Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Aug 1.
Published in final edited form as: Biomaterials. 2014 Jun 11;35(26):7429–7440. doi: 10.1016/j.biomaterials.2014.05.037

The influence of extracellular matrix composition on the differentiation of neuronal subtypes in tissue engineered innervated intestinal smooth muscle sheets

Shreya Raghavan 1,2, Khalil N Bitar 1,2
PMCID: PMC4086147  NIHMSID: NIHMS600703  PMID: 24929617

Abstract

Differentiation of enteric neural stem cells into several appropriate neural phenotypes is crucial while considering transplantation as a cellular therapy to treat enteric neuropathies. We describe the formation of tissue engineered innervated sheets, where intestinal smooth muscle and enteric neuronal progenitor cells are brought into close association in extracellular matrix (ECM) based microenvironments. Uniaxial alignment of constituent smooth muscle cells was achieved by substrate microtopography. The smooth muscle component of the tissue engineered sheets maintained a contractile phenotype irrespective of the ECM composition, and generated equivalent contractions in response to potassium chloride stimulation, similar to native intestinal tissue. We provided enteric neuronal progenitor cells with permissive ECM-based compositional and viscoelastic cues to generate excitatory and inhibitory neuronal subtypes. In the presence of the smooth muscle cells, the enteric neuronal progenitor cells differentiated to functionally innervate the smooth muscle. The differentiation of specific neuronal subtypes was influenced by the ECM microenvironment, namely combinations of collagen-I, collagen-IV, laminin and/or heparan sulfate. The physiology of differentiated neurons within tissue engineered sheets was evaluated. Sheets with composite collagen and laminin had the most similar patterns of Acetylcholine-induced contraction to native intestinal tissue, corresponding to an increased protein expression of choline acetyltransferase. An enriched nitrergic neuronal population, evidenced by an increased expression of neuronal nitric oxide synthase, was obtained in tissue engineered sheets that included collagen-IV. These sheets had a significantly increased magnitude of electrical field stimulated relaxation, sensitive maximally to nitric oxide synthase inhibition. Tissue engineered sheets containing laminin and/or heparan sulfate had a balanced expression of contractile and relaxant motor neurons. Our studies demonstrated that neuronal subtype was modulated by varying ECM composition. This observation could be utilized to derive enriched populations of specific enteric neurons in vitro prior to transplantation.

Keywords: Neural stem cell, neural tissue engineering, extracellular matrix, adult stem cell, enteric nervous system

1. Introduction

An uninterrupted enteric nervous system with the preservation of myenteric ganglia is required for intestinal motility and function [1, 2]. Motor neurons of the myenteric ganglia pre-dominantly express acetylcholine/tachykinins (excitatory) or nitric oxide/inhibitory peptides/purines (inhibitory) to mediate smooth muscle contraction and relaxation [3, 4]. Partial, selective or total loss of neurons is reported in several disorders including, but not limited to, Hirschsprung's disease, achalasia, and inflammation [5-8]. Neural-crest derived enteric neuronal progenitor cells have been isolated from adult mammalian guts, including ganglionic bowel of patients with Hirschsprung's disease [9-12]. These cells have the ability to differentiate into neuronal and glial phenotypes [13-15]. However, there is little information and understanding of microenvironment driven differentiation and limited studies describing subsequent functional behavior of these differentiated neurons in vitro [10, 11].

The ECM has been long known to provide permissive and non-permissive environmental cues for migration and differentiation of neuronal progenitor cells [16]. Collagen IV, laminin and heparan sulfate proteoglycans play an important role in guiding the migration of neural-crest derived cells and formation of the enteric nervous system [17-19]. Moreover, myenteric ganglia are surrounded by an ECM composed of type IV Collagen, laminin and a heparan sulfate proteoglycan [20].

Type I fibrillar collagen has favorable biological properties and has been widely used in neural tissue engineering [21, 22]. Collagen IV has been demonstrated to promote neurite outgrowth in neuroepithelial progenitor cells and in sympathetic peripheral neurons in vitro [23, 24]. Collagen IV aids the colonization of the embryonic gut and modulates selective neurotrophic signaling [20, 25]. The IKVAV peptide in laminin interacts with a neurally acquired receptor on post-migratory neural crest stem cells promoting their differentiation [26]. In the gut, heparan sulfate proteoglycan is required for GDNF signaling [27]. Glycosaminoglycan – growth factor interactions additionally stabilize and increase local bioavailability of growth factors, thus inducing neurite outgrowth [28]. The ECM can provide a background upon which cell-cell and cell-matrix signaling can work to regulate phenotypes of differentiating enteric neuronal progenitor cells.

In previous studies using 2D culture substrata, we demonstrated that varying ECM composition of 2D culture substrata influenced neuroglial differentiation of adult enteric neuronal progenitor cells [29]. The number of neurons, neurite lengths and preliminary neuronal network formation were all enhanced in culture substrata that contained collagen IV, laminin and heparan sulfate. The previous studies, however, were carried out on 2D coated glass coverslips. Moreover, we examined neuronal and glial differentiation, without documenting the phenotype or functionality of the differentiated neurons. In the present studies, we utilize tissue engineering as a tool to provide enteric neuronal progenitor cells with 3D viscoelastic ECM microenvironments. Given that the cues for differentiation arising from a contractile phenotype of smooth muscle cells would be constant, we hypothesized that ECM composition could differentially regulate the generation of specific neuronal subsets. The objective of the current undertaking was to determine the presence of differentiated motor neurons obtained and investigate their ability in mediating smooth muscle contraction/relaxation.

We utilize a previously described tissue engineered model of intestinal longitudinal smooth muscle sheets, where uniaxial alignment of a smooth muscle monolayer was facilitated by the use of substrate microtopography [30]. We innervated these tissue engineered sheets using enteric neuronal progenitor cells embedded with hydrogels of varying ECM composition. Differentiated neuronal composition (cholinergic, nitrergic, peptidergic) within tissue engineered sheets was evaluated. Functional neuronal physiology mediating smooth muscle contraction/relaxation was also evaluated in the tissue engineered sheets using real-time force generation measurements.

2. Materials and Methods

2.1 Materials

All tissue culture reagents (including media, supplements, and natural mouse laminin) were purchased from Invitrogen (Carlsbad, CA). Rat tail type I collagen and mouse collagen type IV were purchased from BD Biosciences (Bedford, MA). Heparan sulfate was purchased from Celsus (Cincinnati, OH). Growth factors were purchased from Stemgent (Cambridge, MA). All primary antibodies were purchased from Abcam (Cambridge, MA), unless specified otherwise.

2.2 Isolation and primary culture from adult rabbit GI tissues

Enteric neuronal progenitor cells were isolated from jejunal biopsies of adult New Zealand White rabbits using a collagenase/dispase digestion technique and cultured in neuronal growth media, as described previously [29]. Cells aggregated to form enteric neurospheres. Longitudinal smooth muscle cells were isolated from the adult rabbit sigmoid colon as described previously [30]. Isolated longitudinal smooth muscle cells were expanded in culture until confluency.

2.3 Composition and characterization of ECM hydrogels

ECM hydrogels were made with the following components:

  1. Collagen I gels (800-1600μg/ml);

  2. Collagen I (800μg/ml) and Collagen IV (200μg/ml) gels composite gels;

  3. Collagen I and Collagen IV with laminin (5-10μg/ml);

  4. Collagen I and Collagen IV with laminin and heparan sulfate (10-20μg/ml).

Other components of the gel included: 1% fetal calf serum, 0.1X antibiotics in Dulbecco's modified Eagle's medium. 0.1N Sodium hydroxide was used to adjust pH to ~7.4 for gelation.

