Age-related remodeling of the intestinal mucus barrier in pigs: implications for post-weaning diarrhea

Highlights

• •Postnatal maturation drives major changes in goblet cells and mucus characteristics.
• •MUC2 and X409 show distinct age- and segment-dependent developmental patterns.
• •Disease-associated differences were predominantly identified in the jejunum.
• •Diarrheic pigs had fewer luminal goblet cells and reduced X409 labeling at day 67.
• •Timely mucus barrier maturation may be important for resilience post weaning.

Abstract

The intestinal mucus barrier plays a central role in maintaining intestinal integrity, yet its development during early life and weaning in pigs remain incompletely understood. This study characterized age-related changes in goblet cell populations and mucus properties and investigated their association with diarrhea.

Thirty Danish crossbred piglets from a longitudinal cohort were stratified by age (4, 14, 25, 49, and 67 days) and diarrhea status (cases vs. controls). Jejunal and colonic tissues were evaluated by histology and immunohistochemistry for MUC2 and MUC5AC, while mucus glycosylation was assessed using the bacteria-derived mucin-binding probe X409. Digital image analysis was performed at crypt level.

Marked age-dependent remodeling occurred in both intestinal segments. Crypt-villus unit depth and goblet cell abundance generally increased with age, accompanied by age-related changes in MUC2 and X409-labeling. X409 staining intensity varied with age and position along the crypt-villus axis, indicating age-dependent changes in the spatial distribution of mucus-associated O-glycosylation. Disease-associated differences were limited. In jejunum at 67 days of age, cases had fewer luminal goblet cells and reduced X409 labeling than controls, while colonic crypt depth differed between cases and controls at selected ages. No MUC5AC expression was observed.

These findings demonstrate substantial age-dependent quantitative and qualitative remodeling of the intestinal mucus barrier during postnatal development. Limited disease-associated differences suggest that diarrhea was not characterized by widespread alterations in mucus properties, although localized differences, particularly in older piglets, may be associated with intestinal vulnerability. These results highlight the importance of mucus barrier maturation during the developmental period arround weaning.

Graphical abstract

Implications

Post-weaning diarrhea is a well-known challenge in pig production and is influenced by multiple factors beyond infection. This study provides new insight into how the intestinal mucus barrier changes during this period, showing that both the number of mucus-producing cells and the quality of the mucus are markedly altered during postnatal development and across the weaning period. These changes may affect how well piglets cope with the transition and help explain why some animals are more susceptible to diarrhea than others. Supporting gut barrier development – potentially through improved weaning strategies – could possibly help reduce disease and reliance on antibiotic treatments.

1. Introduction

The intestinal mucosal barrier provides the first line of defense against luminal microorganisms and other potentially harmful substances while simultaneously allowing for nutrient uptake (Weström et al., 2020). The barrier comprises four interacting layers including 1) the luminal layer of symbiotic microorganisms, 2) the mucus layer, 3) the mechanical layer formed by epithelial cells and tight junctions, and 4) the underlying immunological layer of gut-associated immune cells (Song et al., 2023). Together, these components are essential for maintaining barrier integrity, and deficiencies in one layer can lead to impaired barrier function, making the host more susceptible to gastrointestinal diseases (Camilleri et al., 2012). During early life, these components undergo substantial structural and functional maturation. In piglets, intestinal epithelial and mucus-associated cell populations are immature at birth and continue to develop throughout the suckling period and after weaning (Modina et al., 2021; Weström et al., 2020).

Weaning represents a particularly important transition in this process. The abrupt change from milk to solid feed, exposure to environmental and social stressors, and concurrent changes in the intestinal microbiota impose substantial challenges on the developing intestinal barrier (Moeser et al., 2017). This period coincides with a high incidence of post-weaning diarrhea (PWD), which remains an important cause of reduced productivity and antimicrobial use in commercial pig production (Eriksen et al., 2021, 2023). Although PWD is multifactorial and involves interactions between infectious agents, host susceptibility, nutrition, microbiota and environmental factors, impaired maturation or disruption of the intestinal barrier may increase susceptibility to enteric disease (Eriksen et al., 2023; Rydal et al., 2025; Son & Kim, 2025). Thus, understanding how the intestinal barrier develops during the transition from suckling to post-weaning life may provide insight into why some piglets are more resilient to diarrhea than others.

The mucus layer is an important component of the intestinal barrier. It separates the intestinal epithelium from the luminal microbiota and contributes to microbial homeostasis by providing nutrients and attachment sites for mutualistic bacteria while limiting access of pathogenic microorganisms to the epithelial surface (Gustafsson & Johansson, 2022). Mucins are the principal structural components of mucus, with the secreted gel-forming mucin MUC2 being predominant in the intestine (Bansil & Turner, 2006; Johansson et al., 2014). MUC2 is synthesized and secreted by goblet cells, and changes in goblet cell abundance, mucus secretion and mucin expression can therefore substantially affect barrier function (Cornick et al., 2015; Johansson et al., 2014; Van Der Post et al., 2019). In pigs, such alterations have been described in response to both infectious and non-infectious challenges. These include Lawsonia intracellularis (Bengtsson et al., 2015; Zapata et al., 2015), porcine circovirus type 2 (Jung & Saif, 2017; Zlotowski et al., 2009), porcine reproductive and respiratory syndrome virus (Lin et al., 2023; Quintana-Hayashi et al., 2015; Zhao et al., 2021), and porcine delta coronavirus infections (Zhang et al., 2023), as well as medicinal and dietary interventions (Desantis et al., 2019; Sawaed et al., 2024; Tang et al., 2024). Some enteric pathogens can additionally induce expression of mucins not normally present in the intestine, such as MUC5AC (Hasnain et al., 2011; Kim et al., 2009; Lin et al., 2023; Quintana-Hayashi et al., 2015).

Beyond mucin abundance and expression, mucus function is influenced by post-translational modification of mucins, particularly O-glycosylation (Bansil & Turner, 2006; Chatterjee et al., 2023; Fekete & Buret, 2023; Xia et al., 2022a). The glycans attached to the mucin backbone contribute to mucus structure and stability and provide binding sites and nutrient sources for intestinal microorganisms (Johansson et al., 2011; Kamphuis et al., 2017). Altered mucin glycosylation can therefore affect interactions between the mucus barrier and the intestinal microbiota and may influence susceptibility to enteric disease (Fekete & Buret, 2023; Padra et al., 2018; Quintana-Hayashi et al., 2018). In pigs, weaning has been associated with changes in mucin O-glycan composition (Xia al., 2022), while dietary interventions and enteric pathogens can also alter mucin glycosylation patterns (Desantis et al., 2019; Venkatakrishnan et al., 2017). These findings indicate that mucus maturation involves not only changes in the number and composition of mucins but also changes in their structural properties.

