Dual-strain Bacillus subtilis 597 and 600 complex improves productive performance and nutrient digestibility, egg quality and physiological status in early-to-peak phase laying hens

Abstract

Efficacy of a dual-strain Bacillus subtilis complex (597 and 600) on productive performance, egg quality, apparent ileal digestibility (AID), serum parameters, selected cecal bacterial populations, tibia traits, and economic efficiency in laying hens was investigated. A total of 324 Hy-Line Brown hens at 18 wk of age were randomly assigned to two dietary treatments with 18 replicates of 9 birds per cage for a 22-wk trial divided into Phase 1 (wk 18 to 29) and Phase 2 (wk 30 to 40). Diets included a control diet (CON) and CON supplemented with 0.01% probiotic complex (PRO; 1.3 × 10⁶ CFU/g feed). During Phase 1, productive performance was unaffected (P > 0.10). During Phase 2, PRO increased hen-day egg production, egg weight, and mass, with improved feed conversion ratio and egg loss rate (P < 0.05). PRO enhanced eggshell breaking strength, darkened shell color, and reduced yolk cholesterol across both phases (P < 0.01), whereas albumen height, Haugh unit, and shell thickness increased during Phase 2 (P < 0.05). At wk 29 and 40, PRO increased ileal villus height and villus height to crypt depth ratio (P < 0.05). At wk 40, PRO increased the AID of DM, gross energy, CP, Ca, and P (P < 0.05). Serum triglycerides and very low-density lipoprotein cholesterol were lower in PRO across both phases, with lower total cholesterol and corticosterone and elevated serotonin at wk 40 (P < 0.05). Selected cecal bacterial populations showed higher Lactobacillus spp. and Bifidobacterium spp., with reduced Clostridium spp. (P < 0.05) and Escherichia coli (P < 0.001, wk 40). Tibia weight, breaking strength, and Ca content were improved in the PRO at wk 29, with additional improvements in tibia density and ash content by wk 40 (P < 0.05). Economically, PRO increased egg income and return on investment during Phase 2 (P < 0.05). Collectively, dual-strain B. subtilis complex improved hen productive performance, and egg quality by modulating the gut environment and nutrient digestibility.

Introduction

Modern layers have been genetically selected for high egg production, targeting approximately 500 eggs over a 100-wk production cycle (Hanlon et al., 2022). Achieving this extended persistence depends on successfully managing the critical transition from the pullet phase to onset of lay. This early-lay period represents a profound physiological challenge, during which hens undergo drastic hormonal reorganization, rapid follicular development, and a surge in metabolic demand and nutrient turnover for egg production (Ponsuksili et al., 2023; Xu et al., 2024). It is reasoned that priming the intestinal mucosal barrier and establishing a resilient gut environment during this transitional window is imperative to ease the metabolic shift and safeguard early lay performance. Consequently, nutritional strategies designed to prime gut functionality and reinforce physiological resilience from onset of lay have gained considerable attention in alleviating management transition and combating physiological stressors (Kiarie et al., 2024).

Among dietary strategies, probiotics (i.e., direct-fed microbials) have emerged as promising tools to enhance gut health and performance without conventional antimicrobial growth promoters (AGPs) (Abd El‐Hack et al., 2020; Jha et al., 2020). Their functional benefits involve complex host-microbe interaction, including competitive exclusion, microbial cross-feeding, and pathogen antagonism (Gaggìa et al., 2010; Oketch and Heo, 2025). Spore-forming Bacillus subtilis is particularly advantageous due to its stability during feed processing and gastric transit (Forte et al., 2016; Ramlucken et al., 2020). In addition, B. subtilis secrete extracellular enzymes and antimicrobial peptides that can improve nutrient utilization and mineral utilization (e.g., Ca and P), which support skeletal integrity and eggshell quality (Bilal et al., 2021; Khogali et al., 2022; Yaqoob et al., 2022).

Probiotic efficacy, however, is highly strain-dependent (Jha et al., 2020; Oketch et al., 2024). It is reported that B. subtilis 597 directly modulates host immune signaling independently of microbiota changes (Winther et al., 2024), whereas B. subtilis 600 exhibits a distinct carbohydrase profile targeting indigestible oligosaccharides and non-starch polysaccharides (Sandvang et al., 2021). Combining these complementary strains is targeted to synergistically enhance nutrient utilization and gut health (Teo and Tan, 2005; Sandvang et al., 2021). Nevertheless, data regarding the dual-strain B. subtilis 597 and 600 complex in laying hens during early-to-peak lay remain limited.

Therefore, this study aimed to investigate the effects of a dual-strain probiotic complex containing B. subtilis 597 and 600 on productive performance, egg quality, apparent ileal digestibility (AID), blood biochemical profiles, intestinal histomorphology, selected cecal bacterial populations, tibia characteristics, and economic outcomes in early-to-peak phase laying hens. It was hypothesized that probiotic supplementation would favorably modulate the intestinal ecosystem and nutrient utilization capacity, potentially contributing to improved physiological resilience, egg quality, and productive performance.

Materials and methods

The experiment was approved by the Animal Ethics Committee of Chungnam National University (Protocol Number; 202510A-CNU-231). The study was conducted at the Cheongyang Research Station affiliated with Chungnam National University.

Birds, diets, and housing

A total of three hundred and twenty-four Hy-Line Brown laying hens were sourced from a commercial supplier (Yulgeuk Livestock Farm, Yeoju-si, Gyeonggi-do, Republic of Korea). Upon arrival, birds were individually weighed to verify body weight uniformity. Birds were distributed into 36 cages (9 hens per cage) in a completely randomized design with 18 replicates per treatment. The birds were then allowed a two-week adaptation period where the egg production was monitored and subsequently used to allocate the birds. After a two-wk adaptation period (wk 16 to 18), hens were provided with basal diets in mash form designed based on a two-phase feeding program (Phase 1: wk 18 to 29; Phase 2: wk 30 to 40). The basal diet (CON) met the recommended requirements of Hy-Line International (2025). The ingredient and nutrient composition of the diets are presented in Table 1. The probiotic group (PRO) was prepared by incorporating the CON with 0.01% probiotics in powder form (GalliPro Form 5; Novonesis, Copenhagen, Denmark). The test product, GalliPro® Form 5, is a commercially available probiotic containing viable spores of naturally occurring B. subtilis strains 597 and 600 at an approximately 1:1 ratio, with a guaranteed minimum concentration of 1.3 × 10¹⁰ CFU/g, providing a final concentration of approximately 1.3 × 10⁶ CFU/g in feed. To ensure uniform distribution of the probiotic complex at 0.01% in the final diet, a small portion of the basal diet was first pre-mixed with the probiotic powder before being thoroughly blended into the remaining batch using a mechanical feed mixer.

Table 1.

Ingredients and chemical composition of basal diets.

