Gut microbial community assembly exhibits compartment-specific responses to selection for high body weight in female chickens Authors: Fei Zhang, Yao Wang, Jiale Qian, Shuai Li, Xutao Deng, Mengya Lu, Yanli Liu, Xin Yang, Jianqin Zhang, Zhouzheng Ren, Xiaojun Yang Published: 2026 ABSTRACT AbstractConventional poultry breeding primarily targets host growth traits, yet breeding may also reshape the gut microbiota through host–microbe interactions. Given the spatial heterogeneity of the chicken gastrointestinal tract, whether divergent selection for body weight is associated with compartment-specific microbial assembly and differences in the digestive ecosystem remains unclear. Here, using female primitive line and a high body weight (HBW) selected line, we systematically characterized differences in host gastrointestinal traits and gut microbial ecology associated with divergent selection history. Significant differences in gut morphology and local environmental conditions were observed between the two lines, together with region-specific shifts in microbial community assembly. Specifically, crop weight was greater and microbial assembly exhibited greater stochasticity in the HBW line, potentially reflecting differences in feeding-related gastrointestinal dynamics. In contrast, the small intestine displayed improved villus morphology and altered microbial assembly patterns, accompanied by enrichment of potentially beneficial taxa associated with nutrient utilization, including Monoglobus in the duodenum and Blautia in the ileum, indicating coordinated changes in host intestinal development and microbial ecological adaptation. The cecal community showed relatively stable assembly patterns but was enriched in metabolically relevant taxa such as Lachnospiraceae and Faecalibacterium, together with differences in predicted functional potential related to nutrient metabolism. Overall, this study shows that the HBW line is associated with distinct, compartment-specific patterns of gut microbial assembly, providing a conceptual framework for considering the gut microbiota in poultry breeding strategies.Keywords: Gut microbiota, High body weight selection, Community assembly, Gastrointestinal compartmentalization, Host–microbiome interaction Introduction Poultry are economically important livestock, whose production traits have been continuously optimized through long-term artificial breeding, resulting in diverse phenotypes with improved growth, egg production, and environmental adaptability (Wang et al., 2023; Zuidhof, et al., 2014). During this process, changes in the host gastrointestinal environment selectively shape microbial communities (Wu, et al., 2025). It is now well established that gut microbiota play key roles in nutrient digestion (Singh, et al., 2026), immune modulation (Kong, et al., 2026), and metabolic homeostasis (Zhang, et al., 2026c) in poultry. However, traditional breeding programs have predominantly focused on host phenotypic traits, such as growth rate and productivity, and physiological parameters, including intestinal villus morphology and digestive enzyme activity, largely neglecting microbial communities (Cheema, et al., 2003; de Verdal, et al., 2011; Havenstein, et al., 2003; Kinstler, et al., 2024). Recent studies indicate that gut microbiota are influenced by the host genetic background, suggesting that integrating host–microbiome interactions into poultry breeding could represent a promising strategy to enhance production performance and health (Kpodo, et al., 2026; Zhao, et al., 2026). Previous studies have shown that poultry lines subjected to different breeding histories exhibit distinct gut microbial profiles. For example, cecal microbial composition and function vary among different Chinese indigenous chicken breeds, potentially influenced by host genetic factors, which in turn regulate muscle metabolic traits (Feng, et al., 2022). Similarly, fecal microbiota differed in high- versus low-abdominal fat lines, correlating with lipid storage and metabolism (Ding, et al., 2016). Despite these findings, a systematic understanding of gut microbial differences associated with poultry breeding remains limited. Most studies have focused on single gut segments, leaving the spatial heterogeneity among different gastrointestinal regions poorly understood. The chicken digestive tract displays marked functional differentiation, with segments acting both independently and cooperatively in nutrient digestion and metabolism (Wen, et al., 2021). The crop functions in storage and preliminary fermentation (Cutler, et al., 2005; Feigenbaum, et al., 1962), the small intestine mediates nutrient absorption (Mushtaq, et al., 2025), and the cecum plays a key role in fiber fermentation and energy recovery (Zhang, et al., 2022). Such functional differentiation may drive region-specific microbial responses. Moreover, prior work has mainly emphasized shifts in microbial composition, whereas the assembly mechanisms underlying community dynamics—governed by deterministic processes (e.g., environmental filtering) and stochastic processes (e.g., dispersal limitation)—have received little attention (Zhou and Ning, 2017). Whether spatially structured microbial assembly patterns differ among chicken lines with distinct selection histories remains unclear. Our previous study characterized differences in gastrointestinal structure and microbial community composition between male chickens from the primitive and HBW-selected lines (Zhang, et al., 2026a). In the present study, we focused on female chickens and extended this work by examining compartment-specific microbial assembly, inter-compartment microbial connectivity, and the ecological processes underlying community organization across the gastrointestinal tract. In this study, we used a primitive chicken line and its derived high body weight (HBW) selected line to characterize differences in host gastrointestinal traits, luminal environments, microbial community structure, and community assembly processes across distinct gut compartments, including the crop, small intestine, and cecum. We hypothesized that differences between the two lines would vary across gut compartments and be associated with region-specific differences in the host gastrointestinal