Cell Surface Architecture Research Articles

Cell-surface changes during in vitro differentiation of pluripotent embryonal carcinoma cells

When aggregates of HM-1 embryonal carcinoma (EC) cells were exposed to 10 −6 M retinoic acid for 2 days and cultured in medium lacking retinoic acid, they differentiated to nerve cells, endoderm cells, and myoblasts. Cells 2 days after initial exposure to retinoic acid were not significantly different from the parental EC cells, as judged by cell-surface architecture and by reactivity to lectins. On the fourth day, the surface of the aggregates was covered with two kinds of cells distinguishable from the parental cells. The round cells with short villi seemed to be precursors to endoderm cells. Receptors for Dolichos biflorus agglutinin (DBA) newly appeared and receptors for peanut agglutinin (PNA) were still expressed on their surfaces. The other cells, which were round cells with a few processes, might be precursors to nerve cells. PNA receptors had disappeared from their surfaces, and DBA receptors were not expressed. On the sixth day of differentiation, possible precursors to myoblasts were detected; they were flat cells with smooth surfaces. These cells lacked cell-surface receptors for the two lectins, while the precursor cells and the myoblasts excreted intercellular fibers reacting with PNA. HM-1 cells synthesized much embryoglycan, the structure of which was similar to that of the glycan isolated from quasinullipotent F9 cells. The only difference was that the glycan from HM-1 cells lacked DBA binding sites. Synthesis of fucosylated embryoglycan mainly decreased between the second and fourth day of differentiation. As above, cell-surface changes occurred mainly between the second and fourth day. The period seems to be important in determining the fate of the cells, since endoderm cells were scarcely seen among differentiated cells which had been continuously exposed to 10 −6 M retinoic acid during the period.

Cell surface and cytoskeletal interactions in a temperature-sensitive strain of SV40-transformed mouse fibroblasts.

In order to assess the role of cytoskeletal structure in modulating cell surface topography during cell transformation, cytoskeletal organization of 3T3 mouse cells transformed with a tsA mutant of simian virus 40 (SV40) was studied in detail by correlative light and electron microscopy. Detergent-extracted, critical-point dried whole cells observed in the electron microscope were seen to contain well-organized microfilament bundles (stress fibers) traversing the longitudinal axis of cells grown at the restrictive temperature (39 degrees C). When grown at the permissive temperature (32 degrees C), cells prepared in this manner were not observed to contain such structures. However, when semithin sections (0.5 micron) were viewed by transmission electron microscopy at 120 kV, short microfilament bundles were seen in 32 degrees C-grown cells. There was an alteration in the morphology of these structures at sites of attachment to the substratum (focal contacts), and they were shorter in length than microfilament bundles of 39 degrees C-grown cells. A difference was also observed between the two phenotypes in the layer of microfilaments associated with the dorsal cell surface. Since it is this layer that directly determines cell surface architecture, it is proposed that changes in microfilament bundle-generated surface tension are responsible for alterations of this layer, leading to an altered cell surface morphology. Tension may be modified by disturbances in focal contacts (or adjacent regions) or altered actin-associated protein(s).

FITC-labeled lectins and calcofluor white ST as probes for the investigation of the molecular architecture of cell surfaces. Studies on conjugatophycean species

In a screening program with 7 FITC-labeled lectins as probes, ConA receptors were identified in all of the 28 members of theConjugatophyceae, being under investigation. In nearly all of them RCA120 receptors, too, are expressed. In 3 species only, PNA receptors, and in 2 species UEA receptors have been detected. No binding of DBA, SBA, and WGA was observed. The receptors for ConA, RCA120, and UEA were shown to be associated with different molecules. Each lectin exhibits a unique and specific binding pattern, both chemically, as well as with regard to the topographic distribution on cell surfaces. While ConA receptors predominantly are associated with constituents of the cell wall, RCA120 receptors mostly form part of the surrounding mucilage; the same holds for UEA receptors. Besides a variability of topographic distribution and species-to-species variation, a cell-to-cell variation exists in many species, suggesting that the expression of a lectin receptor is due to the developmental state of the cell and/or depends on external stimuli. In conclusion, we may point out, that FITC-labeled lectins turned out to be extremely useful probes for the investigation of the molecular architecture of cell walls. Calcofluor white ST binding to fibrillar polysaccharides (most probably cellulose) was shown to be inhibited by external incrustations of the cell wall. One species does not show any reaction with calcofluor white ST at all.

