The experimental trial was conducted at the Swine Laboratory of the Department of Animal Science, Center for Human, Social, and Agrarian Sciences (CCHSA) of the Universidade Federal da Paraíba (UFPB), Campus III, in Bananeiras, Paraíba, Brazil (6°45'12" S and 35°38'57" W). All procedures in this research were approved by the Ethics Committee on Animal Use of the Universidade Federal da Paraíba (CEUA-UFPB), under protocol number 5476260521.

Sixty weaned piglets at 28±1 days of age were used, comprising 30 castrated males and 30 females of the same commercial lineage, sourced from a commercial farm, with an initial average weight of 6.43 ± 0.25 kg. The number of animals used followed the determinations of the CEUA/UFPB, and the sample size calculation was carried out as described by Sakomura and Rostagno (2016). The animals were housed in suspended nursery cages with slatted plastic floors, equipped with nipple drinkers and semi-automatic feeders.

The animals were distributed in a randomized block design (RBD) to control for differences in initial weight. The piglets were divided into six treatments (experimental diets), with five replicates, and each experimental unit consisted of two animals (one male and one female).

The experimental diets (Table 1) were formulated to meet the minimum nutritional requirements of the animals, according to Rostagno et al. (2017). The treatments evaluated were enzymatic complexes in combination with Yucca schidigera extract, where the additives replaced a fraction of the inert component of the diets. The treatments were: CD: control diet; CDY: CD with 125 g ton⁻¹ of Yucca schidigera extract (YSE) (De-Odorase^®; guarantee level: 65 mg kg^-1 of sapogenin); CDE: CD with 200 g ton⁻¹ of enzyme complex (EC) (Alltech^® Allzyme SSF e+C; guarantee levels: 700 HUT g⁻¹ of protease, 300 SPU g⁻¹ of phytase and 40 CMCU g⁻¹ of cellulase); CDME: CD with 400 g ton⁻¹ of multienzyme complex with emulsifier (Alltech^® Allzyme Allsotution; guarantee levels: 640 AJDU g⁻¹ pectinase, 386 HUT g⁻¹ protease, 60 SPU g⁻¹ phytase, 32 BGU g⁻¹ of β-glucanase, 16 XU g⁻¹ of xylanase, 8 CMCU g⁻¹ of cellulase, 4.5 FAU g⁻¹ of amylase and 52 g kg⁻¹ of glyceryl polyethyleneglycol ricinoleate - PEGR); CDE+Y1: CD + EC (200 g ton⁻¹) + YSE (125 g ton⁻¹); CDE+Y2: CD + YSE (200 g ton⁻¹) + 18 YSE (250 g ton⁻¹).

Table 1

Item	Phases
	I	II	III
Ingredient1
Yellow corn (7.9%)	39.04	50.32	65.38
Soybean meal (45.4%)	22.00	22.00	22.00
Inactivated whole soy (37.3%)	14.47	10.17	6.53
Whey powder (12.3%)	17.25	10.06	-
Soy oil	2.02	2.22	0.72
Dicalcium phosphate	1.74	1.75	1.63
Limestone	1.05	0.95	0.77
L-lysine HCL	0.52	0.54	0.42
DL-methionine	0.23	0.21	0.13
L-arginine	0.22	0.21	-
L-tryptophan	0.04	0.06	0.03
L-threonine	0.26	0.26	0.16
L-valine	0.11	0.12	0.02
Mineral and vitamin supplement	0.50	0.50	0.50
Salt	0.49	0.41	0.47
BHT	0.02	0.02	0.02
Inert2	0.06	0.18	0.23
Total	100.00	100.00	100.00
Calculated values
Metabolizable energy (kcal/kg)	3,400	3,375	3,250
Crude protein (%)	21.42	19.87	18.06
Calcium (%)	1.07	0.97	0.79
Crude fiber (%)	2.51	2.47	2.54
NDF (%)	10.46	11.40	12.95
ADF (%)	4.40	4.32	4.43
Available phosphorus (%)	0.53	0.48	0.39
Digestible tryptophan (%)	0.28	0.28	0.21
Digestible lysine (%)	1.45	1.35	1.12
Digestible methionine (%)	0.52	0.48	0.38
Digestible methionine+cystine (%)	0.81	0.75	0.64
Digestible threonine (%)	0.97	0.90	0.73

Phases: I - 21 to 32 (5.5 to 9 kg) of age; II - 33 to 42 (9.1 to 15 kg) of age; III - 43 to 63 (15.1 to 25 kg) of age.