environment. Materials and methods Animal ethics statement The animal management and experimental procedures in this study were performed in accordance with the Guidelines for Care and Use of Laboratory Animals and have been approved by the Animal Ethics and Welfare Committee of Northwest A&F University (protocol number DK2022009). Information on the two chicken lines The experimental chickens included females from the primitive line and the HBW line of Lueyang Black-boned chickens. The primitive line has been maintained through natural mating without artificial breeding. The HBW line was derived from the primitive line by selecting for high body weight as the sole breeding criterion over seven generations. The average body weights of the two lines were as follows (16 weeks of age): primitive line males averaged 1387.97 g and females 1169.19 g, while HBW line males averaged 2056.31 g and females 1530.43 g. Thus, the 7th generation of the HBW line showed an increase of approximately 668.34 g in males and 361.24 g in females compared with the primitive line. These results indicate that substantial genetic progress in body weight had been achieved through directional breeding by the 7th generation. Chickens, housing and sampling All chickens were reared at the Institute of Poultry–Crop Integrating Systems, Northwest A&F University. A total of twenty 1-day-old female Lueyang Black-boned chickens were used in this study, including ten from the primitive line and ten from the HBW line. The selected chickens were healthy and had body weights close to the population mean. Chickens were housed individually in cages (40 × 40 × 40 cm), with one bird per cage and ten cages per line at the beginning of the experiment. Thus, each bird occupied a cage floor area of 0.16 m². Chickens from the two lines were randomly assigned to cage positions and were maintained under the same environmental and management conditions. All chickens were raised from 1 to 180 days of age, during which the body weight of each bird was recorded continuously. During the rearing period, three birds died of disease (two from the primitive line and one from the HBW line), and two birds were removed from the experiment due to intestinal disorders (one from each line). All chickens had ad libitum access to feed (composition shown in Table 1) and water, following the feeding standards recommended by the Shaanxi Engineering Research Center of Lueyang Black Chicken. At 180 days of age, the chickens were euthanized and samples were collected as follows: serum (stored at −80 °C for biochemical analysis), intestinal tissues (fixed in 4% formaldehyde for morphology analysis), and gastrointestinal digesta (stored at −80 °C for 16S rRNA sequencing and short-chain fatty acid analyses). Gastrointestinal organ indices were also recorded during necropsy (Table 2). Table 1Ingredient composition of diet (as-fed basis, %).IngredientsContent (%)Nutrient(analyzed, unless noted)Content (%)d1-42d43-180d1-42d43-180Corn59.7261.83Metabolizable energy (calculated, kcal/kg)28002850Soybean meal (43%)22.0018.00Crude protein (calculated/analyzed)20.00/19.8518.00/17.9Wheat bran7.80/Calcium (calculated/analyzed)1.00/1.181.00/1.05Corn germ meal/7.00Total phosphorus (calculated/analyzed)0.70/0.680.60/0.58Cottonseed meal (46%)5.005.00Available phosphorus (calculated)0.400.32DDGS1/2.50Lysine (calculated)1.201.10Dicalcium phosphate1.401.00Methionine (calculated)0.450.42Soy oil1.001.00Threonine (calculated)0.850.75Fine limestone1.001.50Premix11.001.00L-Lysine (70%)0.500.65Salt0.200.15DL-methionine0.160.15Sodium carbonate0.100.10Choline chloride (60%)0.100.10Antioxidant0.020.02Total100.00100.00Abbreviations: DDGS, distillers dried grains with soluble.1 Mineral and vitamin Premix (Provided per kg of diet) : Vitamin A, 12,000 IU; Vitamin D3, 2,500 IU; Vitamin E, 20.00 mg; Vitamin K3, 2.80 mg; Vitamin B1, 2.00 mg; Vitamin B2, 7.00 mg; Vitamin B6, 6.80 mg; Vitamin B12, 0.025 mg; Biotin, 0.033 mg; Niacin, 55.00 mg; Folic acid, 1.40 mg; Pantothenic acid, 10.00 mg; Manganese, 80.00 mg; Zinc, 80.00 mg; Iron, 60.00 mg; Iodine, 0.40 mg; Copper, 9.00 mg; Selenium, 0.20 mg. Ingredient composition of diet (as-fed basis, %). Abbreviations: DDGS, distillers dried grains with soluble. 1 Mineral and vitamin Premix (Provided per kg of diet) : Vitamin A, 12,000 IU; Vitamin D3, 2,500 IU; Vitamin E, 20.00 mg; Vitamin K3, 2.80 mg; Vitamin B1, 2.00 mg; Vitamin B2, 7.00 mg; Vitamin B6, 6.80 mg; Vitamin B12, 0.025 mg; Biotin, 0.033 mg; Niacin, 55.00 mg; Folic acid, 1.40 mg; Pantothenic acid, 10.00 mg; Manganese, 80.00 mg; Zinc, 80.00 mg; Iron, 60.00 mg; Iodine, 0.40 mg; Copper, 9.00 mg; Selenium, 0.20 mg. Table 2Serum biochemistry of the two chicken lines.ItemTreatmentSEMP-valuePrimitive lineHBW lineTP, g/L37.6541.135.400.531GLU, mmol/L10.1511.660.930.128TC, mmol/L2.342.600.370.495TG, mmol/L4.796.402.470.525HDL, mmol/L1.050.890.360.658LDL, mmol/L0.330.340.080.908UA, mmol/L0.26b0.41a0.03< 0.0011 Abbreviations: GLU, glucose; HBW, high body weight; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TC, total cholesterol; TG, triglyceride; TP, total protein; UA, uric acid; SEM, standard error of the mean.a, b Means within a row lacking a common superscript differ (P ≤ 0.05). Serum biochemistry of the two chicken lines. 1 Abbreviations: GLU, glucose; HBW, high body weight; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TC, total cholesterol; TG, triglyceride; TP, total protein; UA, uric acid; SEM, standard error of the mean. a, b Means within a row lacking a common superscript differ (P ≤ 0.05). Chemical composition of the diet The diets were formulated according to the recommended level of Chinese Feeding Standard of Chicken (NY/T 33-2004). The dietary metabolizable energy (ME) was calculated as the sum of the product of ingredient inclusion levels and their corresponding ME values: where Xi represents the inclusion level of ingredient i (kg/kg diet), and MEi is the metabolizable energy value of ingredient i obtained from the 32nd edition of the Chinese Feed Composition and Nutritional Value Table (Xiong, 2021). Premix was excluded due to negligible energy contribution. The crude protein (N × 6.25) content was determined using the Kjeldahl method 990.03 (AOAC, 1995) with a nitrogen analyzer (Model: