Structural changes at the myogenic cell surface during the formation of myotendinous junctions

Summary Myotendinous junctions are sites which are morphologically and molecularly specialized for force transmission between intracellular and extracellular structural proteins. In the present investigation, the formation of these specialized junctions is studied in chicken embryos from 9 days following fertilization to 1 day posthatching, using light and electron microscopy. Observations indicate that the first discernible event in myotendinous junction formation is the appearance of basement membrane at the incipient junction at 9–10 days postfertilization, concomitant with the aggregation of fibroblasts at the junctional regions of myogenic cells. Subsequently, subsarcolemmal densities appear at sites opposite basement membrane locations by 13 days postfertilization. Myofibrils insert into subsarcolemmal densities by day 15 and invaginations of the cell membrane are initiated at those insertions. Type I collagen fibers appear at the cell surface at day 17. Junctional structure at day 17 qualitatively resembles that of adult myotendinous junctions. Changes in junctional structure following day 17 are primarily increases in the amount of subsarcolemmal densities, myofibril-membrane associations, and amount of junctional membrane folding.     

Desmin at myotendinous junctions

Myofibrils are linked to the cell membrane at myotendinous junctions located at the ends of muscle fibers, and at costameres, sites positioned periodically along lateral surfaces of muscle cells. Both of these sites are enriched in proteins that link active components of myofibrils to the cell membrane. Costameres are also enriched in desmin intermediate filaments that link passive components of myofibrils to the lateral surfaces of muscle cells. In this study, the possibility that desmin is also found between the terminal Z-disk of myofibrils and the myotendinous junction membrane is examined by immunocytochemistry and by KI-extraction procedures. Data presented show that desmin is located in the filamentous core of cellular processes at myotendinous junctions at sites 30 nm or more from the membrane. This core lies deep to subsarcolemmal material previously shown to contain talin, vinculin, and dystrophin. The distance from desmin to the membrane suggests desmin does not interact directly with membrane proteins at the junction. Immunoblots and indirect immunofluorescence of junctional regions of muscle compared to nonjunctional regions show no apparent enrichment of desmin at junctional sites, although vinculin, another costameric and junctional component, is significantly enriched at junctional regions. These findings show that passive elements of myofibrils may be continuous from myotendinous junctions of muscle origin to insertion via desmin filaments located between terminal Z-disks and the junctional membrane. This can provide a system in parallel to that involving thin filaments, vinculin, and talin for linking myofibrils to the cell membrane at myotendinous junctions.     

Compartmental nitrate concentrations in barley root cells measured with nitrate-selective microelectrodes and by single-cell sap sampling

Nitrate-selective microelectrodes were used to measure intracellular nitrate concentrations (as activities) in epidermal and cortical cells of roots of 5-d-old barley (Hordeum vulgare L.) seedlings grown in nutrient solution containing 10 mol · m^–3 nitrate. Measurements in each cell type grouped into two populations with mean (±SE) values of 5.4 ± 0.5 mol · m^–3 (n=19) and 41.8 ± 2.6 mol · m^–3 (n = 35) in epidermal cells, and 3.2 ± 1.2 mol · m^–3 (n = 4) and 72.8 ± 8.4 mol · m^–3 (n = 13) in cortical cells. These could represent the cytoplasmic and vacuolar nitrate concentrations, respectively, in each cell type. To test this hypothesis, a single-cell sampling procedure was used to withdraw a vacuolar sap sample from individual epidermal and cortical cells. Measurement of the nitrate concentration in these samples by a fluorometric nitrate-reductase assay confirmed a mean vacuolar nitrate concentration of 52.6 ± 5.3 mol · m^–3 (n = 10) in epidermal cells and 101.2 ± 4.8 mol · m^–3 (n = 44) in cortical cells. The nitrate-reductase assay gave only a single population of measurements in each cell type, supporting the hypothesis that the higher of the two populations of electrode measurements in each cell type are vacuolar in origin. Differences in the absolute values obtained by these methods are probably related to the fact that the nitrate electrodes were calibrated against nitrate activity but the enzymic assay against concentration. Furthermore, a 28-h time course for the accumulation of nitrate measured with electrodes in epidermal cells showed the apparent cytoplasmic measurements remained constant at 5.0 ± 0.7 mol · m^–3, while the vacuole accumulated nitrate to 30–50 mol · m^–3. The implications of the data for mechanisms of nitrate transport at the plasma membrane and tonoplast are discussed.Symbol _n ² Chi-squared with n degrees of freedom R.-G.Z. was awarded a Sino-British Friendship Scholarship sponsored by the British Council and H.-W.K. was supported by an AFRC Linked Research Grant to A.D.T for collaboration with R.A.L. We wish to thank Dr. K. Goulding for advice on ion chromatography, Dr. K. Moore for assistance with statistical analysis and Dr. J.H. Williams for advice on the microsample analysis.     