2.3.1 Rheological characterization of ECM hydrogels

Oscillatory rheometry (ATS RheoSystems) was used to measure viscoelastic moduli of ECM gels. 20mm parallel base plates were used to perform a stress sweep of the sample at 1Hz. ECM gels were allowed to gel in situ between the parallel plates at 37°C. The viscoelastic modulus was obtained from a linear region of the stress-strain curve, at strains lower than 10%, within the sensitivity ranges for torque and strain of the rheometer. 3-5 individually manufactured ECM gels were measured to determine an average viscoelastic modulus. Compositions that resulted in a matrix viscoelasticity within the range of 150-300Pa were utilized for further experimentation, so as not to let stiffness be a variable in influencing neuroglial differentiation.

2.3.2 Characterization of ultrastructure of ECM hydrogels

Sample preparation of ECM hydrogels for scanning electron microscopy was adapted from Stuart et al. [31]. Gels were dehydrated through graded ethanol (10% to 100%). Hydrogels were dried at critical point using carbon dioxide exchange. The resulting dehydrated ECM discs were mounted onto metallic stubs with conducting carbon tape, sputter coated with gold, and visualized using an AMRAY 1910 Field Emission Scanning Electron Microscope. Constant working distance and magnification were maintained to image all samples. NIH Image J was used to measure and compare fiber diameters. Porosity was determined using Image J from micrographs obtained from at least three-independent samples of dehydrated ECM gels.

2.4 Tissue engineering innervated intestinal smooth muscle sheets

The tissue engineering process was adapted from Raghavan et al. [30]. Briefly, 500,000 longitudinal smooth muscle cells were aligned uniaxially for 4 days on 35mm diameter circular Sylgard molds containing wavy microtopographies. Enteric neurospheres were treated with Accutase to obtain single cell suspensions. 200,000 cells were resuspended in the appropriate ECM solution and overlaid on the aligned smooth muscle monolayer. Upon gelation, neuronal differentiation medium (neurobasal-A) was added, supplemented with B27 and 1% fetal bovine serum. Differentiation medium was exchanged every second day. Enteric neuronal progenitor cells were allowed to differentiate within the hydrogel for a period of 10 days. Smooth muscle cells compacted the ECM hydrogel over the next 10 days, forming ~1cm long innervated smooth muscle sheets, anchored between silk sutures. Phase microscopy was used to image neuronal differentiation at the edge of the tissue engineered sheets.

2.5 Biochemical characterization of neuroglial composition in tissue engineered sheets

At day 10, tissue engineered sheets were harvested in radioimmunoprecipitation buffer to isolate protein. Protein concentration was estimated spectrophotometrically using the Bradford assay. 20μg of protein from each sample was resolved electrophoretically and transferred to polyvinylidene difluoride membranes. Membranes were blotted with antibodies for neuronal βIII Tubulin, neuronal nitric oxide synthase (nNOS), choline acetyltransferase (ChAT), and Smoothelin. β-Actin was used to confirm equal loading. HRP-conjugated secondary antibodies were used to visualize proteins using enhanced chemiluminescence.

2.6 Immunohistochemical characterization of neuron composition in tissue engineered sheets

Tissue engineered sheets were fixed in 4% formaldehyde and washed extensively in glycine buffer. Immunohistochemical staining was performed following previously established protocols utilized for staining differentiated neurons within bioengineered tissues [32]. Sheets were blocked with 10% horse serum and permeabilized in 0.15% Triton-X for 45 minutes. Permeabilized sheets were incubated with primary antibodies directed against Vasoactive Intestinal Peptide (VIP), ChAT and nNOS for 60 minutes at room temperature. Following antibody incubation, sheets were washed three times with phosphate buffered saline, pH 7.4. Tissue engineered sheets were incubated with appropriate fluorophore conjugated secondary antibodies for 45 minutes, washed in phosphate buffered saline and imaged using an inverted fluorescence microscopy (Nikon Ti-E, Japan). For a negative control, incubation with the primary antibody was skipped, and only fluorophore conjugated secondary antibodies were used to visualize background fluorescence.

2.7 Measurement of physiological function in innervated tissue engineered sheets

Myogenic and neuronal functionality were assessed using real-time force generation as previously described [30, 33]. 4-5 individual tissue engineered sheets for each ECM composition were tested. Tissue engineered sheets were anchored between a stationary pin and measuring pin of a force transducer (Harvard Apparatus, Holliston MA) at 0% stretch. The organ bath maintained temperature at 37°C. An additional 10% stretch was applied using a vernier control. Tissues were immersed in 4ml of medium, which was exchanged at the end of every experiment following a brief wash with fresh medium. Peak contraction or maximal relaxation was quantified following pharmacological or electrical stimuli, and compared between tissue engineered sheets with varying ECM compositions.

Before each treatment, tissues were washed in fresh warm medium and allowed to equilibrate to a baseline. The following stimuli were used independently to assess physiological functionality of the tissue engineered sheets: 1) 60mM Potassium chloride to assess electromechanical coupling integrity of the smooth muscle; 2) 1μM Acetylcholine (contractile agonist); 3) Electrical field stimulation (5Hz, 0.5ms, 40V) applied using parallel plate platinum electrodes. Preincubation with neuronal blocker, tetrodotoxin (TTX) was used to dissect myogenic and neuronal components of contraction/relaxation. Pre-incubation with specific inhibitors were used to identify functional neuronal subtypes: 1) nNOS-blocker Nω-Nitro-L-arginine methyl ester hydrochloride (L-NAME; 300μM); and 2) VIP-receptor antagonist [D-p-Cl-Phe6, Leu17]-Vasoactive Intestinal Peptide (VIP-Ra; 2μM). Following stimulation and subsequent contraction/relaxation and recovery, tissues were washed with fresh medium, and allowed to re-establish a baseline before the next treatment. Equilibrated baseline was arbitrarily set to zero, to measure contraction/relaxation due to a stimulus.

2.8 Data Analysis and statistical methods

Densitometry on western blots was performed using BioRad Quantity One (Hercules, CA). Raw data was acquired from the force transducer at 1000 samples/second. Second order Savitsky-Golay smoothing was applied to data using GraphPad Prism 5.0 for Windows (GraphPad Software, San Diego, CA). Area under the curve (AUC) was measured from the time of addition of pharmacological agonist/electrical field to the end of the contraction/relaxation response. Extent of inhibition by pharmacological inhibitors was calculated by expressing the AUC of contraction/relaxation in the presence of the inhibitor as a percentage of the AUC in the absence of the inhibitor. One way ANOVAs with Tukey post-tests were used to compare means using GraphPad Prism. p ≤ 0.05 was considered significant. Physiological evaluation and densitometry was carried out between 3-5 tissue engineered sheets within each experimental set; all values are expressed as mean ± SEM.

3. Results

3.1 Ultrastructure and viscoelastic properties of ECM hydrogels

All compositions of ECM hydrogels gelled at 37°C within 30 minutes. Scanning electron micrographs revealed a fibrous structure in type I Collagen gels (Fig 1A). The fibers were randomly oriented, with diameters averaging at 478.3 ± 19.31μm. With the addition of type IV Collagen, network-like structures were observed (Fig 1B). Cables of fibers within the networked structures were thicker, with average diameters of 714.8 ± 36.67μm. Addition of laminin to the hydrogels did not alter the ultrastructure or the networked suprastructure (Fig 1C). With the addition of heparan sulfate, the fibers within the networked structures were pulled more tightly together and cabled (Fig 1D). The dehydrated ECM gels displayed a porous appearance, with average porosity ranging from 40.77% - 43.95% (Table 1).

Figure 1. Scanning electron micrographs of dehydrated ECM gels.

Figure 1

Open in a new tab

Images were obtained at constant magnification and constant working distance. (A) Collagen I fibers were randomly oriented, and presented a dense fibrous structure; (B) Composite Collagen I/IV sheets demonstrated evidence of the formation of network-like structures; (C) There was no difference in ultrastructure with the addition of laminin; (D) Evidence of cabling and cross-linking was observed with the addition of heparan sulfate. Average porosity was determined and summarized in Table 1. Viscoelastic modulus was calculated using oscillatory rheometry of ECM gels in their hydrated state, and tabulated in Table 1. Scale bar 10μm.

