Equine Enteric Glial Culture and Application to the Study of a Neural Inflammatory Mechanism in Equine Colic

Summary

Enteric glia are becoming increasingly recognized for their roles in intestinal homeostasis and disease processes, including postoperative complications. Equine patients recovering from emergency exploratory laparotomy suffer from a high risk of inflammatory postoperative conditions, highlighting the importance of establishing repeatable equine enteric glial primary cell culture for study.

Abstract

Inflammatory postoperative conditions of equine colic (acute abdomen) contribute not only to increased client cost, patient discomfort, and hospitalization time, but in many cases, prove to be life-threatening. A unique population of intestinal cells, enteric glia, are increasingly acknowledged for their roles in sensing the gastrointestinal environment and communicating with surrounding cell types. Interactions between enteric glia and intestinal epithelia may prove critical in establishing how equine enteric glia can alter the mucosal barrier to modulate inflammation in health and colic.

To study this interaction, we present a method of establishing primary equine enteric glial cultures from equine jejunum and exposing the cultures to inflammatory conditions known to be present in colic. Primary enteric glial cultures were obtained from adult horses euthanized for reasons unrelated to colic. Intestinal villi and lamina propria were micro-dissected to expose the submucosa. The isolated submucosa underwent enzymatic digestion with collagenase, protease, and bovine serum albumin for 2-3 h. Next, mechanical digestion involving centrifugation, pipetting, and cell strainers (40-100 µm), yielded a pellet used for plating on 0.05 mg/mL poly-L-lysine-coated wells at a concentration of ~400,000 cells/300 µL of media.

Following confluence and first passage, the enteric glial cells were then exposed to equine recombinant IL-1β (0, 10, 25 ng) for 24 h. To model epithelial-glial interactions at the time of colic, medium conditioned by either control or treated enteric glia was added directly to confluent equine jejunal monolayers while measuring transepithelial electrical resistance (TEER) using a dual-electrode EndOhm chamber. These data demonstrate just one of many potential impactful applications of equine enteric glial culture.

Introduction

Equine colic is the most prevalent medical presenting complaint for emergency consultation¹. With up to 17% of those horses requiring surgical correction, efforts to increase postoperative outcomes should be at the forefront of equine medical research². Currently, postoperative colic patients experience a high risk of several life-threatening disorders including sepsis/endotoxemic shock (12.3% of patients) and postoperative ileus (13.7% of patients)³. Despite progress in treatment for postoperative complications, there continues to be a need for advanced treatments for preventing or treating these conditions.

Recent research has highlighted the local and systemic inflammatory status of colic patients⁴. For example, proinflammatory proteins such as tumor necrosis factor (TNF) alpha, interleukin 1β (IL-1β), interleukin 6 (IL-6), and monocyte chemoattractant protein-1, have all been shown to be significantly increased in expression in colic intestinal mucosa versus normal intestinal tissue⁴. From a systemic point of view, it has been demonstrated that increased TNF alpha in the intestinal tissue is correlated with an increased risk of postoperative nasogastric reflux greater than 2 L, a general measurement of postoperative dysmotility⁴. It is also known that the administration of interleukin IL-1β and TNF alpha is capable of inducing clinical signs of septic shock⁵.

A potential explanation for the inflammatory status of colic intestinal tissue and its link to postoperative complications is the intestinal epithelial barrier. In health, the tight junction complexes linking the single layer of columnar epithelium that lines the intestinal tract provide a functional barrier to limit the luminal content and its bacterial components from reaching the submucosal space and bloodstream. However, the distension and damage caused by colic-associated intestinal obstruction and intestinal manipulation during surgery may disrupt this intestinal barrier function.

In terms of the functional components of the intestinal wall as a whole, the submucosal enteric glial network within the enteric nervous system has been shown to be crucial in the pathophysiological development of postoperative complications associated with inflammatory pathways and may provide a specific therapeutic target^6,7,8,9. Not only are enteric glia present throughout the entire gastrointestinal tract, but they act as sensors of the intestinal environment, influence signaling with numerous cell types of the intestinal wall, and directly regulate the intestinal barrier⁶. It is, therefore, justifiable to presume that these potent sensors of the intestine would be activated by injury and inflammation and could produce an acute response such as alterations in barrier permeability.

This study is the first to describe the culture of equine enteric glia and, more specifically, the role of inflammatory equine enteric glia on intestinal barrier function. Here, we present methods of primary culture of equine submucosal enteric glia and their response to exposure to inflammatory IL-1β, evaluation of the effect of enteric glial products post IL-1β exposure on the permeability of equine enterocyte monolayers, and possible blockade through the application of equine serum.