The corresponding values of each component of the additives per kg of diet were calculated (Table 2).

Table 2

Item	Experimental diets1
	CD	CDY	CDE	CDME	CDE+Y1	CDE+Y2
Protease (HUT kg−1)	-	-	140.0	154.4	140.0	140.0
Phytase (SPU kg−1)	-	-	60.0	24.0	60.0	60.0
Cellulase (CMCU kg−1)	-	-	8.0	3.2	8.0	8.0
Pectinase (AJDU kg−1)	-	-	-	256.0	-	-
β-glucanase (BGU kg−1)	-	-	-	12.8	-	-
Xylanase (XU kg−1)	-	-	-	6.4	-	-
Amylase (FAU kg−1)	-	-	-	1.8	-	-
PEGR (g kg−1)	-	-	-	20.8	-	-
Sapogenin (mg kg−1)	-	8.1	-	-	8.1	16.3

¹ CD: control diet; CDY: CD with 125 g ton⁻¹ of Yucca schidigera extract (YSE) (De-Odorase^®; guarantee level: 65 mg kg⁻¹ of sapogenin); CDE: CD with 200 g ton⁻¹ of enzyme complex (EC) (Alltech^® Allzyme SSF e+C; guarantee levels: 700 HUT g⁻¹ of protease, 300 SPU g⁻¹ of phytase and 40 CMCU g⁻¹ of cellulase); CDME: CD with 400 g ton⁻¹ of multienzyme complex with emulsifier (Alltech^® Allzyme Allsotution; guarantee levels: 640 AJDU g⁻¹ pectinase, 386 HUT g⁻¹ protease, 60 SPU g⁻¹ phytase, 32 BGU g⁻¹ of β -glucanase, 16 XU g⁻¹ of xylanase, 8 CMCU g⁻¹ of cellulase, 4.5 FAU g⁻¹ of amylase and 52 g kg⁻¹ of glyceryl polyethyleneglycol ricinoleate - PEGR); CDE+Y1: CD + EC (200 g ton⁻¹) + YSE (125 g ton⁻¹); CDE+Y2: CD + YSE (200 g ton⁻¹) + YSE (250 g ton⁻¹).

The animals were weighed at the beginning and end of each phase (0 to 7 days, 0 to 16 days, and 0 to 32 days), as well as the feed leftovers, obtaining the daily feed intake (DFI), daily weight gain (DWG), and feed conversion ratio (FCR).

During the performance evaluation, the incidence of diarrhea in the piglets was assessed. Fecal scores were recorded during the first 21 days of the experimental period. Fecal scores were recorded during the first 21 days of the experimental period. Fecal consistency was visually analyzed twice daily, at 08:00 and 17:00 h, according to the following scores: 1 = normal feces; 2 = pasty feces; 3 = watery feces. Scores 1 and 2 were considered non-diarrheic feces, and 3 were considered diarrhea. These identifications were always performed by the same observer.

To evaluate the digestibility of the experimental diets, the partial feces collection method was used with the inclusion of 1% Celite as a source of acid-insoluble ash (AIA) in the diet from days 43 to 60 of age as an indigestibility marker (Liu, 2022). After the start of feed consumption with the marker, the feed was provided for three days to maintain the indicator flow through the digestive tract, followed by feces collection for four days.

The partial feces collection method was used, with two feces collections per day (morning and afternoon), directly from the animals’ rectum. After collection, the feces were homogenized and stored in plastic bags at −18 °C for later analysis. Feed samples were also collected and stored for analysis.

The feces were thawed at room temperature, homogenized by pen, pre-dried at 55 °C, and ground for the determination of dry matter, organic matter, ash, crude protein, gross energy, neutral detergent fiber, and acid detergent fiber (AOAC, 2005). The gross energy of the feces and feed was determined using a Parr 6100 bomb calorimeter.

Phosphorus content was analyzed by spectrophotometry using a UV-VIS spectrophotometer (UV-5100; Metash Instruments, Shanghai, China) according to the methodology proposed by Rangana (1979).

For calcium determination, samples were analyzed using an atomic absorption spectrometer with a flame atomizer (iCE 3500; Thermo Scientific, Cambridge, UK). A hollow cathode lamp containing Ca (Photron, Victoria, Australia) was used as the primary radiation source, and background correction was performed with a deuterium lamp coupled to the equipment.

