Endothelial cells drive organ fibrosis in mice by inducing expression of the transcription factor SOX9.

Fibrosis is a hallmark of chronic disease. Although fibroblasts are involved, it is unclear to what extent endothelial cells also might contribute. We detected increased expression of the transcription factor Sox9 in endothelial cells in several different mouse fibrosis models. These models included systolic heart failure induced by pressure overload, diastolic heart failure induced by high-fat diet and nitric oxide synthase inhibition, pulmonary fibrosis induced by bleomycin treatment, and liver fibrosis due to a choline-deficient diet. We also observed up-regulation of endothelial SOX9 in cardiac tissue from patients with heart failure. To test whether SOX9 induction was sufficient to cause disease, we generated mice with endothelial cell-specific overexpression of Sox9, which promoted fibrosis in multiple organs and resulted in signs of heart failure. Endothelial Sox9 deletion prevented fibrosis and organ dysfunction in the two mouse models of heart failure as well as in the lung and liver fibrosis mouse models. Bulk and single-cell RNA sequencing of mouse endothelial cells across multiple vascular beds revealed that SOX9 induced extracellular matrix, growth factor, and inflammatory gene expression, leading to matrix deposition by endothelial cells. Moreover, mouse endothelial cells activated neighboring fibroblasts that then migrated and deposited matrix in response to SOX9, a process partly mediated by the secreted growth factor CCN2, a direct SOX9 target; endothelial cell-specific Sox9 deletion reversed these changes. These findings suggest a role for endothelial SOX9 as a fibrosis-promoting factor in different mouse organs during disease and imply that endothelial cells are an important regulator of fibrosis.

Impact of cryopreservation on viability, gene expression and function of enteric nervous system derived neurospheres.

Introduction: Impairment of both the central and peripheral nervous system is a major cause of mortality and disability. It varies from an affection of the brain to various types of enteric dysganglionosis. Congenital enteric dysganglionosis is characterized by the local absence of intrinsic innervation due to deficits in either migration, proliferation or differentiation of neural stem cells. Despite surgery, children's quality of life is reduced. Neural stem cell transplantation seems a promising therapeutic approach, requiring huge amounts of cells and multiple approaches to fully colonize the diseased areas completely. A combination of successful expansion and storage of neural stem cells is needed until a sufficient amount of cells is generated. This must be combined with suitable cell transplantation strategies, that cover all the area affected. Cryopreservation provides the possibility to store cells for long time, unfortunately with side effects, i.e., upon vitality. Methods: In this study we investigate the impact of different freezing and thawing protocols (M1-M4) upon enteric neural stem cell survival, protein and gene expression, and cell function. Results: Freezing enteric nervous system derived neurospheres (ENSdN) following slow-freezing protocols (M1-3) resulted in higher survival rates than flash-freezing (M4). RNA expression profiles were least affected by freezing protocols M1/2, whereas the protein expression of ENSdN remained unchanged after treatment with protocol M1 only. Cells treated with the most promising freezing protocol (M1, slow freezing in fetal calf serum plus 10% DMSO) were subsequently investigated using single-cell calcium imaging. Freezing of ENSdN did not alter the increase in intracellular calcium in response to a specific set of stimuli. Single cells could be assigned to functional subgroups according to response patterns and a significant shift towards cells responding to nicotine was observed after freezing. Discussion: The results demonstrate that cryopreservation of ENSdN is possible with reduced viability, only slight changes in protein/gene expression patterns and without an impact on the neuronal function of different enteric nervous system cell subtypes, with the exception of a subtle upregulation of cells expressing nicotinergic acetylcholine receptors. In summary, cryopreservation presents a good method to store sufficient amounts of enteric neural stem cells without neuronal impairment, in order to enable subsequent transplantation of cells into compromised tissues.

Degradation of intestinal mRNA: a matter of treatment.

