Cholinoceptor Activation Subserving the Effects of Interferon Gamma on the Contractility of Rat Ileum

Recombinant rat interferon γ stimulated the contractility of isolated rat ileum at doses of 4–12 units/ml. Muscarinic cholinoceptors were involved, as treatment of the tissue with atropine prevented the contractile response of the ileum. Furthermore, interferon γ increased the affinity of carbachol for the cholinoceptors and did not change its maximum effect. Neurogenic pathways were also involved since pretreatment of ileum with hexamethonium, hemicholinium or tetrodotoxin impaired the contractile effect of interferon γ. In contrast to the action of exogenous carbachol, the effects of interferon γ are indirect. They appear to involve a G protein regulating phosphoinositide turnover and cytoskeletal structures since they could not be induced in ileum strips that were pretreated with pertussis toxin, phospholipase C inhibitors (2-nitro-carboxyphenyl, NN-diphenyl carbamate and neomycin), cytochalasine B or colchicine.


Introduction
Interferon gamma (IFN,) is a glycoprotein with antiviral activity that is involved in a variety of immunoregulatory processes. In addition to its effects on immune cell function, it can modulate smooth muscle" and intestinal epithelial cell growth.
IFNT is known to induce HLA class II antigen expression on endothelial cells, smooth muscle cells and enterocytes, and this could be important for the induction of an autoimmune or inflammatory response. The intestine normally contains abundant IFNT-producing T-lymphocytes, and IFNT release may occur locally and increase after antigen challenge. Through its action on tight junction permeability it could affect intestinal epithelial cell contacts and/or influence the function of the gut by modifying electrolyte transport. 1 Recently we have shown that recombinant rat IFNT interacts with isolated rat atria, mimicking the action of a muscarinic cholinergic agonist. Thus, incubation of rat atria with IFN, decreased tension and cAMP synthesis and increased cGMP production. 11 In this study we investigated if INF,, could also alter the mechanical behaviour of the intestine.
Our results suggest that IFNT imitates the action of carbachol, a muscarinic cholinergic agent, inducing the contraction of isolated rat ileum. In contrast to the exogenously added agonist carbachol, the effects of IFN, on ileum also involve a nerve-mediated pathway and are indirect. They require the integrity of the cytoskeleton and involve a regulatory G protein, indicating that IFN, reaction with the cholinergic receptor is complex and probably involves common signalling pathways. performed by preincubating rIFNT (1 000 U/ml) with an equal volume of M anti-IFNT (10 000 U/ml) during 30 min at 37C. According to the providers (Amgen Biologicals) the activity of IFN, (units/ml) was defined by an antiviral assay whereas one unit of M anti-IFNT was the amount of antibody sufficient for neutralization of one unit of rat IFNT antiviral activity.
The reaction mixture was used thereafter, adjusting the dilution to achieve the final concentration desired in the organ bath. Carbachol, hexamethonium, hemicholinium, tetrodotoxin, 2-nitro-carboxyphenyl, NN-diphenyl carbamate (NCDC), neomycin, cytochalasine B, colchicine, pertussis toxin and atropine were obtained from Sigma (St Louis, MI). Isolated ileum preparations: Ileum strips were obtained from male albino rats of the Wistar strain weighing between 200-250 g. The animals were sacrificed by decapitation and after excision of the abdomen coat, the ileum fragments were carefully dissected. They were placed in Petri dishes filled with a modified Krebs-Ringer-bicarbonate (KRB) solution. 11 The ileum was opened longitudinally and small portions (-1 cm long) were used. The ileum preparations were mounted in an organ bath containing 15 ml of KRB solution gassed with 5% CO2 in oxygen and kept at 37C and pH 7.4. One end of the the other was connected to a force transducer (Stathan UC3 Gold Cell) coupled to an ink-writing oscillograph (San Ei 180). A constant resting tension of 750 mg was applied to the tissue preparations by means of a micrometric device. Only ileum strips that presented spontaneous contractions were used and the isometric developed tension (in mg) was recorded. Preparations were allowed to equilibrate for 20 min. The contractile tension recorded at this moment (before delivering IFN, or the drugs) was considered the initial control. These tension values (expressed in mg) were obtained by measuring the amplitude of all the contractions recorded over a 10 min period and calculating their mean value. These initial magnitudes were compared with the experimental values and the variations induced by IFN, or drugs were expressed as percentage changes with respect to the initial control. The control values of tension at the end of equilibrium and before the addition of IFN, or drugs was: 200 + 18 mg (n 50).
Concentration-response curves were constructed according to van Rossum. 12 Single doses were delivered in volumes of 0.01 to 0.025 ml of an appropriate isotonic solution. The total amount of vehicle added to the bath never exceeded 0.25 ml. The time interval between doses was that needed by each dose to produce a maximum effect. Atropine, hexamethonium, hemicholinium, tetrodotoxin, pertussis toxin, neomycin, cytochalasine B, colchicine or NCDC were added to the organ bath 30 min before delivering IFN, or carbachol. At the concentrations used, these drugs did not alter the baseline.