The standard curve was prepared with a calcium standard solution (Specsol, São Paulo, Brazil). Instrumental parameters were used according to the manufacturer’s recommendations, and the data were processed using SOLAAR^® software (Thermo Scientific, Cambridge, UK).

The apparent total tract digestibility coefficients (ATTDC) of nutrients, energy, and Ca and P availability were calculated according to Adeola (2001), as follows:

Retained and excreted phosphorus and nitrogen were calculated as follows:

At the end of each phase (35, 44, and 60 days of age), approximately 10 mL of blood was collected directly from the jugular vein from one animal per experimental unit, selected according to the average weight closest to the treatment average weight. Blood samples were centrifuged at 2500 rpm for 10 minutes to obtain serum, and aliquots were stored in plastic microtubes, identified, and kept at −20 °C until analysis.

All analyses were performed using an automated biochemical analyzer, model Labmax 240 (Labtest^® Diagnóstica S.A; Brazil), using Labtest^® commercial kits. The serum metabolites analyzed were urea by the enzymatic method (Ureia UV Liquiform (Ref.: 104)) and creatinine by the colorimetric method (Creatinina K (Ref.: 96)). The hepatic serum markers analyzed were: aspartate aminotransferase (AST; AST Liquiform (Ref.: 109)), alanine aminotransferase (ALT), and alkaline phosphatase (ALP; ALP (Ref.: 76)) by the kinetic method, and gamma-glutamyl transferase (GGT; GGT Liquiform (Ref.: 105)) by the modified Szasz method. Serum minerals calcium (Ca Arsenazo Liquiform (Ref.: 95)) and phosphorus (Phosphorus UV Liquiform (Ref.: 12)) were analyzed by the arsenazo and UV photometry methods, respectively.

At the end of the experimental period (60 days of age) the piglets were slaughtered after an 6-hour fast by electrical stunning followed by exsanguination in accordance with the procedures approved by CEUA/UFPB. The piglets were slaughtered following the humane slaughter protocol.

Immediately after slaughter, the animals’ abdomens were opened, and the viscera removed. With the aid of forceps and a scalpel, fragments of the duodenum, jejunum, and ileum were collected as described below for subsequent analyses.

For the study of the small intestine structure, samples of approximately 1–10 cm from the proximal duodenum and 25–35 cm from the proximal jejunum and ileum were collected. The intestinal segments were fixed in a solution containing 60% methanol, 30% chloroform, and 10% acetic acid for twelve hours, and kept refrigerated. Shortly after, the solution was replaced with 70% alcohol for the histological analyses.

The samples remained in 70% alcohol for 24 hours, then washed in running water for five minutes, and subsequently dehydrated in a series of increasing alcohol concentrations, passed through a xylene bath, before being embedded in paraffin. Later, microtomy of the paraffin blocks was performed to prepare the histological slides.

The small intestine slides were stained using hematoxylin-eosin staining to determine the following parameters: villus height, crypt depth, and villus width. Based on this data, the villus height:crypt depth ratio (VH:CD), mucosal thickness, and absorptive area were calculated according to a modified methodology described by Moreira Filho et al. (2015).

The count of goblet cells present in the intestinal villi was performed after staining with periodic acid-Schiff (PAS) + hematoxylin. To evaluate the number of goblet cells, eight counts per sample were performed, with a line of 500 µm per villus drawn, totaling 4000 µm per sample. The result was expressed as the number of goblet cells per 2000 µm. For slide evaluation, a light microscope model Olympus BX53 and a Zeiss Axiocam camera coupled with CellSens Dimension digital image capture program was used.

After slaughter, jejunum fragments of approximately 1 cm were collected, washed in saline solution (0.9% NaCl), and small cuts were made in the fragments using a scalpel and scissors. Samples were stored in 2 mL microtubes, and frozen at −80 °C until mRNA isolation.

The mRNA was extracted from the samples using the Qiagen RNeasy^® Mini kit (Cat. No.74106), and cDNA synthesis was performed using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA), according to the manufacturers’ recommendations.

Primers were used for the mRNA expression of the following genes: tumor necrosis factor (TNF-α), sodium-glucose cotransporter type 1 (SGLT-1), dipeptide and tripeptide transporter in enterocytes (PEPT-1), mucin type 2 (MUC-2), sodium-dependent phosphate transporter type 2 (NaPi-IIb), and the reference genes β-actin (ACTB) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Table 3).

