Introduction

Crohn’s disease (CD) is associated both (i) with polymorphisms in the card15 gene, coding for the cytosolic NOD2 innate immunity receptor, recognizing bacterial peptidoglycan (PG), and (ii) with a decrease in Faecalibacterium, an anti-inflammatory intestinal bacterium¹. Oral administration of Faecalibacterium could represent a novel therapeutic strategy for CD patients, which is currently under investigation with F. prausnitzii EXL01 strain (human clinical trial #NCT05542355). Indeed, Faecalibacterium exhibits anti-inflammatory effects both in vitro and in vivo in colitis models^1-6 and we recently discovered that it directly induced the production of Interleukin-10 (IL-10) in CD14+ human monocytes⁷. However, the bacterial molecules, the host receptors and signaling pathways involved are still poorly understood. Therefore, the aim of the current study was to decipher the role of the interaction between Faecalibacterium, its PG and NOD2 in health and CD.

Aims & Methods

The PG structure of F. prausnitzii EXL01 was determined by ultra high-performance liquid chromatography and mass spectrometry. NOD2 activation was evaluated using HEK-human NOD2 reporter cell line. Following stimulation with either bacteria or PG, secretion of IL-10 and TNF-a by i) WT and NOD2 KO THP1 human monocytic cell line and ii) by dendritic cells (DC) from WT and NOD2^-/- mice and from human patients with CD harboring 0 or 2 NOD2 loss-of-function frameshift mutations were evaluated by ELISA.

Results

Faecalibacterium PG presents an amino acid substitution in the minimal motif recognized by NOD2 (data patented). PG of F. prausnitzii activates NOD2, but not NOD1, in a dose-dependent manner and has a higher ability to activate NOD2 than the one from other Gram positive bacteria. We showed that the production of IL-10 by THP1 monocytes and murine DC upon stimulation either with F. prausnitzii or its PG was decreased in absence of NOD2, indicating that secretion of IL-10 by human and murine immune cells upon Faecalibacterium stimulation is dependent on NOD2. Interestingly, no significant effect was observed for TNF-a secretion, leading to a significant decrease of the IL-10/TNF- a ratio in NOD2 KO cells. In addition, secretion of IL-10 upon stimulation with F. prausnitzii or its PG was lower in DC from CD patients with 2 NOD2 loss-of-function frameshift mutations compared to DC from CD patients without NOD2 mutation. Levels of secreted TNF-a were higher in DC from CD patients with 2 NOD2 loss-of-function frameshift mutations compared to DC from CD patients without NOD2 mutation, leading to decreased IL-10/TNF- a ratio in absence of functional NOD2. The pathways downstream of NOD2 as well as the effects of Faecalibacterium and its PG on NOD2 in vivo are currently under investigation.

Conclusion

Our study showed that Faecalibacterium and its PG activate NOD2, and therefore the production of IL-10, in a different way than other Gram-positive bacteria. These results provide new insight into the molecular pathways underlying the immunomodulatory effects of F. prausnitzii, and highlight NOD engagement as a key immunological feature. Identification of PG as a molecule responsible for the therapeutic effects paves the way for the next generation of Faecalibacterium-based drugs, whether whole-cell or component-driven, in inflammatory diseases such as CD.

References

1. Sokol H, Pigneur B, Watterlot L, et al. Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identified by gut microbiota analysis of Crohn disease patients. Proc Natl Acad Sci U S A. 2008; 105: 16731-6.
2. Lenoir M, Martin R, Torres-Maravilla E, et al. Butyrate mediates anti-inflammatory effects of Faecalibacterium prausnitzii in intestinal epithelial cells through Dact3. Gut Microbes. 2020; 12: 1-16.
3. Quevrain E, Maubert MA, Michon C, et al. Identification of an anti-inflammatory protein from Faecalibacterium prausnitzii, a commensal bacterium deficient in Crohn's disease. Gut. 2016; 65: 415-25.
4. Rossi O, van Berkel LA, Chain F, et al. Faecalibacterium prausnitzii A2-165 has a high capacity to induce IL-10 in human and murine dendritic cells and modulates T cell responses. Sci Rep. 2016; 6: 18507.
5. Sarrabayrouse G, Bossard C, Chauvin JM, et al. CD4CD8 lymphocytes, a novel human regulatory T cell subset induced by colonic bacteria and deficient in patients with inflammatory bowel disease. PLoS Biol. 2014; 12: e1001833.
6. Touch S, Godefroy E, Rolhion N, et al. Human CD4+CD8alpha+ Tregs induced by Faecalibacterium prausnitzii protect against intestinal inflammation. JCI Insight. 2022; 7.
7. https://www.biorxiv.org/content/10.1101/2024.10.06.616495v1

Disclosure

HS reports lecture fees, board membership, or consultancy from Carenity, AbbVie, Astellas, Danone, Ferring, Mayoly Spindler, MSD, Novartis, Roche, Tillots, Enterome, BiomX, Takeda, Biocodex, has stocks from Enterome and is co-founder of Exeliom Biosciences. PL report lecture fee, board membership, or consultancy from Biose, Biostime, Boiron, Bonduelle, BMS, Bromatech, IPSEN, iTaK, Lallemand, Lesaffre, L’Oréal, Mayoli, Merck, Procter and Gamble, Second Genome, Therascience and URGO and is co-founder of Exeliom Biosciences. Other authors declare no competing interests. DS, LH, PR and BH are employees of Exeliom Biosciences.