Results
IFN, induced a concentration-dependent increase in the contractile activity of isolated ileum. An original tracing showing the pattern of stimulation is shown in Fig. 1. The stimulatory effect of IFN, developed gradually, reaching a plateau at 5-10 min. For this reason the ileum was exposed for a period of 8-10 min to each concentration ( Moreover, IFN, potentiated the action of carbachol by increasing the affinity of the cholinoceptor for its agonist (carbachol) while the maximum effect (gmax) remained unchanged ( cholinergic IFN, effect was nerve mediated, hexamethonium (10 -6 M), hemicholinium (2 x 10 -5 M) and tetrodotoxin (5 x 10 -7 M) were used as inhibitors. Figure 3A shows a significant reduction of the stimulatory effect of IFN, when the ileum was preincubated either with the inhibitor of nicotinic cholinoceptors (hexamethonium) or with the acetyl choline (AcCh) synthesis inhibitor (hemicholinium). Likewise, tetrodotoxin, an inhibitor of propagated action potentials, reduced the response of the ileum segment to IFN,. In contrast, these agents did not modify the response to exogenous carbachol (Fig.  3B). To ascertain whether G regulatory proteins participated in the muscarinic cholinergic action of IFN,, the ileum strips were treated with pertussis toxin. 13 The results shown in Fig. 4A demonstrate that the stimulatory effect of IFN, was abrogated by pertussis toxin. In contrast the mechanical response of pertussis toxin-treated tissue to carbachol remained unchanged ( Fig. 4B). At least one pathway of cholinoceptor-triggered phosphoinositide is regulated by pertussis toxin-sensitive G proteins that control activation of phospholipase C. To determine if phospholipase C was involved in the cholinergic action of IFN, we performed experiments in which the phospholipase C activation was inhibited by NCDC TM or neomycin. 15 The results of Fig. 4A demonstrate that incubation of ileum with NCDC at a concentration of 10-M, which is known to inhibit phospholipase C activity, TM prevented the effect of IFN,. On the other hand, NCDC did not alter the inotropic effect of carbachol (Fig. 4B). The same results were obtained preincubating ileum with 10 -M neomycin (data not shown).
In order to study if cytoskeletal structures were involved in the cholinomimetic action of IFN,, ileum strips were incubated with the microtubule disrupting agent colchicine, or with cytochalasine B to prevent microfilament polymerization. As shown in Fig. 5A, both colchicine (10 -M) and cytochalasine B (3 x 10 -M) impaired the stimulatory action of IFN,. In contrast, neither agent modified the response to carbachol (Fig. 5B)  IFN' (U/ml) 10   which release AcCh to act on nicotinic postsynaptic cholinoceptors; these in turn, could stimulate the release of AcCh to act on smooth muscle receptors.
IFNy induced firing of propagated action potentials, as tetrodotoxin impaired its activity. Nevertheless, as the inhibitory action of tetrodotoxin was only partial, a direct effect of IFNT on the smooth muscle could also be suggested. Therefore, the interactions of IFNT with the cholinoceptor pathways of signal transduction appear to be complex. G proteins could be a link between IFNT and the metabolic pathways triggered by muscarinic cholinergic stimulation. Muscarinic cholinergic receptors (mAcChR) belong to the family of receptors that are coupled to GTPbinding regulatory proteins (G proteins). 19 A single mAcChR can activate more than one type of G protein to regulate several signal transduction pathways. 19 Thus, events that follow muscarinic agonist binding to mAcChR can be the result of pertussis toxin sensitive or insensitive G protein coupled pathways. 2,21 Our results (Fig. 4A) demonstrate that a G regulatory protein is involved in the interaction be- tween IFNy and the cholinergic system in the intestine, as pertussis toxin treatment of the tissue prevented the reaction. This contrasts with pertussis toxin insensitivity of the carbachol-induced stimulatory effect in the same experimental system (Fig. 4B), suggesting that G protein insensitive pathways predominate in the contractile effect of carbachol. Cholinoceptor activation is associated with phosphoinositide (PI) turnover through phospholipase C activation 13 (Fig. 4A). Again, the stimulatory action of carbachol was insensitive to phospholipase C inhibition (Fig. 4B), indicating that different metabolic pathways resulting in similar mechanical effects may be followed by IFN, and carbachol. Many reactions mediated by G proteins share common features. 23 Tubulin, a main component of the cytoskeleton, has a GTP binding site and GTPase activity, that are necessary for the polymerization of the microtubules. 23 To determine if the cytoskeleton was involved in the reaction triggered by IFN, on the ileum, we studied the effect of drugs that interfere with cytoskeletal function (cytochalasine B to prevent microfilament polymerization and colchicine for microtubule disruption) ( Fig. 5A 18 and in intestinal epithelial cells subtle changes in cytoskeletal rearrangement were observed. However, the amount of IFN, necessary to obtain these effects was 100 times higher than the dose used to enhance contractility in this study.
In summary, we have shown that IFN, can trigger the mechanical response of isolated ileum. This is the consequence of a complex and indirect interaction between IFN, and the cholinergic pathway that involves neurogenic and myogenic pathways, the cytoskeleton, pertussis sensitive G regulatory proteins and PI turnover. The participation of the gut mucosal immune system in the regulation of intestinal function has been demonstrated. 26 In the normal gut there is a balance in the cytokine network. Under inflammatory conditions or chronic antigenic stimulation at the local level, this balance is disrupted. 27 Lymphokines produced by T-lymphocytes could synergize or antagonize the effects of neurotransmitters or other cytokines 11,28 and thus influence the inflammatory response of the intestine. Thus, the contractile effects of IFN, described in this study could play a role in chronic inflammatory bowel diseases or in HIV diarrhoea, when the balance of the local immune system is disturbed. 29