Characterization of transmitters involved in electrically-evoked NANC relaxations
The presence of NO and VIP in myenteric neurones and fibres supplying smooth muscle cells suggests the possibility that both inhibitory transmitters participate in neurogenic NANC relaxations, accounting for the reservoir function of this stomach region. Nevertheless, the colocalization of different transmitters within a neural structure does not necessarily imply their concomitant release at any level of electrical or mechanical stimulation. NO-mediated relaxations are preferentially activated by shorter trains or lower frequencies of electrical field stimulation in both gastric (Li & Rand, 1990) and extra-gastric isolated preparations (Maggi & Giuliani, 1996). Conversely, VIP is released either by brief stimuli at high frequency or by sustained stimuli at low frequency (Lefebvre et al., 1995; Currò & Preziosi, 1998). In the human isolated gastric fundus strips, our results with L-NOARG indicate that electrically-induced NANC relaxations are mediated by NO in the frequency range of 0.3–3 Hz with trains lasting 5 s. Even with longer trains (5 min) at low frequency (1 Hz) NO seems the only inhibitory transmitter involved. In this respect, the characteristics of stimulation for NO release parallel those of other established transmitters, such as acetylcholine or noradrenaline, which require a lower level of stimulation than any other accompanying peptide transmitter (Bartfai et al., 1988). By contrast, L-NOARG only significantly reduced (without suppressing) relaxations obtained in the range of 10–50 Hz, suggesting the participation of other inhibitory transmitter(s) in addition to NO.
A VIP desensitization procedure was used to assess whether VIP participates in the L-NOARG-resistant component of NANC relaxation. The attenuation of receptor function by means of agonist-induced receptor desensitization is considered a reliable alternative to the use of antagonists, especially when the latter lack potency and selectivity, such as the C-terminal fragment of VIP, VIP(10–28), which may (Crist et al., 1992; Jin et al., 1994) or may not antagonize relaxations to applied VIP (Morris & Murphy, 1989; Maggi & Giuliani, 1993). When desensitization was induced to the relaxant effect of VIP, it also markedly reduced the relaxant effects of electrical field stimulation (range 10–50 Hz) in the presence of L-NOARG. Compared to control, the relevant inhibition of NANC relaxations observed under these conditions (90%) suggests that at frequencies 10 Hz, relaxations are mostly mediated by the concomitant release of NO and VIP, which are costored in neurones innervating smooth muscle cells, as observed in our immunohistochemical experiments. In our hands, VIP desensitization probably involved VIP receptors specifically (and not other receptor types or post-receptor mechanisms), since isoprenaline-induced relaxation, which is mediated via cyclic AMP formation like that of VIP (Bitar & Makhlouf, 1982), was not affected by this procedure. Furthermore, the finding that the relaxation induced by VIP was insensitive to L-NOARG could be taken as evidence that in human gastric fundus strips, like in pig or canine fundus strips (Lefebvre et al., 1995; Bayguinov et al., 1999) or guinea-pig isolated whole stomach (Desai et al., 1994), VIP does not seem to produce part of its relaxant effect by releasing NO from neurones or directly from isolated smooth muscle cells (Grider et al., 1992; Murthy et al., 1993; Chakder & Rattan, 1996; Jin et al., 1996). However, since human intestinal and rabbit gastric smooth muscle cells were recently found to express a constitutive endothelial NOS (Teng et al., 1998), experiments with human dispersed smooth muscle cells from fundus are required to establish whether VIP is able to promote NO release under these conditions.
With regard to the L-NOARG-resistant component of NANC relaxations evoked at the highest frequencies (50 Hz), it was slightly, but significantly reduced by suramin, a non-selective P2 purinoceptor antagonist (Hoyle et al., 1990). Although this could be taken as evidence for the involvement of ATP as an additional mediator of NANC relaxations, we believe that ATP plays no role in our experimental conditions for several reasons. Firstly, combined L-NOARG and VIP desensitization virtually abolished NANC relaxations; secondly, in our hands suramin antagonized VIP-induced relaxations, as observed by Briejer et al. (1995) in the guinea-pig proximal colon. Therefore, the suramin-sensitive component of NANC relaxation could be tentatively ascribed to partial VIP inhibition. In our hands, however, this minor inhibition by suramin was detectable only after suppression of the nitrergic function, since suramin was ineffective when the nitrergic innervation was fully operating.
At frequencies 3 Hz, L-NOARG unmasked an early contractile response that was substantially reduced by TTX and hyoscine. This suggests that endogenous NO may tonically inhibit the release of excitatory transmitters. Prejunctional modulation of cholinergic transmission by endogenous NO has been reported in the guinea-pig and dog intestine and in rabbit gastric corpus (Knudsen & Tøttrup, 1992; Baccari et al., 1993; Hryhorenko et al., 1994), while evidence for a postjunctional mechanism was obtained in the guinea-pig (Milenov & Kalfin, 1996) and pig gastric fundus (Leclere & Lefebvre, 1998). In addition, as observed in isolated intestinal preparations from several mammalian species (Li & Rand, 1990; Meulemans et al., 1993; Ciccocioppo et al., 1994), L-NOARG enhanced the basal tone of human fundus strips through a mechanism that was partially sensitive to TTX and hyoscine. This suggests that L-NOARG acts by removing a tonic nitrergic inhibition both on smooth muscle cells and excitatory cholinergic neurones.
Relaxations to graded mechanical distension
Electrical field stimulation is generally regarded as a reliable procedure to evoke transmitter release from enteric nerve terminals, although its physiological relevance is still debated. In our study, we also resorted to mechanical distension to assess the role of the neural mechanisms underlying NANC relaxation.
Although relaxations of gastric fundus strips to graded distension were mainly due to passive elongation of smooth muscle cells, a neurogenic component (∼25%) apparently mediated only by NO was observed at 2 g distension. Based on our results, NO is released upon mild electrical or mechanical stimulation. Therefore, NO could be viewed as the primary inhibitory transmitter initiating accommodation in the human gastric fundus, as already observed in other species (Desai et al., 1991) and in other parts of the gut (Waterman et al., 1994; Ciccocioppo et al., 1994).