CNS-2000, LECO Corporation, St. Joseph, MI, USA). With reference to the previous methodology, the actual phosphorus content in the diet was analyzed using a spectrophotometer (Model: UV-2700, Shimadzu Corporation, Kyoto, Japan) in accordance with GB/T 6437-2018, while the calcium content was measured using flame atomic absorption spectrophotometry (Model: Zeenit700P, Analytik Jena AG, Jena, Germany) according to GB/T 6436-2018. The samples were ashed at 600 °C for 12 h using a muffle furnace, and the minerals were determined using inductively coupled plasma mass spectrometry (Model: 7400, Thermo Fisher Scientific, Waltham, MA, USA). The calculated nutritional values in Table 1 referred to the 32nd edition Chinese Feed Composition and Nutritional Value Table (Xiong, 2021). Serum biochemistry analysis Blood samples were collected from the brachial vein of each chicken (Primitive line, n = 7; HBW line, n = 8) into serum separation tubes. The samples were allowed to clot for 2 h at room temperature (25 °C). Then, the clotted blood was centrifuged (15 min, 4 °C, 500 × g) for serum samples (stored at –80 °C for future analysis). The levels of total protein (Catalog No.: 70112, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China), glucose (Catalog No.: 70129, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China), triglyceride (Catalog No.: 701133, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China), total cholesterol (Catalog No.: 701132, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China), high-density lipoprotein (Catalog No.: 713134, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China), low-density lipoprotein (Catalog No.: 713135, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China), and uric acid (Catalog No.: 71325, Shandong Boke Biotechnology Industry Co., Ltd., Jinan, China) in serum were measured using a Hitachi Clinical Analyzer (Model: 7180, Hitachi High-Technologies, Tokyo, Japan) at Yangling Demonstration Zone Hospital in Yangling, China (Wang, et al., 2024). Organ traits The organ indices were calculated as follows (Primitive line, n = 7; HBW line, n = 8) (Huang, et al., 2022; Zhang, et al., 2026b): 1)Breast muscle yield (%) = total weight of bilateral deboned breast muscles / body weight × 100%. Breast muscle yield (%) = total weight of bilateral deboned breast muscles / body weight × 100%. 2)Thigh muscle yield (%) = total weight of bilateral deboned thigh muscles / body weight × 100%. Thigh muscle yield (%) = total weight of bilateral deboned thigh muscles / body weight × 100%. 3)Abdominal fat yield (%) = total weight of abdominal fat (including abdominal and gizzard fat) / body weight × 100%. Abdominal fat yield (%) = total weight of abdominal fat (including abdominal and gizzard fat) / body weight × 100%. 4)Digestive system weight: The sum of the weights of the crop, proventriculus, gizzard, duodenum, jejunum, ileum, cecum, pancreas, and liver. Digestive system weight: The sum of the weights of the crop, proventriculus, gizzard, duodenum, jejunum, ileum, cecum, pancreas, and liver. 5)Organ index (% of body weight): Organ weight / body weight × 100%. Organ index (% of body weight): Organ weight / body weight × 100%. 6)Organ index (% of digestive system weight): Organ weight / digestive system weight × 100%. Organ index (% of digestive system weight): Organ weight / digestive system weight × 100%. Intestinal histomorphology Intestinal tissue samples (duodenum, jejunum, and ileum; approximately 2 cm in length) were collected from the midpoint of each small intestinal segment (Primitive line, n = 7; HBW line, n = 8). The samples were gently flushed with sterile 0.9% saline to remove intestinal contents and subsequently fixed in 4% paraformaldehyde (Catalog No.: BL539A, Biosharp Life Sciences, Hefei, China) for 48 h. Fixed samples were then dehydrated, embedded in paraffin (Catalog No.: JB-P5, Wuhan Junjie Electronics Co., Ltd., Wuhan, China), sectioned (5 μm) using a microtome (Model: RM2016, Leica Microsystems, Wetzlar, Germany), stained with hematoxylin and eosin (H&E) (Catalog No.: G1076, Servicebio, Wuhan, China), and examined under a light microscope (Model: Eclipse E100, Nikon, Tokyo, Japan). Villus height (VH, µm) was measured from the tip of the villus to the villus-crypt junction (the top of the lamina propria), while crypt depth (CD, µm) was defined as the distance between the base of the crypt and the villus-crypt junction in crypts. For each biological replicate, six fully intact and measurable villus–crypt units were included in the morphometric analysis, and their mean value was calculated (Liu, et al., 2026). Intestinal digesta pH Chickens were processed without a feed withdrawal period. After euthanasia, the gastrointestinal digesta were collected in separate centrifuge tubes. Then, a pH probe of a digital pH meter (Catalog #: PH-STAR, Metz GMBH) was placed directly into the digesta, and pH was recorded. The probe was rinsed in deionized water and wiped dry, after each reading (Cherian, et al., 2013). Intestinal tissue sIgA The concentration of sIgA in intestinal tissues was analyzed using Chicken-specific double antibody sandwich sIgA ELISA kit (Catalog #: RX600526CH, Quanzhou Ruixin Biotechnology Co., LTD) (Abdelhady, et al., 2021). Approximately 0.1 g of gastrointestinal tissue were weighed into 2 mL centrifuge tubes. Then, 900 μL of pre-cooled phosphate buffered saline (0.01 M, pH = 7.4) and two ceramic beads was added. Samples were homogenized for eight cycles of 30 seconds at 65 Hz at 4 °C. Afterward, the samples were centrifuged at 4 °C for 15 minutes at 5000 × g, and the supernatant was collected. ELISA was conducted according to the manufacturer's instructions. After the ELISA reaction, absorbance was measured at 450 nm using a microplate reader (Synergy HT, Biotek (China) Co., LTD). Then, the concentration of sIgA was calculated according to the standard curve. Standard curve: sIgA concentration (μg/g) = 115.090×OD value −10.791. Short chain fatty acids analysis The concentrations of short-chain fatty acids (SCFAs) in gastrointestinal digesta (duodenum, jejunum, ileum, and cecum; Primitive line, n = 7; HBW line, n = 8) were analyzed using gas chromatography (Model: 7890B, Agilent Technologies, Santa Clara, CA, USA), as previously