Protocol

Equine enteric glial primary cultures were obtained from three horses humanely euthanized with an overdose of a barbiturate for reasons unrelated to this study. The horses selected for the culture studies were adult horses with no current history of gastrointestinal disease.

1. Submucosal equine enteric glial primary culture

1. Grow equine epithelial monolayers from jejunal crypts obtained from three separate adult equine horses.
   1. Coat 24-well plates with coating solution for 2 h at 37 °C or overnight at room temperature, wash them 2x with sterile water, and store them at room temperature until the cells are ready for plating.
   2. Following humane euthanasia, within 60 min, remove a section of the jejunum and place it in cold Ringer's solution on ice.
   3. Remove a 10 cm portion from the jejunum and open it at the antimesenteric border to keep non-lymphoid regions at the center of the tissue. Trim the portion to an approximately 7 cm x 7 cm section. Rinse the tissue well in fresh Ringer's solution prior to dissection.
   4. Pin the tissue on a silicone elastomer-coated Petri dish in cold Ringer's solution or PBS mucosal side up, stretching the tissue as much as possible (Figure 1A).
   5. Remove the intestinal villi with curved forceps (Figure 1B) followed by the lamina propria layer from the approximately 5 cm x 5 cm non-lymphoid region (Figure 1C).
      NOTE: These layers separate easily from the submucosa (Figure 1B).
   6. Remove the submucosa in one sheet by freeing it with microdissection scissors at one corner and observe the natural separation while peeling the submucosa away from the inner circular muscle layer of the jejunum (Figure 1D).
   7. Mince the collected submucosal layer into 2-5 mm pieces and place them into a 50 mL conical tube containing 5 mL of "Organo-FBS" (Table 1), 0.825 mg of collagenase, 5 mg of protease, and 20 mg of bovine serum albumin. Incubate the tissue for 2-3 h at 37 °C for enzymatic digestion.
   8. Following incubation, add 10 mL of room temperature "Organo + FBS" to stop the enzyme action. Pipette the tissue mixture up and down approximately 15x with a 10 mL serological pipette to dissociate cells from the tissue and then centrifuge at 3,000 × g for 3 min at room temperature.
   9. Discard the supernatant and resuspend the pellet with 10 mL room temperature PBS by pipetting up and down again approximately 15x with a 10 mL serological pipette to dissociate cells from the tissue.
   10. Centrifuge the conical tube at 3,000 × g for 3 min at room temperature. Discard the supernatant and resuspend the pellet with 10 mL of room temperature "Organo + FBS" by pipetting up and down approximately 15x with a 10 mL serological pipette to dissociate cells from the tissue.
   11. Filter the cells serially through a 100 µm, a 70 µm, and then a 40 µm pore-size cell strainer.
   12. Centrifuge the filtered cells and media at 3,000 × g for 3 min at room temperature. Discard the supernatant and resuspend the pellet in 1 mL of "Organo + FBS."
   13. Stain with trypan blue to identify live cells and count them with a hemocytometer.
   14. Seed approximately 400,000 cells in 300 µL of "Organo + FBS" in each well of a 24-well plate and add growth factors to each well: N2 5 µL/well, G5 5 µL/well, and B27 10 µL/well. Place the 24-well plate in a 37 °C incubator with 5% CO₂ to allow the cells to adhere. Twenty-four hours later, change the media to "glia media" (Table 1) with the above growth factors and change every 48 h (Figure 2).

2. Immunofluorescent cytology

1. Grow the enteric glial cells from the primary culture (P0) (step 2.1.14) to 70-80% confluence and passage them once (P1).
2. Fix the primary culture (P0) cells in the wells using 4% paraformaldehyde for 5 min prior to the application of 0.1% PBS azide.
3. Prepare saturation solution consisting of PBS azide, 4% donkey serum, and 0.5% Triton-X. Add 300 µL of saturation solution to the wells to permeabilize the cells for 2 h at room temperature.
4. Add the primary antibody for the glial marker, glial fibrillary acidic protein (GFAP), and the primary antibody for the fibroblast marker, alpha smooth muscle actin, at 1:1,000 dilutions in saturation solution for overnight incubation at 4 ^oC.
5. Wash 3x with PBS prior to adding secondary antibody for donkey anti-rabbit IgG Alexa Fluor 594 and goat anti-mouse IgG Alexa Flour 488 at a 1:500 dilution in saturation solution for 2-3 h.
6. Stain the nuclei with 4',6-diamidino-2-phenylindole (DAPI) at a 1:1,000 dilution in PBS for 5 min.
7. Following three washes with PBS, store the cells in 0.1% PBS azide.

