Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Prejunctional (presynaptic) histamine H₃ receptors and alpha-2 adrenoceptors, localized on nerve endings of the peripheral nervous system, negatively modulate the release of different neuromediators, including ACh from the myenteric plexus (Trzeciakowski, 1987; Poli et al., 1991; Coruzzi et al., 1991) and NE from cardiac nerves or sympathetic neurons (Lipscombe et al., 1989; Malinowska and Schlicker, 1993; Endou et al., 1994). The activation of such receptor systems limits both the action potential upstroke and the availability of free Ca⁺⁺ at the axoplasmic level, possibly by reducing Ca⁺⁺ transport through the neuronal-type Ca⁺⁺ channels (Endou et al., 1994; Poli et al., 1994; Schlicker et al., 1994) or by activating ATP-sensitive K⁺ channels (Ohkubo and Shibata, 1995). Irrespective of the mechanism, the activation of H₃ and alpha-2 receptors results in an attenuation of the exocytotic neurotransmitter release and, consequently, of the postsynaptic responses associated with nerve stimulation.

It has recently been demonstrated that histamine H₃ receptors and alpha-2 adrenoceptors, occurring at the noradrenergic nerve endings of the brain cortex, interact with each other in the modulation of NE release, showing a negative cooperativity when simultaneously activated by selective agonists (Schlicker et al., 1992). Moreover, the effect induced by an H₃ receptor agonist is potentiated when the alpha-2 adrenoceptor is simultaneously blocked, which suggests that the negative cooperativity also occurs between an exogenously added H₃ receptor agonist and endogenous NE (Schlicker et al., 1992).

We have recently employed a functional test to study peripheral histamine H₃ receptors and alpha-2 adrenoceptors in the gut (Coruzzi et al., 1991; Poli et al., 1993; Poli et al., 1994). These receptors negatively modulate the cholinergic nerve activity in the isolated, electrically driven guinea pig duodenum without affecting muscle contractions evoked by exogenous ACh (Coruzzi et al., 1991). It was also shown that the activation of H₃ receptors and alpha-2 adrenoceptors, but not of adenosine A₁ receptors, results in a Ca⁺⁺-dependent inhibition of neurogenic responses (Poli et al., 1994). These finding suggested a postreceptor mechanism of prejunctional inhibition coupled with H₃ and alpha-2 receptors, possibly consistent with a restriction of Ca⁺⁺ fluxes through neuronal-type channels (Poli et al., 1994).

Thus the purposes of the present study were 1) to test the hypothesis that alpha-2 adrenoceptors and H₃ receptors interact in modulating the activity of intestinal cholinergic nerves and 2) to demonstrate possible interactions between the former two receptors and adenosine A₁ receptors.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

The inhibitory effects mediated by histamine H₃ receptors and other prejunctional receptor systems on the cholinergic transmission were studied, using the isolated guinea pig duodenum, according to a previously described technique (Coruzzi et al., 1991; Poli et al., 1993; Poli et al., 1994).

Animal procedure. Male albino guinea pigs (300-400 g) were killed by cervical dislocation. The whole duodenum was rapidly removed and placed into a dissection disc containing oxygenated (95% O₂ and 5% CO₂) Krebs-Henseleit solution of the following composition (mM): NaCl 113; KCl 4.7; CaCl₂ · 2H₂O 2.5; KH₂PO₄ 1.2; MgSO₄ · 7H₂O 1.2; NaHCO₃ 25 and dextrose 11.5 (pH 7.4-7.5).

EFS. A pair of coaxial platinum electrodes were positioned 10 mm apart the longitudinal axis of the preparations. Single square-wave pulses (0.5 msec in duration, 50 V at 150-200 mA intensity) were delivered to the tissues every 10 sec. For each experiment, the intensity was adjusted to the level that gave 70% to 80% of the maximum tissue response to EFS. In these conditions, tetrodotoxin- and atropine-sensitive twitch responses were obtained (Coruzzi et al., 1991), a result that suggests the involvement of ACh released from cholinergic neurons of the myenteric plexus.

Evaluation of drug effects. Agonists were cumulatively administered by addition of logarithmically increasing concentrations to the bath fluid, or by single administration of EC₅₀ concentrations (those that gave 50% of the absolute maximum effect obtainable with the agonist, irrespective of its efficacy). When MHA was used, the experiments were carried out in the presence of H₁ and H₂ receptor blockers (pyrilamine and famotidine 1 µM) to prevent the activation of H₁ and H₂ receptors by concentrations of the compound higher than 10 µM (Arrang et al., 1987; Endou et al., 1994). To avoid possible desensitization phenomena at the level of H₃ receptors (Coruzzi et al., 1991), a single concentration-response curve to MHA was carried out in the same preparation. In the case of clonidine, UK 14,304, adenosine and CPA, it was possible to construct several concentration-response curves in the same preparation, provided that 60 to 90 min elapsed between drug administrations.