However, the development of the intestinal mucus barrier across the weaning transition remains poorly characterized, particularly in relation to susceptibility to diarrhea. Previous studies have largely assessed mucus composition or glycosylation using ex vivo samples, providing limited information on the properties of the mucus barrier within intact tissue. Combining quantitative assessment of goblet cell populations and mucin expression with in-situ analysis of mucin glycosylation using the bacteria-derived mucin-binding molecule X409 (Nason et al., 2021), which binding is dependent on the presence of clustered O-glycans specific to mucins (Jaroentomeechai et al., 2025), provides an opportunity to characterize complementary aspects of mucus barrier development within their tissue context.

In the present study, we therefore investigated age-related changes in goblet cell populations and mucus characteristics within these cells in the jejunum and colon across early life and the weaning period and assessed whether these patterns differed between piglets with diarrhea and clinically unaffected individuals. The selected ages encompass key stages of intestinal development including early suckling (4 days), established suckling (14 days), the immediate pre-weaning period (25 days), and the post-weaning period (49 and 67 days). The two oldest age groups represent the developing intestinal barrier several weeks after weaning, when the immediate physiological response to weaning is expected to have subsided and long-term maturation processes can be examined. Goblet cell populations and mucin expression were assessed by immunohistochemistry for MUC2 and MUC5AC, while mucus glycosylation was evaluated using X409-labeling. By characterizing both quantitative and qualitative aspects of the mucus barrier across this developmental window, the study aimed to improve the understanding of intestinal barrier maturation and its potential relationship with susceptibility to PWD.

2. Materials and methods

2.1. Animals and study design

At a conventional Danish pig herd, 2,550 piglets (Danish crossbred: Duroc x Yorkshire/Landrace) from 170 litters were followed from birth until 67 days of age across 11 batches. From each litter, 15 newborn piglets were randomly selected, ear-tagged, and remained with their biological dam until weaning at day 26 of age. Individual daily clinical observations, including unthriftiness, perianal fecal staining, and any potential need for treatment, were recorded throughout the study period. In addition, diarrhea was assessed weekly using a fecal scoring system previously described by Eriksen et al. (Eriksen et al., 2024), in which scores 1 and 2 denote firm and soft feces, respectively, whereas scores 3 and 4 denote runny and watery fecal consistency.

A total of 145 piglets representing both sexes were randomly selected from seven of the 11 batches using the RAND() function in Excel. Piglets were selected in the following age groups: 20 piglets at 4 days, 25 piglets at 14 days, 25 piglets at 25 days, 45 piglets at 49 days, and 30 piglets at 67 days of age, respectively. The age groups 4, 14, and 25 days represented different stages of early life and the pre-weaning period, whereas the two later time points (49 and 67 days of age) represented later stages of intestinal barrier maturation. Within each age group, three case piglets and three control piglets were subsequently selected for the present study.

For the age groups euthanized at 4, 49 and 67 days of age, case and control status were defined according to the fecal score recorded at euthanasia. Piglets with fecal scores of 3 or 4 were classified as cases, whereas piglets with fecal scores of 1 or 2 were classified as controls. Among piglets euthanized at 14 and 25 days of age, none had a fecal score of 3 or 4 at euthanasia. Therefore, case piglets in these age groups were selected based on prior clinical observations of diarrhea, while control piglets had no recorded history of diarrhea before euthanasia. None of the selected piglets had received any antibiotic treatment throughout the study period.

2.2. Sample collection and tissue fixation

The piglets were transported for approximately one hour from the herd of origin to University of Copenhagen, Frederiksberg, Denmark, on the morning of euthanasia. Animals were anesthetized upon arrival via intramuscular injection of Zoletil (0.1 mL/kg) and subsequently euthanized by intracardiac injection of pentobarbital. Following evisceration, tissue samples were collected from the mid-jejunum and from the apex of the colonic spiral. Each sample was secured by excision, and the unopened segments were then immediately submerged in Carnoy’s solution to ensure rapid fixation and preservation of the mucus layer. All tissue samples were secured within 15 minutes of euthanasia.

2.3. Tissue processing

Samples were fixed in Carnoy’s solution for 24 hours and subsequently transferred to 99% ethanol until further processing. Tissues were trimmed, dehydrated, embedded in paraffin, sectioned, and mounted on glass slides according to standard histological protocols.

2.4. Immunohistochemistry and histochemical labeling

For direct histological visualization of mucus type and quality, immunohistochemistry targeting the mucin backbone was combined with X409 labeling of mucin-restricted clustered O-glycans to achieve high specificity for mucin-containing structures. A schematic overview of the differences in target recognition between mucin-directed immunohistochemistry, X409 labeling, and the more commonly applied lectin staining methods is shown in Fig. 1.

Fig. 1.

Schematic comparison of molecular targets recognized by different mucus visualization approaches. Mucin-directed antibodies bind epitopes within the mucin core protein backbone. X409 recognizes clustered mucin-associated O-glycans, whereas lectins bind specific terminal carbohydrate motifs also present on non-mucin glycoproteins. Created in BioRender. Brandt becker, C. (2026).

2.4.1. MUC2 and MUC5AC immunohistochemistry

Immunohistochemical staining for MUC2 was performed using a polyclonal anti-MUC2 antibody (Thermo Fisher Scientific, PA5-79702), while MUC5AC labeling was performed using a monoclonal anti-MUC5AC antibody (clone 45M1, Thermo Fisher Scientific, MA5-12178). The anti-MUC5AC antibody was reported by the manufacturer to have species cross-reactivity to pigs, while the anti-MUC2 antibody was tested on porcine tissues in our own lab for relevant staining prior to this study. Immunohistochemical staining for MUC2 and MUC5AC was performed using identical staining protocols, differing only in the antigen retrieval pre-treatment step and detection system as follows.

After deparaffinization, tissue sections underwent heat-induced epitope retrieval for 60 minutes at 95°C using either Tris–EDTA buffer for MUC2 or citrate buffer for MUC5AC. Sections were subsequently cooled and rinsed in Tris-buffered saline (TBS). Endogenous peroxidase activity was quenched by incubation with 3% hydrogen peroxide for 10 minutes, followed by rinsing and blocking with 2.5% normal goat serum for 5 minutes. Sections were then incubated with the respective primary antibodies overnight at 4°C. After rinsing, sections were incubated for 30 minutes with an HRP-conjugated polymer detection system (Vector laboratories, MP-7451 or MP-7452), followed by visualization using an aminoethyl carbazole (AEC) peroxidase substrate. Slides were rinsed in tap water, counterstained with Mayer’s hematoxylin, and mounted using glycerol gelatine.

Detailed staining protocols for both antibodies are provided in the Supplementary material S1 and S2.