Table 1.

ECM composition Viscoelastic modulus (Pa) Porosity (%)
Col1 182.3 ± 2.6 43.95 ± 0.47
Col1/Col4 236 ± 13.5 43.95 ± 0.79
Col1/4/Laminin 220.7 ± 16.27 42.28 ± 0.51
Col1/4/Laminin/Heparan sulfate 287 ± 20.11 40.77 ± 0.29
Open in a new tab

Viscoelastic moduli were measured in hydrated ECM gels using oscillatory rheometry. Type I Collagen gels had increasing viscoelastic moduli with increasing collagen concentration ranging from 72.6 ± 4.86Pa (800μg/ml) to 182.3 ± 2.6 (1600μg/ml) to 424 ± 2 Pa (3200μg/ml). The addition of 200μg/ml collagen IV to 800μg/ml collagen I increased the modulus of the gels to 236 ± 13.53 Pa. The addition of laminin had no effect on viscoelastic moduli (compare 236 ± 13.53 Pa to 220.7 ± 16.27 Pa). 10μg/ml of heparan sulfate caused an increase in the modulus of ECM hydrogels (287 ± 20.11 Pa, p<0.05). Table 1 summarizes that the final ECM gels evaluated had viscoelastic moduli ranging from 182Pa to 287Pa.

3.2 Neuronal differentiation in engineered innervated intestinal smooth muscle sheets

Uniaxially aligned smooth muscle cells compacted overlaying ECM hydrogels over 10 days in culture as described before [30]. The resultant tissue engineered sheets were ~1cm long, and a few cell layers thick. In the presence of smooth muscle, the enteric neuronal progenitor cells differentiated within the ECM hydrogel. Neuronal differentiation was identified morphologically by microscopic examination at day 10, demonstrating similar differentiation profiles expressed by enteric neuronal progenitor cells, both in vitro [29] and in tissue engineered constructs [32]. Several differentiated neurons were observed in tissue engineered sheets, regardless of the ECM composition (Fig 2). Arrows in the figures indicate numerous instances of neuronal clustering and preliminary neuronal networking.

Figure 2. Neuronal differentiation within tissue engineered sheets.

Figure 2

Open in a new tab

Phase contrast micrographs were obtained at the edge of the tissue engineered sheets. Evidence of neuronal differentiation and initiation of network formation was observed in all ECM compositions at day 10. The arrows indicate instances of preliminary neuronal networking. Scale bar 200μm.

3.3 Neuronal composition in engineered innervated intestinal smooth muscle sheets

Immunoblotting was used to assess neuronal composition within tissue engineered innervated intestinal smooth muscle sheets. Blotting for β-actin demonstrated that equal amounts of protein were assayed. Representative blots for each protein are shown, indicating the approximate molecular weight at which they appear on the gels (Fig 3E). Contractile phenotype of constituent smooth muscle was demonstrated by the similar expression of smoothelin, within the tissue engineered sheets (Fig 3B). The expression of Smoothelin was constant, regardless of the ECM composition of the sheets, indicating that the constituent smooth muscle cells maintained a contractile phenotype.

Figure 3. Immunoblot analysis of tissue engineered longitudinal sheets.

Figure 3

Open in a new tab

Sheets were assessed for expression of neuronal differentiation, constituent smooth muscle phenotype, and excitatory and inhibitory neural markers. Densitometry was used to quantify band intensities, to quantify and compare expression. (A) Neuronal βIII Tubulin expression was similar amongst all four matrices suggesting that all ECM compositions supported neuronal differentiation; (B) Constituent smooth muscle within the tissue engineered sheets maintained contractile phenotype, demonstrated by similar Smoothelin expression; (C) Choline acetyltransferase (ChAT) expression was significantly (*p<0.05) elevated in Col I and Col I/IV/Laminin sheets compared to Col I/IV and Col I/IV/Lam/HS sheets; (D) Neuronal nitric oxide synthase (nNOS) expression was significantly lower (**p<0.001) in Col I sheets compared to elevated levels in all tissue engineered sheets containing Col4 with or without laminin and/or heparan sulfate. (E) Representative immunoblots are provided along with β Actin, demonstrating equal loading.

3.3.1 Neuronal differentiation

Pan neuronal marker βIII Tubulin expression was similar amongst all tissue engineered sheets, despite the ECM composition (Fig 3A). This suggested that irrespective of the ECM composition, neuronal differentiation of enteric neurospheres proceeded similarly in the presence of smooth muscle cells. βIII Tubulin expression ranged from 21.65± 1.43AU – 28.98 ±0.85 AU. βIII Tubulin expression was similar amongst various ECM gel compositions (ns; Fig 3A), indicating similar neuronal differentiation.

3.3.2 Cholinergic neurons

Choline acetyltransferase (ChAT) expression was used to detect the presence of cholinergic neurons (Fig 3C). Collagen I (33.73 ± 1.13 AU) and collagen I/IV/laminin (28.82 ± 1.21 AU) sheets had a significantly elevated expression of ChAT compared to sheets with composite collagen and/or heparan sulfate. Immunoblotting demonstrated an enriched cholinergic neuron population in tissue engineered sheets manufactured with collagen I only or composite collagen I/IV with laminin. The presence of cholinergic neurons was additionally confirmed using immunohistochemistry (Fig 4E-H).

Figure 4. Immunofluorescence for differentiated neurons within tissue engineered sheets.

Figure 4

Open in a new tab

Differentiated neurons within tissue engineered sheets were stained with markers for vasoactive intestinal peptide (VIP; red), choline acetyltransferase (ChAT; red) or neuronal nitric oxide synthase (nNOS; green). (A-D) Numerous differentiated VIP-ergic neurons were present in tissue engineered sheets. (E-H) Differentiated excitatory cholinergic neurons expressing ChAT were present within tissue engineered sheets; (I-L) Differentiated inhibitory nitrergic neurons expressing nNOS were present within tissue engineered sheets. Scale bar 100μm.

3.3.4 Nitrergic inhibitory motor neurons

Neuronal nitric oxide synthase (nNOS) expression was used to detect the presence of inhibitory nitrergic motor neurons (Fig 3D). Sheets with collagen IV (with or without laminin/heparan sulfate) had a significantly higher nNOS expression ranging from 26.37 ± 1.29 AU – 28.15 ± 2.69 AU. Conversely to ChAT, collagen I sheets had minimal nNOS expression (11.33 ± 2.85 AU). Presence of nNOS was additionally confirmed using immunohistochemistry (Fig 4I-L).

3.3.4 VIP-ergic inhibitory motor neurons

Vasoactive intestinal peptide (VIP) motor neurons were identified using immunohistochemistry. VIP neurons were abundant, with increased immunofluorescence in composite hydrogels with laminin and heparan sulfate (Fig 4A-D).

3.4 Agonist-induced contractility of tissue engineered innervated smooth muscle sheets

3.4.1: Potassium Chloride-induced contraction

Electromechanical coupling integrity of constituent smooth muscle cells was first evaluated using potassium chloride (KCl). KCl treatment elicited rapid contractions that were sustained for ~5 minutes (Fig 5). Peak maximal contraction in response to KCl was similar between the different tissue engineered sheets (Fig 5A-D), ranging from 279.5 ± 4.79μN to 296.5 ± 6.26μN. This correlated with the equivalent expression of contractile smooth muscle marker, Smoothelin, indicating that the constituent smooth muscle cells within the tissue engineered sheets maintained a contractile phenotype regardless of ECM composition. Furthermore, KCl-induced contractions in tissue engineered sheets were similar to native rabbit intestinal tissues (Fig 5E) in time course, but slightly reduced in magnitude. Peak KCl-induced contractions in native tissue averaged 373.5 ± 10.63μN. KCl-induced contraction was unaffected by pre-treatment with neuronal blocker TTX (red traces; Fig 5), indicating myogenic electromechanical coupling integrity. Figure 5F demonstrates that the area under the curve of contraction was similar in all tissue engineered sheets, and was significantly higher in native rabbit intestinal tissues. Although reduced in magnitude compared to native tissue, KCl-induced contractions were similar among the different tissue engineered sheets, indicating a robust contractile smooth muscle phenotype unaffected by the ECM composition.