3. Submucosal equine enteric glial primary culture IL-1β exposure

1. When cells from the first passage (P1) reach 70-80% confluence, perform the IL-1β exposure experiment. Expose the wells to 0 ng, 10 ng, and 25 ng of IL-1β for 24 h. Following 24 h, bank and store the media from each well at -80 °C.

4. Equine organoid culture for production of enterocyte monolayers

NOTE: Following the Stewart et al. protocol published in 2018, banked equine intestinal crypts and buds that had been cryobanked were expanded for 2D enteroid studies¹⁰.

1. Use approximately 700-1,000 buds/crypt fragments (40,000-50,000 cells when dissociated into single cells) per transwell insert. Coat the transwell inserts with 42 µL of 0.05% Matrigel prior to the addition of the enterocytes. Allow the Matrigel to be fixed on the insert by incubating for 1 h at 37 ^oC. Aspirate the extra medium and allow the transwells to dry for 30 min uncovered. Add 200 µL of DMEM-F12 medium to each transwell insert and store the plate at 37 °C until use.
2. Wash the Matrigel patties containing the organoids with PBS and then expose them to 500 µL of Cell Recovery while on ice. Pipette them repeatedly to ensure full collection of cells into a 15 mL tube.
3. Centrifuge the tube at 300 × g for 5 min at 4 °C., remove the supernatant, and suspend the cell pellet in the target volume of intestinal epithelial stem cell (IESC) medium. Calculate the target volume by multiplying the desired number of monolayers by 200 µL of medium. (Table 1).
4. Precoat a 1 mL syringe in IESC media and aspirate the cell suspension through a 16 G needle. Replace the 16 G needle with a 28 G needle and eject the cell suspension through it to encourage separation of the organoids.
5. Place 200 µL of the resulting cell suspension into the apical side of the coated transwells and 500 µL of IESC as well as previously described growth factors on the basal side of the transwell insert¹⁰. Change the media and growth factors every 24-48 h until confluence is reached. Define confluence as the cell density corresponding to a TEER of 1,000 ohmsxcm^2,11. For measurement, see step 5.1.

5. Equine enterocyte monolayer permeability following exposure to inflammatory cytokines and equine enteric glial products

1. Use a dual-electrode TEER-measuring chamber to quantify the TEER across the monolayers and establish a base reading across each transwell permeable membrane. Ensure that the IESC media and transwell have 15 min to come to room temperature prior to taking measurements. Fill the culture cup with 1.5 mL of IESC media and ensure exactly 200 µL of media is present within the transwell insert so that the fluid lines match at the time of measurement.
2. Expose the basolateral aspect of the transwell inserts to either inflammatory cytokines of 25 ng IL-1β, 10 ng and 25 ng interleukin 6 (IL-6) per well, or glial products of cultures exposed to 10 ng or 25 ng IL-1β. Expose the basolateral aspect of the control transwell inserts to glia media, and glia media from cultures not exposed to IL-1β.
3. Measure the TEER for 45 min at 10-15 min intervals by placing each transwell into the culture cup.

6. Statistics

1. Calculate the fold change of the TEER of each transwell for the time points: 0, 10, 20, 30, and 45 min.
2. Utilizing the software of choice (see Table of Materials), compare the correlations between decreased TEER and basolateral exposure to IL-1β (25 ng), IL-6 (10 ng, 25 ng, and 100 ng), and glia media (exposed to 0 ng IL-1β, 10 ng IL-1β, and 25 ng IL-1β) to control media using an unpaired one-tailed t-test.

Results

Microdissection of equine jejunum to the submucosal layer (Figure 1) with further enzymatic and mechanical digestion, could produce viable cell cultures of equine enteric glia. The cells demonstrated a pleomorphism with a dominance of spindle-shaped cells consistent with the enteric glia of other species (Figure 2A). The cultures were positive for the selective glial marker, glial fibrillary acidic protein (GFAP), with low fibroblast contamination (alpha SMA, Figure 2B). In some instances, microdissection of the submucosal layers was not controlled to a high enough degree, and cultures were contaminated with high percentages of fibroblasts observed through immunofluorescence of fibroblast marker smooth muscle actin (Figure 3).