Mechanisms That Regulate the Structural and Functional Architecture of Cell Surfaces

Publisher Summary This chapter emphasizes on the search for mechanisms controlling cell surface molecular and functional topography. The asymmetric surface topography is a predictable correlate of membrane deformation and microfilament asymmetry, independent of how asymmetry is induced and which cell type is studied. Tthe chapter demonstrates that receptors can be segregated out of and into regions of microfilament accumulation. The topographical heterogeneity is not restricted to membrane molecular determinants, but extends to a range of endocytic functions and to a macromolecular assembly, which is the coated pit. There is also a discussion about the arrest of dynamic surface events during mitosis. The chapter describes a working model to explain that the topographical data depends on initial recruitment of microfilaments to incipient protuberance, uropod, pseudopod, or cleavage furrow membranes. The resulting microfilament-membrane interaction acting inward and along the cell surface leads to the generation of tension that is propagated over the cell surface as a wave. Microfilament polarization is required to establish and maintain membrane structural asymmetry and to eventually relieve the accumulation of membrane components by endocytosis.

Chapter 8 The Brush Border of Intestinal Epithelium: A Model System for Analysis of Cell-Surface Architecture and Motility

Publisher Summary This chapter examines a model system for analysis of cell-surface architecture and motility of the brush border of intestinal epithelium. The apical surface of absorptive epithelial cells, such as those lining the lumen of the intestine or proximal tubule of the kidney, consists of a tightly packed array of microvilli variously referred to as the striated, microvillus, or brush border. The brush-border membrane contains a variety of membrane proteins, most of which are enzymes involved in the processing and transport of the nutrients absorbed by the small intestine. Brush borders are isolated from dissociated intestinal epithelial cells by homogenization and differential centrifugation. For cytological experiments, one chicken is used, and for biochemical studies, 5–10 animals are used. Microvilli are freed from the brush borders by homogenization in a Dounce homogenizer, fitted with a tight pestle. Homogenization is continued until the brush borders begin to fragment. As is the case for brush-border isolation, it is important to monitor the progress of homogenization by light microscopy. It is suggested that the contraction of the terminal web was caused by sarcomere-like myosin interactions with the basal ends of the microvillar core filaments and the actin filaments extending from the zonula adherens.

Pit formation and rapid changes in surface morphology of sympathetic neurons in response to nerve growth factor.

Scanning and transmission electron microscope studies were carried out on the rapid cell surface responses of cultured newborn rat sympathetic neurons to nerve growth factor (NGF), a substance that promotes their survival and differentiation. The somas of sympathetic neurons continuously exposed to NGF or deprived of the factor for 4-5 h have a very smooth surface. After readdition of NGF to the latter type of cultures, there is rapidly initiated a transient, sequential change in the cell surface. Microvilli and small ruffles appear within 30 s and are most prominent by 1 min. By 3 min of exposure, the microvilli and ruffles decrease in prominence, and by 7 min the somal surface is again smooth. By 30 s after NGF readdition, as increase in the number of 60- tp 130-nm coated pits is also detectable. This increase reaches a maximum of about threefold from 0.5 to 3 min and then gradually decreases. Alterations in the surface did not occur on the nonneuronal cell types present in the cultures and were not observed in response to another basic protein (cytochrome c) or to physical manipulation. Changes in cell surface architecture induced by NGF in normal sympathetic neurons and, as previously described, in PC12 pheochromocytoma cells indicate that such responses may present or reflect primary events in the mechanism of the factor's action.