Ingredients	Amount, %
	Phase 1 (18 to 29 weeks of age)	Phase 2 (30 to 40 weeks of age)
Corn	50.02	50.02
Wheat bran	5.35	6.72
Soybean meal, 44%	28.38	27.61
Vegetable oil	3.98	3.45
Limestone	9.71	9.74
Mono-calcium phosphate	1.28	1.28
Salt	0.50	0.50
Lys-HCl	0.05	0.02
DL-methionine	0.20	0.18
Threonine	0.05	0.04
Tryptophan	0.02	0.00
Cysteine	0.09	0.09
Valine	0.07	0.05
Vitamin-mineral premix1	0.30	0.30
Calculated nutrient composition, %
ME, kcal/kg	2850	2800
Crude protein	17.80	17.60
Crude fat	2.25	2.29
Crude fiber	3.66	3.76
SID Lysine	0.88	0.84
SID Methionine	0.44	0.42
SID Methionine + Cysteine	0.78	0.76
SID Threonine	0.60	0.59
SID Tryptophan	0.20	0.18
SID Valine	0.77	0.74
Calcium	4.18	4.20
Total P	0.72	0.72
Available P	0.39	0.39

ME, metabolizable energy; SID, standardized ileal digestibility; Ca, calcium; P, phosphorus.

Ingredient and additive prices: The calculated basal diet costs were $0.3138/kg for Phase 1 (wk 18–29) and $0.3201/kg for Phase 2 (wk 30–40). The probiotic additive (GalliPro Form 5) was priced at $15.67/kg. Based on the 0.01% (100 g/ton) top-dressing inclusion rate, the additive contributed an additional cost of $0.001567/kg ($1.567/ton) to the PRO diet.

¹Provided per kilogram of diet: vitamin A, 12,000 IU (trans-retinyl acetate); vitamin D₃, 2,500 IU (cholecalciferol); vitamin E, 30 IU (DL-α-Tocopherol acetate); vitamin K₃, 3 mg; D-pantothenic acid, 15 mg; nicotinic acid, 40 mg; choline, 400 mg; and vitamin B₁₂, 12 μg; Fe (from iron sulfate), 90 mg; Cu (from copper sulfate), 8.8 mg; Zn (from zinc oxide), 100 mg; Mn (from manganese oxide), 54 mg; I (from potassium iodide), 0.35 mg; Se (from sodium selenite), 0.30 mg.

Regarding hen housing and management, animals and facilities were inspected twice daily for general health status, animal welfare, mortalities, constant water supply, temperature, lighting, humidity, ventilation, and unexpected events. Hens were maintained in a controlled facility with temperature and humidity maintained at around 20 to 22°C and 40 to 60% respectively, under a 16L:8D photoperiod. The birds were housed in enriched cages (90 × 90 × 90 cm) furnished with perches and nesting areas, complying with European Union standards (European Commission, 1999). Each cage was equipped with four nipple drinkers and a detachable feeder, providing ad libitum access to allocated feed and water throughout the experimental period.

Experimental procedures and sample collection

Individual body weight was recorded at wk 18, 29, and 40 to assess probiotic effects on body weight. Productive performance parameters, including hen-day egg production (HDEP), egg weight, and feed intake (FI), were assessed on a bi-weekly basis. Hen’s FI was determined by subtracting the residual feed from the total amount offered. Feed conversion ratio (FCR) and egg mass were subsequently calculated. The egg loss rate was derived by counting unmarketable eggs (i.e., dirty, rough, misshapen, pimpled, cracked, white-banded, pale-shelled, soft-shelled, or shell-less) classified according to Nannoni et al. (2022). For egg analysis, 4 eggs per replicate cage (n = 72) were randomly sampled at wk 21, 25, 29, 33, 37, and 40, excluding unmarketable eggs. Internal quality parameters (i.e., albumen height, Haugh unit, yolk color, yolk percentage, and albumen percentage) and external quality parameters (i.e., eggshell color, eggshell breaking strength, eggshell thickness, and eggshell percentage) were measured as stipulated by Oketch et al. (2026).

To evaluate AID for nutrients, diets were mixed with chromic oxide (0.3%) for one wk prior to sampling at wk 29 and 40. One bird representing the median body weight of each cage (n = 36) was selected for sampling. Blood was then collected from the brachial vein into non-heparinized tubes for serum analysis immediately prior to euthanasia via carbon dioxide asphyxiation (Ravindran et al., 2017). Following abdominal incision, a segment of the mid-ileum (distal to Meckel’s diverticulum) was excised, and gently rinsed with phosphate-buffered saline at pH 7.4. Ileal tissue samples were then placed in labeled containers containing 10% formaldehyde and stored for subsequent histomorphological analysis, including villus height (VH), villus width, crypt depth (CD), villus height to crypt depth ratio (VH:CD ratio), and villus absorptive surface area (VSA). After tissue sampling, ileal digesta from birds receiving the same dietary treatment were gently flushed with distilled water, collected per cage, and stored at −20°C until analysis of AID of DM, gross energy, CP, Ca, and P. Fresh cecal digesta were collected aseptically from each sacrificed hen, placed in sterile containers, and stored at −80°C for subsequent enumeration of specific cecal microbiota. Additionally, the left tibia was excised from each selected hen, cleaned of adherent soft tissue, and dried for subsequent tibia analyses according to the procedure described by Abdelqader et al. (2013).

Sample processing and laboratory analysis

Within 24 h of collection, egg quality parameters were analyzed. Breaking strength was determined using a texture analyzer (TA.XTplusC, Stable Micro Systems, Godalming, Surrey, UK), and Haugh unit, albumen height and shell color were measured using an egg multi-tester (QCM+ Range, TSS, Dunnington, York, UK). Following the recommendation of Jones (2012), albumen height was measured at a distance of at least 1 cm from the yolk. Yolk color was scored using a DSM Yolk Color Fan on a scale from 1 (light yellow) to 15 (orange). Specific gravity was assessed via the saline flotation method described by Peebles (2004). For component analysis, the yolk, albumen, and eggshell were separated and weighed to calculate their relative percentages. Yolk cholesterol was analyzed according to Yalçın et al. (2012). For eggshell ratio determination, eggshells were rinsed off adhering albumen, air-dried at room temperature for 24 h, and weighed. Eggshell thickness (with intact membranes) was measured at sharp, equatorial, and blunt regions using a micrometer (Digimatic MDC-MX Series, Mitutoyo, Aurora, IL, USA).

Blood samples were centrifuged at 3,000 × g for 10 min at 4°C using a LABOGENE 1248R (Gyrozen Co., Ltd., Daejeon, Korea). The separated serum was stored at −80°C in the deep freezer (UniFreez U 400, DAIHAN Scientific Co., Ltd., Wonju, Korea) prior to analysis. An automated biochemistry analyzer (Hitachi 7020, Tokyo, Japan) was employed to determine the concentration of high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-c), triglycerides (TG), and alkaline phosphatase (ALP). Very low-density lipoprotein cholesterol (VLDL-c) and total cholesterol (TC) were subsequently determined by calculation. Additionally, specific markers including calcitonin, serotonin, corticosterone, total antioxidant capacity (T-AOC), and malondialdehyde were quantified using chicken-specific ELISA kits (MyBioSource, San Diego, CA, USA) following the manufacturer’s protocols.