Table 3

Gene	Primer sequence (5'-3')		PS (bp)1	Access no.
	Foward	Reverse
TNF-α	tcaacctcctctctgccatc	cccaggtagatgggttcgta	91	JF831365.1
MUC-2	tccacggggactgctactac	caggtctgcttgtctgtgga	103	XM_021082584.1
NaPi-IIb	tatcccttcagctgggtgac	gggtcagagtcgacgagaac	100	JN983501.1
SGLT-1	cggttggagcttctctgttt	attccattcaaagccaccag	107	MW280290.1
PEPT-1	ccatgttctgggctttgttt	tgatccggctggattttaag	100	AY180903.1
GAPDH	acatggcctccaaggagtaaga	gatcgagttggggctgtgact	101	NM_001206359.1
β-actin	ctggcaccacaccttctaca	gggtcatcttctcacggttg	107	DQ178122.1

¹ Product size in base pairs.

Relative gene expression of mRNA was determined by real-time polymerase chain reaction (qPCR) using SYBR Power SYBR^® Green Master Mix (Thermo Fisher Scientific) and specific primers. After verifying that the primers amplified with an efficiency of approximately 100%, qPCR tests were performed.

The final volume was calculated for 20 μL of the reaction, containing the following components: 10 μL of SYBR Green QPCR Master mix; 4 μL of each primer; 0.3 μL of diluted reference dye; 0.7 μL of distilled and deionized water, and 5 μL of cDNA. The polymerase chain reaction was performed under the following conditions: 95 °C for 3 min (1 cycle), 95 °C for 15 s, and 60°C for 20 s (40 cycles), 95 °C for 1 min, 55 °C for 30 s, and 95 °C for 30 s (1 cycle).

After each test, a melting curve analysis was generated to verify the specificity and purity of all qPCR products, and β-actin and GAPDH were chosen as reference genes to normalize cDNA loading. The qPCR cycles were performed in a thermocycler, and the relative expression was calculated based on the 2^- method (Livak and Schmittgen, 2001), using reference genes for expression normalization.

Normality and homogeneity of variances were tested using the Cramer-von Mises and Levene tests, respectively. Variables were subjected to analysis of variance using the GLM procedure of SAS statistical software (SAS On Demand for Academics). The data was evaluated for outliers. The pen was considered the experimental unit except for blood data, for which the animal was considered the experimental unit. The data were analyzed considering treatment and BW block as fixed and random effects, respectively.

The statistical model used for the analyses was:

in which Y = observation for each parameter analyzed; μ = overall mean; Ti = treatment effect, where i = 1, 2, 3, 4, 5 and 6; Bj = block effect, where j = 1, 2, 3, 4, and 5; eij = experimental error associated with each observation.

In case of significant difference, means were compared by Tukey’s test (P<0.05) and considered a tendency when 0.05 < P < 0.10.

From day 0 to 7, piglets fed the CDE+Y2 diet showed higher average daily feed intake (ADFI) and average daily gain (ADG), a beneficial outcome considering the typical decline in performance due to post-weaning stress (Shi et al., 2014). However, higher feed conversion ratios observed in piglets fed CDY and CDE+Y2 from days 0 to 16 and 0 to 32 may be linked to unknown effects of Yucca extract (YSE). Research on YSE inclusion is inconsistent—some studies report slight reductions in body weight (Gebhardt et al., 2019), whereas others show improved feed conversion (Fan et al., 2022). Further investigation is needed to assess long-term impacts, particularly regarding potential toxicity from excessive saponins (Dos Reis et al., 2016).

The similar growth performance observed in piglets fed CDE and CDME compared with the control diet is likely due to the use of highly digestible ingredients. This may have reduced the effectiveness of enzyme supplementation, as enzyme benefits are typically more pronounced in diets with high levels of antinutritional factors or nutrient limitations (Barros et al., 2014; Ao, et al., 2020; Brandão Melo et al., 2020).

Notably, piglets fed CDE+Y1 showed improved feed conversion from days 0 to 32. Supporting this, a meta-analysis by Torres-Pitarch et al. (2017) found that 61% of studies on enzyme supplementation in weaned piglets reported improvements in growth performance and feed efficiency. These preliminary results suggest that the high digestibility of the basal diet may have masked the potential benefits of supplementation.

