Hassan A. Khalil

Type I Collagen as an Extracellular Matrix for the In Vitro Growth of Human Small Intestinal Epithelium

Background We previously reported in vitro maintenance and proliferation of human small intestinal epithelium using Matrigel, a proprietary basement membrane product. There are concerns over the applicability of Matrigel-based methods for future human therapies. We investigated type I collagen as an alternative for the culture of human intestinal epithelial cells. Methods Human small intestine was procured from fresh surgical pathology specimens. Small intestinal crypts were isolated using EDTA chelation. Intestinal subepithelial myofibroblasts were isolated from a pediatric sample and expanded in vitro. After suspension in Matrigel or type I collagen gel, crypts were co-cultured above a confluent layer of myofibroblasts. Crypts were also grown in monoculture with exposure to myofibroblast conditioned media; these were subsequently sub-cultured in vitro and expanded with a 1∶2 split ratio. Cultures were assessed with light microscopy, RT-PCR, histology, and immunohistochemistry. Results Collagen supported viable human epithelium in vitro for at least one month in primary culture. Sub-cultured epithelium expanded through 12 passages over 60 days. Histologic sections revealed polarized columnar cells, with apical brush borders and basolaterally located nuclei. Collagen-based cultures gave rise to monolayer epithelial sheets at the gel-liquid interface, which were not observed with Matrigel. Immunohistochemical staining identified markers of differentiated intestinal epithelium and myofibroblasts. RT-PCR demonstrated expression of α-smooth muscle actin and vimentin in myofibroblasts and E-Cadherin, CDX2, villin 1, intestinal alkaline phosphatase, chromogranin A, lysozyme, and Lgr5 in epithelial cells. These markers were maintained through several passages. Conclusion Type I collagen gel supports long-term in vitro maintenance and expansion of fully elaborated human intestinal epithelium. Collagen-based methods yield familiar enteroid structures as well as a new pattern of sheet-like growth, and they eliminate the need for Matrigel for in vitro human intestinal epithelial growth. Future research is required to further develop this cell culture system for tissue engineering applications.

Long-term renewable human intestinal epithelial stem cells as monolayers: A potential for clinical use.

PURPOSE Current culture schema for human intestinal stem cells (hISCs) frequently rely on a 3D culture system using Matrigel™, a laminin-rich matrix derived from murine sarcoma that is not suitable for clinical use. We have developed a novel 2D culture system for the in vitro expansion of hISCs as an intestinal epithelial monolayer without the use of Matrigel. METHODS Cadaveric duodenal samples were processed to isolate intestinal crypts from the mucosa. Crypts were cultured on a thin coat of type I collagen or laminin. Intestinal epithelial monolayers were supported with growth factors to promote self-renewal or differentiation of the hISCs. Proliferating monolayers were sub-cultured every 4-5days. RESULTS Intestinal epithelial monolayers were capable of long-term cell renewal. Less differentiated monolayers expressed high levels of gene marker LGR5, while more differentiated monolayers had higher expressions of CDX2, MUC2, LYZ, DEF5, and CHGA. Furthermore, monolayers were capable of passaging into a 3D culture system to generate spheroids and enteroids. CONCLUSION This 2D system is an important step to expand hISCs for further experimental studies and for clinical cell transplantation. LEVEL OF EVIDENCE 1 Experimental.

Isolation of Primary Myofibroblasts from Mouse and Human Colon Tissue

The myofibroblast is a stromal cell of the gastrointestinal (GI) tract that has been gaining considerable attention for its critical role in many GI functions. While several myofibroblast cell lines are commercially available to study these cells in vitro, research results from a cell line exposed to experimental cell culture conditions have inherent limitations due to the overly reductionist nature of the work. Use of primary myofibroblasts offers a great advantage in terms of confirming experimental findings identified in a cell line. Isolation of primary myofibroblasts from an animal model allows for the study of myofibroblasts under conditions that more closely mimic the disease state being studied. Isolation of primary myofibroblasts from human colon tissue provides arguably the most relevant experimental data, since the cells come directly from patients with the underlying disease. We describe a well-established technique that can be utilized to isolate primary myofibroblasts from both mouse and human colon tissue. These isolated cells have been characterized to be alpha-smooth muscle actin and vimentin-positive, and desmin-negative, consistent with subepithelial intestinal myofibroblasts. Primary myofibroblast cells can be grown in cell culture and used for experimental purposes over a limited number of passages.

