Prostaglandin dependence of membrane order changes during myogenesis in vitro.

Myogenic differentiation in vitro involves at least three events at the cell surface: binding of prostaglandin to cells, cell-cell adhesion, and fusion of the myoblast membranes into syncytia. Previous work has suggested that binding of prostaglandin is causal to the change in cell-cell adhesion and that both are accompanied by a characteristic reorganization of the myoblast membrane detected as a transient increase in membrane order by electron paramagnetic resonance. We show here that this membrane order change, which reaches a maximum at 38 h of development in vitro, was the last membrane order change before bilayer fusion which begins several hours later. This membrane order change, which accompanies the change in cell-cell adhesion, was dependent on the availability of prostaglandin. In myoblasts maintained in indomethacin, where further differentiation is known to be blocked at the prostaglandin binding step, the membrane order change did not occur. However, if myoblasts are provided with exogenous prostaglandin, the membrane order change occurred and differentiation proceeded. The results indicate that the basis of the membrane order change was the reorganization of myoblast membranes to allow increased adhesion and prepare the membrane for bilayer fusion. They also demonstrate that, like the increase in myoblast adhesion, the membrane order change was dependent on prostaglandin being available to bind to its receptor.

Changes in myoblast membrane order during differentiation as measured by EPR.

The events which make possible the characteristic fusion of the cell membranes of embryonic myoblasts are known to involve modification of the cell membrane (Hausman, R.E., Dobi, E.T., Woodford, E.J., Petrides, S., Ernst, M. and Nichols E.B. (1986) Dev. Biol. 113, 40-48). Myoblasts from chick embryos were allowed to differentiate in gyrotory aggregate culture and the order of their membranes was measured by EPR. Two spin-labels which insert at different depths into the lipid bilayer were used. Measurement with the 5-nitroxystearate label showed an increase in myoblast membrane order (2T' parallel) from 0-15 h of culture and again from 26-38 h of culture. Measurement with the 12-nitroxystearate label showed the 0-15 h increase in order but the second increase was greatly reduced and shifted in time. While the specific sources of these changes in membrane order cannot yet be identified, the changes observed correlated well with known events of myogenic differentiation in vitro. The initial increase in membrane order occurred while the myoblasts were recovering from the effects of trypsin dissociation and undergoing gyrotory aggregation. The second increase in membrane order occurred during the known period of prostaglandin receptor activity and increased cell-cell adhesion.

Changes in myoblast membrane electrical properties during cell-cell adhesion and fusion in vitro.

Characteristic of the process of myogenesis are the changes in the composition and organization of the cell membrane. While poorly understood, these changes have biochemical and biophysical relevance. Recently, changes in molecular order of the myoblast membrane which accompany differentiation in vitro have been observed (Santini, M.T., Indovina, P.L. and Hausman, R.E. (1987) Biochim. Biophys. Acta 896, 19-25). To further investigate these cell fusion processes we have examined additional physical parameters: conductivity and permittivity of the myoblast membrane during differentiation which reflect the molecular arrangement of the membrane. The determination of these parameters is possible because in the radio frequency range suspensions of cells in an electrolyte buffer show a characteristic conductivity dispersion due to the interfacial polarization. An analysis of our experimental data based on a 'single-shell' model showed that conductivity and permittivity of the membrane of pre- and post-fusion myoblasts varied significantly and abruptly. The conductivity of the cell interior (cytosol) remained constant. We discuss the significance of the observed changes in these membrane parameters for myogenesis.

Rescue of the Li+-induced delay of embryonic myogenesis in vitro by added inositol.

Signaling between embryonic myoblasts to coordinate gene expression is part of normal skeletal muscle development in the embryo. An unanswered question is the nature of the second messengers carrying the information to the nucleus. We have investigated the cell membrane events associated with the binding of prostaglandin to a transient receptor on the embryonic chick myoblast membrane in vitro. The membrane events include a transient change in membrane order seen by electron paramagnetic resonance (EPR), a change in cell-cell adhesion, a rapid decrease in membrane permeability and fusion of the membrane bilayers. The addition of 20 mM Li+, an inhibitor of inositol phosphate phosphatase, perturbed the transient change in membrane order and delayed the change in cell-cell adhesion and conductivity for 2-6 h. Other alkali metal ions had no such effects. The addition of inositol to the culture medium in the continued presence of Li+ restored the normal timing of the two latter events. We interpret this as evidence for an inositol phosphate second messenger system which might connect the activation of the prostaglandin receptor with the change in cell-cell adhesion, the changes in membrane conductivity and perhaps bilayer fusion. We suggest that Li+, by blocking the regeneration of polyphosphatidylinositol from inositol phosphate, reduced the efficiency of the second messenger system such that further differentiation of the myoblast membrane was delayed. The exogenous inositol provided an alternative source and membrane differentiation was unaffected.