Next, as an example of the usefulness of equine enteric glial culture, the impact of IL-1β-exposed enteric glia on intestinal epithelial barrier function was assessed. Basolateral exposure to media from equine submucosal enteric glia culture exposed to 10 ng and 25 ng IL-1β for 24 h significantly increased permeability (p = 0.05 and 0.04, respectively, mean of control media = 0.9, mean of glial 10 ng IL-1β group = 0.8, 95% confidence interval = -0.2 to 0.02, mean of glial 25 ng IL-1β group = 0.8, 95% confidence interval = -0.1995 to 0.01060, n = 3-4) (Figure 4A). However, exposure of equine epithelial monolayers to basolateral 10-25 ng equine IL-6, an inflammatory cytokine demonstrated to both be produced by enteric glia exposed to IL-1β and to induce increased enterocyte permeability in rodent models¹², did not significantly increase equine enterocyte monolayer permeability (Figure 4B)

Figure 1: Isolation of enteric glia from equine jejunum by microdissection and enzymatic and mechanical digestion. (A) The jejunum was cut along the antimesenteric border and pinned mucosa side up. (B) The intestinal villi were visualized by microscopy and removed with gentle scraping until the wispy lamina propria was visualized. (C) The lamina propria layer was then removed until the smooth muscle fibers were just visible through the translucent submucosa. (D) The submucosa was tented and peeled back from the muscular layer, making cuts to fibrous tags as needed. Images were obtained by a 10x dissection microscope (B,C). Please click here to view a larger version of this figure.

Figure 2: Imaging of the equine submucosal enteric glia cultures. (A) The stellate (arrows), fusiform (open arrowhead) and ganglionic (closed arrowhead) morphology of the equine enteric glia was appreciated by phase contrast microscopy. (B) Purity of cultures was assessed by staining for GFAP, and for the fibroblast marker, smooth muscle actin Scale bars = 100 mm (A), 400 mm (B). Abbreviations: GFAP = glial fibrillar acidic protein; DAPI = 4',6-diamidino-2-phenylindole. Please click here to view a larger version of this figure.

Figure 3: Imaging of unsuccessful equine submucosal enteric glia cultures. Purity of cultures was assessed by staining for GFAP (red), and for the fibroblast marker, smooth muscle actin (green), demonstrating a large percentage of fibroblast contamination. Yellow arrowhead denotes GFAP and white arrowhead smooth muscle actin. Scale bar = 400 mm. Abbreviation: GFAP = glial fibrillar acidic protein. Please click here to view a larger version of this figure.

Figure 4: Equine epithelial monolayer TEER following inflammatory cytokine and inflammatory glial products exposure. (A) Basolateral exposure to media from equine submucosal enteric glia culture, exposed to 10 ng and 25 ng of IL-1β for 24 h, significantly increased permeability (*p = 0.05 and 0.04, respectively, n = 3-4). (B) Equine enterocyte monolayer acute exposure to basolateral 25 ng of IL-1β and 10-25 ng of equine IL-6 did not significantly increase equine epithelial monolayer permeability (n = 2-4). Please click here to view a larger version of this figure.

Table 1: Reagents for isolation and culture. The following reagents were utilized to formulate media for equine enteric glia isolation, enteric glial culture, and equine intestinal epithelial stem cell culture. Please click here to download this Table.

Discussion

The aim of this study was to develop a repeatable method of primary culture of equine submucosal enteric glia and demonstrate its application to model epithelial-glial interactions at the time of colic. Enteric glia isolation and culture, which is novel in the horse, has proven beneficial in understanding intestinal disease pathways in pig and rodent models and in humans^6,7,13,14. Studying this cell population is potentially important because of their role as subepithelial sensors of the intestinal environment, as well as their ability to interact with and communicate with the surrounding cells within the intestinal wall. With their keen sensing ability, it is not difficult to imagine a potentially critical role of glia in equine colic, where intestinal tissue is stressed by distension, ischemia, and potentially surgical manipulation⁶.

Previous work by Lisowski et al. demonstrated the heightened inflammatory status of equine intestinal tissue at the time of colic surgery. The importance of mitigating this inflammatory environment is highlighted by rodent enteric glia cultures that have proven to be capable of synthesizing products with the ability to disrupt intestinal barrier function^7,12. Leakage of intestinal luminal contents into the intestinal wall would not only lead to an inflammatory response but also potential infiltration into the circulation, which could then induce endotoxemic shock/sepsis. Equine colic postoperative patients are unfortunately plagued by numerous inflammatory-based postoperative complications, including sepsis and postoperative ileus, that not only reduce survival and patient comfort but also increase client costs and emotional distress³. Establishing a model of epithelial-glial interactions at the time of colic may elucidate a potential pathway of equine intestinal barrier permeability in colic, and possible prevention.

