Vagal afferent neurons contain a variety of neurochemical markers and neuroactive substances, most of which are present also in dorsal root ganglion cells. To test for the suitability of the calcium-binding protein calretinin as a specific marker for vagal afferent fibers in the periphery, immunocytochemistry for this protein was combined with retrograde tracing. Nerve fibers in the rat esophagus, as well as vagal and spinal sensory neurons innervating the esophagus, were investigated for co-localization of calretinin with calbindin, calcitonin gene-related peptide, and NADPH diaphorase. The results indicated that calretinin immunocytochemistry demonstrates neuronal structures known as vagal afferent from other studies, in particular intraganglionic laminar endings. A few enteric neurons whose distribution was unrelated to intraganglionic laminar endings also stained for calretinin. Strikingly, calretinin immunoreactivity was absent from spinal afferent neurons innervating the rat esophagus. In intraganglionic laminar endings and nodose ganglion cells calretinin was highly co-localized with calbindin but not with calcitonin gene-related peptide. On the other hand, calbindin was also found in spinal afferents to the esophagus where it was co-localized with calcitonin gene-related peptide. Vagal afferent neurons innervating the esophagus were never positive for NADPH diaphorase. Thus, calretinin appears to be a more specific marker for vagal afferent structures in the esophagus than calbindin, which is expressed by both vagal and spinal sensory neurons. Calretinin immunocytochemistry may be utilized as a valuable tool for investigations of subpopulations of vagal afferents in certain viscera.

NADPH-diaphorase histochemistry was combined with demonstration of acetylcholinesterase and immunocytochemistry for calcitonin gene-related peptide to study esophageal innervation in the rat. Most of the myenteric neurons stained positively for NADPH-diaphorase, as did numerous varicose nerve fibers in the myenteric plexus, among striated muscle fibers, around arterial blood vessels, and in the muscularis mucosae. A majority of motor endplates (as demonstrated by acetylcholinesterase histochemistry or calcitonin gene-related peptide immunocytochemistry) were associated with fine varicose NADPH-diaphorase-positive nerve fibers. Analysis of brainstem nuclei, sensory vagal, spinal, and sympathetic ganglia in normal and neonatally capsaicin-treated rats, and comparison with anterogradely labeled vagal branchiomotor, preganglionic and sensory fibers led to the conclusion that NADPH-diaphorase-positive fibers on motor endplates originate in esophageal myenteric neurons. No association of NADPH-diaphorase-positive nerve fibers with motor endplates was found in other organs containing striated muscle. These results suggest extensive, presumably nitrergic, co-innervation of esophageal striated muscle fibers by enteric neurons. Thus, control of peristalsis in the esophagus of the rat may be more complex than hitherto assumed.

Langerhans cells play an important part in the immune surveillance of the human epidermis. Therefore, a certain distribution and numerical relationship to other epidermal cells can be expected. To quantify epidermal Langerhans cells population extensive studies have been performed using two-dimensional quantification methods on vertical sections or epidermal sheet preparations. Whereas methods using vertical sections were complicated considerably by the sampling procedure, the dendritic shape, and the suprabasal, nonrandom distribution of Langerhans cells, epidermal sheet preparations have their limitations regarding the numerical relationship of Langerhans cells to total epidermal cells and the epidermal morphology as such. In order to improve the validity of data the three-dimensional dissector method combined with confocal laser scanning microscopy has been applied to quantify the number of Langerhans cells and other epidermal cell nuclei per volume unit in cryosections of 24 punch biopsies of normal breast skin of eight women. Furthermore, the ratio of Langerhans cells to other epidermal cells, their number per biopsy, and per skin surface area were calculated. To minimize the bias by shrinkage the reference volume was estimated using Cavalieri's principle. A constant ratio of one Langerhans cells to 53 other epidermal cells was identified in breast skin (interindividual correlation coefficient: 0.952, p < 0.0001). Thus, Langerhans cells represent 1.86% of all epidermal cells; however, a wide interindividual range was found for the number of Langerhans cells per mm2 (912-1806; mean +/- SD 1394 +/- 321) and other epidermal cells per mm2 (47,315-104,588; mean +/- SD 73,952 +/- 19,426). This explains the conflicting results achieved by conventional morphometric assessments relating cell numbers to skin surface area, ignoring the varying thickness of the epidermis. The surprisingly constant relationship of Langerhans cells to other epidermal cells stresses the hypothesis of an epidermal Langerhans cells unit where one Langerhans cells seems to be responsible for the immune surveillance of 53 epidermal cells.

The existence of a distinct ganglionated myenteric plexus between the two layers of the striated tunica muscularis of the mammalian esophagus represented an enigma for quite a while. Although an enteric co-innervation of vagally innervated motor endplates in the esophagus has been repeatedly suggested, it was not possible until recently to demonstrate this dual innervation. Ten years ago, we were able to demonstrate that motor endplates in the rat esophagus receive a dual innervation from both vagal nerve fibers originating in the brain stem and from varicose enteric nerve fibers originating in the myenteric plexus. Since then, a considerable amount of data could be raised on enteric co-innervation and its occurrence in a variety of species, including humans, its neurochemistry, spatial relationships on motor endplates, ontogeny, and possible roles during esophageal peristalsis. These data underline the significance of this newly discovered innervation component, although its function is still largely unknown. The aim of this review is to summarize current knowledge about enteric co-innervation of esophageal striated muscle and to provide some hints as to its functional significance.

The development of vagal and enteric innervation of esophageal motor endplates was examined in perinatal and adult BALB/c and NMRI mice using immunocytochemistry and confocal laser scanning microscopy. Nicotinic acetylcholine receptors were demonstrated with fluorochrome-tagged alpha-bungarotoxin, vagal motor terminals with antisera against vesicular acetylcholine transporter and calcitonin gene-related peptide, and enteric nerve terminals with antisera against neuronal nitric oxide synthase, vasoactive intestinal peptide and galanin. Results demonstrated that enteric and vagal innervations of striated esophageal muscle fibers develop in close spatiotemporal relationship, but with different courses. Connections between VAChT-positive vagal nerve terminals and growing acetylcholine receptor clusters were established from E17 to reach 100% motor endplate innervation at P14 and were maintained throughout adult life. CGRP immunoreactivity developed with a delay of several days after the appearance of VAChT in vagal terminals. From P14 to adulthood CGRP was colocalized with VAChT in almost all motor endplates. In contrast, enteric co-innervation rates increased from E17 to a maximum of 70-80% at P4, while their incidence at motor endplates progressively declined over the following 5 months to lower levels maintained throughout adulthood. Whereas adult enteric co-innervation rates in BALB/c and NMRI mice differed significantly (approximately 30% versus approximately 10%, respectively), their increase and reduction, respectively, during development showed an identical time course. These results suggest a well-ordered sequence of attraction of enteric nerve fibers to, and removal from motor endplates in the developing mouse esophagus. Thus, enteric co-innervation may subserve a functional role in the development and control of perinatal striated esophageal muscle rather than representing an unspecific "hangover" from the smooth muscle past of this organ.