Mast cell tryptase and proteinase-activated receptor 2 induce hyperexcitability of guinea-pig submucosal neurons

Abstract

Mast cells that are in close proximity to autonomic and enteric nerves release several mediators that cause neuronal hyperexcitability. This study examined whether mast cell tryptase evokes acute and long-term hyperexcitability in submucosal neurons from the guinea-pig ileum by activating proteinase-activated receptor 2 (PAR2) on these neurons. We detected the expression of PAR2 in the submucosal plexus using RT-PCR. Most submucosal neurons displayed PAR2 immunoreactivity, including those colocalizing VIP. Brief (minutes) application of selective PAR2 agonists, including trypsin, the activating peptide SL-NH2 and mast cell tryptase, evoked depolarizations of the submucosal neurons, as measured with intracellular recording techniques. The membrane potential returned to resting values following washout of agonists, but most neurons were hyperexcitable for the duration of recordings (> 30 min-hours) and exhibited an increased input resistance and amplitude of fast EPSPs. Trypsin, in the presence of soybean trypsin inhibitor, and the reverse sequence of the activating peptide (LR-NH2) had no effect on neuronal membrane potential or long-term excitability. Degranulation of mast cells in the presence of antagonists of established excitatory mast cell mediators (histamine, 5-HT, prostaglandins) also caused depolarization, and following washout of antigen, long-term excitation was observed. Mast cell degranulation resulted in the release of proteases, which desensitized neurons to other agonists of PAR2. Our results suggest that proteases from degranulated mast cells cleave PAR2 on submucosal neurons to cause acute and long-term hyperexcitability. This signalling pathway between immune cells and neurons is a previously unrecognized mechanism that could contribute to chronic alterations in visceral function.

Figure 1. Detection of proteinase-activated receptor 2 (PAR2) mRNA in the ileal submucosal plexuses by RT-PCR

Lanes 1–3 correspond to three guinea-pigs (gp) and lane 4 is a control (no RNA).

Figure 2. Localization of immunoreactive PAR2

A, localization of PAR2 in sections (i-iii) and whole mounts (iv-viii) of guinea-pig ileum. Panels i-vi were stained with PAR2 antibody B5 and vii-viii with PAR2 antibody C17. Note PAR2 immunoreactivity at the apical membrane (i, arrow heads) and basolateral membranes (ii, arrow) of enterocytes of the villi and crypts, in the submucosal plexus (sub. plexus, ii, iv, v, vii, arrows) and myenteric plexus (mye. plexus, iii, arrows). Panels vi and vii are whole mounts of submucosal plexus incubated with preabsorbed PAR2 antibodies. Scale bar = 20 μm for i-iv, 30 μm for v-viii. B, localization of PAR2 (B5 antibody, i, iv) and VIP (ii, v) in sections of guinea-pig ileum. Note immunoreactive PAR2 in neurons of the submucosal plexus that were stained with the VIP antibody (arrows). VIP-positive fibres in the musculature were not stained for PAR2 (arrow heads). Scale bar = 25 μm.

Figure 3. Trypsin and the tethered ligand of rat PAR2 (SL-NH₂) depolarize submucosal neurons through PAR2

A, representative depolarizations to superfusion of trypsin (10 nm) or SL-NH₂ (50 μm, solid bars), which desensitized during the 3 min application. The break lines refer to a break in the trace during which time the cell returned to the resting membrane potential, as indicated by the trace after the break lines. Resting membrane potentials were −61 mV for the neuron superfused with trypsin and −51 mV for the neuron superfused with SL-NH2. B, representative recording showing depolarization to SL-NH₂ (50 μm). Superfusion for 3 min desensitized the response to trypsin (10 nm). The resting membrane potential was −52 mV. C, N-ethylmaleimide (NEM, 10 μm) suppressed the inhibitory and slow excitatory synaptic potentials evoked by electrical fibre-tract stimulation (triangle). When trypsin (10 nm) was applied with NEM it did not depolarize the membrane. The action of NEM was not reversible. The resting membrane potential was −58 mV. D, summary of mean depolarizations evoked by PAR2 agonists alone, SL-NH₂ following trypsin desensitization, trypsin following SL-NH₂ desensitization, trypsin incubated with soybean trypsin inhibitor (SBTI), the reverse sequence of the activating peptide (LR-NH₂) and trypsin with NEM. E, concentration–response analysis of the magnitude of depolarization to graded concentrations of trypsin. n= number of neurons.

