Carbachol induces TGF-alpha expression and colonic epithelial cell proliferation in sensory-desensitised rats
- First Online:
- Cite this article as:
- Bulut, K., Felderbauer, P., Hoeck, K. et al. Int J Colorectal Dis (2010) 25: 335. doi:10.1007/s00384-009-0856-2
- 54 Views
Signals for the expression of the peptide growth factors epidermal growth factor and transforming growth factor-α (TGFα) in the gastrointestinal mucosa are largely unknown. We have shown earlier that extrinsic afferents in the gastrointestinal tract induce TGFα expression in colonic mucosa via the deliberation of neurotransmitters substance P and calcitonin gene-related peptide. The aim of our present study was to determine the effects of carbachol on mucosal TGFα expression and epithelial cell proliferation in vivo.
Rats were divided in three groups. Group 1 was treated with vehicle only, group 2 received one single subcutaneous injection of 250 μg/kg of carbachol and animals in group 3 were sensory-desensitised prior to the injection of 250 μg/kg carbachol. TGFα expression and epithelial cell proliferation was evaluated by polymerase chain reaction, Western blot analysis and bromodeoxyuridine staining.
Carbachol induced a significant increase in mucosal epithelial cell proliferation and TGFα expression. Sensory desensitisation did neither abolish the increased TGFα expression nor the increase in epithelial cell proliferation.
Parasympathetic pathways are involved in the control of TGFα expression in gastrointestinal mucosa as well as in epithelial cell proliferation.
KeywordsGastrointestinal nervous systemTGFαEGFExperimental model of colitisCarbacholParasympathetic nervesExtrinsic primary afferents
Over the last decade, epidermal growth factor (EGF), the EGF family of related peptides and their common binding site, the so called EGF receptor, have been shown to be key constituents in the maintenance and repair of the gastrointestinal mucosa.
In experimental models of colitis in rats, exogenously administered EGF protected and healed the intestinal mucosa [1, 2]. Furthermore, after extirpation of submandibular glands in rodents, the main source of intraluminal EGF in the stomach, a marked reduction in the rate of healing of experimentally induced gastric ulcers was observed. Exogenously administered EGF completely reversed the deleterious effect and enhanced healing in these animals [3–5]. Transforming growth factor-α (TGFα) knockout mice and mice harbouring a defective EGF receptor showed an increased susceptibility to the dextran sulphate-induced colitis in mice [6, 7]. In the mutant TGFα knockout mice, exogenously applied TGFα completely reversed this effect. In accordance to this, mice over-expressing TGFα showed a decreased susceptibility in the same animal model . More recently, Sinha and colleagues demonstrated a positive effect of EGF given as enemas in a small cohort of patients with mild to moderate ulcerative colitis .
Within the kindred of the EGF family of growth factors, TGFα appears to be the main ligand for the EGF receptor in the mucosa of the gastrointestinal tract in vivo. An increased expression of both, TGFα and EGF receptor, has been demonstrated following mucosal injury in a variety of experimental animal models of stomach ulcers or chemically induced colitis [10–14].
Although a multitude of data is available regarding the role of EGF, TGFα and EGF receptor in mucosal protection and repair as well as in carcinogenesis , no sufficient information exists, how expression of the growth factors is regulated. According to the literature, EGF and related peptides are able to induce their own expression and the expression of the EGF receptor . Furthermore, the endogenous tumour promoter desoxycholate induces TGFα mRNA expression in transformed colonic epithelial cells in vitro .
We reported earlier that TGFα mRNA and protein are increasingly expressed following stimulation of colonic mucosa with the neurotoxin capsaicin, indicating that activation of extrinsic capsaicin-sensitive primary afferent nervous fibres induce TGFα expression in vivo . More recently, we detected an increased TGFα expression in the non-inflamed duodenum during the course of a trinitrobenzene sulphonic acid (TNBS)-induced experimental colitis in rats . The increased TGFα expression in the duodenum was not readily explained by the known signals for the TGFα expression. Possible signals for the increased duodenal expression of TGFα presumably were fasting, parasympathetic or adrenergic parts of the enteric nervous system.
The parasympathetic agonist carbachol was shown to be able to activate the EGF receptor in human and rat conjunctival goblet cells and to induce the basolateral release of TGFα from T84 cells in vitro [20, 21]. The aim of our present study therefore was to evaluate possible effects of the activation of parasympathetic pathways with carbachol on TGFα expression in the rat colon in vivo.