In vivo and in vitro effects of proteases upon adenylate cyclase activity of liver cells and plasma membranes.

1. 1. Several hours after an intraperitoneal injection of papain (3 mg) the marked stimulation of adenylate cyclase activity was observed as well as the significant increase in cyclic AMP content of rat livers. The fluoride-sensitive adenylate cyclase activity was stimulated more than the basal activity. Trypsin, which is not mitogenic, was much less effective on both events, as compared to papain which is mitogenic to the liver in vivo. 2. 2. Papain stimulated the basal adenylate cyclase activity of isolated liver cells in vitro at concentrations larger than 2.5 μg papain per ml, but the sensitivity of adenylate cyclase activity to fluoride was completely lost upon increasing the papain concentration. Trypsin, on the other hand, gave no effect on the basal adenylate cyclase activity of isolated liver cells at enzyme concentrations from 0 to 500 μg/ml. The fluoride sensitivity was lost upon increasing trypsin concentration similarly to the case of papain. 3. 3. Adenylate cyclase activity (both basal and fluoride-sensitive) of isolated liver plasma membranes was stimulated by papain (but not by trypsin) at concentrations as low as about 0.25 μg papain/ml, similarly to the in vivo effect of proteases. Upon increasing the protease (papain, trypsin) concentration further, both the basal and fluoride-sensitive activities of plasma membranes were markedly reduced. 4. 4. Acid mucopolysaccharides of liver plasma membranes were released almost completely by treatment with papain at concentrations larger than 25 μg/ml, whereas trypsin was less effective in degrading proteoglycans associated with plasma membranes. 5. 5. The results were discussed in relation to the possible involvement of protease in mitogenesis of liver cells through the alteration of cell surface architecture and function.

Simian virus 40 T-antigen-related cell surface antigen: serological demonstration on simian virus 40-transformed monolayer cells in situ.

Simian virus 40 (SV40)-transformed monolayer cells were analyzed in situ by indirect immunofluorescence microscopy for the postulated cell surface location of SV40 T-antigen-related molecules. With antisera prepared against purified, sodium dodecyl sulfate-denatured SV40 T-antigen, positive surface staining was obtained when the cells had been treated with formaldehyde before immunofluorescence analysis. In contrast, living SV40-transformed cells analyzed in monolayer were surface fluorescence negative. The fixation procedure developed in this study combined with a double staining immunofluorescence technique allowed the simultaneous analysis of the same cells for the expression of both SV40 T-antigen-related surface antigen and nuclear T-antigen. The localization of SV40 T-antigen-related surface antigen on the outer surface of the plasma membrane of formaldehyde-fixed SV40-transformed cells was demonstrated directly by the protein A-mediated binding of Staphylococcus aureus bacteria on formaldehyde-fixed SV40-transformed cells precoated with antiserum against sodium dodecyl sulfate-denatured T-antigen. Both cell surface staining and S. aureus binding were found to be highly specific for SV40 T-antigen-related binding sites. These results indicate that T-antigen-related molecules in a cryptic form are located on the surface of SV40-transformed monolayer cells and can be detected in situ after modification of the cell surface architecture.

Culture density and age-dependent interactions of 3T3 AND SV3T3 cells with immobilized and soluble Lens culinaris lectin.

Short term binding of both Balb/c derived 3T3 cells and SV3T3 cells to 2B-Sepharose coated with Lens culinaris lectin (LCL) was compared with their LCL-induced agglutinability in relation to culture density and age. The cells were grown for 4 days either to low (LD) or to high density (HD); for certain experiments HD-cells were stimulated (HDS) by a short trypsin treatment and/or by addition of fresh medium. HD-3T3 cells bound somewhat faster to immobilized LCL than LD-3T3 cells, although LD-3T3 cells agglutinated at lower LCL concentrations. In the case of SV 3T3 cells, binding was much less pronounced for HD than for LD cells. The agglutinability of HD-SV 3T3 cells, however, was greater than that of the other transformed groups. Upon stimulation, binding and agglutinability data of both HD cell lines began to resemble the results obtained with LD-cells. Taken together, the date revealed in all cases an inverse relationship between binding and agglutinability, i.e., high agglutinability was closely correlated with slower binding and vice versa. The results indicate that culture density and age-dependent differences in cell surface architecture can be detected by short term binding to immobilized LCL.

