Association Between Zinc Status and Tight Junction Integrity and Proinflammatory Cytokine Expression in Crohn's Disease

ABSTRACT

Aims

Zinc deficiency is common in patients with Crohn's disease (CD) and has been associated with adverse clinical outcomes. However, the relationship between zinc status and intestinal barrier integrity remains unclear. We investigated the association between serum zinc levels and the expression of tight junction (TJ) proteins and proinflammatory cytokines using paired intestinal biopsies from patients with CD.

Methods and Results

This retrospective observational study included patients with CD who underwent colonoscopic biopsies under both low zinc (< 80 μg/dL) and normal zinc (≥ 80 μg/dL) conditions between March 2017 and December 2023. Immunohistochemical staining was performed to assess TJ proteins (E‐cadherin, zonula occludens‐1 [ZO‐1], occludin, and claudin‐1) and proinflammatory cytokines (tumor necrosis factor‐α, interleukin [IL]‐17, IL‐12, and IL‐23). Twelve patients were included in the main analysis, while eight who did not receive treatment intensification were included in the sub‐analysis. In the main analysis, E‐cadherin, ZO‐1, and claudin‐1 expression levels were significantly lower during the low zinc state, whereas IL‐17 expression was significantly higher. Similar trends were observed in the sub‐analysis. Additionally, the CD Activity Index was significantly higher, and hemoglobin and albumin levels were significantly lower during the low zinc state.

Conclusion

Low zinc status was associated with altered expression of TJ proteins and IL‐17 in the intestinal mucosa of patients with CD. These findings suggest that zinc status may be linked to epithelial barrier integrity and mucosal immune responses in CD. Although further validation is required, zinc supplementation may represent a simple and accessible adjunct to current therapies.

1. Introduction

Inflammatory bowel disease (IBD), comprising Crohn's disease (CD) and ulcerative colitis, is a chronic inflammatory disorder of the gastrointestinal tract characterized by repeated cycles of relapse and remission [1]. Within the IBD spectrum, CD is distinguished by discontinuous, transmural inflammation that can affect any part of the gastrointestinal tract; it is frequently complicated by strictures, fistulae, and the need for surgical intervention [2]. Despite advances in medical therapy, including biologics and small‐molecule agents [3, 4], a substantial proportion of patients experience disease progression. This persistent challenge highlights the need for a better understanding of modifiable factors that influence intestinal inflammation and mucosal integrity.

Malnutrition and micronutrient deficiencies are common in patients with CD due to chronic inflammation, impaired absorption, and increased metabolic demand [5, 6, 7, 8]. Among these, zinc deficiency is particularly prevalent, affecting ~40% of patients [9]. Zinc is the second most abundant trace element in the human body after iron and is essential for the activity of numerous enzymes and transcription factors, playing critical roles in growth, taste perception, and immune function [10, 11].

Clinical studies have shown that zinc deficiency is associated with an increased risk of hospitalization, surgery, and disease‐related complications in patients with CD [9]. Furthermore, we have previously reported that zinc supplementation improves disease activity in patients with CD and hypozincemia [12].

Experimental studies have suggested potential mechanisms underlying these clinical observations. Zinc deficiency is linked to the increased expression of proinflammatory cytokines, such as interleukin (IL)‐17 and IL‐23, and the exacerbation of intestinal inflammation in animal models [13, 14]. In addition, zinc plays a critical role in maintaining intestinal barrier function. Tight junctions (TJs), composed of proteins such as claudins, occludin, and zonula occludens (ZO)‐1, regulate paracellular permeability, while E‐cadherin, a key component of adherens junctions, further contributes to epithelial integrity [15]. In vitro studies have demonstrated that zinc deficiency reduces the expression of these TJ proteins [16], suggesting that impaired barrier function increases the translocation of luminal antigens and bacteria.

However, despite these experimental findings, the relationship between zinc status, intestinal barrier integrity, and mucosal immune responses has not been adequately investigated in clinical samples from patients with CD. In particular, there is a lack of evidence based on paired intestinal biopsy specimens obtained under different zinc conditions within the same patients.

