Multi-protease supplementation enhances growth performance and promotes gut health in broilers fed low-protein, low-amino acid diets

Abstract

The objective of this study was to assess the impact of dietary supplementation with a multi-protease enzyme on broiler diets with reduced levels of protein and amino acids. One-day-old male Cobb-500 chicks (2250 birds) were assigned to 10 treatments (9 pens/treatment). The treatments included (1) a nutritionally adequate Positive Control (PC); (2) PC + 200 g/t of a multi-protease (KZP); (3) a Negative Control (NC1) reformulated to—4% PC and amino acids; (4) NC1 + 100 g/t KZP; (5). NC1 + 200 g/t KZP; (6). NC1 + 300 g/t KZP; (7) a Negative Control (NC2) reformulated to - 8% PC and amino acids; (8). NC2 + 100 g/t KZP; (9) NC2 + 200 g/t KZP; and (10) NC2 + 300 g/t KZP. NC1 and NC2 showed the worst growth performance, with supplementation of KZP able to restore performance to the level of the PC at 200 g/t and 300 g/t dosage. There was no effect on carcass yield. The inclusion of multi-protease enhanced intestinal morphology, with increased villus height and villus: crypt ratio, along with reduced crypt depth. Protease supplementation promoted descriptive taxonomic shifts in cecal microbiota composition. Multi-protease enhanced bird performance, gut morphology, and influenced microbial community composition in amino acid-reduced diets.

• Multi-protease supplementation allowed the reduction of dietary crude protein and amino acid levels without compromising broiler performance, restoring growth to the level of the positive control and contributing to lower feed costs and reduced environmental impact.

• Improvements in intestinal morphology and descriptive shifts in microbial community composition indicate greater digestive efficiency, reinforcing the enzyme’s potential as a sustainable nutritional strategy in broiler production.

Introduction

Corn and soybean meal are the primary ingredients used in broiler diets worldwide. However, the nutritional value of soybean meal may be limited by the presence of antinutritional factors, including trypsin inhibitors, allergenic proteins, non-starch polysaccharides, and phytate, which can impair nutrient utilization and reduce bird performance (Hemetsberger et al. 2021; Rocha et al. 2014; Troni et al. 2016).

Protein and digestible amino acids represent a substantial proportion of feed costs, making precise formulation essential for maximizing production efficiency. Diets formulated according to the ideal protein concept can improve amino acid balance while reducing crude protein levels, thereby decreasing nitrogen excretion and improving nutrient utilization efficiency (Vasconcellos et al. 2011; Oliveira et al. 2012; Wu 2014). In this context, exogenous proteases have been widely incorporated into poultry diets to enhance protein digestion and amino acid availability (Shad et al. 2022; Vieira et al. 2023).

The use of exogenous proteases in animal diets has been a significant nutritional approach in improving protein utilization and reducing nitrogen loss, improve performance, carcass yield, gut morphology and intestinal microbiome. These improvements in efficiency, as well as in carcass weight and yield, are likely driven largely by enhanced AA digestion and protein deposition (Wealleans et al. 2017a). In a study conducted by Duque-Ramírez et al. (2023), it was found that protease improves animal performance by enhancing parameters such as weight gain, feed conversion, meat quality, and carcass yield.

The exact mechanisms through which a protease supports gut health remain unclear and are likely the result of multiple interacting factors (Cowieson and Roos 2016). Previous studies have shown that exogenous proteases can alter the availability of substrates in the intestine, potentially influencing bacterial growth (Bedford and Patridge 2001; Malo et al. 2010).

Previous studies have shown that protease supplementation can improve feed efficiency, growth performance, nitrogen utilization, and protein digestibility, resulting in greater amino acid availability for protein deposition and improved productive efficiency in broilers (Oxenboll et al. 2011; Matias et al. 2015; Wealleans et al. 2017a; Cowieson et al. 2018).

Although proteases are designed to offset decreases in crude protein and amino acids, research findings have been mixed. Further investigation is necessary to elucidate the non-protein-related benefits of proteases, including their impact on gut morphology and changes in the microbiome.

Therefore our study examined how increasing multi-protease levels in broiler diets with lower amino acid content affects bird performance, carcass traits, gut morphology, and intestinal microbiome composition.

Materials and methods

All experimental procedures in this study were approved by the Animal Use Research Committee of the Federal University of Lavras, Minas Gerais, Brazil. The study was conducted at the CPTA (Center for Research in Poultry Technology) affiliated with the Federal University of Lavras—MG (Brazil).

Animals and housing

A total of 2250 one-day-old male Cobb-500 chicks, obtained from a commercial hatchery, were used. The chicks were properly vaccinated against Marek’s disease and vectorized gumboro. The birds were raised in a conventional poultry house equipped with internal and external side curtains, a ceiling lining, fans, and thermometers for maintaining and recording the ambient temperature according to the age, as indicated in the lineage manual (Cobb-500, 2016).

The feed was provided in mash form in tubular feeders, and water (nipple drinkers) and feed were provided ad libitum throughout the experimental period. Heating of the chicks up to 14 days was carried out using gas heaters with automatic temperature control.

