Growth performance, bone mineralization, nutrient digestibility, and fecal microbial composition of multi-enzyme-supplemented low-nutrient diets for growing-finishing pigs

Abstract

A study evaluated the effects of adding multi-enzyme mixture to diets deficient in net energy (NE), standardized ileal digestible (SID) amino acids (AA), standardized total tract digestible (STTD) P, and Ca on growth performance, bone mineralization, nutrient digestibility, and fecal microbial composition of grow-finish pigs. A total of 300 pigs (initial body weight [BW] = 29.2 kg) were housed by sex and BW in 45 pens of 7 or 6 pigs and fed 5 diets in a randomized complete block design. Diets were positive control (PC), and negative control 1 (NC1) or negative control 2 (NC2) without or with multi-enzyme mixture. The multi-enzyme mixture supplied at least 1,800, 1,244, 6,600, and 1,000 units of xylanase, β -glucanase, arabinofuranosidase, and phytase per kilogram of diet, respectively. The PC was adequate in all nutrients. The NC1 diet had lower content NE, SID AA, STTD P, and Ca than PC diet by about 7%, 7%, 32%, and 13%, respectively. The NC2 diet had lower NE, SID AA, STTD P, and Ca than PC diet by 7%, 7%, 50%, and 22%, respectively. The diets were fed in four phases based on BW: Phase 1: 29–45 kg, Phase 2: 45–70 kg, Phase 3: 70–90 kg, and Phase 4: 90–120 kg. Nutrient digestibility, bone mineralization, and fecal microbial composition were determined at the end of Phase 1. Pigs fed PC diet had greater (P < 0.05) overall G:F than those fed NC1 diet or NC2 diet. Multi-enzyme mixture increased (P < 0.05) overall G:F, but the G:F of the multi-enzyme mixture-supplemented diets did not reach (P < 0.05) that of PC diet. Multi-enzyme mixture tended to increase (P = 0.08) femur breaking strength. Multi-enzyme mixture increased (P < 0.05) the ATTD of GE for the NC2 diet, but unaffected the ATTD of GE for the NC1 diet. Multi-enzyme mixture decreased (P < 0.05) the relative abundance of the Cyanobacteria and increased (P < 0.05) relative abundance of Butyricicoccus in feces. Thus, the NE, SID AA, STTD P, and Ca could be lowered by about 7%, 7%, 49%, and 22%, respectively, in multi-enzyme mixture-supplemented diets without negative effects on bone mineralization of grow-finish pigs. However, multi-enzyme mixture supplementation may not fully restore G:F of the grow-finish pigs fed diets that have lower NE and SID AA contents than recommended by 7%. Since an increase in content of Butyricicoccus in intestine is associated with improved gut health, addition of the multi-enzyme mixture in diets for pigs can additionally improve their gut health.

Lay Summary

A study evaluated the effects of supplementing a multi-enzyme mixture that contain fiber degrading enzymes and phytase on the growth performance, bone strength, and fecal microbial composition of grow-finish pigs fed corn-wheat-wheat bran-based diets. Five diets fed were a positive control (PC) diet, and two negative control (NC1 and NC2) diets without or with the multi-enzyme mixture. The PC diet was adequate in all nutrients and had greater available (net) energy and digestible amino content than NC1 diet or NC2 diet by 7%, and greater digestible P content than the NC1 diet (by 32%) and NC2 diet (by 50%). The diets were fed from 30 to 120 kg body weight. Feed efficiency for PC diet was greater than that for NC1 diet or NC2 diet. Multi-enzyme mixture improved feed efficiency, bone strength, and fecal concentration of beneficial micro-organisms (known as Butyricicoccus) for NC1 and NC2 diets. However, feed efficiency for the NC1 and NC2 diets did not reach that for the PC diet. Thus, multi-enzyme mixture can fully restore bone strength (but not feed efficiency) and improve health of grow-finish pigs fed corn-wheat-wheat bran-based diets in which available energy, amino acids, and P contents have been reduced by the afore-mentioned margins.

Supplementation of corn-wheat-wheat bran-based diets that are deficient in net energy and digestible amino acids and P with multi-enzyme mixture can result in improved feed efficiency, bone mineralization, energy and nutrient digestibility; and increased fecal concentration of Butyricicoccus microorganisms (that are beneficial with regard to gut health) of grow-finish pigs.

Introduction

Nutrient availability in non-ruminant diets that are based on corn, wheat, and their co-products can be improved through supplementation with phytase, xylanase, and β-glucanase that target phytic acid, arabinoxylans, and β-glucans, respectively (Woyengo and Nyachoti, 2011). For this reason, the net energy (NE), standardized ileal digestible (SID) amino acids (AA), and standardized total tract digestible (STTD) P values may be reduced in diets that are supplemented with the fore-mentioned enzymes without negative impacts on growth performance and bone mineralization of pigs. However, there is limited information on the effects of supplementing diets that are reduced in nutrients on growth performance and bone mineralization of pigs. It is well known that wheat and its co-products contain variable amount of phytase which is sensitive to heat treatment (Slominski et al., 2004; Woyengo et al., 2008; Woyengo and Nyachoti, 2011). The extent to which digestible P level in phytase-supplemented wheat-wheat bran-based diets can be reduced can vary depending on the amount of endogenous phytase in the diets (Schlemmer et al., 2001). In addition, the addition of fiber-degrading enzymes to diets can affect gut microbiota and gut health of pigs, and their effects may vary depending the ingredient composition of the diet and types of supplemental fiber-degrading enzymes used (Zhang et al., 2018). However, information is lacking on the effects of reducing the STTD P values in the corn-based diets that contain heat-treated wheat and wheat-bran, and supplementing the resulting low P diets with the product that contain phytase, xylanase, arabinofuranosidase, and β-glucanase on growth performance, bone mineralization, and nutrient digestibility of pigs. Additionally, fermentation of arabinoxylans compared with fermentation of other types of dietary fiber such as resistant starch yields volatile fatty acids that have a greater molar proportion of butyric acid that is associated with improved gut health of pigs (Nielsen et al., 2014). Xylanase can hydrolyze arabinoxylans into arabinoxylan-oligosaccharides that are readily fermented in the hindgut of pigs (Pedersen et al., 2015; Petry et al., 2021). However, information is lacking on the effects of supplementing corn-wheat-wheat bran-based diets for pigs with the product that contain phytase, xylanase, arabinofuranosidase, and β-glucanase on indicators of gut health.

It was hypothesized that supplementation of the low energy and AA diets with a combination of phytase, xylanase, arabinofuranosidase, and β-glucanase can improve growth performance and nutrient digestibility of pigs to those for the energy and AA adequate diets. It was also hypothesized that supplementation of corn-wheat-based diets can result in degradation of arabinoxylans into short fragments that are more fermentable leading to increased abundance of butyric acid producing microorganisms. The objective of this study was to evaluate the effects of reducing NE, digestible AA, and mineral contents in corn-based diets (containing heat-treated wheat and wheat bran) with and without phytase, xylanase, arabinofuranosidase, and β-glucanase on growth performance, bone mineralization, nutrient digestibility, and fecal microbial composition of growing-finishing pigs.

