Intercellular adhesive selectivity. I. An improved assay for the measurement of embryonic chick intercellular adhesion (liver and other tissues)

An improved assay for measuring intercellular adhesive selectivity of embryonic chick liver cells is described. Three major improvements over earlier procedures are noted: (a) enhanced reproducibility of liver cell-liver cell aggregate adhesion (homotypic adhesion) was achieved; (b) 25-70% of the input cells adhered to the collecting aggregates during the course of routine experiments as compared to the 0.25% in earlier assays. This increase in cellular adhesion suggests that the observed cell pick-up is a characteristic of the majority of the dissociated liver cell population; (c) the rate of intercellular adhesion was increased 1,000-fold. The main feature of the assay is that it measures the tissue adhesive selectivities of the dissociated cell population. Studies were undertaken on three embryonic chick tissues (liver, neural retina, and mesencephalon) to determine the tissue selectivity of intercellular adhesion of these dissociated cell types. Some general properties of liver cell homotypic adhesion have been studied and are reported.     

AN ASSAY FOR INTERCELLULAR ADHESIVE SPECIFICITY

A modification of an assay for intercellular adhesive specificity is described. The method involves the collection of radioactively labeled cells by aggregates of the same (isotypic aggregates) or different (heterotypic aggregates) types of tissue and determination of the number of cells collected by liquid scintillation counting. The use of ³²P to label the tissues permitted a much more rapid estimation of cell collection than was obtained previously. With the use of chick embryo neural retina, liver, forebrain, pectoral muscle, and heart ventricle tissue, it was shown that isotypic was always greater than heterotypic collection. Labeled neural retina cell collection by neural retina aggregates was studied as a function of time, cell suspension density, aggregate diameter, temperature, and aggregate number. Neural retina aggregates were treated with certain enzymes in an attempt to determine whether specific changes on the surface of the aggregates would interfere with labeled neural retina cell collection. Of the various proteases and glycosidases tested, only β-galactosidase rendered the surface more nonspecific.     

Transformation-induced alterations in adhesion : Binding of pre-formed cell aggregates to cell layers

The formation of intercellular adhesions by mouse 3T3 cells and their SV40-transformed derivatives is analysed by measuring the binding of pre-formed aggregates of these cells to cell layers or to a plastic substratum. The rationale for this procedure is to reduce the effects of cell dissociation on quantitative assessments of adhesive interactions. The fibroblasts within the aggregates retain the growth characteristics these cells show in monolayer culture. The proportion of aggregates binding is independent of the number of aggregates added and changes with time in a manner consistent with a first-order process, allowing the percent aggregates binding per unit time to serve as a parameter of intercellular adhesion. The rate of binding in homologous adhesive interactions is slower than in heterologous ones, binding in 3T3SV interactions is slower than in 3T3 interactions, and binding to cellular substrata is slower than to plastic. Binding of 3T3SV aggregates is readily distinguished from binding of 3T3 aggregates by the presence of a brief lag in binding rate, the formation of irregular projections from the bound aggregate, and a differential effect on binding rates of varying the temperature or of treating a single reactant with glutaraldehyde. Thus, there are quantitative and qualitative differences in the adhesive interactions of normal and transformed cells. The distinct binding properties of 3T3SV aggregates and the greater binding rates in heterologous interactions may be relevant to the invasive behavior of transformed cells in vivo.     

Intercellular adhesive selectivity. III. Species selectivity of embryonic liver intercellular adhesion

A species difference in the intercellular adhesive selectivity of mixtures of embryonic liver cells is reported. This is first quantitative assessment of species differences in the intercellular adhesive properties of embryonic cells. A collecting aggregate assay, a new double-label assay procedure, and histological and autoradiographic procedures were used to elucidate the intercellular adhesive selectivity of developing mammalian and avian liver cells. Evidence is presented that the reported adhesive differences are not due to the different cell types composing the respective embryonic mammalian and avian livers. Finally, such heterolgous-homotypic selectivity of adhesion is not a property of all tissues, since it is shown that developing brain cells (mesencephalon) do not exhibit the avove intercellular adhesive selectivity (mammalian vs. avian). These findings provide further support for the hypothesis that generic identity as well as cell type may play an important part in determining the intercellular adhesive behavior of heterologous-homotypic mixtures of embryonic cells. A possible evolutionary divergence of morphogenetic mechanisms is discussed.     