The use of combined exogenous enzymes in pig diets is an effective strategy to reduce antinutritional factors, enhance nutrient digestion and absorption, and improve overall performance (Duarte et al., 2019). Supplementation with Yucca schidigera extract (YSE) at 120–125 g ton⁻¹ has also been linked to improved piglet performance, likely due to enhanced intestinal barrier function (Colina et al., 2001; Gebhardt et al., 2019; Fan et al., 2022; Yang et al., 2021). These benefits are attributed to the glyco-components and saponins present in YSE (Chen et al., 2021), which, when combined with enzyme complexes, contribute to better productive outcomes. However, no effect on diarrhea incidence was observed, differing from previous studies that reported reduced diarrheic scores with enzyme or YSE supplementation (Zhang et al., 2014; Long et al., 2021; Yang et al., 2021). This discrepancy may be due to elevated oxidative stress during weaning, which could have exceeded the mitigating capacity of YSE (Chen et al., 2021).

The improvements in dry matter (DM), organic matter (OM), NDF, ADF, and energy digestibility observed with CDE and CDME supplementation are attributed to the hydrolysis of otherwise indigestible feed substrates, which become accessible targets for exogenous enzymes (Adeola et al., 2011). The carbohydrases used - amylase, cellulase, β-glucanase, pectinase, and xylanase - are designed to break down compounds primarily found in the cell walls of plant-based ingredients (Ao et al., 2020).

Additionally, CDE, CDME, and CDE+Y1 enhanced nutrient availability and retention while reducing phosphorus (P) excretion, a result consistent with findings of Trindade Neto et al. (2021). This effect is partly due to the inclusion of exogenous protease, which helps degrade the protein matrix surrounding starch granules. When combined with phytase, which hydrolyzes phytate, this enzymatic synergy further improves nutrient and energy release.

Phytase supplementation is well documented as an effective nutritional strategy to reduce dietary and excreted phosphorus (P), addressing both environmental and economic concerns (Torres-Pitarch et al., 2017). Given that inorganic P is a limited, non-renewable resource, and that manure use as fertilizer is a major contributor to P contamination, optimizing dietary P levels in livestock is crucial (Menezes-Blackburn et al., 2015).

Combining enzymes in diets with high in non-starch polysaccharides (NSP) and fiber has shown positive effects on growth performance and feed efficiency in piglets (Kim et al., 2005; Cowieson et al., 2017; Zeng et al., 2018). These enzymes degrade β-glucans and arabinoxylans in the plant cell wall, reducing digesta viscosity in the small intestine. This reduction allows endogenous enzymes to act more effectively, as high viscosity can otherwise hinder their access to substrates (Owusu-Asiedu, 2010; Passos et al., 2015; Li et al., 2018).

The improved nutrient digestibility following enzyme complex supplementation may result from the effective degradation of non-starch polysaccharides (NSP) (Duarte et al., 2019; Li et al., 2021). Giang et al. (2010) suggested that exogenous enzymes enhance nutrient digestibility in the first two weeks post-weaning due to the underdeveloped proteolytic and amylolytic digestive system in piglets, which matures around the six week of age.

Guo et al. (2022) reported that multi-enzyme complexes improved feed, amino acid, and energy utilization in corn- and soybean-based diets for piglets during the early post-weaning period. These findings suggest that enzyme supplementation may enable newly weaned piglets to extract nutrients and energy from simpler diets more efficiently.

In our study, combining enzyme supplementation with a high level of Yucca schidigera extract (YSE, 250 g ton⁻¹) resulted in lower nutrient digestibility, reduced retention, and increased phosphorus excretion. However, supplementation with 125 g ton⁻¹ of YSE has been shown to improve ammonia nitrogen utilization, likely enhancing nutrient digestibility (Fan et al., 2022).

The saponins in YSE reduce ammonia levels and support nutrient digestion and absorption at doses up to 200 g ton⁻¹, though higher levels may have adverse effects (Yen and Pond, 1993). The mechanism behind YSE’s action remains unclear, as it is believed not to be absorbed by the digestive tract, suggesting no direct metabolic effects (Espinosa-Muñoz et al., 2008). Kaya et al. (2003) proposed that YSE’s surfactant activity, particularly from saponins, reduces cell membrane tension in microvilli, facilitating nutrient absorption. However, Shi et al. (2014) cautioned that the efficacy of saponins is dose-dependent, with low levels being beneficial but high levels potentially acting as antinutritional factors.