2.4.2. X409 mucin-binding assay

Mucus glycosylation was assessed using the recombinant mucin-binding molecule X409, previously described by others (Nason et al., 2021). X409 corresponds to the mucin-binding domain of the mucinase “secreted protease of C1 esterase inhibitor” (StcE), a zinc metalloprotease produced by enterohemorrhagic E. coli. Binding of X409 is dependent on the presence of clustered O-linked glycan specific to mucins but is independent of specific glycan structures (Jaroentomeechai et al., 2025). The purified recombinant X409 (residues 797-898), containing N-terminal a MHHHHHHSSHENLYFQG linker, was obtained from Copenhagen Center for Glycomics (Nason et al., 2021). Binding of X409 to porcine mucins was confirmed using a porcine mucin array and by staining porcine gastrointestinal tissues prior to the study (data not shown). Tissue staining was performed according to the protocol described in Supplementary material S3. In brief, after deparaffinization, tissue sections underwent heat-induced pretreatment in citrate buffer (pH 6.0), after which sections were rinsed in TBS. Endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide, followed by blocking with normal horse serum. To reduce non-specific binding, an avidin-biotin blocking step was subsequently applied. Sections were incubated overnight at 4°C with 5µg/mL 6xHis-tagged X409 (Copenhagen Center for Glycomics, 1:200 dilution). After washing, bound X409 was detected using a mouse anti-His tag antibody (AD1.1.10, Santa Cruz Biotechnology, sc-53703), followed by incubation with a biotinylated horse anti-mouse IgG secondary antibody kit (Vector laboratories, PK-6102). Signal amplification was performed using avidin-biotin-peroxidase complex (ABC). Visualization was achieved using an AEC peroxidase substrate. Slides were counterstained with Mayer’s hematoxylin and mounted with glycerol gelatine. An overview of the applied primary reagents is provided in Table 1.

Table 1.

Primary reagents applied for histopathological analysis.

Reagent	Type	Host or recombinant expression system	Target	Manufacturer	Dilution
MUC5AC	Monoclonal, clone 45M1 [MA5-12178]	Mouse	Mucin 5AC	Thermo Fisher Scientific	1:160 000
MUC2	Polyclonal [PA5-79702]	Rabbit	Mucin 2	Thermo Fisher Scientific	1:1000
X409	Recombinant mucin-binding domain from a mucinace (StcE)	Chemically competent E. coli	O-glycan patches on mucins	Copenhagen Center for Glycomics	1:200

2.5. Slide digitization

All MUC2 and X409-stained histological slides were digitized by whole-slide scanning using a Zeiss Axioscan Z1 scanner equipped with a 20x/0.8 Plan-Apochromat objective. Scanner settings were optimized for brightfield microscopy of intestinal tissues.

2.6. Histopathological and digital image analysis

Histopathological evaluation was performed using conventional brightfield microscopy for the MUC5AC-labelled specimens, and digital image analysis for the MUC2- and X409-labelled specimens. Whole-slide images were analyzed using the open-source software QuPath (Bankhead et al., 2017), and all tissue samples were assessed blinded to clinical status. A detailed workflow for slide evaluation and quantification is provided in Supplementary material S4.

2.6.1. MUC5AC assessment

MUC5AC expression was assessed qualitatively by conventional brightfield microscopy using a Leica DMLB microscope. Slides were evaluated for the presence of de novo MUC5AC expression in goblet cells, and samples were assigned a binary score indicating absence (-) or presence (+) of detectable staining.

2.6.2. Crypt selection and manual quantification

MUC2 and X409 staining were assessed using digital image analysis to enable quantitative evaluation. For each piglet, staining, and intestinal segment, ten longitudinally sectioned crypts or villus-crypt units were selected for quantitative analysis (Fig. 2). Crypts were selected based on complete representation from crypt base to luminal opening, adequate tissue preservation, and absence of sectioning artefacts. Crypts with substantial tissue damage, oblique sectioning, incomplete representation, or other artefacts interfering with identification were excluded and replaced by another eligible crypt. The same selection criteria were applied across staining procedures and intestinal segments.

Fig. 2.

Schematic illustration of crypt compartmentalization used for quantitative analysis. Intestinal crypts (colon) and crypts with associated inter-villous space (jejunum) were divided longitudinally from the luminal surface (0%) to the crypt base (100%). For quantitative analyses, crypts and villus-crypt units were separated into luminal (0–50%) and basal (50–100%) compartments. For X409 staining, these lengths were further subdivided into five depth intervals (<20%, 20–40%, 40–60%, 60–80%, and >80%) to assess depth-dependent variation in staining intensity. Goblet cells and staining parameters were quantified within the defined compartments. Illustration created in BioRender. Brandt Becker, C. (2026). https://BioRender.com/9xhue1g.

Selected crypts or villus-crypt units were manually annotated using the brush tool (Fig. 3D). The length of each crypt or villus-unit, defined as the distance from the luminal surface to the crypt base, was measured using the polyline tool. The crypt or villus-crypt unit was subsequently divided into luminal and basal compartments based on crypt length. In the jejunum, the intervillous space was included in the luminal compartment (Fig. 2).

Fig. 3.

Identification and quantitative assessment of goblet cells and mucus staining. Representative images illustrating the criteria and workflow used for goblet cell identification and quantitative image analysis. A–C) Goblet cells showing different degrees of staining intensity: A) negative, B) faintly positive, and C) strong staining, scale bar = 20 µm. Goblet cells are indicated by black arrows. D) Representative colonic crypt manually annotated for quantitative analysis. The crypt outline is shown in yellow, with crypt length and the division into luminal and basal compartments indicated by red lines, scale bar = 100µm. E) Representative colonic crypt showing the estimated positive staining area generated by the machine-learning classifier, outlined in yellow, scale bar = 100 µm. F) Representative optical density image from a colonic crypt illustrating selection of cells for optical density measurements. Crypt outlines are shown in pale pink, crypt length and compartment boundaries in red, and selected cells are indicated by yellow cellular outlines. Scale bar = 100 µm.

Goblet cells were identified morphologically and manually quantified using the point annotation tool. The total number of goblet cells and the number of goblet cells showing positive MUC2 or X409 staining were recorded for each compartment. The criteria used for morphological identification of goblet cells, including morphologically distinct goblet cells without detectable cytoplasmic MUC2- or X409-staining, are illustrated in Fig. 3A-C.

2.6.3. Quantification of positive staining area

In addition to cell-based quantification, the proportion of each annotated crypt or villus-crypt unit showing positive staining (MUC2, X409) was quantified. Positive staining was identified using a trained artificial neural network (ANN) pixel classifier in QuPath (Fig. 3E). The classifier was trained on representative images to recognize positive staining based on Gaussian, Laplacian of Gaussian, structure tensor eigenvalue (maximum), and Hessian determinant features. Classification was performed at a pixel resolution of 0.44 µm. Classifier performance was evaluated by visual inspection of the classified images across the analyzed specimens, which demonstrated consistent identification of positive staining.

2.6.4. X409 staining intensity

For X409, staining intensity was additionally evaluated by measuring the optical density sum of selected goblet cells, normalized to the area of each cell, at defined positions along the crypt or villus-crypt unit (Fig. 3F). The unit was divided into five regions according to the distance from the luminal surface: <20%, 20-40%, 40-60%, 60-80%, and >80% of total crypt length (Fig. 2). Optical density represents the amount of light absorbed or scattered by the stain and was calculated as: $OD=Lo{g}_{10}\left(\frac{I_{background}}{I_{stainedarea}}\right)$. The resulting OD values were converted into a semi-quantitative staining intensity score: <0.2 (0, negative), 0.2-0.4 (1, weak), 0.4-0.6 (2, moderate), and >0.6 (3, strong).