Consequently, in this study, we aimed to investigate the association between serum zinc levels and the expression of TJ proteins and proinflammatory cytokines in intestinal tissues from patients with CD using paired biopsy specimens.

2. Materials and Methods

2.1. Study Design

This retrospective observational study was conducted at the Department of Gastroenterology and Hepatology of our institution. We included patients with CD who underwent colonoscopy with intestinal biopsy on at least two occasions under different serum zinc conditions between March 2017 and December 2023. We aimed to evaluate the association between serum zinc levels and the expression of TJ proteins and proinflammatory cytokines using these paired intestinal biopsy specimens.

2.2. Description of Participants

The inclusion criteria were as follows:

1. A diagnosis of CD is based on clinical, endoscopic, and histological findings [17].

2. The availability of paired colonoscopic mucosal biopsy specimens obtained under both hypozincemic (serum zinc < 80 μg/dL) and normozincemic (serum zinc ≥ 80 μg/dL) conditions

Biopsy specimens were included if they were obtained within 28 days of the corresponding serum zinc measurement. The order of sampling was not restricted.

The exclusion criteria were as follows:

1. Intestinal resection was performed between the two biopsy time points.

2. A history of total colectomy or the presence of a stoma.

3. The absence of colonic biopsy specimens.

To assess the potential influence of treatment intensification, a sub‐analysis was performed excluding patients who underwent treatment escalation between the two biopsy time points. Treatment escalation was defined as the initiation or dose increase of immunomodulators, corticosteroids, or biologics.

2.3. Definitions

According to the criteria of the Japanese Society of Clinical Nutrition [10], serum zinc levels of 80–130 μg/dL are considered normal, 60–80 μg/dL indicate latent zinc deficiency, and < 60 μg/dL indicate zinc deficiency requiring treatment. In this study, hypozincemia was defined as a serum zinc concentration < 80 μg/dL, while concentrations ≥ 80 μg/dL were defined as normal.

2.4. Data Collection

At both the low and normal zinc time points, the following data were collected:

• Demographics and clinical characteristics: Sex, age, body mass index, smoking history, disease location, history of intestinal resection, and the presence of strictures, fistulas, or perianal lesions.

• Medications: Elemental diet, immunomodulators, corticosteroids, biologics, and zinc supplementation.

• Clinical disease activity: CD activity index (CDAI) [18].

• Endoscopic disease activity: Simple Endoscopic Score for CD (SES‐CD) [19].

• Biopsy site: To standardize sampling, biopsy specimens were primarily obtained from the rectum; if rectal biopsies were unavailable, specimens were collected from the colonic segment closest to the rectum.

• Laboratory data: Complete blood count, albumin, C‐reactive protein (CRP), and serum zinc levels.

2.5. Immunohistochemistry

Immunohistochemical staining was performed to evaluate the expression of TJ proteins (E‐cadherin, ZO‐1, occludin, and claudin‐1) and proinflammatory cytokines (tumor necrosis factor [TNF]‐α, IL‐17, IL‐12, and IL‐23). IL‐17 and IL‐23 were included because zinc deficiency has been shown to increase their expression in experimental colitis models, indicating activation of the IL‐23/T‐helper 17 (Th17) axis [13]. IL‐12 and TNF‐α were analyzed as they represent major therapeutic targets in IBD and are central to mucosal immune regulation.

The primary antibodies used are summarized in Table 1. The secondary antibody was Histofine Simple Stain MAX‐PO (MULTI) (Nichirei, Tokyo, Japan). Antigen retrieval was performed by heating sections in a pH −9.0 retrieval solution at 95°C for 20 min. Specimens were then incubated with the respective primary antibodies: E‐cadherin and claudin‐1 at room temperature for 60 min, and ZO‐1, occludin, IL‐17, IL‐12, and IL‐23 at 4°C overnight. For TNF‐α, sections were incubated overnight at 4°C, followed by amplification with a linker reagent at room temperature for 15 min. All sections were subsequently incubated with the secondary antibody at room temperature for 30 min.

TABLE 1.

Primary antibodies used for immunohistochemistry.