On day seven, only the liver enzyme GGT differed significantly among treatments, with higher levels observed in diets containing YSE. However, these values remained within the normal physiological range (Kaneko, 1989). As biochemical markers reflect animal performance, and serum creatinine is closely linked to body weight and muscle mass in pigs (Wang et al., 2021), the absence of changes in creatinine levels suggests that amino acid needs were adequately met for optimal muscle growth.

At day 32, increased AST levels were observed in diets supplemented with YSE, except for CDE+Y1, which maintained the level equal to CD. However, AST levels are considered normal (32 to 84 U L⁻¹) according to Kaneko (1989). According to Kurtz and Travlos (2017), the enzymes GGT and AST are considered important markers of liver function and assessment of the enzymatic metabolic profile. Liver tissue damage causes these enzymes to leak into systemic circulation, thereby altering their serum levels.

Yang et al. (2021) found that including 120 g ton⁻¹ of YSE in weaned piglet diets did not affect serum ALT levels, indicating its safety at this dosage. However, higher YSE levels led to increased ALT and potential side effects (Dos Reis et al., 2016). Although saponins are not inherently toxic, excessive intake can cause adverse effects. The elevated serum ALP levels observed in piglets fed CDY and CDE+Y2 may reflect compromised intestinal integrity and increased permeability (Bilski et al., 2017; Celi et al., 2019; Wang et al., 2022).

Serum phosphorus (P) levels are known to increase in proportion to dietary P content (Dellaert et al., 1990; Eeckhout et al., 1995). Additionally, phytase supplementation enhances serum P concentrations (Gentile et al., 2003; Beaulieu et al., 2007). Lu et al. (2016) also reported that piglets fed diets richer in calcium (Ca) and phosphorus showed higher serum levels of both minerals.

Enzyme supplementation did not affect serum Ca concentrations. However, we observed lower serum P when there was a higher inclusion of YSE (250 g ton⁻¹) at the end of the second phase (16th day). ATTD of P was 28.03% lower in CDE+Y2 compared with CDE, suggesting that supplementation with YSE at the level of 250 g ton⁻¹ negatively impacted mineral utilization. This is corroborated by the 24.97% lower serum P under the same condition.

Post-weaning stress is linked to negative changes in intestinal morphology (Spreeuwenberg et al., 2001; Leonard et al., 2011) and impaired intestinal barrier function (Carey et al., 1994). Key indicators of small intestine absorptive capacity include villus height and width, crypt depth, VH:CD ratio, mucosal thickness, and absorptive area (Yang et al., 2021). Structural alterations can lead to the loss of brush border enzymes—such as aminopeptidases and carbohydrases—reducing nutrient digestibility (Kluess et al., 2010). Weaned piglets typically show reduced villus height, limiting absorption (Long et al., 2021). However, piglets receiving enzymatic supplementation, with or without plant extracts, showed improved villus morphology and mucosal thickness. Crypt depth increased proportionally, resulting in no change to the VH:CD ratio.

Jejunal villi development improved with enzymatic supplementation combined with plant extracts, particularly with the inclusion of a multi-enzyme complex plus emulsifier (CDME). Unlike in the duodenum, the VH:CD ratio increased in piglets fed CDE, CDME, and CDE+Y1 compared with the control (CD).

Treatments with CDE and CDME also resulted in wider villi, with CDME promoting greater mucosal thickness and absorptive area—likely due to its higher enzyme concentration enhanced by the emulsifier. Similar improvements were observed in the ileum, supporting the conclusion that combining exogenous enzymes with plant extracts enhances intestinal morphology, consistent with improved digestibility and performance outcomes.

Prior studies have reported positive effects of enzyme supplementation on intestinal structure in weaned piglets (Valverde Piedra et al., 2009; Yi et al., 2013; Jiang et al., 2015; Duarte et al., 2019; Long et al., 2021; Yang et al., 2021; Fan et al., 2022).

Maintaining intestinal integrity is essential for growth and health in piglets, as the villus height-to-crypt depth (VH:CD) ratio reflects nutrient absorption and digestive efficiency (Li et al., 2021). A higher VH:CD ratio is associated with better nutrient utilization (Montagne et al., 2003). These benefits may be attributed to both the improved digestibility and increased post-weaning feed intake observed in piglets fed the supplemented diets, aligning with findings by Zhu et al. (2022) on the relationship between feed intake and intestinal development.