Acquisition of synchronous beating between embryonic heart cell aggregates and layers

Synchronous beating between chick embryonic heart cell aggregates and heart cell layers was used to study the relationship between intercellular adhesion and ionic coupling. Adhesion was measured by counting the proportion of aggregates which were not to be removed from cell layers by gentle washing after a 30 min incubation. Synchrony between bound aggregates and contiguous layers was assessed by phase microscopy. The first evidence of synchrony was seen 1.5 h after addition of aggregates to layers, following which there was an increase in the percentage of aggregates beating synchronously, reaching over 50% at 7 h and slowly increasing to a maximum of 65% by 24 h. Scanning electron microscopy and autoradiography of thymidine-labeled cells suggest that synchrony does not depend on cell movement at the interface between aggregate and layer. Acquisition of synchrony can be prevented completely by inhibiting protein synthesis, although pulsation of aggregates and layers continues in proportions unchanged from controls. After reversal of protein synthesis inhibition, synchrony is acquired at a rate and to an extent closely resembling that of newly adherent controls. These data indicate that ionic coupling is neither an inevitable nor an immediate consequence of adhesion. Since ionic coupling has been shown to correlate with the presence of gap junctions, the findings suggest that gap junctions are not involved in the initial events responsible for intercellular adhesion in vitro and that their formation following adhesion in this system may depend upon protein synthesis.     

Freeze-fracture study on the erythrocyte membrane during malarial parasite invasion

A new method is presented for the quantitative analysis of intercellular adhesive specificity. In this assay, two cell types are mixed, one unlabeled and the other labeled with the fluorescent dye, fluorescamine [4-phenylspiro(feran-2[3H],1'-phthalan)-3,3'-dione]. The resulting aggregates are analyzed by fluorescence microscopy to determine the number of labeled and unlabeled cells per aggregate. Random (nonspecific) aggregation was characterized by a binomial distribution, and adhesive specificity was accordingly quantified by the deviation (as determined by a chi-square test) from the calculated binomial distribution. The labeling procedure was simple and rapid, and experiments with 18 different cell types showed that it did not affect cell viability, morphology, rate and extent of adhesion, plating efficiency, and the capability of myogenic cells to undergo terminal differentiation. Most important, assays with morphologically identifiable cell pairs indicated that the fluorescent label neither induced apparent nor destroyed existing adhesive specificity. The most pronounced adhesive specificities were observed with freshly explanted cells from adult tissues and also with mixtures of simian virus 40-transformed and nontransformed BALB/c 3T3 cells. A glucosamine-6-phosphate N-acetylase-deficient mutant 3T3 line (AD6), however, aggregated randomly with parental 3T3 cells. Lectin-resistant mutant Chinese hamster ovary (CHO) cells displayed marginal adhesive specificity when mixed with normal CHO cells.     

Adhesive specificity in normal and transformed mouse fibroblasts

Adhesive specificity was studied in normal and transformed $\text{Balb}\text{c}$ mouse fibroblasts by comparing the number of labeled cells collected from a suspension of these cells by aggregates of various cell types. Aggregates of the two malignant cells examined collected either very many cells (aggregates of SV3T3 cells) or very few cells (aggregates of 3T12 cells). In addition, the relative adhesive behavior of these two aggregate types did not vary according to the cell suspension in which they were circulated. These data make it unnecessary to assume that malignancy is always accompanied by a decrease in intercellular adhesion.The adhesive behavior of normal 3T3 cell aggregates, compared to the aggregates composed of either malignant cell type, varied according to the type of cells in the suspension. Aggregates of 3T3 cells collected an appreciable number of SV3T3 cells but few 3T12 cells. Collection of 3T3 cells by 3T3 aggregates was also low if the 3T3 cells of the suspension were harvested from confluent cultures. However, collection of 3T3 cells by 3T3 aggregates increased significantly, as compared to collection by SV3T3 and 3T12 aggregates in the same cell suspension, if the 3T3 suspension was prepared from sparse cultures.Flat-revertants of SV3T3 cells were also studied. These cells behave like nonmalignant 3T3 cells rather than like the SV3T3 cells from which they were derived.We suggest that malignancy may not be caused by decreased intercellular adhesion as compared to normal cells but, perhaps, by decreased intercellular recognition.     