Target	Clone/type	Manufacturer	Catalog no	Dilution
E‐cadherin	NCH38/monoclonal	DAKO	M3612	1:100
ZO‐1	−/polyclonal	Abcam	ab96587	1:1000
Occludin	EPR20992/monoclonal	Abcam	ab216327	1:200
Claudin‐1	−/polyclonal	Abcam	ab15098	1:400
IL‐17	−/polyclonal	GeneTex	GTX133781	1:1000
IL‐12	−/polyclonal	GeneTex	GTX66715	1:200
IL‐23	−/polyclonal	Abcam	ab45420	1:1000
TNF‐α	RM1005/multiclonal	Abcam	ab307164	1:50

Abbreviations: IL, interleukin; TNF, tumor necrosis factor; ZO, zonula occludens.

Human tonsillar tissue was used as a positive control for immunostaining of proinflammatory cytokines (Figures S1–S4). Negative controls were prepared by omitting the primary antibody. For TNF‐α, preliminary optimization was performed at a dilution of 1:100, after which a dilution of 1:50 was used for the final staining protocol (Figure S1).

Digital image analysis for staining assessment was performed as previously described, with modifications [20, 21]. The regions of interest were defined as 0.04 mm² for TJ proteins and 0.0625 mm² for cytokines. Ten fields were randomly selected from each specimen, and the positive area was quantified for each field using WinROOF image analysis software (Mitani Co., Fukui, Japan). For quantitative analysis of immunohistochemical staining, a threshold for positive staining was predetermined for each marker using several representative stained slides before the formal analysis. The threshold was adjusted to distinguish positive staining from background staining. Once established, the same threshold was applied uniformly to all slides for each marker throughout the analysis, without further modification. The staining rate was calculated as the proportion of the positively stained area relative to the total area of the region of interest. The median value of the 10 fields was used as the representative value for each specimen.

2.6. Outcome Measures

The primary outcome was the change in the expression (staining rate) of TJ proteins and proinflammatory cytokines in biopsy specimens between the low and normal zinc states.

Secondary outcomes included changes in clinical activity indices, laboratory data, and endoscopic activity between the two states.

2.7. Statistical Analysis

All statistical analyses were performed using JMP Student Edition version 18.2.0 (SAS Institute, Cary, NC, USA). Graphs comparing low and normal zinc states were generated using GraphPad Prism version 10 (GraphPad Software Inc., San Diego, CA, USA). Comparisons between the two states were performed using the Wilcoxon signed‐rank test. Statistical significance was set at p < 0.05.

Additional exploratory Spearman's rank correlation analyses were performed using the changes (Δ; normal‐zinc state minus low‐zinc state) in TJ protein expression, serum zinc levels, CDAI, and proinflammatory cytokine expression. Correlations between CDAI and laboratory parameters (hemoglobin and albumin) were evaluated using the values obtained at each sampling time point.

3. Results

3.1. Patient Selection

A total of 15 patients with CD met the initial screening criteria (Figure 1). Of these, two patients were excluded because they had no colonic biopsy specimens available, and one patient was excluded because they underwent intestinal resection between the two biopsy time points. Consequently, 12 patients were included in the main analysis.

FIGURE 1.

Patient selection flow chart. CD, Crohn's disease.

From this cohort, four patients who underwent treatment intensification (initiation of immunomodulator therapy, n = 1; initiation of biologic therapy, n = 3) between the two biopsy time points were excluded, leaving eight patients for the sub‐analysis.

3.2. Patient Characteristics

The baseline characteristics of the 12 patients with CD are summarized in Table 2. Eight patients (66.7%) were male, and the median age at the time of the low zinc state was 40.5 years. Disease location was ileal in two patients, colonic in four patients, and ileocolonic in six patients. Regarding disease complications, two patients had a history of small bowel resection, three had strictures, one had a fistula, and six had perianal disease.

TABLE 2.

Patient characteristics.