Fixed ileal tissues were dehydrated and embedded in paraffin wax. Transverse sections (5 µm thickness) were prepared, stained with hematoxylin and eosin, and visualized using a light microscope (Olympus CX23, Olympus Corporation, Tokyo, Japan). Morphometric indices, including VH, villus width, and CD, were determined using image analysis software (NIS-Elements Viewer, version 4.20; Nikon, USA). Measurements were taken from 10 intact villi exhibiting a distinct lamina propria per sample at 100 × magnification, and the mean value was used as the experimental unit. One bird per cage was analyzed, resulting in 18 experimental units per treatment at each sampling time. To determine AID, frozen digesta samples were thawed and oven-dried at 55°C for 24 h. The dried samples were subsequently ground to pass through a 0.75-mm sieve using an ultra-centrifugal mill (ZM 200, Retsch GmbH & Co., KG, Haan, Germany). Diet and ileal digesta samples were analyzed for DM, gross energy, CP, Ca, P, and chromic oxide concentrations. Dietary dry matter (DM) was determined by oven drying at 105°C to constant weight according to AOAC (2005). Crude protein was analyzed using AOAC method 990.03, and gross energy was determined using bomb calorimetry. Calcium and P concentration were determined according to AOAC (2005) method 985.01. The concentration of chromic oxide was quantified following the method of Fenton and Fenton (1979).

To assess the counts of Lactobacillus spp., Bifidobacterium spp., Clostridium spp., Enterococcus spp., Escherichia coli (E. coli), and coliforms in cecal contents, a fresh cecal digesta sample was homogenized in 0.85% sterile saline at a 1:9 (w/v) ratio for 2 min using a stomacher (Digital Power Embossing LS-400, BNF KOREA, Gimpo, Korea). Then, a series of 10-fold serial dilutions was prepared using 0.85% sterile saline, and 0.1 mL of each dilution was inoculated into appropriate agar media, and 1 mL of each dilution was inoculated into petrifilm plates (Petrifilm™ 3 M, St. Paul, MN, USA). Lactobacillus spp., Bifidobacterium spp., and Clostridium spp. were cultured on Lactobacillus Selection Agar supplemented with glacial acetic acid (MB Cell, Seoul, Korea), TOS-MUP agar (MB Cell, Seoul, Korea) containing MUP supplement (MB Cell, Seoul, Korea), and Reinforced Clostridial Agar (MB Cell, Seoul, Korea), respectively. Lactobacillus spp., Bifidobacterium spp., and Clostridium spp. plates were placed into anaerobic jars supplemented with gas-generating sachets (AnaeroGen 2.5 L; Oxoid, Basingstoke, UK) and incubated under anaerobic conditions at 37°C for 48 h. Enterococcus spp., E. coli, and coliforms were cultured on Enterococcosel Agar (MB Cell, Seoul, Korea), Eosin Methylene Blue agar (MB Cell, Seoul, Korea), and petrifilm coliform count plates, respectively, and incubated under aerobic conditions at 37°C for 24 h. The results were expressed as log CFU/g.

The left tibia was obtained, dried using adsorbent paper, weighed, and its volume was quantified using the water displacement technique. Briefly, the bone was immersed in a graduated cylinder containing a known volume of distilled water, and the volume displacement was recorded from the difference between the initial and final volume (Oketch et al., 2024). Tibia density was calculated as the ratio of weight to volume. Subsequently, tibia breaking strength was measured using a universal testing machine (UTA-500 N; Yeonjin S-Tech, Yongin-si, Korea) employing a 3-point bending test to measure tibia resistance to bending or breakage. Total ash, Ca, and P concentrations were determined according to AOAC (2005) procedures; method 942.05 for ash and 985.01 for Ca and P.

Statistical analysis

All data were analyzed using IBM SPSS Statistics (version 29.0; IBM Corp., Armonk, NY, USA). For parameters involving repeated measurements over time, including body weight, productive performance, and egg quality, a linear mixed model was used. The model included dietary treatment, bird age (wk), and their interaction as fixed effects, with cage included as the experimental unit and random effect. Repeated measures within the same cage over time were modeled using a first-order autoregressive covariance structure [AR(1)]. Initial body weight was included as a covariate when appropriate. When the interaction between treatment and age was significant (P < 0.05), Tukey’s post-hoc test was used for mean comparisons. In the absence of significant interaction, main effects of treatment and age were interpreted. Percentage data, including hen-day egg production and egg loss rate, were checked for normality and homogeneity of variance; arcsine square-root transformation was applied when required. As transformed and untransformed analyses produced similar interpretations, untransformed means are presented for clarity.

For parameters evaluated specifically at the conclusion of each feeding phase (at wk 29 or 40) including blood biochemical indices, selected cecal bacterial populations, intestinal morphology, AID, and tibia traits, the data were analyzed separately within each time point using an independent t-test to compare the means between the CON and PRO. The cage was defined as the experimental unit for productive performance, egg quality, and economic analysis. For selected cecal bacterial populations, intestinal morphology, AID and tibia traits, a single bird selected per cage (one bird per cage) served as the experimental unit. Statistical significance was declared at P < 0.05, and tendencies were considered at 0.05 ≤ P < 0.10.

Results

Growth and productive performance

No differences were observed in body weight between the CON and PRO groups throughout the experimental period (P > 0.10). An age effect was observed, with body weight increasing as the birds aged (P < 0.001). There was no interaction between treatment and age for body weight (P > 0.10). In phase 1, dietary treatments did not affect HDEP, egg loss rate, egg weight, egg mass, FI, or FCR (P > 0.10). All evaluated productive parameters were influenced by the age of production (P < 0.001). No interactions between treatment and age were observed for any parameters during this phase (P > 0.10). In phase 2, the PRO group had higher laying performance (P = 0.037), egg weight (P = 0.020), and egg mass (P < 0.001) compared with the CON group. The egg loss rate (P = 0.012) and FCR (P = 0.008) were lower in the PRO-fed layers compared with those fed CON. Feed intake did not differ between the treatment groups (P > 0.10). An age effect was observed for egg loss rate (P < 0.001), egg weight (P < 0.001), egg mass (P = 0.041), FI (P = 0.001), and FCR (P < 0.001). There were no interactions between treatment and age for any of the productive performance parameters during phase 2 (P > 0.10).

Internal egg and eggshell quality

In phase 1, interactions between treatment and age were observed for shell thickness (P = 0.003), shell percentage (P = 0.045), and yolk cholesterol (P < 0.001) (Table 4). Specifically, the PRO maintained greater shell thickness at wk 21 and 25 (P < 0.05) but not at wk 29 (P > 0.05), whereas yolk cholesterol was lower in the PRO group across all measured ages in Phase 1 (P < 0.05). For parameters without significant interactions, the PRO-fed birds had darker shell color (P < 0.001) and greater breaking strength (P = 0.004) compared with the CON group. Albumen height (P = 0.062) and yolk color (P = 0.086) tended to be higher in PRO than in CON. An age effect was observed for shell color, percentage and thickness; albumen height; Haugh unit and yolk color, percentage, and cholesterol (P < 0.05).

Table 4.

Effect of the dual-strain Bacillus subtilis supplementation on egg quality of laying hens.