Distribution of R-cognin in late embryonic and post-hatching chicken retina.

The histological distribution of R-cognin in chick retinas was determined from embryonic day 8 through 13 wk post-hatching by indirect immunofluorescence using polyclonal anticognin. On embryonic day 8, at a developmental stage without distinct retina layers, most of the cells within the tissue exhibited fluorescence. By embryonic day 12, when the strata of the retina are delineating and initial synapses are beginning to form, R-cognin fluorescence became concentrated in the nascent ganglion cell, and to a lesser extent, inner nuclear layers. By embryonic day 16 (extending through post-hatching day 26), the staining of R-cognin was specific for the ganglion cell and nerve fiber layers. Fluorescence was prominent on the cell somal membranes but also was present on the processes of these cells at the ganglion cell layer-inner plexiform layer interface. Pre-immune serum and anticognin after preabsorption with R-cognin exhibited no fluorescence. The results demonstrated that the known decrease in R-cognin found in the retina during the latter half of embryonic development in the chick is not uniform across the retina, but that R-cognin is preferentially retained on cells within the ganglion cell layer. While cells within the ganglion cell layer also exhibit alpha-bungarotoxin binding, the majority of the latter is found in the inner plexiform layer. Thus, the observations are consistent with a role for R-cognin in the formation or maintenance of functional cell-cell connections within the entire retina prior to developmental day 11, or in retinal ganglion cell layer formation or stability subsequent to embryonic day 11.

Antisense inhibition of R-cognin expression modulates differentiation of retinal neurons in vitro.

PURPOSE: Retina cognin (R-cognin) is a 50 kDa membrane-associated polypeptide expressed during retinogenesis where it is involved in mediating tissue-specific cell-cell interactions. In addition to its intercellular role in aggregation, R-cognin may act as a cell surface signaling molecule. An antisense oligonucleotide was used to inhibit R-cognin expression and to investigate the effects of this inhibition on subsequent neuronal differentiation. METHODS: Cultures of retina cells were prepared from 6 day (E6) and 8 day (E8) chicken embryos and were incubated with a deoxyoligonucleotide complimentary to 20 bases of the sequence encoding R-cognin or random oligonucleotides. The levels of choline acetyltransferase (ChAT) and glutamic acid decarboxylase (GAD), markers of cholinergic and GABAergic differentiation, respectively, were detected by Western blots on protein extracts from treated cultures. RESULTS: The antisense treatment inhibited ChAT levels at E6 and GAD levels at E8. The treatment resulted in no decrease in the level of the enzyme glyceraldehyde 3-phosphate dehydrogenase. A random oligonucleotide did not affect the levels of any of the proteins. CONCLUSIONS: These results confirm the cell recognition role of R-cognin and suggest that it is important in intracellular signaling cascades necessary for normal retina development.

Determination of fractal dimension of physiologically characterized neurons in two and three dimensions.

Although there is a growing interest in the application of fractal analysis in neurobiology, questions about the methodology have restricted its wider application. In this report we discuss some of the underlying principles for fractal analysis, we propose the cumulative-mass method as a standard method and we extend the applicability of fractal analysis to both 2 and 3 dimensions. We have examined the relationship between the method of log-log Sholl analysis and fractal analysis and have found that they correlate well. Measurements of physiologically characterized retinal ganglion cells indicate that different cell types can have significantly different fractal dimensions. Such differences may allow the correlation of the physiological type of a neuron with its morphological fractal dimension.

The cell recognition molecule, cognin, mediates choline acetyltransferase activity in embryonic chick retina.