Serum zinc level	< 80 μg/dL	≥ 80 μg/dL	p
Sex, male	8 (66.7)
Age, years	40.5 (23–53)
Body mass index	22.5 (17.0–28.3)	22.5 (16.8–29.6)	0.48
Current smoker	2 (16.7)
Disease location
L1 ileal	2 (16.7)
L2 colonic	4 (33.3)
L3 ileocolonic	6 (50.0)
History of intestinal resection	2 (16.7)
Stricture	3 (25.0)
Fistula	1 (8.3)
Perianal disease	6 (50.0)
Treatment
Enteral nutrition	5 (41.7)	4 (33.3)
Immunomodulator	7 (58.3)	9 (75.0)
Prednisolone	1 (8.3)	0 (0.0)
Anti TNF‐α	5 (41.7)	6 (50.0)
Anti IL‐12/23	3 (25.0)	3 (25.0)
Anti α4β7 integrin	1 (8.3)	1 (8.3)
Zinc supplement	2 (16.7)	4 (33.3)
Disease activity
CDAI	102 (17–280)	56.5 (12–298)	0.035
SES‐CD	5 (0–21)	4.5 (0–9)	0.24
Laboratory data
White blood cell (/μL)	5850 (3600–9400)	6000 (4000–8000)	0.24
Hemoglobin (g/dL)	12.9 (9.0–16.8)	13.8 (11.8–17.3)	< 0.01
Platelet (/μL)	30.1 (19.5–51.1)	26.4 (21.2–47.3)	0.15
Albumin (g/dL)	3.9 (2.6–4.6)	4.2 (3.7–5.0)	0.021
C‐reactive protein (mg/dL)	0.03 (0.0–5.1)	0.02 (0.0–0.7)	0.31
Zn (μg/dL)	73.7 (52.0–78.6)	90.4 (80.3–107.4)	< 0.01
Biopsy site
Cecum	0 (0)	1 (8.3)
Descending colon	1 (8.3)	1 (8.3)
Rectum	11 (91.7)	10 (83.3)
Biopsy interval (month)	12.5 (3–32)

Note: Values are presented as medians (ranges) or numbers (%). Comparisons between two states were performed using the Wilcoxon signed‐rank test.

Abbreviations: CDAI, Crohn's disease activity index; IL, interleukin; SES‐CD, simple endoscopic score for Crohn's disease; TNF, tumor necrosis factor.

As for treatment, the use of immunomodulators and biologics was comparable between the two zinc states, while zinc supplementation was more frequent during the normal zinc state.

Disease activity was higher during the low zinc state, with a significantly higher CDAI than in the normal zinc state (102 vs. 56.5, p = 0.035). In contrast, the SES‐CD did not differ significantly between the two zinc states (5 vs. 4.5, p = 0.24).

Hemoglobin and albumin levels were significantly lower during the low zinc state (12.9 vs. 13.8 g/dL, p < 0.01; 3.9 vs. 4.2 g/dL, p = 0.021), while CRP levels were comparable (0.03 vs. 0.02 mg/dL, p = 0.31).

Biopsy specimens during the low zinc state were obtained from the descending colon in one patient and from the rectum in 11 patients. During the normal zinc state, specimens were obtained from the cecum in one patient, the descending colon in one patient, and the rectum in 10 patients. The median interval between the two biopsy procedures was 12.5 months (range: 3–32 months).

3.3. Changes in TJ Protein Expression Between the Low and Normal Zinc States

Immunohistochemical analysis revealed that the expression of TJ proteins was generally lower during the low zinc state than the normal zinc state. Specifically, the expression levels of E‐cadherin, ZO‐1, and claudin‐1 were significantly reduced during the low zinc state (p = 0.026, p = 0.024, and p < 0.01, respectively), whereas occludin showed a trend toward lower expression (p = 0.069) (Figure 2).

FIGURE 2.

Changes in the staining rates of tight junction proteins in the main analysis (n = 12). Data are presented for E‐cadherin, ZO‐1, occludin, and claudin‐1. Paired comparisons were performed using the Wilcoxon signed‐rank test. Whole‐slide images were scanned at ×20 magnification. Scale bars = 50 μm. ZO, zonula occludens.