Goblet cells produce mucins that form a protective mucus layer, which helps prevent pathogens from binding to the intestinal epithelium (Xiong et al., 2019). The thickness of this layer and the number of goblet cells are critical for intestinal defense (Song et al., 2022). Enzymatic supplementation combined with YSE at 125 g ton⁻¹ (CDE+Y1) increased the number of jejunal goblet cells, likely due to improved intestinal morphology. However, a higher YSE level (250 g ton⁻¹) resulted in fewer goblet cells in both the jejunum and ileum.

In the intestine, toxins can trigger inflammation and impair the protective function of the mucus barrier (Song et al., 2022). While saponins in YSE offer benefits at low inclusion levels, higher concentrations may lead to adverse effects and act as antinutritional factors (Shi et al., 2014; Dos Reis et al., 2016).

The combination of exogenous enzymes and plant extracts reduced the relative mRNA expression of TNF-α, a key pro-inflammatory cytokine. While the exact mechanism behind this anti-inflammatory effect remains unclear (Moita and Kim, 2022), the PGRG emulsifier in the CDME diet may have enhanced lipid digestion and energy utilization, amplifying the benefits of the multienzyme complex. Additionally, the breakdown of NSPs and phytate by carbohydrases and phytase produces oligosaccharides and reduces digesta viscosity, both of which may help lower pro-inflammatory cytokine levels and support immune function in piglets (Agyekum et al., 2015; Lee and Bedford, 2016; Chen et al., 2020).

YSE contains resveratrol and saponins, both of which have strong anti-inflammatory properties. Resveratrol can significantly reduce TNF-α levels (Wenzig et al., 2008), while saponins—especially the protopanaxadiol saponin fraction—have been shown to suppress the expression of inflammatory cytokines (Cheeke et al., 2006; Yang et al., 2015). These properties likely explain the reduced inflammation observed in piglets receiving diets supplemented with enzymes and plant extracts during the early post-weaning phase. However, studies on the immune-modulatory effects of enzyme supplementation, with or without plant extracts, remain limited and likely depend on factors such as physiological stage, diet composition, and enzyme dosage.

The intestinal barrier consists primarily of epithelial cells and a mucosal layer (France and Turner, 2017). Fan et al. (2022) found that supplementation with 125 g/ton of YSE increased jejunal MUC-2 expression, suggesting a role for saponins, phenolics, and polysaccharides in enhancing barrier function. However, in our study, higher MUC-2 expression was observed in piglets fed the control diet, possibly indicating increased mucosal turnover due to reduced nutrient digestibility (Lu et al., 2016). Lower MUC-2 levels in enzyme-supplemented diets may reflect better jejunal integrity and reduced epithelial stress, as supported by findings from Lu et al. (2016) and Shekels et al. (2001).

Our results also showed that CDE supplementation increased jejunal NaPi-IIb mRNA expression by 53.1%, likely due to the inclusion of phytase in the enzyme blend, which enhanced phosphorus availability and retention. This supports findings by Vigors et al. (2014). Other treatments had no significant effect on NaPi-IIb expression, likely due to the predominance of passive (paracellular) phosphorus absorption when dietary P is adequate (Stein et al., 2008; Sabbagh et al., 2011).

Enzyme supplementation also influenced nutrient transporter expression. CDE increased SGLT-1 expression by 54.2%, and diets containing enzymes —especially CDY and CDE+Y2 — upregulated PEPT-1. This is likely due to the hydrolysis of antinutritional factors, which increased the availability of glucose and peptides for absorption. Similar effects were reported by Zhu et al. (2022).

Our findings are consistent with those of Fan et al. (2022), who observed increased SGLT-1 expression in piglets fed 125 g ton^-1 of YSE. Although PEPT-1 was unaffected in that study, there were significant increases in other nutrient transporters (e.g., CAT-1, ASCT-1, rBAT, FATP-4), which support nutrient absorption and play key roles in metabolic and signaling pathways.

Given the swine industry’s need for sustainable production strategies, our results highlight the potential of combining enzyme complexes with plant extracts to enhance intestinal health and nutrient metabolism in weaned piglets. Further research is needed to determine optimal inclusion rates, timing, and interactions with other feed additives to fully harness these benefits.