Items		Shell color1	Albumen height (mm)	Haugh unit2	Yolk color	Shell thickness (mm)	Shell breaking strength (kg)	Yolk percentage (%)	Albumen percentage (%)	Shell percentage (%)	Yolk cholesterol (mg/g yolk)
Phase 1
Week 21	CON3	24.06	9.24	97.55	6.92	0.36b	5.68	24.08	60.03	15.89a	13.60a
	PRO4	22.07	9.42	98.01	7.03	0.38a	5.80	23.63	60.24	16.13a	11.56bc
Week 25	CON	24.26	9.18	96.02	6.57	0.36b	5.66	25.37	60.66	13.86b	11.87b
	PRO	23.21	9.56	97.73	6.71	0.38a	5.87	24.82	61.14	13.96b	11.21cd
Week 29	CON	25.97	8.76	93.50	7.58	0.36b	5.62	25.25	60.81	13.68b	11.48bc
	PRO	24.74	8.92	94.45	7.64	0.37b	5.86	25.58	60.14	13.90b	10.69d
SEM5		0.141	0.065	0.320	0.030	0.001	0.033	0.107	0.175	0.111	0.109
Main effects
Treatment	CON	24.76a	9.06	95.69	7.02	0.36	5.65b	24.90	60.50	14.48	12.32
	PRO	23.34b	9.30	96.73	7.13	0.37	5.85a	24.68	60.51	14.66	11.16
Age	Week 21	23.06b	9.33a	97.78a	6.97b	0.37	5.74	23.85b	60.14	16.01	12.58
	Week 25	23.74b	9.37a	96.87a	6.64c	0.37	5.76	25.10a	60.90	13.91	11.54
	Week 29	25.35a	8.85b	93.98b	7.61a	0.36	5.74	25.41a	60.48	13.79	11.09
P-value6
Treatment		< 0.001	0.062	0.106	0.086	< 0.001	0.004	0.305	0.985	0.411	< 0.001
Age		< 0.001	0.002	< 0.001	< 0.001	0.011	0.946	< 0.001	0.155	< 0.001	< 0.001
Treatment × Age		0.380	0.771	0.733	0.869	0.003	0.746	0.127	0.325	0.045	< 0.001
Phase 2
Week 33	CON	26.36	8.40	91.51	6.96	0.39	5.92	26.19	58.92	13.67	12.71
	PRO	25.10	8.42	91.58	6.88	0.39	6.13	26.03	59.64	14.47	11.90
Week 37	CON	27.68	8.35	90.99	7.04	0.35	5.23	30.02	56.38	13.47	12.58
	PRO	26.14	8.65	92.52	7.13	0.36	5.62	29.65	56.69	13.87	11.36
Week 40	CON	27.92	8.18	90.06	7.21	0.36	5.43	25.60	60.75	13.66	12.42
	PRO	26.36	8.66	92.16	7.26	0.37	5.66	25.64	60.36	14.02	11.79
SEM		0.134	0.060	0.308	0.026	0.001	0.030	0.113	0.259	0.122	0.156
Main effects
Treatment	CON	27.32a	8.31b	90.85b	7.07	0.36b	5.53b	27.27	58.68	13.60b	12.70a
	PRO	25.87b	8.58a	92.09a	7.09	0.37a	5.80a	27.11	58.90	14.12a	11.68b
Age	Week 33	25.73b	8.41	91.55	6.92c	0.39a	6.03a	26.11b	59.28b	14.07	12.30
	Week 37	26.91a	8.50	91.76	7.08b	0.35c	5.42b	29.84a	56.54c	13.67	11.97
	Week 40	27.14a	8.42	91.11	7.24a	0.36b	5.54b	25.62b	60.56a	13.84	12.10
P-value
Treatment		< 0.001	0.026	0.046	0.726	< 0.001	< 0.001	0.466	0.451	< 0.001	< 0.001
Age		< 0.001	0.806	0.694	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	0.069	0.312
Treatment × Age		0.892	0.290	0.391	0.311	0.252	0.349	0.756	0.316	0.379	0.392

^a-d Means with a column with different lowercase (a-d) superscripts differ significantly (P < 0.05).

¹Black, 0; white, 82.5.

²Haugh unit = 100 × log(albumen height – 1.7egg weight^0.37 + 7.6).

³Basal diet without probiotics.

⁴Basal diet with probiotics (1.3 × 10⁶ CFU/g feed).

⁵Pooled standard error of the mean.

⁶Statistical significance was determined at P < 0.05.

In phase 2, the PRO group had darker shell color values (P < 0.001) and yolk cholesterol (P < 0.001) compared with CON. Additionally, the eggs from PRO-fed hens had higher albumen height (P = 0.026), Haugh unit (P = 0.046), shell thickness (P < 0.001), shell breaking strength (P < 0.001), and shell percentage (P < 0.001) compared with the CON group. An age effect was observed for shell color, thickness and breaking strength (P < 0.001), yolk color and percentage (P < 0.001), and albumen percentage (P < 0.001). A tendency for an age effect was observed for shell percentage (P = 0.069).

Intestinal histomorphology

At wk 29, the PRO exhibited higher VH, VH: CD ratio, and VSA compared with CON (P < 0.001; see Table 5). Similarly, at wk 40, PRO maintained a higher VH (P = 0.032) and VH: CD ratio (P = 0.002) than CON.

Table 5.

Effect of the dual-strain Bacillus subtilis supplementation on ileal histomorphology of laying hens.

Item	Diets1		SEM2	P-value3
	CON	PRO
Week 29
Villus height (µm)	775.20	938.73⁎⁎⁎	23.081	< 0.001
Crypt depth (µm)	120.35	114.43	3.755	0.438
Villus width (µm)	112.09	114.93	2.419	0.565
Villus height: crypt depth ratio	6.56	8.48⁎⁎⁎	0.292	< 0.001
Villus absorptive surface area (mm2)4	0.27	0.34⁎⁎⁎	0.011	< 0.001
Week 40
Villus height (µm)	912.86	1042.76*	30.092	0.032
Crypt depth (µm)	160.22	148.67	5.439	0.297
Villus width (µm)	110.11	98.45	4.080	0.156
Villus height: crypt depth ratio	5.82	7.16⁎⁎	0.230	0.002
Villus absorptive surface area (mm2)	0.30	0.32	0.007	0.158

^⁎Within a row, means differ significantly at P < 0.05.

^⁎⁎Within a row, means differ significantly at P < 0.01.

^⁎⁎⁎Within a row, means differ significantly at P < 0.001.

¹CON, 0 CFU/g feed; PRO, 1.3 × 10⁶ CFU/g feed.

²Pooled standard error of the mean.

³Statistical significance was determined at P < 0.05.

⁴Villus absorptive surface area = 2π × (Villus width / 2) × Villus height (Oketch et al., 2025).

Apparent ileal digestibility

At wk 29, hens in PRO group tended to exhibit greater AID of Ca and P compared with those in CON (P = 0.095 and P = 0.088, respectively; Table 6). At wk 40, the PRO-fed birds exhibited higher AID for DM (P = 0.048), gross energy (P = 0.032), crude protein (P = 0.042), Ca (P = 0.015), and P (P = 0.031), relative to CON.

Table 6.

Effect of the dual-strain Bacillus subtilis supplementation on nutrient apparent ileal digestibility of laying hens.

Item	Diets1		SEM2	P-value3
	CON	PRO
Week 29
Dry matter	80.96	81.44	0.222	0.288
Gross energy	76.13	77.22	0.390	0.167
Crude protein	82.51	83.14	0.324	0.339
Calcium	68.86	70.39	0.460	0.095
Phosphorus	64.25	65.50	0.367	0.088
Week 40
Dry matter	78.84	79.87*	0.262	0.048
Gross energy	72.30	74.74*	0.576	0.032
Crude protein	80.41	81.87*	0.364	0.042
Calcium	65.04	67.91*	0.604	0.015
Phosphorus	61.54	63.72*	0.512	0.031

^⁎Within a row, means differ significantly at P < 0.05.