described (Liu, et al., 2023b). Briefly, 0.1 g of digesta was weighed and mixed with 500 μL PBS in a 1.5 mL tube. After centrifugation (10,000 × g for 10 min at 4 °C), the supernatant (400 μL, V1) was mixed with 80 μL (V2) of a 25% H3PO4 solution (Catalog No.: M466284, Aladdin Biochemical Technology Co., Ltd., Shanghai, China) and stored at 4 °C for 4 h. The sample was then centrifuged again (13,500 × g for 15 min at 4 °C), and the supernatant (400 μL, V3) was mixed with 80 μL (V4) of 25% crotonic acid (Catalog No.: C104150, Aladdin Biochemical Technology Co., Ltd., Shanghai, China). The sample was then thoroughly mixed and stored at 4 °C for 1 h. The mixed liquor was filtered through an organic filter (0.22 μm) into a gas phase bottle. The concentration values of SCFAs in the standard solution were as follows: acetic acid (Catalog No.: A433221, Aladdin Biochemical Technology Co., Ltd., Shanghai, China), 83.5799 mmol/L; propionic acid (Catalog No.: P110443, Aladdin Biochemical Technology Co., Ltd., Shanghai, China), 40.2656 mmol/L; isobutyric acid (Catalog No.: I103521, Aladdin Biochemical Technology Co., Ltd., Shanghai, China), 22.6989 mmol/L; butyric acid (Catalog No.: B274612, Aladdin Biochemical Technology Co., Ltd., Shanghai, China), 26.9124 mmol/L; isovaleric acid (Catalog No.: I108280, Aladdin Biochemical Technology Co., Ltd., Shanghai, China), 19.5829 mmol/L; valeric acid (Catalog No.: V108272, Aladdin Biochemical Technology Co., Ltd., Shanghai, China), 19.7087 mmol/L; and crotonic acid. SCFA levels were calculated based on the peak results from standards and samples. Formula (1): CSCFAs = The concentration of SCFAs in the standard solution, mmol/L; Ccrotonic acid = The concentration of crotonic acid in the standard solution, mmol/L; Scrotonic acid = The peak area value of crotonic acid determined by the standard solution; SSCFAs = The peak area value of SCFAs determined by the standard solution. Formula (2): KSCFAs = The SCFAs coefficient determined by the standard substance; Ccrotonic acid = The concentration value of crotonic acid in the sample, mmol/L; Scrotonic acid = The peak area value of crotonic acid determined by the sample, mmol/L; SSCFAs = The peak area value of SCFAs determined by the sample, mmol/L. 16 s rRNA sequencing analysis Total microbial genomic DNA was extracted from gastrointestinal digesta (crop, duodenum, ileum and cecum; Primitive line, n = 7; HBW line, n = 8) samples using the E.Z.N.A.® intestinal digesta DNA Kit (Catalog No.: T10-100, MJYH, shanghai, China) according to manufacturer’s instructions. The quality and concentration of DNA were determined by 1.0% agarose gel electrophoresis and a NanoDrop spectrophotometer (Model: 2000, Thermo Fisher Scientific, Wilmington, DE, USA) and kept at −80 °C prior to further use. The hypervariable region V3-V4 of the bacterial 16S rRNA gene was amplified with primer pairs 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R(5′-GGACTACHVGGGTWTCTAAT-3′) (Liu, et al., 2016) by Thermal Cycler PCR thermocycler (Model: T100, Bio-Rad Laboratories, Hercules, CA, USA). The PCR reaction mixture including 4 μL 5 × Fast Pfu buffer, 2 μL 2.5 mM dNTPs, 0.8 μL each primer (5 μM), 0.4 μL Fast Pfu polymerase, 10 ng of template DNA, and ddH2O to a final volume of 20 µL. PCR amplification cycling conditions were as follows: initial denaturation at 95 °C for 3 min, followed by 27 cycles of denaturing at 95 °C for 30 s, annealing at 55 °C for 30 s and extension at 72 °C for 45 s, and single extension at 72 °C for 10 min, and end at 4 °C. The PCR product was extracted from 2% agarose gel and purified using the PCR Clean-Up Kit (Catalog No.: C01-10000, Yuhua Biotechnology Co., Ltd., Shanghai, China) according to manufacturer’s instructions and quantified using Qubit 4.0 (Catalog No.: Q33238, Thermo Fisher Scientific, Waltham, MA, USA). Purified amplicons were pooled in equimolar amounts and paired-end sequenced on an Illumina PE300 platform (Illumina Inc., San Diego, CA, USA) according to the standard protocols provided by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The raw sequencing reads were deposited into the NCBI Sequence Read Archive database (Accession Number: PRJNA1229171). After demultiplexing, the resulting sequences were quality filtered with fastp (0.19.6) (Chen, et al., 2018) and merged with FLASH ( v1.2.11) (Magoč and Salzberg, 2011). The high-quality sequences were denoised using the DADA2 (Callahan, et al., 2016) plugin in the Qiime2 (Bolyen, et al., 2019) (version 2020.2) pipeline with recommended parameters, which obtains single nucleotide resolution based on error profiles within samples. DADA2 denoised sequences are typically referred to as amplicon sequence variants (ASVs). To minimize the effects of sequencing depth on alpha and beta diversity measures, the number of sequences from each sample was rarefied to 20,000, which still yielded an average Good’s coverage of 97.90%. Taxonomic assignment of ASVs was performed using the Naive bayes consensus taxonomy classifier implemented in Qiime2 and the SILVA 16S rRNA database (v138). Bioinformatic analysis of the intestinal digesta microbiota was carried out using the Majorbio Cloud platform (https://cloud.majorbio.com). Based on the ASVs information, rarefaction curves and alpha diversity indices, including observed ASVs, Chao1 richness, Shannon index, and Good’s coverage, were calculated using Mothur v1.30.1 (Schloss, et al., 2009). The similarity among microbial communities was visualized by principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity using the vegan package (v2.5-3). Differences in microbial community composition between the primitive and HBW lines within each gastrointestinal segment were assessed using analysis of similarities (ANOSIM) based on Bray–Curtis dissimilarity with 999 permutations. The ANOSIM R statistic and corresponding P value were reported. Linear discriminant analysis (LDA) effect size (LEfSe) (Segata, et al., 2011) (http://huttenhower.sph.harvard.edu/LEfSe) was performed to identify the significantly abundant taxa (phylum to genera) of bacteria among the different groups (LDA score > 2, P < 0.05). PICRUSt2 was employed for predicting the bacterial function. Differences were assessed for significance using the Wilcoxon rank sum