AIM: To characterize the influence of location, species and treatment upon RNA degradation in tissue samples from the gastrointestinal tract. METHODS: The intestinal samples were stored in different medium for different times under varying conditions: different species (human and rat), varying temperature (storage on crushed ice or room temperature), time point of dissection of the submucous-mucous layer from the smooth muscle (before or after storage), different rinsing methods (rinsing with Medium, PBS, RNALater or without rinsing at all) and different regions of the gut (proximal and distal small intestine, caecum, colon and rectum). The total RNA from different parts of the gut (rat: proximal and distal small intestine, caecum, colon and rectum, human: colon and rectum) and individual gut layers (muscle and submucosal/mucosal) was extracted. The quality of the RNA was assessed by micro capillary electrophoresis. The RNA quality was expressed by the RNA integrity number which is calculated from the relative height and area of the 18 S and 28 S RNA peaks. From rat distal small intestine qPCR was performed for neuronal and glial markers. RESULTS: RNA obtained from smooth muscle tissue is much longer stable than those from submucosal/mucosal tissue. At RT muscle RNA degrades after one day, on ice it is stable at least three days. Cleaning and separation of gut layers before storage and use of RNALater, maintains the stability of muscle RNA at RT for much longer periods. Different parts of the gut show varying degradation periods. RNA obtained from the submucosal/mucosal layer always showed a much worse amplification rate than RNA from muscle tissue. In general RNA harvested from rat tissue, either smooth muscle layer or submucosal/mucosal layer is much longer stable than RNA from human gut tissue, and RNA obtained from smooth muscle tissue shows an increased stability compared to RNA from submucosal/mucosal tissue. At RT muscle RNA degrades after one day, while the stability on ice lasts at least three days. Cleaning and separation of gut layers before storage and use of RNALater, maintains the stability of muscle RNA at RT for much longer periods. Different parts of the gut show varying degradation periods. The RNA from muscle and submucosal/mucosal tissue of the proximal small intestine degrades much faster than the RNA of distal small intestine, caecum or colon with rectum. RNA obtained from the submucosal/mucosal layer always showed a much more reduced amplification rate than RNA from muscle tissue [ß-Tubulin III for muscle quantification cycle (Cp): 22.07 ± 0.25, for ß-Tubulin III submucosal/mucosal Cp: 27.42 ± 0.19]. CONCLUSION: Degradation of intestinal mRNA depends on preparation and storage conditions of the tissue. Cooling, rinsing and separating of intestinal tissue reduce the degradation of mRNA.

The human gastrointestinal tract, a potential autologous neural stem cell source.

Stem cell therapies seem to be an appropriate tool for the treatment of a variety of diseases, especially when a substantial cell loss leads to a severe clinical impact. This is the case in most neuronal cell losses. Unfortunately, adequate neural stem cell sources are hard to find and current alternatives, such as induced programmed stem cells, still have incalculable risks. Evidence of neurogenesis in the adult human enteric nervous system brought up a new perspective. In humans the appendix harbors enteric neuronal tissue and is an ideal location where the presence of neural stem cells is combined with a minimal invasive accessibility. In this study appendices from adults and children were investigated concerning their neural stem cell potential. From each appendix tissue samples were collected, and processed for immunohistochemistry or enteric neural progenitor cell generation. Free-floating enteric neurospheres (EnNS's) could be generated after 6 days in vitro. EnNS's were either used for transplantation into rat brain slices or differentiation experiments. Both transplanted spheres and control cultures developed an intricate network with glia, neurons and interconnecting fibers, as seen in primary enteric cultures before. Neuronal, glial and neural stem cell markers could be identified both in vitro and in vivo by immunostaining. The study underlines the potential of the enteric nervous system as an autologous neural stem cell source. Using the appendix as a potential target opens up a new perspective that might lead to a relatively unproblematic harvest of neural stem cells.

FGF2 deficit during development leads to specific neuronal cell loss in the enteric nervous system.