Cell signaling and cell-cell interactions play an important role in neuronal differentiation in the embryonic CNS. Previous work (Hausman, R.E., Vivek Sagar, G.D. and Shah, B.H., Dev. Brain Res., 59 (1991) 31-37) had shown that cholinergic differentiation in the embryonic chick retina depends on insulin and neuron-neuron interactions. Here, we pursued the molecular nature of that dependence on cell interactions. The embryonic chick retina is known to contain several cell adhesion or recognition molecules. We asked if retina cognin, a 50 kDa cell surface-associated protein, played a role in controlling cholinergic differentiation in the developing chick retina. As previously, cholinergic differentiation was measured by two markers: choline acetyltransferase (ChAT) activity and high-affinity choline uptake. We used polyclonal antibody to cognin to determine if blocking cognin-mediated cell interactions would affect the normal embryonic increases in these cholinergic markers. We demonstrated a 40% inhibition of the normal developmental appearance of ChAT activity in retina neuronal cultures from early development, but no effect in cultures from more differentiated retina. The inhibition was selective for retina, since it was not seen in neural tissues like cerebrum and cerebellum that also express ChAT. In contrast to the effect of insulin, choline uptake was not affected by treatment with cognin antibody. Antibodies to two other cell recognition molecules present in the retina (Ng-CAM and N-cadherin) did not block the normal developmental appearance of ChAT. These results suggest that cognin-mediated interactions play a unique role in the control of one aspect of cholinergic differentiation in the developing chick retina.

Role of the cell recognition molecule, cognin, in GABAergic differentiation in chick retina.

Previous work showed that GABAergic differentiation in developing chick retina depends on insulin and cell interactions. Here, we investigated whether it depended on cell signaling mediated by retina cognin, a 50 kDa cell recognition molecule. Cognin mediates cell adhesion in vitro and occurs on retinal neurons that become both GABAergic and cholinergic. We investigated two markers of GABAergic differentiation: glutamate decarboxylase (GAD) activity and high-affinity GABA uptake. Both increase during differentiation of retinal neurons in culture and can be easily measured. We blocked cognin-mediated cell signaling with cognin antibody and found a reduction of the developmental increase in GAD activity in cultures of retinal neurons from 7 and 11 day chick embryos. There was no reduction of high-affinity GABA uptake. This suggested that cognin-mediated signaling was necessary for the normal developmental increase in GAD but not for high-affinity GABA uptake. These results contrasted with our previous observations on cholinergic differentiation in cultured retinal neurons. We found that cognin antibody blocked the normal developmental increase in choline acetyltransferase (ChAT) only if the cells were exposed before embryonic day 7. Thus, while both GAD and ChAT activity appear to be controlled by cell signaling involving cognin, the periods of developmental sensitivity for the two differentiation markers are different. Antibodies to other adhesion molecules, Ng-CAM, and N-cadherin, did not similarly affect GAD activity. Antibodies to laminin at a 10-fold higher concentration inhibited GAD activity only in early embryonic retina. Tests for protein synthesis and "housekeeping" enzyme activity demonstrated that the cognin antibody effect was selective for neuronal differentiation pathways.(ABSTRACT TRUNCATED AT 250 WORDS)

Topological correlation between the cell-recognition protein, R-cognin and α-bungarotoxin receptor in retinal plasma membrane.

The amount of the neural retina cell recognition protein, R-cognin, in the plasma membrane of chick embryo neural retina cells declined 43% between 10 and 17 days of embryonic development. Over this period there was a 27% increase in the plasma membrane content of the α-bungarotoxin receptor. Plasma membranes of both these ages were sonicated into vesicles and these vesicles partitioned on α-bungarotoxin agarose beads into those which contained detectable α-bungarotoxin receptor and those which did not. At 10 days, approximately 6% of the plasma membrane vesicles contained receptor. At 17 days, <2% did. At 10 days, 60% of the R-cognin was found in the α-bungarotoxin receptor-containing vesicles, at 17 days 86%. At 17 days, 6% of the retina membrane with a high concentration of both α-bungarotoxin receptor and R-cognin was of a density indicative of it being of synaptic origin. These results suggested that R-cognin and α-bungarotoxin receptor occurred close together in the plasma membrane of retina cells. However, the lack of competition between R-cognin gamma globulin and specific α-bungarotoxin binding indicated that the α-bungarotoxin receptor and R-cognin were not the same protein. Thus, R-cognin and the α-bungarotoxin receptor appear to be separate proteins which occur in close proximity on the retina plasma membrane.