Materials and Methods

The experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee at South Dakota State University (#18-086E).

Experimental animals

A total of 300 growing pigs (initial body weight [BW] of 29.2 ± 5.1 kg; Large White-Landrace female × Large White-Hampshire male; Pig Improvement Company) were obtained from Swine Education and Research Facility, South Dakota State University (Brookings, SD). The pigs were then individually weighed and housed in 45 pens of 7 or 6 pigs (30 pens with 7 pigs per pen and 15 pens with 6 pigs per pen). Pens (1.8 × 2.4 m) had fully slated-concrete floors, metal spindle walls (1.0 m high), and solid polyvinyl chloride gates. Each pen was equipped with a cup drinker and a double-space dry feeder. Room temperature was maintained at 22 ± 2 °C throughout the experiment.

Experimental diets

Five experimental diets based on corn, wheat, and wheat bran were fed in this study (Table 1). The diets included a positive control diet (PC), and negative control diet 1 (NC1) and negative control diet 2 (NC2) without or with multi-enzyme mixture in 2 × 2 factorial arrangement. The multi-enzyme mixture was added to diets at 100 g/metric ton, supplying at least 1,800, 1,244, 6,600, and 1,000 units of xylanase, β-glucanase (that hydrolyses β-glucans that contains β 1-3 and β 1-4 linkages), arabinofuranosidase, and phytase per kilogram of diet, respectively. The PC diet was formulated to be adequate in all nutrients according to NRC (2012) recommendations. The NC1 diet was reduced in NE, SID AA, STTD P, and Ca contents by 7.0%, 7.0%, 0.08 percentage points, and 0.07 percentage points, respectively, compared to the PC diet. The NC2 diet was the same as the PC diet but with lower contents in NE, SID AA, STTD P, and Ca by 7.0%, 7.0%, 0.13 percentage points, and 0.12 percentage points, respectively. The reduction in NE value and nutrient content in the NC1 and NC2 diets was achieved by a partial replacement of corn, soybean meal (SBM), soybean oil, crystalline AA, calcium carbonate, and monocalcium phosphate in PC diet with wheat bran and soybean hulls. Ground wheat and wheat bran were heat-treated with steam at 71–76 °C for 5 s to inactivate endogenous phytase. After heat treatment, the steamed wheat and wheat bran were dried for 12 h to allow the moisture to evaporate before their inclusion in diets.

Table 1.

Ingredient and calculated chemical composition of the basal diets (%, as-fed basis)¹

Item	Phase 1: 29–45 kg BW			Phase 2: 45–70 kg BW			Phase 3: 70–90 kg BW			Phase 4: 90–120 kg BW
	PC	NC1	NC2	PC	NC1	NC2	PC	NC1	NC2	PC	NC1	NC2
Ingredients, %
Corn	59.17	45.64	46.15	62.13	49.64	50.39	68.11	54.91	55.41	68.82	56.06	56.35
Soybean meal 48%	13.93	11.05	10.97	11.47	7.66	7.47	6.31	3.23	3.15	5.62	2.50	2.50
Wheat2	10.00	10.00	10.00	10.00	10.00	10.00	10.00	10.00	10.00	10.00	10.00	10.00
Wheat bran2	10.00	25.00	25.00	10.00	25.20	25.20	10.00	25.40	25.40	10.00	25.60	25.60
Soybean hulls	1.00	3.00	3.00	1.35	3.10	3.00	1.50	3.00	3.00	2.00	3.00	3.00
Soybean oil	2.53	2.53	2.38	2.20	2.04	1.84	1.45	1.34	1.20	1.40	1.17	1.07
Calcium carbonate	1.02	1.00	0.98	0.93	0.91	0.89	0.82	0.81	0.79	0.73	0.71	0.63
Monocalcium phosphate	0.98	0.53	0.27	0.81	0.36	0.10	0.71	0.25	-	0.57	0.11	-
l-lysine HCl 98%	0.57	0.52	0.52	0.47	0.46	0.47	0.47	0.45	0.45	0.33	0.32	0.32
dl-methionine 99%	0.11	0.09	0.09	0.06	0.04	0.04	0.04	0.02	0.02
l-threonine	0.16	0.15	0.15	0.12	0.12	0.12	0.12	0.12	0.12	0.07	0.07	0.07
l-tryptophan	0.06	0.04	0.04	0.05	0.04	0.04	0.05	0.04	0.04	0.03	0.02	0.02
l-valine	0.05	0.04	0.04	-	-	-	-	-	-	-	-	-
Mineral premix3	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15	0.15
Vitamin premix4	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.05	0.05
Salt	0.23	0.23	0.23	0.23	0.23	0.23	0.24	0.24	0.24	0.24	0.24	0.24
Calculated nutrients
Crude protein, %	14.56	14.52	14.53	13.04	12.87	12.88	11.71	11.64	11.64	11.23	11.24	11.26
Ether extract, %	5.50	5.50	5.38	5.05	4.95	4.82	4.64	4.54	4.42	4.55	4.41	4.32
NE, kcal/kg	2,475	2,302	2,302	2,475	2,302	2,302	2,475	2,302	2,302	2,475	2,302	2,302
SID4 indispensable AA, %
Arg	0.781	0.765	0.765	0.716	0.674	0.671	0.575	0.554	0.554	0.558	0.537	0.538
His	0.342	0.327	0.327	0.322	0.300	0.300	0.280	0.264	0.265	0.275	0.260	0.260
Ile	0.510	0.474	0.474	0.474	0.424	0.422	0.396	0.357	0.357	0.387	0.348	0.348
Leu	1.169	1.056	1.059	1.128	1.000	1.000	1.037	0.922	0.925	1.027	0.914	0.917
Lys	0.980	0.911	0.911	0.850	0.791	0.791	0.730	0.679	0.679	0.610	0.567	0.567
Met	0.324	0.288	0.288	0.268	0.239	0.238	0.221	0.195	0.195	0.184	0.171	0.172
Met + Cys	0.550	0.512	0.512	0.480	0.466	0.446	0.420	0.391	0.391	0.380	0.365	0.366
Phe	0.618	0.580	0.580	0.580	0.527	0.526	0.498	0.458	0.458	0.488	0.448	0.449
Thr	0.590	0.549	0.549	0.520	0.484	0.484	0.460	0.428	0.428	0.400	0.372	0.372
Trp	0.170	0.158	0.158	0.150	0.140	0.140	0.130	0.121	0.121	0.110	0.102	0.102
Val	0.640	0.595	0.595	0.556	0.511	0511	0.480	0.448	0.448	0.471	0.439	0.440
Calcium, %	0.660	0.589	0.541	0.590	0.519	0.471	0.520	0.449	0.401	0.460	0.389	0.341
Total P, %	0.578	0.589	0.528	0.532	0.536	0.483	0.498	0.503	0.451	0.467	0.474	0.451
STTD5 P, %	0.310	0.230	0.176	0.270	0.190	0.136	0.240	0.160	0.107	0.210	0.130	0.106
Sodium, %	0.100	0.100	0.100	0.100	0.100	0.100	0.100	0.100	0.100	1.000	1.000	1.000

BW, body weight; PC, positive control diet; NC1, negative control diet 1 with lower in NE, standardized ileal digestible AA, standardized total tract digestible P, and Ca than PC diet by 7%, 7%, 0.11%–0.15%, and 0.18%–0.26%, respectively; and NC2, negative control diet 2 with lower in NE, standardized ileal digestible AA, standardized total tract digestible P, and Ca than PC diet by 7%, 7%, 0.26%–0.38%, and 0.43%–0.55%, respectively.