Materials and methods
For all experiments, adult male Wistar rats with body weights between 250 and 300 g were used. The animal committees of the University of Essen and the government of North Rhine-Westphalia in Germany approved the experiments.
Animals were fasted overnight with free access to drinking water. Fifty-five rats were divided in three groups. Animals in group 1 were vehicle-treated (= control, 15 rats), group 2 were carbachol-treated (20 rats) and group 3 were sensory-desensitised and carbachol-treated (20 rats).
In group 1, animals received one single subcutaneous injection of 0.75 ml phosphate buffered saline (PBS). In group 2, animals received 250 μg/kg body weight carbachol dissolved in 0.75 ml of phosphate buffered saline (vehicle) in one single subcutaneous injection. The dose of 250 μg/kg body weight was shown previously to sufficiently induce intestinal goblet cells mucin secretion in vivo . The third group was sensory-desensitised prior to the subcutaneous injection of 250 μg/kg body weight carbachol in 0.75 ml of phosphate buffered saline. Animals (three animals in the control group and four animals in groups 2 and 3) were sacrificed by CO2 asphyxia at each time point at 6, 8, 12, 24 and 48 h after injection of carbachol or vehicle, and the tissues of the distal colons were dissected. Additionally, three animals were sacrificed without treatment.
Representative strips of the colon of about 2 cm of length (1.5–3.5 cm from the anus) were fixed in 4% zinc formalin in 0.9% NaCl. The remaining parts of the distal colon were shock-frozen in liquid nitrogen and kept frozen at −80°C until the day when the extraction of RNA or protein was performed.
Ablation of sensory nerves (sensory desensitisation)
As described previously, ablation of sensory nerves (sensory desensitisation) was performed by subcutaneous injection of capsaicin (125 mg/kg total dose; 1% solution in 10% ethanol, 10% Tween 80 and 80% in 0.9% NaCl; Sigma, St. Louis, MO, USA) or vehicle on two consecutive days . Two weeks after sensory desensitisation, animals were tested for the effectiveness of desensitisation by monitoring the protective wiping movements after instillation of one drop of 0.1 N NaOH into one eye . The disappearance of the wiping reflex was used as an indirect indication of a successful capsaicin pre-treatment and animals that still showed wiping movements after this treatment were excluded from the study.
Application of BrdU
All animals received 50 mg/kg body weight bromodeoxyuridine (BrdU; Sigma, Deisenhofen, Germany) diluted in 0.9% NaCl solution in a volume of 0.4 ml via a tail vein 1 h prior to the end of each experiment.
Histologic evaluation of tissue damage
Colonic specimens were fixed in buffered 4% zinc formalin, embedded in paraffin wax blocks and cut in 5-µm-thick sections. The sections were consecutively stained with haematoxylin and eosin and evaluated for tissue damage, the number of goblet cells per crypt as well as for the length and width of the colonic crypts as described previously .
Immunohistochemistry for BrdU-immunostaining nuclei in epithelial cells
Paraffin wax blocks were cut in 5-µm-thick sections and deparaffinised and rehydrated. Sections were digested with 0.1% protease solution (Sigma, Deisenhofen, Germany) and treated with 2 N HCl. Immunohistochemical analysis was performed by the avidin–biotin complex (ABC) technique using a standardised procedure (Vectastain ABC®, Vector, Burlingame, CA, USA) as described previously . In brief, after rehydration, sections were washed in 0.1 M PBS containing 0.3% Triton-X 100 and pre-incubated for 1 h at room temperature with normal horse serum (diluted 1:30 in 0.1% (v/v) PBS) and then incubated with anti-BrdU in a dilution of 1:1,000 (Sigma, Deisenhofen, Germany) for 12 to 15 h at room temperature. After several washes in 0.1 M PBS/0.3% Triton-X 100, sections were incubated with the secondary antibody (horse anti-mouse diluted 1:100 in 0.1 M PBS) for 2 h at room temperature. Endogenous peroxidase was quenched in ice-cold methanol with 1% hydrogen peroxide for 5 min. Binding of the antibody was then visualised after incubation with the avidin–biotin complex for 2 h at room temperature with hydrogen peroxide in the presence of diaminobenzidine. Finally, sections were slightly counterstained with eosin and mounted with Permount® medium (Sigma, Deisenhofen, Germany). All pictures were taken with a Zeiss photomicroscope using a Kodak® colour print film. The total numbers of immunoreactive BrdU-positive nuclei in epithelial cells were counted at a magnification of 1:100 by two independent investigators blinded to the various experimental subsets.