Cognin distribution during differentiation of embryonic chick retinal cells in vitro.

Differentiation of individual retina neurons is closely linked to development of retina function. This differentiation may be intrinsic to the cell or determined by the position of the cell within the developing tissue. Retina cognin, a cell-cell recognition protein, which may itself mediate position-dependent cell interactions in vivo exhibits a characteristic change in distribution during embryonic chick development. Cognin is progressively lost from the outer retina in a manner which appears position-dependent. We asked if this change in cognin distribution was actually position-dependent or intrinsic to the retina cells. Neural retina cells from 8-day-old chick embryos were cultured in vitro. Continued differentiation of the cultured cells was demonstrated by neurite outgrowth and characteristic increases in choline acetyltransferase and glutamic acid decarboxylase activity. In such cultures, the characteristic developmentally related disappearance of retina cognin occurred as in vivo. This indicated that this aspect of retina neuronal differentiation was independent of position within the tissue and likely intrinsic to individual cells after 8 days of embryonic development.

Effect of insulin on GABAergic development in the embryonic chick retina.

We investigated the role of insulin in GABAergic differentiation in the embryonic chick retina at different embryonic ages using glutamate decarboxylase (GAD) and high-affinity GABA uptake as developmental markers. Both these GABAergic markers exhibit developmentally programmed increases in activity during retinogenesis that also occur in culture. Insulin stimulated GABA uptake in retina neurons at all embryonic ages in a dose-dependent manner and GAD activity by 30% in embryonic retina neurons after 11 days of development. The stimulation of GABA uptake by insulin was blocked by addition of ouabain suggesting a role for the Na+,K+ ATPase. The same concentration of insulin caused a 76% stimulation of protein synthesis in these retinal cells, and previous work demonstrated that insulin also stimulates cholinergic differentiation in the chick retina (Hausman et al., Dev. Brain. Res. 59, (1991) 31-37). Thus, there was no selective stimulation of GABAergic differentiation by insulin but likely a neurotrophic effect. The increase in GAD activity in neurons from post-11-day embryonic neurons contrasts with our previous findings at embryonic days 6-7 where there is little change in GAD activity after addition of insulin. It is possible that the failure of insulin to stimulate GAD activity during early retina development is due to the increased accumulation of GABA in the presence of insulin. GABA levels were increased more than two-fold by 100 ng/ml insulin.

Initial cholinergic differentiation in embryonic chick retina is responsive to insulin and cell-cell interactions.

Previous work [Kyriakis et al., Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 7463-7467] had shown that insulin, when added during a window of binding from embryonic days 9-11, stimulates the normal developmental increase in choline acetyltransferase (ChAT) activity (a marker for cholinergic differentiation) in cultured embryonic chick retinal neurons. Here, we investigated the effect of insulin and IGF 1 on embryonic chick retinal neurons at the stage of development (embryonic day 6) when ChAT activity is first expressed. We investigated insulin peptide effects in retinal tissue developing in vitro as well as in cultures of retinal cells. We show that insulin also stimulated the initial embryonic increase in ChAT activity but had no stimulatory effect on glutamic acid decarboxylase activity (a marker for GABAergic differentiation), an enzyme whose activity also increases developmentally in the same retinal neurons. In fact, insulin inhibited the expression of GAD activity in the retina. The insulin-mediated increase in ChAT activity was independent of normal cell-cell interactions but could not replace them. Insulin also stimulated choline uptake but only after a two day delay, suggesting that the normal program for cholinergic differentiation in the chick retina was induced by insulin. IGF 1 did not have any effect on either cholinergic or GABAergic differentiation. We conclude that cholinergic differentiation in chick embryo retinal neurons is dependent on both insulin- and cell contact-mediated signals.

Thioreductase activity of retina cognin and its role in cell adhesion.

