Precision cut intestinal slices, a novel model of acute food allergic reactions

2.1 Source of human intestinal tissue

Human intestinal tissue was obtained from 16 children (median age: 4 months) undergoing clinically indicated resection of the small or large intestine (tissue donor information in Table S1). Patients with systemic inflammatory disorders, metabolic disorders, or systemic immunosuppression were excluded. Samples from patients undergoing surgery due to non-inflammatory disorders (e.g., stoma closure after an initial surgical correction of Hirschsprung's disease) were eligible. The use of human tissue for research was approved by the SickKids Research Ethics Board (REB #1000059282). All experiments were carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving human subjects. Privacy rights of all human subjects were observed, and informed consent was obtained from the parents/guardians.

2.2 Preparation of PCIS

PCIS were prepared as previously published in detail by de Graaf et al. with modifications.^{21, 23} In short, a piece of full-thickness human intestinal tissue (approximately 1.0 × 0.5 × 0.2 cm) was cut from the healthy region of a surgically resected specimen by a trained pathologist directly following surgical removal. The sample was immediately placed into oxygenated, chilled (4°C, 95% O₂, 5% CO₂) Krebs Henseleit Buffer (KHB), and transported to the laboratory within one hour. The sample was then dissected into strips approximately 0.5 × 0.2 cm and embedded in low melting temperature agarose (37°C, 3% w/v, 0.9% NaCl; MilliporeSigma, Darmstadt, Germany). Once the agarose solidified, the tissue was cut into 400 μm thick slices using a Krumdieck Tissue Slicer (Alabama Research and Development, Munford, AL, USA), or a VT1200S Vibratome (Leica Microsystems, Nussloch, Germany). PCIS were then incubated in 12-well culture plates in 1.2 ml Williams' medium E (WME) containing L-glutamine and supplements (D-glucose 14 mM, gentamycin 50 μg/ml; Gibco, MA, USA) in an oxygenated incubator (37°C, 90% O₂, 5% CO₂; Thermo Fisher Scientific, MA, USA) with gentle shaking (Figure 1A).

2.3 Viability assays

The viability of PCIS were assessed using lactate dehydrogenase (LDH) release (cytotoxicity) assays and water-soluble tetrazolium salt (WST-1) (metabolic) assays (Roche, Mannheim, Germany) from 0 to 24 h in culture. Treatment with the detergent Triton X-100 (MilliporeSigma, Darmstadt, Germany) was included to show maximum LDH release and as a dead-control.

2.4 Single nucleus RNA sequencing of human colon tissue

Colon tissue samples (n = 4), also used to generate PCIS, were snap-frozen, and stored at -80°C prior to single nucleus RNA sequencing (sNuc-Seq). After library generation and quality control, the samples were deep sequenced using a NovaSeq (target 10,000 nuclei, 80,000 reads/nuclei). FASTQ files were aligned to a reference human transcriptome using 10X Genomics Cell Ranger (CA, USA). The resulting gene expression data were merged using the Seurat R package.^{24, 25} Batch effects were removed by “regressing out” variables, and clustering was performed using the Leiden algorithm implemented in Seurat. Cell identities were assigned and confirmed using a list of pre-defined marker genes in reference to single cell gene expression databases and the top distinct genes expressed by each cluster.^{26, 27}

2.5 Passive sensitization

PCIS were passively sensitized in a 1:10 dilution of human plasma from clinically confirmed peanut allergic donors or peanut-sensitized non-allergic donors (Table S2) in WME for 90–120 min (37°C, 90% O₂, 5% CO₂) with gentle shaking.^{28, 29} Human plasma was derived from ongoing and completed trials including the Markers of Nut Allergy Study (MONAS) cohort at the Hospital for Sick Children.³⁰ The use of human plasma for research was approved by the SickKids REB for MONAS (#1000053791) and the Low Dose Multi-OIT (LoMo) study (#1000060633).