Experimental procedures

The animals were distributed in a completely randomized design in 90 experimental plots containing 25 birds each, assigned to ten treatments with nine pen replicates per treatment. The treatments included (1) a nutritionally-adequate Positive Control (PC); (2) PC + 200 g/t of a multi-protease (KZP); (3) a Negative Control 1 (NC1) reformulated to—4% PC and amino acids; (4) NC1 + 100 g/t KZP; (5) NC1 + 200 g/t KZP; (6) NC1 + 300 g/t KZP; (7) a Negative Control 2 (NC2) reformulated to—8% PC and amino acids; (8) NC2 + 100 g/t KZP; (9) NC2 + 200 g/t KZP; and (10) NC2 + 300 g/t KZP.

The diets consisted of corn and soybean meal, following a nutritional program with starter (1 to 21 days), grower (22 to 35 days), and finisher (36 to 42 days) phases formulated according to the nutritional recommendations of Bertechini (2020), performance variables were evaluated and presented as cumulative periods (1–7, 1–21, 1–35, and 1–42 d). The program included protein and amino acid nutritional challenges, with one diet having a 4% reduction and another diet having an 8% reduction (Tables 1 and 2). The experimental diets were formulated to maintain similar metabolizable energy, electrolyte balance, vitamin and mineral levels across treatments, with crude protein and amino acid concentrations being the primary variables modified in the negative control diets. The multi-protease used in this study was KEMZYME^®Protease from Kemin Industries, Des Moines, USA (KZP) is a patented thermostable multi-protease containing acid, neutral, and alkaline proteases. The enzyme is designed for targeted release along the gastrointestinal tract, allowing protein hydrolysis under different pH conditions and improving amino acid digestibility. Due to its formulation, KZP remains active throughout the gastrointestinal tract. Its innovative enteric coating protects the neutral, acid, and alkaline proteases in the gastric phase and releases protease activity at neutral and alkaline regions of the gastrointestinal tract. Additionally, a thermostable coating protects all the three types of proteases from the heat generated during the feed milling process. No other enzymes were included in the experimental diets.

Table 1.

Composition of experimental diets for the 1 to 21 days phase.

Ingredients	Positive control (PC)	Negative Control 1 (NC1)	Negative Control 2 (NC2)
Corn 7, 88%	60.00	60.00	65.45
Soybean meal 45%	34.60	34.71	30.00
Oil	1.33	1.38	0.45
Dicalcium phosphate	1.88	1.88	1.91
Limestone	0.68	0.68	0.70
Salt	0.37	0.40	0.33
Sodium bicarboante	0.13	0.10	0.20
Mineral premix2	0.10	0.10	0.10
Vitamin premix1	0.10	0.10	0.10
DL-Methionine 99%	0.32	0.29	0.29
L-Lysine 78%	0.27	0.20	0.28
L-Threonine 98%	0.09	0.05	0.08
Salinomycin 12%	0.05	0.05	0.05
Inert (Kaolin)	0.03	0.03	0.03
Total	100	100	100
CP (%)	21.18	20.33	19.48
ME, kcal	2950	2950	2950
Ca (%)	0.85	0.85	0.85
AvP (%)	0.45	0.45	0.45
tP (%)	0.55	0.68	0.67
Na (%)	0.20	0.20	0.20
K (%)	0.84	0.82	0.75
Cl (%)	0.39	0.35	0.32
Lys dig (%)	1.20	1.15	1.10
Met dig (%)	0.60	0.57	0.55
M + C (%)	0.88	0.85	0.81
Thr dig (%)	0.79	0.75	0.72

¹Supplementation per kilogram of feed: vitamin A. 12.000 IU; vitamin D3. 2.500 IU; vitamin E. 30 IU; vitamin B1. 2 mg; vitamin B6. 3 mg; calcium pantothenate. 10 mg; biotin. 0.07 mg; vitamin K₃. 3 mg; folic acid. 1 mg; nicotinic acid. 35 mg; vitamin B₁₂. 15 μg; selenium. 0.300 mg.

²Supplementation per kg of feed: manganese. 80 mg; iron. 50 mg; zinc. 50 mg; copper. 10 mg; cobalt. 1 mg; iode. 1 mg.

Table 2.

Composition of experimental diets for the 22 to 42 days phase.

Ingredients	Positive control (PC)	Negative control 1 (NC1)	Negative control 2 (NC2)
Corn (7.88%)	65.63	66.79	73.38
Soybean meal (45%)	28.92	27.77	26.61
Oil	2.41	2.41	2.41
Dicalcium Phosphate	0.84	0.84	0.84
Limestone	0.82	0.82	0.82
Salt	0.28	0.28	0.28
Sodium bicarbonate	0.20	0.20	0.20
Minerals premix2	0.10	0.10	0.10
Vitamins premix1	0.10	0.10	0.10
DL-Methionine 99	0.28	0.24	0.20
L-Lisine 78	0.20	0.15	0.10
L-Threonine 98	0.02	0.00	0.00
Salinomycin	0.05	0.05	0.05
Inert	0.10	0.21	0.31
Total	100.00	100.00	100.00
CP (%)	18.66	17.92	17.18
ME kcal/kg	3100	3100	3100
Ca (%)	0.76	0.76	0.76
P disp (%)	0.38	0.38	0.38
P total (%)	0.47	0.47	0.47
Na (%)	0.18	0.18	0.18
Lis dig (%)	1.04	0.99	0.95
Met dig (%)	0.55	0.52	0.50
Met + cis dig (%)	0.81	0.77	0.74
Treo dig (%)	0.67	0.64	0.61