¹CON, 0 CFU/g feed; PRO, 1.3 × 10⁶ CFU/g feed.

²Pooled standard error of the mean.

³Statistical significance was determined at P < 0.05.

Blood biochemical indices

At wk 29, PRO had higher serum concentrations of HDL (P = 0.008), ALP (P < 0.001), and serotonin (P < 0.001) compared with CON (Table 7). Conversely, concentrations of TG (P = 0.002) and VLDL-c (P = 0.002) were lower in PRO than in the CON group. A tendency for a reduction in LDL-c concentration was observed in PRO (P = 0.054). At 40 wk of age, PRO had lower serum concentrations of LDL-c (P = 0.027), VLDL-c (P = 0.005), TC (P = 0.013), TG (P = 0.005), and corticosterone (P = 0.001) compared with CON. Serum concentrations of serotonin (P < 0.001) and T-AOC activity (P = 0.033) were higher in PRO than in CON.

Table 7.

Effect of the dual-strain Bacillus subtilis supplementation on blood parameters of laying hens.

Item	Diets1		SEM2	P-value3
	CON	PRO
Week 29
HDL-c (mg/dL)	46.36	57.15⁎⁎	2.063	0.008
LDL-c (mg/dL)	35.94	30.58	1.397	0.054
VLDL-c (mg/dL)4	47.17	35.91⁎⁎	1.907	0.002
Total cholesterol (mg/dL)5	129.47	123.64	2.875	0.318
Triglyceride (mg/dL)	235.85	179.56⁎⁎	9.533	0.002
Calcitonin (pg/mL)	0.43	0.34	0.032	0.169
Alkaline phosphatase (IU/mL)	2.92	5.30⁎⁎⁎	0.313	< 0.001
Serotonin (ng/mL)	403.61	569.41⁎⁎⁎	25.580	< 0.001
Corticosterone (ng/mL)	5.86	5.29	0.255	0.278
T-AOC activity (U/ml)	1.74	1.81	0.134	0.824
Malondialdehyde (μmol/L)	4.77	4.38	0.167	0.242
Week 40
HDL-c (mg/dL)	48.62	50.18	1.149	0.506
LDL-c (mg/dL)	84.94	71.19*	3.162	0.027
VLDL-c (mg/dL)	47.60	42.54⁎⁎	0.938	0.005
Total cholesterol (mg/dL)	181.16	163.91*	3.542	0.013
Triglyceride (mg/dL)	237.98	212.72⁎⁎	4.689	0.005
Calcitonin (pg/mL)	0.37	0.36	0.034	0.811
Alkaline phosphatase (IU/mL)	2.37	2.47	0.118	0.678
Serotonin (ng/mL)	397.91	467.31⁎⁎⁎	8.688	< 0.001
Corticosterone (ng/mL)	3.31	2.35⁎⁎	0.153	0.001
T-AOC activity (U/ml)	1.41	1.82*	0.096	0.033
Malondialdehyde (μmol/L)	2.58	2.53	0.090	0.762

HDL-c, high density lipoprotein cholesterol; LDL-c, low density lipoprotein cholesterol; VLDL-c, very low density lipoprotein cholesterol; T-AOC activity, total antioxidant capacity activity.

^⁎Within a row, means differ significantly at P < 0.05.

^⁎⁎Within a row, means differ significantly at P < 0.01.

^⁎⁎⁎Within a row, means differ significantly at P < 0.001.

¹CON, 0 CFU/g feed; PRO, 1.3 × 10⁶ CFU/g feed.

²Pooled standard error of the mean.

³Statistical significance was determined at P < 0.05.

⁴VLDL-c (mg/dL) = Triglycerides / 5 (Friedwald et al., 1972).

⁵Total cholesterol (mg/dL) = HDL-c + LDL-c + VLDL-c (Friedwald et al., 1972).

Selected cecal bacterial populations

Hens supplemented with PRO increased the populations of Lactobacillus spp. (P = 0.007) and Bifidobacterium spp. (P = 0.006) compared to CON (Table 8). Conversely, the count of Clostridium spp. was reduced in PRO (P = 0.004). A marginal reduction was observed in the population of E. coli in response to probiotics (P = 0.091). By wk 40, PRO maintained higher counts of Lactobacillus spp. (P < 0.001) and Bifidobacterium spp. (P = 0.011). Furthermore, both Clostridium spp. and E. coli populations were lower in PRO compared with CON (P < 0.001).

Table 8.

Effect of the dual-strain Bacillus subtilis supplementation on cecal microbiota of laying hens.

Item	Diets1		SEM2	P-value3
	CON	PRO
Week 29
Lactobacillus spp. (Log CFU/g)	8.86	9.63⁎⁎	0.147	0.007
Bifidobacterium spp. (Log CFU/g)	8.30	9.05⁎⁎	0.137	0.006
Clostridium spp. (Log CFU/g)	9.53	8.60⁎⁎	0.170	0.004
Escherichia coli (Log CFU/g)	3.27	2.42	0.251	0.091
Coliforms (Log CFU/g)	3.61	3.69	0.101	0.696
Enterococcus spp. (Log CFU/g)	7.51	7.25	0.097	0.184
Week 40
Lactobacillus spp. (Log CFU/g)	10.17	11.84⁎⁎⁎	0.174	< 0.001
Bifidobacterium spp. (Log CFU/g)	8.62	9.52*	0.182	0.011
Clostridium spp. (Log CFU/g)	10.69	9.32⁎⁎⁎	0.158	< 0.001
Escherichia coli (Log CFU/g)	5.35	4.07⁎⁎⁎	0.136	< 0.001
Coliforms (Log CFU/g)	4.85	4.82	0.228	0.963
Enterococcus spp. (Log CFU/g)	7.89	7.97	0.132	0.769

^⁎Within a row, means differ significantly at P < 0.05.

^⁎⁎Within a row, means differ significantly at P < 0.01.

^⁎⁎⁎Within a row, means differ significantly at P < 0.001.

¹CON, 0 CFU/g feed; PRO, 1.3 × 10⁶ CFU/g feed.

²Pooled standard error of the mean.

³Statistical significance was determined at P < 0.05.

Tibia bone characteristics

At wk 29, the PRO exhibited higher tibia weight (P = 0.042), breaking strength (P = 0.034), and Ca content (P = 0.048) compared to CON (Table 9). Tendencies for increased tibia P (P = 0.086) and ash content (P = 0.080) were also observed at wk 29. At wk 40, PRO increased tibia weight (P = 0.017), density (P = 0.042), breaking strength (P = 0.048), Ca content (P = 0.037), and ash content (P = 0.046). A trend toward higher tibia P was also noted at wk 40 (P = 0.054).

Table 9.

Effect of the dual-strain Bacillus subtilis supplementation on tibia characteristics of laying hens.