2.7. Statistical analysis

All statistical analyses were performed separately for jejunum and colon using R software (version 4.5.1) (R Core Team, 2024). Because 10 crypts were evaluated within each piglet, crypt-level observations were treated as repeated measurements within each piglet. Linear mixed-effects models were fitted using the lme4 package (Bates et al., 2015), with piglet ID included as a random effect. Age, diarrhea status, and the age x diarrhea status interaction were included as fixed effects. Age was treated as a categorical variable with five levels (4, 14, 25, 49, and 67 days).

For parameters describing goblet cell numbers, MUC2 and X409-positive cells, staining fractions, and positive staining area, the mixed effects models were used to assess the effects of age, diarrhea status, and their interaction. For X409 optical density, measurements obtained at five predefined positions along the crypt axis were analyzed using an extended mixed-effects model including age, diarrhea status, crypt depth, and all two- and three-way interactions as fixed effects, with piglet ID and crypt number included as random effects.

Model assumptions were assessed by inspection of model residuals using residual-versus-fitted and normal Q-Q plots. Estimated marginal means were calculated using the emmeans package (Lenth & Piaskowski, 2026). Where interactions involving diarrhea status were statistically significant, pairwise comparisons between cases and controls were performed within each age group. P-values for pairwise comparisons were adjusted for multiple testing using the Benjamini-Hochberg procedure. Statistical significance was defined as p < 0.05.

3. Results

3.1. Study population and sample overview

A total of 30 piglets were included across five age groups, with equal representation of cases and controls. Jejunal and colonic samples were collected from all animals, and ten crypts per segment and staining were evaluated. Quantitative analyses were performed at crypt level, and a summary of data is presented in Table 2, Table 3.

Table 2.

Summary statistics of quantitative parameters (goblet cell numbers and area percentages) sorted by age-group and diarrhea status (group mean ± standard deviation, n = 3 pigs/30 crypts for each group).

	4 days		14 days		25 days		49 days		67 days
	case	control	case	control	case	control	case	control	case	control
JEJUNUM
Goblet cells
basal compartment	15.9 ± 4.9	14.6 ± 3.9	17.9 ± 5.5	16.4 ± 4.6	16.7 ± 6.7	16.7 ± 7.1	26.2 ± 7.3	21.6 ± 7.1	24.6 ± 18.6	38.4 ± 11.0
luminal compartment	16.7 ± 2.9	8.7 ± 3.3	12.7 ± 5.7	8.8 ± 4.8	8.1 ± 4.4	5.4 ± 3.5	10.4 ± 4.4	10.2 ± 5.8	7.7 ± 6.2	15.2 ± 5.0
total	23.1 ± 6.0	23.2 ± 5.4	30.6 ± 9.1	25.2 ± 8.2	24.7 ± 10.1	22.1 ± 9.6	36.6 ± 8.8	31.8 ± 10.5	32.4 ± 23.9	53.6 ± 13.5
MUC2
basal compartment	15. 4 ± 5.0	14.5 ± 3.9	17.9 ± 5.4	16.2 ± 4.7	15.0 ± 8.2	16.4 ± 7.1	26.2 ± 7.3	21.6 ± 7.1	24.6 ± 18.7	38.0 ± 11.2
luminal compartment	6.8 ± 2.9	8.7 ± 3.3	12.4 ± 5.2	8.4 ± 4.9	6.9 ± 5.0	4.7 ± 3.7	10.4 ± 4.3	10.0 ± 5.8	7.7 ± 6.3	15.2 ± 5.0
total	22.2 ± 5.9	23.2 ± 5.5	30.3 ± 8.8	24.6 ± 8.5	21.8 ± 12.3	21.1 ± 9.9	36.5 ± 8.8	31.7 ± 10.5	32.3 ± 24.0	53.2 ± 13.7
% positive area	24.6 ± 7.6	24.6 ± 7.6	36.7 ± 9.9	43.9 ± 21.9	19.8 ± 10.2	14.6 ± 5.6	20.0 ± 6.4	16.9 ± 4.6	10.6 ± 7.5	24.2 ± 7.2
X409
basal compartment	9.5 ± 4.9	11.1 ± 3.9	12.3 ± 4.0	10.5 ± 3.6	15.0 ± 6.4	13.9 ± 5.4	27.0 ± 8.0	18.7 ± 8.1	22.4 ± 17.6	37.6 ± 8.4
luminal compartment	1.1 ± 1.3	2.2 ± 2.4	1.6 ± 2.0	0.5 ± 0.8	3.2 ± 4.3	1.6 ± 1.9	6.2 ± 3.8	2.4 ± 1.9	2.0 ± 2.1	9.9 ± 5.8
total	10.5 ± 5.7	13.3 ± 5.5	13.9 ± 5.2	10.9 ± 3.8	18.2 ± 8.4	15.5 ± 5.6	33.2 ± 10.3	21.2 ± 7.6	24.4 ± 19.0	47.5 ± 11.3
% positive area	3.9 ± 3.1	6.4 ± 4.8	4.9 ± 2.4	2.8 ± 1.7	12.2 ± 7.1	6.0 ± 3.2	11.8 ± 3.8	7.3 ± 4.2	8.8 ± 6.8	22.2 ± 7.8
COLON
Goblet cells
basal compartment	22.8 ± 9.1	16.0 ± 4.4	18.0 ±12.7	25.6 ± 10.5	25.8 ± 8.1	25.3 ± 4.2	46.0 ± 14.0	31.7 ± 9.2	43.9 ± 16.3	49.1 ± 10.2
luminal compartment	12.4 ± 4.2	12.0 ± 3.6	12.0 ± 6.7	16.2 ± 6.0	13.6 ± 5.8	16.0 ± 3.4	24.1 ± 6.5	20.3 ± 7.0	24.6 ± 7.6	31.3 ± 6.6
total	35.2 ± 11.8	28.0 ± 6.1	30.0 ± 19.2	41.8 ± 14.1	39.4 ± 11.6	41.3 ± 5.6	70.2 ± 17.7	52.0 ± 14.1	68.5 ± 20.3	80.4 ± 11.6
MUC2
basal compartment	20.7 ± 9.8	12.9 ± 4.6	17.4 ± 12.7	23.7 ± 11.5	23.9 ± 9.8	15.8 ± 9.7	26.6 ± 11.9	28.4 ± 10.7	37.1 ± 14.7	45.6 ± 10.7
luminal compartment	10.4 ± 4.9	8.5 ± 3.3	11.1 ± 6.4	14.4 ± 6.0	12.6 ± 5.8	12.7 ± 5.0	21.6 ± 6.9	17.6 ± 7.5	23.2 ± 7.8	30.3 ± 7.1
total	31.1 ± 12.8	21.4 ± 5.5	28.5 ± 18.5	38.1 ± 15.2	36.4 ± 13.5	28.5 ± 13.3	48.2 ± 17.2	46.0 ± 16.2	60.2 ± 19.5	75.9 ± 13.1
% positive area	58.5 ± 13.1	24.3 ± 14.0	42.9 ± 23.9	58.9 ± 9.5	38.9 ± 17.3	43.4 ± 22.0	17.9 ± 10.8	37.5 ± 21.9	47.0 ± 16.2	39.4 ± 16.7
X409
basal compartment	22.0 ± 6.2	18.0 ± 4.4	14.6 ± 11.8	23.7 ± 9.4	23.3 ± 5.5	27.8 ± 6.6	49.1 ± 13.8	32.4 ± 9.1	47.7 ± 20.3	55.1 ± 13.1
luminal compartment	11.1 ± 3.6	11.7 ± 3.3	9.4 ± 7.5	15.2 ± 4.5	13.0 ± 4.3	14.0 ± 4.4	26.0 ± 6.9	20.5 ± 7.1	27.9 ± 10.5	30.3 ± 6.7
total	33.1 ± 8.2	29.6 ± 7.2	24.1 ± 18.3	39.0 ± 12.4	36.3 ± 8.0	41.8 ± 9.4	75.1 ± 18.1	52.9 ± 13.2	75.7 ± 26.1	85.4 ± 15.7
% positive area	43.7 ± 14.4	43.2 ± 14.5	20.4 ± 15.0	33.0 ± 14.3	25.9 ± 6.4	47.2 ± 18.7	39.4 ± 8.2	43.7 ± 11.8	58.1 ± 11.3	55.2 ± 10.1