In the sub‐analysis excluding patients who underwent treatment intensification, similar trends were observed. E‐cadherin and claudin‐1 expression remained significantly lower during the low zinc state (p < 0.01 and p = 0.028, respectively), whereas the differences for ZO‐1 and occludin did not reach statistical significance (Figure 3).

FIGURE 3.

Changes in the staining rates of tight junction proteins in the sub‐analysis (n = 8). Data are presented for E‐cadherin, ZO‐1, occludin, and claudin‐1. Patients who underwent treatment intensification were excluded from this analysis. Paired comparisons were performed using the Wilcoxon signed‐rank test. ZO, zonula occludens.

3.4. Changes in Proinflammatory Cytokine Expression Between the Low and Normal Zinc States

Immunohistochemical analysis of cytokines showed that IL‐17 expression was significantly elevated during the low zinc state compared to the normal zinc state (p < 0.01) (Figure 4). In contrast, no significant differences were observed in the expression of TNF‐α, IL‐12, or IL‐23 (p = 0.58, p = 0.067, and p = 0.44, respectively).

FIGURE 4.

Changes in the staining rates of proinflammatory cytokines in the main analysis (n = 12). Data are presented for TNF‐α, IL‐17, IL‐12, and IL‐23. Paired comparisons were performed using the Wilcoxon signed‐rank test. Whole‐slide images were scanned at ×20 magnification. Scale bars = 50 μm. TNF, tumor necrosis factor; IL, interleukin.

In the sub‐analysis excluding patients who underwent treatment intensification, IL‐17 expression remained significantly higher during the low zinc state (p = 0.024), whereas the levels of TNF‐α, IL‐12, and IL‐23 showed no significant differences (Figure 5).

FIGURE 5.

Changes in the staining rates of proinflammatory cytokines in the sub‐analysis (n = 8). Data are presented for TNF‐α, IL‐17, IL‐12, and IL‐23. Patients who underwent treatment intensification were excluded from this analysis. Paired comparisons were performed using the Wilcoxon signed‐rank test. TNF, tumor necrosis factor; IL, interleukin.

Representative low‐magnification images of the entire biopsy specimens from the low‐ and normal‐zinc states for each TJ protein and cytokine are provided in Figures S5–S12.

3.5. Correlation Analyses

To further explore the relationships among changes in TJ protein expression, serum zinc levels, disease activity, and proinflammatory cytokines, exploratory Spearman correlation analyses were performed (Tables S1 and S2). No significant correlations were observed between changes in serum zinc levels or CDAI and changes in TJ protein expression. Among the cytokines analyzed, only changes in IL‐12 expression showed significant inverse correlations with changes in E‐cadherin (ρ = −0.64, p = 0.030) and ZO‐1 (ρ = −0.73, p < 0.01); no significant correlations were observed for TNF‐α, IL‐17, or IL‐23. In addition, CDAI was not significantly correlated with hemoglobin or albumin levels (Table S2).

4. Discussion

We investigated the association between serum zinc levels, intestinal barrier function, and mucosal immune responses in patients with CD using paired biopsy specimens. We found that the expression of TJ proteins—including E‐cadherin, ZO‐1, and claudin‐1—was significantly lower during the low zinc state than the normal zinc state. In addition, IL‐17 expression was significantly increased during the low zinc state, whereas no significant differences were observed in TNF‐α, IL‐12, or IL‐23 expression. These findings were largely preserved in the sub‐analysis excluding patients who underwent treatment intensification. Taken together, our results suggest that alterations in serum zinc status are associated with changes in epithelial barrier integrity and mucosal immune responses in patients with CD.

These findings can be explained, at least in part, by the established role of zinc in maintaining epithelial barrier integrity. Previous studies have demonstrated that zinc deficiency reduces TJ protein expression and increases epithelial permeability [16, 22]. Animal models of colitis have further shown that dietary zinc deficiency exacerbates intestinal inflammation, accompanied by increased mucosal expression of proinflammatory cytokines [13].

Clinically, zinc deficiency is common in patients with CD [6] and has been associated with increased risks of hospitalization, surgery, and disease‐related complications [9]. In addition, small clinical studies, including our previous report, have suggested that zinc supplementation may improve disease activity in patients with IBD who have low zinc levels [12, 23].