Augmentation of proliferation and generation of specific cytotoxic cells in human mixed lymphocyte culture reactions by colchicine

Cellular proliferation and generation of cytotoxic lymphocytes in mixed lymphocyte culture reactions (MLC) results from cellular interactions initiated by individual differences in cell surface structures determined by the HLA-D locus. Cellular microtubular assemblies (MTA) modulate cell shape and surface architecture and have also been implicated in mediating stimulatory signals from the lymphocyte plasma membrane to intracellular sites. To assess the role of MTA in alloactivation, we tested the effect of colchicine, a microtubule-disrupting alkaloid, on the MLC. Diametrically opposite results were observed depending upon the colchicine treatment protocol. Brief exposure of stimulating cells to 10 −6 M colchicine resulted in an increase in [ 3H]thymidine incorporation from 30, 694 ± 2787 to 47,345 ± 4361 cpm/culture (mean ± SEM) ( P < 0.001), exposure of responding cells to 10 −6 M colchicine resulted in an increase from 33,054 ± 4012 to 46,790 ± 5458 cpm/culture ( P < 0.01), and exposure of both cells to 10 −6 M colchicine resulted in an increase from 33,054 ± 4012 to 52,685 ± 6720 cpm/culture ( P < 0.01). However, direct addition of colchicine to a final concentration of 10 −6 M to MLCs at different times resulted in complete suppression of proliferation when added as late as 96 hr, and 63% suppression when added at 120 hr. Pretreatment of stimulating, responding, or both cells with lumicolchicine did not enhance proliferation. Pretreatment of cells with 10 −6 and 10 −4 M colchicine enhanced proliferation, while pretreatment with 10 −2 M colchicine prevented blastogenesis. The potentiation of proliferation induced by colchicine was evident as early as 48 hr after the initiation of the MLC. Generation of specific cytotoxic cells in the MLC was also enhanced by exposing lymphocytes to colchicine prior to the proliferative phase in six out of eight experiments (specific chromium release = 168% of control) despite constant effector:target ratios. These findings indicate that early disruption of microtubules leads to enhanced cellular proliferation and generation of cytotoxic lymphocytes in allogeneic one-way MLCs and suggest that the state of polymerization of the MTA may modulate immune responses involving cell-cell interactions.

Chapter 10 Molecular and Lectin Probe Analyses of Neuronal Differentiation

This chapter discusses the changes occurring during neuronal maturation in neuroblastoma (NB) and dorsal root ganglia (DRG) in culture. The general properties and advantages of the NB system are reviewed. One may envisage that the sequence of events in the course of neuronal maturation would be an external change in the environment produces a change in the cell surface architecture, resulting in the activation of adenylate cyclase and, therefore, in an increase in the intracellular cAMP level. This increase in cAMP induces global physiological changes, including the induction of neurospecific enzymes. During development, examples of changes in the surface carbohydrate-containing molecules are numerous, and glycocomponents have been implicated in many processes including specific cell interaction and recognition. NB cultures have been widely used as a model system in approaching a number of problems related to developmental neurobiology. The possibility of culturing undifferentiated and differentiated cells has allowed the study of various molecular events occurring during differentiation. Differentiated NB cells are often characterized by neurites, a higher RNA and protein content, and a higher activity of neurospecific enzymes. All these molecular events are regulated at a posttranscriptional level, but the mechanisms have not been clarified.

Rapid, sequential changes in surface morphology of PC12 pheochromocytoma cells in response to nerve growth factor.

