Taube P. Rothman

Guinea pig 5-HT transporter: cloning, expression, distribution, and function in intestinal sensory reception

Studies of the guinea pig small intestine have suggested that serotonin (5-HT) may be a mucosal transmitter that stimulates sensory nerves and initiates peristaltic and secretory reflexes. We tested the hypothesis that guinea pig villus epithelial cells are able to inactivate 5-HT because they express the same 5-HT transporter as serotonergic neurons. A full-length cDNA, encoding a 630-amino acid protein (89.2% and 90% identical, respectively, to the rat and human 5-HT transporters) was cloned from the guinea pig intestinal mucosa. Evidence demonstrating that this cDNA encodes the guinea pig 5-HT transporter included 1) hybridization with a single species of mRNA ( approximately 3.7 kb) in Northern blots of the guinea pig brain stem and mucosa and 2) uptake of [3H]5-HT by transfected HeLa cells via a saturable, high-affinity (Michaelis constant 618 nM, maximum velocity 2.4 x 10(-17) mol . cell-1 . min-1), Na+-dependent mechanism that was inhibited by chlorimipramine > imipramine > fluoxetine > desipramine > zimelidine. Expression of the 5-HT transporter in guinea pig raphe and enteric neurons and the epithelium of the entire crypt-villus axis was demonstrated by in situ hybridization and immunocytochemistry. Inhibition of mucosal 5-HT uptake potentiates responses of submucosal neurons to mucosal stimulation. The epithelial reuptake of 5-HT thus appears to be responsible for terminating mucosal actions of 5-HT.

Development of the interstitial cell of Cajal: origin, Kit dependence and neuronal and nonneuronal sources of Kit ligand

Kit is a marker for interstitial cells of Cajal (ICC). ICCs interact with enteric neurons and are essential for gastrointestinal motility. The roles of neural crest‐derived cells, neurons, Kit, and Kit ligand (KL) in ICC development were analyzed. ICC development lagged behind that of neurons and smooth muscle. Although mRNA encoding Kit and KL was detected at E11, Kit‐immunoreactive ICCs did not appear until E12 in foregut and E14 in terminal hindgut. Transcripts of Kit and KL and Kit‐immunoreactive cells were found in aganglionic gut from ls/ls and c‐ret −/− mice. ICCs also developed in crest‐free cultures of ls/ls terminal colon. ICCs appeared in cultures of noncrest‐ but not those of crest‐derived cells isolated from the fetal bowel by immunoselection with antibodies to p75NTR. KL immunoreactivity was coincident in cells with neuronal or smooth muscle markers. The development of ICCs in cultures of mixed cells dissociated from the fetal gut was dependent on plating density. No ICCs appeared at ≤80,000 cells/ml, but many cells, including filamentous ICCs, appeared at ≥200,000 cells/ml. Exogenous KL partially substituted for a high plating density. These data support the ideas that mammalian ICCs are neither derived from the neural crest nor developmentally dependent on neurons. ICC differentiation/survival requires KL, which can be provided by neurons or cells in a smooth muscle lineage. Neurons may be needed for development of myenteric ICCs and the mature ICC network. J. Neurosci. Res. 59:384–401, 2000

Bone Morphogenetic Protein-2 and -4 Limit the Number of Enteric Neurons But Promote Development of a TrkC-Expressing Neurotrophin-3-Dependent Subset

