Anti-H. pylori growth activity of C. sativa leaf extract and castalagin

The anti-H. pylori growth activity of C. sativa leaf extract and castalagin was evaluated on the cagPAI-positive H. pylori strains PMSS1, G27, and P12. The differences of the three H. pylori strains in terms of virulence factors and susceptibility/resistance to antibiotics are shown in Table S1. The results were particularly promising for both the extract (Fig. 1A, B, C) and the pure molecule (Fig. 1D, E, F), showing a relevant inhibitory activity on bacterial growth at value of 50 μg/mL and 6 μM, respectively, for all H. pylori strains after 72 h of growth. Interestingly, this antibacterial activity was found to be comparable across the three strains examined, and consistent with previous results obtained with H. pylori strain 26695 (Piazza et al. 2023). Tetracycline, selected as a reference antibiotic, completely inhibited the bacterial growth at 0.25 μg/mL or 0.56 μM (Fig. 1G, H, I).

Fig. 1

The anti-H. pylori growth activity of C. sativa leaf extract and castalagin. H. pylori PMSS1, G27, and P12 strains were cultivated in the presence of extract (CL) or pure molecule (C). Firstly, the anti-H. pylori growth activity of the extract (A, B, C) and pure molecule (D, E, F) at different concentration was measured by microbroth dilution assay after 72 h. Tetracycline (T) was utilised as the reference antibiotic (G, H, I), exhibiting a MIC of 0.06 μg/mL. Secondly, the anti-H. pylori growth activity of the extract (CL, 100 μg/mL, grey bars) and pure molecule (C, 10 μM, white bars) was measured by microbroth dilution assay over time, for 24, 48, or 72 h (J, K, L). The results are presented as the mean ± SEM (n = 4) and expressed as the relative percentage compared to H. pylori growth (black bars), which was arbitrarily assigned the value of 100%. The one-way ANOVA statistical test was employed, followed by the Bonferroni post-hoc test. * p < 0.05, **p < 0.01, and ***p < 0.001 vs. H. pylori

To gain a better understanding of the timing required for the action and efficacy of the extract and the pure molecule, the anti-H. pylori growth activity was additionally assessed over the time at the selected concentration of 100 μg/mL for chestnut leaf extract and 10 μM for castalagin (Fig. 1J, K, L). The time course experiment demonstrated that the antibacterial activity was already present at 24 h, although this effect was not statistically significant for the extract. The effect was significant at 48 h, reaching a maximum H. pylori growth inhibition at 72 h. Again, the activity observed among the three distinct H. pylori strains was comparable.

In vitro anti-H. pylori colonisation and anti-inflammatory activity of C. sativa leaf extract and castalagin

The mechanisms of pathogenicity of H. pylori and related gastric inflammation have been widely studied in gastric epithelial cell lines, including the human GES-1 cells, by our research group (Martinelli et al. 2023a). Despite the numerous advantages offered by cell lines, they are not entirely representative of the complex environment of H. pylori infection. Gastric organoids are considered to be the in vitro model that most closely resembles the physiology of complete organs, mimicking the cellular diversity and the architectural complexity of the stomach (Idowu et al. 2022; Seidlitz et al. 2021).

The cagPAI-positive PMSS1 H. pylori strain was selected for organoid cell infection (MOI of 50). The chestnut leaf extract (concentration range: 50–200 μg/mL) and castalagin (concentration range: 25–100 μM) did not exhibit signs of cytotoxicity towards organoid cells after 72 h of treatment, as determined by the Trypan blue assay (Fig. S1).

Following a 6-h exposure of 2D-seeded organoid cells to H. pylori, the extract (100 μg/mL) and the pure molecule (25 μM) were added to the co-culture for a further 6, 24, 48, or 72 h. From 24 h onward, both the extract and the pure molecule inhibited H. pylori viability in this model (Fig. 2A, B). This inhibition was clearly evident and statistically significant at 48 h (Fig. 2C) and at 72 h of treatment (Fig. 2D).

