Morphological characteristics and immunohistochemical detection of nicotinic acetylcholine receptors on intestinofugal afferent neurones in guinea-pig colon

Intestinofugal afferent neurones (IFANs) provide excitatory synaptic input to abdominal prevertebral ganglion neurones. Input is greatly reduced during blockade of nicotinic acetylcholine receptors (nAChRs) in the wall of the colon, suggesting two projection pathways: a direct pathway without synaptic interruption and an indirect pathway interrupted by at least one nicotinic cholinergic synapse. This study aimed to characterize the morphology of IFANs and examine the distribution of nAChRs on them. We identified IFANs in guinea-pig colon by retrograde labelling with fluorescent tracer DiI placed either on the lumbar colonic nerves in vitro or inferior mesenteric ganglion in vivo. Confocal laser scanning microscopy and computerized image-processing software were used for 3D image reconstruction. Approximately 70% of identified IFANs had Dogiel type I-like morphology, the remainder were Dogiel type II-like. In vivo labelled IFANs were injected with Lucifer Yellow and immunostained for nAChRs using monoclonal antibody MAb35. Approximately 3% of total plasma membrane surface of IFANs with Dogiel type I morphology had MAb35-IR. In contrast, <1% of membrane surface of IFANs with Dogiel type II morphology had MAb35-IR. The finding that IFANs displayed immunostaining for nAChRs suggests the presence of putative nicotinic synapses.

Prevertebral ganglia and intestinofugal afferent neurones

Intestinofugal afferent neurones (IFANs) are a unique subset of myenteric ganglion neurones that regulate normal gastrointestinal function. The IFANs relaying mechanosensory information to sympathetic neurones of the prevertebral ganglion (PVG) function as volume detectors. It is possible that mechanosensory information arriving in the PVG via axon collaterals of visceral spinal afferent nerves can be modulated entirely within the PVG itself.

Physiological study of interstitial cells of Cajal identified by vital staining

Interstitial cells of Cajal (ICC) form networks that intercalate between the enteric nervous system and smooth muscle cells and play a fundamental role in the control of gastrointestinal motility by initiating rhythmic electrical activity. In this report, we used a method to examine the physiological and morphological properties of ICC in living, intact tissues. ACK2, an anti-Kit antibody, was conjugated to a fluorescent probe and used to identify individual ICC for intracellular electrical recordings, to record changes in intracellular calcium concentration using fluorescent dyes and for confocal microscopy. Cyclic changes in intracellular calcium concentration were recorded in ICC with a frequency similar to the electrical slow wave. In addition, injection of a fluorescent dye into single ICC enabled the three-dimensional reconstruction of single myenteric plexus ICC within the intact network. The data show that ICC in intact networks from the myenteric plexus region in living tissues in the guinea-pig antrum exhibit an electrical slow wave, and that intracellular calcium oscillates at a frequency similar to the slow wave.

The three-dimensional structure of neurons in the guinea pig inferior mesenteric and pelvic hypogastric ganglia

The three-dimensional (3-D) morphology of sympathetic inferior mesenteric ganglion (IMG) neurons and sympathetic-parasympathetic pelvic hypogastric ganglion (PHG) neurons was studied using confocal laser scanning microscopy. Cell bodies of IMG neurons were disc-shaped and were arranged orderly in layers. The dendritic arbor of individual neurons was confined to a plane with a thickness that did not exceed the thickness of the parent cell body. The actual dendritic surface area (71,400 micron 2) and volume (81,500 micron 3) of the IMG neurons were up to 100-fold larger than previously reported for similar sympathetic neurons using data of 2-D measurements and estimations of the third dimension. PHG neurons had a much smaller dendritic surface area (4100 micron 2) and volume (2400 micron 3) compared to IMG neurons. The ratio dendritic/somal surface area for individual IMG and PHG neurons ranged from 5:1 to 14:1 and from 0.1:1 to 6:1, respectively. The total dendritic path-length was 8-42 times greater for IMG than for PHG neurons. Neurons in the IMG were either stellate with radiating dendrites or bipolar-shaped with dendrites emerging from the two poles of the cell body. Neurons in the PHG were of two morphological types. One type (nearly 2/3 of all the imaged PHG neurons) had two to seven relatively long dendrites and an axon; the other type had only one to three short unbranched dendrites and an axon. The spatial organization of neurons within the ganglia and the structural features of individual neurons are likely to have important implications regarding connectivity patterns between neurons within the ganglion as well as on how information is processed by the ganglion.