The hypothesis that BMPs (bone morphogenetic proteins), which act early in gut morphogenesis, also regulate specification and differentiation in the developing enteric nervous system (ENS) was tested. Expression of BMP-2 and BMP-4, BMPR-IA (BMP receptor subunit), BMPR-IB, and BMPR-II, and the BMP antagonists, noggin, gremlin, chordin, and follistatin was found when neurons first appear in the primordial bowel at embryonic day 12 (E12). Agonists, receptors, and antagonists were detected in separated populations of neural crest- and noncrest-derived cells. When applied to immunopurified E12 ENS precursors, BMP-2 and BMP-4 induced nuclear translocation of phosphorylated Smad-1 (Sma and Mad-related protein). The number of neurons developing from these cells was increased by low concentrations and decreased by high concentrations of BMP-2 or BMP-4. BMPs induced the precocious appearance of TrkC-expressing neurons and their dependence on neurotrophin-3 for survival. BMP-4 interacted with glial cell line-derived neurotrophic factor (GDNF) to enhance neuronal development but limited GDNF-driven expansion of the precursor pool. BMPs also promoted development of smooth muscle from mesenchymal cells immunopurified at E12. To determine the physiological significance of these observations, the BMP antagonist noggin was overexpressed in the developing ENS of transgenic mice under the control of the neuron-specific enolase promoter. Neuronal numbers in both enteric plexuses and smooth muscle were increased throughout the postnatal small intestine. These increases were already apparent by E18. In contrast, TrkC-expressing neurons decreased in both plexuses of postnatal noggin-overexpressing animals, again an effect detectable at E18. BMP-2 and/or BMP-4 thus limit the size of the ENS but promote the development of specific subsets of enteric neurons, including those that express TrkC.

Accumulation of components of basal laminae: Association with the failure of neural crest cells to colonize the presumptive aganglionic bowel of lsls mutant mice☆

Aganglionosis occurs in the terminal colon of the ls/ls mouse because an intrinsic defect of the presumptive aganglionic tissue prevents the entry and colonization of this portion of the bowel by migrating neural crest cells. The current study was undertaken to determine if abnormalities of the extracellular matrix could be identified in this segment that might account for migratory failure. Since basal laminae of the muscularis mucosa are overproduced in the aganglionic segment of adult ls/ls mice, we examined components of basal laminae in fetal gut from Day E 11 to Day E 16 of gestation. This period spans the time of enteric ganglion formation. Laminin and collagen type IV were studied by immunocytochemistry and proteoglycans by staining glycosaminoglycans with Alcian blue. Abnormalities of each of these components occur during development of the presumptive aganglionic bowel in the ls/ls mouse and could be detected as early as Day E 11. These defects consist mainly of an overabundance of these materials, both in defined basal laminae and throughout the extracellular space of the mesenchyme. Electron microscopic observations in the presumptive aganglionic ls/ls colon revealed a thickening of basal laminae and exceptionally wide intercellular spaces between smooth muscle myoblasts that contained an irregular fibrillar material, consisting of 4.5- to 6.0-nm filaments associated with 14- to 20-nm granules. Fibrillar and flocculant material was continuous with formed basal laminae, and was concentrated in the same areas found to have an overabundance of laminin immunoreactivity. These observations indicate that there is an accumulation of extracellular matrix material, including components of basal laminae, that (i) precedes the formation of enteric ganglia, (ii) is in the path through which enteric neural precursors from the crest would have to migrate, and (iii) is limited to the aganglionic and hypoganglionic ls/ls bowel. These data are consistent with the hypothesis that components of basal laminae contribute to the inability of crest cells to colonize the terminal bowel of ls/ls mice.

The ?1 subunit of laminin-1 promotes the development of neurons by interacting with LBP110 expressed by neural crest-derived cells immunoselected from the fetal mouse gut

