also found a dose-dependent effect of phytase with 67.6% and 78.1% InsP₆ disappearance at the terminal ileum when 500 and 2,500 FTU/kg phytase was used, respectively.

The ileal myo-inositol concentration was greatest in the NC + 2,000 FTU treatment and 7.4-fold higher compared with NC. In addition, the blood myo-inositol concentrations were highest in the NC + 2,000 FTU fed piglets. This was especially noticeable and significant in fed plasma myo-inositol concentration in the NC + 2,000 FTU treatment that was 1.6-fold higher compared with that in NC piglets. Others (Guggenbuhl et al., 2016; Laird et al., 2018) have also found increased plasma inositol concentrations with increased dietary phytase application. A study done by Cowieson et al. (2017) reported an immediate increase of plasma myo-inositol concentration after oral ingestion of myo-inositol, and 3,000 FTU phytase/kg feed significantly increased plasma myo-inositol 6 h after feed intake. Therefore, the increased myo-inositol concentration in the fed status at 2,000 FTU/kg was likely due to the released myo-inositol by phytase. However, only a numerical increase of portal vein myo-inositol concentration was observed in the present study with a large variability in the data set. This is in contrast to Laird et al. (2016), who found higher myo-inositol concentration in portal, compared with, peripheral pig plasma. However, such a difference was not seen in another trial carried out by Laird et al. (2018). Taken together, the inconsistency in the effect of phytase on plasma myo-inositol concentrations could be due to differences in experimental conditions in different studies, and perhaps due to variability in data sets. Factors such as timing of sampling relative to feeding, differences in dietary phytate, P, and Ca level could contribute to this variability. Additional experiments are needed to investigate effects of these factors, and the potential role that saturation of the transporter for myo-inositol (Na⁺-dependent transporter SMIT2; Sasseville et al., 2014) may play in regulating plasma myo-inositol concentrations.

Muscle Plasma Membrane GLUT4 Concentration

Longissimus dorsi muscle is a good tissue target for measuring energy sensing and insulin signaling (Manjarín et al., 2016). Insulin is a peptide hormone that activates PI 3-kinase signaling, which, through conversion of PIP₂ to PIP₃, recruits PDK1 to the plasma membrane with resultant phosphorylation of AKT and subsequent translocation of GLUT4 to the plasma membrane of skeletal muscle. As a metabolic precursor of PIP3, it can be speculated that myo-inositol may have distal insulin-like effects. Indeed, although myo-inositol supplementation reduces fasting blood glucose level in humans with type 2 diabetes (Pintaudi et al., 2016), there was no treatment effect of phytase on fasting blood glucose and insulin concentrations in this piglet study. The lack of a phytase effect on blood glucose concentration in the present study suggests that physiological blood glucose concentrations may not be under significant regulation by inositol levels generated in this study. In addition, treatment had no effect on the level of phosphorylated-Akt. However, GLUT4 concentration in muscle plasma membranes was increased by high phytase level (NC + 2,000 FTU, P < 0.05) supplementation. Chukwuma et al. (2016) reported that myo-inositol could promote rat muscle glucose uptake. Therefore, although phytase did not affect serum glucose and insulin levels in this study, possibly due to a complex metabolic regulation of these levels, it still might affect certain aspects of insulin signaling, such as an increase in membrane GLUT4 concentration. This effect has a potential to increase muscle tissue glucose uptake to support tissue growth. Dang et al. (2010) reported that oral injection of myo-inositol could induce the translocation of GLUT4 to the membrane of skeletal muscle in mice. In a subsequent study, the same group (Yamashita et al., 2013) showed that 1 g/kg oral myo-inositol raised plasma myo-inositol from <0.1 mM to 2.67 ± 0.72 mM also induced GLUT4 translocation. It is plausible, therefore, that the increased GLUT4 level in muscle plasma membranes of NC + 2,000 compared with NC piglet might be due to species variation in sensitivity to inositol and the increased myo-inositol release from phytate. However, there may also be an interaction between P and myo-inositol in regulating GLUT4 abundance as higher GLUT4 level was also found in the PC treatment, despite the lower blood plasma myo-inositol concentration in this treatment. Thus, the regulation of GLUT4 by myo-inositol and P requires further investigation.

We conclude that super-dosed phytase (2,000 FTU/kg feed) improved growth performance of weanling pigs and caused an almost complete hydrolysis of phytate (InsP₆, 97.8%; 93.6% InsP_6-2). The increased released of myo-inositol with high phytase supplementation may explain some of the extra phosphoric effect seen at this level, and this may suggest increased muscle glucose uptake. Future studies will be needed to better understand the effects of myo-inositol, its absorption, utilization, and regulatory factors on growth performance of weanling pigs.