¹Supplementation per kilogram of feed: vitamin A. 12.000 IU; vitamin D3. 2.500 IU; vitamin E. 30 IU; vitamin B1. 2 mg; vitamin B6. 3 mg; calcium pantothenate. 10 mg; biotin. 0.07 mg; vitamin K₃. 3 mg; folic acid. 1 mg; nicotinic acid. 35 mg; vitamin B₁₂. 15 μg; selenium. 0.300 mg.

²Supplementation per kg of feed: manganese. 80 mg; iron. 50 mg; zinc. 50 mg; copper. 10 mg; cobalt. 1 mg; iode. 1 mg.

The birds and the leftover feed were weighed at 7, 21, 35, and 42 days of age for evaluation of weight gain (WG), feed intake (FI), feed conversion (FC), mortality rate, and productive efficiency index. At 42 days of age, two birds per experimental unit were euthanized by cervical dislocation after an 8-hour fast to evaluate carcass yield, breast, thigh + drumstick, back and wings. The birds were slaughtered according to the average weight of each replication, within a variation range of ± 10% of the average weight of each plot, this criterion was adopted to ensure that the sampled birds were representative of their respective experimental units and to minimize the effects of individual weight variation on the measured parameters. The birds were identified, individually weighed, and placed in boxes. After fasting, the birds were weighed again to calculate the carcass and cut yields relative to the slaughter weight. Carcass yield (%) was obtained by the ratio between the weight of the cold carcass (without feet, head, and neck) and the fasting weight. The yields of breast, thigh, drumstick, thigh + drumstick, wing, and back (%) were obtained by the ratio between the weight of these parts and the cold carcass weight.

The same birds were subsequently used for intestinal health assessment. Samples from the mid-jejunum and ileum were collected for histomorphometric analyses, focusing on villus height, crypt depth, and the villus: crypt (V: C) ratio. Histological slides were prepared from 5 μm tissue sections obtained with a microtome after paraffin embedding. Slides were stained with hematoxylin and eosin, following the method described by Nuñez et al. (1996), and analyzed under an optical microscope (Olympus) using ProgRes CapturePro software (Version 2.7; Jenoptik, Jena, Germany).

Measurements were performed with ImageJ software, considering fifteen well-oriented and vertically aligned villi per section. Villus height was measured from the tip to the villus–crypt junction, while crypt depth was measured from the crypt base to the same junction. The V: C ratio was calculated by dividing mean villus height by mean crypt depth. These data enabled the evaluation of intestinal structure integrity and development in response to the tested treatments.

Intestinal digesta was collected at the end of the performance trial, along with samples from different intestinal segments, for the determination of the intestinal microbiota profile. Samples were immediately stored under appropriate conditions until analysis. Microbial DNA was extracted from the intestinal digesta, and the bacterial community was characterized through 16S rRNA gene sequencing. Raw sequence data were subjected to quality control procedures, including filtering and removal of low-quality reads, followed by clustering into Operational Taxonomic Units (OTUs) and taxonomic assignment using a reference database.

To read sequence files, aggregate OTUs, and analyze microbial communities, the packages phyloseq (McMurdie and Holmes 2013), vegan (Oksanen et al. 2007), and microbiome (Lahti and Shetty 2018) were used in R software (R Foundation for Statistical Computing, Vienna, Austria). Alpha diversity was evaluated using the Observed Richness and Shannon diversity indices at the phylum, family, and species taxonomic levels. Beta diversity was assessed using Bray–Curtis dissimilarity and visualized through principal coordinates analysis (PCoA). Taxonomic aggregation was performed separately for phylum-, family-, and species-level analyses to evaluate microbial community responses to dietary treatments.

Statistical analysis

At the end of each phase and in the total phase of bird production, data were analyzed using analysis of variance (ANOVA) with the SISVAR statistical package (Ferreira 2011). The experiment was conducted as a completely randomized design with ten independent dietary treatments and nine replicate pens per treatment, with the pen considered the experimental unit. Prior to analysis, data were evaluated for normality of residuals and homogeneity of variances to verify the assumptions of ANOVA. When significant treatment effects were detected (P < 0.05), means were compared using the Student-Newman-Keuls (SNK) test at the 5% significance level.

Microbiome alpha diversity was evaluated using the Shannon diversity index and Observed Richness at the phylum, family, and species taxonomic levels. Linear regression analyses were performed to assess the dose-response effect of KZP supplementation within the NC1 and NC2 dietary treatments, with KZP inclusion level considered a continuous variable. Beta diversity was assessed using Bray–Curtis dissimilarity and visualized by principal coordinates analysis (PCoA). Due to the limited number of observations available for each treatment, PCoA results were interpreted descriptively and were not subjected to inferential statistical testing. Statistical significance was declared at P ≤ 0.05.