Evaluation of receptor interactions Protocol I. The interaction between two different prejunctional receptors was evaluated on the basis of the effects elicited by the respective agonists, administered alone or in combination. Preparations were stimulated as described above. Then ED₅₀s of two agonists (agonist A and agonist B) were tested separately, and the respective effects (effect A and effect B) were measured. After washout and complete recovery of contraction to EFS (60-90 min), the agonists were coadministered, and we obtained the combined effect (effect [A + B]). Then "effect [A + B]" was compared with the algebraic sum "effect A + effect B," which represents the expected effect for additivity. Significant deviations from additivity, determined by statistical comparison (see below), were taken as evidence of cooperativity between receptors.

Protocol II. Other kinds of experiments were performed to test whether previous activation of a prejunctional receptor could modify the effects evoked by activation of a second receptor. In these experiments, the current strength of the electrical pulses was calibrated at a level (~150 mA) that gave 55% to 60% of the maximum tissue responsivity to EFS. Typically, we constructed the concentration-response curve of an agonist and then left the tissue to recover after washout before applying a single concentration of a second agonist. Once the maximum effect was reached, the current strength was raised (usually from 150 to 270 mA) to restore a predrug level of twitch response, after the application of the second agonist, and then the concentration-response curve of the former agonist was reconstructed in the presence of the second agonist (see fig. 3 for more details). The adjustment of the current strength was made to compensate for inhibition of the electrically evoked response by one of the two interacting agonists and hence to reproduce a predrug level of neurogenic contraction. This procedure was designed to exclude the possibility of experimental artifacts due to the decrease of neurogenic response and to make it possible to evaluate a true receptor interaction (Schlicker et al., 1992).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Typical experiment showing the effects of UK 14,304 before and after near-maximum concentrations of MHA and reversal of inhibitory effect of MHA by raising current strength (from 150 to 270 mA). Concentrations of drugs are expressed in nanomoles per liter. Vertical calibration represents the isotonic shortening of duodenal muscle in response to EFS. x: stop stimulation and washout.

Statistical analysis. Results are given as means ± S.E.M. Comparisons between two sets of data were made by Student's t test for paired or unpaired data. When more than two groups were compared, the analysis of variance was used. In all these tests, P < .05 was considered statistically significant. The potency of the different agonists on EFS-induced contractions was expressed by the EC₅₀ value (see above). Confidence limits 95% (CL₉₅) of the geometric mean of the individual EC₅₀ values were then provided.

Drugs. The following drugs and chemicals were used in this study: yohimbine hydrochloride, pyrilamine maleate (mepyramine), atropine sulfate, tetrodotoxin (citrate buffer) and adenosine (base) (Sigma Chemical Co., St. Louis, MO), THEO (Merck, Darmstadt, FRG) and CPA, UK 14,304 (5-bromo-N-(4,5-dihydro-1H-imidazol-2-y1)-6-quinoxalinamine, also known as bromoxidine), idazoxan hydrochloride, 8-cyclopentyl-1,3-dipropylxantine (free base) and thioperamide maleate (RBI, Natick, MA). Famotidine (free base) was a gift of Sigma-Tau, Roma, Italy; (R)--methylhistamine dihydrochloride was generously provided by Dr. J. C. Schwartz (Centre Paul Broca de l'INSERM, Paris, France); and the H₃ receptor ligands immepip dihydrobromide (VUF 4708) and clobenpropit dihydrobromide (VUF 9153) were synthesized by Prof. H. Timmermann (Vrije Universiteit, Amsterdam, the Netherlands).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects mediated by prejunctional H₃, alpha-2 and A₁ receptors. The activation of H₃ alpha-2 and A₁ receptors by a series of specific or nonspecific agonists induced a concentration-dependent inhibition of EFS-evoked contractions of the guinea pig duodenum. The efficacy of the H₃ receptor agonist MHA (fig. 1 and table 1) was lower (~60% of maximum inhibition of EFS-evoked twitch responses), compared with the alpha-2 agonist UK 14,304 or with the A₁ agonist CPA, both of which compounds produced an almost complete damping of the cholinergic response. Also, the new agonist for H₃ receptors, immepip (Barnes et al., 1993; Vollinga et al., 1994), incompletely suppressed the cholinergic response. Conversely, a complete inhibition was attained with the endogenous A₁/A₂ receptor agonist adenosine, and the alpha-2 agonist clonidine showed a partial agonistic activity, compared with UK 14,304. The potency and the efficacy of these agonists are listed in table 1. None of these compounds modified the contractile effect of exogenous ACh, even at the maximum inhibitory concentrations affecting EFS-evoked responses. An example is shown in figure 1, which considers only the effect of CPA.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Left plot: Inhibitory effect of CPA, UK 14,304, MHA and immepip on the amplitude of twitch responses of the isolated guinea pig duodenum. Data are the mean ± S.E.M. of 5 to 8 experiments. On abscissa: molar concentration of drugs; on ordinate: percent of the predrug contractions evoked by EFS, taken as 100%. Right panel: Original recordings showing the effects of CPA on the cholinergic responses evoked by EFS (upper) and by exogenous ACh. Vertical calibration represents the isotonic shortening of duodenal muscle in response to EFS. Concentrations of drugs are expressed in nanomoles per liter. x: stop stimulation and washout; w: washout.