Figure 5. Potassium chloride induced contraction of tissue engineered sheets.

Figure 5

Open in a new tab

60mM Potassium chloride (KCl) was used to examine the electromechanical coupling integrity of the constituent smooth muscle cells within the tissue engineered sheets. The black traces indicate the contraction in response to the addition of KCl. The red traces indicate the addition of KCl in the presence of a neuronal blocker, TTX. Pre-treatment with TTX did not inhibit KCl-induced contraction. The ECM composition of the tissue engineered sheets did not affect smooth muscle contraction, evidenced by similar contractile patterns in response to KCl stimulation. A robust and immediate contraction was observed upon addition of KCl (indicated by the arrows) in all tissue engineered sheets (A-D), similar to native rabbit intestinal tissue (E). Peak contraction in response to KCl ranged between 279.5μN and 296.5μN in tissue engineered sheets, and averaged at 373.3 ± 10.63 in native tissue. (F) The area under the curve of KCl-induced contraction was quantified to demonstrate no significant (ns) difference in contraction in tissue engineered sheets, with a slightly elevated (*p<0.05) magnitude in native tissue.

3.4.2: Acetylcholine-induced contraction

Exogenous addition of 1μM Acetylcholine (Ach) was used to simulate agonist-induced contraction. All tissue engineered sheets contracted in response to Ach, and sustained contractions up to ~ 5 minutes post stimulation with Ach (Fig 6A-D). Tissue engineered sheets with composite collagen I/IV with laminin had a significantly elevated peak maximal Ach-induced contraction (Fig 6C; 232.9 ± 8.167μN), as well as an elevated area under the curve of contraction (47606 ± 2054 AU). Magnitude of Ach-induced contraction was still significantly lower compared to contraction in native tissue (342.6 ± 3.15μN; 70448 ± 5876 AU). However, the time course of contraction was very similar to native tissue in tissue engineered sheets containing laminin, reaching maximal contraction within a minute of agonist stimulation. Collagen I sheets also had an elevated Ach-induced contraction (Fig 6A; 238.9 ± 13.72 μN; 42668 ± 2172 AU) corresponding to the elevated ChAT protein expression. However, the kinetics of contraction did not match native tissue.

Figure 6. Acetylcholine induced contraction.

Figure 6

Open in a new tab

Addition of 1μM Acetylcholine (Ach; arrow) resulted in contraction of tissue engineered sheets, as well as native tissue. Red traces demonstrate Ach treatment in the presence of neuronal blocker, TTX. (A-E) Representative tracings of Ach-induced contraction in tissue engineered sheets and native tissue. Magnitude of Ach-induced contraction varied between tissue engineered sheets. Comparison of the area under the curve of contraction demonstrated that tissue engineered sheets approached 31.5% (Col4) - 67.6% (Laminin) of contraction observed in native tissue. In the presence of TTX, magnitude of Ach-induced contraction was attenuated. Quantification of inhibition (F) revealed that the degree of inhibition with TTX varied amongst the tissue engineered sheets with different ECM compositions. Highest inhibition was observed in Col i (72.77 ± 2.5%) and Col I/IV/Laminin (60.58 ± 1.7%) sheets, indicating an elevated presence of cholinergic neurons contributing to Ach-induced contraction. Significantly lower inhibition (*p<0.05; 48.36 ± 4.3 – 50.31 ± 4.2%) was seen in Col I/IV and Col I/IV/Lam/Heparan sulfate sheets. TTX pre-treatment attenuated Ach-induced contraction by 72.73 ± 3.7% in native tissue.

In order to estimate the smooth muscle (myogenic) component of Ach-induced contraction, neurotoxin TTX was used as a pretreatment (red traces, Fig 6). Area under the curves of contraction was compared with and without TTX pre-treatment in order to estimate %inhibition (Fig 6F). %Inhibition of Ach-induced contraction in the presence of TTX was highest in two ECM conditions: i) collagen I sheets (72.77 ± 2.45%); and ii) collagen I/IV/laminin sheets (60.58 ± 1.66%). These values of %inhibition were similar to that observed in native tissue (72.73 ± 3.66%) upon TTX-pretreatment. This increased neuronal contribution to Ach-induced contraction also correlated with the elevated protein expression of ChAT in collagen I and composite collagen I/IV/laminin sheets (Fig 3C). TTX-pretreatment inhibited Ach-induced contraction to a significantly lower extent in collagen I/IV ± heparan sulfate sheets, ranging from 48.36 ± 4.36 % (Heparan sulfate) to 50.31 ± 4.22% (collagen IV; Fig 6F).

3.5 Relaxation in engineered innervated sheets in response to electrical field stimulation

Electrical field stimulation (EFS) at 5Hz, 0.5ms was used to stimulate neurons within the tissue engineered sheets to produce relaxation of smooth muscle (Fig 7). The extent of relaxation was quantified as area under the curve of relaxation. Extent of relaxation significantly varied amongst the tissue engineered sheets with varying ECM compositions. Sheets bioengineered with collagen IV, which displayed elevated nNOS expression, had higher relaxation compared to sheets bioengineered with collagen I only (compare 109693 ± 8465 AU in collagen I/IV sheets to 23142 ± 4921 in collagen I sheets). Sheets containing laminin and/or heparan sulfate also had significantly elevated relaxation compared to collagen I sheets (68395-69025 AU). In response to EFS, native tissues relaxed generating 101550 ± 11279 AU. Tissue engineered sheets with collagen IV and/or laminin and/or heparan sulfate additionally had a time course of relaxation most similar to native tissue. Maximal relaxation was achieved within 2 minutes of EFS, and a subsequent recovery of basal force was complete within 10 minutes. Upon pre-treatment with TTX, EFS-induced relaxation was inhibited entirely (red traces, Fig 7).

Figure 7. Electrical Field Stimulation induced relaxation in tissue engineered sheets.

Figure 7

Open in a new tab

Electrical Field stimulation (EFS; shaded gray area) was used to stimulate relaxation in tissue engineered sheets (A-D) and native tissue (E). Red traces indicate TTX pre-treatment. EFS induced relaxation was significantly attenuated by TTX-pretreatment (90.9 ± 2.41% - 94.41 ± 0.93%), indicating that differentiated neurons within the tissue engineered sheets were capable of evoking smooth muscle relaxation. The magnitude of relaxation varied amongst the tissue engineered sheets (summarized in Table 3). (F) Quantification of the area under the curve of relaxation indicated that Col I sheets had a significantly low (***p<0.001; 23142 ± 4921AU) magnitude of relaxation. Relaxation in Col I/IV sheets (109693 ± 8465AU) were similar to those observed in native tissue (101550 ± 11279AU) in response to the electrical field, indicating the presence of elevated levels of inhibitory motor neurons capable of mediating relaxation. Relaxation was higher in Col I/IV/Laminin (69025 ± 7154AU) and Col I/IV/Lam/Heparan sulfate (68395 ± 8228AU) sheets, when compared to Col I, also indicating a similar increase in the presence of inhibitory motor neurons.