Intercellular adhesive selectivity. II. Properties of embryonic chick liver cell-cell adhesion

Studies directed at understanding the molecular basis of liver cell homotypic adhesion are presented. An assay which measures the rate of adhesion of isotopically labeled (32PO4) embryonic chick liver cells to liver cell aggregates, described in a companion paper, has been used to investigate the problem of intercellular adhesive selectivity. Cation requirements, the effects of various inhibitors of metabolism and protein synthesis, of chelators (EDTA and EGTA), and the effects of temperature on liver cell adhesion are reported. Two mechanisms of inhibition of liver intercellular adhesion are suggested. One involves destruction of cell-surface adhesion receptors (sensitivity to proteases); the other is an energy-dependent step which may involve alterations in plasma membrane conformation and/or membrane fluidity. Finally, a model is suggested for liver cell-cell adhesion that incorporates the early tissue selectivity of intercellular adhesion previously reported, followed by a multistep process which leads to histogenic aggregation.     

Can retinal adhesion mechanisms determine cell-sorting patterns: a test of the differential adhesion hypothesis

Embryonic chick neural retina cells possess two classes of adhesion mechanism, one Ca2+-independent, one Ca2+-dependent, responsible for short-term cell aggregation. This study investigates the role of these mechanisms in the long-term cell sorting potentially relevant to in vivo histogenesis. Retina cells are prepared either with both (E cells) or with only one mechanism (TC cells, CD; LTE cells, CI), respectively. The two types of cell preparations are differentially labelled using fluorescein or rhodamine isothiocyanate, mixed and allowed to aggregate in the presence or absence of cycloheximide at 0.5 microgram ml-1 to retard metabolic recovery of the removed adhesive mechanism. When observed by fluorescence and phase-contrast microscopy, the aggregates formed in cycloheximide show cell sorting, the cells with both mechanisms assuming a more interior position relative to those with a single adhesion mechanism. In parallel hanging-drop experiments, preformed aggregates of cells with a single adhesion mechanism are seen to spread upon aggregates of cells with both mechanisms. No sorting occurs amongst cells from a given stage prepared using any single dissociation protocol. The observed cell sorting would thus seem to derive exclusively from differential cell adhesiveness dependent upon the different dissociation conditions and maintained in the presence of cycloheximide. The experiments support the hypothesis that the dual CI and CD adhesion mechanisms in question can play a central role in governing cell-sorting behaviour during normal histogenesis.     

Isolation of an adhesion-mediating protein from chick neural retina adherons

Adherons are high molecular weight glycoprotein complexes which are released into the growth medium of cultured cells. They mediate the adhesive interactions of many cell types, including those of embryonic chick neural retina. The cell surface receptor for chick neural retina adherons has been purified, and shown to be a heparan sulfate proteoglycan (Schubert, D., and M. LaCorbiere, 1985, J. Cell Biol., 100:56-63). This paper describes the isolation and characterization of a protein in neural retina adherons which interacts specifically with the cell surface receptor. The 20,000-mol-wt protein, called retinal purpurin (RP), stimulates neural retina cell-substratum adhesion and prolongs the survival of neural retina cells in culture. The RP protein interacts with heparin and heparan sulfate, but not with other glycosaminoglycans. Monovalent antibodies against RP inhibit RP-cell adhesion as well as adheron-cell interactions. The RP protein is found in neural retina, but not in other tissues such as brain and muscle. These data suggest that RP plays a role in both the survival and adhesive interactions of neural retina cells.     

Cell adhesion dynamics and actin cytoskeleton reorganization in HepG2 cell aggregates

The temporal dependence of cytoskeletal remodelling on cell-cell contact in HepG2 cells has been established here. Cell-cell contact occurred in an ultrasound standing wave trap designed to form and levitate a 2-D cell aggregate, allowing intercellular adhesive interactions to proceed, free from the influences of solid substrata. Membrane spreading at the point of contact and change in cell circularity reached 50% of their final values within 2.2 min of contact. Junctional F-actin increased at the interface but lagged behind membrane spreading, reaching 50% of its final value in 4.4 min. Aggregates had good mechanical stability after 15 min in the trap. The implication of this temporal dependence on the sequential progress of adhesion processes is discussed. These results provide insight into how biomimetic cell aggregates with some liver cell functions might be assembled in a systematic, controlled manner in a 3-D ultrasound trap.     