Wheat and wheat bran were heat-treated with steam before diet mixing to inactivate endogenous phytase.

Provided the following per kilogram of diet: 2,226 IU vitamin A, 340 IU vitamin D₃, 11.3 IU vitamin E, 0.01 mg vitamin B₁₂, 0.91 mg menadione, 2.04 mg riboflavin, 12.5 mg pantothenic acid, 11.3 mg niacin, 0.23 mg folic acid, 0.68 mg pyridoxine, 0.68 mg thiamine, and 0.04 mg biotin.

Provided the following per kilogram of diet: 75 mg Zn as ZnSO₄, 75 mg Fe as FeSO₄; 7 mg Cu as CuSO₄, and 20 mg Mn as MnSO₄.

Experimental design and procedure

The five diets were allotted to the 45 pens (9 pens/diet) within a randomized complete block design. The diet were fed in four phases based on BW: Phase 1: 29–45 kg, Phase 2: 45–70 kg, Phase 3: 70–90 kg, and Phase 4: 90–120 kg. Titanium dioxide (0.3%) was added as an indigestible marker in each diet during the last week of the first phase of feeding for determination of apparent total tract digestibility (ATTD) of dry matter (DM), organic matter (OM), crude protein (CP), ether extract (EE), crude ash, neutral detergent fiber (NDF), acid detergent fiber (ADF), P, and gross energy (GE) in Phase 1 diets by indicator method (Stein et al., 2007). During the experimental period, diets and fresh water were offered to pigs ad libitum. Pig BW and feed intake were determined by phase to calculate average daily gain (ADG), average daily feed intake (ADFI), and gain-to-feed ratio (G:F). At the end of the first phase of feeding, one pig (per pen) with BW that was close to the pen average BW was selected and then euthanized by captive bolt followed by exsanguination. Right and left femurs were excised from each euthanized pig and stored at −20 °C for determination of bone mineralization and bone breaking strength, respectively. Fresh fecal sample was collected directly from each pig’s rectum after euthanization. The collected fecal samples were immediately snap-frozen using liquid N and stored at −80 °C for later analysis of microbial composition. Fresh fecal samples were collected from each pen during the last 2 d of first phase of feeding and immediately stored frozen at −20 °C for the determination of apparent total tract digestibility (ATTD) of energy and nutrients.

Sample preparation and analyses

Femurs for determining bone ash were defleshed by autoclaving at 121 °C for 30 min, cleaned, and subsequently dried in an oven at 135 °C for 2 h. Fat was extracted from the dried bones using petroleum ether (E139-4, Fischer Scientific, Pittsburgh, PA) as solvent in a Jumbo Soxhlet extraction apparatus (Chemglass Life Sciences, Vineland, NJ). After the fat extraction, the bones were left in a fume hood for 24 h to allow the petroleum ether to evaporate. Bones were then dried in an oven at 135 °C for 2 h to determine their fat-free weight and ashed at 600 °C in a muffle furnace for 12 h for the determination of bone ash. Femurs for determining bone breaking strength were defleshed by scraping muscle tissues from the bones using kitchen knives. Maximal breaking load was measured using an MTS Insight 5 equipment (MTS, Eden Prairie, MN) at room temperature by subjecting each bone to a 3-point bending test (Tuner and Burr, 1993). Force was applied to the center of the bone held by supports 3.3 cm apart. The crosshead speed was set at 50 mm/min and the sample rate was 10 points/s. Final strength was determined from load-displacement curves.

Fecal samples were pooled by pen and air-dried in an oven at 60 °C for 4 d. The dried fecal samples together with diet samples were ground through a 0.75-mm screen in a centrifugal mill (model ZM200; Retsch GmbH, Haan, Germany). The ground samples were analyzed as follows: Phase 1 diets and fecal samples for DM, GE, N (N × 6.25 = CP), EE, P, ADF, NDF, and titanium contents; the samples were analyzed for DM by oven drying at 135 °C for 2 h (method 930.15), CP by a combustion procedure (method 990.03) as per AOAC (2007); and for ADF and NDF by van Soest method (model 3000, Labconco, Kansas city, MO). The GE was analyzed using an adiabatic bomb calorimeter (model AC600, Leco, St. Joseph, MI). Titanium dioxide in samples was determined by spectrophotometry (model Spectra MAX 190, Molecular Devices, Sunnyvale, CA) at 408 nm after ashing at 525 °C for 10 h (Myers et al., 2004). The EE was analyzed using a goldfisch fat extraction apparatus (model 35001, Labconco, Kansas city, MO). The samples were not acid-hydrolyzed before ether extraction. Phosphorus was analyzed using a spectrophotometer (Lambda 25, PerkinElmer, Waltham, MA) at 400 nm according to method 965.17 of AOAC (2007).

For analysis of the fecal microbial composition, total microbial DNA was extracted from fecal samples using the PowerFecal Pro DNA kit (QIAGEN, MD) following the manufacturer’s instructions. The quality of the DNA was determined using NanoDrop one (Thermo Fisher Scientific, DE) and quantified using Qubit Fluorometer 3.0 (Invitrogen, CA). The extracted DNA samples were used for the sequencing of the hypervariable V3-V4 regions of the bacterial 16S rRNA using the Illumina MiSeq platform. The library preparation for metagenomic sequencing was performed using 0.3 ng of DNA with a Nextera XT library preparation kit (Illumina, San Diego, CA) and sequenced on the MiSeq platform. The variations in bacterial communities within the feces of pigs were analyzed using 16S rRNA microbial community analysis package in Quantitative Insights into Microbial Ecology framework (QIIME, Version 2.0). Briefly, 45 samples were quality filtered, demultiplexed, and denoised using dada2. The alpha diversity indices including Ace, Chao1, Shannon, and Simpson diversity index were used to estimate the α-diversity index and the Bray NMDS dissimilarity index was used to calculate the β-diversity index. The taxonomy was assigned to Amplicon Sequence Variants using the dada2 package to implement the naïve Bayesian classifier method against GreenGenes (http://greengenes.lbl.gov). The operating taxonomic units (OTUs) were clustered with 97% similarity cutoff using USEARCH and Chimeric sequences and subsequently filtered out to obtain OTUs for species classification. The sequences have been deposited into the NCBI database, accession number PRJNA719859.

Calculations and statistical analysis

The ATTD of energy and nutrients was calculated using the indicator method (Stein et al., 2007). Data were subjected to analysis of variance using the MIXED procedure (SAS Inst. Inc., Cary, NC). The pen was considered as the experimental unit. The model included diet, sex, diet × sex interaction, and initial body weight, which was a covariate. Main effects of negative control diet type and multi-enzyme mixture and their interactions were determined. Means were separated by the probability of difference in order to compare PC diet with other diets and describe interactions (when they were significant) between negative control diet type and multi-enzyme mixture. To test the hypotheses, P < 0.05 was considered significant. If pertinent, trends (0.05 < P ≤ 0.10) are also reported.