3.5.1 Inhibition of nitric oxide synthase

In order to identify the presence and functionality of nitrergic neurons, an inhibitor of nitric oxide synthase (L-NAME) was used (green traces, Fig 8). %inhibition was determined by comparing areas under the curves of maximal relaxation with and without the L-NAME pre-treatment. %inhibition with L-NAME treatment was the lowest in collagen I sheets (33.37 ± 8.37%; green trace, Fig 8A). This corresponded to the low protein expression of nNOS in collagen I sheets compared to sheets containing collagen IV (Fig 3D). In contrast, the inhibition of nNOS activity attenuated relaxation upto 61.71 ± 2.82% (green trace, Fig 8B) in collagen I/IV sheets. In sheets containing laminin and heparan sulfate, %inhibition with L-NAME varied between 62.28 ± 2.75% (laminin, Fig 8C) to 57.16 ± 1.91% (heparan sulfate). This inhibition is significantly elevated compared to collagen I sheets, corresponding to the increased expression of nNOS observed in the collagen I/collagen IV sheets (Fig 3C). Native tissues had a higher %inhibition with L-NAME (78.02 ± 2.85%).

Figure 8. Inhibition of relaxation with L-NAME.

Figure 8

Open in a new tab

The functionality of nitrergic neurons was studied by inhibiting EFS-induced relaxation with L-NAME, a non-metabolizable substrate for nNOS. The green traces indicate EFS in the presence of L-NAME. Pretreatment with L-NAME attenuated EFS-induced relaxation in all tissue engineered sheets (A-D) and native tissue (E). (F) Quantification of the area under the curve for relaxation indicated that the extent of L-NAME inhibition varied amongst the tissue engineered sheets. Col I sheets had a significantly lower %inhibition with L-NAME (*p<0.05; 33.4 ± 8.4%) corresponding to the lowest immunoblot expression of nNOS. The degree of L-NAME inhibition was higher in tissue engineered sheets containing Col I/IV and/or laminin and/or heparan sulfate (57.16% - 62.28%), corresponding to the higher immunoblot expression of nNOS. Attenuation of relaxation in the presence of L-NAME was 78 ± 2.9% in native tissue.

3.5.2: Inhibition of the VIP-receptor

The functionality of VIP-ergic neurons was assessed using a VIP receptor antagonist (VIP-Ra). Pre-treatment with VIP-Ra inhibited maximal relaxation in all tissue engineered sheets to varying extents ranging from 55.55 ± 3.92% - 65.92 ±5.38% (pink traces, Fig 9). Inhibition of EFS-induced relaxation indicated the presence of differentiated VIP-ergic neurons in tissue engineered sheets.

Figure 9. Inhibition of relaxation with VIP-Ra.

Figure 9

Open in a new tab

The functionality of VIP-ergic neurons was studied by inhibiting EFS-induced relaxation with VIP-Ra, a VIP-receptor antagonist. Purple traces indicate EFS in the presence of VIP-Ra. In the presence of VIP-Ra, EFS-induced relaxation was attenuated in tissue engineered sheets (A-D) and in native tissue (E), indicating the presence of functional VIP-ergic neurons capable of mediating smooth muscle relaxation upon electrical field stimulation. Area under the curve of relaxation was quantified, to calculate the %inhibition of relaxation in the presence of VIP-Ra (F). Extent of VIP-Ra induced inhibition of relaxation varied from 56.55 ± 3.12% - 63.11 ± 3.2% in tissue engineered sheets, and averaged at 73.32 ± 3.23 % in native tissue.

4. Discussion

Neural stem cell transplantation is a promising therapeutic approach to repopulate neurons within enteric ganglia. A complete loss of neurons is reported in HSCR, and a partial loss of selective neuronal subtypes is documented in achalasia and stenosis [7, 8, 34]. Several groups have injected enteric neuronal progenitor cells into experimental models of aganglionosis, demonstrating the feasibility of transplantation [9, 12, 35]. However, there is inadequate focus on differentiation of progenitor cells into mature neuronal subtypes, and subsequent assessment of functionality. Here, we describe a method to bias differentiation of enteric neuronal progenitor cells in vitro, prior to transplantation.

The ECM microenvironment, consisting of collagens, laminin and proteoglycans, not only acts as a structural framework for cells, but also plays an active role in aiding neurotrophic signaling [16, 36]. For these studies, we chose to evaluate four ECM components (collagen I, collagen IV, laminin and heparan sulfate), three of which are known to be present in adult myenteric ganglia [37]. Collagen IV has been documented to be favorable for neurite outgrowth and neuronal differentiation [23, 24, 38]. Laminin has long been known for its neurite promoting activity, in central, peripheral, and enteric neurons [26, 29, 39]. The role of the heparan sulfate proteoglycan in neuronal differentiation is also well documented, both developmentally and in regenerative medicine applications [40-42]. Fibrillar Collagen I was used additionally in these studies for ease of gelation and incorporation of other ECM components within a 3D hydrogel.

Apart from composition, substrate elasticity has been demonstrated to affect the differentiation of adult neural stem cells, with neuronal differentiation reported between 100-500Pa [43, 44]. ECM hydrogel compositions were adjusted in order to maintain their viscoelastic modulus within the range suitable for neuronal differentiation (Table 1). Structural architecture was verified using scanning electron microscopy, wherein the addition of collagen IV demonstrated the presence of network structures, similar to self-assembled collagen IV in the mammalian basement membrane [45]. The addition of laminin did not alter the ultrastructure, because it was expected to coat collagen fibers evenly. Additionally, laminin was also not expected to alter the stiffness/viscoelasticity of the gels, given the manner of its interaction with the collagens [46-48]. The glycosaminoglycan chains of heparan sulfate are documented to cross link between laminin and collagen IV, thereby pulling fibers into a more compact structure, and slightly increasing the viscoelasticity of ECM gels [45, 49].

4.1 Smooth muscle cells within tissue engineered sheets drive the differentiation of enteric neuronal progenitor cells

Tissue engineered sheets provided a good modality to assess variability of differentiated neurons due to ECM composition as well as the functionality of differentiated neurons. The proximity to smooth muscle promoted the differentiation of enteric neuronal progenitor cells extensively. In vitro differentiation of neural stem cells in the presence of gut-derived factors has been demonstrated previously by us and others [32, 33, 50]. Neurotrophic factors (NT-3, Neurturin, GDNF) and morphogens (BMP-2/4) capable of driving enteric neuronal progenitor cell proliferation and differentiation have been demonstrated to arise from the smooth muscle and mesenchyme of the developing and adult gut [51-53]. Recently, the postnatal bowel was demonstrated to support the differentiation of enteric neuronal progenitor cells, strengthening the fact that cues for differentiation can be derived from the postnatal gut [54, 55]. Hence, it was expected that smooth muscle cells within the tissue engineered sheets would drive the differentiation of enteric neuronal progenitor cells. We evaluated all tissue engineered sheets to ensure that the constituent smooth muscle cells demonstrated a contractile phenotype expressing Smoothelin (Fig 3B). Smoothelin expression has been previously demonstrated to be essential for contractility of smooth muscle [56]. In line with the equivalent expression of smoothelin, myogenic electromechanical coupling integrity was also equivalent in the tissue engineered sheets (Fig 5A-D). Similar patterns of contractions were observed in tissue engineered sheets in response to KCl, regardless of ECM composition.

4.2 ECM modulates differential neuronal subtypes while supporting overall smooth muscle-driven neuronal differentiation

In the presence of the smooth muscle, enteric neuronal progenitor cells differentiated, and expressed similar amounts of pan-neuronal marker βIII Tubulin (Figure 3A), suggesting that smooth muscle derived factors and substrate viscoelasticity were suitable for overall neuronal differentiation. However, on closer examination of neural subtypes, there was a differential expression of excitatory and inhibitory markers within tissue engineered sheets with varying ECM compositions (Fig 3).