test. Significance was noted with a P < 0.05. Microbial source tracking was performed at the ASV level using SourceTracker, a Bayesian source-tracking approach, with upstream gastrointestinal compartments defined as potential sources and downstream compartments as sinks. Community assembly processes were evaluated in R (v3.3.1) using the iCAMP package (v1.5.12). For βNTI analysis, a null model with 999 randomizations was used, and ecological processes were classified according to βNTI and RCBray​ as homogeneous selection (βNTI < −2), heterogeneous selection (βNTI > 2), dispersal limitation (|βNTI| < 2 and RCBray​ > 0.95), homogenizing dispersal (|βNTI| < 2 and RCBray < −0.95), or undominated processes (|βNTI| < 2 and |RCBray​| < 0.95). iCAMP analysis was further performed using a phylogenetic-bin-based null-model framework, with a minimum bin size of 24 taxa; ecological processes within each bin were inferred using βNRI and RCBray​. ASVs with prevalence ≥ 0 were retained for both βNTI and iCAMP analyses. Statistical analysis Data analysis was performed using SPSS version 23.0 (IBM Corp., Chicago, IL, USA). The individual chicken was considered the experimental unit. Data were tested for normality using the Shapiro–Wilk test prior to analysis. Comparisons between two groups were conducted using two-tailed Student’s t-test. For analyses involving more than two groups, normally distributed data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test when appropriate. Data that did not meet the assumptions of normality were analyzed using nonparametric tests. Results are presented as means ± SEM. Statistical significance was set at P < 0.05, and trends were discussed when 0.05 < P ≤ 0.10. Results Host physiological and intestinal differences between the two lines The HBW line had significantly greater body weight and absolute breast and leg muscle weights than the primitive line (P < 0.05), whereas relative muscle indices did not differ significantly (Fig. 1B–D). Meanwhile, the HBW line exhibited higher serum uric acid levels (P < 0.05), whereas glucose, high-density lipoprotein, low-density lipoprotein, total cholesterol, triglyceride, and total protein levels remained unchanged (Table2). Consistent with these changes, HBW chickens exhibited higher weights of the crop, gizzard, ileum, and liver (P < 0.05). Villus height-to-crypt depth ratios (VH/CD) in the duodenum and ileum were also increased (P < 0.05), indicating morphological differences in the small intestine. Trends toward a higher gizzard index and a lower cecal index were observed (P < 0.10). In contrast, no significant differences were detected in intestinal length, sIgA levels, digesta pH, or total short-chain fatty acid concentrations (Fig. 1E). Host physiological and intestinal responses to HBW selection. (A) Experimental design. (B) Body weight. (C) Absolute weights of pectoral muscle, leg muscle, and abdominal fat. (D) Relative indices of pectoral muscle, leg muscle, and abdominal fat. (E) Organ indices and intestinal functional parameters. Values are shown as log₂-transformed ratios of the mean value in the HBW line relative to that in the primitive line. White boxes indicate missing values. Data are presented as mean ± SEM. Statistical significance was assessed using Student’s t-test (*P < 0.05, ⁎⁎P < 0.01, #P < 0.1). Differences in body weight between the two lines at each sampled age were assessed separately using two-tailed Student’s t-tests. HBW, high body weight; SEM, standard error of the mean. Region-specific microbial compositional differences between the two lines To characterize microbial differences between the primitive and HBW lines, we first examined spatial variation in community composition along the gastrointestinal tract. Clear regional differences were observed, with the crop dominated by Lactobacillus, the duodenum and ileum enriched in Lactobacillus and Bacillus, and the cecum characterized by Bacteroides and Alistipes (Fig. 2A–D). Despite these pronounced spatial patterns, no statistically significant differences in α-diversity or β-diversity were detected between the primitive and HBW lines across intestinal regions (Fig. 2E–J). Gastrointestinal microbial community shifts associated with HBW selection. (A–D) Relative abundances of the top 10 bacterial genera across gastrointestinal segments. (E) Chao1 richness index. (F) Shannon diversity index. Statistical significance in (E, F) was assessed using the Wilcoxon rank-sum test. (G–J) Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarities showing microbial community variation between chicken lines across four gastrointestinal segments. Data are presented as mean ± SEM. HBW, high body weight; SEM, standard error of the mean; ANOSIM, analysis of similarities. We next assessed compositional differences at ASV level. Differential taxa were detected across all intestinal segments, with the highest number observed in the cecum (Fig. 3A). In the crop, Lactobacillus 30A (ASV12) was enriched in the primitive line, whereas unclassified chloroplast (ASV117) was enriched in the HBW line. In the duodenum, the primitive line was characterized by uncultured bacterium genus Clostridia UCG-014 (ASV368), while the HBW line was enriched with Acidovorax (ASV1636), uncultured bacterium genus Monoglobus (ASV1696), Dietzia (ASV1664), unclassified RF39 (ASV3822), and unclassified chloroplast (ASV117). In the ileum, no differential ASVs were detected in the primitive line, whereas the HBW line was enriched with Blautia (ASV29 and ASV75) and uncultured bacterium genus Subdoligranulum (ASV840). In the cecum, the primitive line was enriched with Ruminococcus torques_group (ASV252), Ruminococcaceae (ASV563), and Enterococcus (ASV2345). By contrast, the HBW line was enriched with multiple taxa, including Alistipes (ASV225), uncultured bacterium genus Faecalibacterium (ASV260), Bacteroides fragilis (ASV3769), Blautia (ASV313), Christensenellaceae R-7 group (ASV911), members of Lachnospiraceae, as well as Firmicutes, Clostridium, and related uncultured or unclassified taxa. Microbial taxa and predicted functions associated with HBW selection. (A) Differentially abundant ASVs across four gastrointestinal segments identified by LEfSe analysis (LDA score > 2, P < 0.05). LDA scores are displayed