Complement-mediated lesion of sympathetic ganglia in vitro with acetylcholinesterase antibodies

When administered to rats, antibodies against acetylcholinesterase (AChE) selectively destroy presynaptic inputs to sympathetic ganglia. To investigate the mechanism of this immunolesion, we created an in vitro system in which relevant components could be manipulated. Freshly dissected rat superior cervical ganglia (SCG) were incubated 15-20 h at 37 degrees C in fresh human serum (a potent source of complement) with continuous oxygenation. More than 96% of neurons in six control ganglia retained synaptic inputs, as defined by action potentials or excitatory postsynaptic potentials (EPSP) upon stimulation of the preganglionic trunk. However, when anti-AChE antibodies were present (0.16 mg/ml), none of 61 neurons from six incubated ganglia showed synaptic responses although membrane potential and input resistance remained normal. Staining for AChE and synaptophysin (a synaptic vesicle marker) was also disrupted in ganglia exposed to AChE antibodies in complement-sufficient serum. When complement was eliminated by substituting serum that was heat-inactivated or deficient in C3, synaptic input was retained in 60-90% of neurons incubated with AChE antibodies. Choline acetyltransferase activity (ChAT), an enzymatic marker of cholinergic cytoplasm in sympathetic ganglia, was largely lost after incubation with AChE antibodies and serum. However, incubation with AChE antibodies in heat-inactivated serum, or serum that was deficient in C3 or C8, caused no measurable loss of ganglionic ChAT activity. These findings strongly implicate the complement cascade in the destruction of preganglionic sympathetic terminals that follows binding of AChE antibodies.

PACAP27 and other neuropeptides in the inferior mesenteric ganglion

The presence and location of PACAP27-like immunoreactivity (PACAP27-LI) in the colon-inferior mesenteric ganglion (IMG) reflex pathway and the effect of exogenously administered PACAP27 on the excitability of IMG are reported. The results provide morphological and electrophysiological support for the hypothesis that PACAP modulates reflex activity between the large intestine and IMG. The intense excitatory effect would be expected to increase the rate of action potential discharge in IMG neurons, increasing sympathetic drive to the colon thereby decreasing of colonic activity.

Patch-clamp recordings of membrane currents evoked during natural synaptic activity in sympathetic neurons

Membrane currents elicited by colonic distension and by electrical stimulation of the intermesenteric nerve containing colonic afferent nerve fibres were recorded from neurons of the mouse superior mesenteric ganglion at 20 degrees C with the whole-cell patch-clamp method. Electrically-evoked excitatory postsynaptic currents reversed at -3.5 mV. At membrane holding voltages of -70 mV and -110 mV, the excitatory postsynaptic currents were characterized by a single exponential decay with a mean (+/- S.E.M.) time-constant of 17.5 +/- 1.3 ms and 15.5 +/- 2.3 ms, respectively. Colonic distension evoked a series of the excitatory postsynaptic currents which ranged in amplitude from 10 to 700 pA (at a membrane holding voltage of -70 mV). Hexamethonium (100 microM) applied only to the ganglion abolished both electrically- and distension-evoked excitatory postsynaptic currents, suggesting activation of nicotinic acetylcholine receptors. The decay time-course of distension-evoked single excitatory postsynaptic currents was characterized by one, or, less commonly, by two exponentials. The decay time-constant histograms of distension-evoked single excitatory postsynaptic currents exhibited main kinetic components of 8.1 +/- 2.3 ms and 8.2 +/- 2.5 ms (peak +/- S.D.) at -70 and -110 mV membrane holding voltages, respectively. Longer time-constants ranging up to 51 ms were also observed. The number of the distension-evoked excitatory postsynaptic currents with a decay time-constant higher than 20 ms, as well as their mean amplitude, were significantly lower at -110 mV than at -70 mV membrane potential levels, in contrast to the currents with a decay time-constant lower than or equal to 20 ms. The results suggest that colonic afferent nerve fibres activate in the mouse superior mesenteric ganglion neurons a few populations of the postsynaptic nicotinic acetylcholine receptors with different channel kinetics, which are characterized by a lack of voltage sensitivity within -70 to -110 mV membrane potential range, except those with comparatively slow channel kinetics, which are possibly blocked by membrane hyperpolarization.