Sheets containing laminin had a balanced expression of both ChAT and nNOS. Kinetics of Ach-induced contraction in laminin sheets was most similar to native tissue, indicating the presence of an increased viable cholinergic neuronal component in composite collagen/laminin sheets. Furthermore, attenuation of EFS-induced relaxation by L-NAME (~62%) was also kinetically similar to native tissue (~78%), indicating the presence of a nitrergic neuronal component.

Collagen I, in the absence of any other matrix components, was the ECM of choice when an enriched cholinergic neuronal population was required, with a significantly diminished nitrergic neuronal population (Fig 3C-D). While these sheets demonstrated a robust TTX-sensitive Ach-induced contraction commensurate with the heightened ChAT protein expression, relaxation in response to an electrical field was diminished. Furthermore, there was minimal attenuation of relaxation upon the inhibition of nNOS, correlating with the low nNOS expression.

Composite Collagen I/IV sheets had an enhanced nNOS protein expression, with an associated increase in EFS induced relaxation (Fig 7B). AUC of relaxation in composite Collagen I/IV sheets was comparable to native intestinal tissue. However, both ChAT expression and contraction was lower in composite collagen sheets.

4.3 ECM is a framework upon which smooth muscle derived factors regulate differentiation of neural subtypes

Overall, the studies in this paper contribute towards understanding the impact of environmental and molecular mechanisms associated with enteric neural subset activation. Here we demonstrate a critical role of collagen I and collagen IV containing ECM environments in promoting excitatory and inhibitory motor neurons, respectively. The ECM microenvironment plays a role in modulating neurotrophic as well as morphogenetic signaling. Morphogenetic signaling via the BMP family expressed in fetal gut is important for the phenotypic diversity of enteric ganglia, including nitrergic and VIP-ergic neuron differentiation [57] [58] [59, 60]. Collagen IV is documented to modulate BMP signaling, and heparan sulfate modulates GDNF signaling in the gut [25, 27]. Immunoreactivity of neurotrophic factor, NT-3, has been observed in ganglia and in the ECM molecules surrounding them, suggesting a role for a Collagen IV-based ECM to modulate NT-3 signaling [61]. The constituent smooth muscle phenotype in tissue engineered sheets was contractile, expressing smoothelin, and generating contractions and relaxations approaching ~60% of those generated by native intestinal tissue. Differentiation cues arising from the constituent smooth muscle cells drove enteric neuronal differentiation. Furthermore, it is likely that the ECM could act as a framework for smooth muscle-derived factors, enhancing or inhibiting their effects, resulting in the generation of differential neuronal phenotypes.

5. Conclusions

The studies in this paper used tissue engineering as a tool to evaluate the effect of ECM composition on the differentiation of adult enteric neuronal progenitor cells into mature neuronal subtypes. Furthermore, our studies indicated that different neuronal subtypes emerged within microenvironments that varied in their ECM composition. A combination of biochemical, immunohistochemical and physiological analyses revealed that several functional differentiated neuronal subtypes were present in tissue engineered intestinal sheets, capable of mediating smooth muscle contraction/relaxation. Neuronal populations varied from being highly cholinergic (collagen I), highly nitrergic (collagen IV), or balanced between the two (laminin and/or heparan sulfate). Additionally, peptidergic neurons with VIP immunoreactivity were also detected. Our studies indicate that neuronal differentiation was modulated by varying the composition of ECM microenvironments. Enriched populations of differentiated neurons can be derived within transplantable tissue engineered sheets, using ECM microenvironments. ECM microenvironments may also facilitate adequate trophic support and phenotype maintenance of differentiated neurons.

Table 2.

ECM/Force ΔForce (μN) Area under the Curve (AU) Inhibition of contraction by TTX (%AUC)
Collagen I 238.9 ± 13.72* 42668 ± 2172* 72.77 ± 2.47
Collagen I/IV 170.5 ± 7.56 22201 ± 2507 50.31 ± 4.23
Collagen I/IV/Laminin 232.9 ± 8.17* 46409 ± 2361* 60.58 ± 1.68
Collagen I/IV/Laminin/Heparan sulfate 185.1 ± 10.55 37447 ± 2534 48.36 ± 4.36
Native intestinal tissue 342.6 ± 3.15** 70448 ± 5876** 72.73 ± 3.66
Open in a new tab

Table 3.

ECM/Force ΔForce (μN) Area under the Curve (AU) Inhibition of relaxation by TTX (%AUC) Inhibition of relaxation by L-NAME (%AUC) Inhibition of relaxation by VIP-Ra (%AUC)
Collagen I −146.9 ± 10.63 23142 ± 4921 90.99 ± 2.41 33.37 ± 8.37 56.55 ± 3.12
Collagen I/IV −389.7 ± 12.19 109693 ± 8465 94.41 ± 0.93 61.71 ± 2.82 63.11 ± 3.2
Collagen I/IV/Laminin −250.8 ± 32.02 69025 ± 7154 92.09 ± 1.24 62.28 ± 2.75 60.96 ± 5.19
Collagen I/IV/Laminin/Heparan sulfate −239.5 ± 20.28 68395 ± 8228 92.78 ± 0.93 57.16 ± 1.91 61.92 ± 4.77
Native intestinal tissue −355.8 ± 17.47 101550 ± 11279 93.05 ± 0.71 78.02 ± 2.85 73.32 ± 3.23
Open in a new tab