Item	Diets1		SEM2	P-value3
	CON	PRO
Week 29
Tibia weight (g)	12.29	12.95*	0.161	0.042
Tibia length (cm)	12.36	12.32	0.056	0.771
Tibia volume (cm3)4	9.56	9.89	0.323	0.614
Tibia density (g/cm3)5	1.36	1.33	0.040	0.706
Tibia breaking strength (kgf)	22.42	23.81*	0.329	0.034
Tibia P (%)6	10.67	10.90	0.068	0.086
Tibia Ca (%)6	21.32	22.19*	0.220	0.048
Tibia ash (%)6	55.41	56.08	0.188	0.080
Week 40
Tibia weight (g)	12.76	13.67*	0.195	0.017
Tibia length (cm)	12.53	12.55	0.063	0.863
Tibia volume (cm3)	10.06	10.17	0.087	0.531
Tibia density (g/cm3)	1.27	1.35*	0.018	0.042
Tibia breaking strength (kgf)	23.28	24.25*	0.245	0.048
Tibia P (%)	11.01	11.44	0.111	0.054
Tibia Ca (%)	22.28	23.28*	0.241	0.037
Tibia ash (%)	56.13	57.26*	0.288	0.046

^⁎Within a row, means differ significantly at P < 0.05.

¹CON, 0 CFU/g feed; PRO, 1.3 × 10⁶ CFU/g feed.

²Pooled standard error of the mean.

³Statistical significance was determined at P < 0.05.

⁴Tibia volume (cm³) = water volume after tibia immersion – initial water volume before tibia immersion (Oketch et al., 2024).

⁵Tibia density (g/cm³) = Tibia weight / Tibia volume (Oketch et al., 2024).

⁶Values are expressed on a defatted dry basis.

Economic analysis

Economic analysis revealed clear treatment-related differences during specific feeding phases (Table 10). During Phase 2 (wk 30 to 40), hens in PRO exhibited higher egg income (P = 0.020) and a greater ROI (P = 0.010) compared with those in CON group. Over the entire experimental period (wk 18 to 40), a trend toward an increased ROI was also maintained in PRO (P = 0.094). In contrast, during Phase 1 (wk 18 to 29), productive economic parameters including egg income, feed cost, and ROI remained unaffected by the dietary treatments (P > 0.10), and cumulative feed cost over the entire period showed no differences (P > 0.10).

Table 10.

Economic analysis of dual-strain Bacillus subtilis supplementation in laying hens.

Item	Diets1		SEM2	P-value3
	CON	PRO
Week 18 to 29 (USD/hen)
Egg income4	1.39	1.43	0.020	0.357
Feed cost 5	0.49	0.49	0.003	0.814
Return on investment 6	0.90	0.93	0.018	0.284
Week 30 to 40 (USD/hen)
Egg income	1.55	1.58*	0.006	0.020
Feed cost	0.54	0.53	0.001	0.200
Return on investment	1.01	1.04*	0.006	0.010
Week 18 to 40 (USD/hen)
Egg income	1.47	1.50	0.012	0.171
Feed cost	0.51	0.51	0.002	0.731
Return on investment	0.95	0.99	0.010	0.094

^⁎Within a row, means differ significantly at P < 0.05.

¹CON, 0 CFU/g feed; PRO, 1.3 × 10⁶ CFU/g feed.

²Pooled standard error of the mean.

³Statistical significance was determined at P < 0.05.

⁴Egg income was calculated using hen-day egg production, average egg weight, and the price of an egg over two weeks.

⁵Feed costs were estimated based on local ingredient prices and the average feed intake per hen over two weeks.

⁶Calculated as the difference between egg income and feed cost.

Discussion

The present study demonstrated that supplemental dual-strain B. subtilis complex (597 and 600) improved hen productive performance, egg quality, nutrient digestibility, intestinal morphology, bone characteristics, and selected physiological indicators during early-to-peak production. These outcomes align with the requirement of AGP-free poultry production systems, which prioritize sustainable productivity, intestinal health, and nutrient utilization. In this study, body weight was unaffected by probiotic supplementation, likely reflecting the metabolic priority of modern layers to allocate nutrient toward egg production rather than somatic growth. This observation concurs with previous Bacillus-based probiotic studies in laying hens (Oketch et al., 2024; Rodriguez-Sanchez et al., 2026).

As to productive performance, probiotic-mediated effects were not observed during Phase 1 (wk 18 to 29), reflecting the time required for probiotic strains to establish stable intestinal colonization and exert measurable effects (Zhang et al., 2021; Zhao et al., 2025). Furthermore, early-lay physiological stress, driven by rapid follicular development and endocrine reorganization, may temporarily obscure probiotic benefits. In contrast, more pronounced responses were observed during phase 2 (wk 30 to 40), where probiotic-supplemented hens exhibited higher HDEP, egg weight, and egg mass, together with lower FCR and egg loss rate. These findings indicate that probiotic benefits emerge more clearly once hens reach a metabolically stable production stage, concurring with evidence that prolonged supplementation allows cumulative improvements in gut functionality and nutrient utilization to translate into production outcomes (Tajudeen et al., 2024; Zhao et al., 2025; Rodriguez-Sanchez et al., 2026).

Regarding egg quality, both the internal and external eggshell quality were improved with probiotics, consistent with previous Bacillus-based probiotic studies (Oketch et al., 2024; Zhao et al., 2025; Rodriguez-Sanchez et al., 2026). Enhanced albumen quality likely stems from improved intestinal histomorphology and nutrient absorption, as was observed in the current study. Darker shell color, as observed in PRO group, may reflect increased deposition of protoporphyrin IX, potentially supported by enhanced serum T-AOC, as oxidative stress downregulates δ-aminolevulinic acid synthase 1, a key enzyme in protoporphyrin IX biosynthesis (Samiullah et al., 2015; Lu et al., 2021; Fu et al., 2025). The greater eggshell breaking strength and thickness correspond to increased ileal Ca and P digestibility in the current study. While not directly measured in this study, previous literature indicates that B. subtilis may support shell formation by modulating Ca transport pathways, matrix protein expression, or luminal mineral availability (Nishiyama et al., 2021; Wang et al., 2021; Zhu et al., 2026). These observations complement broader nutritional strategies aimed at optimizing mineral utilization and eggshell integrity in modern laying hens (Salehi et al., 2025).

Improved intestinal morphology likely expanded the epithelial surface area for mineral uptake, supporting nutrient absorption and subsequent eggshell mineralization (Rodriguez-Sanchez et al., 2026). In addition, enhanced lactic acid-producing bacteria in the cecum, such as Lactobacillus and Bifidobacterium species, can lower luminal pH through organic acid fermentation (Engberg et al., 2009; Gholiabad et al., 2014; Khan et al., 2020). Reduced intestinal pH is postulated to increase the solubility and ionization of dietary Ca (Oketch et al., 2024), facilitating both passive paracellular diffusion and active transcellular absorption of $\mathrm{C}{\mathrm{a}}^{2+}$ across the intestinal epithelium (Gholiabad et al., 2014; Jiang et al., 2023; Lee et al., 2025). This pH-mediated increase in mineral bioavailability directly supports the elevated ileal Ca and P digestibility observed in in PRO at wk 40, further reinforcing evidence that lowered intestinal pH benefits physiological processes related to mineral absorption and subsequently, eggshell quality (Świątkiewicz et al., 2010; Gultemirian et al., 2022). Furthermore. probiotic-induced shifts in gut microbial metabolism may indirectly modulate shell matrix protein expression, further contributing to optimal shell ultrastructure (Lu et al., 2025).