Table 3.

Summary statistics of X409 staining intensity (mean optical density) in goblet cells (group mean ± standard deviation, n = 3 pigs/30 crypts for each group).

	4 days		14 days		25 days		49 days		67 days
	case	control	case	control	case	control	case	control	case	control
JEJUNUM
Optical density
< 20% from lumen	0.14 ± 0.06	0.18 ± 0.10	0.17 ± 0.07	0.14 ± 0.03	0.26 ± 0.24	0.19 ± 0.08	0.44 ± 0.23	0.22 ± 0.09	0.20 ± 0.20	0.40 ± 0.29
20-40% from lumen	0.16 ± 0.09	0.24 ± 0.12	0.17 ± 0.06	0.12 ± 0.06	0.24 ± 0.18	0.20 ± 0.10	0.53 ± 0.27	0.30 ± 0.26	0.32 ± 0.37	0.62 ± 0.35
40-60% from lumen	0.18 ± 0.10	0.28 ± 0.17	0.26 ± 0.13	0.14 ± 0.08	0.36 ± 0.27	0.26 ± 0.10	0.71 ± 0.30	0.58 ± 0.32	0.45 ± 0.37	1.10 ± 0.57
60-80% from lumen	0.21 ± 0.10	0.41 ± 0.24	0.32 ± 0.12	0.23 ± 0.12	0.50 ± 0.27	0.53 ± 0.23	0.89 ± 0.34	0.79 ± 0.29	0.82 ± 0.29	1.27 ± 0.44
>80% from lumen	0.77 ± 0.33	0.91 ± 0.35	0.67 ± 0.14	0.55 ± 0.13	0.93 ± 0.35	0.78 ± 0.26	1.19 ± 0.28	1.07 ± 0.36	1.37 ± 0.37	1.53 ± 0.57
Optical density score
< 20% from lumen	0.16 ± 0.37	0.42 ± 0.58	0.31 ± 0.47	0.06 ± 0.25	0.78 ± 1.01	0.44 ± 0.58	1.63 ± 1.06	0.63 ± 0.62	0.50 ± 0.85	1.39 ± 1.12
20-40% from lumen	0.27 ± 0.45	0.71 ± 0.60	0.33 ± 0.48	0.09 ± 0.29	0.69 ± 0.93	0.43 ± 0.66	1.93 ± 0.75	0.93 ± 1.12	0.80 ± 1.28	2.00 ± 1.08
40-60% from lumen	0.31 ± 0.54	0.90 ± 0.92	0.79 ± 0.82	0.15 ± 0.37	1.23 ± 1.24	0.70 ± 0.61	2.53 ± 0.78	1.97 ± 1.12	1.36 ± 1.18	2.57 ± 0.90
60-80% from lumen	0.56 ± 0.64	1.57 ± 0.92	1.07 ± 0.78	0.63 ± 0.67	1.76 ± 1.09	2.00 ± 0.79	2.73 ± 0.58	2.60 ± 0.67	2.73 ± 0.63	2.97 ± 0.18
>80% from lumen	2.43 ± 0.82	2.87 ± 0.43	2.67 ± 0.48	2.27 ± 0.74	2.87 ± 0.35	2.63 ± 0.56	2.93 ± 0.25	2.87 ± 0.35	3.00 ± 0.00	2.97 ± 0.18
COLON
Optical density
< 20% from lumen	0.37 ± 0.09	0.39 ± 0.18	0.29 ± 0.09	0.36 ± 0.09	0.37 ± 0.12	0.39 ± 0.22	0.69 ± 0.20	0.75 ± 0.24	1.21 ± 0.45	1.22 ± 0.36
20-40% from lumen	0.54 ± 0.17	0.60 ± 0.27	0.35 ± 0.10	0.46 ± 0.10	0.48 ± 0.11	0.53 ± 0.23	0.72 ± 0.15	0.79 ± 0.22	1.27 ± 0.42	1.51 ± 0.33
40-60% from lumen	0.67 ± 0.16	0.72 ± 0.29	0.43 ± 0.11	0.50 ± 0.11	0.50 ± 0.12	0.60 ± 0.23	0.72 ± 0.15	0.83 ± 0.23	1.27 ± 0.40	1.46 ± 0.33
60-80% from lumen	0.72 ± 0.21	0.74 ± 0.32	0.46 ± 0.12	0.55 ± 0.15	0.47 ± 0.13	0.63 ± 0.22	0.72 ± 0.16	0.86 ± 0.21	1.20 ± 0.48	1.28 ± 0.32
>80% from lumen	0.92 ± 0.29	0.81 ± 0.28	0.49 ± 0.13	0.49 ± 0.16	0.47 ± 0.14	0.71 ± 0.26	0.76 ± 0.21	0.84 ± 0.31	1.13 ± 0.63	1.06 ± 0.27
Optical density score
< 20% from lumen	1.32 ± 0.55	1.45 ± 0.83	1.04 ± 0.56	1.33 ± 0.55	1.32 ± 0.61	1.48 ± 1.06	2.60 ± 0.56	2.67 ± 0.61	2.87 ± 0.35	3.00 ± 0.00
20-40% from lumen	2.03 ± 0.67	2.17 ± 0.70	1.20 ± 0.65	1.83 ± 0.59	1.87 ± 0.63	2.07 ± 0.94	2.70 ± 0.47	2.83 ± 0.38	2.97 ± 0.18	3.00 ± 0.00
40-60% from lumen	2.67 ± 0.48	2.40 ± 0.56	1.68 ± 0.61	2.03 ± 0.72	2.03 ± 0.61	2.27 ± 0.87	2.70 ± 0.53	2.87 ± 0.43	3.00 ± 0.00	3.00 ± 0.00
60-80% from lumen	2.63 ± 0.61	2.40 ± 0.67	1.70 ± 0.61	2.17 ± 0.75	1.93 ± 0.69	2.30 ± 0.92	2.77 ± 0.50	2.87 ± 0.43	2.83 ± 0.46	3.00 ± 0.00
>80% from lumen	2.83 ± 0.38	2.70 ± 0.47	1.95 ± 0.65	1.90 ± 0.80	1.83 ± 0.75	2.53 ± 0.68	2.77 ± 0.50	2.70 ± 0.60	2.43 ± 0.94	2.97 ± 0.18