A plasmalemmal protein, LBP110, which binds to the alpha1 chain of laminin-1, is acquired by the neural crest-derived precursors of enteric neurons after they colonize the gut. We tested the hypothesis that laminin-1 interacts with LBP110 to promote enteric neuronal development. The effects of laminin-1 on neuronal development were studied in cultures of cells immunoselected from fetal mouse gut (E14-15) with antibodies to LBP110 or p75NTR, a marker for enteric crest-derived cells. No matter which antibody was used, the development of cells expressing neuronal markers was increased three- to fourfold by culturing the cells on a laminin-1-containing substrate. To determine whether this effect of laminin-1 is due to the selective adherence of a neurocompetent subset of precursors, immunoselected cells were permitted to preadhere to poly-D-lysine. Addition of soluble laminin-1 24 h later promoted neuronal but not glial development. The laminin-1-induced increment in neuronal development was abolished both by a peptide containing the sequence of the LBP110-binding domain, IKVAV, and by antibodies to laminin alpha1 that recognize the IKVAV domain. Neither reagent affected the total number of cells. In contrast, the response to laminin-1 was not affected by control peptides, preimmune sera, or antibodies to laminin beta1. Laminin-1 transiently induced the expression of nuclear Fos immunoreactivity; this action was blocked specifically by the IKVAV peptide. These data are consistent with the hypothesis that LBP110 interacts with the IKVAV domain of laminin alpha1 to promote the differentiation of neurons from enteric crest-derived precursors.

Promotion of the development of enteric neurons and glia by neuropoietic cytokines: Interactions with neurotrophin-3

Neurotrophin-3 (NT-3) is known to promote enteric neuronal and glial development. Ciliary neurotrophic factor (CNTF) and leukemia inhibitory factor (LIF) were investigated to test the hypothesis that the development of subsets of enteric neurons and/or glia is also affected by a neuropoietic cytokine, by itself, or together with NT-3. Crest-derived cells, immunoselected from the fetal rat gut (E14) with antibodies to p75NTR, were found by RT-PCR and immunocytochemistry (after culture) to express both alpha (CNTER alpha) and beta components (gp130 and LIFR beta) of the tripartite CNTF receptor. In situ, myenteric ganglia below the esophagus were CNTFR alpha-immunoreactive by E16-E18. In vitro, CNTF and LIF induced in crest-derived cells nuclear translocation of STAT3 (signal transducer and activator of transcription 3), a concentration-dependent increase in expression of neuronal or glial markers, and a decrease in expression of the precursor marker, nestin. LIFR beta was expressed by more cells than CNTFR alpha; therefore, although the factors were equipotent, the maximal effect of LIF > CNTF. The cytokines and NT-3 were additive in promoting neuronal but not glial development. Specifically, the development of neurons expressing NADPH-diaphorase activity (an early marker found in inhibitory motor neurons) was promoted by CNTF and NT-3. These observations support the idea that a ligand for the tripartite CNTF receptor complex plays a role in ENS development.

Development of the enteric nervous system.

The mature enteric nervous system (ENS) is characterized by a degree of neuronal phenotypic diversity and independence of central nervous system control unequaled by any other region of the peripheral nervous system. Studies that have utilized the immunocytochemical demonstration of neurofilament protein and explanation of primordial gut with subsequent growth in culture have indicated that the neural crest precursors of enteric neurons are already committed to the neuronal lineage when they colonize the bowel; however, neuronal phenotypic expression occurs within the gut itself. It is likely that precursors able to give rise to each type of neuron found in the mature ENS are present among the earliest neural crest emigres to reach the bowel. In mice a proximodistal wave of neuronal phenotypic expression occurs that does not appear to reflect the descent of neuronal precursors. This observation, the known plasticity of developing neural crest-derived neurons, and the demonstration of a persistent population of proliferating neuroblasts in the gut raise the possibility that enteric neuronal phenotypic expression is influenced by the enteric microenvironment.

Development of the monoaminergic innervation of the avian gut: transient and permanent expression of phenotypic markers