Results

The analyzed CP and P values for the Phase 1 diets in Table 2 were similar to the calculated CP values for the Phase 1 diets in Table 1. Data on effect of multi-enzyme mixture supplementation on the growth performance of pigs are presented in Table 3. Pigs fed PC diet had greater (P < 0.05) BW than pigs fed NC1 diet or NC2 diet at the end of Phases 1, 3, and 4. Also, pigs fed PC diet had greater (P < 0.05) ADG and G:F than those of pigs fed NC1 diet or NC2 diet during Phase 1, 3, and entire study period. Pigs fed NC1 diet had greater (P < 0.05) BW than pigs fed NC2 diet at the end of Phase 4. However, the NC1 and NC2 diets did not differ in ADG, ADFI and G:F. The NC diet type and multi-enzyme mixture supplementation did not interact on BW, ADG, and ADFI. However, NC diet type and multi-enzyme mixture supplementation interacted (P < 0.05) on G:F during Phase 1 such that supplementation of NC1 diet with multi-enzyme mixture did not affect the G:F, whereas supplementation of the NC2 diet with multi-enzyme mixture increased (P < 0.05) the G:F of pigs. The addition of multi-enzyme mixture to the NC2 diet increased (P < 0.05) the G:F overall study period. The overall G:F value for multi-enzyme mixture supplemented diets was lower (P < 0.05) than that for the PC diet.

Table 2.

Analyzed composition of Phase 1 diets as fed

Item	Diets1
	PC	NC1	NC2
Dry matter, %	86.08	86.48	86.41
Organic matter, %	94.84	94.84	94.76
Crude ash, %	5.16	5.17	5.24
Ether extract, %	7.25	7.29	6.07
Gross energy, kcal/kg	3,967	4,040	3,995
Crude protein, %	15.36	14.97	15.20
Neutral detergent fiber, %	14.71	21.51	20.64
Acid detergent fiber, %	4.53	7.00	6.47
Total P, %	0.59	0.62	0.56

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2.

Table 3.

Effect of dietary treatments on growth performance

Item1	Diets2					SEM	P-value3
	PC	NC1	NC2	NC1+E	NC2+E		Diet	NC	E	NC×E
Body weight, kg
Initial	29.2	29.1	29.2	29.2	29.2	1.73	–	–	–	–
Phase 1	45.8a	43.7b	43.7b	43.7b	45.8a	0.47	<0.01	0.04	0.04	0.07
Phase 2	68.6	66.6	65.7	66.8	66.8	0.73	0.10	0.73	0.40	0.71
Phase 3	90.6a	85.9bc	83.7c	86.8b	86.6b	0.84	<0.01	0.54	0.12	0.66
Phase 4	120.4a	114.6b	111.0c	115.2b	114.1bc	1.17	<0.01	0.31	0.23	0.68
Average daily gain, kg
Phase 1	0.837a	0.724b	0.727b	0.726b	0.828a	0.023	<0.01	0.04	0.04	0.06
Phase 2	0.941	0.956	0.913	0.960	0.870	0.026	0.11	0.08	0.68	0.40
Phase 3	0.956a	0.837bc	0.786c	0.874b	0.862bc	0.028	<0.01	0.52	0.09	0.74
Phase 4	1.105	1.063	1.018	1.051	1.016	0.029	0.21	0.28	0.91	0.95
Overall	0.953a	0.891b	0.865b	0.907ab	0.898b	0.017	<0.01	0.06	0.16	0.11
Average daily feed intake, kg
Phase 1	1.888	1.792	1.799	1.766	1.880	0.031	0.07	0.66	0.55	0.98
Phase 2	2.344	2.387	2.395	2.413	2.386	0.051	0.89	0.96	0.76	0.66
Phase 3	2.979	2.906	2.891	3.0052	2.921	0.063	0.67	0.78	0.37	0.56
Phase 4	3.920	4.015	3.973	3.910	3.944	0.084	0.90	0.85	0.76	0.94
Overall	2.765	2.759	2.782	2.790	2.774	0.042	0.98	0.60	0.15	0.46
Gain:Feed, kg/kg
Phase 1	0.441a	0.406b	0.407b	0.412b	0.467a	0.009	<0.01	<0.01	<0.01	<0.01
Phase 2	0.403a	0.400a	0.379ab	0.398a	0.364b	0.009	0.01	<0.01	0.40	0.45
Phase 3	0.321a	0.290b	0.272b	0.291ab	0.297ab	0.011	0.04	0.64	0.24	0.31
Phase 4	0.284	0.266	0.258	0.269	0.259	0.007	0.09	0.21	0.85	0.99
Overall	0.363a	0.340bc	0.329c	0.342bc	0.347b	0.004	<0.01	<0.01	0.03	0.52

There were 7 or 6 pigs per pen in Phase 1 and 6 or 5 pigs per pen in Phases 2 to 4.

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2; NC1+E, negative control diet 1 plus multi-enzyme mixture; NC2+E, negative control diet 2 plus multi-enzyme mixture.

NC, main effects of negative control diet type; E, main effect of multi-enzyme mixture; NC×E, interaction between negative control diet type and multi-enzyme mixture.

Within a row, means without a common superscript differ (P < 0.05).

Data on effects of multi-enzyme mixture supplementation on bone DM, ash, and bone breaking strength are presented in Table 4. Pigs fed PC diet had greater (P < 0.05) crude ash content (%) of femur than that of pigs fed NC2 diet, but not NC1 diet. Pigs fed NC1 diet had greater (P < 0.05) femur bone crude ash content (%) than that of pigs fed NC2 diet. The PC, NC1, and NC2 diets did not differ in femur bone strength. The NC diet type and multi-enzyme mixture supplementation did not interact on bone DM, crude ash, and breaking strength. However, multi-enzyme mixture supplementation increased (P < 0.05) the DM content in femur of pigs regardless of NC diet type. Also, multi-enzyme mixture supplementation tended to increase (P = 0.078) bone breaking strength regardless of NC diet type. The femur bone DM, crude ash, and breaking strength values for multi-enzyme mixture supplemented diets did not differ from that of PC diet.

Table 4.

Effect of dietary treatments on bone mineralization

Item	Diets1					SEM	P-value2
	PC	NC1	NC2	NC1+E	NC2+E		Diet	NC	E	NC×E
Femur content
Dry matter, %	73.76	66.78	63.62	74.21	70.11	3.42	0.16	0.33	0.04	0.96
Crude ash, %	42.30a	39.92a	33.73b	41.41a	39.13ab	2.05	0.04	0.06	0.16	0.43
Crude ash, g	27.0	26.5	26.8	27.2	27.4	1.31	0.99	0.68	0.40	0.76
Femur breaking strength
Max load, N	2,002	1,960	1,821	2,162	1,998	117	0.38	0.46	0.08	0.82
Max stress, Mpa	1,173	1,148	1,067	1,266	1,171	69	0.38	0.46	0.08	0.82

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2; NC1+E, negative control diet 1 plus multi-enzyme mixture; NC2+E, negative control diet 2 plus multi-enzyme mixture.