Colorectal Emergencies: Perforated Diverticulitis (Operative and Nonoperative Management)

Diverticulitis is the most common manifestation of diverticulosis. It accounts for roughly 150,000 hospital admissions in the US per year, resulting in ∼ 50,000 bowel resections. It is defined simply as inflammation due to perforation of a diverticulum. Diverticulosis, defined as the presence of diverticula in the colon, was rare prior to the 1900s. Since that time there has been a significant increase in the incidence of diverticulosis and diverticulitis, across all age groups.[1, 2] This rising incidence coincided with the advent of industrial roller milling in the late 1800s, a process that turned whole grains into refined white flour and led to a significant reduction in fiber consumption in western societies. We currently estimate that roughly 5-10% of the US population has diverticula present in their colon by the age of 40. This incidence increases to 60-70% by the age of 80. Of all patients that have diverticulosis, 10-25% will become symptomatic. Low dietary fiber results in diminished stool bulk, resulting in slower gastrointestinal (GI) transit and changes in colonic motility, leading to elevated regional colonic (intraluminal) pressures as the colon must contract more vigorously to propel stool. Diverticula develop due to these increased intraluminal pressures, forming herniations at areas of weakness where the vasa recta, terminal branches of the marginal artery, penetrate the bowel wall to provide arterial blood to the mucosal layer. Colonic diverticula are pulsion, or false diverticula, composed of mucosa, muscularis mucosa, and peritoneum. The sigmoid colon is the most commonly affected segment of bowel. Other risk factors for the development of diverticulosis include age, obesity, and lack of physical activity. Diverticulitis develops following perforation of a diverticulum. Older theories surmised that diverticulitis was caused by an initial obstruction at the neck of the diverticulum, leading to distention and eventual perforation. This led to the recommendation to avoid certain foods such as popcorn, seeds, and nuts, to minimize the risk that they might get stuck and lead to perforation. This theory has largely fallen out of favor and the current thinking is that diverticulitis simply develops when the colonic pressure exceeds the wall tension of a diverticulum.

A durable model of Hirschsprung's colon.

INTRODUCTION Hirschsprungs disease is characterized by colonic aganglionosis, curable only by surgical correction. Stem cells may offer regenerative benefits while preventing surgical risks. Existing Hirschsprungs model systems are limited by alimentary compromise and spontaneous ganglionic reconstitution. We endeavored to generate a model of permanent colonic aganglionosis to support longitudinal cell therapy studies. METHODS Among adult female Lewis rats (n=11), laparotomy was performed and one-centimeter segments of descending colon were isolated from continuity and denervated by trans-serosal benzalkonium chloride (BAC) exposure. Postoperative weights were plotted. The colon segments were retrieved after 50 or 100days. Immunohistochemical staining (IHC) for beta-III tubulin (TUJ1) and glial fibrillary acid protein (GFAP) revealed colonic ganglia. Muscle layer diameter and the presence of ganglia were contrasted between normal and denervated segments. RESULTS All animals survived, experienced 5% weight loss after one week, and then consistently gained weight. Isolated segments had significantly hypertrophied smooth muscle layers compared to normal colon. Ganglia were identified by IHC in normal colonic segments, and denervated colonic segments had no IHC evidence of myenteric ganglia. CONCLUSION Colonic segmental isolation and denervation result in an effective model of irreversible colonic aganglionosis. Animals retain alimentary function. Muscularis hypertrophy, myenteric denervation, and normal animal longevity are suitable for long-term studies of cell therapy.

Bioengineering functional smooth muscle with spontaneous rhythmic contraction in vitro

Oriented smooth muscle layers in the intestine contract rhythmically due to the action of interstitial cells of Cajal (ICC) that serve as pacemakers of the intestine. Disruption of ICC networks has been reported in various intestinal motility disorders, which limit the quality and expectancy of life. A significant challenge in intestinal smooth muscle engineering is the rapid loss of function in cultured ICC and smooth muscle cells (SMC). Here we demonstrate a novel approach to maintain the function of both ICC and SMC in vitro. Primary intestinal SMC mixtures cultured on feeder cells seeded electrospun poly(3-caprolactone) scaffolds exhibited rhythmic contractions with directionality for over 10 weeks in vitro. The simplicity of this system should allow for wide usage in research on intestinal motility disorders and tissue engineering, and may prove to be a versatile platform for generating other types of functional SMC in vitro.

