Msh homeobox genes regulate cadherin-mediated cell adhesion and cell-cell sorting.

Msx-1 and Msx-2 are two closely related homeobox genes expressed in cephalic neural crest tooth buds, the optic cup endocardial cushions, and the developing limb [Hill and Davidson, 1991; Monaghan et al., 1991; Robert et al., 1991]. These sites correspond to regions of active cell segregation and proliferation under the influence of epithelial-mesenchymal cell interactions [Brown et al., 1993; Davidson et al., 1991], suggesting that Msx-1 and Msx-2 regulate cell-cell interactions. We have investigated the potential relationship between expression of the Msh homeobox genes (Msx-1 and Msx-2) and cadherin-mediated cell adhesion and cell sorting. We report that cell lines stably expressing Msx-1 or Msx-2 differentially sort on the basis of Msh gene expression. We demonstrate in vitro that initial cell aggregation involves calcium-dependent adhesion molecules (cadherins) and that Msh genes regulate cadherin-mediated adhesion. These results support the hypothesis that Msh genes play a role in the regulation of cell-cell adhesion and provide a link between the genetic phenomena of homeobox gene expression and cellular events involved in morphogenesis, including cell sorting and proliferation.

KSA antigen Ep-CAM mediates cell-cell adhesion of pancreatic epithelial cells: morphoregulatory roles in pancreatic islet development.

Cell adhesion molecules (CAMs) are important mediators of cell-cell interactions and regulate cell fate determination by influencing growth, differentiation, and organization within tissues. The human pancarcinoma antigen KSA is a glycoprotein of 40 kD originally identified as a marker of rapidly proliferating tumors of epithelial origin. Interestingly, most normal epithelia also express this antigen, although at lower levels, suggesting that a dynamic regulation of KSA may occur during cell growth and differentiation. Recently, evidence has been provided that this glycoprotein may function as an epithelial cell adhesion molecule (Ep-CAM). Here, we report that Ep-CAM exhibits the features of a morphoregulatory molecule involved in the development of human pancreatic islets. We demonstrate that Ep-CAM expression is targeted to the lateral domain of epithelial cells of the human fetal pancreas, and that it mediates calcium-independent cell-cell adhesion. Quantitative confocal immunofluorescence in fetal pancreata identified the highest levels of Ep-CAM expression in developing islet-like cell clusters budding from the ductal epithelium, a cell compartment thought to comprise endocrine progenitors. A surprisingly reversed pattern was observed in the human adult pancreas, displaying low levels of Ep-CAM in islet cells and high levels in ducts. We further demonstrate that culture conditions promoting epithelial cell growth induce upregulation of Ep-CAM, whereas endocrine differentiation of fetal pancreatic epithelial cells, transplanted in nude mice, is associated with a downregulation of Ep-CAM expression. In addition, a blockade of Ep-CAM function by KS1/4 mAb induced insulin and glucagon gene transcription and translation in fetal pancreatic cell clusters. These results indicate that developmentally regulated expression and function of Ep-CAM play a morphoregulatory role in pancreatic islet ontogeny.

A model for central synaptic junctional complex formation based on the differential adhesive specificities of the cadherins.

Cadherins control critical developmental events through well-documented homophilic interactions. In epithelia, they are hallmark constituents of junctions that mediate intercellular adhesion. Brain tissue expresses several cadherins, and we now show that two of these, neural (N)- and epithelial (E)-cadherin, are localized to synaptic complexes in mutually exclusive distributions. In cerebellum, N-cadherin is frequently found associated with synapses, some of which are perforated, and in hippocampus, N- and E-cadherin-containing synapses are found aligned along dendritic shafts within the stratum lucidum of CA3. We propose that the cadherins function as primary adhesive moieties between pre- and postsynaptic membranes in the synaptic complex. According to this model, once neurites have been guided to the vicinity of their cognate targets, it is the differential distribution of cadherins along the axonal and dendritic plasma membranes, and ultimately cadherin self-association, that "locks in" nascent synaptic connections.

Neural cells from dogfish embryos express the same subtype-specific antigens as mammalian neural cells in vivo and in vitro.