as a heatmap, with positive values indicating enrichment in the HBW line and negative values indicating enrichment in the primitive line. (B–E) Predicted microbial functional profiles based on KEGG level 3 pathways in the crop, duodenum, ileum, and cecum. Statistical significance between lines was assessed using the Wilcoxon rank-sum test (*P < 0.05, ⁎⁎P < 0.01). ASV, amplicon sequence variant; HBW, high body weight; LEfSe, linear discriminant analysis effect size; LDA, linear discriminant analysis; KEGG, Kyoto Encyclopedia of Genes and Genomes. To further explore potential functional differences associated with these compositional shifts, we performed functional prediction analysis. HBW selection was associated with enrichment of the Fanconi anemia pathway in the crop (Fig. 3B; P < 0.05), caffeine metabolism in the duodenum (Fig. 3C; P < 0.05), and both amino acid and secondary metabolite biosynthesis in the cecum (Fig. 3E; P < 0.05), while a trend toward enrichment of amino acid biosynthesis was observed in the ileum (Fig. 3D; P = 0.054). Inter-compartment microbial connectivity and community assembly differ between the two lines To examine whether relationships among microbial communities differed between the primitive and HBW lines, we next examined compositional similarity and microbial source contributions. In the primitive line, the highest similarity was observed between the crop and ileum, whereas the cecum showed the lowest similarity with other regions. In contrast, in the HBW line, higher similarity was detected among the crop, duodenum, and ileum, with moderate similarity between the cecum and other compartments (Fig. 4A). Notably, crop–ileum similarity was lower in the HBW line than in the primitive line. Consistent with these patterns, source tracking analysis revealed shifts in microbial contributions along the gastrointestinal tract (Fig. 4B, C). The contribution of the crop to the duodenum decreased from 75.28% to 69.20%, and to the ileum from 47.42% to 37.49%. In parallel, the contribution of the duodenum to the ileum increased (50.30% to 55.88%), whereas its contribution to the cecum decreased (47.59% to 35.31%). Reconfiguration of gastrointestinal microbial community assembly associated with HBW selection. (A) Pairwise Bray–Curtis dissimilarities of microbial communities among gastrointestinal segments at the ASV level. (B, C) SourceTracker analysis showing microbial source contributions among the crop, duodenum, ileum, and cecum in the two chicken lines. (D, E) βNTI-based null model and iCAMP analyses assessing microbial community assembly processes at different analytical resolutions. Statistical significance between lines was assessed using Student’s t-test (*P < 0.05, ⁎⁎P < 0.01), and differences among gastrointestinal segment combinations were evaluated using ANOSIM. ASV, amplicon sequence variant; HBW, high body weight; SEM, standard error of the mean; ANOSIM, analysis of similarities; βNTI, beta nearest taxon index; iCAMP, infer community assembly mechanisms by phylogenetic-bin-based null model analysis. To further assess the ecological processes underlying these changes, we quantified microbial community assembly mechanisms using βNTI-based null model analysis and iCAMP (Fig. 4D–E). The βNTI analysis showed that microbial communities in the crop and small intestine were primarily governed by homogeneous selection, whereas the cecum was mainly associated with heterogeneous selection (Fig. 4D). Compared with the primitive line, HBW selection induced moderate shifts in assembly processes across gastrointestinal compartments. In the crop, the contribution of homogeneous selection decreased from 81.25% to 73.47%, accompanied by an increase in undominated processes from 18.75% to 26.53%, indicating a shift toward more stochastic assembly. In the small intestine, βNTI-based analysis revealed increased contributions of homogeneous selection in both the duodenum (from 46.88% to 55.56%) and ileum (from 61.11% to 77.55%), together with reduced undominated processes. Meanwhile, heterogeneous selection in the duodenum increased from 3.13% to 11.11%, suggesting that HBW selection altered the balance of deterministic processes within the small intestinal environment. In the cecum, assembly patterns showed relatively limited changes, with slight increases in homogeneous selection (from 6.25% to 8.16%) and undominated processes (from 18.75% to 22.45%), accompanied by a decrease in heterogeneous selection (from 71.88% to 65.31%). iCAMP analysis further revealed compartment-specific shifts in assembly processes at the phylogenetic-bin level (Fig. 4E). Homogeneous selection remained a major process in the crop and small intestine, whereas dispersal limitation dominated assembly in the cecum. Following HBW selection, the crop and small intestine exhibited reduced contributions of homogeneous selection (crop: 92.79% to 80.43%; duodenum: 62.50% to 54.98%; ileum: 70.64% to 66.79%), accompanied by increases in dispersal limitation and undominated processes. In contrast, the cecum showed an increased contribution of homogeneous selection (from 14.65% to 22.18%), together with decreases in heterogeneous selection (from 7.16% to 4.04%) and dispersal limitation (from 69.58% to 64.60%). Although βNTI-based and iCAMP approaches differed in the estimated contribution of homogeneous selection in the small intestine, both analyses indicated that HBW selection modified microbial community assembly patterns across gastrointestinal compartments. These results suggest that HBW selection reshaped the relative contributions of deterministic and stochastic processes, with assembly responses depending on both gut compartment and analytical resolution. Ecological associations link host phenotypes with key microbial taxa To link microbial variation with host phenotypes, we examined associations between differential taxa and gastrointestinal traits (Fig. 5). In the duodenum, uncultured_bacterium_g__Monoglobus (ASV1696) was positively associated with VH: CD, whereas uncultured_bacterium_o__Clostridia_UCG−014 (ASV368) showed a negative association. In the ileum, Blautia (ASV29 and ASV75) was positively associated with VH/CD. In the cecum, uncultured_bacterium_o__Clostridia_vadinBB60_group (ASV1821) was positively