However, previous findings have been largely limited to experimental models or indirect clinical associations; to our knowledge, no study has directly examined the relationship between serum zinc status and the expression of TJ proteins or proinflammatory cytokines in human intestinal biopsy specimens. Therefore, our study extends previous experimental observations to the clinical setting, demonstrating that zinc status is associated with alterations in barrier‐related proteins and mucosal cytokine expression in patients with CD.

To further explore these associations, we performed exploratory correlation analyses using changes in each parameter. No significant correlations were observed between changes in serum zinc levels and changes in TJ protein expression. These findings suggest that the association between zinc status and epithelial barrier integrity may not represent a simple dose‐dependent relationship. Nevertheless, the paired comparison consistently demonstrated lower expression of several TJ proteins during the low‐zinc state, supporting an association between zinc status and mucosal barrier alterations.

In the present study, IL‐17 expression was significantly increased during the low zinc state. IL‐17 is a key cytokine in the Th17 pathway and plays a central role in the pathogenesis of CD. Previous studies have suggested that zinc deficiency promotes Th17 responses, leading to increased production of IL‐17 [13]. These findings are consistent with our results and suggest that fluctuations in zinc status may modulate mucosal immune responses in CD. Among the cytokines examined, only changes in IL‐12 expression were inversely correlated with changes in E‐cadherin and ZO‐1, whereas no significant correlations were observed for TNF‐α, IL‐17, or IL‐23. As IL‐12 is a key cytokine involved in Th1 differentiation, this finding may suggest that Th1‐associated inflammation contributes to epithelial barrier dysfunction. However, given the exploratory nature of these analyses and the limited sample size, this observation should be interpreted cautiously.

We observed that disease activity and nutritional status differed between the low and normal zinc states, as reflected by higher CDAI scores and lower hemoglobin and albumin levels during the low zinc state. These findings suggest that serum zinc levels may, at least in part, reflect underlying disease activity and systemic nutritional status. Therefore, the observed differences in TJ protein and cytokine expression may have been influenced by clinical disease activity. However, exploratory correlation analyses demonstrated no significant correlations between changes in CDAI and changes in TJ protein expression, and no significant correlations were observed between CDAI and hemoglobin or albumin levels. These findings suggest that disease activity alone is unlikely to fully account for the observed alterations in TJ protein expression. Nevertheless, the present study cannot determine whether altered zinc status is a cause or a consequence of mucosal changes in patients with CD. This issue should be addressed in future prospective interventional studies.

Despite these considerations, the paired design of this study enabled intra‐individual comparison under different zinc conditions, thereby reducing confounding from inter‐individual variability. In addition, similar trends were observed in the sub‐analysis excluding patients with treatment intensification, supporting an association between zinc status and mucosal changes.

This study has some limitations. First, the sample size was relatively small, which may have limited the statistical power to detect modest associations, particularly in the exploratory correlation analyses. In addition, the retrospective, single‐center design may limit the generalizability of the findings. Second, although efforts were made to standardize biopsy sampling, the sites were not completely uniform, which may have influenced the assessment of mucosal expression. Third, as disease activity and nutritional status differed between the low and normal zinc states, potential confounding effects cannot be fully excluded. Fourth, circulating cytokine levels were not evaluated because residual serum samples were unavailable, precluding assessment of the relationship between mucosal cytokine expression and systemic cytokine profiles. Fifth, the present study relied primarily on immunohistochemical assessment. Although quantitative image analysis was performed using a predefined and uniform thresholding protocol, complementary validation methods, such as Western blotting, gene expression analyses, or fluorescence‐based immunostaining, were not feasible because of the limited amount of archived biopsy tissue. Therefore, the present findings should be interpreted with appropriate caution and require confirmation in larger prospective studies using complementary methodologies.

Despite these limitations, our findings provide clinical evidence linking zinc to mucosal homeostasis. Prospective studies are warranted to further clarify the clinical impact of zinc supplementation in patients with CD.