NC, main effects of negative control diet type; E, main effect of multi-enzyme mixture; NC×E, interaction between negative control diet type and multi-enzyme mixture.

Within a row, means without a common superscript differ (P < 0.05).

Data on effects of multi-enzyme mixture supplementation on ATTD of DM, OM, crude ash, EE, NDF, ADF, CP, GE, and P are presented in Table 5. Pigs fed the PC diet had greater (P < 0.05) ATTD of DM, OM, CP, GE, EE, and crude ash than those of pigs fed NC1 diet or NC2 diet. Also, pigs fed the PC diet had greater (P < 0.05) ATTD of P than that of pigs fed NC2 diet, and tended to have greater (P < 0.10) ATTD of P than that of pigs fed NC1 diet. The NC1 and NC2 diets did not differ in ATTD of DM, OM, CP, GE, EE, and crude ash. The ATTD of P for NC1 diet was numerically greater (P < 0.10) than that for NC2 diet. The NC diet type and multi-enzyme mixture supplementation interacted (P < 0.05) on ATTD of CP and crude ash such that the addition of multi-enzyme mixture did not affect the ATTD of CP and crude ash for NC1 diet, but increased (P < 0.05) ATTD of CP and crude ash for NC2 diet. Also, the NC diet type and multi-enzyme mixture supplementation tended to interact (P < 0.10) on ATTD of DM, GE, and OM such that the addition of multi-enzyme mixture did not affect the ATTD of DM, GE, and OM for NC1 diet, but increased (P < 0.05) ATTD of DM, GE, and OM for NC2 diet. The ATTD of DM, GE, CP, OM, and crude ash values for multi-enzyme mixture supplemented NC2 diet did not differ from that of the PC diet. The NC diet type and multi-enzyme mixture supplementation did not interact (P < 0.10) on ATTD of P. Multi-enzyme mixture supplementation increased (P < 0.05) on ATTD of P regardless of NC diet type. The ATTD of P values for multi-enzyme mixture supplemented diets did not differ from that of the PC diet.

Table 5.

Effect of dietary treatments on apparent total tract digestibility (ATTD) of energy and nutrients

ATTD, %	Diets1					SEM	P-value2
	PC	NC1	NC2	NC1+E	NC2+E		Diet	NC	E	NC×E
Dry matter	83.99a	78.73bc	78.20c	77.02c	81.37ab	1.56	<0.01	0.19	0.60	0.06
Gross energy	83.91a	79.41bc	78.87c	77.35c	81.43ab	1.53	<0.01	0.21	0.91	0.06
Crude protein	82.12a	76.30b	75.72b	73.90b	79.87a	1.66	<0.01	0.05	0.52	<0.01
Organic matter	87.58a	83.42bc	83.10c	82.12c	85.44ab	1.23	<0.01	0.17	0.63	0.06
Ether extract	82.40a	79.17ab	74.98b	76.91b	76.15b	2.09	0.01	0.11	0.69	0.37
Crude ash	61.27a	51.75b	49.49b	48.72b	59.61a	3.29	<0.01	0.20	0.26	0.04
Neutral detergent fiber	60.96	59.64	59.14	60.14	64.22	3.20	0.45	0.64	0.25	0.35
Acid detergent fiber	42.59	43.30	40.46	40.96	42.19	5.17	0.98	0.61	0.86	0.65
Phosphorus	40.13ab	30.65bc	21.90c	37.05ab	41.42a	3.61	0.02	0.60	<0.01	0.12

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2; NC1+E, negative control diet 1 plus multi-enzyme mixture; NC2+E, negative control diet 2 plus multi-enzyme mixture.

NC, main effects of negative control diet type; E, main effect of multi-enzyme mixture; NC×E, interaction between negative control diet type and multi-enzyme mixture.

Within a row, means without a common superscript differ (P < 0.05).

Shannon index and Simpson index of fecal microbiome of pigs fed PC diet were lower (P < 0.05) than those of pigs fed NC2 diet (Table 6). The NC diet type and multi-enzyme mixture supplementation tended to interact (P < 0.10) on Shannon index and Simpson index of fecal microbiome such that the addition of multi-enzyme mixture tended to increase (P < 0.10) the Shannon index and Simpson index of fecal microbiome for NC1 diet, but tended to decrease (P < 0.10) the Shannon index and Simpson index of fecal microbiome for NC2 diet. The NC diet type and multi-enzyme mixture supplementation did not affect the alpha diversity indices of fecal microbiome in pigs. The β-diversity of fecal microbiome differed (P < 0.05) among dietary treatments (Figure 1). Data on effects of multi-enzyme mixture supplementation on relative abundance of fecal bacteria at phylum and genus levels are presented in Tables 7 and 8, respectively. Pigs fed the NC2 diet tended to have greater (P < 0.10) relative abundance of Proteobacteria phylum in feces than pigs fed NC1 diet regardless of multi-enzyme mixture supplementation (Table 7 and Figure 2). Multi-enzyme mixture supplementation decreased (P < 0.05) the relative abundance of Cyanobacteria phylum in feces regardless of NC diet type. At genus level, the relative abundance of Lactobacillus in fecal microbiome of pigs fed PC diet tended to be lower (P = 0.05) than that of pigs fed NC1 diet (Table 8). Pigs fed the NC2 diet had a greater (P < 0.05) relative abundance of Faecalibacterium and tended to have greater (P < 0.10) relative abundance of Roseburia and Campylobacteria and have less (P < 0.10) relative abundance of Methanosphaera in feces than pigs fed the NC1 diet regardless of multi-enzyme mixture supplementation. Multi-enzyme mixture supplementation decreased (P < 0.05) relative abundance of Lactobacillus and tended to decrease (P < 0.10) relative abundance of Bifidobacterium, but increased (P < 0.05) relative abundance of Butyricicoccus in feces regardless of diet type. The relative abundance of Butyricicoccus in feces for multi-enzyme mixture supplemented NC diets did not differ from that for PC diet.

Table 6.

Alpha diversity indices of fecal microbiome of pigs fed diets for Phase 1

Item	Diets1					SEM	P-value2
	PC	NC1	NC2	NC1+E	NC2+E		Diet	NC	E	NC×E
Ace	322.8	354.0	366.8	377.1	363.9	20.3	0.39	0.99	0.55	0.45
Chao1	322.7	354.5	367.1	377.7	363.9	20.3	0.38	0.97	0.56	0.44
Shannon	3.66b	3.96ab	4.36a	4.22a	3.94ab	0.16	0.04	0.86	0.81	0.07
Simpson	0.89b	0.92ab	0.95a	0.94a	0.92ab	0.01	0.04	0.87	0.81	0.09

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2; NC1+E, negative control diet 1 plus multi-enzyme mixture; NC2+E, negative control diet 2 plus multi-enzyme mixture.

NC, main effects of negative control diet type; E, main effect of multi-enzyme mixture; NC×E, interaction between negative control diet type and multi-enzyme mixture.

Within a row, means without a common superscript differ (P < 0.05).

Figure 1.