2.6 Video microscopy

PCIS muscle contraction was filmed using a stereomicroscope with a camera attachment (Walter Products, ON, Canada; TCapture Software, Tucsen Photonics, Fuzhou, China) for 10 min without the addition of stimuli to establish baseline movement of the individual sample. The slices were then filmed for an additional 10 min upon addition of either peanut extract (PE) (1 μg/ml, ALK-Abelló, Hørsholm, Denmark), serotonin (10 μg/ml, MilliporeSigma, Darmstadt, Germany), FcɛRI-crosslinker (10 μl, Bühlmann, Amherst, NH, USA), histamine (10 μg/ml, ALK-Abelló, Hørsholm, Denmark), or ovalbumin (OVA; 1 μg/ml, InvivoGen, San Diego, CA, USA) as an irrelevant allergen. Any samples not passing quality control criteria (Table S3) were excluded from analysis.

2.7 Analysis of PCIS smooth muscle contractions

Statistical analysis was conducted using GraphPad Prism 6 (GraphPad Software, San Diego, CA, USA). Responses between different stimulations were compared using the Mann–Whitney U test with a p value < 0.05 considered statistically significant.

A novel video analysis software was developed internally (available at: https://github.com/celalp/video_parser). This software measures muscle contraction using individual pixel movement per frame (Figure 2). Videos were analyzed using a semi-supervised approach with variables kept consistent between the unstimulated control and stimulated sample. Videos were processed using OpenCV (Open Source Computer Vision Library) and scikit-image Python packages.^{31, 32} Pixels that changed from frame x-1 to frame x were calculated using a Gaussian mixture-based background/foreground segmentation algorithm.^{33, 34} A background subtraction algorithm was chosen for this analysis as videos consisted of one large central object with a mostly static background, meaning that any observed movement was captured as foreground. The selection of the specific algorithm (MOG2 in OpenCV) was influenced by the large variability of exposure levels (e.g., glare) between different slices. A dynamic Gaussian Mixture Method allowed for the same background removal method for all experiments. Measurements were validated and confirmed by an independent assessor. Overall pixel movement per frame of the video was plotted for each stimulation and total Area Under the Curve (AUC) was calculated and compared for each contraction response using GraphPad Prism 6.

2.8 Tryptase quantification in cell free supernatant

Culture supernatant from PCIS passively sensitized and stimulated with PE (1 μg/ml) or controls was collected 10 min post-stimulation and stored at −80°C. Tryptase levels in supernatant were measured using a tryptase beta-2 ELISA (MilliporeSigma, Burlington MA, USA) after concentration using centrifugal filter devices (Amicon Ultra-0.5 10 K device, MilliporeSigma, Burlington, MA, USA).

2.9 Treatment with antihistamines

Passively sensitized PCIS were incubated with diphenhydramine hydrochloride (10 μg/ml), cetirizine dihydrochloride (1 μg/ml), and rupatadine (1 μg/ml) (MilliporeSigma, Darmstadt, Germany) for 10 min prior to stimulation with histamine (10 μg/ml) and filmed for 10 min post-stimulation. Concentrations were chosen based on published reports and dose–response experiments.^35-38

2.10 Effect of immunomodulation from successful OIT patients

PCIS were passively sensitized with plasma (1:10 in WME) and stimulated with PE and controls to assess differences in pre- (oral food challenge [OFC] positive) and post-successful peanut OIT (ingestion of 5× the baseline eliciting dose).

3.1 PCIS maintain viability for at least 24 h

This model was generated to study acute, IgE-mediated allergic reactions, which are known to occur within seconds of exposure to a relevant allergen. To confirm that PCIS remained viable during the incubation process, they were monitored from 0–24 h post-slicing. Cell death (LDH) and metabolism (WST-1) were assessed in each sample and demonstrated consistent viability of PCIS during the investigated 24 h in culture (Figure 1B,C).