Introduction

Inflammatory Bowel Disease (IBD)^1,2 requires close monitoring to detect flare ups early and guide treatment. While endoscopy is the gold standard, it is invasive, costly, and uncomfortable^1,2. Intestinal Ultrasound (IUS) is a non-invasive alternative with strong correlation to endoscopic findings—especially bowel wall thickness (BWT), a marker of disease activity³. Nevertheless, IUS remains limited due to operator dependence and a steep learning curve⁴.
AI has improved consistency in other domains, e.g. endoscopic evaluation, but its application to IUS in IBD remains underexplored. Prior models lack interpretability and require manual image cropping, or are trained on ideal, selected data, limiting clinical use^5,6.

Aims & Methods

We aim to develop a deep-learning model that automatically identifies and paints the bowel wall and measures BWT directly from raw clinical IUS images.
A training dataset of 570 images from 144 IUS videos and a testing dataset of 55 images from 55 separate exams were created. All images were extracted from previously performed IUS examinations reflecting real-world variation, including e.g. unclear boundaries and artefacts.
All images were annotated by International Bowel Ultrasound Group (IBUS) certified experts, including outline-paintings of the inner and outer bowel wall and two BWT measurements.
The AI consists of a combination of Convolutional neural networks and other image processing algorithms.
Evaluation included BWT error against the expert mean, classification accuracy using the standard IBUS 3 mm threshold³, and a leave-one-out comparison with individual doctors.

Results

The model produced predictions on 54 of 55 test images.
The model deviated from the gold standard mean by 0.98 mm (SD 1.1 mm) per image on the regression task. The average distance to expert-defined bounds was 0.44 mm (SD 0.89 mm), with 59% of predictions inside this range.
For classification (using a 3 mm threshold), the model reached an accuracy of 0.77, sensitivity of 0.69, specificity of 0.94, and Cohen’s Kappa of 0.56.
In the leave-one-out analysis, expert performance ranged from 0.89–0.93 accuracy and 0.79–0.83 Kappa. Depending on which expert was excluded from the gold standard, the model achieved 0.74–0.80 accuracy and 0.49–0.60 Kappa. Experts stayed within expert-defined bounds in 72–81% of cases, while the model did so in 46–55%.

Conclusion

Clinically, the model performed well. An error of 0.5–1 mm is negligible in practice and matches typical variation in manual measurements. Many real-world test images had BWT values near the 3 mm threshold, so small deviations led to misclassifications. In practice, a 2.5 mm reading could still raise concerns based on symptoms³.
While the system alone cannot provide expert-level BWT measurements, it could provide assistance to experts as well as non-expert and junior doctors, especially in locating the bowel, which is an essential part of IUS.
The model works on unprocessed clinical data, with the only selection criteria being that there is an identifiable bowel segment, measurable by an IBUS-certified doctor. Allowing the images to reflect realistic conditions
In summary, we developed an AI model that identifies the bowel and measures BWT with acceptable accuracy. We have already started the processes of collecting video data, to extend the AI’s functionality to a fully clinical setting.

References

1. Torres J, Mehandru S, Colombel JF, Peyrin-Biroulet L. Crohn’s disease. The Lancet. 2017;389(10080):1741-1755. doi:10.1016/S0140-6736(16)31711-1
2. Ungaro R, Mehandru S, Allen PB, Peyrin-Biroulet L, Colombel JF. Ulcerative colitis. Lancet. 2017;389(10080):1756-1770. doi:10.1016/S0140-6736(16)32126-2
3. Novak KL, Group I, Nylund K, et al. Expert Consensus on Optimal Acquisition and Development of the International Bowel Ultrasound Segmental Activity Score [IBUS-SAS]: A Reliability and Inter-rater Variability Study on Intestinal Ultrasonography in Crohn’s Disease. J Crohns Colitis. 2021;15(4):609-616. doi:10.1093/ECCO-JCC/JJAA216
4 Bove L, Meyer J, Collins M, Frampas E, Bourreille A, Le Berre C. Understanding the learning curve of intestinal ultrasound in inflammatory bowel disease: A comparative study between novice, regular, and expert. Clin Res Hepatol Gastroenterol. 2025;49(3):102548. doi:10.1016/J.CLINRE.2025.102548
5. Kumaralingam L, Le May K, Dang VB, Alavi J, Huynh HQ, Le LH. Artificial intelligence-assisted approach to assessing bowel wall thickness in pediatric inflammatory bowel disease using intestinal ultrasound images. J Crohns Colitis. 2025;19(4). doi:10.1093/ECCO-JCC/JJAF037
6. Carter D, Albshesh A, Shimon C, et al. Automatized Detection of Crohn’s Disease in Intestinal Ultrasound Using Convolutional Neural Network. Inflamm Bowel Dis. February 2023:1-6. doi:10.1093/IBD/IZAD014