Results and discussion

Performance

Exogenous enzymes such as carbohydrases, phytase, and protease are widely used to improve nutrient digestibility in broiler diets. Proteases enhance protein degradation and amino acid utilization (Chambers et al. 2001), and may improve nutrient digestibility even in low-protein diets (Bertechini et al. 2020). However, responses to protease supplementation have been inconsistent, likely due to differences in experimental conditions and diet formulation (Cardinal et al. 2019). In the present study, no differences (P> 0.05) were observed in feed intake, weight gain, or feed conversion during the first week, which may be related to the immature digestive system of newly hatched chicks (Maiorka and Dahlke 2006), as well as the limited capacity of young birds to respond to dietary protease supplementation during early gastrointestinal development.

During the 1 to 21-day phase (Table 3), dietary amino acid reduction and protease supplementation significantly affected bird performance (P < 0.05). Feed intake differed only between treatment 4 (NC1 + 100 g/t KZP) and treatment 9 (NC2 + 200 g/t KZP), whereas no differences were observed among the remaining treatments (P > 0.05). The reduced amino acid levels in NC1 and NC2 negatively affected weight gain compared with the positive control (P < 0.05).

Table 3.

Impacts of dietary amino acid levels and multi-protease supplementation on broiler performance for days 1–7. 1–21 and 1–35¹.

Diet	Protease level, g/t	Feed intake, g	Weight gain, g	Feed conversion
d1-7
PC	0	149	122	1.22
	200	155	127	1.22
NC1	0	156	131	1.19
	100	151	124	1.21
	200	150	125	1.20
	300	153	126	1.21
NC2	0	156	131	1.19
	100	154	128	1.20
	200	156	128	1.21
	300	151	126	1.19
CV.%		5.18	6.79	3.78
P-value		0.49	0.44	0.53
SEM		0.02	0.02	0.01
d1-21
PC	0	1274ab	938ab	1.35a
	200	1278ab	971a	1.31a
NC1	0	1280ab	856c	1.49b
	100	1264b	947ab	1.33a
	200	1277ab	953a	1.34a
	300	1279ab	928b	1.37a
NC2	0	1285ab	830c	1.54b
	100	1276ab	913b	1.39a
	200	1293a	933ab	1.38a
	300	1282ab	953a	1.34a
CV.%		3.15	2.29	3.36
P-value		0.01	0.01	0.01
SEM		0.04	0.09	0.01
d1-35
PC	0	3526	2304ab	1.53a
	200	3580	2416a	1.48a
NC1	0	3529	2210bc	1.59c
	100	3592	2314ab	1.55ab
	200	3523	2367a	1.48a
	300	3549	2396a	1.48a
NC2	0	3612	2190c	1.64c
	100	3577	2247bc	1.59ab
	200	3592	2370a	1.51a
	300	3613	2399a	1.50a
CV %		2.91	3.63	3.74
P-value		0.44	0.01	0.01
SEM		0.03	0.02	0.01

¹Means with different letters in the column differ statistically by the test SNK (P < 0.05).

SEM, standard error of the mean.

Differences in feed intake among treatments may have influenced nutrient consumption and, consequently, contributed to the observed performance responses. Therefore, the results obtained likely reflect the combined effects of dietary amino acid levels, protease supplementation, and variations in nutrient intake among treatments.

Protease supplementation at 100 g/t significantly improved weight gain in birds fed the negative control diets (NC1 and NC2), whereas higher inclusion levels did not result in additional significant improvements. Protease supplementation to the positive control diet showed numerical improvements in body weight gain and FCR (P > 0.05). Lowering dietary amino acids greatly impacted feed conversion, leading to higher FCR in the negative controls NC1 and NC2. However, since the differences between NC1 and NC2 were not significant (P > 0.05), it indicates that reducing amino acid levels by more than 4% did not affect bird performance during this period. The supplementation of protease at a minimum dosage of 100 g/t effectively restored the performance of broilers fed negative control diets to a level comparable with that of positive control treatment during this phase of broiler growth. No significant improvement in body weight gain and FCR was noted with higher protease dosages, despite some numerical differences.

For the 1 to 35-day phase, there was an effect of the treatments (P < 0.05) on weight gain and feed conversion, with no effects on feed intake (Table 3). The two negative control treatments (NC1 and NC2) showed significantly lower weight gains (P < 0.05) compared to other treatments. Similarly, NC1 + 100 g/t and NC2 + 100 g/t also had lower gains, indicating that adding 100 g/t of KZP does not adequately compensate for the 4% and 8% reductions in dietary amino acids by improving digestion. The PC treatment with the nutritionally adequate levels of AA according to the feed tables showed similar weight gain to NC1, which had a 4% reduction in the levels of limiting amino acids. This fact indicates that the nutritional levels of PC have a safety margin of at least 4% in amino acids.

The highest weight gains for the 1 to 35-day phase of the birds were observed for PC + 200 g/t as well as NC2 + 200 g/t and NC2 + 300 g/t. This implies that at these dosages, the protease has the potential to restore bird weight gain despite an up to 8% reduction in amino acids, suggesting significant improvements in the digestibility of these nutrients.

