The patterns and extent of vagal preganglionic divergence and convergence within the gastrointestinal tract of the rat were characterized with the anterograde tracer Phaseolus vulgaris-leucoagglutinin (PHA-L). Three weeks after tracer was iontophoretically injected into two to four sites within the dorsal motor nucleus of the vagus, wholemounts of perfused gut organs (stomach, duodenum, cecum) were prepared, counterstained with Cuprolinic blue, and processed for PHA-L using the avidin biotin complex with diaminobenzidine. Controls included animals injected with PHA-L after intracranial deafferentations. Well-positioned injections labeled an extremely dense and intricate network of varicose efferent axons throughout the gastric myenteric plexus (including that of the fundus). Individual fibers collateralized extensively, forming a variety of pericellular arborizations and terminal complexes made up of both en passant and end swellings. Single axons frequently innervated subsets of neurons within ganglia. Most enteric neurons were contacted by varicosities of more than one vagal fiber. The patterns of vagal preganglionic fibers in the duodenal and cecal myenteric plexuses resembled the organization in the stomach in many aspects, but the projections in each organ had distinctive characteristics, and label was less dense in the intestines than in the stomach. Vagal preganglionic fibers directly innervated submucosal ganglia, although sparsely. Brainstem injections of PHA-L retrogradely labeled a few myenteric neurons in the corpus, fundus, and duodenum: These "gastrobulbar" and "duodenobulbar" neurons received reciprocal vagal preganglionic innervation. Finally, the PHA-L that spread to the nucleus of the solitary tract occasionally produced transganglionic labeling of afferent intramuscular arrays (gastric fundus). The results of this paper provide strong evidence that the traditional "command neuron" or "mother cell" hypotheses of vagal-enteric organization should be abandoned for an integrative neural network model.

We have developed a dorsal intracranial surgery that is minimally invasive and gives excellent access to either afferent or efferent vagal rootlets to produce selective deafferentations or deefferentations in the rat. We have combined this new unilateral afferent rhizotomy with a contralateral celiac branch cut (to completely deafferent the intestines) and a duodenal catheter placement 4 cm distal to the pylorus. Animals were maintained with 17 h/day access to a nutritionally complete liquid diet. Measures of first meal size, daily intake, and body weight before and after both surgeries indicated that animals with unilateral vagal deafferentiations recovered as fast and completely as sham-operated controls. Intraduodenal oleate (1.2 kcal) infusions reduced the size of the first meal in surgical controls (by 64%; P < 0.01) but not in the deafferented rats. A dual wheat germ agglutinin-horseradish peroxidase/Fluorogold protocol provides verification of sensory and motor lesions. The selective vagal deafferentation provided by the new surgery offers a useful model for determining gastrointestinal sites of nutrient detection and separating pre- and postabsorptive consequences of a meal.

No abstract

The arrival of nutrients in the gastrointestinal tract suppresses intake. To specify the neural pathways and receptor locations of this feedback, we examined the effects of intraduodenal infusions of 10 nutrients plus saline on short-term food intake of rats with selective deafferentations of vagal celiac branches. Three response profiles were observed: 1) isotonic saline, 5.6% glycerol, and 3% fructose did not inhibit intake of controls or selectively deafferented animals; 2) 3% glucose, 3% maltose, 3% L-phenylalanine, 12% Isocal, and 1.4% oleic acid suppressed intake of controls, but this inhibition was eliminated by vagal celiac deafferentation; and 3) 3% casein hydrolysate and 24% Isocal suppressed intake of controls and rats with selective vagotomies, although the latter exhibited significantly less suppression. In addition, elimination of celiac afferents chronically reduced meal size (i.e., first 30-min intake) without reducing daily food intake or body weight. Furthermore, D-phenylalanine infusions produced a delayed suppression of food intake in controls (possibly from intraluminal irritation); however, this reduction was eliminated with celiac deafferentation. Overall, this experiment indicates that vagal celiac afferents are critical for preabsorptive detection of some energy-yielding molecules or properties of nutrient solutions (as well as, perhaps, intraluminal inflammation), but not others, which are still detected, although only partially.

Three-dimensional reconstruction protocols in confocal microscopy are typically considered in terms of rendering separate stacks of optical sections. Single stacks, however, include volumes that are often too small to permit descriptions of entire neurons, complete axonal arbors, or complex neural networks. Furthermore, traditional tissue preparation protocols generally yield specimens too limited to permit reconstructions of complex neural systems. For 3-D analyses of extensive networks such as the autonomic nervous system projections within the viscera, it is critical to incorporate appropriate tissue techniques, including suitable tracer protocols, into the reconstruction strategy. This report summarizes complementary technologies, including whole mount procedures, tracer techniques for identifying single fibers in situ, and methods of examining stacks of optical images, which make it practical to describe the complete terminal field of an individual axon in the gastrointestinal tract. Such methods establish that vagal motor axons travel long distances within their target organs, collateralize frequently, and ramify extensively. Vagal afferents have extensive, complex, and, in some cases, polytopic arbors within target tissues.