The largest part of the peripheral nervous system is the enteric nervous system (ENS). It consists of an intricate network of several enteric neuronal subclasses with distinct phenotypes and functions within the gut wall. The generation of these enteric phenotypes is dependent upon appropriate neurotrophic support during development. Glial cell line-derived neurotrophic factor (GDNF) and fibroblast growth factor-2 (FGF2) play an important role in the differentiation and function of the ENS. A lack of GDNF or its receptor (Ret) causes intestinal aganglionosis in mice, while fibroblast growth factor receptor signaling antagonist is identified as regulating proteins in the GDNF/Ret signaling in the developing ENS. Primary myenteric plexus cultures and wholemount preparations of wild type (WT) and FGF2-knockout mice were used to analyze distinct enteric subpopulations. Fractal dimension (D) as a measure of self-similarity is an excellent tool to analyze complex geometric shape and was applied to classify the subclasses of enteric neurons concerning their individual morphology. As a consequence of a detailed analysis of subpopulation variations, wholemount preparations were stained for the calcium binding proteins calbindin and calretinin. The fractal analysis showed a reliable consistence of subgroups with different fractal dimensions (D) in each culture investigated. Seven different neuronal subtypes could be differentiated according to a rising D. Within the same D, the neurite length revealed significant differences between wild type and FGF2-knockout cultures, while the subclass distribution was also altered. Depending on the morphological characteristics, the reduced subgroup was supposed to be a secretomotor neuronal type, which could be confirmed by calbindin and calretinin staining of the wholemount preparations. These revealed a reduction up to 40 % of calbindin-positive neurons in the FGF2-knockout mouse. We therefore consider FGF2 playing a more important role in the fine-tuning of the ENS during development as previously assumed.

Anastomotic sealing with a fibrin-coated collagen patch in small-diameter bowel.

PURPOSE: The aim of this study was to evaluate the complication rates and inflammatory response in TachoSil™-sealed small-diameter anastomoses with conventional and reduced suture number as a model for neonatal bowel surgery. METHODS: Ileo-ileal anastomoses were performed in 73 rats. In the control group, the anastomosis was accomplished with the conventional technique, using nine interrupted sutures. In the other groups with nine, six, and three interrupted sutures, the anastomotic line was additionally sealed with a fibrin-coated collagen patch (TachoSil™). The rats were sacrificed on days 0, 2, and 10. Clinical and functional parameters included the rates of ileus, insufficiency and death, operating time, adhesions, bursting pressure, and preanastomotic dilatation. The histological examination of the anastomoses concentrated on assessing the inflammatory cell infiltration of the TachoSil™ patch and the intestinal wall. RESULTS: Severe preanastomotic dilatation was observed in additionally sealed ileo-ileal anastomoses with conventional suture number and high complication rates (ileus, perforation, death) occurred in additionally sealed anastomoses with reduced suture number. We found a massive microabscess-forming inflammation in additionally sealed anastomoses. Inflammatory cell infiltration was highest in the collagen layer of the sealing patch (p < 0.05 vs. fibrin layer of the sealing patch and vs. intestinal wall). CONCLUSIONS: As a result of our findings, additional sealing of small-diameter intestinal anastomoses with TachoSil™ cannot be recommended.

Expression of intermediate filament proteins and neuronal markers in the human fetal gut.

The human enteric nervous system (ENS) derives from migrating neural crest cells (NCC) and is structured into different plexuses embedded in the gastrointestinal tract wall. During development of the NCC, a rearrangement of various cytoskeletal intermediate filaments such as nestin, peripherin, or alpha-internexin takes place. Although all are related to developing neurons, nestin is also used to identify neural stem cells. Until now, information about the prenatal development of the human ENS has been very restricted, especially concerning potential stem cells. In this study the expression of nestin, peripherin, and alpha-internexin, but also of neuronal markers such as protein gene product (PGP) 9.5 and tyrosine hydroxylase, were investigated in human fetal and postnatal gut. The tissue samples were rapidly removed and subsequently processed for immunohistochemistry or immunoblotting. Nestin could be detected in all samples investigated with the exception of the 9th and the 12th week of gestation (WOG). Although the neuronal marker PGP9.5 was coexpressed with nestin at the 14th WOG, this could no longer be observed at later time points. Alpha-internexin and peripherin expression also did not appear before the 14th WOG, where they were coexpressed with PGP9.5. This study reveals that the intermediate filament markers investigated are not suitable to detect early neural crest stem cells.