Considering only the effects of the protease in each basal diet (NC1, NC2) studied, it can be observed that as the enzyme inclusion increased, there was also a linear increase (P < 0.05) in the weight gain responses (Figs. 1 and 2). This indicates that up to the use of 300 g/t of the enzyme, there was substrate (protein) for the enzyme action, resulting in the observed increase in weight gain. For FCR, the worst values were observed for the NC1 and NC2 treatments, while the best responses were seen for birds fed 200 and 300 g/t KZP, regardless of the nutritional reductions. In vitro studies conducted by Peñuela-Sierra et al. (2024) using the same multi-protease also reported a significant increase in the release of free amino acids from feed ingredients and compound feeds as the enzyme level increased. The highest inclusion level 2400 U/g (equivalent to 300 g/t), released on average 76.6% more free amino acids compared to 1200 U/g.

Figure 1.

Weight gain from 1 to 35 days according to the inclusion of protease with 4% nutritional reductions (NC1).

Figure 2.

Weight gain from 1 to 35 days according to the inclusion of protease with 8% nutritional reductions (NC2).

Few in vivo studies are available to assess the dose response to protease. Angel et al. (2011) and Walk et al. (2019), reported no significant dose response to increasing levels of protease, while other studies have demonstrated significant increases (Duque-Ramírez et al. 2022). In pigs, Peng et al. (2024), reported significant improvements in average daily gain between 200 and 400 g/t of multi-protease, and a tendency for improved feed efficiency. This study is therefore a valuable addition to the literature, demonstrating for the first time a linear dose response to multi-protease in broilers.

Across the whole study (1 to 42 days), the lowest weight gains (P < 0.05) can be observed for the NC1 and NC2, as well as for NC2 + 100 g/t KZP (Table 4). This indicates that, at 8% AA reduction, the 100 g/t dose was insufficient to maintain growth rates. By contrast, at the 4% reduction in NC, 100 g/t KZP was able to significantly improve both BWG (2833 g NC vs 3011 g NC1 + 100 g/t KZP, P < 0.05) and FCR (1.707 NC1 vs 1.607 NC1 + 100 g/t KZP, P < 0.05).

Table 4.

Impacts of dietary amino acid levels and multi-protease supplementation on broiler performance for days 1–42¹.

Diet	Protease level, g/t	Feed intake, g	Weight gain, g	Feed conversion
d1-42
PC	0	4825	3080a	1.56ab
	200	4838	3149a	1.53a
NC1	0	4837	2833cd	1.70c
	100	4840	3011ab	1.60b
	200	4842	3040ab	1.59ab
	300	4836	3070a	1.57ab
NC2	0	4903	2772d	1.76d
	100	4854	2893c	1.67c
	200	4890	3081a	1.59ab
	300	4885	3102a	1.57ab
CV %		2.50	3.17	2.98
P-value		0.89	0.01	0.01
SEM		0.04	0.03	0.01

¹Means followed by different letters in the column differ statistically by the test SNK (P < 0.05).

SEM, standard error of the mean.

For feed conversion, NC1 and NC2 resulted in the worst values, with the 8% reduction in amino acids more impactful than the 4% reduction (1.567 PC vs 1.707 NC1 v 1.769 NC2, P < 0.05). The marked increase in FCR observed in NC2 indicates reduced efficiency of nutrient utilization compared with the PC diet. However, when the multi-protease was supplemented, the negative effects of the nutrient reduction were mitigated. On top of the NC1 diet, 100 g/t of the multi-protease was able to restore statistical equivalence to negative control, with further numerical improvements at 200 and 300 g/t. On top of the NC2 diet, statistical equivalence to the PC diet was achieved at 200 g/t KZP. This response in broiler chickens is consistent with the findings of Mohammadigheisar and Kim (2018), Wealleans et al. (2017a), and Zavelinski et al. (2024). They observed that adding protease to low-protein diets formulated with reduced levels of crude protein and amino acids significantly mitigated the negative effects typically associated with such nutritional restrictions in broiler chickens, resulting in improved weight gain and reduced FCR. Similar linear effects of multi-protease supplementation across the whole study can be observed in Figs. 3 and 4 for NC1 and NC2, respectively.

Figure 3.

Weight gain from 1 to 42 days of age according to the levels of protease inclusion with 4% nutritional reductions (NC1).

Figure 4.

Weight gain from 1 to 42 days of age according to the levels of protease inclusion with 8% nutritional reductions (NC2).

Moreover, diets with reduced crude protein levels, especially when supplemented with enzymes or amino acids, can significantly decrease nitrogen excretion, aligning this nutritional approach with current sustainability goals in poultry production (Cho et al. 2024).

Carcass yield

Carcass yield is closely linked to the adequate nutrition of broiler chickens, as when animals have an adequate nutrient intake, they will efficiently deposit muscle (Mendes et al. 2003). In the present study, no significant effects (P > 0.05) of the experimental treatments on the carcass characteristics of the birds were observed (Table 5), including no significant effect of the nutritional reductions. However, numerically, as the level of protease in the chicken’s diet increased, a linear trend towards greater carcass yield and meat cuts was observed. Cho et al. (2020) observed increased carcass weights and breast yields with multi-protease in both adequate and reformulated diets. Xu et al. (2017) found that dietary multi-protease supplementation increased breast muscle weight.

