Reorganization of the ependyma during axolotl spinal cord regeneration: changes in intermediate filament and fibronectin expression.

Changes in intermediate filament content and extracellular matrix material showed that the injury response of ependymal cells in lesioned axolotl spinal cord involves an epithelial-to-mesenchymal transformation, and that fibrous astrocytes are excluded from the remodeling lesion site. Antibody localization was used to visualize cytokeratin-, vimentin-, and glial fibrillary acidic protein- (GFAP-) containing intermediate filaments, as well as the adhesive glycoprotein fibronectin. In normal axolotl spinal cord cytokeratins were found near the apical surface of the ependymal cells. Transmission electron microscopic examination suggested that these cytokeratins were in tonofilaments. Cytokeratin expression was lost and vimentin production was initiated in ependymal cells 2-3 weeks following spinal cord injury. There was a period of approximately 1-2 weeks when cytokeratins and vimentin were co-expressed in vivo. This co-expression was maintained in vitro by culture on a fibronectin-coated substratum. As the central canal reformed, vimentin expression was lost. Ependymal cells lacked GFAP intermediate filaments, but GFAP was present in fibrous astrocytes of the neuropil and white matter. Following injury, GFAP localization showed that fibrous astrocytes disappeared from the remodeling lesion site and reappeared only after the ependymal epithelium reformed and newly myelinated axons were found. Fibronectin expression closely followed the expression of vimentin during mesenchymal ependymal cell outgrowth. These results suggest that the ependymal cell outgrowth requires changes in cell shape followed by changes in production of extracellular matrix.

Structural changes in the proximal tubule of the short-toes axolotl mutant.

A recessive lethal mutation in axolotls that involves the kidneys, the Mullerian ducts and the limbs was described by Humphrey (1967). In the present experiments, we have examined the structural defects that lead to kidney malfunction and subsequent death in homozygous mutants and compared the defects with those observed in other axolotls lacking this mutant gene. The ultrastructure of the mesonephric kidney was studied in homozygous s/s short-toes axolotls with ascites and/or edema and hemorrhages (group 1a); in s/s short-toes axolotls not yet expressing kidney malfunction symptoms (group 1b); in normal siblings, either +/+ or +/s, without affected limbs but possibly heterozygous for the 's' gene (group 1c); in +/+ animals with malfunctioning kidneys that lack the short-toes gene (group 2a) and in normal +/+ animals from different gene pools (group 2b). In all of the short-toes animals that expressed pathological phenotypes, the proximal tubule showed abnormal morphology. There was no morphological evidence of kidney abnormality in the siblings having normal limbs or in the group 1b axolotls examined. However, mesonephroi from animals (group 2a) of other gene pools that had ascites exhibited a different and more pronounced pathology. On the basis of the dramatically distinct proximal tubule pathology of the sporadic floaters, we conclude that this phenotype is more likely caused by infection than to a variant of the short-toes gene.

Marginal neurons in the urodele spinal cord and the associated denticulate ligaments.

Marginal neurons have been described in the spinal cords of a variety of vertebrates including lamprey, reptiles, birds, and mammals but not in amphibians. There has been speculation about a motor function for these neurons but recent experimental evidence in lampreys indicates that they are intraspinal mechanoreceptor neurons. Additional evidence on reptiles and birds demonstrates that the marginal neurons are closely associated with the denticulate ligaments. In the present investigation, we have examined the spinal cords of Necturus, Ambystoma tigrinum, and A. mexicanum with light and electron microscopic techniques. Marginal nuclei were found in the ventrolateral position immediately internal to the pia and to the denticulate ligament. The marginal neurons were scattered in a continuous column of neuropil without segmental accumulation. They were approximately 30 to 50 microns in diameter and fusiform with dendrites extending from the poles, parallel with the length of the spinal cord. Neuronal fingerlike processes, like those found in peripheral mechanoreceptors and in the marginal nuclei of reptiles, were also found in the three species of urodeles studied. The structure of the denticulate ligaments, similar in the three different amphibians, was composed of collagen, elastin, and fibroblasts, all of which were concentrated in the segmental lateral processes.

Primary culture of axolotl spinal cord ependymal cells.

In order to examine the role of ependymal cells in the spinal cord regeneration of urodele amphibians, procedures were established to identify and culture these cells. Cell isolation and culture conditions were determined for ependymal cells from larval and adult axolotls (Ambystoma mexicanum). Dissociated cells prepared from intact spinal cords were cultured on fibronectin- or laminin-coated dishes. Dissociated cells attached more rapidly to fibronectin, but attached and spread on both fibronectin and laminin. Essentially pure populations of ependymal cells were obtained by removing 2 week old ependymal outgrowth from lesion sites of adult spinal cords. These ependymal outgrowths attached and grew only on fibronectin-coated dishes. Growth and trophic factors were tested to formulate a medium that would support ependymal cell proliferation. The necessary peptide hormones were PDGF, EGF, and insulin. TGF-beta(1) affected the organization of cell outgrowth. Initially, longterm culture required the presence of high levels of axolotl serum. Addition of purified bovine hemaglobin in the culture medium reduced the serum requirement. Outgrowth from expiants was subcultured by transferring groups of cells. Intrinsic markers were used to identify ependymal cells in culture. The ependymal cells have characteristic ring-shaped nucleoli in both intact axolotl spinal cords and in culture. Indirect immunofluorescence examination of intermediate filaments showed that ependymal cells were glial fibrillary acidic protein (GFAP) negative and vimentin positive in culture. Identification of dividing cells was made using (3)H-thymidine incorporation and autoradiography, and by the presence of mitotic figures in the cultured cells.

Accessory limb production by nerve-induced cell proliferation.

The deviation of large limb nerves to a more proximal skin wound yielded a high proportion of accessory limb responses in different age groups of Ambystoma mexicanum (axolotls). In some instances the deviated nerve was positioned on skin previously grafted from an animal of different age and pigmentation from that of the host. Grafts were found not to be a necessary prerequisite for accessory limb induction, but the presence of wound epithelium was required. The rule of distal morphogenesis was expressed in reference to the level at which the nerve was cut, not in reference to the wound site where the accessory actually developed. The upper arm proved to be a more favorable site for accessory limb production than the flank or the leg under the conditions of the present experiments, in which little or no damage was done to the underlying muscles. The orientation of the accessory limb was extremely varied despite the uniformity of the surgical procedure.

Morphological characterization of actively fusing L6 myoblasts.

We have characterized morphologically the surface of L6 myoblasts at the time of active cell fusion using transmission electron microscopy. Two subclones of the L6 line were used in these studies: the L6Cl55 line that fuses to form multinucleated syncytia and the NF44 non-fusing variant. Ultrastructural analysis revealed an electron-opaque material at localized points of cell-cell apposition in actively fusing L6Cl55 cells. This material may be transported by and secreted from smooth-surfaced cytoplasmic vesicles with an electron-dense core. In contrast to L6Cl55 cells, the electron-dense plaques were seen infrequently in cultures of the NF44 non-fusing variant. This previously unidentified substance may be associated with cell-cell recognition or adhesion, both necessary prerequisites for myoblast membrane fusion. Alternatively, the electron-dense plaques may be directly involved in the fusion event.