Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Enteric immune/inflammatory cells are putative sources of paracrine signals to the enteric nervous system (ENS). Most is known about signaling between mast cells and the neural elements of the local microcircuits of the ENS. Mast cells contain a variety of preformed chemical mediators, including histamine. They are stimulated by antigens to secrete histamine. Antigen stimulation involves receptors for antibodies on the mast cells. When the receptors are occupied by antibodies to a sensitizing antigen, and cross-linking occurs by interaction of the sensitizing antigen with the bound antibody, the mast cells release histamine. Intestinal mast cells proliferate during infection of the intestine with nematode parasites such as Trichinella spiralis and Nippostrongylus brasiliensis. Animal models infected with these parasites as well as food allergy models using hypersensitivity to milk protein have proved informative in studies on mast cell involvement in enteric immunoneural communication (Frieling et al., 1994a,b). In these models, recognition of the antigen by antibodies bound to the sensitized mast cells triggers release of histamine and other mediators. The mediators then become messengers to the ENS, which responds by suppressing other programs in its library and running a program for intestinal behavior adapted for elimination of the antigen from the lumen. The neural program integrates copious mucosal secretion of H₂O and electrolytes with powerful motor propulsion (Wood, 1993, 1998). In this respect, intestinal mast cells are uniquely equipped and situated to recognize agents that threaten whole body integrity and signal the ENS to program an appropriate defensive response.

Several mast cell-derived mediators share common neuropharmacologic actions on electrical and synaptic behavior of the ENS. These include histamine, 5-hydroxytryptamine (Wood and Mayer, 1978), interleukin-1B, interleukin-6 (Xia et al., 1999), leukotrienes (Frieling et al., 1997), prostaglandins (Dekkers et al., 1997), tumor necrosis factor  (Xia et al., 1995), and platelet-activating factor (Xia et al., 1996b). Histamine, which is the focus of this study, is not localized to any extent in enteric neurons and is not considered as a putative neurotransmitter in ENS microcircuits (Panula et al., 1985). Mast cells are the principal source of histamine in the intestine.

Wood and Mayer (1975) reported histamine-evoked excitation of neurons in the myenteric plexus of cat small intestine. Subsequent work in the guinea pig found that the excitatory effects of histamine mimic slow excitatory postsynaptic potential (sEPSP) in after hyperpolarization (AH)-type enteric neurons (Nemeth et al., 1984; Tamura and Wood, 1992; Frieling et al., 1993). The changes in electrical behavior during the sEPSP include depolarization of the membrane potential, increase in the electrical resistance of the membrane, and enhanced excitability reflected by repetitive spike discharge. In addition, long-lasting hyperpolarizing afterpotentials of several seconds duration in AH-type neurons are suppressed to permit repetitive spike discharge. Transduction of slow synaptic signals involves activation of adenylate cyclase and second messenger function of cAMP (Palmer et al., 1986, 1987).

Histamine H₂ receptors are the mediators of the slow excitatory response to histamine in cell bodies of enteric neurons in the guinea pig (Nemeth et al., 1984; Tamura and Wood, 1992; Frieling et al., 1993). Exposure to histamine elevates levels of cAMP in myenteric ganglia, and this action is also blocked by selective histamine H₂ receptor antagonists and mimicked by selective agonists (Xia et al., 1996a).

The ENS neurons that respond to histamine are known to express multiple K⁺ channels, including a delayed rectifier, an A-type channel, a Ca²⁺-activated channel, and an inward rectifier (Hirst et al., 1985; North and Tokimasa, 1987; Galligan et al., 1989; Zholos et al., 1999). These channels are collectively responsible for setting resting membrane potential and determining action potential frequency and duration (reviewed by Wood, 1989, 1994). Voltage-activated K⁺ conductance appears to determine the duration of the action potential in AH neurons because the repolarization phase is prolonged by either 4-aminopyridine (4-AP) or tetraethylammonium (Tamura and Wood, 1989). Stimulation of Ca²⁺-activated K⁺ current in AH neurons accounts for long-lasting hyperpolarizing afterpotentials that lengthen the refractory period and thereby limit the frequency of spike discharge by the somal membrane (North, 1973; Wood and Mayer, 1978). Conversion from low to high excitability in AH neurons during sEPSPs has been suggested to involve suppression of A-current (Wood, 1989, 1994). Nevertheless, the physiological significance of A-type K⁺ current (I_A) in the neurons of the ENS is not well understood.