Table 5.

Impacts of dietary amino acid levels and multi-protease supplementation on broiler carcass characteristics at 42 days of age.

Diet	Protease level, g/t	Carcass yield %
		Carcass	Breast	Thigh + drumstick	Wings
PC	0	68.60	39.90	30.39	10.77
	200	70.60	38.59	30.48	10.90
NC1	0	69.10	38.09	30.09	11.05
	100	66.11	38.00	31.10	10.85
	200	69.43	40.98	30.50	11.08
	300	69.50	39.49	30.20	10.78
NC2	0	68.56	39.10	29.98	10.58
	100	69.90	38.76	30.31	10.98
	200	69.30	39.37	30.21	11.20
	300	70.00	40.10	30.31	10.50
CV %		3.86	6.10	6.20	6.75
P		0.06	0.08	0.80	0.53
SEM		0.88	0.80	0.62	0.24

SEM, standard error of the mean.

As with performance studies, carcass response to protease supplementation is variable. For example, Rehman et al. (2018) and Alhotan et al. (2024) did not notice any significant change in carcass yield or abdominal fat, while Law et al. (2015) found that mono-component protease supplementation increased carcass yields regardless of the level of crude protein in the diet. Similarly, in low-protein diets, protease supplementation also showed a significant effect, improving carcass yield (Vojtěch et al. 2014) while Wealleans et al. (2024b), Cho et al. (2020), Xu et al. (2017) and Akter et al. (2022), found that the use of multi-protease was able to restore carcass yield in amino acid deficient diets to a greater extent than monocomponent proteases, with particular increases in breast weight.

Overall, no significant effects of protease supplementation on carcass yield were observed in the present study. Although positive effects on carcass traits have been reported in some studies, responses appear to depend on dietary composition, enzyme characteristics, and experimental conditions. Further research is needed to better understand the factors influencing the effectiveness of protease supplementation on carcass traits in broilers.

Intestinal morphology

There was a significant difference among treatments for jejunal intestinal morphology (Table 6). Villus height was improved by protease supplementation, particularly in birds fed amino acid-reduced diets. The positive response observed in the NC2 treatments suggests that protease supplementation helped preserve intestinal morphology and absorptive capacity under conditions of greater amino acid restriction. Lu et al. (2020) and Yaseen et al. (2024) also observed greater villus height when supplementing protease in broiler diets.

Table 6.

Villus height, crypt depth, and villus: crypt ratio of the jejunum in broilers at 42 days of age according to treatments.

Treatments	Villus height, µm	Crypt depth, µm	Villus: crypt ratio
1.PC	628.42c	105.51ab	5.99 cd
2.PC + 200 g/t	655.93c	102.72ab	6.40bcd
3.NC1–4% aas	609.96c	103.64ab	5.90 cd
4.NC1 + 100 g/t	729.41b	117.82a	6.21 cd
5.NC1 + 200 g/t	640.17c	95.17b	6.72bcd
6.NC1 + 300g/t	658.26c	95.06b	6.92bc
7.NC2–8% aas	607.63c	110.53a	5.53d
8.NC2 + 100 g/t	748.39b	99.05b	7.54b
9.NC2 + 200 g/t	740.50b	97.10b	7.63b
10.NC2 + 300 g/t	950.71a	97.48b	9.75a
P	0.01	0.05	0.01
CV %	8.30	5.30	5.55
SEM	5.30	3.90	0.39

Different letters in the same column indicate significant differences according to the SNK test at 5%.

SEM, standard error of the mean.

The treatments NC1 + 100 g/t and NC2—8% AAs presented greater crypt depth, whereas the other treatments, particularly NC1 + 200 g/t, NC1 + 300 g/t, NC2 + 100 g/t, NC2 + 200 g/t, and NC2 + 300 g/t, showed shallower crypts.

The villus: crypt ratio also varied significantly among treatments. The villus: crypt ratio differed significantly among treatments. Birds fed the NC2 diet supplemented with 300 g/t of protease exhibited the highest villus: crypt ratio, while NC2 + 100 g/t and NC2 + 200 g/t also showed higher values than the control treatments. In contrast, PC, NC1—4% AAs, and NC2—8% AAs presented lower villus: crypt ratios, indicating that protease supplementation improved intestinal morphology, particularly under the greater amino acid reduction. The increased villus ratio observed in protease-supplemented birds was associated with taller villi and a more favorable intestinal structure, suggesting an enhanced capacity for nutrient absorption in the jejunum. Similar improvements in intestinal morphology following enzyme supplementation have been reported by Hazrati et al. (2020). Therefore, the higher villus ratios observed in the protease-supplemented treatments indicate that the enzyme helped maintain intestinal integrity and absorptive potential in birds fed amino acid-reduced diets, supporting the relationship between intestinal morphometry and gut functionality described by Choct (2006).