Flow cytometric analysis and modeling of cell-cell adhesive interactions: the neutrophil as a model

The immune function of granulocytes, monocytes, lymphocytes, and other specialized cells depends upon intercellular adhesion. In many cases the molecules mediating leukocyte cell adhesion belong to the Leu-CAM superfamily of adhesive molecules. To elucidate the events of homotypic aggregation in a quantitative fashion, we have examined the aggregation of neutrophils stimulated with formyl peptides, where aggregate formation is a transient reversible cell function. We have mathematically modeled the kinetics of aggregation using a linear model based on particle geometry and rates of aggregate formation and breakup. The time course was modeled as a three-phase process, each phase with distinct rate constants. Aggregate formation was measured on the flow cytometer; singlets and larger particles were distinguished using the intravital stain LDS-751. Aggregation proceeded rapidly after stimulation with formyl peptide (CHO-nle-leu-phe-nle-tyr-lys). The first phase lasted 30-60 s; this was modeled with the largest aggregation rate and smallest rate of disaggregation. Aggregate formation plateaued during the second phase which lasted up to 2.5 min. This phase was modeled with an aggregation rate nearly an order of magnitude less than that of the initial fast phase, whereas the disaggregation rate for this phase did not change significantly. A third phase where disaggregation predominated, lasted the remaining 2-3 min and was modeled with a four to fivefold increase of the disaggregation rate. The mechanism of cell-cell adhesion in the plateau phase was probed with the monoclonal antibody IB4 to the CD18 subunit of the adhesive receptor CR3. Based on these studies it appears that new aggregates do not form to a large degree after the first phase of aggregate formation is complete. However, new adhesive contact sites may form within the contact region of these adherent cells to keep the aggregates together.     

Enzymatic dissection of embryonic cell adhesive mechanisms

In this paper we describe a kinetic assay for cell adhesion which measures the formation of cell clusters. Cluster formation is dependent on both calcium and protein synthesis, two parameters essential for the formation of histotypic aggregates. We also describe modifications of the stndard method for trypsinization of tissues which result in populations of single cells that appear to bear intact and functional cell surface adhesive systems. These modifications involve the use of chymotrypsin and the inclusion of calcium during enzyme digestion of tissues with trypsin and chymotrypsin. Using the cluster formation assay and the modified tissue dissociation techniques, we demonstrate the presence of two functionally distinct adhesive systems operating among embryonic chick neural retina cells. These two systems differ in proteolytic sensitivity, protection by calcium against proteolysis, dependence on calcium for function and morphogenetic potential. Cells possessing one of these intact adhesive systems are capable of extensive morphogenetic interactions in the absence of protein synthesis.     

Motility of endodermal epithelial cells plays a major role in reorganizing the two epithelial layers in Hydra

Cell-cell interactions and cell rearrangements play important roles during development. Aggregates of Hydra cells reorganize into the two epithelial layers and subsequently form a normal animal. Examination of the formation of the two layers under various situations, indicates that the motility of endodermal epithelial cells, but not the differential adhesive forces of the two types of epithelial cells, plays the critical role in setting up the two epithelial layers. (1) When aggregates of ectodermal cells and of endodermal cells were placed in direct contact, the endodermal cells migrated into the interior of the ectodermal aggregate. This process was completely inhibited by cytochalasin B although initial firm attachment between the two aggregates was not blocked. (2) A single endodermal epithelial cell placed in contact with an ectodermal aggregate, actively extended pseudopod-like structures and migrated toward the center of the ectodermal aggregate. In contrast, an ectodermal epithelial cell remained in contact with an endodermal aggregate and never exhibited migratory behavior. Cytochalasin treatment of only endodermal epithelial cells abolished the migration. (3) One to 4 endodermal epithelial cells and/or ectodermal epithelial cells were placed in contact with one another forming up to 4-cell aggregates. Endodermal epithelial cells exhibited high motility that can be attributed to the migratory movement described above. Finally, formation of actin bundles, as visualized with rhodamine-phalloidin, was always correlated with pseudopod formation in endodermal epithelial cells during early and mid stages of aggregate formation.     