Myomuscular junctions in the gill-sac muscle of the Atlantic hagfish,Myxine glutinosa: analogy with myotendinous junctions

Summary Myomuscular junctions between muscle fibers in the gill sacs of the Atlantic hagfish, Myxine glutinosa, were examined by electron microscopy. According to the presence of sarcolemmal differentiations typical of myotendinous junctions, the myomuscular junctions can be described as a symmetric myotendinous junction     

Gap junctions are non-randomly distributed inDrosophila wing discs

Summary The distribution of gap junctions in mature larvalDrosophila melanogaster wing discs was analyzed by means of quantitative electron microscopy. Gap junctions are non-randomly distributed in the proximal-distal disc axis and in the apical-basal cell axis of the epithelium. In the epithelial cells, the surface density, number and length of gap junctions are greatest in the apical cell region and distal disc region. The average gap junction surface density is 0.0572 m^–1 and 2.77% of the lateral cell surface is composed of gap junctions. In the adepithelial cells, the gap junction surface density is 0.0005 m^–1 and 0.06% of the cell surface is composed of gap junctions. No gap junctions were observed between epithelial cells and adepithelial cells. The absolute area of gap junctions was estimated in a proximal-distal strip of cells in the disc and is considerably less in the folded regions of the epithelium compared to the flat notum and wing pouch regions. The results are discussed with respect to pattern formation and growth control in imaginal discs.     

Adhesive strength of single muscle cells to basement membrane at myotendinous junctions

Whole muscles loaded to failure frequently fail at or near myotendinous junctions. The present investigation was directed toward determining the breaking stress and failure site of intact and injured myotendinous junction preparations consisting of muscle cells dissected free from surrounding parallel structures but still attached to tendon collagen fibers. These tests show that the breaking stress for intact myotendinous units is 2.7 x 10(5) N/m2, expressed relative to cell cross-sectional area. Failure occurs immediately external to the junction membrane between the cell membrane and lamina densa of the basement membrane. Site and stress at failure are independent of strain and strain rate over a biologically relevant range. Breaking stress in the plane of the membrane, corrected for membrane folding, is 1.2 X 10(4) N/m2. This value is not significantly greater than stress at maximum isometric tension for these cells at these sarcomere lengths. After compression injury, cells fail within the compression site at significantly lower stress (1.9 X 10(5) N/m2). These findings suggest that, in muscle strain injuries that occur under conditions simulated here, failure occurs at myotendinous junctions unless the muscle has suffered previous compression injury leading to failure within the muscle.     

The hydraulic conductivity as a criterion for the membrane integrity of protoplasts fused by an electric field pulse

The hydraulic conductivity of the membrane, Lp, of fused plant protoplasts was measured and compared to that for unfused cells, in order to identify possible changes in membrane properties resulting from the fusion process. Fusion was achieved by an electric field pulse which induced breakdown in the membranes of protoplasts in close contact. Close membrane contact was established by dielectrophoresis. In some experiments pronase was added during field application; pronase stabilizes protoplasts against high field pulses and long exposure times to the field. The Lp-values were obtained from the shrinking and swelling kinetics in response to osmotic stress. The Lp-values of fused mesophyll cell protoplasts of Avena sativa L. and of mesophyll and guard cell protoplasts of Vicia faba L. were found to be 1.9±0.9·10^-6, 3.2±2.2·10^-6, and 0.8±0.7·10^-6 cm·bar^-1·s^-1, respectively. Within the limits of error, no changes in the Lp-values of fused protoplasts could be detected in comparison to unfused protoplasts. The Lp-values are in the range of those reported for walled cells of higher plants, as revealed by the pressure probe.Abbreviations GCP guard cell protoplast - Lp hydraulic conductivity - MCP mesophyll cell protoplast     

Investigation of electrophysiological responses of Neurospora crassa to blue light

The influence of light in a spectrum range of 350–500 nm 20 W m^-2 (20,000 erg · cm^-2 · s^-1) has been studied in the mycelial cells of Neurospora crassa. Light-induced input resistance and membrane potential changes can be measured by means of intracellular microelectrodes. The value of the input resistance reached maximum after a 2–5 min illumination. The maximum hyperpolarization of the cell membrane reaching 30–40 mV was observed after 20–25 min illumination, when the input resistance values did not differ significantly in the illuminated and non-illuminated cells.     