Effects of experimental diets on beta diversity index (Bray–Curtis index) of fecal microbiome of pigs.

Table 7.

The distribution of bacterial communities at phylum level of pigs fed diets for Phase 1

Item, %	Diets1					SEM	P-value2
	PC	NC1	NC2	NC1+E	NC2+E		Diet	NC	E	NC×E
Firmicutes	83.63	85.95	79.52	79.40	83.24	3.02	0.48	0.87	0.49	0.11
Bacteroidetes	12.48	9.12	14.52	15.05	12.67	2.56	0.50	0.73	0.30	0.15
Spirochaetes	1.69	0.94	2.22	1.82	2.22	0.93	0.86	0.49	0.63	0.58
Verrucomicrobia	0.06	0.04	1.15	1.72	0.04	0.90	0.55	0.74	0.75	0.13
Actinobacteria	0.79	2.72	0.90	0.69	0.72	0.69	0.18	0.31	0.20	0.27
Proteobacteria	0.41	0.17	0.47	0.27	0.28	0.12	0.44	0.09	0.79	0.13
Euryarchaeota	0.62	0.72	0.88	0.65	0.59	0.26	0.94	0.80	0.67	0.87
Cyanobacteria1	0.12	0.21	0.14	0.11	0.09	0.03	0.11	0.26	0.03	0.46
Tenericutes2	0.01	0.02	0.05	0.02	0.01	0.01	0.19	0.79	0.26	0.11
TM7	0.08	0.04	0.05	0.07	0.07	0.02	0.36	0.69	0.10	0.70
Chlamydiae	0.09	0.06	0.07	0.19	0.05	0.06	0.46	0.29	0.31	0.19
Elusimicrobia3	0.01	0.01	0.04	0.01	0.00	0.01	0.31	0.41	0.22	0.18

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2; NC1+E, negative control diet 1 plus multi-enzyme mixture; NC2+E, negative control diet 2 plus multi-enzyme mixture.

NC, main effects of negative control diet type; E, main effect of multi-enzyme mixture; NC×E, interaction between negative control diet type and multi-enzyme mixture.

Within a row, means without a common superscript differ (P < 0.05).

Table 8.

The distribution of bacterial communities at genus level of pigs fed diets for Phase 1

Item, %	Diets1					SEM	P-value2
	PC	NC1	NC2	NC1+E	NC2+E		Diet	NC	E	NC×E
Streptococcus	41.01	35.86	28.53	34.25	39.70	5.20	0.48	0.87	0.59	0.34
Clostridium	15.52	13.58	13.40	16.11	15.18	2.58	0.93	0.75	0.34	0.92
Lactobacillus	12.99b	24.01a	15.29ab	11.25b	10.09b	2.65	0.05	0.33	0.01	0.37
Prevotella	9.47	6.08	12.36	11.90	10.93	2.45	0.39	0.35	0.25	0.15
Ruminococcus	5.47	5.72	5.91	6.22	6.39	0.97	0.97	0.78	0.72	0.97
Treponema	2.39	1.32	4.20	3.06	3.61	1.64	0.77	0.40	0.68	0.50
Gemmiger	1.59	1.70	2.03	1.72	1.55	0.34	0.87	0.90	0.51	0.47
Barnesiella	1.03	1.28	1.18	1.32	1.11	0.48	0.99	0.67	0.90	0.97
Akkermansia	0.02	0.03	2.08	3.04	0.04	1.72	0.59	0.78	0.78	0.14
Butyricicoccus	1.05abc	0.24c	0.75bc	1.56ab	1.73a	0.33	0.02	0.41	0.01	0.58
Coprococcus	0.96	0.53	0.72	0.96	0.70	0.17	0.36	0.85	0.21	0.13
YRC22 3	0.28b	0.4b	1.48a	0.90ab	0.48ab	0.37	0.13	0.48	0.73	0.09
Succinispira	0.65	0.29	1.32	0.66	0.57	0.43	0.53	0.34	0.93	0.32
Methanobrevibacter	0.30	0.33	1.53	0.53	0.56	0.55	0.49	0.33	0.70	0.48
Eubacterium 2	0.50ab	1.53a	0.44b	0.41b	0.44b	0.37	0.16	0.21	0.19	0.21
Oscillospira	0.82	0.60	0.82	0.55	0.41	0.24	0.70	0.93	0.25	0.43
Faecalibacterium	0.48	0.37	0.76	0.46	0.73	0.15	0.31	0.03	0.78	0.91
Roseburia	0.29	0.41	0.84	0.40	0.75	0.20	0.27	0.08	0.92	0.98
Methanosphaera	0.67	0.71	0.31	0.58	0.46	0.15	0.32	0.08	0.98	0.24
Olsenella	0.14	2.31	0.16	0.10	0.14	0.80	0.21	0.31	0.27	0.28
Vestibaculum	0.62	0.32	0.94	0.30	0.37	0.25	0.32	0.22	0.23	0.26
Dorea	0.52	0.46	0.47	0.41	0.54	0.18	0.98	0.61	0.75	0.52
Bulleidia 5	0.32b	0.55a	0.35ab	0.35ab	0.37ab	0.08	0.28	0.22	0.24	0.23
Peptococcus	0.26	0.19	0.50	0.35	0.28	0.12	0.46	0.27	0.99	0.22
Sharpea	0.15	0.66	0.16	0.06	0.30	0.23	0.40	0.69	0.43	0.10
Bifidobacterium 6	0.31ab	0.39a	0.26ab	0.12b	0.23ab	0.09	0.36	0.95	0.07	0.22
Enterococcus	0.13	0.18	0.34	0.28	0.25	0.09	0.55	0.51	0.73	0.56
Catenibacterium	0.09	1.00	0.01	0.11	0.12	0.36	0.28	0.28	0.39	0.28
Desulfovibrio	0.28	0.12	0.33	0.21	0.23	0.13	0.83	0.28	0.79	0.51
Anaerorhabdus	0.12	0.17	0.26	0.24	0.27	0.12	0.88	0.55	0.55	0.96
Sporobacter	0.32	0.01	0.32	0.19	0.22	0.14	0.52	0.13	0.48	0.38
Campylobacter	0.34	0.07	0.34	0.12	0.12	0.11	0.25	0.08	0.26	0.09
Selenomonas	0.09	0.21	0.76	0.15	0.24	0.09	0.58	0.40	0.72	0.81
Bacteroides	0.16	0.11	0.28	0.15	0.34	0.09	0.32	0.18	0.78	0.86

PC, positive control diet; NC1, negative control diet 1; NC2, negative control diet 2; NC1+E, negative control diet 1 plus multi-enzyme mixture; NC2+E, negative control diet 2 plus multi-enzyme mixture.

NC, main effects of negative control diet type; E, main effect of multi-enzyme mixture; NC×E, interaction between negative control diet type and multi-enzyme mixture.

YRC22 for NC2 was greater than that for PC (P = 0.03).

Eubacterium for NC1 was greater than those for NC1+E (P = 0.04) and NC2+E (P = 0.04).

Bulleidia for NC1 was greater than that for PC (P = 0.04).

Bifidobacterium for NC1 was greater than that for NC1+E (P = 0.04).