Figure 4. PAR2 agonists evoke long-term excitation

A, representative trace showing threshold current injection (lower trace) to elicit a single action potential (upper trace) at time 0. Superfusion of trypsin (10 nm) for 3 min evoked typical depolarization. One hour following the washout of trypsin, the intracellular current injection elicited five action potentials. The break lines refer to a brake in the trace during which time the cell returned to the resting membrane potential of −51 mV. B, representative trace showing that SL-NH₂ (50 μm) had a similar action to trypsin, as shown in A. The resting membrane potential was −50 mV. C, one hour following superfusion of trypsin, as in (A), single pulse fibre-tract stimulation (triangle) evoked a fast EPSP that now triggered an action potential. D-F, summary of the mean number of action potentials (D), the percent change in input resistance (E) and the increase in fast EPSPs (increased amplitude of action potential discharge (F) under control conditions and following stimulation with PAR2 agonists. *P < 0.05, **P < 0.01, #P= 0.055.

Figure 5. PAR2-evoked changes in fast EPSPs correspond to alterations in the postsynaptic neuronal membrane

Control responses (A–D, left panels) were obtained from one neuron at time 0 and compared to responses from the same neuron 30 min after superfusion of trypsin (30 nm, 3 min; A–D, right panels). Exposure to trypsin depolarized the membrane potential by 7 mV; the membrane returned to resting potential after washout (not shown). A, threshold current injection (lower trace) elicited a single action potential (upper trace) at time 0 (left). Thirty minutes after trypsin, the same intracellular current injection elicited three action potentials (right). B, input resistance was measured from the electrotonic hyperpolarization (lower trace) evoked by a hyperpolarizing current injection (upper trace) at time 0 (right). Thirty minutes after trypsin, the same hyperpolarizing current evoked a much greater hyperpolarization, reflecting increased input resistance. C, the amplitude of the fast EPSP elicited by fibre-tract stimulation also increased after trypsin exposure and now triggered an action potential. D, the nicotinic agonist 1,1-dimethyl-4-phenyl-piperazinium (DMPP) was applied by pressure-pulse application (5–10 ms, 100 μm) at time 0 and at 30 min after trypsin to determine whether changes in the fast ‘nicotinic’ EPSPs were due to pre- or postsynaptic effects. The amplitude of the depolarization was markedly increased following application of trypsin, demonstrating a postsynaptic action. The resting membrane potential was −55 mV. Representative tracing of three experiments.

Figure 6. Mast cell tryptase and degranulation activate PAR2 on submucosal neurons

A, pressure-pulse application (‘puff’) of tryptase evoked a slow depolarization. Intracellular current injection (lower trace) sufficient to elicit a single action potential (upper trace) prior to tryptase stimulation, 1 h later evoked multiple action potentials. Resting membrane potential was −62 mV. B, tryptase-like activity was barely detectable in aliquots from unchallenged milk-sensitized tissue, but levels were significantly elevated in aliquots from β-lactoglobulin (β-LG)-treated preparations. Activity was significantly inhibited by leupeptin (leu, 1 mm) and di-isopropyl fluorophosphates (DIPF, 1 mm; P < 0.05). n= number of animals. Tryptase activity is expressed as picomoles of p-nitroanilide generated per minute (PNA min⁻¹). C, representative trace showing β-LG-evoked depolarization desensitized responses to trypsin. The resting membrane potential was −56 mV.

Figure 7. Mast cell degranulation evokes long-term excitation

A, superfusion of β-LG (10 μm), which degranulates mast cells in milk-sensitized tissue, evoked a slow depolarization in the presence of cimetidine (cim; 100 μm), pyrilamine (pyr; 1 μm), tropisetron (trop; 1 μm) and indomethacin (ind; 10 μm). The intracellular current injection elicited multiple action potentials 2 h after washout of the β-LG, as seen with tryptase (Fig. 6A) and other PAR2 agonists (Fig. 4). The resting membrane potential was −61 mV. B–D, summary of mean action potential discharge (B), input resistance (C) and magnitude of the fast EPSP (increased amplitude or action potential discharge; D), for controls and following stimulation with β-LG alone or with antagonists. ***P < 0.001, **P < 0.01,*P < 0.05, #P= 0.053.