Fig. 2

In vitro anti-H. pylori colonisation and anti-inflammatory activity of C. sativa leaf extract and castalagin on H. pylori PMSS1 cultured on murine gastric organoid cells. Firstly, 2D-organoid-derived monolayers were infected with H. pylori (Hp) for a period of 6 h and then were treated with the extract (CL, 100 μg/mL) or pure molecule (C, 25 μM) for a period of 6 (A), 24 (B), 48 (C), or 72 (D) hours. Thereafter, cells were harvested and plated into horse blood petri dishes. Following a period of 4 days, the formation of H. pylori colonies on the petri dishes was meticulously enumerated. The results are presented as the mean ± SEM of the absolute number of colonies counted per well following treatment with the extract or the pure molecule in a minimum of three experiments. Secondly, 2D-organoid-derived monolayers were infected with H. pylori for a period of 6 h and then were treated with the extract or pure molecule for a period of 24 (E, F, G, H) or 48 (I, J, K, L) hours. The gene expression on gastric tissue was measured with qPCR using the TaqMan method. The results were indicated as the mean ± SEM of the relative gene expression, calculated using the 2ΔC(t) method in a minimum of three experiments. Gene expression levels for each sample were normalized to HPRT expression. The one-way ANOVA statistical test was employed, followed by the Dunnett test. Tetracycline (T, 30 μg/mL) was utilised as the reference antibiotic. CTRL, untreated tissues, *p < 0.05, **p < 0.01, ***p < 0.001 vs. H. pylori

Since inhibition of IL-8 secretion and NF-κB translocation in H. pylori-infected GES-1 cells was previously demonstrated (Piazza et al. 2023), the inhibition of inflammatory genes responsible for leukocytes recruitment, particularly those linked to the NF-κB pathway and involved in CXCR2 and CXCR3 activation was evaluated. The expression of inflammatory genes in organoids, including Cxcl1, Cxcl2, Cxcl10, and Ccl2, has recently been evaluated in murine gastric organoids following stimulation of the NF-κB pathway with TNFα or ADP heptose (Wizenty et al. 2022), while it has never been evaluated in the H. pylori infected-WT murine gastric organoids.

Co-cultures of WT murine gastric organoids and H. pylori PMSS1, were treated with the extract (100 μg/mL) and the pure molecule (25 μM) for 24 h (Fig. 2E, F, G, H) or 48 h (Fig. 2I, J, K, L), followed by qRT-PCR. The results demonstrated that Cxcl1, the functional murine homologue of human CXCL8 gene, was significantly inhibited at 24 h in respect to infected tissues (Fig. 2E), while Cxcl2, another functional murine homologue of human CXCL8 gene, was statistically inhibited at the later time points (48 h) (Fig. 2J). Inhibitory effects were further observed on the expression of Ccl2 and Cxcl10 genes, which was more pronounced and significant after 48 h of treatment with castalagin (Fig. 2K, L).

Cxcl11 and Il6 were not expressed at 24 h, while they were slightly expressed and inhibited by the extract and pure molecule at 48 h (Fig. S2); the Cxcl9 gene was not expressed at any time point considered. The expression levels of the genes under investigation were also evaluated after 72 h of treatment. The results showed that the Cxcl1, Cxcl2, Ccl2, and Cxcl10 were expressed by H. pylori-infected gastric organoids, slightly reduced 72 h of treatment (Fig. S3). In contrast, the Cxcl11, Il6, and Cxcl9 genes were not expressed.

These findings from our in vitro model of H. pylori-infected gastric organoids corroborated those previously obtained in H. pylori-infected GES-1 cells, thus confirming the antibacterial and the anti-inflammatory activities of C. sativa leaf extract and castalagin.