The C. elegans MAGI-1 protein is a novel component of cell junctions that is required for junctional compartmentalization

Cell junctions are essential to maintain polarity and tissue integrity. Epithelial cell junctions are composed of distinct sub-compartments that together ensure the strong adhesion between neighboring cells. In Caenorhabditis elegans epithelia, the apical junction (CeAJ) forms a single electron-dense structure, but at the molecular level it is composed of two sub-compartments that function redundantly and localize independently as two distinct but adjacent circumferential rings on the lateral plasma membrane. While investigating the role of the multi PDZ-domain containing protein MAGI-1 during C. elegans epidermal morphogenesis, we found that MAGI-1 localizes apical to both the Cadherin/Catenin (CCC) and AJM-1/DLG-1 (DAC) containing sub-domains. Removal of MAGI-1 function causes a loss of junctional compartmentalization along the lateral membrane and reduces the overall robustness of cell-cell adhesion mediated by either type of cell junctions. Our results suggest that MAGI-1 functions as an “organizer” that ensures the correct segregation of different cell adhesion complexes into distinct domains along the lateral plasma membrane. Thus, the formation of stable junctions requires the proper distribution of the CCC and DAC adhesion protein complexes along the lateral plasma membrane.     

Different ratios of sucrose/raffinose-induced membrane depolarizations in the mesophyll of species with symplasmic (Catharanthus roseus, Ocimum basilicum) or apoplasmic (Impatiens walleriana, Vicia faba) minor-vein configurations

In mesophyll cells of species with a symplasmic (Ocimum basilicum, Catharanthus roseus, Magnolia denudata) or an apoplasmic (Vicia faba, Impatiens walleriana, Bellis perennis) minor-vein configuration, membrane depolarizations in response to 20 or 200 mol·m^–3 raffinose and sucrose were measured. Ageing period and resting potential marginally affected the degree of depolarization. The symplasmic species showed similar depolarization responses to 20 and 200 mol·m^–3 sucrose or raffinose. In the apoplasmic species, depolarization increased statistically significantly from 20 to 200 mol·m^–3 sucrose, whereas the depolarization response to raffinose was equal at both concentrations. In the apoplasmic species, moreover, the depolarization response to raffinose was significantly weaker than to sucrose at all concentrations. A major difference between symplasmic and apoplasmic species seems to lie in the scantiness of raffinose carriers in the mesophyll plasma membrane of species with the apoplasmic mode of phloem loading.Abbreviations 20R(200R) 20(200) mol·m^–3 raffinose - 20S(200S) 20(200) mol·m^–3 sucrose     

The kinetics of P515 in relation to the lipid composition of the thylakoid membrane

Flash-induced P515 absorbance changes have been studied in dark-adapted chloroplasts isolated from spinach plants grown under two different light intensities. The slow component (reaction 2), normally present in the P515 response of chloroplasts isolated from plants grown at an intensity of 60 W · m^–2, was largely reduced in chloroplasts isolated from plants grown at an intensity of 6 W · m^–2. This reduction of the slow component in the P515 response appeared to be coincident with an alteration in the lipid composition of the thylakoid membrane. Mainly the ratio monogalactosyldiacylglycerol to digalactosyldiacylglycerol appeared to be altered. In thylakoids from plants grown at 6 W · m^–2, the ratio was approximately 35% lower than that of plants grown at 60 W · m^–2. The amount of both cytochromeb ₅₆₃ and cytochromef was largely reduced in chloroplasts isolated from plants grown at low light intensity. These results may indicate a possible correlation between structural organization of the thylakoid membrane and the kinetics of the flash-induced P515 response.     