The three-dimensional structure of myenteric neurons in the guinea-pig ileum

Myenteric neurons of the guinea-pig ileum were intracellularly filled with the fluorescent dye Lucifer Yellow, optically sectioned with a confocal microscope and volume reconstructed to recreate 3-D images of the cells. The resulting images provide information not evident from regular microscopy. The somata varied in cross-section from flat-oval to nearly circular, and their surface membranes were marked by invaginations and protrusions significantly increasing the surface area of the somatic membrane. The neurons could be divided into four morphological classes: Dogiel type I, Dogiel type II, filamentous, and intermediate. There was no clear correlation between cell class and the shape of the soma in cross-section. The dendritic processes of all the neurons studied extended in an orad-caudad or circumferential direction of the bowel wall. When the filled neurons were viewed edge-on, the spatial arrangement of the processes was confined to a plane that had a thickness less than the thickness of the parent soma. The broad, short dendrites of Dogiel type I neurons were oval or nearly circular in cross-section. Directly measured quantitative data were obtained for the volume and surface area of the somata and visible processes. The structural details reported herein are likely to have important implications regarding the functional properties of individual enteric ganglion neurons and circuits of enteric ganglion neurons.

Heme oxygenase 2 is present in interstitial cell networks of the mouse small intestine

BACKGROUND & AIMS

The interstitial cell (IC) network may be of fundamental importance in regulating gastrointestinal motility. Intestinal smooth muscle cells are depolarized in the absence of ICs, and there are no spontaneous slow waves. The messenger molecules between IC network and smooth muscle are unknown. Exogenous administration of CO relaxes the opossum internal anal sphincter and the guinea pig ileum, and it modulates potassium current and membrane potential of circular smooth muscle cells of the human jejunum. The aim of this study was to determine whether heme oxygenase (HO)-1 and HO-2, enzymes that catalyze the production of CO, are present in the IC network of the mouse small intestine.

METHODS

Antibodies specific for c-Kit, HO-1, and HO-2 were used for immunohistochemistry. Confocal images were obtained and were volume rendered, and the images were converted into three-dimensional images.

RESULTS

HO-2-like but not HO-1-like immunoreactivity was found in IC networks associated with the myenteric plexus and the deep muscular plexus.

CONCLUSIONS

HO-2 but not HO-1 is present in the IC cell network of the mouse small intestine. The enzymatic activity of HO-2 will result in the endogenous production of CO in IC networks of the mouse small intestine.

Selective disruption of neurotransmission by acetylcholinesterase antibodies in sympathetic ganglia examined with intracellular microelectrodes

Antibodies to acetylcholinesterase (AChE) induce adrenergic dysfunction in rats by selective, complement-mediated destruction of preganglionic sympathetic nerve terminals. To analyze this phenomenon at the neuronal level, monoclonal antibodies to AChE (1.6 mg) were injected via the tail vein, and superior cervical ganglia (SCG) or inferior mesenteric ganglia (IMG) were studied in vitro. In control SCG, all impaled neurons generated action potentials during direct injection of depolarizing current or indirect stimulation through the preganglionic nerve. Current injection remained effective in ganglia from treated rats, but preganglionic stimulation was greatly impaired: at 12 h and 3 d, less than 10% of the neurons responded, even to a maximal stimulus (150 V); at 9 d, only 25% responded. By contrast, in IMG, synaptic transmission was much less affected by antibody exposure: 60% or more of examined neurons responded to preganglionic stimulation. Differences in antibody access did not explain differing sensitivities of SCG and IMG since immunohistochemistry showed rapid accumulation of IgG deposits in both ganglia. These results are believed to reflect widespread but subtotal preganglionic sympathectomy by AChE antibodies. Current information indicates that paravertebral ganglia are all antibody-sensitive, but some prevertebral ganglia are resistant, suggesting immunochemical differences between them.

[Activity of the neurons of the abdominal-aortic plexus in the guinea pig during preservation of its connections with the large intestine]

With intracellular recording, peripheral inputs from the mechanoreceptor C--afferents of the colon which signal the motor status of its isolated oral and aboral segments were shown to converge on neurons of the abdominal aortic plexus of the guinea pig. A proportion (19.8%) of synapthcally activated neurons of this plexus send axons to the periphery within mesenteric nerves, thus forming the efferent link of an extracentral intestio--intestinal reflex.