Neural cells are classically identified in vivo and in vitro by a combination of morphological and immunocytochemical criteria. Here, we demonstrate that antibodies used to identify mammalian oligodendrocytes, neurons, and astrocytes recognize these cell types in the developing spiny dogfish central nervous system and in cultures prepared from this tissue. Oligodendrocyte-lineage-specific antibodies O1, O4, and R-mAb labeled cells in the 9 cm dogfish brain stem's medial longitudinal fascicle (MLF) and in areas lateral to it. Process-bearing cells, cultured from the dogfish brain stem, were also labeled with these antibodies. An anti-lamprey neurofilament antibody (LCM), which recognized 60 and 150 kDa proteins in dogfish brain stem homogenates, labeled axons and neurons in the brain stem and axons in the cerebellum of the dogfish embryo. It also labeled cell bodies and/or processes of some cultured cerebellar cells. An anti-bovine glial fibrillary acidic protein antibody, which recognized 42-44 kDa protein(s) in dogfish brain stem homogenates, labeled astrocyte-like processes in the brain stem and cerebellum of the dogfish embryo and numerous large and small flat cells in the cerebellar cultures. These results demonstrate that dogfish oligodendrocytes, neurons, and astrocytes express antigens that are conserved in mammalian neural cells. The ability to culture and identify neural cell types from cartilaginous fish sets the stage for studies to determine if proliferation, migration, and differentiation of these cell types are regulated in a similar fashion to mammalian cells.

Considerations on the folding topology and evolutionary origin of cadherin domains.

Cell-cell adhesion in zonula adherens and desmosomal junctions is mediated by cadherins, and recent crystal structures of the first domain from murine N-cadherin provide a plausible molecular basis for this adhesive action. A structure-based sequence analysis of this adhesive domain indicates that its fold is common to all extracellular cadherin domains. The cadherin folding topology is also shown to be similar to immunoglobulin-like domains and to other Greek-key beta-sandwich structures, as diverse as domains from plant cytochromes, bacterial cellulases, and eukaryotic transcription factors. Sequence similarities between cadherins and these other molecules are very low, however, and intron patterns are also different. On balance, independent origins for a favorable folding topology seem more likely than evolutionary divergence from an ancestor common to cadherins and immunoglobulins.

Novel E-cadherin-mediated adhesion in peripheral nerve: Schwann cell architecture is stabilized by autotypic adherens junctions.

Previous studies (Blank, W. F., M. B. Bunge, and R. P. Bunge. 1974. Brain Res. 67:503-518) showed that Schwann cell paranodal membranes were disrupted in calcium free medium suggesting that cadherin mediated mechanisms may operate to maintain the integrity of the paranodal membrane complex. Using antibodies against the fifth extracellular domain of E-cadherin, we now show by confocal laser and electron immunomicroscopy that E-cadherin is a major adhesive glycoprotein in peripheral nervous system Schwann cells. E-Cadherin is not found, however, in compact myelin bilayers. Rather, it is concentrated at the paranodes, in Schmidt-Lanterman incisures, and at the inner and outer loops. At these loci, E-cadherin is associated with subplasmalemmal electron densities that coordinate in register across several cytoplasmic turns of a single Schwann cell. F-Actin and beta-catenin, two proteins implicated in cellular signaling, also co-localize to E-cadherin positive sites. These complexes are autotypic adherens-type junctions that are confined to the plasma membrane synthesized by a single Schwann cell; E-cadherin was never observed between two Schwann cells, nor between Schwann cells and the axon. Our findings demonstrate that E-cadherin and its associated proteins are essential components in the architecture of the Schwann cell cytoplasmic channel network, and suggest that this network has specialized functions in addition to those required for myelinogenesis.

Structural basis of cell-cell adhesion by cadherins.

Crystal structures of the amino-terminal domain of N-cadherin provide a picture at the atomic level of a specific adhesive contact between cells. A repeated set of dimer interfaces is common to the structure in three lattices. These interactions combine to form a linear zipper of molecules that mirrors the linear structure of the intracellular filaments with which cadherins associate. This cell-adhesion zipper may provide a mechanism to marshal individual molecular adhesive interactions into strong bonds between cells.

Isolation and identification of proteolipid proteins in jimpy mouse brain.

