All animal testing protocols were approved by Animal Care and Use at Universitas Indonesia following the Universitas Indonesia Regulations of Animal Experimentation (ethical number: 471/UN2.F1/ETIK/PPM.00.02/2022).

Male Sprague-Dawley (SD) rats weighing 150–200 g were purchased from the Indonesian Food and Drug Authority, Indonesia and housed under conventional conditions maintained under a 12 h light/dark cycle with water and food provided ad libitum. One cage contained one rat to facilitate the calculation of feces' weight every day.

Thirty-six SD rats were randomly divided into a sham-operated group (sham) (n = 6) and a 0.1% BAC group (n = 30). The BAC group was further divided into five groups based on the time of termination. Week 1 (W1), Week 3 (W3), Week 5 (W5), Week 8 (W8), and Week 12 (W12) postintervention. To minimize suffering, all rats were anesthetized with 87 mg ketamine/kg and 13 mg xylazine/kg. Subsequently, laparotomy was performed using a 2-cm-long midline incision to identify the sigmoid colon. To prevent bowel dehydration, the paper gauze was wetted with three drops of 0.1% of BAC solution every 5 min. After 30 min of dripping BAC, the paper gauze was released, and the sigmoid colon was washed with 0.9% normal saline. The abdomen was then closed, and all animals were individually housed with a standard diet and tap water ad libitum. The sham group rats underwent a sham operation, and the sigmoid colon was washed with 0.9% normal saline for 30 min.

Abdominal circumference measurements were carried out in all groups of 36 rats, including the sham group (n = 6) and the BAC group (n = 30). Before sacrifice, abdominal circumference was measured in the largest zone of the rat's abdomen in the ventral position using a plastic non extensible measuring tape with an accuracy of 0.1 cm.

All rats were weighed once a week, and their feces were weighed daily. All rats were euthanized by intraperitoneal injection with an overdose of ketamine/xylazine and were sacrificed by exsanguination under deep anesthesia at W1, W3, W5, W8, and W12 after BAC application. The isolated segment of the sigmoid colon was excised from each rat and divided into two parts. One part was promptly fixed in 10% neutral buffered formaldehyde solution for further histological examination, and the other part was washed with cold phosphate-buffered saline (PBS), snap-frozen in liquid nitrogen and stored at −80°C for different experiments. Fresh feces of rats in all groups were collected in sterile tubes and stored at −80°C until further analysis.

For light microscopy observation, the sigmoid colon segment from all rats fixed in 10% neutral buffered formaldehyde solution was dehydrated, embedded in paraffin, and then cut into 5-μm sections. These sections were then stained with hematoxylin-eosin and analyzed by two pathologists blinded to the study. The number of ganglionic cells per ganglia was counted and recorded. Teiltelbaum's scoring was used to analyze the degree of enterocolitis histologically. Grade 0, normal intestinal mucosa; Grade 1, crypt dilation and mucin retention; Grade 2, cryptitis or crypt abscesses; Grade 3, multiple crypt abscesses; Grade 4, crypt hyperplasia and inflammatory cell infiltration; and Grade 5, crypt necrosis (19). Sigmoid colon tissue was used to detect Paneth-like cells based on their eosinophilic granules in the Lieberkühn crypts with hematoxylin-eosin staining using light microscope at a magnification of 1000× (22).

For immunohistochemistry (IHC), formalin-fixed, paraffin-embedded sigmoid colon tissue sections were used. After deparaffinization and hydration, the slides were washed in Tris-buffered saline (TBS; 10 mM/L Tris-HCl, 0.85% NaCl, pH 7.2). Endogenous peroxidase activity was quenched by incubating the slides with 0.3% H₂O₂ in methanol. The slides were incubated overnight with mouse monoclonal α-defensin 1 antibody (Novusbio, NBP2-75406, Centennial, USA) as the primary antibody, diluted at a ratio of 1:50 at 4°C. Subsequently, the slides were washed in TBS, added with HRP-conjugated recombinant anti-mouse antibody, incubated at 20–22°C for 45 min, then washed again with TBS, incubated with diaminobenzidine tetrahydrochloride as a substrate, and then counterstained with hematoxylin. For semi-quantitative analysis of α-defensin 1, the number of positive cells with strong expression was counted.

Cryosections (5-μm) were cut using a microtome (Bright Instrument Company Ltd., Huntington, UK) housed within a cryostat at −25°C. The sections were collected onto uncoated, precleaned glass slides fixed in 3.7% formaldehyde for 1 h. After rinsing in double distilled water three times for 30 s each, the slides were permeabilized in 0.5% Triton X-100 for 5 min. Subsequently, the slides were washed in PBS (137 mmol/L sodium chloride, 3 mmol/L potassium chloride, 8 mmol/L disodium hydrogen phosphate, and 3 mmol/L potassium dihydrogen phosphate, pH 7.4) for three times 5 min each. The sections were incubated in anti-peripherin antibody (ab4666, Abcam, Cambridge, UK) for 2 h. Following washing in PBS three times for 5 min each, a secondary Alexa Fluor fluorescent-conjugated antibody was applied for 30 min. Secondary antibody controls were included as negative controls. Images were taken with a fluorescence microscope (Olympus, BX43, Japan).