Within a row, means without a common superscript differ (P < 0.05).

Figure 2.

Effects of experimental diets on heatmap at phylum level of fecal microbiome of pigs.

Discussion

The growth performance for PC diet was greater than that for NC1 diet or NC2 diet, implying that at least one of these diet components (NE, SID AA, Ca, and digestible P) was limiting growth performance. The similarity between NC1 diet and NC2 diet with regard to the growth performance of pigs implies that the reduction in Ca and digestible P did not impact ADG and feed efficiency of pigs. Similarly, Woyengo et al. (2008) did not observe reduction in ADG of grow-finish pigs fed wheat-based diets due to reduction in Ca and available P contents in PC diet by 0.10 and 0.07 percentage points, respectively.

Multi-enzyme mixture supplementation to NC2 diet improved G:F for Phase 1 and hence for the entire study to that of PC diet, but not in NC1 diet, which could have been due to the increase in ATTD of nutrients at the end of Phase 1 of feeding due to addition of multi-enzyme to NC2 diet, but not NC1 diet. The overall G:F for the multi-enzyme mixture supplemented NC1 diet or NC2 diet was lower than that for PC diet, implying that the multi-enzyme mixture did not fully correct for the nutrient deficiency in the NC2 diet. Thus, the differences between the current study and that of Jerez-Bogota et al. (2020) with regard to the ability of the multi-enzyme mixture to restore growth performance of pigs fed nutrient-deficient diets could be due to differences in magnitude of reduction in NE and SID AA in diets (−7% vs. −5%) to which the multi-enzyme mixture was supplemented.

Femur ash for PC diet did not differ from that for NC1 diet but differed from that for NC2 diet, implying that the reduction in Ca and digestible P by 0.12 and 0.13 percentage points, respectively, was sufficient to affect bone mineralization of pigs. Jerez-Bogota et al. (2020) reported reduction in femur ash of grow-finish pigs due to reduction in Ca and digestible P by 0.12 and 0.13 percentage points, respectively, in diets that contained wheat and wheat bran. In the current study, but not in the study of Jerez-Bogota et al. (2020), dietary wheat and wheat bran were subjected to heat treatment with the goal of inactivating endogenous phytase. Endogenous phytase can increase the digestibility of Ca and P. Thus, we speculated that the reduction in Ca and digestible P by 0.07 and 0.08 percentage points, respectively, would result in reduced femur ash content due to presence of insignificant amount of endogenous phytase in the basal diet. Thus, it appears that the reduction in dietary Ca and digestible P by 0.07 and 0.08 percentage points, respectively, is not sufficient to affect bone mineralization of pigs fed wheat and wheat bran containing diets regardless of whether the latter 2 feedstuffs have subjected to heat treatment or not. The difference between NC1 diet and NC2 diet with regard to femur ash but not growth performance could be attributed to the fact that dietary requirement of Ca and digestible P for bone mineralization is greater than that for growth. The similarity between PC diet and multi-enzyme mixture-supplemented NC2 diet with regard to femur ash imply that the multi-enzyme mixture was able to correct for Ca and digestible P deficiencies in the NC2 diet. As expected, the ATTD of GE, CP, and P for the PC diet was higher than those for the NC1 diet or NC2 diet. Also, as expected the ATTD of GE and CP for the NC1 diet did not differ from those for the NC2 diet. The tendency of ATTD of P for the NC2 diet to be lower than that for the NC1 diet is attributed to the fact that the former was formulated to be lower in digestible P than the latter.

The ATTD of EE was unaffected by NC diet type and multi-enzyme mixture supplementation, which could partly be attributed to the fact that diet and fecal samples were not acid-hydrolyzed before ether extraction. Ether extract analysis of samples that have not been acid-hydrolysed may result in EE values that are not so accurate (NRC, 2012). Multi-enzyme mixture supplementation improved ATTD of GE and CP for NC2 diet, but not for NC1 diet, which was surprising. It had been assumed that the multi-enzyme mixture would have similar effects on NC1 diet and NC2 diet with regard to ATTD of GE and CP because these diets were formulated to similar NE and SID AA levels. Because the NC1 diet and NC2 diet differed in Ca and digestible P, the multi-enzyme mixture supplementation interacted with these minerals on nutrient digestibility. Thus, there is need to do further research to establish whether or not such interactions exist.

Partial replacement of corn and SBM with wheat or wheat bran in diets for pigs resulted in increased alpha diversity of gut microbiota in cecum (Zhang et al., 2018), colon (Kraler et al., 2016), or rectum (Zhao et al., 2018), implying that dietary wheat bran increased species and richness of microorganisms in the hindgut of pigs. Moreover, Kraler et al. (2016) reported that partial replacement of corn, barley, and SBM with 15% wheat bran in weaning pigs’ diet resulted in increased Chao1 index of colonic microbiota. Zhang et al. (2018) reported that pigs fed wheat-based diet containing 25% wheat co-products had greater Shannon and Simpson indices representing the beta diversity of microbiota, for cecal microbiota than pigs fed corn-corn DDGS-based diet. Corn-based diet has a greater proportion of insoluble arabinoxylans and lower proportion of soluble arabinoxylans than wheat-based diet (Ndou et al., 2015) because arabinoxylans in corn forms more linkages with other non-starch polysaccharides (NSP) components (arabinoxylan, lignin) (Fengler and Marquardt, 1988; Zhang et al., 2018). Thus, wheat NSP are more fermentable in the hindgut of pigs, leading to higher microbial diversity in the hindgut (Zhang et al., 2018). An increase in levels of Ca and P in diets for pigs can result in reduced growth of some microorganisms in gastrointestinal tract (Metzler-Zebeli et al., 2011) because sources of these minerals in diets (such as limestone and mono-calcium phosphate) have high gastric acid binding capacity (Lawlor et al., 2005) that results in modified gastrointestinal tract environment and hence growth of microorganisms in the gastrointestinal tract. In the current study, the PC diet contained more corn, Ca and P, and less wheat bran than NC1 diet or NC2 diet, whereas the NC2 diet contained less Ca and P than the NC1 diet. Thus, the higher microbial diversity in feces of pigs fed NC2 diet than in feces of pigs fed PC diet could be attributed to the higher content of fermentable fiber and lower content of Ca and P in the NC2 diet than in the PC diet.

The presence of bacteria of genus Faecalibacterium in gastrointestinal tract of animals is positively associated with good gut health partly because these microorganisms are involved in butyric acid production (Duncan et al., 2002; Louis et al., 2004). In the current study, the relative abundance of Faecalibacterium for NC2 diet was greater than that for the NC1 diet, implying that reduction in dietary Ca and P resulted in increased relative abundance of Faecalibacterium. However, the actual mechanisms by which dietary Ca and P could affect the relative abundance of Faecalibacterium have not been well established. Metzler-Zebeli et al. (2011) reported that reduction in dietary Ca and P content in diets for growing pigs resulted in increased gene copy numbers of butyrate-production pathway genes in the stomach, but not in cecum or colon, implying that dietary Ca and P had limited effect on the butyric acid producing bacteria in the hindgut and hence feces.