In conclusion, the normal zinc state was associated with increased expression of TJ proteins and reduced expression of IL‐17 in patients with CD. These findings suggest that changes in zinc status are associated with alterations in epithelial barrier integrity and mucosal immune responses. Although further validation is required, zinc supplementation may represent a simple and accessible adjunctive therapy to current treatment strategies.

Funding

The authors have nothing to report.

Ethics Statement

This study was approved by the Ethics Committee of our institution (IRB no. 023‐0322) and conducted in accordance with the Ethical Guidelines for Medical and Biological Research Involving Human Subjects.

Consent

The study protocol was published on our institutional website, and written informed consent was obtained from all participants.

Conflicts of Interest

Naoya Sakamoto has received lecture fees from Gilead Sciences, AbbVie, Chugai Pharmaceutical Co. Ltd., and AstraZeneca PLC, as well as research grants from Chugai Pharmaceutical Co. Ltd., Sumitomo Pharma Co. Ltd., Shionogi & Co. Ltd., and Eisai Co. Ltd. Shinsuke Otagiri received a research grant from Mochida Pharmaceutical Co. Ltd. The other authors declare no conflicts of interest.

Supporting information

Figure S1: Control and optimization images for tumor necrosis factor‐alpha (TNF‐α) immunohistochemistry. Representative images of human tonsillar tissue are shown for the negative control, preliminary optimization at a dilution of 1:100, and the final positive control at a dilution of 1:50. Negative controls were prepared by omitting the primary antibody. Whole‐slide images were scanned at ×20 magnification. Scale bars = 800 μm (left) and 50 μm (right).

Figure S2: Positive and negative controls for interleukin‐17 (IL‐17) immunohistochemistry. Representative images of human tonsillar tissue are shown for the negative and positive controls. Negative controls were prepared by omitting the primary antibody. Whole‐slide images were scanned at ×20 magnification. Scale bars = 800 μm (left) and 50 μm (right).

Figure S3: Positive and negative controls for interleukin‐12 (IL‐12) immunohistochemistry. Representative images of human tonsillar tissue are shown for the negative and positive controls. Negative controls were prepared by omitting the primary antibody. Whole‐slide images were scanned at ×20 magnification. Scale bars = 800 μm (left) and 50 μm (right).

Figure S4: Positive and negative controls for interleukin‐23 (IL‐23) immunohistochemistry. Representative images of human tonsillar tissue are shown for the negative and positive controls. Negative controls were prepared by omitting the primary antibody. Whole‐slide images were scanned at ×20 magnification. Scale bars = 800 μm (left) and 50 μm (right).

Figure S5: E‐cadherin immunostaining in paired biopsy specimens. Representative low‐magnification images of E‐cadherin immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S6: Zonula occludens‐1 (ZO‐1) immunostaining in paired biopsy specimens. Representative low‐magnification images of ZO‐1 immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S7: Occludin immunostaining in paired biopsy specimens. Representative low‐magnification images of occludin immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S8: Claudin‐1 immunostaining in paired biopsy specimens. Representative low‐magnification images of claudin‐1 immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S9: Tumor necrosis factor‐alpha (TNF‐α) immunostaining in paired biopsy specimens. Representative low‐magnification images of TNF‐α immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S10: Interleukin‐17 (IL‐17) immunostaining in paired biopsy specimens. Representative low‐magnification images of IL‐17 immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S11: Interleukin‐12 (IL‐12) immunostaining in paired biopsy specimens. Representative low‐magnification images of IL‐12 immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Figure S12: Interleukin‐23 (IL‐23) immunostaining in paired biopsy specimens. Representative low‐magnification images of IL‐23 immunostaining in paired biopsy specimens obtained during the low‐ and normal‐zinc states are shown. Whole‐slide images were scanned at ×20 magnification. Scale bars = 500 μm.

Table S1: Correlation between tight junction proteins and clinical/pathological parameters.

Table S2: Correlation analysis between CDAI and laboratory parameters.

Data Availability Statement

The full‐slide images of all analyzed specimens are available from the corresponding author upon reasonable request.