The progressive reduction in yolk cholesterol observed in the probiotic-supplemented group suggests a modulation of lipid metabolism, which aligns with the parallel reduction in serum LDL-c, VLDL-c, and TC at wk 40. Probiotic-mediated reductions in yolk cholesterol are attributed to complementary mechanisms along the gut-liver axis. Specifically, probiotic-associated proliferation of commensal Lactobacillus and Bifidobacterium species in the cecum may increase the intraluminal production of bile salt hydrolase (BSH), an enzyme responsible for deconjugating conjugated bile acids (Bustos et al., 2016; Yaqoob et al., 2022). Because deconjugated bile acids are more hydrophobic, they are less efficiently reabsorbed via enterohepatic circulation, potentially accelerating fecal excretion (Choi et al., 2015; Ishimwe et al., 2015; Oketch et al., 2024). To replenish the diminished bile acid pool, the liver utilizes endogenous cholesterol as a precursor for de novo bile acid synthesis, thereby diverting circulating cholesterol from VLDL-c secretion and subsequent yolk deposition (Jones et al., 2013; Joyce et al., 2014; Choi et al., 2015). Although BSH activity was not directly quantified in the present study, this pathway is widely recognized as a primary hypocholesterolemic mechanism of probiotic bacterial in poultry (Yaqoob et al., 2022), consistent with prior reports demonstrating reduced cholesterol through Bacillus-based probiotic supplementation in laying hens (Sobczak and Kozłowski, 2015; Oketch et al., 2024; Sihamok et al., 2026).

As to ileal morphology, probiotics increased VH, VH:CD ratio and the relative VSA, in line with previous studies involving Bacillus probiotic blends (Zhao et al. 2025; Rodriguez-Sanchez et al. 2026). Increased VH:CD ratio reflects enhanced mucosal structural integrity, including a higher proportion of VSA relative to the proliferative zone, which is generally associated with improved nutrient absorptive capacity (Khogali et al., 2022). Proposed mechanisms by which B. subtilis improved intestinal morphology include the secretion of anti-inflammatory metabolites, modulation of tight junction protein expression, reduction of pathogenic colonization, and production of digestive enzymes that mitigate mucosal irritants (Chen et al., 2020; Zhang et al., 2021; Zou et al., 2022). These structural enhancements align with the functional capacity of B. subtilis 597 to support mucosal barrier integrity, epithelial architecture preservation, and host immune response (Winther et al., 2024), potentially facilitating the improved digestibility of DM, GE, CP, Ca, and P. Particularly, the enhanced Ca and P digestibility provides a physiological rationale supporting the observed improvements in eggshell quality and skeletal mineralization observed in the PRO group of hens (Table 2, Table 3).

Table 2.

Effect of the dual-strain Bacillus subtilis supplementation on body weight of laying hens.

Items		Body weight (g)
Week 16	CON1	1413.12
	PRO2	1402.69
Week 18	CON	1623.48
	PRO	1620.51
Week 29	CON	1791.35
	PRO	1797.27
Week 40	CON	1918.79
	PRO	1919.13
SEM3		3.581
Main effects
Treatment	CON	1686.69
	PRO	1684.90
Age	Week 16	1407.91d
	Week 18	1622.00c
	Week 29	1794.31b
	Week 40	1918.96a
P-value4
Treatment		0.803
Age		< 0.001
Treatment × Age		0.857

^a-d Means within a column with different superscripts differ significantly (P < 0.05).

¹Basal diet without probiotics.

²Basal diet with probiotics (1.3 × 10⁶ CFU/g feed).

³Pooled standard error of the mean.

⁴Statistical significance was determined at P < 0.05.

Table 3.

Effect of the dual-strain Bacillus subtilis supplementation on productive performance of laying hens.

Items		Hen-day egg production (%)1	Egg loss rate (%)2	Average egg weight (g)3	Egg mass (g/day/hen)4	Average daily feed intake (g)5	Feed conversion ratio (g feed/g egg)6
Phase 1
Week 18 to 19	CON7	61.64	4.33	48.23	30.00	95.63	1.98
	PRO8	63.27	3.67	48.04	30.48	95.14	1.98
Week 20 to 21	CON	89.37	1.98	50.08	44.78	105.41	2.11
	PRO	90.13	1.81	50.43	45.40	105.70	2.10
Week 22 to 23	CON	90.34	0.93	52.46	47.40	114.86	2.19
	PRO	93.74	0.86	52.54	49.25	114.49	2.18
Week 24 to 25	CON	91.67	0.64	54.77	50.21	119.70	2.19
	PRO	94.80	0.61	54.80	51.99	119.29	2.18
Week 26 to 27	CON	93.08	0.81	55.87	52.00	119.54	2.14
	PRO	95.72	0.28	56.12	53.71	119.29	2.13
Week 28 to 29	CON	94.36	0.95	56.57	53.30	119.08	2.11
	PRO	96.52	0.93	57.01	54.98	119.05	2.10
SEM9		0.796	0.110	0.153	0.447	0.577	0.009
Main effects
Treatment	CON	86.74	1.61	53.00	46.28	112.37	2.12
	PRO	89.03	1.36	53.16	47.63	112.16	2.11
Age	Week 18 to 19	62.46c	4.00a	48.13e	30.24e	95.38d	1.98c
	Week 20 to 21	89.75b	1.89b	50.26d	45.09d	105.56c	2.10b
	Week 22 to 23	92.04ab	0.90c	52.50c	48.32c	114.66b	2.18a
	Week 24 to 25	93.23ab	0.62c	54.78b	51.10bc	119.49a	2.18a
	Week 26 to 27	94.40ab	0.55c	56.00a	52.85ab	119.42a	2.13b
	Week 28 to 29	95.44a	0.94c	56.79a	54.14a	119.06a	2.11b
P-value10
Treatment		0.157	0.268	0.610	0.136	0.857	0.567
Age		< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Treatment × Age		0.997	0.914	0.988	0.994	0.920	0.999
Phase 2
Week 30 to 31	CON	93.85	2.56	58.88	55.26	119.58	2.03
	PRO	95.69	2.47	59.37	56.80	119.48	2.01
Week 32 to 33	CON	93.55	3.46	60.70	56.73	119.44	1.97
	PRO	95.39	2.50	61.21	58.38	119.24	1.95
Week 34 to 35	CON	92.96	2.18	61.67	57.36	118.75	1.93
	PRO	94.44	1.80	62.21	58.74	118.15	1.90
Week 36 to 37	CON	92.76	4.38	61.69	57.18	119.23	1.93
	PRO	93.95	3.52	62.26	58.47	118.92	1.91
Week 38 to 40	CON	92.06	4.06	61.80	56.89	119.41	1.93
	PRO	93.06	3.33	62.55	58.25	119.31	1.91
SEM		0.348	0.114	0.119	0.207	0.084	0.004
Main effects
Treatment	CON	93.04b	3.33a	60.95b	56.68b	119.28	1.96a
	PRO	94.52a	2.73b	61.52a	58.13a	119.02	1.94b
Age	Week 30 to 31	94.77	2.51bc	59.13c	56.03b	119.53a	2.02a
	Week 32 to 33	94.47	2.98b	60.96b	57.56a	119.34a	1.96b
	Week 34 to 35	93.70	1.99c	61.94a	58.05a	118.45b	1.91c
	Week 36 to 37	93.35	3.95a	61.97a	57.82a	119.07a	1.92c
	Week 38 to 40	92.61	3.69a	62.18a	57.56a	119.36a	1.92c
P-value
Treatment		0.037	0.012	0.020	< 0.001	0.124	0.008
Age		0.336	< 0.001	< 0.001	0.041	0.001	< 0.001
Treatment × Age		0.993	0.757	0.997	0.999	0.887	0.997

^a-e Means with a column with different lowercase (a-e) superscripts differ significantly (P < 0.05).