Total RNA was extracted from sigmoid colon tissue using a High Pure RNA Isolation kit (Roche Applied Science, Penzberg, Germany) according to the manufacturer's instructions. Nanodrop 1000 Spectrophotometer (Thermo Scientific) was used to measure RNA concentration at a wavelength of 260 nm. To synthesize complementary DNA, 1 μg of RNA, and the Transcriptor First Strand cDNA Synthesis kit (Roche Applied Science) were used according to manufacturer's instructions. Quantitative real-time PCR (qRT-PCR) was conducted using the LightCycler® 480 Instrument (Roche Applied Science) with FastStart Essential DNA Green Master Mix (Roche Life Science). All reactions were performed similarly: 95°C for 10 s, followed by 45 cycles of 95°C for 15 s and 60°C for 1 min. Relative gene expression was calculated using the 2^−ΔΔCT method, with β-actin as the housekeeping gene. The following primer pairs were used: for β-actin: Forward: 5′-CTGGTCGTACCACAGGCATT-3′ Reverse: 5′-CTCTTTGATGTCACGCACGA-3′; for α-defensin (cryptdins-1): Forward: 5′-CCG AGA GTG CTT CCT AAA CTA C-3′ Reverse: 5′-AAA GTC TCA GGT GGG ATG TTA G-3′; for lysozyme: Forward: 5′-GAA TGG GAT GTC TGG CTA CTA TG-3′ Reverse: 5′-GTC TCC AGG GTT GTA GTT TCT G-3′; for sPLA2: Forward: 5′-ACT CAT GAC CAC TGC TAC AAT C-3′ Reverse: 5′-GTA TGA GTA CGT GTT GGT GTA GG-3′; and for IL-1β: Forward: 5′-CCA GGA TGA GGA CCC AAG CA-3′ Reverse: 5′-TCC CGA CCA TTG CTG TTT CC-3′.

The frozen sigmoid colonic tissues were weighed and homogenized in an ice-cold Tris buffer (50 mM Tris-HCl, pH 7.4, 200 mM NaCl, 20 mM NaF, 1 mM Na₃VO₄, 1 mM 2-mercaptoethanol, 0.01 mg/mL leupeptin, 0.01 mg/ml aprotinin). Homogenates were subsequently centrifuged (3,000× g, 10 min, 4°C), and the supernatants were collected and stored at −80°C. The total protein concentration in each sample was measured by the bicinchoninic acid method. Equal amounts of protein extract (50 μg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Bio-Rad, CA, USA) and electrophoretically transferred to nitrocellulose membranes. Then, the membranes were blocked with 5% skim milk in TBS Tween 20 (20 mM Tris, pH 7.6, 137 mM NaCl and 0.1% Tween 20). Subsequently, the membranes were incubated overnight at 4°C and agitated with primary antibodies against acetylcholinesterase (68 kDa, 1:500, ab97299; Abcam, Cambridge, UK), butyrylcholinesterase (75 kDa, 1:500, AF9024; R&D systems, USA), and occludin (60 kDa, 1:1,000, sc-133256; Santa Cruz, CA, USA). The membranes were washed three times in TBS-T for 15 min and incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (1:5,000, Santa Cruz) or rabbit anti-mouse IgG secondary antibody (1:5,000, Santa Cruz) for 1 h at room temperature. After a final wash, signals were detected using a chemiluminescence system (ECL western blotting substrate, Life Technologies). To normalize the expression of these antibodies, the protein expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in the same sample was required. Finally, the membranes were scanned and analyzed using ImageJ software.

Bacterial DNA was extracted from 200 mg rat feces per sample using the ZymoBIOMICS™ DNA Miniprep Kit (catalog #D4300) according to the manufacturer's protocols and then amplified with 341F (CCTAYGGGRBGCASCAG) and 806R (GGACTACNNGGGTATCTAAT) primers, including the V3–V4 region of the bacterial 16S rRNA gene. The PCR products with the proper size were selected by 2% agarose gel electrophoresis. The exact amount of PCR products from each sample was pooled, end-repaired, A-tailed, and ligated with illumine adapters. To generate 250 bp paired-end raw reads, libraries were sequenced on a paired-end Illumina platform. The library was checked with Qubit and real-time PCR for quantification and a bioanalyzer for size distribution detection. Quantified libraries were pooled and sequenced on the Illumina platform according to the required effective library concentration and data amount. Principle component analysis (PCA) was used to analyze the beta diversity, whereas Chao1 index and Shannon index were used to analyze the alpha diversity of gut microbiota among groups. PCA is a statistical procedure to extract principle components and structures in data by using orthogonal transformation and reducing data dimensionalities.

gDNA samples were amplified with target-specific primer (16SV3-V4). Library preparation was performed using the final PCR products. The final library was sequenced on Illumina platform to generate paired-end raw reads. Adapter and PCR primer sequences from the paired-end reads were removed using Cutadapt. DADA2 was used to correct sequencing errors and remove low-quality sequences and chimera errors. The resulting ASVs data was used for taxonomic classification against SILVA (silva_nr99_v138.1).