Specific cellular accumulation of [3H]5-hydroxytryptamine ([3H]5-HT) occurs during development of the avian gut. This accumulation is transient in extraganglionic mesenchymal cells (TES cells) but is a permanent characteristic of enteric serotonergic neurons (ESN). Species-specific differences were found in the location of TES cells and ESN. In chicks TES cells surrounded myenteric ganglia and ESN were restricted to the myenteric plexus. In quails TES cells surrounded submucosal ganglia and [3H]5-HT-labeled submucosal as well as myenteric neurons. [3H]Norepinephrine accumulated only in noradrenergic terminals and not in TES cells or ESN. The origins of TES cells and ESN were studied in chimeras, in which neuraxis from appropriate or inappropriate axial levels was grafted from quail to chick. Both types of chimeric bowel contained TES cells and ESN. Most TES cells in chimeras were chick in origin and distributed as in chicks (around myenteric ganglia); however, some TES cells and all ESN were quail cells. To test whether crest cells are required for development of TES cells and ESN, aneuronal chick hindgut was explanted and grown alone, or with quail neuraxis, as chorioallantoic membrane (CAM) grafts. TES cells appeared in CAM grafts whether or not crest cells were present; however ESN only appeared in explants when quail neuraxis was included. In addition, an ectopic [3H]5-HT-labeled chromaffin-like cell, also of quail origin, was found in enteric plexuses in these combined explants of crest and gut. Most TES cells, therefore, are neither derived from nor dependent on the presence of crest cells in the gut wall. Since even an inappropriate axial level of crest was found to produce ESN when it was experimentally induced to colonize the bowel the enteric microenvironment probably plays a critical role in serotonergic neural development. The species-specific location of TES cells and ESN is consistent with the hypothesis that TES cells constitute an important component of this microenvironment.

Effects of extended periods of reserpine and α-methyl-p-tyrosine treatment on the development of the putamen in fetal rabbits

Developing nigrostriatal neuroblasts exhibit catecholamine‐induced fluorescence before their axons have left the vicinity of the cell bodies. To evaluate possible developmental effects of dopamine, we have used reserpine and α‐methyl‐p‐tyrosine to deplete dopamine chronically during the development of these axons. We found that dopamine‐induced fluorescence was either absent or markedly decreased in the fetal putamens. To determine whether the absence of fluorescence was due to a reduction of dopamine terminals, the uptake of tritium‐labeled dopamine was measured in the putamen. Uptake of labeled dopamine was significantly depressed in reserpine‐treated fetuses to 70% of that of controls; however, no depression of labeled dopamine was found in the α‐methyl‐p‐tyrosine‐treated fetuses. After both drug treatments, the striatal perikarya were less mature than those of controls. Although we cannot rule out possible non‐specific or toxic effects of the drugs, these observations support the conclusion that presynaptic dopamine may be important for development of target neurons in the neostriatum.

The effect of back-transplants of the embryonic gut wall on growth of the neural tube

Experiments in which the developing gut of avian embryos was back-transplanted to permit the bowel to interact with the developing neural tube were undertaken. Segments of intestine from 4-day quail embryos were implanted between the somites and neural tubes of chick embryos of 7 to 24 somites. The spinal cord responded to the presence of the bowel by enlarging unilaterally on the side of the graft. This effect encompassed both gray and white matter and was accompanied by the extension of neuritic projections from the spinal cord into the enteric grafts. The growth-promoting effect of enteric transplants was manifest at all levels of the neural tube where the grafts were made and led to enlargement of the brain as well as the spinal cord; however, truncal neural crest derivatives in the region of the grafts, such as developing sympathetic and spinal ganglia, were unaffected. Neither sham operations nor grafts of ciliary ganglion, lung, pancreas, mesonephros, or rudiment of the eye mimicked the action of the gut. The effect of the bowel was manifest as early as 24 hr following back-transplantation and was found to be due to an increase in the number of cells in the neuroepithelium. The cell responsible for the ability of the gut wall to enhance neuroepithelial proliferation was not identified, but the effect lacked species specificity and could be elicited in the absence of endoderm or neural crest derivatives in the explant. We propose that the musculoconnective tissue of the gut produces a short-range diffusible factor that induces mitogenic activity in the neuroepithelial cells of the neural tube, but not in the crest cells that form sympathetic or sensory ganglia. Since the gut is not normally in apposition to the neural tube, we suggest that the physiological targets of this factor are the specialized crest cells that colonize the bowel and give rise to the enteric nervous system.