In vivo anti-H. pylori colonisation and anti-inflammatory activity of C. sativa leaf extract and castalagin

To study potential antibacterial and anti-inflammatory activities of chestnut leaf extract and castalagin in an in vivo murine model, C57BL/6 WT mice were infected with the H. pylori PMSS1 strain for a period of 5 weeks. This period is necessary for the development of stable bacterial colonisation, which induces gastritis and triggers both local and systemic immune responses (Arnold et al. 2011; Artola-Borán et al. 2025). Thereafter, the mice were treated for a period of 10 days with the chestnut leaf extract (100 or 500 mg/kg) or castalagin (25 mg/kg). Gastric tissue of infected mice was used to assess H. pylori colonisation, the production of inflammatory mediators, and immune cell infiltration.

The extract (100 mg/kg) and the pure molecule exhibited a marked and significant inhibition of bacterial colonisation (Fig. 3A), corroborating the previously observed antibacterial activity.

Fig. 3

In vivo anti-H. pylori colonisation and anti-inflammatory activity of C. sativa leaf extract and castalagin in infected mice. WT mice were infected with H. pylori PMSS1 strain (Hp) for a period of 5 weeks and then were treated with the extract (CL, 100 mg/kg and 500 mg/kg) or pure molecule (C, 25 mg/kg) for the last 10 days of the infection. Thereafter, the stomachs were harvested, homogenized, and plated onto horse blood petri dishes for the evaluation of anti-H. pylori colonisation. H. pylori were enumerated 7 days later. Horizontal bars indicate medians. The non-parametric one-way ANOVA statistical test (Kruskal–Wallis) was employed, followed by the Dunn’s post-hoc test (A). The inhibition of inflammatory gene expression was measured on gastric tissue with qPCR using the TaqMan method. The results are presented as the mean ± SEM of the relative gene expression, calculated using the 2ΔC(t) method. Gene expression levels for each sample were normalized to HPRT expression. One-way ANOVA statistical test was employed, followed by the Dunnett test (B, C, D, E, F, G). The inhibitory activity of the extract and castalagin was measured on neutrophil, monocyte, and lymphocyte infiltration into gastric tissue. The quantification of absolute counts per stomach by spectral flow cytometry is shown. The samples were then acquired on a 5-Laser Cytek© Aurora. The Kruskal-Wallis non-parametric test was employed, followed by the Dunn’s multiple comparisons test (H, I, J, K, L, M, N, O, P, Q, R, S). *p < 0.05, **p < 0.01, *** p < 0.001 vs. H. pylori

Cxcl1 was inhibited following treatment, with statistical significance for castalagin (Fig. 3B). Interestingly, Tnfa, Cxcl9, Cxcl10, and Ifng were also reduced, albeit nor significantly (Fig. 3D, E, F, G,). Other genes such as Ccl2, Il1b, Il6, and Il17A, were poorly expressed in the gastric mucosa of mice following 5 weeks of H. pylori infection.

To ascertain if the chestnut leaf extract and castalagin may have exerted a direct effect on the transcription factor NF-κB, two NF-κB inhibitor genes were also investigated. The expression of Nfkbia was unaffected by the treatments; however, the extract exhibited a slight inhibition of Nfkbiz gene expression, as an indirect consequence of NF-κB pathway inhibition (Fig. S4).

The immune response to infection, particularly the early recruitment of monocytes, neutrophils, macrophages, and dendritic cells to the site of inflammation, is directed by epithelial activation of NF-κB target genes. To study the most relevant myeloid, lymphocytic, T-cell, and granulocytic populations, we performed a flow cytometric analysis on single cell suspensions from gastric tissue for a comparison with colonisation data shown in Fig. 3A and Fig. S5. Mice were treated with chestnut leaf extract (500 mg/kg) or castalagin (25 mg/kg). As expected, the initial response to H. pylori was characterized by the presence of neutrophils and monocytes, which was effectively prevented by castalagin, while with a less extent by the chestnut leaf extract (Fig. 3H, I, J). Lymphocyte populations (e.g., Ki67 CD4 + , cMAF CD4 + , NKT, CD8 + , Tregs, Th17) were not affected by treatment (Fig. 3K, L, M, N, O, P, and Fig. S6).