Interactions among H₃, alpha-2 and A₁ receptors. From the concentration-response curves of figure 1, we extrapolated the concentrations of prejunctional receptor agonists giving 10% to 15% of the respective maximum effects, and the following values were chosen for MHA, CPA and UK 14,304, respectively: 3 nM, 0.5 nM and 3 nM.

View larger version (41K): [in this window] [in a new window]   Fig. 2.   Inhibitory effect of CPA (grid columns), UK 14,304 (UK) (white columns) and MHA (horizontally barred column) obtained by single administration or by coadministration (black columns) of two of these agonists. A) The following concentrations, which represent the approximate EC10-15 of each agonist, were used: MHA: 3 nM; UK 14,304: 3 nM; CPA: 0.5 nM. B) The EC50 value of each drug was used (see table 1). Hatched columns represent the inhibition value expected for additivity of the individual effects. Data are the mean ± S.E.M. of 6 to 9 experiments. On ordinate: percent of response to EFS, considering the predrug level to be 100%. * indicates significant difference (P < .05) from the value expected for additivity.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Inhibitory effect of CPA, UK 14,304 and MHA, first given alone and subsequently administered in the presence of each of the other two agonists. Control curves (Co) were constructed to low-current strength (typically 150 mA); the combined effect was tested after reversal of inhibition by raising the current strength at 270 mA, according to the scheme represented in figure 3. Data are the mean ± S.E.M. of 5 to 8 experiments. On abscissa: molar concentration of drugs; on ordinate: percent of the predrug contractions evoked by EFS. * indicates significant difference (P < .05) from the corresponding control value.

Effect of specific blockers on the agonist-induced prejunctional effects. To ascertain whether the cooperativity between CPA and UK 14,304 or MHA reflects cross-reactions at the recognition sites of receptors, we tested each of these agonists in the presence of the corresponding receptor blockers and in the presence of blockers acting at the other receptors.

Effects of the H₃, alpha-2 and A₁ receptor antagonists on the EFS-evoked cholinergic response. The possible role of endogenous ligands in modulation of the ACh release, as a factor in the cooperativity between A₁ and alpha-2 or H₃ receptors, was tested with specific blockers for prejunctional receptors on the contractile response to EFS.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Effect of THIO, CLO, DPCPX, THEO, IDAZ and yohimbine (YO) on the amplitude of muscle contractions evoked by EFS. Data represent the mean ± S.E.M. of 5 to 12 experiments. On ordinate: percent of the cholinergic response to EFS, considering the predrug level to be 100%. The concentrations of drugs are expressed in micromoles per liter. n.s.: not statistically significant; *: P < .05, compared with the respective control (Co) value.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The use of specific agonists to activate H₃ receptors, alpha-2 adrenoceptors and A₁ receptors reveals the occurrence of negative cooperativity between some, but not all, of the presynaptic heteroreceptors that control ACh release from the intrinsic excitatory nerves of the isolated guinea pig duodenum. In fact, alpha-2 adrenoceptors and H₃ receptors do not interact with each other, whereas adenosine A₁ receptors interact with both of them. This is suggested by the additivity of the inhibitory actions of the specific alpha-2 adrenoceptor agonist UK 14,304 and the specific H₃ receptor agonist MHA, and by a reduced inhibition when both agonists were combined with the A₁ receptor agonist CPA. These interactions occur when agonists are applied at EC₅₀ or more; at lower concentrations (near to threshold), no evidence of cooperativity between receptors emerged.