The aim of this study was to test the hypothesis that the sEPSP-like actions of histamine on enteric neurons include effects on I_A. A preliminary report of the results has been published in abstract form (Starodub et al., 1998).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Enteric Neural Cultures. Short-term myenteric neuronal cultures were used for the patch-clamp studies. Initiation of the cultures was based on established methodology for the adult guinea pig (Hanani et al., 1994). The small intestine was removed from male Hartley albino guinea pigs (300-350 g), sectioned into 5-cm-long pieces, and placed over glass rods to facilitate removal by microdissection of the longitudinal muscle together with the adherent myenteric plexus. Myenteric ganglia were enzymatically dissociated from longitudinal muscle-myenteric plexus preparations as described earlier (Xia et al., 1991). Strips of longitudinal muscle with myenteric plexus attached were placed in an enzyme solution composed of 20 mg of collagenase type IA, 15 mg of protease type IX, and 5 mg of deoxyribonuclease I in 15 ml of Krebs' solution. Digestion and dissociation were allowed to proceed for 15 to 25 min at 37°C in a shaker bath. The digested tissue was washed several times with ice-cold Krebs' solution containing 5% antibiotic-antimycotic mixture, and then transferred into polystyrene Petri dishes (15 mm × 100 mm). Dissociated ganglia were collected with suction pipettes under microscopic control (20× magnification, Wild Heerbrugs M4 stereomicroscope; Wild Heerbrugs, Basel, Switzerland). The ganglia, with no visible smooth muscle present, were transferred into medium 199 supplemented with 15% L-glutamate, 10% heat-inactivated fetal calf serum, 33 mM glucose, 1% Penn-Strep solution (10,000 U penicillin and 10 mg/ml¹ streptomycin), and 0.5% gentamycin. The ganglia were then transferred onto 22 × 22 mm coverslips at the bottom of 33-mm plastic Petri dishes. Each dish received 30 to 50 ganglia. About 50% of the ganglia attached to the surface of the coverslip overnight and were used for patch clamp studies the next day.

Patch-Clamp Methods. The coverslips with attached ganglia were washed free of culture medium and placed into a custom made recording cell mounted on the stage of an inverted microscope with Hoffmann modulation contrast optics, epi-fluorescence attachments, and a 35-mm camera (Diaphot 300; Nikon, Tokyo, Japan). The perforated patch configuration (Horn and Marty, 1988) was used to record whole cell currents. Preservation of the intraneuronal milieu was accomplished by permeabilizing the membrane with amphotericin B after gigaseal formation. Serial resistances below 5 M were achieved within 5 to 15 min after stable contact between the cell membrane and pipette tip. No swelling of the cells was ever observed during electrophysiological recording. Pipettes were fabricated from borosilicate glass capillary tubes (7052; World Precision Instruments, Sarasota, FL) on a Flaming/Brown Model P-97 micropipette puller (Sutter Instruments, San Francisco, CA). Tip resistances were ~1 M. An Ag-AgCl reference electrode was connected to the bath through an agar bridge saturated with KCl solution. Ionic currents were recorded and voltage clamp test pulses were applied with an Axopatch 200 amplifier and Labmaster interfaced to a 486 MHz PC computer with pClamp software (Axon Instruments, Foster City, CA). The experiments were done at room temperature (22-25°C). Averaged data are given as the mean ± S.E. Student's paired t test was used for statistical comparison and differences were accepted as significant for P < .05.