Data were analyzed using Prism GraphPad version 9.5.1 software (Boston, MA 02110). Data were reported as the mean ± standard error of the mean. Statistical significance was determined using the one-way ANOVA followed by Tukey's post hoc test for multiple comparisons. Differences among groups were considered significant at p ≤ 0.05.

All rats survived until the end of the study. At the end of the intervention, the body weight and feces weight were significantly lower in the BAC groups than in the sham group (Figure 1A). However, the difference of feces weight was significant only in W3 BAC group (Figure 1B). Hypoganglionosis caused abdominal distention, which increased according to the time of termination was associated with BAC administration, as supported by the increase in abdominal circumference (Figure 1C).

The number of ganglion cells per ganglia was reduced in the BAC group compared to that in the sham group (Figure 1D). The quantitative analysis showed that the mean number of ganglion cells per ganglia decreased with the time of termination (Figure 1E). Application of BAC successfully resulted in narrow and dilated segments.

All groups had peripherin, an intermediate filament specifically expressed to the peripheral nervous system including enteric ganglion cells. Compared to the sham group, the BAC group showed a decreased expression of peripherin according to the time of termination (Figure 1F).

Next, to evaluate enteric nervous system status, we checked the protein expression of AChE and BChE by Western blotting method. The ratio of AChE/BChE was higher in the BAC groups than in the sham group, mainly at Week 5 postintervention (Figure 2A). This indicates an extrinsic nerve fiber hypertrophy in the sigmoid colon segment. Next, to evaluate tight junction proteins, we examined the expression of occludin protein, one of the tight junction proteins that regulates colonic permeability, in sigmoid colon tissue. Compared to the sham group, the BAC group showed a decreased expression of occludin according to the time of termination (Figure 2B). The functions of the intestinal epithelium are associated with tight junction integrity. To analyze the presence of an inflammatory process due to the loss of intestinal tight junction barrier function, we measured the gene expression of IL-1β, which was increased in the BAC group, mainly at Week 8 postintervention (3-fold) compared to that in the sham group (Figure 2C).

On hematoxylin-eosin staining and immunohistochemical analysis of α-defensin to detect PLCs in the sigmoid colon tissue, while the α-defensin-positive PLCs were almost absent in the sham group, these continued to increase in the BAC groups according to the time of termination and peaked at Week 5 postintervention, which proved that PLCs metaplasia occurred in the sigmoid colon tissue of hypoganglionosis rat models (Figures 3A,B). However, no changes were noted in the PLCs in the small intestine adjacent to the colon as they were not mitotic (Supplementary Figures S1, S2).

The gene expression levels of α-defensin (cryptdin), sPLA2, and lysozyme started to increase at Week 1 postintervention and then decreased until the end of the study (Figures 3C–E).

The composition of luminal microbiota in the rectosigmoid of the BAC and sham groups was assessed by 16S rDNA sequencing each of four animals. Hierarchical clustering analysis was performed at each taxonomic level from phylum through genus at the time of termination points. At the phylum level, Firmicutes and Bacteroidetes were the two major bacterial phyla in the gastrointestinal tract of both the sham and the BAC groups (Figure 4A). The Firmicutes/Bacteroidetes (F/B) ratio decreased in the BAC group compared to the sham group according to the time of termination (Figure 4B). To further understand the phyla contributing to these grouping, the top three phyla found at each time point of termination were analyzed between groups (Figure 4C). In the sham group, the representation of Bacteroidetes and Proteobacteria decreased while that of Firmicutes increased. Conversely, in the BAC group, Bacteroidetes and Proteobacteria increased, and Firmicutes decreased, mainly at Weeks 8–12 postintervention, but without significant differences between groups (Figure 4C).

At the genus level, the top ten genera in each group are shown in Figure 5A. Prevotella is the most common genus found in the sham and BAC groups. Lactobacillus, a species considered to provide intestinal mucosal protection, was higher in the sham group and decreased in the BAC group. In contrast, Bacteroides increased significantly at Week 8 postintervention compared to the other groups (Figure 5B).

Richness (Chao1index) and diversity (Shannon index) were similar between the sham group and BAC group (Figure 5C). However, the richness and diversity of microbial communities were significantly different between early weeks (Week 1 and Week 3) and late weeks (Week 5–Week 12) postintervention of BAC. PCA described the β diversity, revealing a separation between the sham group and the BAC group at Weeks 8–12 (Figure 5D). PCA showed that PC1 and PC2 accounted for 49.40% and 26.75% variance, respectively, and PC1 clearly separated the samples in the BAC group in the early and late weeks. This implies differences in the structure of the gut microbiota in early and advanced hypoganglionosis.

As shown in Figures 6A,B, the levels of abundance of microbiota phyla and genera did not differ in the sham group sacrificed at various time points, while the microbiota diversity at Week 12 was different between the sham group and the BAC-treated group (Figures 6C,D).

The degree of enterocolitis Grade 1–5 was determined in the BAC group (Figure 7A). Grade 1 enterocolitis began to occur in Week 1 postintervention and increased according to the time of termination (Figure 7B).