Acknowledgements

This work was supported by NIH R01DK071614 and NIH R01DK042876. The authors would like to thank Dr. Chris Pernell at the North Carolina State Food Science Rheology Laboratory for his expert technical assistance with rheometry. The authors declare no conflicts of interest and no competing financial interests.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Furness JB. The enteric nervous system and neurogastroenterology. Nat Rev Gastroenterol Hepatol. 2012;9:286–94. doi: 10.1038/nrgastro.2012.32. [DOI] [PubMed] [Google Scholar]
  • 2.Kunze WA, Furness JB. The enteric nervous system and regulation of intestinal motility. Annu Rev Physiol. 1999;61:117–42. doi: 10.1146/annurev.physiol.61.1.117. [DOI] [PubMed] [Google Scholar]
  • 3.Furness JB. Types of neurons in the enteric nervous system. J Auton Nerv Syst. 2000;81:87–96. doi: 10.1016/s0165-1838(00)00127-2. [DOI] [PubMed] [Google Scholar]
  • 4.Sang Q, Young HM. Chemical coding of neurons in the myenteric plexus and external muscle of the small and large intestine of the mouse. Cell Tissue Res. 1996;284:39–53. doi: 10.1007/s004410050565. [DOI] [PubMed] [Google Scholar]
  • 5.De Giorgio R, Guerrini S, Barbara G, Stanghellini V, De Ponti F, Corinaldesi R, et al. Inflammatory neuropathies of the enteric nervous system. Gastroenterology. 2004;126:1872–83. doi: 10.1053/j.gastro.2004.02.024. [DOI] [PubMed] [Google Scholar]
  • 6.Kessmann J. Hirschsprung's disease: diagnosis and management. Am Fam Physician. 2006;74:1319–22. [PubMed] [Google Scholar]
  • 7.Takahashi T, Nakamura K, Itoh H, Sima AA, Owyang C. Impaired expression of nitric oxide synthase in the gastric myenteric plexus of spontaneously diabetic rats. Gastroenterology. 1997;113:1535–44. doi: 10.1053/gast.1997.v113.pm9352855. [DOI] [PubMed] [Google Scholar]
  • 8.Kraichely RE, Farrugia G. Achalasia: physiology and etiopathogenesis. Dis Esophagus. 2006;19:213–23. doi: 10.1111/j.1442-2050.2006.00569.x. [DOI] [PubMed] [Google Scholar]
  • 9.Almond S, Lindley RM, Kenny SE, Connell MG, Edgar DH. Characterisation and transplantation of enteric nervous system progenitor cells. Gut. 2007;56:489–96. doi: 10.1136/gut.2006.094565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Lindley RM, Hawcutt DB, Connell MG, Almond SL, Vannucchi MG, Faussone-Pellegrini MS, et al. Human and mouse enteric nervous system neurosphere transplants regulate the function of aganglionic embryonic distal colon. Gastroenterology. 2008;135:205–16. e6. doi: 10.1053/j.gastro.2008.03.035. [DOI] [PubMed] [Google Scholar]
  • 11.Metzger M, Bareiss PM, Danker T, Wagner S, Hennenlotter J, Guenther E, et al. Expansion and differentiation of neural progenitors derived from the human adult enteric nervous system. Gastroenterology. 2009;137:2063–73. e4. doi: 10.1053/j.gastro.2009.06.038. [DOI] [PubMed] [Google Scholar]
  • 12.Schäfer Kh, Micci Ma, Pasricha PJ. Neural stem cell transplantation in the enteric nervous system: roadmaps and roadblocks. Neurogastroenterology & Motility. 2009;21:103–12. doi: 10.1111/j.1365-2982.2008.01257.x. [DOI] [PubMed] [Google Scholar]
  • 13.Schafer KH, Hagl CI, Rauch U. Differentiation of neurospheres from the enteric nervous system. Pediatr Surg Int. 2003;19:340–4. doi: 10.1007/s00383-003-1007-4. [DOI] [PubMed] [Google Scholar]
  • 14.Kruger GM, Mosher JT, Bixby S, Joseph N, Iwashita T, Morrison SJ. Neural crest stem cells persist in the adult gut but undergo changes in self-renewal, neuronal subtype potential, and factor responsiveness. Neuron. 2002;35:657–69. doi: 10.1016/s0896-6273(02)00827-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Bondurand N, Natarajan D, Thapar N, Atkins C, Pachnis V. Neuron and glia generating progenitors of the mammalian enteric nervous system isolated from foetal and postnatal gut cultures. Development. 2003;130:6387–400. doi: 10.1242/dev.00857. [DOI] [PubMed] [Google Scholar]
  • 16.Simon-Assmann P, Kedinger M, De Arcangelis A, Rousseau V, Simo P. Extracellular matrix components in intestinal development. Experientia. 1995;51:883–900. doi: 10.1007/BF01921739. [DOI] [PubMed] [Google Scholar]
  • 17.Letourneau PC, Condic ML, Snow DM. Interactions of developing neurons with the extracellular matrix. J Neurosci. 1994;14:915–28. doi: 10.1523/JNEUROSCI.14-03-00915.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Rauch U, Schafer KH. The extracellular matrix and its role in cell migration and development of the enteric nervous system. Eur J Pediatr Surg. 2003;13:158–62. doi: 10.1055/s-2003-41265. [DOI] [PubMed] [Google Scholar]
  • 19.Tamiolakis D, Papadopoulos N, Hatzimichael A, Lambropoulou M, Tolparidou I, Vavetsis S, et al. A quantitative study of collagen production by human smooth muscle cells during intestinal morphogenesis. Clin Exp Obstet Gynecol. 2002;29:135–9. [PubMed] [Google Scholar]
  • 20.Bannerman PG, Mirsky R, Jessen KR, Timpl R, Duance VC. Light microscopic immunolocalization of laminin, type IV collagen, nidogen, heparan sulphate proteoglycan and fibronectin in the enteric nervous system of rat and guinea pig. J Neurocytol. 1986;15:733–43. doi: 10.1007/BF01625191. [DOI] [PubMed] [Google Scholar]
  • 21.Stang F, Fansa H, Wolf G, Reppin M, Keilhoff G. Structural parameters of collagen nerve grafts influence peripheral nerve regeneration. Biomaterials. 2005;26:3083–91. doi: 10.1016/j.biomaterials.2004.07.060. [DOI] [PubMed] [Google Scholar]
  • 22.Dewitt DD, Kaszuba SN, Thompson DM, Stegemann JP. Collagen I-matrigel scaffolds for enhanced Schwann cell survival and control of three-dimensional cell morphology. Tissue Eng Part A. 2009;15:2785–93. doi: 10.1089/ten.TEA.2008.0406. [DOI] [PubMed] [Google Scholar]
  • 23.Ali SA, Pappas IS, Parnavelas JG. Collagen type IV promotes the differentiation of neuronal progenitors and inhibits astroglial differentiation in cortical cell cultures. Brain Res Dev Brain Res. 1998;110:31–8. doi: 10.1016/s0165-3806(98)00091-1. [DOI] [PubMed] [Google Scholar]
  • 24.Lein PJ, Higgins D, Turner DC, Flier LA, Terranova VP. The NC1 domain of type IV collagen promotes axonal growth in sympathetic neurons through interaction with the alpha 1 beta 1 integrin. J Cell Biol. 1991;113:417–28. doi: 10.1083/jcb.113.2.417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wang X, Harris RE, Bayston LJ, Ashe HL. Type IV collagens regulate BMP signalling in Drosophila. Nature. 2008;455:72–7. doi: 10.1038/nature07214. [DOI] [PubMed] [Google Scholar]
  • 26.Pomeranz HD, Sherman DL, Smalheiser NR, Tennyson VM, Gershon MD. Expression of a neurally related laminin binding protein by neural crest-derived cells that colonize the gut: relationship to the formation of enteric ganglia. J Comp Neurol. 1991;313:625–42. doi: 10.1002/cne.903130408. [DOI] [PubMed] [Google Scholar]
  • 27.Barnett MW, Fisher CE, Perona-Wright G, Davies JA. Signalling by glial cell line-derived neurotrophic factor (GDNF) requires heparan sulphate glycosaminoglycan. J Cell Sci. 2002;115:4495–503. doi: 10.1242/jcs.00114. [DOI] [PubMed] [Google Scholar]
  • 28.Hynes RO. The extracellular matrix: not just pretty fibrils. Science. 2009;326:1216–9. doi: 10.1126/science.1176009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Raghavan S, Gilmont RR, Bitar KN. Neuroglial differentiation of adult enteric neuronal progenitor cells as a function of extracellular matrix composition. Biomaterials. 2013;34:6649–58. doi: 10.1016/j.biomaterials.2013.05.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Raghavan S, Lam MT, Foster LL, Gilmont RR, Somara S, Takayama S, et al. Bioengineered three-dimensional physiological model of colonic longitudinal smooth muscle in vitro. Tissue Eng Part C Methods. 2010;16:999–1009. doi: 10.1089/ten.tec.2009.0394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Stuart K, Panitch A. Characterization of gels composed of blends of collagen I, collagen III, and chondroitin sulfate. Biomacromolecules. 2009;10:25–31. doi: 10.1021/bm800888u. [DOI] [PubMed] [Google Scholar]