Neuronal Identification. At the end of each electrophysiological experiment, a map illustrating the position of the neurons was sketched. The neurons were injected with Lucifer yellow from a separate pipette and photographed (Zholos et al., 1999). The presence of calbindin, as a marker for AH/Dogiel morphologic type II neurons, was determined with standard immunohistochemical methods, as described previously for our patch clamp studies (Zholos et al., 1999). Coverslips with the attached ganglia were fixed in a solution of 2% formaldehyde, 0.2% picric acid, and 0.1 M sodium phosphate buffer at pH 7.0 overnight at 4°C. The fixative was removed by three washes in dimethyl sulfoxide (10 min) followed by three washes in phosphate-buffered saline (10 min; pH 7.0). The preparations were then incubated overnight at 37°C with anticalbindin antibody (monoclonal anticalbindin-D28K; Sigma, St. Louis, MO) diluted 1:150. The antibody complex was visualized as insoluble end product produced with a 3-amino-9-ethylcarbazole staining kit (Sigma).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

A-type K⁺ Current. I_A was identified by clamping the voltage at 50 mV with the neurons in Ca²⁺- and Na⁺-free external bathing solution. Depolarizing voltage steps from 40 to 50 mV evoked an outward current identified in an earlier study as delayed rectifying K⁺ current (Zholos et al., 1999; Fig. 1A). A series of depolarizing steps from 40 to 50 mV starting at the end of hyperpolarizing prepulses to 110 mV resulted in a family of transient outward currents (I_A) in addition to the delayed rectifier current (Fig. 1B). The transient outward current traces were separated from the delayed rectifier current by subtracting the traces obtained by depolarizing steps from a holding potential of 50 mV from traces obtained after a 1-s hyperpolarizing prepulse to 110 mV (Fig. 1C). Unmasking in this manner revealed I_A in both calbindin-positive and -negative neurons.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Methods used for analysis of properties of IA in myenteric neurons of guinea pig small intestine. A, family of outward currents activated by a series of depolarizing steps from a holding potential of 50 mV. B, family of outward currents activated after a prepulse to 110 mV for 1 s. C, A-type current obtained by digital subtraction of A from B. D, suppression of A-type current by 10 mM 4-AP. A, B, and C were obtained from the same neuron in a bathing solution with depleted Ca2+ and Na+.

Histamine. I_A was activated at 15-s intervals by stepping to 20 mV after hyperpolarizing preconditioning pulses. Application of histamine in the bathing solution reversibly suppressed I_A (Fig. 2, A and B). This occurred in 31 of the 52 neurons studied. Calbindin immunoreactivity was present in 27 of the 31 neurons in which histamine suppressed I_A. This indicated that only 2 of the 21 neurons, in which histamine did not suppress I_A, expressed immunoreactive calbindin. The majority of the neurons, for which histamine suppressed I_A, were presumably AH-type neurons based on the presence of calbindin, which is a generally accepted marker for this kind of myenteric neuron (Iyer et al., 1988). Apparently, the calbindin-negative neurons that did not show effects of histamine were S-type myenteric neurons. Histamine is known not to mimic slow synaptic excitation in S-type neurons (Nemeth et al., 1984; Tamura and Wood, 1992).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Suppression of IA by histamine in myenteric neurons from guinea pig small intestine. A, superposition of a series of A-type currents in control and in the presence of 10 nM, 100 nM, and 1 µM histamine. Suppression of the current was concentration-dependent with maximum suppression at 1 µM. B, A-type current was activated at 15-s intervals by stepping the membrane potential from a holding potential of 50 to 20 mV for 200 ms after a conditioning prepulse to 110 mV for 1 s. Data points for each 15-s interval merge into a single trace. Histamine suppression of the current developed within 5 min after each increase in concentration from 10 nM to 1 µM.