RNA extraction and reverse transcription
Total cellular RNA was extracted by a modified acidic guanidinium thiocyanate method as described previously . Five micrograms of the extracted total RNA was incubated with DNAse at 37°C in a water bath to remove DNA contamination. After DNA digestion, the samples were heated for 10 min at 99°C to inactivate the DNAse, and samples were chilled on ice for 5 min. One microgram of the RNA was transferred to a new tube and heated for 10 min at 70°C. Reverse transcription (RT) of the mRNA was then performed at 42°C using a standardised procedure (Promega, Inc., Madison, WI, USA). After reverse transcription, cDNA samples were heated at 99°C in a water bath for 5 min to inhibit enzyme activity and were stored at −20°C.
RT-PCR for the assessment of GAPDH and TGFα mRNA expression
Rat glyceraldehyde 3-phosphate dehydrogenase (GAPDH)  and rat TGFα  primers were synthesised upon the previously published sequences. Primers for TGFα were sense 5′-GCAGTGGTGTCTCACTTCAA-3′ and anti-sense 5′-CACTGCCAGGAGATCTGCATGCTC-3′, and primers for GAPDH were sense 5′-AATGCATCCTGCACCACCAA-3′ and anti-sense 5′-GTAGCCATATTCATTGTCATA-3′ (Eurogentec, Seraing, Belgium). RT-cDNA samples were diluted in ten times polymerase chain reaction (PCR) buffer containing MgCl2. PCR for GAPDH was performed by heating the samples to 94°C for 3 min, then 25 cycles with 1 min at 94°C, 1 min at 60°C and 2 min at 72°C. After amplification, samples were heated at 72°C for 7 min and stored at −20°C. For TGFα, samples were diluted in ten times PCR buffer containing MgCl2, heated for 3 min at 94°C, then 35 cycles with 1 min at 94°C, 2 min at 65°C and 1 min at 72°C. After amplification, samples were heated for 7 min at 72°C and stored at −20°C. In each PCR, samples containing only water contamination and samples containing RNA without reverse transcription were amplified to ensure no PCR infidelity and to estimate persistent DNA contamination. PCR products were separated on 1% to 1.5% agarose gels with a DNA ladder in adjacent lanes. Changes of GAPDH (515 bp) and TGFα (158 bp) expression were estimated by analysing the bands from the agarose gel with a scanning laser densitometer with the integrated area expressed as the average of five individual determinations.
Western blot analysis
Colonic tissues were homogenised (250 mg/ml, w/v) in 68 mM Tris-HCl buffer, pH 6.8, containing sodium dodecyl sulphate (2%), urea (2 M), 10% glycerol, distilled water and proteinase inhibitors (aprotinin, pefabloc, leupeptin, phenylmethylsulphonyl fluoride and soybean trypsin inhibitor) using a tissuemizer (Tekmar, Ohio). The homogenates were diluted 1:2 with Laemmli’s electrophoretic sample buffer containing β-mercaptoethanol and heated in a boiling water bath for 10 min . The amount of protein in the tissue extracts was evaluated using the Biuret method. Samples were cooled and centrifuged, and aliquots equivalent to 20 µg of total protein were subjected to electrophoresis on 8% or 16% Tris-glycine separating gels. The proteins were electrophoretically transferred to polyvinylidene fluoride membranes (Millipore, Milford, MA, USA) as described by Towbin et al. . After incubation with the primary antibody to rat TGFα (1:500), membranes were washed in PBS/Tween 20. Immunoreactive bands were detected after incubation with the secondary horseradish peroxidase antibody (1:1,000) by the chemiluminescence method as described previously .
The quantity of TGFα mRNA expression from ethidium bromide-stained agarose gels and TGFα immunoreactive protein on Western blots was calculated using scanning laser densitometry (BioRad, Hercules, CA, USA) with the integrated area expressed as the average of five individual determinations. The two-tailed unpaired Student’s t test or the Mann–Whitney U test was used as appropriate for comparison of the data, and a p value of <0.05 was considered significant.
Carbachol treatment did not lead to any structural damage like erosions or ulcers of the colonic mucosa. In response to carbachol, no changes of crypt length, number of goblet cells or mucosal crypt associated inflammatory cells were obtained (data not shown).