Retina cognin (R-cognin) is a 50-kDa protein on the surface of embryonic chick retina cells that mediates cell-cell recognition and neuronal differentiation. It is developmental stage- and tissue-specific in its expression. The partial cDNA clone for R-cognin is nearly identical to that of chicken protein disulfide isomerase (chicken PDI) and enzyme with thioreductase activity. The R-cognin clone extends from beyond the 3' polyadenylation site up to the boundary between PDI exons 1 and 2, with the putative R-cognin equivalent of PDI exon 1 remaining uncloned. The question posed here was whether the sequence-specific properties of PDI were significant in the action of R-cognin. We show that R-cognin, like PDI, has thioreductase activity as revealed by RNase renaturation enzymatic assays. We then asked if this thioreductase activity was involved in the mediation of cell adhesion and recognition in developing chick retina. We show, through cell aggregation assays, that both R-cognin and chicken PDI enhance chick retina cell aggregation but not that of cells from other CNS tissues. We also show that treating R-cognin and chicken PDI with the thioreductase inhibitor 5,5'-dithio-bis (2-nitrobenzoic acid), which covalently binds to the functional cysteines of the thioreductase active sites, reduces the enhancement of cell aggregation. Thus R-cognin acts, in part, by catalyzing a covalent protein-protein linkage at the cell surface.

Developmental localization of retina cognin synthesis by in situ hybridization.

Retina cognin (R-cognin) is a 50 kDa protein involved in cell recognition and neuronal differentiation during development of the embryonic chick retina. Initial characterization of a partial cDNA encoding R-cognin revealed a striking similarity to the cDNA encoding protein disulfide isomerase (PDI), a 57 kDa multifunctional protein. The exact nature of the relationship between R-cognin and PDI is not known; however, both proteins appear to be encoded by the same gene. In the present study, we developed cRNA probes to examine the expression of R-cognin and PDI transcripts in embryonic chick retina and liver. In the retina, the amount of transcript decreased with embryonic age, in parallel to a similar decrease in R-cognin protein. In the liver, where PDI is prominently expressed, the amount of transcript was not developmentally regulated. The spatial and temporal pattern of expression of the R-cognin-encoding retinal transcript was examined by in situ hybridization. R-cognin mRNA was expressed in cells across the retina early in retinogenesis, but became restricted to the cells of the inner retina later in development. This pattern of expression was the same as the developmental pattern of R-cognin protein [Dobi et al., Invest. Ophthalmol. Vis. Sci. 27, (1986) p. 323-329], thus, demonstrating that this secreted protein functions at the surface of the cells where it is transcribed.

Effect of viscosity on neurite outgrowth and fractal dimension.

The growth mechanism by which neurons achieve their characteristic ramified morphology has long been of interest, but determining whether physical parameters, such as viscosity, are important has been difficult due to a lack of useful hypotheses and standard reproducible techniques. We have recently shown that neurons exhibit fractal behavior and that their fractal dimension (df) is consistent with a physical process called diffusion-limited aggregation (DLA). We suggested that this DLA behavior might stem from viscosity differences, chemical gradients or electrical fields (Caserta et al., Phys. Rev. Lett., 64 (1990) 95-98). DLA is a model for a large family of growth processes. In order for a process to fit the DLA model, the growth rate must be proportional to the gradient of a field at a point on the growing structure (Feder, Plenum, New York, 1988, Ch. 4). Chemical, electrical, or fluid pressure fields can fit the model depending on the particular physical system under study. Here, we studied growth of retinal neurons from chick embryos in culture media of various fluid viscosities. Thus, we test whether DLA in this system was based on a fluid pressure field. As viscosity was increased from 1 to 4.3 cps, the number of neurite branches decreased 98%. However, there was no effect on df. Over this range of viscosities, total cellular protein synthesis decreased only 17%. The results indicate that, while differences in viscosity between the interior and exterior of the cell affect neurite outgrowth, they do not affect the fractal behavior of neurons. Thus, viscosity differences are not the basis for the DLA pattern of neuronal arborization.

Increase in homotypic aggregation of metastatic Morris hepatoma cells after fusion with membranes from non-metastatic cells.