1

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effects of histamine on inactivation kinetics of IA in myenteric neurons from guinea pig small intestine. Steady-state inactivation properties were determined by applying 1-s conditioning steps to potentials between 120 and 60 mV, followed by activating voltage steps to 20 mV. Normalized current is plotted as a function of the conditioning voltage. A, inactivation curve. Data points represent mean normalized value of peak current ± S.E. for eight neurons. The Boltzmann equation was used to fit the curve to the data points. Inactivation kinetics were voltage-dependent and accelerated with stronger depolarizing steps. B, histamine in concentrations of 10 nM, 100 nM, and 1 µM did not significantly alter the slope factors for steady-state inactivation curves.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   The selective histamine H2 receptor agonist, dimaprit, suppressed IA in myenteric neurons from guinea pig small intestine. A, A-type current was activated at 15-s intervals and the data plotted as in Fig. 2B. Dimaprit-evoked suppression of the current developed within 4 min after application of 10 nM and an increase to 100 nM dimaprit. B, superposition of a series of A-type currents in control and in the presence of 10 and 100 nM dimaprit. Suppression of the current was concentration-dependent with maximum suppression at 100 nM. C, dimaprit in concentrations of 10 nM, 100 nM, and 1 µM did not significantly alter the slope factors for steady-state inactivation curves.

View larger version (8K): [in this window] [in a new window]   Fig. 5.   Concentration-effect curves for suppression of IA by histamine and dimaprit were similar. A, concentration-effect curve for histamine; EC50 was 10 nM. B, concentration-effect curve for dimaprit; EC50 was 16 nM. Data points are means ± S.E. for eight neurons.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Suppression of IA by histamine and dimaprit was blocked by the selective histamine H2 receptor antagonist cimetidine. A-type current was activated at 15-s intervals, and the data was plotted as in Fig. 2B. A, cimetidine (100 µM) was added to the bathing solution 5 min before addition of 100 nM histamine. Cimetidine blocked suppression of IA by histamine. Addition of 100 nM histamine alone, after washout of the previous application of cimetidine and histamine, suppressed IA. B, cimetidine (100 µM) was added to the bathing solution 5 min before addition of 100 nM dimaprit. Cimetidine blocked suppression of IA by dimaprit. Addition of 100 nM dimaprit alone, after washout of the previous application of cimetidine and dimaprit, suppressed IA.

Elevation of cAMP. Several lines of evidence suggest that the histamine H₂ receptor on myenteric neurons is a metabotropic receptor coupled by G-proteins to adenylate cyclase (Wood et al., 1994). Histamine elevates cAMP in dissociated guinea pig myenteric ganglia, and this effect is potentiated by pretreatment with phosphodiesterase inhibitors (Xia et al., 1996a). Intracellular studies with "sharp" microelectrodes found that elevation of cAMP by histamine or forskolin, or exposure to membrane-permeant analogs of cAMP elevated excitability in myenteric neurons and closely mimicked slow synaptic excitation (Nemeth et al., 1984; Palmer et al., 1986). This led us to test, in a preliminary way, whether elevations of cAMP by forskolin, treatment with IBMX, or exposure to the membrane-permeant analog CPTcAMP had effects on I_A.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Stimulation of adenylate cyclase with forskolin suppressed IA without effects on steady-state inactivation kinetics in a manner similar to the action of histamine or dimaprit. A, suppression of IA by 10 µM forskolin. B, steady-state inactivation curves in the presence and absence of 10 µM forskolin. Inactivation curves were determined as described for Fig. 3. The slope factor for the inactivation curve was unchanged during suppression of IA by forskolin. Data points are means ± S.E. for eight neurons.