3.2 sNuc-Seq of gut tissue confirms the presence of mast cells and other cell types

The cellular composition of PCIS was analyzed using sNuc-Seq. Single cell gene expression profiles from colon tissue donors (n = 4) were compiled to reveal the cellular composition of PCIS (Figure 1D). Expression profiles aligned with the cellular identities of each cluster and were confirmed using single cell gene expression databases.^{26, 27} All expected cell types in colon tissue were observed (Figure 1D, Table S4) including epithelial cells (enterocytes, goblet cells, enteroendocrine cells, and endothelial cells), stromal cells (fibroblasts, smooth muscle cells), and immune cells (T cells, γδT cells, B cells, mast cells, dendritic cells, and macrophages).

3.3 PCIS contract in response to antigen-independent stimuli

To validate the model and determine the sensitivity of PCIS to non-specific antigen stimulations, FcɛRI-crosslinker, histamine, and serotonin were selected as positive controls. All conditions elicited a strong, reproducible smooth muscle contraction response, elevated in comparison with the unstimulated control (Figure 3A; Videos S1–S6).

Based on original video data, a novel software was designed to quantify smooth muscle contractions using object and movement detection of irregular shapes. This was achieved by tracking the overall movement of individual pixels within each frame (Figure 2) which were then quantified and graphed resulting in a curve of the contraction response. Total AUCs of each response per stimuli were compared, showing a significant increase in movement between the unstimulated slices and PCIS stimulated with the positive controls (Figure 3B).

3.4 Allergen-specific induction of PCIS contractions reflects allergic status of the donor

In order to establish PCIS as a model of FA, responses must be reproducible, allergen-specific, and clinically relevant. PCIS passively sensitized with plasma from clinically confirmed peanut allergic children displayed a strong contractile response after stimulation with PE (Figure 4A; Videos S7 and S8). This response was strictly allergen-specific, as stimulation with a clinically irrelevant food allergen (OVA) did not result in an increased response (Figure 4A; Videos S9 and S10).

To evaluate whether response patterns were linked to allergen-specific IgE or match clinical reactivity at the individual level, PCIS were passively sensitized with plasma from peanut-sensitized non-allergic donors. These donors had measurable peanut-specific IgE but were asymptomatic upon peanut exposure. PCIS sensitized with this plasma did not show significant muscle contraction responses following stimulation with PE, while maintaining reactivity to positive controls (Figure 4A; Videos S11 and S12). The total AUCs of each contraction response revealed a significantly greater response following stimulation with a relevant allergen (PE), in comparison to an irrelevant allergen (OVA) in slices sensitized with peanut allergic plasma, as well as to PE stimulation in sensitized non-allergic tissue (Figure 4B).

Tryptase was measured in the culture supernatant following stimulation to confirm the release of mast cell derived mediators (Figure 4C). Tryptase concentrations in the supernatant of PCIS were significantly greater following exposure to PE and elevated in response to FcɛRI-crosslinker in tissue sensitized with peanut allergic plasma. Unstimulated tissue or stimulation with an irrelevant allergen (OVA) did not result in similar tryptase levels, demonstrating the specificity of the response and linkage to degranulation (Figure 4C). PE stimulation in slices sensitized with peanut sensitized non-allergic plasma did not result in a significant change in tryptase concentration, reflecting their lack of reactivity to PE exposure (Figure 4C).

3.5 PCIS can be used to study anti-allergic drugs

Antihistamines were used to demonstrate the suitability of PCIS as a model of FA to assess specific mechanisms of action. Pre-treatment of the tissue with the antihistamines diphenhydramine hydrochloride, cetirizine dihydrochloride, and rupatadine resulted in significant suppression of contraction responses to histamine stimulation (Figure 4D).

3.6 Peanut OIT suppresses allergen-specific responses

As a promising treatment option for FA, the suppression of allergic responses following OIT was also investigated using the PCIS model. PCIS passively sensitized with plasma from a peanut allergic individual pre-peanut OIT, stimulated with PE resulted in a strong smooth muscle contraction response (Figure 5). Following completion of the OIT protocol, plasma from the same individual resulted in a diminished response after stimulation with PE, which is reflective of the clinical phenotype of the plasma donor. Co-incubation of pre- and post-OIT plasma also resulted in a reduced contraction response in comparison with the pre-OIT condition, reflecting the sensitivity of the model (Figure 5).