associated with cecal length, norank_f__UCG−010 (ASV531) was positively associated with digesta pH, and norank_o__Clostridia_UCG−014 (ASV3772) was negatively associated with cecal length. Several cecal taxa, including Ruminococcaceae (ASV1199), Blautia (ASV313), and Christensenellaceae_R − 7_group (ASV911), were also positively associated with body weight. Ecological associations link host phenotypes with key microbial taxa. Spearman correlation analysis between LEfSe-identified differential ASVs and body weight as well as intestinal parameters. Correlation coefficients are shown within squares, and significance levels are indicated by asterisks (*P < 0.05, ⁎⁎P < 0.01). ASV, amplicon sequence variant; LEfSe, linear discriminant analysis effect size. Discussion This study revealed marked differences in gastrointestinal structure and microbial ecology between the primitive and HBW selected lines across multiple regions of the chicken gastrointestinal tract. Previous studies have shown that host genetic background and long-term selection for growth- or metabolism-related traits can alter the chicken gut microbiota, but these studies have generally focused on microbial differences within selected intestinal segments. In contrast, the present study characterized multiple gastrointestinal compartments and further examined not only region-specific shifts in microbial composition, but also changes in community assembly processes and microbial connectivity among compartments (Feng, et al., 2022; Yang, et al., 2022). These findings provide a more integrated view of how HBW selection is associated with spatial reorganization of the gastrointestinal microbiota. The crop, functioning as a temporary storage and initial fermentation chamber, plays a key role in regulating digesta flow and maintaining intestinal function, and is closely associated with feed intake and feed efficiency (Cutler, et al., 2005; Feigenbaum, et al., 1962). Compared with the primitive line, the HBW line had greater crop weight, accompanied by greater stochasticity in microbial community assembly. A larger crop may be associated with differences in feed intake in high-body-weight chickens, consistent with the positive correlation between crop size and body weight (Kierończyk, et al., 2016). However, the increased stochasticity, together with the absence of enrichment of representative fermentative taxa, suggests that this pattern may be consistent with greater variability in feeding-related digesta dynamics, although feed intake and digesta turnover were not directly measured in this study. Moreover, the lack of enrichment of fermentation-associated taxa indicates that crop fermentation capacity may not be a limiting factor for host growth under the current selection regime. In the gastric compartments, proventriculus weight remained unchanged, whereas gizzard weight increased, suggesting that the greater gizzard weight may be consistent with increased mechanical processing capacity, although its functional consequences for downstream digestion were not directly assessed. These patterns in crop and gastric indices are consistent with observations in males (Zhang, et al., 2026a). The small intestine represents the primary site of nutrient digestion and absorption, and HBW selection induced notable changes in this compartment. In the present study, HBW chickens exhibited increased ileal weight and higher villus height-to-crypt depth ratios in the duodenum and ileum, suggesting improved intestinal structural development and potentially enhanced absorptive capacity. Correspondingly, HBW selection altered microbial community assembly in the small intestine, although βNTI and iCAMP showed contrasting trends. βNTI indicated increased homogeneous selection in the duodenum and ileum, whereas iCAMP revealed reduced homogeneous selection together with greater contributions of dispersal limitation and undominated processes. This discrepancy likely reflects differences in analytical scale, with βNTI characterizing whole-community phylogenetic turnover and iCAMP resolving lineage-specific assembly processes. Importantly, these patterns are not necessarily ecologically contradictory. Different microbial lineages within the same intestinal compartment may respond differently to host-derived selection and dispersal constraints; such lineage-specific heterogeneity can be obscured when assembly is summarized at the whole-community level but becomes detectable after phylogenetic binning in iCAMP. Thus, rather than uniformly strengthening environmental filtering, HBW selection appears to have reorganized small-intestinal microbial communities through concurrent changes in host-associated selection and microbial dispersal. The improved intestinal morphology may have altered local habitat conditions and favored specific microbial lineages. Consistent with this interpretation, HBW chickens showed enrichment of potentially beneficial taxa, including uncultured Monoglobus (Kim, et al., 2019) in the duodenum and Blautia (Liu, et al., 2023a) in the ileum, which were associated with nutrient metabolism and intestinal adaptation. PICRUSt2 analysis further suggested altered microbial functional potential in the small intestine, with enrichment of predicted caffeine metabolism in the duodenum and a trend toward increased amino acid biosynthesis in the ileum. Although these predictions do not directly indicate microbial metabolic activity, they further suggest compartment-specific shifts in microbial metabolic potential accompanying the taxonomic and ecological changes. Meanwhile, the reduced contribution of upstream compartments to the small intestine observed in SourceTracker analysis suggests altered microbial transmission along the gastrointestinal tract, providing a possible explanation for the increased dispersal limitation detected by iCAMP. Together, these findings indicate that HBW selection reshaped small-intestinal microbial assembly by modifying both local host–microbe interactions and inter-compartment microbial connectivity, rather than simply strengthening deterministic environmental filtering. The cecum served as the primary site of microbial fermentation and plays a critical role in nutrient utilization and host metabolic regulation. In the HBW