Similarly, Sun et al. (2025) also reported greater villus height and villus: crypt (V: C) in broilers supplemented with the highest level of Bacillus subtilis producing protease (3.60 × 10⁸ CFU/kg) in the diet.

These positive effects are attributed to the enhanced availability of digestible peptides promoted by protease activity, which supports villus development and intestinal absorptive capacity, thereby contributing to improved intestinal health in broilers (Hao et al. 2025).

The observed increase in villus height and villus: crypt ratio in birds supplemented with multi-protease indicates a functional enhancement of the intestinal mucosa. These structural modifications reflect a larger absorptive surface area and a lower energetic cost for tissue renewal, allowing better utilization of nutrients released by enzymatic activity. According to Cho et al. (2020) and Cowieson et al. (2018), improved epithelial integrity is directly associated with enhanced feed efficiency and weight gain, as it reduces endogenous nutrient losses and supports active amino acid transport. Thus, the morphological modulation promoted by multi-protease not only sustains digestion but also acts as a central axis of nutritional conversion efficiency in protein-reduced diets.

There was a significant difference in villus height among treatments for ileal morphology (P < 0.0001) (Table 7). Protease supplementation influenced villus height, with the highest value observed in the NC1 + 100 g/t treatment. Intermediate responses were observed in PC + 200 g/t, NC1—4% AAs, NC1 + 200 g/t, and NC1 + 300 g/t, whereas the control diet and most NC2 treatments showed lower values. Overall, these results suggest a positive effect of protease supplementation on intestinal morphology; however, they do not support the identification of an optimal enzyme inclusion level. The highest villus height values were observed in birds fed the NC1 + 100 g/t protease diet, although similar results were also found for the PC + 200 g/t, NC1—4% AAs, and NC1 + 200 g/t treatments. These results suggest that moderate protease supplementation under a mild amino acid reduction may contribute to improved intestinal morphology.

Table 7.

Villus height, crypt depth, and villus: crypt ratio of the ileum in broilers at 42 days of age according to treatments.

Treatments	Villus height, µm	Crypt depth, µm	Villus: crypt ratio
1.PC	477.59c	91.77a	5.20b
2.PC + 200 g/t	506.19ab	86.14ab	5.87ab
3.NC1–4% aas	530.88ab	97.36a	5.45b
4.NC1 + 100 g/t	567.57a	78.25bc	7.25a
5.NC1 + 200 g/t	532.22ab	78.51bc	6.78a
6.NC1 + 300g/t	442.53bc	75.35ab	5.87ab
7.NC2–8% aas	467.04c	88.37ab	5.29b
8.NC2 + 100 g/t	476.47c	86.56ab	5.54ab
9.NC2 + 200 g/t	415.79bc	74.15c	5.62ab
10.NC2 + 300 g/t	442.19bc	75.5c	5.85ab
P	0.01	0.01	0.01
CV %	7.97	5.01	6.46
SEM	6.30	4.28	0.38

Different letters in the same column indicate significant differences according to the SNK test at 5%.

SEM, standard error of the mean.

Crypt depth was significantly affected by the treatments (P < 0.01). Greater crypt depth values were observed in the PC and NC1—4% AAs treatments; however, these treatments did not differ from the PC + 200 g/t, NC1 + 300 g/t, NC2—8% AAs, and NC2 + 100 g/t treatments, which showed intermediate results. Lower crypt depth values were observed in birds receiving NC1 + 100 g/t, NC1 + 200 g/t, NC2 + 200 g/t, and NC2 + 300 g/t diets. These results suggest that protease supplementation may reduce intestinal cell turnover and improve intestinal health by decreasing epithelial renewal activity.

The villus: crypt ratio differed among treatments (P < 0.01). The highest values were observed in birds receiving NC1 diets supplemented with 100 or 200 g/t protease; however, the PC + 200 g/t treatment also showed similar intermediate results. In addition, improvements were observed in NC2 diets supplemented with protease at all inclusion levels compared with the unsupplemented NC2 treatment. These results suggest that protease supplementation may contribute to improved intestinal functionality and absorptive capacity.

Similar results in ileal morphology were reported by Lee et al. (2023), where protease supplementation at 50 g/t in broiler diets increased villus height, reduced crypt depth, and improved the V:C ratio. Tajudeen et al. (2022) also observed greater villus height and higher V:C ratios with protease supplementation in broilers. In a similar study, Sun et al. (2025) found improved ileal V:C ratio following protease addition. Moreover, studies by Kamel et al. (2015) and Nabizadeh et al. (2017) demonstrated that exogenous enzyme supplementation positively contributes to intestinal morphology in poultry. Consistently, protease supplementation in the present study may have contributed to improvements in intestinal morphology, resulting in enhanced intestinal structure in broilers.

The differences observed in the jejunum and ileum are consistent with previous findings, as the values in the jejunum are higher due to the intense digestion and absorption of nutrients, which are greater than in the ileum.