  • 32.Gilmont RR, Raghavan S, Somara S, Bitar KN. Bioengineering of physiologically functional intrinsically innervated human Internal Anal Sphincter constructs. Tissue Eng Part A. 2014 doi: 10.1089/ten.tea.2013.0422. epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Raghavan S, Gilmont RR, Miyasaka EA, Somara S, Srinivasan S, Teitelbaum DH, et al. Successful implantation of bioengineered, intrinsically innervated, human internal anal sphincter. Gastroenterology. 2011;141:310–9. doi: 10.1053/j.gastro.2011.03.056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Takahashi T. Pathophysiological significance of neuronal nitric oxide synthase in the gastrointestinal tract. J Gastroenterol. 2003;38:421–30. doi: 10.1007/s00535-003-1094-y. [DOI] [PubMed] [Google Scholar]
  • 35.Kulkarni S, Becker L, Pasricha PJ. Stem cell transplantation in neurodegenerative disorders of the gastrointestinal tract: future or fiction? Gut. 2012;61:613–21. doi: 10.1136/gut.2010.235614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Adams JC, Watt FM. Regulation of development and differentiation by the extracellular matrix. Development. 1993;117:1183–98. doi: 10.1242/dev.117.4.1183. [DOI] [PubMed] [Google Scholar]
  • 37.Bannerman PC, Mirsky R, Jessen K, Timpl R, Duance V. Light microscopic immunolocalization of laminin, type IV collagen, nidogen, heparan sulphate proteoglycan and fibronectin in the enteric nervous system of rat and guinea pig. J Neurocytol. 1986;15:733–43. doi: 10.1007/BF01625191. [DOI] [PubMed] [Google Scholar]
  • 38.Tomaselli KJ, Damsky CH, Reichardt LF. Interactions of a neuronal cell line (PC12) with laminin, collagen IV, and fibronectin: identification of integrin-related glycoproteins involved in attachment and process outgrowth. J Cell Biol. 1987;105:2347–58. doi: 10.1083/jcb.105.5.2347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Rogers SL, Letourneau PC, Palm SL, McCarthy J, Furcht LT. Neurite extension by peripheral and central nervous system neurons in response to substratum-bound fibronectin and laminin. Dev Biol. 1983;98:212–20. doi: 10.1016/0012-1606(83)90350-0. [DOI] [PubMed] [Google Scholar]
  • 40.Yamaguchi Y. Heparan sulfate proteoglycans in the nervous system: their diverse roles in neurogenesis, axon guidance, and synaptogenesis. Semin Cell Dev Biol. 2001;12:99–106. doi: 10.1006/scdb.2000.0238. [DOI] [PubMed] [Google Scholar]
  • 41.Mammadov B, Mammadov R, Guler MO, Tekinay AB. Cooperative effect of heparan sulfate and laminin mimetic peptide nanofibers on the promotion of neurite outgrowth. Acta Biomaterialia. 2012;8:2077–86. doi: 10.1016/j.actbio.2012.02.006. [DOI] [PubMed] [Google Scholar]
  • 42.Sariola H, Saarma M. Novel functions and signalling pathways for GDNF. J Cell Sci. 2003;116:3855–62. doi: 10.1242/jcs.00786. [DOI] [PubMed] [Google Scholar]
  • 43.Banerjee A, Arha M, Choudhary S, Ashton RS, Bhatia SR, Schaffer DV, et al. The influence of hydrogel modulus on the proliferation and differentiation of encapsulated neural stem cells. Biomaterials. 2009;30:4695–9. doi: 10.1016/j.biomaterials.2009.05.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Saha K, Keung AJ, Irwin EF, Li Y, Little L, Schaffer DV, et al. Substrate modulus directs neural stem cell behavior. Biophys J. 2008;95:4426–38. doi: 10.1529/biophysj.108.132217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Kleinman HK, McGarvey ML, Hassell JR, Star VL, Cannon FB, Laurie GW, et al. Basement membrane complexes with biological activity. Biochemistry. 1986;25:312–8. doi: 10.1021/bi00350a005. [DOI] [PubMed] [Google Scholar]
  • 46.Deister C, Aljabari S, Schmidt CE. Effects of collagen 1, fibronectin, laminin and hyaluronic acid concentration in multi-component gels on neurite extension. J Biomater Sci Polym Ed. 2007;18:983–97. doi: 10.1163/156856207781494377. [DOI] [PubMed] [Google Scholar]
  • 47.Charonis AS, Tsilibary EC, Yurchenco PD, Furthmayr H. Binding of laminin to type IV collagen: a morphological study. J Cell Biol. 1985;100:1848–53. doi: 10.1083/jcb.100.6.1848. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Swindle-Reilly KE, Papke JB, Kutosky HP, Throm A, Hammer JA, Harkins AB, et al. The impact of laminin on 3D neurite extension in collagen gels. J Neural Eng. 2012;9:046007. doi: 10.1088/1741-2560/9/4/046007. [DOI] [PubMed] [Google Scholar]
  • 49.Yurchenco PD, Schittny JC. Molecular architecture of basement membranes. FASEB J. 1990;4:1577–90. doi: 10.1096/fasebj.4.6.2180767. [DOI] [PubMed] [Google Scholar]
  • 50.Kulkarni S, Zou B, Hanson J, Micci MA, Tiwari G, Becker L, et al. Gut-derived factors promote neurogenesis of CNS-neural stem cells and nudge their differentiation to an enteric-like neuronal phenotype. Am J Physiol Gastrointest Liver Physiol. 2011;301:G644–55. doi: 10.1152/ajpgi.00123.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Rossi J, Herzig KH, Voikar V, Hiltunen PH, Segerstrale M, Airaksinen MS. Alimentary tract innervation deficits and dysfunction in mice lacking GDNF family receptor alpha2. J Clin Invest. 2003;112:707–16. doi: 10.1172/JCI17995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Peters RJ, Osinski MA, Hongo JA, Bennett GL, Okragly AJ, Haak-Frendscho M, et al. GDNF is abundant in the adult rat gut. J Auton Nerv Syst. 1998;70:115–22. doi: 10.1016/s0165-1838(98)00044-7. [DOI] [PubMed] [Google Scholar]
  • 53.Saffrey MJ, Burnstock G. Growth factors and the development and plasticity of the enteric nervous system. J Auton Nerv Syst. 1994;49:183–96. doi: 10.1016/0165-1838(94)90165-1. [DOI] [PubMed] [Google Scholar]
  • 54.Hotta R, Stamp LA, Foong JP, McConnell SN, Bergner AJ, Anderson RB, et al. Transplanted progenitors generate functional enteric neurons in the postnatal colon. J Clin Invest. 2013;123:1182–91. doi: 10.1172/JCI65963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Hagl CI, Rauch U, Klotz M, Heumuller S, Grundmann D, Ehnert S, et al. The microenvironment in the Hirschsprung's disease gut supports myenteric plexus growth. Int J Colorectal Dis. 2012;27:817–29. doi: 10.1007/s00384-012-1411-0. [DOI] [PubMed] [Google Scholar]
  • 56.Niessen P, Rensen S, van Deursen J, De Man J, De Laet A, Vanderwinden JM, et al. Smoothelin-a is essential for functional intestinal smooth muscle contractility in mice. Gastroenterology. 2005;129:1592–601. doi: 10.1053/j.gastro.2005.08.018. [DOI] [PubMed] [Google Scholar]
  • 57.Chalazonitis A, Pham TD, Li Z, Roman D, Guha U, Gomes W, et al. Bone morphogenetic protein regulation of enteric neuronal phenotypic diversity: relationship to timing of cell cycle exit. J Comp Neurol. 2008;509:474–92. doi: 10.1002/cne.21770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Anitha M, Shahnavaz N, Qayed E, Joseph I, Gossrau G, Mwangi S, et al. BMP2 promotes differentiation of nitrergic and catecholaminergic enteric neurons through a Smad1-dependent pathway. Am J Physiol Gastrointest Liver Physiol. 2010;298:G375–83. doi: 10.1152/ajpgi.00343.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Hendershot TJ, Liu H, Sarkar AA, Giovannucci DR, Clouthier DE, Abe M, et al. Expression of Hand2 is sufficient for neurogenesis and cell type-specific gene expression in the enteric nervous system. Dev Dyn. 2007;236:93–105. doi: 10.1002/dvdy.20989. [DOI] [PubMed] [Google Scholar]
  • 60.Morikawa Y, Zehir A, Maska E, Deng C, Schneider MD, Mishina Y, et al. BMP signaling regulates sympathetic nervous system development through Smad4-dependent and -independent pathways. Development. 2009;136:3575–84. doi: 10.1242/dev.038133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Hoehner JC, Wester T, Pahlman S, Olsen L. Localization of neurotrophins and their high-affinity receptors during human enteric nervous system development. Gastroenterology. 1996;110:756–67. doi: 10.1053/gast.1996.v110.pm8608885. [DOI] [PubMed] [Google Scholar]

RESOURCES