¹Hen-day egg production, % = Total number of eggs produced in a day / Total number of laying hens in a day $\times$100.

²Egg loss percentage, % = Number of spoilt eggs / Number of total eggs × 100.

³Average egg weight, g = Total egg weight / number of eggs.

⁴Egg mass, g/hen/day = Hen-day egg production × egg weight / 100.

⁵Average daily feed intake, g/hen/day = Total feed intake in a week / Days × Number of laying hens.

⁶Feed conversion ratio, g/hen/day = Feed intake per day per hen / Average egg weight.

⁷Basal diet without probiotics.

⁸Basal diet with probiotics (1.3 × 10⁶ CFU/g feed).

⁹Pooled standard error of the mean.

¹⁰Statistical significance was determined at P < 0.05.

Shifts in hen metabolic and skeletal status, indicative of enhanced lipid homeostasis, active bone remodeling, and improved physiological resilience, were also observed in present study. Reduced in serum VLDL-c, LDL-c, and TG, alongside elevated HDL-c, reflects a favorable rebalancing of systemic lipid metabolism via the gut-liver axis. These shifts concur with the known hypolipidemic properties of Bacillus-based probiotics (Fathi et al., 2018; Zhao et al., 2025), which are hypothesized to be driven by BSH activity promoting fecal cholesterol excretion and SCFA-mediated regulation of hepatic lipogenesis (Khan et al., 2020; Zhao et al., 2025). Concurrently, the early elevation of serum ALP at wk 29 corresponds to the intense skeletal remodeling and Ca mobilization required during the transition from cortical to medullary bone reserves (Zou et al., 2021; Onbaşılar et al., 2025). As a biochemical marker of osteoblastic activity, the enhanced ALP response in the PRO supports active bone formation and enhanced mineral transport capacity. Improvements in tibial structural strength and mineral retention emerged during Phase 1 and expanded to include bone density and total ash content by Phase 2. These findings concur with previous observations demonstrating probiotic-mediated modulation of mineral utilization and bone mineralization, which may cushion hens against age-related bone depletion (Yan et al., 2019; Oketch et al., 2024; Zhu et al., 2026). The observed tibial improvements correspond with improved Ca and P digestibility and serum ALP, providing direct evidence suggestive of greater intraluminal mineral availability to support skeletal mineral accretion and bone formation.

A distinctive neuroendocrine finding in the present study was in the divergent changes in reduced serum corticosterone but improved serotonin concentrations with probiotics. Corticosterone serves as a sensitive biomarker for chronic physiological stress and metabolic disturbance (Hu et al., 2018). Lower serum corticosterone concentration with probiotics reflects improved physiological stability and the attenuation of systemic stress. Approximately 95% of systemic serotonin is synthesized by mucosal enterochromaffin cells in response to specific microbial metabolites and SCFAs (Hu et al., 2018). As a principal mediator of the gut-brain axis, peripheral serotonin regulates gastrointestinal motility, luminal secretion, and FI-related signaling pathways. Elevated serum serotonin in probiotic-supplemented hens could be associated with the stabilized FI patterns and productive performance recorded during Phase 2. Furthermore, significant improvements in serum T-AOC indicate an enhanced antioxidant defense system, which aligns with previous evidence that functional bio-additives and probiotic strains effectively attenuate systemic oxidative stress (Dong et al., 2025; Zhao et al., 2025).

Dietary dual-strain B. subtilis complex also improved cecal populations of Lactobacillus spp. and Bifidobacterium spp., while reducing Clostridium spp. These microbial shifts concur with evidence that Bacillus-based probiotics reshape the intestinal microbiome by selectively proliferating beneficial commensal anaerobes while suppressing potentially pathogenic bacteria (Nishiyama et al., 2021; Zhang et al., 2021; Tajudeen et al., 2024). Commensal taxa such as Lactobacillus and Bifidobacterium spp. ferment carbohydrates into organic acids, lowering intestinal pH and generating SCFAs, which serve as energy substrate for colonocytes and are hypothesized to support epithelial barrier integrity (Khan et al., 2020; Puncharoen et al., 2024). Reduced Clostridium spp. is of pathological relevance, given that B. subtilis 597 is characterized by its target-specific antagonism against Clostridium perfringens via secretion of antimicrobial peptides (Teo and Tan, 2005; Liu et al., 2021). Importantly, stable populations of Enterococcus spp. and total coliforms across treatments demonstrate that the probiotic complex selectively modulated targeted taxa without disrupting the overarching microbial equilibrium.

Regarding economic analysis, egg income, feed cost, and ROI were unaffected during Phase 1, mirroring the comparable productive performance during this initial stage. Conversely, during Phase 2, probiotic-supplemented group yielded higher egg income and ROI, along with a trend toward increased cumulative ROI. These improvements were primarily attributed to enhance egg production and weight, together with a reduced egg loss rate during Phase 2, which augmented the proportion of eggs qualifying for premium, higher-value weight categories. Notably, net feed costs did not differ between the groups despite the expense of probiotic inclusion owing to the identical FI between treatments, indicating that economic benefits were derived from production-related improvements. The concentration of economic benefits in Phase 2 underscores the critical role of supplementation duration, as full commercial value could manifest after an establishment phase. These outcomes validate the economic incentive for supplemental probiotics, particularly during peak production where the compounding benefits of microbiota modulation, enhanced nutrient digestibility, and superior egg quality can translate into potential financial returns (Darsi and Zhaghari, 2021; Souza et al., 2021).

Conclusively, supplemental dual-strain B. subtilis complex (597 and 600) improved productive performance, nutrient digestibility, bone mineralization, cecal microbiota and physiological status in laying hens. These responses indicate enhanced intestinal health and physiological homeostasis through modulation of the cecal microbiota, antioxidant status, and neuroendocrine-associated parameters. Collectively, these findings demonstrate the potential of the B. subtilis complex to improve gut health, physiological resilience, egg quality, and sustainable productivity during peak laying. Future studies should further elucidate the mechanisms underlying observed probiotic responses by characterizing the cecal microbiota using high throughput 16S rRNA gene sequencing and evaluating bile salt hydrolase activity, intestinal barrier gene expression, and long-term intestinal histomorphological adaptations. Additionally, extending supplementation into the late laying phase would help determine the persistence of these benefits throughout the laying cycle.

Disclosures

The authors declare that Antoine Meuter, Shirley Mok, and In Sun Yu are employees of Novonesis. Novonesis provided financial support and the experimental product that was evaluated. The funder had no further role in data collection, data analysis, or interpretation. The authors therefore declare no financial or personal relationships that could inappropriately influence the findings reported in this paper.

Author Contribution

Sehyeok Oh conceptualized the study, developed the methodology, carried out the investigation, analyzed and visualized the data, and drafted as well as edited the manuscript. Elijah Ogola Oketch contributed to the methodology and manuscript editing. Yongtae Han, Hae In Yong, and Young-Joo Yi contributed to the investigation and manuscript reviewing. Antoine Meuter, Shirley Mok, and In Sun Yu provided necessary resources, secured funding, and reviewed the manuscript. Jung Min Heo conceptualized and supervised the overall project, managed project administration, validated the results, and edited the manuscript.