View larger version (10K): [in this window] [in a new window]   Fig. 8.   Elevation of intraneuronal cAMP, either with the membrane-permeant analog CPTcAMP or the phosphodiesterase inhibitor IBMX, suppressed IA. A, suppression of IA by 100 µM CPTcAMP. B, suppression of IA by 0.5 mM IBMX.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Histamine is known to be released from enteric mast cells during type I hypersensitivity responses and to have paracrine effects that mimic slow synaptic excitation in myenteric neurons (Wood, 1993, 1998). The mechanism of its action includes suppression of Ca²⁺-activated K⁺ conductance (Nemeth et al., 1984; Baidan and Wood, 1993) and activation of Cl conductance (Starodub and Wood, 1999, 2000). The repetitive spike discharge that occurs during histamine-evoked excitation in AH neurons was postulated to involve suppression of I_A (Wood, 1989, 1994). Our testing of this hypothesis found that histamine acts on I_A in enteric neurons differentially. It suppresses I_A in AH-type neurons (i.e., calbindin-positive neurons) while sparing I_A in S-type neurons (i.e., calbindin-negative neurons). The effect of histamine was concentration-dependent and equivalent in strength to the effect of a selective H₂ agonist, dimaprit. The selective H₂ receptor antagonist, cimetidine, blocked the effects of both histamine and dimaprit. These results suggest that the action of histamine on I_A is mediated by the H₂ receptor subtype.

The potency of histamine in suppression of I_A is equivalent to its potency in the elevation of cAMP levels in myenteric ganglia, which also involves the histamine H₂ receptor subtype (Xia et al., 1996). Our finding that elevation of intraneuronal cAMP appeared to mimic the effects of histamine on I_A is consistent with the evidence that the signal transduction mechanism for activation of histamine H₂ receptors on AH-type myenteric neurons involves stimulation of adenylate cyclase and second messenger function of cAMP. Nevertheless, this conclusion is tentative because the results with forskolin, IBMX, and CPTcAMP are equivocal until studies are done to show that intraneuronal blockade of protein kinase A suppresses the action of histamine. Work of this nature was beyond the scope of this study. On the other hand, it is known that suppression of adenylate cyclase by adenosine A₁ receptor agonists prevents elevation of cAMP by histamine in myenteric ganglia (Xia et al., 1997).

Suppression of I_A by histamine is expected to have the effect of contributing to the enhanced sEPSP-like excitability that occurs in AH-type neurons in response to histamine. Augmented excitability in response to histamine is reflected by membrane depolarization, repetitive spike discharge, and increased amplitude and lengthening of fast nicotinic EPSPs. Steady-state activation and inactivation curves for I_A overlap at membrane potentials of ~50 mV in calbindin-positive myenteric neurons (A. Starodub and J. Wood, unpublished data). This suggests that I_A might contribute to the resting potential. If so, suppression of the current could contribute to the depolarization and increased input resistance evoked by histamine.

Action potentials that make-up the repetitive firing seen during both slow synaptic excitation and the sEPSP-like action of histamine are preceded by ramp-like prepotentials (Tamura and Wood, 1992; Wood, 1992; Frieling et al., 1993). The prepotentials fall within the lower range of activation voltage for I_A. Consequently, suppression of I_A by histamine would be expected to facilitate the rates of rise of the prepotentials to action potential threshold and thereby contribute to increased frequency of repetitive spike discharge.

Some, but not all, AH-type myenteric neurons receive fast excitatory nicotinic synaptic input (e.g., Grafe et al., 1979; Tamura and Wood, 1989; Wood, 1989). These depolarizing responses extend into the range of activation potentials for I_A where it would be expected to truncate the EPSPs. Suppression of I_A would remove some of the braking action of the outward current and thereby lead to augmentation of the EPSP.

Conclusions

Histamine acts at histamine H₂ receptors to suppress I_A in AH-type myenteric neurons in guinea pig intestine. Suppression of I_A is part of the ionic mechanism responsible for elevation of excitability during slow synaptic excitation and sEPSP-like responses evoked by excitatory paracrine mediators, such as histamine.