Relation between the rate of cell movement under agarose and the positioning of cells in heterotypic aggregates

Single-cell suspensions prepared from 9-day-old chick tissues (skeletal muscle, liver, and neural retina) were used to investigate a possible relationship between intrinsic mobilities of different cell types and their positioning behavior in mixed (heterotypic) cellular aggregates. The relative mobilities of the three cell types, determined by comparing their ability to migrate under an agarose layer, was muscle greater than liver greater than neural retina. The gyratory shaker method was employed to produce heterotypic aggregates from mixed suspensions of muscle, liver, and neural retina cells and the tissue-specific positioning of cells after 24 h in culture was determined from histological and autoradiograph sections. The hierarchy for "inside" positioning of segregated cells was muscle greater than liver greater than neural retina cells, correlating with the rate of movement of these cells in the migration assay. The implication of the results is that relative speed of movement may determine the positioning of cells in heterotypic aggregates.     

A requirement for trypsin-sensitive cell-surface components for cell-cell interactions of embryonic neural retina cells

A quantitative assay was used to measure the rate of collection of a population of embryonic neural retina cells to the surface of cell aggregates. The rate of collection of freshly trysinized cells was limited in the initial stages by the rate of replacement of trypsin-sensitive cell- surface components. When cells were preincubated, or "recovered," and then added to cell aggregates, collection occurred at a linear rate and was independent of protein and glycoprotein synthesis. The adhesion of recovered cells was temperature and energy dependent, and was reversibly inhibited by cytochalasin B. Colchicine had little effect on collection of recovered cells. Antiserum directed against recovered cell membranes was shown to bind to recovered cells by indirect immunofluorescence. The antiserum also was shown to inhibit collection of recovered cells to aggregates, suggesting that at least some of the antigens identified might be involved in the adhesion process. The inhibitory effect of the antiserum was dose dependent . Freshly trypsinized cells absorbed neither the immunofluorescence activity nor the adhesion-inhibiting activity. Recovered cells absorbed away both activities. In specificity studies, dorsal neural retina cells adhered to aggregates of ventral optic tectum in preference to aggregates of dorsal optic tectum. The adhesive specificity of the dorsal retina cells was less sensitive to trypsin than the adhesive specificity of ventral retina cells which adhered preferentially to dorsal tectal aggregates only after a period of recovery.     

An instructive role for the interstitial matrix in tissue patterning: tissue segregation and intercellular invasion

Intercellular invasion is the intrusion of the cells of one tissue into space occupied by a second tissue. The alternative situation to invasion, one characteristic of most coherent tissues, is segregation, with identifiable boundaries existing between contiguous tissues. The interfaces between mesenchymal and myocardial tissues in the developing avian heart show a profoundly different character in different regions of the heart: the interface between epicardial mesenchyme and heart wall myocardium is planar, without intermingling of the two cell types, whereas the interface between endocardial cushion mesenchyme and myocardium is diffuse, with extensive invasion of both tissue types across the border to produce intermingling of the two tissues. Thus, invasion and tissue segregation coexist in different regions of the mesenchyme-myocardium contact zone. Investigation of the involvement of the interstitial matrix in invasion and segregation has been conducted by maintaining the two tissues in mutual contact in organ culture. Investigation of the mechanisms by which the two cell types sort out in randomized chimeric tissue reaggregates has provided insight into the conditions for tissue segregation. We have modeled invasion in organ culture by fusing aggregates of myocardial cells with aggregates of cardiac mesenchymal cells. Cells of both tissues invaded the partner aggregate during a period of 1-3 d of coculture. Both invasion and segregation in the aggregates appear to depend on the presence or absence of a fibronectin-rich interstitial matrix elaborated by the cardiac mesenchyme. During sorting, the matrix appears selectively in regions occupied by the mesenchyme. Under conditions of culture that are nonpermissive for matrix deposition, sorting fails to occur. Stimulation of matrix deposition by addition of serum, transforming growth factor beta, or isolated matrix itself is accompanied by sorting out of the two tissues. Sorting out is blocked reversibly by inclusion of the fibronectin adhesion site peptide, GRGDSP. Invasion of fused aggregates is preceded by a redistribution of the fibronectin-containing matrix of the mesenchymal aggregate such that matrix-poor regions come to occupy the interface with the myocardial partner aggregate. The invasion that ensues involves mesenchymal cells emigrating from, and myocardial cells intruding into, matrix-poor regions of the mesenchymal aggregate.(ABSTRACT TRUNCATED AT 400 WORDS)     