Such results at least in part contrast with the view that two (or more) inhibitory presynaptic (prejunctional) receptor systems, occurring at the same nerve endings, display a negative cooperativity. Although the exact molecular mechanism(s) is (are) lacking, examples of such cooperativity can be found between receptors in central as well as in peripheral tissues, including dopamine D₂ receptors and alpha-2 adrenoceptors in perivascular nerves (Friedman and Duckles, 1995) and adenosine A₁ and kappa opioid receptors in cortical nerve endings (Limberger et al., 1988).

Schlicker et al. (1992) found a negative cooperativity between presynaptic alpha-2 adrenoceptors and H₃ receptors that control the release of ³H-NE from adrenergic nerves of the brain cortex. Whether such discrepancy with our results is a consequence of the different experimental conditions (contraction vs. release experiments) cannot be easily established. It is possible that prejunctional/presynaptic receptor interactions preferentially involve the respective autoreceptor and a heteroreceptor, rather than two heteroreceptors.

One factor that could contribute significantly to this difference is the state of activation of the interacting receptors, because it has been shown that the block of activated, but not of nonactivated, alpha-2 receptors potentiates (Schlicker et al., 1989) or unmasks (Schlicker et al., 1990) the inhibitory effect mediated by H₃ receptors on the release of NE from rat and pig tissues. Histamine and purines (either ATP or its metabolite adenosine), together with NE, play a role as inhibitory neuromediators in the gut (for reviews see Burks, 1994; Wood, 1994). Therefore, these substances, when released from noncholinergic nerves or tissue stores, could influence the activity of exogenously applied agonists at the respective receptor.

To investigate this possibility, we performed experiments with antagonists acting at H₃, alpha-2 and A₁ receptors in an attempt to determine whether such receptors are activated by the corresponding endogenously released agonists. THIO and CLO (H₃) and DPCPX (A₁), have a facilitatory effect on the neurogenic contractions, whereas the nonspecific A₁ receptor blocker THEO (at nonspasmolytic concentrations) did not produce any facilitation. The effect of each H₃ antagonist is apparently consistent with H₃ receptor blockade, being absent in preparations pretreated with the other H₃ antagonist. These results, together with the finding that the nonspecific H₃ blocker impromidine may enhance the release of ACh (Poli et al., 1991), suggest that H₃ receptors are activated by endogenous histamine, released in response to EFS. The use of other specific antagonists for A₁ receptors will clarify whether this system is really activated by endogenous purines or whether DPCPX enhances the cholinergic response through a nonspecific mechanism.

On the other hand, the observed negative cooperativity cannot be simply a consequence of interactions with endogenous ligands, because NE does not appear to be released from enteric nerves in response to EFS (perhaps the frequency of stimuli applied to the preparations is too low to activate sympathetic nerve endings), but alpha-2 receptors, much like H₃ receptors, do interact with A₁ receptors.

Schlicker et al., (1992), but not others (Imamura et al., 1994) found a potentiation of the inhibitory activity mediated by H₃ receptors when an alpha-2 adrenoceptor antagonist was added simultaneously. In our conditions, specific antagonists for alpha-2 and H₃ (and also A₁) receptors, modify only the effect mediated by the corresponding receptor, while leaving unaffected the effects mediated by other receptors. Therefore, H₃ receptors and alpha-2 adrenoceptors occurring at the cholinergic endings of the myenteric plexus represent separate entities, which modulate in the same way, and probably with a similar (but not necessarily identical) molecular mechanism, the release of ACh from excitatory nerves of the gut (Poli et al., 1994).

On the basis of these results, we conclude that the negative cooperativity between A₁ receptors and H₃ or alpha-2 receptors is a consequence of interactions taking place at the postreceptor level. Such cooperativity occurs when agonists are coadministered, or when one kind of receptor is stimulated, before the administration of the agonist for the second receptor. Furthermore, we observed that the interaction is of the nonreciprocal kind. In fact, the activation of A₁ receptors, which apparently interact with both alpha-2 and H₃ receptors, blunts only the effects elicited by the activation of H₃ receptors, whereas alpha-2 receptor activation, in turn, limits those of A₁ receptors. It is evident from these findings that the mechanism underlying the interaction between these prejunctional receptor systems differs with the different receptors (and, consequently, with the postreceptor events) involved.

In conclusion, we provide evidence of negative cooperativity between some, but not all, heteroreceptors involved in the control of ACh release from the myenteric plexus, and we found no evidence of positive cooperativity. These interactions occur when receptors are activated by high concentrations (EC₅₀ or more) of exogenous agonists and seem to be due to cross-talk at the postreceptor level. Whatever the mechanism, these interactions could represent a limiting factor in the pathophysiologic modulation of ACh release by substances occurring in the intestinal neuroimmune system, when simultaneously mobilized by neurohumoral and/or immunological stimuli (Wood, 1994).