Proteolipids were isolated from 20 day old normal and jimpy mouse brain by extraction into chloroform-methanol (2:1, w/v), delipidated by size-exclusion HPLC, and analyzed by SDS-PAGE, Western blots, amino acid analyses, and N-terminal sequencing. SDS-PAGE showed that a major proteolipid from jimpy mouse brain had an apparent molecular weight of approximately 23 kDa, intermediate to that of PLP and DM-20 from normal mouse brain. Western blots with 3 different antibodies which recognize residues 200-224, 116-150, and 270-276 respectively recognized immunoreactive material in normal and jimpy PLP. Since antibody reactive with 270-276 did not recognize jimpy PLP, an altered C-terminus of the jimpy protein is suggested. These results demonstrated that a PLP can be partially purified from jimpy mouse brain. Amino acid analyses failed to show the predicted increase in cysteinyl residues (predicted from cDNA) in jimpy PLP. However, when jimpy brain proteolipids were subjected to N-terminal sequencing, Gly, Leu, Leu, Gly the first four amino acids of PLP were detected. Thus, the partial purification of a proteolipid from jimpy mouse brain, whose characteristics (apparent molecular weight, immunoreactivity, N-terminal sequence and relative net charge) strongly suggested that PLP of altered size is present in jimpy mouse brain.

Characterization of myelin basic protein charge isomers from adult mouse brain.

ADULT mouse MBP charge isomers (C1 or component 1, C2 or component 2, etc.) were purified from an acid soluble brain protein fraction by cation exchange chromatography. They were characterized by their elution profiles, their migration rates in alkaline-urea gels and by SDS-PAGE. Mouse C1 and C2 were both 14 kD in size, while C3, C4 and C5 consisted of the 18.5, 17 and 14 kD isoforms. Comparison of mouse and human MBP charge isomers showed that they were similar in that they both showed extensive charge heterogeneity, but different in that mouse MBP charge isomers were more cationic than their human counterparts. Finally, a possible explanation for the presence of the 14 kD MBP isoform in mouse myelin was suggested.

Myelin basic protein is affected by reduced synthesis of myelin proteolipid protein in the jimpy mouse.

Myelin basic proteins (MBPs) from 6-day-old, 10-day-old, 20-day-old and adult normal mouse brain were compared with those from 20-day-old jimpy (dysmyelinating mutant) mouse brain to determine the effect of reduced levels of proteolipid protein (PLP) on MBPs. Alkaline-urea-gel electrophoresis showed that 6-day-old and 10-day-old normal and jimpy MBPs lacked charge microheterogeneity, since C8 (the least cationic of the components; not be confused with complement component C8) was the only charge isomer present. In contrast, MBPs from 20-day-old and adult normal mouse brain displayed extensive charge microheterogeneity, having at least eight components. A 32 kDa MBP was the major isoform observed on immunoblots of acid-soluble protein from 6-day-old and 10-day-old normal and 20-day-old jimpy mouse brain. There were eight bands present in 20-day-old and adult normal mouse brain. Purified human MBP charge heteromers C1, C2, C3 and C4 reacted strongly with rat 14 kDa MBP antiserum, whereas the reaction with human C8 was weak. This suggested that MBPs from early-myelinating and jimpy mice did not react to MBP antisera because C8 was the major charge isomer in these animals. Purification of MBPs from normal and jimpy brain by alkaline-gel electrophoresis showed that both normal and jimpy MBPs have size heterogeneity when subjected to SDS/PAGE. However, the size isoforms in normal mouse brain (32, 21, 18.5, 17 and 14 kDa) differed from those in jimpy brain (32, 21, 20, 17, 15 and 14 kDa) in both size and relative amounts. Amino acid analyses of MBPs from jimpy brain showed an increase in glutamic acid, alanine and ornithine, and a decrease in histidine, arginine and proline. The changes in glutamic acid, ornithine and arginine are characteristic of the differences observed in human C8 when compared with C1.

Interaction of hepatic asialoglycoprotein receptor with dimyristoyl phosphatidylcholine vesicles.

The hepatocyte membrane asialoglycoprotein receptor (ASGP-R) was extracted from rabbit liver, purified, and then incubated with preformed vesicles of dimyristoyl phosphatidylcholine. The association of protein with lipid was dependent on vesicle size and the best results were achieved with small vesicles of about 20 nm diameter. The ligand binding capacity of ASGP-R-vesicle complexes was also measured and found to be approximately sevenfold greater than free receptor in aqueous buffer and twofold greater than receptor solubilized in Triton X-100. Most likely, the reconstitution procedure used in these experiments does not result in transmembrane insertion of the receptor. ASGP-R probably resides on the surface of the vesicle, held there primarily by weak hydrophobic forces.