In the current study, Campylobacter is the genus of microorganisms under Proteobacteria phylum whose relative abundance in feces tended to be increased by decreasing the dietary level of Ca and P. The results from the current study are in agreement with those from the study of Mann et al. (2014), who reported that a decrease in levels of Ca and P in diets for weaned pigs from excess (190% of requirement) to adequate (100% of requirement) resulted in increased abundance of Campylobacter in the colon by 5.8 units. However, like Faecalibacterium, the actual mechanisms by which dietary Ca and P could affect the relative abundance of Campylobacter have not been established. Proteobacteria is a major phylum of gram-negative bacteria in the intestine of pigs and high abundance of Proteobacteria in the hindgut of mice or human has been associated with gastrointestinal inflammation in response to environmental factors and inflammation-related dysbiosis (Carvalho et al., 2012; Mukhopadhya et al., 2012). Tan et al. (2018) reported that Campylobacter was more abundant in the cecum microbiota of the pigs with low feed efficiency compared to the pigs with high feed efficiency. Thus, in the current study, the lower growth performance of pigs fed NC2 diet than of those fed the NC1 diet at the end of Phase 1 of feeding could be partly be explained by the greater relative abundance of Proteobacteria phylum for the former diet than the latter diet.

The abundance of Methanosphaera genus in feces of humans is positively correlated with dietary level of Ca and P (Kim et al., 2020). Methanosphaera spp. including Methanosphaera stadtmanae utilize methanol (product of pectin degradation by microorganisms) and H₂ to produce methane with ADP phosphorylation (Sparling et al., 1993; Fricke et al., 2006), implying that methanogens require P to synthesize the methane. Thus, a decrease in the relative abundance of Methanosphaera observed in the current study could partly be attributed to the decrease in the dietary Ca and P content.

Cyanobacteria, which are referred to as blue-green algae bacteria, were more abundant in the jejunum and ileum of weaned pigs but were rarely found in the hindgut of weaned pigs (Li et al., 2017). Lin et al. (2019) observed that the abundance of Cyanobacteria in the fecal microbiome of sows tended to increase under the anaerobic condition. Hong et al. (2020) reported that feces from pigs fed the corn-SBM based diet with 20% canola meal tended to have a lower relative abundance of Cyanobacteria than feces from pigs fed the corn-SBM based diet without canola meal. However, there is a lack of information on the mechanisms by which multi-enzyme mixture supplementation can decrease the relative abundance of Cyanobacteria in the fecal microbiome of pigs fed fibrous diets. Thus, it is unclear why the relative abundance of Cyanobacteria was decreased by multi-enzyme mixture supplementation in the current study, and hence, there is a need to establish the mechanisms by which the multi-enzyme mixture can affect the abundance of Cyanobacteria in the gastrointestinal tract of pigs.

The Lactobacillus and Bifidobacterium genera are considered beneficial bacteria for pigs (Dowarah et al., 2017; Yin et al., 2018) because they are involved in production of short-chain fatty acids which result in decreased pH of intestine, promotion of gastrointestinal motility, and inhibition of the growth of nitrate-reducing bacteria (Mäkeläinen et al., 2010; Tana et al., 2010). The Butyricicoccus can contribute to production of butyric acid, which is considered to play a particularly important role as an energy source for colonocytes and in the maintenance of gut health (McIntyre et al., 1993; Pryde et al., 2002; Hedemann and Bach Knudsen, 2007). Metzler-Zebeli et al. (2019) reported increased abundance of Lactobacillus and Bifidobacterium in feces of pigs due to dietary resistant starch, implying that resistant starch is an important substrate for these microorganisms. Arabinoxylans, which are the major non-starch polysaccharides in corn and wheat grains and their co-products, were reported to be a better substrate for butyric acid production by microorganisms in pigs than other types of dietary fiber such as resistant starch (Nielsen et al., 2014). Supplementation of the multi-enzyme mixture to corn-wheat-wheat bran-based diet can result in degradation of NSP (such as arabinoxylans) into small fragments such as arabinoxylan-oligosaccharide (Pedersen et al., 2015; Petry et al., 2021). The degradation of NSP into small fragments can result in increased availability of otherwise NSP-encapsulated nutrients (such as resistant starch) for digestion in the small intestine, leading to their reduced flow to the hindgut; and in increased fermentation of the NSP (including the arabinoxylans) in the hindgut because of the small fragments of NSP that are released from the cell wall matrix by NSP-degrading enzymes can be more fermentable than intact NSP that are still entrapped in cell wall matrix (Pedersen et al., 2015; Petry et al., 2021). Thus, the decrease in relative abundance of Lactobacillus and Bifidobacterium and increase in relative abundance of Butyricicoccus in feces of pigs due to multi-enzyme mixture supplementation could be due to 1) shift in digestion of resistant starch from the hindgut towards small intestine and 2) increase in availability of arabinoxylan fragments in hindgut for fermentation. Because of the positive association between butyric acid production and gut health exists, the improvement in growth performance of pigs in the current study due to multi-enzyme mixture supplementation can partly be attributed to the increase in relative abundance of Butyricicoccus in feces by the multi-enzyme mixture supplementation.

In conclusion, femur ash for NC1 diet did not differ from that for PC diet and was not affected by multi-enzyme mixture supplementation, indicating that the reduction in Ca and digestible P by approximately 0.07 percentage points was not sufficient to impact bone mineralization. The overall ADG and G:F, femur ash content, and ATTD of GE and CP for PC diet were greater than those for the NC2 diet. Multi-enzyme mixture supplementation increased the femur ash content and ATTD of GE and CP for the NC2 diet to those of the PC diet. Multi-enzyme mixture supplementation also increased the overall G:F and the abundance of Butyricicoccus in fecal microbiome, but G:F value for the multi-enzyme mixture supplemented diet did not reach that of the PC diet. Thus, the Ca and digestible P can be lowered by approximately 0.12 percentage points in multi-enzyme mixture-supplemented diets without effects on bone mineralization pigs. However, multi-enzyme mixture supplementation may not fully restore growth performance of pigs fed diets that have been formulated to be lower in NE and SID AA than the NRC (2012) recommended values by 7% through an increase in dietary inclusion of wheat bran and soybean hulls. The increase in relative abundance of Butyricicoccus in feces of pigs due to multi-enzyme mixture supplementation implies that the addition of the multi-enzyme mixture in diets for pigs can additionally improve their gut health.

Glossary

Abbreviations

AA

Amino acids

ADF

acid detergent fiber

ADFI

average daily feed intake

ADG

average daily gain

ATTD

apparent total tract digestibility

BW

body weight

CP

crude protein

DM

dry matter

EE

ether extract

G:F

gain-to-feed ratio

GE

gross energy

NE

net energy

NDF

neutral detergent fiber

NSP

non-starch polysaccharides

OTUs

operating taxonomic units

SBM

soybean meal

SID

standardized ileal digestible

STTD

standardized total tract digestible

Conflict of interest statement

Maamer Jlali, Pierre Cozannet, and Aurelie Preynat are employees of Adisseo France S.A.S. that provided the multi-enzyme mixture product used in the current study. Jinsu Hong, Seidu Adams, Joy Scaria, and Tofuko Woyengo do not have any real or perceived conflicts of interest.