Specificity of intercellular adhesion mediated by various members of the immunoglobulin supergene family

The immunoglobulin supergene family members have been shown to be involved in cell-cell recognition and interaction during cell growth and differentiation. Neural cell adhesion molecule, myelin-associated glycoprotein, and carcinoembryonic antigen (CEA) are immunoglobulin supergene family members which can mediate cell adhesion. We show here that nonspecific cross-reacting antigen (NCA), a closely related CEA family member, is found on the surface of rodent cells transfected with functional NCA complementary DNA in different glycosylated forms, all of which can be deglycosylated to an Mr 35,000 core protein. Furthermore, NCA can mediate Ca2(+)-independent, homotypic aggregation of these NCA-producing transfectant cells. Since CEA has three internal repeated C2-set, immunoglobulin-like domains, whereas NCA has one, only one such domain is required for the intercellular adhesive function. We also demonstrate that NCA- and CEA-producing transfectants can form heterotypic aggregates, whereas mixtures of CEA or NCA transfectants and neural cell adhesion molecule or long form-myelin-associated glycoprotein transfectants sort themselves out into homotypic aggregates. The results suggest that subsets of the immunoglobulin superfamily, such as the CEA family, can be used in both homotypic and heterotypic cellular interactions, whereas less closely related members of the family can be used to separate different cell types by strictly homotypic interactions.     

Utilization of -glutamine in intercellular adhesion: Ascites tumor and embryonic cells

-Glutamine is required for the synthesis of complex carbohydrates required for the intercellular adhesion of mouse teratoma cells. It remained to be seen if these pathways were of general importance in the adhesion of other cell types. In this study, using an electronic particle counter assay to measure cell adhesion, Ehrlich ascites, Sarcoma 180 and Taper liver ascites tumor cells require exogenous -glutamine to aggregate. This effect is concentration dependent and the amino sugar, -glucosamine, replaces the glutamine requirement. Structural analogs of the active compounds are substantially less effective and metabolic inhibitors block the activity of the effective compounds. Two specific glutamine antagonists, DON (6-diazo-5-oxo- -norleucine) and azaserine (O-diazoacetyl-serine) decrease the action of -glutamine but not of -glucosamine. Trypsin dissociated six day old chick embryo neural retina cells do not require -glutamine to reaggregate, though the rate of aggregation is enhanced after preincubation with glutamine. Dissociation of small clumps of neural retina and inhibition of reaggregation of these cells are facilitated by preincubation with azaserine for 3–5 h. -Glutamine reduces the effect of azaserine on retina cells. These results are consistent with known metabolic pathways and suggest that -glutamine is involved in the synthesis of complex carbohydrates necessary for adhesion in a variety of cell types. The defective adhesion of the tumor cells examined may result from inability to produce glutamine synthetase, or effectively store cr transport -glutamine.     

The differential adhesion hypothesis: a direct evaluation

The differential adhesion hypothesis (DAH), advanced in the 1960s, proposed that the liquid-like tissue-spreading and cell segregation phenomena of development arise from tissue surface tensions that in turn arise from differences in intercellular adhesiveness. Our earlier measurements of liquid-like cell aggregate surface tensions have shown that, without exception, a cell aggregate of lower surface tension tends to envelop one of higher surface tension to which it adheres. We here measure the surface tensions of L cell aggregates transfected to express N-, P- or E-cadherin in varied, measured amounts. We report that in these aggregates, in which cadherins are essentially the only cell-cell adhesion molecules, the aggregate surface tensions are a direct, linear function of cadherin expression level. Taken together with our earlier results, the conclusion follows that the liquid-like morphogenetic cell and tissue rearrangements of cell sorting, tissue spreading and segregation represent self-assembly processes guided by the diminution of adhesive-free energy as cells tend to maximize their mutual binding. This conclusion relates to the physics governing these morphogenetic phenomena and applies independently of issues such as the specificities of intercellular adhesives.