Intestinal microbiome

Observed Richness, Shannon alpha diversity, and Bray–Curtis PCoA were used to evaluate taxonomic-level microbiome responses to KZP supplementation under reduced crude protein diets (Fig. 5). Analyses were performed separately at the phylum (Fig. 5a and b), family (Fig. 5c and d), and species (Fig. 5e and f) levels, with KZP dose treated as a continuous variable for alpha-diversity regression. Across all taxonomic levels, neither Observed Richness nor Shannon diversity dose-response regressions reached statistical significance, indicating that KZP did not produce statistically confirmed linear shifts in alpha diversity within the available dataset. The absence of significant effects on alpha diversity suggests that KZP supplementation did not substantially alter overall microbial richness or evenness. Similar findings were reported by Lourenco et al. (2020), who observed limited effects of dietary protease supplementation on alpha-diversity indices despite detecting changes in microbial composition.

Figure 5.

Taxonomic-level microbiome diversity responses to kemzyme protease (KZP) supplementation in broilers fed diets with different crude protein levels. Alpha diversity was evaluated using observed richness and shannon diversity indices at the phylum (a), family (c), and species (e) levels. Beta diversity was assessed using bray–curtis dissimilarity and visualized by principal coordinates analysis (PCoA) at the phylum (b), family (d), and species (f) levels. Regression lines represent dose-response relationships within reduced-protein diets (NC1 and NC2), while PCoA plots illustrate descriptive shifts in microbial community composition across KZP supplementation levels.

At the phylum level (Fig. 5a), Observed Richness showed only minor variation across KZP doses in both NC1 and NC2. Similarly, Shannon diversity remained low in NC1 across increasing KZP doses. In contrast, NC2 showed a stronger positive Shannon trend, driven largely by the 300 g/t treatment, although this response was not statistically significant. Bray–Curtis PCoA (Fig. 5b) indicated that NC1 samples remained tightly grouped and close to the PC baseline structure, whereas NC2 exhibited a clearer trajectory, with the 300 g/t treatment separating from the lower-dose NC2 samples.

These results indicate that while dominant phyla remained relatively stable, subtle changes in community structure may have occurred under greater protein restriction. Previous studies have shown that dietary protein concentration often exerts a stronger influence on gut microbiota composition than enzyme supplementation alone, with reductions in crude protein altering substrate availability and consequently reshaping microbial populations (Apajalahti and Vienola 2016).

At the family level (Fig. 5c), both Observed Richness and Shannon diversity displayed the strongest positive descriptive trends in NC2, while NC1 showed only modest increases across KZP doses. However, none of these relationships reached statistical significance. Family-level Bray–Curtis PCoA (Fig. 5d) revealed greater dispersion than the phylum-level analysis, suggesting that KZP-associated microbiome shifts were more apparent at finer taxonomic resolution. Within NC2, the 100 and 200 g/t treatments shifted away from the baseline, whereas the 300 g/t treatment occupied a distinct ordination position, indicating a dose-associated restructuring of family-level community composition. The greater separation observed at the family level may be explained by changes in substrate availability within the gastrointestinal tract. Exogenous proteases improve protein hydrolysis and amino acid digestibility, potentially reducing the amount of undigested protein reaching the distal intestine. According to Borda-Molina et al. (2019), such changes can modify microbial community composition without necessarily affecting overall diversity metrics.

At the species level (Fig. 5e), Observed Richness tended to increase in NC2 but decrease in NC1 with increasing KZP supplementation. In contrast, Shannon diversity showed negative trends in both NC1 and NC2, although neither relationship was statistically significant. Species-level PCoA (Fig. 5f) showed more complex treatment trajectories than those observed at the phylum level, suggesting that species-level compositional shifts occurred even when overall alpha-diversity responses were modest. Because Shannon diversity incorporates both richness and evenness, the observed decline may indicate that specific bacterial species became relatively more dominant as KZP inclusion increased.

Overall, the data indicate that KZP supplementation produced descriptive taxonomic-level microbiome shifts, particularly in the NC2 reduced crude protein group. Although these changes were not supported by statistically significant alpha-diversity regressions, they were more evident in the Bray–Curtis PCoA trajectories, especially at the family and species levels, where treatment profiles showed clearer separation across KZP doses. These findings suggest that KZP may modulate microbial community composition under reduced crude protein conditions; however, the current dataset should be interpreted descriptively because treatment-level replication was limited and the PCoA ordinations were not assigned inferential p-values.

Similarly, Yi et al. (2024) demonstrated that dietary protease supplementation altered cecal microbiota composition and microbial metabolism in broilers, supporting the hypothesis that exogenous proteases may influence microbial ecology indirectly through changes in nutrient availability. Therefore, although statistical evidence was limited in the present study, the consistent separation observed in PCoA analyses suggests that KZP may contribute to microbiome modulation under reduced crude protein conditions, particularly under more severe protein reductions such as NC2.

Conclusions

In this study, supplementation with a multi-protease improved the growth performance, gut morphology and modulated the intestinal microbiome, particularly under conditions of greater crude protein reduction, where more pronounced shifts in microbial community structure were observed. Using multi-protease supplementation is especially recommended under nutritional conditions with reduced amino acid levels.

Funding

This study was financially supported by Kemin Industries, Inc.

Conflicts of interest

The authors declare that they have no conflict of interest.

Data availability

All data sets obtained and analyzed during the current study are available in the manuscript.