3.2. MUC5AC expression

MUC5AC expression was assessed qualitatively by conventional brightfield microscopy in jejunal and colonic tissue sections from all age groups. No consistent de novo expression of MUC5AC was detected in either of the segments across the study population. Sporadic, weak MUC5AC staining was observed in a small number of colonic goblet cells in two piglets from the 67-day age group, comprising one case and one control animal. Due to the limited extent and low intensity of this staining, MUC5AC expression was not evaluated further.

3.3. Age-related changes in goblet cells and mucus characteristics

Significant age-related changes were observed in crypt-villus unit depth and in the abundance and staining characteristics of goblet cells in both the jejunum and colon (Table 4, Table 5, Fig. 4, Fig. 5). Representative histological images illustrating the age-related development of MUC2 and X409 staining in both intestinal segments are shown in Fig. 4.

Table 4.

Results of mixed-effects models evaluating age-, diarrhea status-, and age x diarrhea-associated differences in goblet cell and mucus characteristics in jejunum. Numbers are p-values of the fixed effects of age, diarrhea status, and their interaction. Piglet identity was included as a random effect to account for repeated measurements from the same animal.

JEJUNUM	Age	Diarrhea status	Age x diarrhea status
Crypt villus unit depth	<0.001*	0.741	0.355
Goblet cells
Basal compartment	0.008*	0.662	0.307
Luminal compartment	0.089	0.717	0.049*
Total	0.018*	0.664	0.189
MUC2
Basal compartment	0.008*	0.600	0.355
Luminal compartment	0.056	0.633	0.069
Total	0.015*	0.594	0.234
Basal positive fraction	0.512	0.267	0.735
Luminal positive fraction	0.365	0.461	0.834
Total positive fraction	0.404	0.311	0.822
% positive area	0.002*	0.476	0.427
X409
Basal compartment	<0.001*	0.674	0.111
Luminal compartment	0.013*	0.569	0.005*
Total	<0.001*	0.603	0.024*
Basal positive fraction	0.023*	0.250	0.677
Luminal positive fraction	0.209	0.422	0.263
Total positive fraction	0.013*	0.157	0.611
% positive area	0.002*	0.700	0.008*

Significant p-values (p < 0.05) are indicated with *.

Table 5.

Results of mixed-effects models evaluating age-, diarrhea status-, and age x diarrhea-associated differences in goblet cell and mucus characteristics in colon. Numbers are p-values of the fixed effects of age, diarrhea status, and their interaction. Piglet identity was included as a random effect to account for repeated measurements from the same animal.

COLON	Age	Diarrhea status	Age x diarrhea status
Crypt villus unit depth	<0.001*	0.859	0.004*
Goblet cells
Basal compartment	<0.001*	0.572	0.190
Luminal compartment	<0.001*	0.220	0.192
Total	<0.001*	0.995	0.156
MUC2
Basal compartment	<0.001*	0.959	0.280
Luminal compartment	<0.001*	0.535	0.191
Total	<0.001*	0.800	0.295
Basal positive fraction	0.252	0.984	0.079
Luminal positive fraction	0.069	0.275	0.683
Total positive fraction	0.308	0.806	0.138
% positive area	0.249	0.959	0.078
X409
Basal compartment	<0.001*	0.995	0.125
Luminal compartment	<0.001*	0.523	0.154
Total	<0.001*	0.845	0.126
Basal positive fraction	0.200	0.345	0.449
Luminal positive fraction	0.196	0.869	0.283
Total positive fraction	0.223	0.468	0.350
% positive area	<0.014*	0.167	0.511

Significant p-values (p < 0.05) are indicated with *.

Fig. 4.

Age-related changes in MUC2 and X409 staining in jejunum and colon. Representative histological images showing MUC2 immunohistochemical staining and X409-labeling in the jejunum and colon at 4, 14, 25, 49, and 67 days of age. MUC2- and X409-positive staining is shown in brown with hematoxylin counterstaining. Jejunal images at 4 and 14 days are shown at lower magnification to accommodate the larger villus-crypt units, whereas later images are shown at higher magnification to better illustrate the changing staining pattern along the villus-crypt axis. All scale bars =100µm.

Fig. 5.

Age-related changes in goblet cell abundance and staining characteristics in jejunum and colon. A-B) Total goblet cell abundance per crypt/crypt-villus unit across age groups in the jejunum (A) and the colon (B). Boxplots show the distribution of crypt-level observations, with individual crypts displayed as points; different point shapes denote individual piglets within each group. C-D) Mean fraction of MUC2- and X409-positive goblet cells across age groups in the jejunum (C) and colon (D). Cases are shown in orange and non-diarrheic controls in blue.

Crypt-villus unit depth increased significantly with age in both jejunum and colon. In the jejunum, goblet cell numbers increased with age in the basal compartment and overall, whereas the age effect in the luminal compartment did not reach statistical significance (Fig. 5A). In the colon, goblet cell numbers increased with age in both compartments and overall (Fig. 5B).

In jejunum, the number of MUC2-positive goblet cells increased significantly with age in the basal compartment and overall, accompanied by an age-related increase in the proportion of MUC2-positive area. The fraction of MUC2-positive goblet cells remained consistently high across age groups (Fig. 5C). In colon, MUC2-positive goblet cell numbers increased with age in both compartments and overall, whereas the proportion of MUC2-positive area did not differ significantly between age groups. Similarly, the fraction of MUC2-positive goblet cells showed only limited variation with age (Fig. 5D).

X409-positive goblet cell numbers increased significantly with age in both intestinal segments, represented both in the basal and luminal compartments and when assessed across the entire villus-crypt unit (Table 4, Table 5). The proportion of X409-positive area also increased with age in both intestinal segments. In contrast to MUC2, the fraction of X409-positive goblet cells changed significantly with age in the jejunum, whereas no significant age-related differences were observed in the colon (Fig. 5C-D). The age-related changes in X409-labeling are also evident in Fig. 4.

3.4. Diarrhea-associated changes in goblet cells and mucus characteristics

The outcome of the mixed-effects model showed no overall effect of diarrhea status for any of the evaluated parameters. However, significant age x diarrhea status interactions were observed for four parameters in the jejunum and for crypt depth in the colon (Table 4, Table 5).