Gap junction ultrastructure in three states of conductance

Summary Junctional conductance between the epidermal cells of the beetle Tenebrio molitor is raised after exposure to the hormone 20-hydroxyecdysone and lowered reversibly by exposure to chlorpromazine. Gap Junctional particle size, density and arrangement associated with these conductance changes were studied. We found no significant difference in particle density in gap junctions of control (2456±471 particles/m², mean ±S.D.) and hormone-treated epidermis (2490±315); however, a significant increase in packing density occurred in chlorpromazine-uncoupled epidermis (3133±665). The particles are randomly arranged in all three states of conductance. Particle size measurements show that the E-face gap junctional particles are heterogeneous with a mean diameter ±S.D. of 15.2±2.0 nm. No significant difference in particle size between controls and experimentals was detected. Although glutaraldehyde irreversibly uncoupled these cells, the absence of glutaraldehyde fixation but presence of glycerol induced marked alterations in the appearance of the gap junctions such that quantification was no longer possible. From this particle-packing data and our previous thin-section data, we estimate that there are 90000 gap junctional particles per cell (within junctional plaques). The conductance of a single gap junctional channel (assuming one population) changes from 94 pS to 213 pS after hormone treatment.     

The permeability of the epidermal cell plasma membrane of barley leaves to abscisic acid

Uptake of ³H-labelled (±)-abscisic acid (ABA) into isolated barley (Hordeum vulgare L.) epidermal cell protoplasts (ECP) was followed over a range of pH values and ABA concentrations. The present results show that ABA uptake is not always linearly correlated with the external concentration of undissociated ABA (ABAH). At pH 7.25, ABA uptake exhibited saturation kinetics with an apparent K _m value of 75 mmol·m^–3 to tal ABA. This saturable transport component was inhibited by pretreating the protoplasts with 1 mol·m^–3 p-chloromercuribenzenesulfonic acid at pH 8.0, conditions that minimized the uptake of this acid sulfhydryl reagent. Moreover, the rate of (±)-[^3]HABA uptake was reduced by addition of 0.1 mol·m^–3 (±)-ABA to 41%, whereas the same concentration of (±)-ABA was approximately half as effective (46% of the inhibitory effect). Thus, it was concluded that only (±)-ABA competes for an ABA carrier that is located in the epidermal cell plasma membrane. The permeability of the epidermal cell plasma membrane was studied by performing a Collander analysis. At pH 6 the overall plasma-membrane permeability of epidermal cells was similar to that of guard cells but was about two times higher than that of mesophyll cells.Abbreviations ABA abscisic acid - ABA^– anion of ABA - ABAH undissociated ABA - 2,4-D 2,4-dichlorophenoxyacetic acid - DMO 5,5-dimethyloxazolidine-2,4-dione - ECP deepidermal cell protoplast - K_r partition coefficient - M_r relative molecular mass - NEM N-ethylmaleimide - PCMBS p-chloromercuriben zenesulfonic acid - P_s permeability coefficient We are grateful to Barbara Dierich for expert technical assistance, to Prof. H. Gimmler (Lehrstuhl für Botanik I, Universität Würzburg, FRG) for helpful discussions and to the Deutsche Forschungsgemeinschaft (SFB 251, TP 3) for financial support.     

Nitrogen fixation in subarctic streams influenced by beaver (Castor canadensis)

Nitrogen fixation was measured in four subarctic streams substantially modified by beaver (Castor canadensis) in Quebec. Acetylene-ethylene (C₂H₂ C₂H₄) reduction techniques were used during the 1982 ice-free period (May–October) to estimate nitrogen fixation by microorganisms colonizing wood and sediment. Mean seasonal fixation rates were low and patchy, ranging from zero to 2.3 × 10^–3 µmol C₂H₄ · cm^–2 · h^–1 for wood, and from zero to 7.0 × 10^–3 µmol C₂H₄ · g AFDM^–1 · h^–1 for sediment; 77% of all wood and 63% of all sediment measurements showed no C₂H₂ reduction. Nonparametric statistical tests were unable to show a significant difference (p > 0.05) in C₂H₂ reduction rates between or within sites for wood species or by sediment depth.Nitrogen contributed by microorganisms colonizing wood in riffles of beaver influenced watersheds was small (e.g., 0.207 g N · m^–2 · y^–1) but greater than that for wood in beaver ponds (e.g., 0.008 g N · m^–2 · y^–1) or for streams without beaver (e.g., 0.003 g N · m^–2 · y^–1). Although mass specific nitrogen fixation rates did not change significantly as beaver transform riffles into ponds, the nitrogen fixed by organisms colonizing sediment in pond areas (e.g., 5.1 g N · m^–2 · y^–1) was greater than that in riffles (e.g., 0.42 g N · m^–2 · y^–1). The annual nitrogen contribution is proportional to the amount of sediment available for microbial colonization. We estimate that total nitrogen accumulation in sediment, per unit area, is enhanced 9 to 44 fold by beaver damming a section of stream.     