line, the cecal microbiota was enriched with taxa associated with efficient nutrient utilization, including Bacteroides fragilis (Zhang, et al., 2025), Faecalibacterium (Lee, et al., 2017), members of Lachnospiraceae and Blautia (Liu, et al., 2023a), Eubacterium (Gao, et al., 2022), Ruminococcus (Ze, et al., 2012), and the Christensenellaceae R-7 group (Pang, et al., 2022). PICRUSt2 analysis further suggested differences in predicted microbial functional potential related to nutrient metabolism; however, these predictions should be interpreted as putative functional capacity rather than direct evidence of microbial metabolic activity. One possible explanation is that differences in substrate delivery to the cecum may contribute to this pattern, positioning it as an important site for additional energy harvest and metabolic regulation. However, despite these predicted functional differences, changes in cecal microbial community assembly were relatively limited, and cecal weight showed a slight decrease. This apparent decoupling between taxonomic shifts and assembly processes is noteworthy. Previous studies have consistently reported differences in cecal microbial composition among chicken lines differing in genetic background, body weight, or selection history, whereas our results suggest that such compositional restructuring does not necessarily require a corresponding large-scale shift in the ecological processes governing community assembly (Ding, et al., 2016; Feng, et al., 2022). Although differences in substrate availability between the two lines could potentially contribute to the observed taxonomic differences, this mechanism was not directly assessed. The relative stability of cecal assembly may reflect factors such as niche saturation or microbial interactions, although these mechanisms require direct testing. Several limitations should be considered when interpreting these findings. First, the comparison was conducted between a female primitive line and an HBW line after seven generations of directional selection. In addition, because only female chickens were included, the present findings should not be directly generalized to males without further validation. Thus, the observed differences cannot be attributed exclusively to genetic selection for body weight, as they may also reflect differences in body weight itself and associated physiological changes. The results should therefore be interpreted as line-associated differences rather than direct causal effects of HBW selection. Second, physiological factors that may link host phenotype to microbial changes, such as feed intake and digesta transit, were not directly measured, limiting mechanistic interpretation of the observed associations. Finally, βNTI, iCAMP, and SourceTracker provide model-based inferences of community assembly and microbial transmission rather than direct measurements of ecological processes. Likewise, PICRUSt2 provides predictions of microbial functional potential based on taxonomic composition rather than direct measurements of gene expression or metabolic activity. These results should therefore be interpreted cautiously and require validation using metagenomic, metatranscriptomic, or metabolomic approaches. Future studies integrating complementary experimental designs with longitudinal physiological and multi-omics measurements will be needed to disentangle the respective contributions of host genetics, body weight, gastrointestinal physiology, and microbial ecological processes. Funding This study was supported by Shaanxi Engineering Research Center of LueYang Black Chicken (S2026-GC—NYBL-0012), Shaanxi Provincial Department of Science and Technology (2022GD-TSLD-46-0302), and Xianyang Science and Technology Bureau (L2024-ZDKJ-ZDCGZH-0007). Author contributions Fei Zhang conceived the study, conducted the animal experiments and sample measurements, analyzed the data, interpreted the results, and drafted the manuscript. Yao Wang participated in the animal experiments and contributed to data analysis and interpretation of the results. Jiale Qian, Shuai Li, Xutao Deng, and Mengya Lu participated in experimental measurements and contributed to data analysis and interpretation of the study findings. Yanli Liu, Xin Yang, and Jianqin Zhang contributed to the interpretation of the results and critically revised the manuscript for important intellectual content. Xiaojun Yang and Zhouzheng Ren contributed to the conception and design of the study, interpretation of the results, and critical development of the scientific content of the manuscript, and supervised the overall research. All authors read and approved the final version of the manuscript and agree to be accountable for the integrity of the work. Ethics declaration Animal subject This study was conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. This study was approved by the Animal Ethics and Welfare Committee of Northwest A&F University .(Approval No. DK2022009) Disclosures The authors affirm the absence of any conflicting interests. Acknowledgements The authors thank members of the Innovative Research Team of Animal Nutrition & Healthy Feeding (Northwest A&F University) for their help in animal care and sample collection. Footnotes Appropriate scientific section for the paper: Microbiology and Food Safety. Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.psj.2026.107712. Contributor Information Fei Zhang, Email: zhangfei98@nwafu.edu.cn. Yao Wang, Email: 3426024109@qq.com. Jiale Qian, Email: 2686779867@qq.com. Shuai Li, Email: 17639883347@163.com. Xutao Deng, Email: dengxutao@nwafu.edu.cn. Mengya Lu, Email: 2396811795@qq.com. Yanli Liu, Email: liuyanli@nwsuaf.edu.cn. Xin Yang, Email: yangx0629@163.com. Jianqin Zhang, Email: zhangjianqin1356@126.com. Zhouzheng Ren, Email: poultryren@nwafu.edu.cn. Xiaojun Yang, Email: yangxj@nwsuaf.edu.cn. Supplementary materials Data availability The 16S rRNA sequencing data have been deposited in the NCBI SRA under accession numbers PRJNA1229171. Other data are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Data Availability Statement The 16S rRNA sequencing data have been deposited in the NCBI SRA under accession numbers PRJNA1229171. Other data are available from the corresponding author upon reasonable request.