In the jejunum, post-hoc comparisons showed that case animals had significantly fewer goblet cells in the luminal compartment at day 67 compared with controls (estimated difference: -7.5 cells/crypt, p=0.0099; Fig. 6). At the same age, cases also had fewer X409-positive goblet cells overall (-23.1 cells/crypt, p=0.0034) and in the luminal compartment (-7.9 cells/crypt, p=0.0007), together with a smaller X409-positive area (-13.4 percentage points, p=0.0014). No significant differences between cases and controls were identified at the remaining ages.

Fig. 6.

Representative MUC2 immunohistochemical staining illustrating disease-associated alterations in goblet cell populations in the jejunum of 67 day old piglets. A) Jejunal villus-crypt units from 67-day-old piglets showing a normal distribution of MUC2-positive goblet cells in a non-diarrheic control animal and B) marked goblet cell depletion in a piglet with diarrhea, showing near-complete loss of identifiable MUC2-positive goblet cells along the crypt-villus unit Quantified villus-crypt units are outlined by a yellow border, and counted cells are marked with blue circles.

In the colon, crypt depth differed between cases and controls in an age-dependent manner. Compared with controls, cases had significantly shorter crypts at day 14 (-123.2 µm, p=0.0138) and at day 67 (-106.3 µm, p=0.0305), whereas crypts were significantly longer in cases at day 49 compared to controls (+119.4 µm, p=0.0165).

3.5. X409 staining intensity along the crypt axis

X409 staining intensity varied significantly with age and according to relative position along the crypt-villus axis in both jejunum and colon. The staining intensity was strongly associated with depth/goblet cell position within the crypt/crypt-villus unit, showing the strongest staining in goblet cells positioned in the crypt base, with progressively fainter staining towards the lumen, especially in the jejunal segments (Table 6, Fig. 7). Significant age x depth interactions demonstrated that the distribution of staining intensity along the axis changed during postnatal development (Table 6, Fig. 4, Fig. 7). In addition, significant interactions involving diarrhea status were identified for staining intensity in both intestinal segments.

Table 6.

Results of mixed-effects model evaluating age-, diarrhea status, crypt depth-, and interaction-associated differences in X409 staining intensity in the jejunum and colon. P-values represent the fixed effects and interactions included in the mixed-effects models. Piglet identity and crypt no. were included as random effects to account for repeated measurements from the same individuals.

Segment	Age	Diarrhea status	Crypt depth	Age x diarrhea	Age x depth	Diarrhea x depth	Age x diarrhea x depth
Jejunum	<0.001*	0.671	<0.001*	0.141	<0.001*	<0.001*	<0.001*
Colon	<0.001*	0.301	<0.001*	0.993	<0.001*	0.014*	<0.001*

Significant p-values (p < 0.05) are indicated with * and bold.

Fig. 7.

X409 staining intensity characteristics in goblet cells of jejunum and colon. A) Distribution of X409 staining intensity of goblet cells along the villus-crypt axis in jejunum (left) and colon (right). Goblet cell position is expressed as relative distance from the gut lumen. Colors indicate age groups (blue shades: pre-weaning; orange shades: post-weaning) and line types indicate disease status. B-C) Representative X409 immunolabeling illustrating differences in goblet cell staining intensity along the crypt axis in a 67-day-old piglet. B) Jejunum, scalebar:50µm. Staining intensity gradually decreases toward the gut lumen. C) Colon, scalebar: 100µm. Goblet cells show a more uniform staining intensity along the crypt axis. Yellow outlines indicate representative crypts and villus-crypt units used for axial intensity assessment.

In the jejunum, the age x diarrhea x depth interaction was significant (Table 6). Post-hoc comparisons showed no significant differences in X409 staining intensity between cases and controls at most age-depth combinations. However, at day 67, staining intensity was significantly lower in cases than in controls for the middle segment of the crypts, namely at 40-60% (estimated difference: -0.63 OD, p=0.0002) and 60-80% depth intervals (-0.42 OD, p=0.0079), whereas no significant differences were identified at the remaining intervals.

In the colon, the age x diarrhea status x depth interaction was also significant. However, post-hoc comparisons did not identify significant differences in staining intensity between cases and controls at any individual age-depth combination after adjustment for multiple comparisons.

4. Discussion

This study provides a detailed characterization of age-related remodeling of the intestinal mucus barrier during early life in pigs, demonstrating marked age-dependent changes in goblet cell populations and mucus characteristics across the weaning period. By integrating quantitative assessment of goblet cell abundance with MUC2 immunohistochemistry and in situ evaluation of mucus-associated glycosylation using X409, the present study demonstrates that maturation of the intestinal barrier involves both quantitative and qualitative changes. Both total goblet cell numbers and MUC2-positive cells increased substantially with age, although the most pronounced shifts occurred across the weaning phase. While dietary transition from milk to solid feed is likely a major contributing factor, previous studies suggest that feed intake and composition alone does not fully explain the observed alterations (Dunsford et al., 1991; Engelsmann et al., 2023). Instead, multiple factors are likely involved, including weaning-associated stressors, age-dependent gut maturation, environmental changes, and the withdrawal of protective interventions such as medicinal zinc oxide supplementation (European Commission, 2017; Tang et al., 2022, 2024). Together, these likely contributed to an amplification of the age-related changes observed in the present study. However, it remains uncertain to what extent the timing of weaning influences this trajectory. For example, it is unclear whether similar changes would be induced by earlier weaning (e.g. at day 21), or whether delaying weaning to a stage where the intestinal barrier is more mature would reduce the magnitude of these alterations. A more developed mucosal barrier at the time of weaning could potentially decrease the need for such pronounced adaptation and thereby mitigate the susceptibility to post-weaning diarrhea. Nevertheless, the consistency of age-related trends across groups suggests that maturation of the intestinal barrier is, at least in part, an intrinsic developmental process.

When comparing pigs with a history of diarrhea to animals without, group differences were most evident after weaning, particularly at day 67. In the oldest age group, significantly lower goblet cells numbers were observed in case animals. It is unclear whether these changes reflect recent intestinal insults or represent later consequences of diarrhea occurring after weaning. However, previous authors have stated that acute intestinal infection is associated with goblet cell hyperplasia and increased MUC2 production, while more chronic inflammatory states are characterized by goblet cell depletion (Sheng & Hasnain, 2022; Xia, Zhong, Wu, et al., 2022), a pattern that aligns well with the findings in the present study.