Epithelial cell proliferation
Transforming growth factor-α expression
Semiquantitative RT-PCR for TGFα mRNA expression
TGFα protein expression
One single subcutaneous injection of 250 μg/kg of the choline ester carbachol (carbamylcholine chloride) significantly induced an increase in the number of BrdU-positive nuclei in the epithelial cell layer along the colonic crypts. The increase in BrdU-positive nuclei indicates an increase in nucleotide turnover and gene expression and is widely accepted as a marker of cell proliferation. These results are in accordance with previously published information where stimulation of parasympathetic pathways with or without carbachol led to the proliferation of cells in the gastrointestinal mucosa [31, 32]. Furthermore, activation of muscarinic receptors is involved in proliferation of colon cancer [33, 34].
The carbachol-induced epithelial cell proliferation in the colonic mucosa was accompanied by an increase in TGFα mRNA and protein expression. In vitro, carbachol induced the basolateral release of TGFα from T84 cells . To our knowledge, our report is the first to show that carbachol induces TGFα expression in the gastrointestinal mucosa in vivo. Within the kindred of the EGF family of growth factors, TGFα appears to be the main ligand for the EGF receptor in vivo. EGF and TGFα are potent mitogens for gastrointestinal cells in vivo and in vitro [16, 35]. EGF, TGFα and their common binding site are involved in mucosal protection and repair as well as in cancer growth [1, 2, 6, 7, 9, 15].
Only few data are available regarding possible signals for TGFα expression. We have shown earlier that activation of extrinsic primary afferents in the rat colon by the neurotoxin capsaicin induced a significant increase in epithelial cell proliferation and TGFα expression . In these experiments, the capsaicin-stimulated epithelial cell proliferation was abolished by specific antagonists to the sensory neurotransmitters substance P or calcitonin gene-related peptide and by sensory desensitisation prior to the experiments. In our current study, the carbachol-induced effect on epithelial cell proliferation was not affected by sensory desensitisation. Although carbachol has mainly muscarinic actions, it retains some substantial nicotinic activity, particularly on autonomic ganglia . However, according to the results of this study, it is unlikely that carbachol acts via the modulation of extrinsic afferent neurons or primary afferents in the colonic mucosa. More recently, we detected an increased TGFα expression in the non-inflamed duodenum during the course of a TNBS-induced experimental colitis in rats . The increased TGFα expression in the duodenum was not readily explained by the known signals for the TGFα expression. Within the possible signals for the increased duodenal TGFα expression, the activation of parasympathetic pathways could be responsible for having mediated the induction of the increased TGFα expression. We now demonstrate that carbachol, a known agonist of the parasympathetic nervous system with mainly muscarinic activity, induces TGFα mRNA and protein expression in the gastrointestinal tract. This information further supports our hypothesis that TGFα expression is mainly modulated by parasympathetic and sensory pathways of the enteric nervous system.
There are some remarkable similarities in the biological activities of carbachol and members of the EGF family of growth factors. Carbachol is capable of producing increases in tone, amplitude of contractions and peristaltic activity of the stomach and intestines, as well as enhanced secretory activity of the gastrointestinal tract . Similarly, EGF and related peptides are able to increase mucosal blood flow, to modulate gastrointestinal motility, to inhibit gastric acid secretion and to stimulate gastric mucus production and secretion [37–41]. In addition, TGFα and EGF are widely distributed in the central nervous system, the spinal cord, in human spleen and lung and in neural tissues in the rat stomach [42–45]. External application of EGF stimulates pituitary adrenocorticotropic hormone, luteinising hormone, prolactin and growth hormone secretion and hypothalamic corticotropin-releasing factor and gonadotropin-releasing hormone suggesting that members of the EGF family can act as neurotransmitters . Therefore, it is possible that TGFα/EGF or a related peptide is released from neuroeffector junctions into the surrounding mucosa and consecutively bind and activate basolateral EGF receptors on epithelial cells. The activation of the EGF receptor in turn could induce TGFα/EGF and EGF receptor expression .
In our experiments, we were not able to detect the depletion of the mucosal goblet cells at 6 h after carbachol treatment. Phillips reported a significant depletion of mucin granules as early as 5 and 60 min after subcutaneous injection of carbachol. The mucin stores were replenished 4 h post-stimulation . Therefore, in our experiments, we may have missed the states of mucin depletion as we started to evaluate colonic tissues beginning at 6 h post-stimulation.
In conclusion, activation of parasympathetic pathways by carbachol induce a significant increase of epithelial cell proliferation and a significant increase in TGFα mRNA and protein expression. The increase in TGFα expression is independent of extrinsic afferent neurons of the gastrointestinal tract, a part of the enteric nervous system already known to induce TGFα expression.