Three types of Morris hepatoma: 7288C, 5123tc, and 9618A were assessed for their ability to grow and produce metastases in the lungs of male buffalo rats. Hepatoma 7288 formed metastatic growths more rapidly than 5123, while 9618 did not form metastatic growths in the lung at all. Cells from the same three hepatomas were then assessed for their ability to aggregate homotypically in vitro. This was used as a measure of potential for detachment from the site of primary growth in vivo, one of the critical steps in metastasis. The results were that 9618 aggregated more effectively than 5123 and 7288 did not aggregate homotypically. Thus, the ability to metastasize and the ability to aggregate homotypically were inversely related in these three tumors, with the most metastatic tumor, 7288, hardly aggregating at all. Was the ability of hepatoma 9618 to aggregate homotypically and the inability of hepatoma 7288 to do so a characteristic of the plasma membrane? Plasma membranes either from the tumor which aggregated best, 9618, or from the liver of non-tumorous neonatal buffalo rats were fused into the surface of the 7288 cells. Either fusion conferred the ability to aggregate on the previously non-aggregating metastatic hepatoma cells. In contrast, fusion of plasma membranes from hepatoma 7288 into 9618 cells failed to affect their ability to aggregate homotypically. The aggregation-conferring effect required actual fusion of the donor membrane, was dependent on the amount of membrane fused and was sensitive to treatment of the donor plasma membranes with trypsin. The results are interpreted as suggesting the presence of homotypic aggregation materials, probably proteins, in the plasma membranes of non-metastatic cells. These may be reduced or absent in the membrane of the highly metastatic hepatomas.

Prostaglandin binding does not require direct cell-cell contact during chick myogenesis in vitro.

Myogenic differentiation in vitro involves at least three events at the cell surface: binding of prostaglandin to the cells, contact-mediated cell-cell recognition, and fusion of the myoblast membranes into myotubes. While the earlier events are thought to be necessary for subsequent fusion, the sequence of events has not been determined. A major impediment to determining the initial event has been the lack of synchrony of cell differentiation in vitro. To overcome this, we cultured chick embryo myoblasts as a suspension of single cells in gyratory rotation in medium without added Ca2+. Under these conditions, myoblasts exhibited characteristic prostaglandin binding at 34 h. Within 30 min, the cells began to aggregate. Because this occurred without change of medium or conditions of rotation, we termed the process autoaggregation. Within 8-10 h. cells within these autoaggregates began to fuse into syncytia. These results suggest that an early cell surface event in embryonic myogenesis is the characteristic binding of prostaglandin to the myoblasts. The results demonstrate that this binding precedes any direct cell-cell contact and suggest that it causes the subsequent change in myoblast cell-cell adhesion.

Vesicle interactions as a model for the retinal cell-cell recognition mediated by R-cognin.

The action of R-cognin, a chick neural retina cell recognition glycoprotein, was investigated in vesicles of retina cell membranes. It was found that the aggregation of the vesicles was dependent on membrane proteins, and specifically R-cognin, as vesicle aggregation was inhibited by R-cognin antibody (Fab'). The R-cognin content of the vesicles, and their ability to aggregate, decreased with increasing embryonic age of the tissue. R-cognin mediated aggregation of the vesicles was not dependent on exogenous calcium. Thus, R-cognin was calcium-independent in its membrane linking activity.

Effects of cell signaling on the development of GABA receptors in chick retina neurons.

R-cognin, a cell recognition molecule, and insulin are known to play significant roles in GABAergic differentiation in the developing chick retina. In the present study, the effects of insulin and R-cognin on post-synaptic (GABAceptive) differentiation were investigated. In ovo binding of [3H]GABA and [3H]flunitrazepam ([3H]Flu) to the GABA and benzodiazepine (BZD) receptors, respectively, remained at low levels during early embryogenesis but increased sharply from mid-embryogenesis through hatching, increases which also occur in cultured neurons from early-embryonic (E7) and mid-embryonic (E11) chick retina. E7 neurons respond to insulin treatment (100 ng/ml) with increased [3H]Flu binding but no change in [3H]GABA binding. Cognin antibody (10 micrograms/ml) treatment of E7 neurons caused no significant inhibition of the developmental increases in binding of either radioligand. Insulin in E11 cultures led to greater developmental increases in binding sites for both radioligands, but exposure to cognin antibody was without significant effect. These data, along with previous studies, indicate that GABAergic differentiation in developing chick retina is regulated, in part, by insulin and cognin-mediated cell signaling. Insulin also regulates post-synaptic (GABAceptive) differentiation whereas cognin-mediated interactions are relatively insignificant.