An additional contribution of the present study was the use of X409 to assess mucus-associated glycosylation in situ. In contrast to conventional approaches that primarily assess physiological properties, composition, and structural profiling of gastrointestinal mucus in ex vivo samples (Barmpatsalou et al., 2021), in situ labeling of tissue sections provides spatial information on mucus distribution and staining intensity within intact crypt architecture. While lectin-based staining has previously been used for histological visualization of mucins (Desantis et al., 2019; Hedemann et al., 2007; Venkatakrishnan et al., 2017), lectins generally recognize terminal carbohydrate structures of both mucin and non-mucin origin, lowering the staining specificity. X409 was previously reported as a selective mucin binding probe targeting a clustered O-glycan motif of 3-4 residues, found only in selected mucins, including porcine MUC2. Additionally, X409 shows higher binding affinity for nascent (mature) mucin glycans as opposed to partly degraded or trimmed down glycans (e.g. immature states), which complement the information provided by traditional immunohistochemistry (Jaroentomeechai et al., 2025). In the present study, X409 revealed patterns that both complemented and extended the findings obtained with MUC2 immunohistochemistry. Whereas MUC2 labeling reflected the abundance of MUC2-containing goblet cells, X409 provided additional information related to mucus conformation or glycosylation within these cells. In the jejunum, a consistent reduction in the fraction of X409-positive goblet cells was observed in case animals compared to controls, and changes in the staining intensity pattern were associated with diarrhea in 67-day old piglets, representing changes not evident when solely investigating the MUC2-stainings. This suggests that the observed differences are unlikely to reflect mucus depletion alone but may instead indicate alterations in mucin glycosylation, conformation, or the secretion of less mature mucins under conditions of increased secretory demand. The presence of immature or incompletely glycosylated mucus could result in a less organized protective layer, reducing the amount of binding sites for beneficial bacteria and thereby impairing several aspects of the intestinal barrier, i.e.: decreased microbial-mediated protection, reduced production of short-chain fatty acids to nourish the epithelium, and a less cohesive mucus layer that may allow easier access for pathogenic microorganisms (Chatterjee et al., 2023; Fekete & Buret, 2023; Quintana-Hayashi et al., 2018). In contrast, colonic X409-labeling largely followed the trends observed for total goblet cell numbers and MUC2-positive cells indicating that variation in this segment is primarily driven by changes in goblet cell abundance rather than mucus glycosylation per se. Taken together, these findings suggest that postnatal development and early life diarrhea in pigs involves not only quantitative changes in goblet cell populations, but also qualitative alterations in mucus properties, particularly in the small intestine, and more studies are needed in the future, in order to further evaluate the mechanisms underlying these changes. The absence of de novo MUC5AC expression – previously reported in several enteric infections (Hasnain et al., 2011; Lin et al., 2023; Quintana-Hayashi et al., 2015) - further indicates that these alterations are unlikely to be driven solely by classical pathogen-induced mucin shifts but rather reflect broader changes in barrier function. Collectively, this points to a phase of substantial functional remodeling of the intestinal barrier during the weaning period, a phase in which subtle impairments in mucus organization may have disproportionate effects on gut homeostasis.

In this context, post-weaning diarrhea should once again be highlighted as a multifactorial condition rather than a disease of single microbiologic etiology. Although highly prevalent in pig production, recent findings from a Danish study indicate that only approximately 25% of diarrhea-affected animals required antibiotic treatment to recover, suggesting that the majority of cases are self-limiting (Morsing et al., 2025). This supports the notion that many diarrheal episodes reflect a transient imbalance between barrier function and environmental challenges during a critical developmental window. The present study demonstrates that, while all animals undergo substantial age-related restructuring of the intestinal barrier, some individuals may be less capable of maintaining appropriate goblet cell and mucus responses during this transition. This raises the question of whether current weaning practices, typically occurring between 21 and 28 days of age, impose an abrupt challenge on an incompletely matured intestinal barrier. Delaying weaning, thereby allowing further development of mucosal defenses, could potentially result in a more gradual adaption and a lower susceptibility to diarrhea. In line with this, production systems with later and more gradual weaning – such as those employed in organic or free-range farming systems – may provide a useful framework for future studies aimed at disentangling the relative contributions of management practices and intrinsic developmental processes to post-weaning intestinal health.

5. Conclusion

In conclusion, the weaning period in pigs is characterized by profound remodeling of goblet cell populations and mucus properties, underscoring the dynamic nature of the intestinal barrier during early life. Notably, goblet cell responses differed with age, with case animals showing reduced goblet cell numbers, and altered staining properties at day 67 compared to controls. The combined quantitative and qualitative alterations identified in this study suggest that compromised mucus integrity may play a role in the pathogenesis of early life diarrhea, including post-weaning diarrhea. These findings provide new insight into the mechanisms underlying this common condition and emphasize the potential need to align management practices with the developmental capacity of the intestinal barrier.

Study limitations

The present study has several limitations that should be considered when interpreting the findings. First of all, the limited sample size reduced the statistical power and, consequently, our ability to identify subtle differences as statistically significant. Readers of this study should also appreciate that the substantial age-associated development of the mucus barrier occurring within the same timeframe as post-weaning diarrhea also complicates disentangling the relative contributions of age and disease to the observed histological findings. Furthermore, allocation of pigs into case and control groups relied primarily on fecal scoring, which may have introduced a risk of misclassification, as described in a recently published study (Becker et al., 2026). Additionally, the study was conducted in piglets from a single herd, potentially limiting the generalizability of the findings to other production systems with different management practices or pathogen pressures. While X409 provides novel insight into mucus glycosylation patterns, it represents an indirect measure of mucus quality and does not allow for detailed characterization of specific glycan structures. Furthermore, the present study did not encompass all components of the mucus barrier. Although goblet cell abundance and intracellular mucus content were evaluated, the secreted mucus layer covering the epithelial surface, which constitutes an important component of the protective barrier, could not be reliably assessed. Complete preservation of this layer could not be ensured using the applied histological techniques, precluding its quantitative evaluation. Moreover, the evaluation of staining intensity in the field of histopathology is associated with a certain degree of inaccuracy due to risk of uneven staining distribution and differences in slide thickness between samples. Finally, the cross-sectional design precludes assessment of temporal changes within individual animals and limits the ability to infer causal relationships from the observed associations.

Ethics approval

All animal procedures were conducted in accordance with national legislation on animal experimentation and were approved by the Danish Animal Experimentation Inspectorate (license number: 2022-15-0201-01324).

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the authors used ChatGPT in order to perform language editing and spell check on the drafted manuscript. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Financial support statement

This study was conducted as part of the PIG-PARADIGM project, funded by the Novo Nordisk Foundation (Grant no. NNFSA210073688).

CRediT authorship contribution statement

C.B. Becker: Writing – original draft, Visualization, Validation, Software, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. J. Elzinga: Writing – review & editing, Visualization, Resources, Methodology, Conceptualization. M.K. Morsing: Writing – review & editing, Resources, Methodology, Investigation. C.B. Jørgensen: Writing – review & editing, Supervision, Resources, Methodology, Data curation. M. Fredholm: Writing – review & editing, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. M.S. Hansen: Writing – review & editing, Visualization, Supervision, Investigation. S.S. Nielsen: Writing – review & editing, Supervision, Formal analysis, Data curation. J.P. Nielsen: Writing – review & editing, Resources, Investigation, Funding acquisition. H.E. Jensen: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The datasets generated and analyzed during the current study are available in the Zenodo repository (DOI: 10.5281/zenodo.20610165). Digital slide images, QuPath annotation files, and the machine-learning model developed for quantitative image analysis are available from the corresponding author upon reasonable request.