Vinculin in subsarcolemmal densities in chicken skeletal muscle: localization and relationship to intracellular and extracellular structures

Using immunocytochemical methods we have studied the distribution of vinculin in the anterior and posterior latissimus dorsi skeletal (ALD and PLD, respectively) muscles of the adult chicken. The ALD muscle is made up of both tonic (85%) and twitch (15%) myofibers, and the PLD muscle is made up entirely of twitch myofibers. In indirect immunofluorescence, antivinculin antibodies stained specific regions adjacent to the sarcolemma of the ALD and PLD muscles. In the central and myotendinous regions of the ALD, staining of the tonic fibers was intense all around the fiber periphery. Staining of the twitch fibers of both ALD and PLD muscles was intense only at neuromuscular junctions and myotendinous regions. Electron microscopy revealed subsarcolemmal, electron-dense plaques associated with the membrane only in those regions where vinculin was localized by immunofluorescence. Using antivinculin antibody and protein A conjugated to colloidal gold, we found that the electron-dense subsarcolemmal densities in the tonic fibers of the ALD contain vinculin; no other structures were labeled. The basal lamina overlying the densities appeared to be connected to the sarcolemma by fine, filamentous structures, more enriched at these sites than elsewhere along the muscle fiber. Increased amounts of endomysial connective tissue were often found just outside the basal lamina near the densities. In tonic ALD muscle fibers, the subsarcolemmal densities were present preferentially over the I-bands. In partially contracted ALD muscle, subsarcolemmal densities adjacent to the Z-disk appeared to be connected to that structure by short filaments. We propose that in the ALD muscle, through their association with the extracellular matrix, the densities stabilize the muscle membrane and perhaps assist in force transmission.     

Major involvement of connexin 43 in seminiferous epithelial junction dynamics and male fertility

In different epithelia, cell membranes contacting one another form intercellular junctional complexes including tight, adherens and gap junctions, which could mutually influence the expression of each other. We have here investigated the role of Cx43 in the control of adherens and tight junction proteins (N-cadherin, β-catenin, occludin and ZO-1) by using conditional Sertoli cell knockout Cx43 (SCCx43KO^−/−) transgenic mice and specific anti-Cx43 siRNA. Gap junction coupling and Cx43 levels were reduced in SCCx43KO^−/− as compared to Wild-type testes. Ultrastructural analysis revealed disappearance of gap junctions, the presence of tight and adherens junctions and persistent integrity of the blood-testis barrier in SCCx43KO^−/− testis. Occludin, N-cadherin and β-catenin levels were enhanced in SCCx43KO^−/− mice as compared to Wild-type animals whereas ZO-1 levels were reduced. Cx43 siRNA blocked gap junction functionality in Sertoli cells and altered tight and adherens protein levels. The Cx43 control of tight and adherens junctions appeared channel-dependent since gap junction blockers (glycyrrhetinic acid and oleamide) led to similar results. These data suggest that the control of spermatogenesis by Cx43 may be mediated through Sertoli cell Cx43 channels, which are required, not only in cell/cell communication between Sertoli and germ cells, but also in the regulation of other junctional proteins essential for the blood-testis barrier.     

Protamine alters structure and conductance ofNecturus gallbladder tight junctions without major electrical effects on the apical cell membrane