The culture of equine submucosal enteric glia involves adaptations and limitations due to the characteristics of the species. Most importantly, the length of time required for effective enzymatic digestion required titrating. The authors found that incubation with the enzymes for 2-3 h at 37 °C was necessary for the successful isolation of equine enteric glia. In addition, allowing equine enteric cultures to reach over 80% confluence prior to passaging appears to be detrimental to culture viability. The main limitation of this method at this date is the sample size. This study only involved tissue/cells that had been obtained from horses undergoing colic surgery or euthanasia for non-study-related reasons. Continued culture of equine enteric glia from additional subjects, as well as developing a protocol for freezing and storage of equine enteric glia will greatly increase the power of this study.

This study introduces the novel utility of equine submucosal enteric glial culture and primary culture of equine enterocytes as potential methodologies to help understand equine colic and gastrointestinal disease. In addition, while equine epithelial monolayers grown from isolated intestinal crypts and organoids have been well established in the past, this study not only continued to showcase their usefulness as a model but also encourages additional studies examining the interaction with the equine intestinal enterocyte and other cell types found within the intestinal wall.

Observed interactions between these cell types contain potential applications to both colic and other inflammatory gastrointestinal disorders such as inflammatory bowel disease (IBD) and colitis. In addition, continued refinement of culture methods for enteric glia and enterocytes will allow the establishment of coculture studies in the future. Furthermore, while the current study focuses on the submucosal enteric glial population due to its proximity to the epithelial monolayer and the focus on intestinal barrier function, additional work may be performed to culture and assess the function of the equine enteric glial from the myenteric plexi of the intestinal wall, which may aid in determining the impact of inflammatory equine enteric glia on motility¹⁵.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1 M HEPES buffer	Gibco	15630-080
10 mM HEPES	Life Technologies	15630-106
2 mM GlutaMAX	Life Technologies	25050-061
4’6-Diaminidino-2-Phenylindol	Invitrogen	D3571
Advanced DMEM/F12	Life Technologies	12634-010
Alpha smooth muscle actin antibody	Abcam	7817
Amphotericin B	Sigma	AA9529	4.4 g/mL stock aliquots, final concentration 1.1 µg/mL
Anti-Antimicotic 1x	Gibco	15240-096
B27	Gibco	12587010
Bovine Serum Albumin	Sigma	AA3311
BSA 50 mg/mL stock solution	Sigma	A3311
CaCL2	ACROS Organiics	206791000	Component of Equine Ringer ‘s Stock 1: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 2 to make complete “Ringer’s Solution”.
Collagenase	Sigma	9891
DMEM-F12 media	Thermo Fisher	11320033
Donkey anti-rabbit IgG Alexa Fluor 594	Invitrogen	21207
EVOM EndOhm dual electrode TEER-measuring chamber	World Precision Instruments	EVM-EL-03-01
EVOM Manual for TEER Measurement	World Precision Instruments	EVM-MT-03-01
G5	Gibco	17503012
Gentamicin solution	Sigma	G1272	Final concentration 20 µg/mL
GFAP antibody	Abcam	4674
Goat anti-mouse IgG Alexa Fluor 488	Invitrogen	28175
IL-1β ELISA	Thermo Fisher	ESIL1B
KCl	Thermo Fisher	P330-500	Component of Equine Ringer’s Stock 1: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 2 to make complete “Ringer’s Solution”.
L-glutamine solution	Corning	25-00-Cl
Matrigel	BD Bioscience	354277
MgCl2	Thermo Fisher	M33-500	Component of Equine Ringer’s Stock 1: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 2 to make complete “Ringer’s Solution”.
N2	Gibco	17502048
Na2HPO4	Thermo Fisher	BP332-1	Component of Equine Ringer’s Stock 2: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 1 to make complete “Ringer’s Solution”.
NaCl	Thermo Fisher	S271-10	Component of Equine Ringer’s Stock 1: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 2 to make complete “Ringer’s Solution”.
NaH2PO4	Thermo Fisher	BP329-500	Component of Equine Ringer’s Stock 2: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 1 to make complete “Ringer’s Solution”.
NaHCO3	Thermo Fisher	S637-212	Component of Equine Ringer’s Stock 2: combine with other ingredients, then add 100 mL of this stock to a graduated cylinder and dilute to 1L with deionized water. Adjust pH to 7.4 with 5% CO2. Combine with Equine Ringer’s Stock 1 to make complete “Ringer’s Solution”.
Pen/Strep solution	Gemini	400-109
Poly-L-lysine	Sigma	P2636	0.5 mg/mL in 1x borate buffer
Prism software	GraphPad
Protease	Sigma	P4630
Sodium bicarbonate solution	Sigma	S8761	7.5% stock solution