Primary Myofibroblasts Maintain Short-Term Viability following Submucosal Injection in Syngeneic, Immune-Competent Mice Utilizing Murine Colonoscopy.

The myofibroblast is an important stromal cell of the gastrointestinal tract. Current in vitro and in vivo models either do not accurately recreate stromal-epithelial interactions or are not specific to myofibroblasts. We sought to create an animal model that would allow the study of myofibroblast-epithelial interactions. We isolated and cultured colonic myofibroblasts from FVB mice. Cells were α-SMA and vimentin positive but desmin negative on immunoblot analysis. We injected the myofibroblasts into the colonic submucosa of syngeneic adult mice (n = 8) via a miniendoscopic system. We then isolated green fluorescent protein (GFP) positive colonic myofibroblasts from C57BL/6-Tg(CAG-EGFP)1Osb/J mice and injected them into the colonic lamina propria of C57BL/6J mice at 1x105 (n = 14), 1x106 (n = 9), or 5x106 cells/mL (n = 4). A subset of mice were injected with serum-free media and ink without cells (n = 3). Mice underwent repeat endoscopy and euthanasia one or 7 days after injection. Colons were isolated and either fixed in 10% formalin or the inked sites were individually excised and lysed for DNA. We assessed the injection sites via histology and immunohistochemical stains for α-SMA and GFP. We used qPCR to quantify GFP DNA transcripts at the lamina propria injection sites. Submucosal injection of myofibroblasts resulted in the formation of a subepithelial wheal on endoscopy, which persisted to day 7. Myofibroblasts injected either into the submucosa or lamina propria maintained viability on post-injection day 7 as evidenced by positive α-SMA staining. qPCR of lamina propria injections showed a dose-dependent increase in GFP DNA transcripts on post-injection day 1, whereas the number of transcripts on day 7 was equivalent for the concentrations injected. We demonstrate short-term survival of primary cultured colonic myofibroblasts in syngeneic mice. This may prove to be a valuable model for studying the role of myofibroblasts in states of health and disease.

Sa1711 Primary Myofibroblasts Maintain Short-Term Viability Following Submucosal Injection in Syngeneic, Immune-Competent Mice Utilizing Murine Colonoscopy

Backround: Myofibroblasts are a subpopulation of stromal cells that are believed to play an important regulatory role in inflammation and cancer in the GI tract. While the inflammatory cell signaling mechanisms that regulate myofibroblast function are being elucidated in vitro, few in vivo models exist to directly study the impact of myofibroblast signal modulation on the overlying epithelium. We report our early experience with a novel technique utilizing murine colonoscopy that may allow for the real-time in vivo study of myofibroblast-epithelial cell interactions in their natural tissue environment of the colon. Methods: Primary colonic myofibroblasts (MFB) were harvested from adult FVB mice and were grown in cell culture. MFB were then suspended in serum-free media (1x103 cells/50μL, 1x104 cells/50μL), mixed with Spot® ink, and were then re-implanted endoscopically in anesthetized immunecompetent syngeneic mice by submucosal injection approximately 4 cm from the anal verge by utilizing small animal colonoscopy (Karl Storz Coloview miniendoscopic system). Mice were put back in their housing and were fed a normal diet. A repeat colonoscopy was performed on post-procedure days 2 and 7. Mice were sacrificed on day 7 and histologic analysis was performed. Results: Submucosal injection of primary myofibroblasts is technically feasible, visualized endoscopically by the appearance of a transient mucosal wheal, and confirmed histologically. Follow up colonoscopy on post-procedure day 2 confirmed that the injection site can be reliably identified and the overlying mucosa appeared intact. Mice tolerate multiple endoscopic procedures and multiple submucosal injections without obvious sequelae. Colonoscopic findings on day 7 were notable for an intact-appearing mucosa with an associated bump at the inked site. Mice were sacrificed on day 7 and histologic analysis suggests that the injected myofibroblasts maintain viability, and do not illicit an inflammatory response or disrupt the normal histologic architecture of the colon wall. Conclusions: Submucosal injection of primary myofibroblasts using murine colonoscopy is technically feasible and well tolerated in mice. The injection site can be easily re-identified, serially evaluated, and directly compared to adjacent tissue under the same experimental conditions. Submucosally-injected primary myofibroblasts maintain short-term viability with minimal disruption to the normal colon architecture of an immune-competent, syngeneic mouse. Successful development of this technique may create a novel platform for the in vivo study of stromal-epithelial cell interactions that is applicable to any existing mouse model.