Summary Protamine is a naturally occurring basic protein (pI; 9.7 to 12.0). We have recently reported that protamine dissolved in the mucosal bath (2 to 20 m), induces about a twofold increase in transepithelial resistance inNecturus gallbladder within 10 min. Conductance decreased concomitantly with cation selectivity.In this leaky epithelium, where >90% of an applied current passes between cells, an increment in resistance of this magnitude suggests a paracellular actiona priori. To confirm this, ionic conductance across the apical cell membrane was studied with microelectrodes. Protamine increased transepithelial resistance without changing apical cell membrane voltage or fractional membrane resistance. Variation in extracellular K concentration (6 to 50mm) caused changes in apical membrane voltage not different from control.To determine if protamine-induced resistance changes were associated with structural alteration of tight junctions, gallbladders were fixedin situ at peak response and analyzed by freeze-fracture electron microscopy. According to a morphometrical analysis, the tight junctional intramembranous domain expands vertically due to incorporation of new strands (fibrils) into the main compact fibrillar meshwork.Since morphologic changes are complete within 10 min, strands are probably recycled into and out of the tight junctional membrane domain possibly by the cytoskeleton either from cytoplasmic vesicles or from intramembranous precursors. Regulation of tight junctional permeability by protamine and other perturbations may constitute a common mechanism by which leaky epithelia regulate transport, and protamine, in concentrations employed in this study, seems reasonably specific for the tight junction.     

Injury to nerves and the initiation of amphibian limb regeneration

The force produced within skeletal muscle fibers is transmitted to the bone via a myotendinous junction. This junctional region was examined by light and electron microscopy in the sartorius muscles of three Rana temporaria. The muscle fibers tapered and inserted at an angle of about 25 degrees with the connective tissue fascia near the bone. The composition of the structures within the last 100 microns of the fiber was analyzed morphometrically. The T-system, terminal cisternae, and caveolae were the same as in the central region of the muscle fiber. However, the mitochondrial content was higher and the volume of longitudinal sarcoplasmic reticulum was lower than elsewhere in the fiber. The membrane at the end of the fiber had extensive villiform processes interdigitating with the tendon. The surface area of the membrane around the villiform processes was estimated with point-counting techniques and calculated from the stereological equations appropriate for partially anisotropic structures. The extra membrane involved in the myotendinous junction was about 32 times that of the cross-sectional area of the fiber. Part of this additional membrane contained specialized adherens junctions through which the contractile proteins of the muscle are anchored to collagen. The increased area at the myotendinous junction presumably provides greater mechanical strength than a flat termination. The high values of membrane capacitance and specific resistance measured electrophysiologically at the end of the fiber also can be attributed to the characteristics of the terminal membrane structure.     

Cell junction and cyclic AMP: III. Promotion of junctional membrane permeability and junctional membrane particles in a junction-deficient cell type

Summary The cyclic nucleotide effect on junction was studied in C1-1D cells, a mouse cancer cell type that fails to make permeable junctions in ordinary confluent culture. Upon administration of cyclic AMP, dibutyryl cyclic AMP, dibutyryl cyclic AMP plus caffeine (db-cAMP-caffeine), or cholera toxin (an adenylate cyclase activator), the cells acquired permeable junctions; they became electrically coupled and transferred fluorescent tracer molecules among each other—a transfer exhibiting the molecular size limit of permeation of normal cell-to-cell channels. The effect took several hours to develop. With the db-cAMP-caffeine treatment, junctional permeability emerged within two hours in one-fifth of the cell opopulation, and within the next few hours in the entire population. This development was not prevented by the cytokinesis inhibitor cytochalasin B. Permeable junctions formed also in two other conditions where the cell-endogenous cyclic AMP level may be expected to increase: serum starvation and low cell density. After three weeks of starving the cells of serum, a junctional permeability arose in confluent cultures, which on feeding with serum disappeared within two to three days. At low cell density, namely below confluency, the cells made permeable junctions, unstarved. In cultures of rather uniform density, the frequency of permeable junctions was inversely related to the average density, over the subconfluent range; at densities of about 1×10⁴ cells/cm², where the cells had few mutual contacts, 80% of the pairs presumed to be in contact were electrically coupled. In cultures with adjoining territories of high (confluent) and low cell density, there was coupling only in the last, and in this low-density state the cells were also capable of coupling with other mammalian cell types (mouse 3T3-BalbC and human Lesch-Nyhan cells).Correlated electron microscopy of freeze-fractured cell junctions showed no membrane differentiation in confluent C1-1D cultures. The junctions acquired differentiations, namely particle clusters of gap junction and strands of tight junction, upon cyclic nucleotide application or serum starvation and in the lowdensity condition. With db-cAMP-caffeine, these differentiations appeared within 4 hr of the treatment (confluent cultures), growing in size over the next hours. Treatment with cycloheximide, but not with cytochalasin B, prevented the development of recognizable gap junction and tight junction in cultures supplied with db-cAMP-caffeine.