The effect of nerve growth factor (NGF), a substance that promotes the differentiation and maintenance of certain neurons, was studied via scanning electron microscopy utilizing the PC12 clonal NGF-responsive pheochromocytoma cell line. After 2-4 d of exposure to NGF, these cells acquire many of the properties of normal sympathic neurons. However, by phase microscopy, no changes are discernible within the first 12-18 h. Since the primary NGF receptor appears to be a membrane receptor, it seemed likely that some of the initial responses to the factor may be surface related. PC12 cells maintained without NGF are round to ovoid and have numerous microvilli and small blebs. After the addition of NGF, there is a rapidly initiated sequential change in the cell surface. Ruffles appear over the dorsal surface of the cells with 1 min, become prominent by 3 min, and almost disappear by 7 min. Microvilli, conversely, disappear as the dorsal ruffles become prominent. Ruffles are seen at the the periphery of cell at 3 min, are prominent on most of the cells by 7 min and are gone by 15 min. The surface remains smooth from 15 min until 45 min when large blebs appear. The large blebs are present on most cells at 2 h and are gone by 4 h. The surface remains relatively smooth until 6-7 h of NGF treatment, when microvilli reappear as small knobs. These microvilli increase in both number and length to cover the cell surface by 10 h. These changes were not observed with other basic proteins, with alpha-bungarotoxin (which binds specifically to PC12 membranes), and were not affected by an RNA synthesis inhibitor that blocks initiation of neurite outgrowth. Changes in the cell surface architecture appear to be among the earlist NGF responses yet detected and may represent or reflect primary events in the mechanism of the factor's action.

Scanning electron microscopy of the respiratory surfaces of Saccobranchus (= Heteropneustes) fossilis (Bloch).

The gill secondary lamellae are generally covered with epithelial cells whose outer surfaces form numerous microvilli. The surface of the primary lamellae is characterised by microridges. A particular type of surface sculpturing seems to be associated with given cell boundaries. Further evidence for the derivation of the air tube and fans which guard its entrance by modification of the basic gill structure has been obtained from both the gross surface architecture and microstructure of the individual cell surfaces. Secondary lamellae are represented by stubby projections which generally have a biserial arrangement. The outer surfaces of the epithelia overlying the capillaries of these respiratory islets are coated with microvilli as in the secondary lamellae. On the other hand, the relatively smooth-surfaced 'lanes' between groups of respiratory islets have a microridged surface similar to that of the primary gill lamellae. It is suggested that previous estimates of surface area, and consequently diffusing capacities of the air-breathing organ, have been low in view of the increased surface, due to both their gross and microstructure. Estimates for gill surface area may need very little correction as the spaces between the microvilli and microridges are probably filled with mucus under normal conditions.

An analysis of con A-mediated agglutination in a chinese hamster ovary subclone which responds morphologically to growth in dibutyryl cyclic amp: III. The role of microvilli in the agglutination process

We have used the H-7w subclone of a Chinese hamster ovary cell line (K1) to investigate the role of cell surface architecture (specifically microvilli, blebs, and sheets) in determining the relative agglutinability of a cell line with Con A. Our evidence clearly demonstrates that no specific, immediately recognizable surface architecture is associated with the agglutinable or non-agglutinable phenotype. Our data suggest that the expression of microvilli on the cell surface is neither necessary to nor sufficient for the phenotype described by enhanced agglutinability with Con A. Furthermore our work demonstrates that cells covered with blebs are as agglutinable as cells covered with microvilli thereby suggesting that the intertwining of microvilli may not be an essential facet of the agglutination phenomenon.

TRANSFORMATION‐SENSITIVE CELL SURFACE PROTEIN: ISOLATION, CHARACTERIZATION, AND ROLE IN CELLULAR MORPHOLOGY AND ADHESION

Cell surface protein is the major cell surface glycoprotein of chick embryo fibroblasts. We have isolated and purified this glycoprotein and find that it is an adhesive protein that increases cell-cell and cell-substratum adhesiveness in a variety cellular adhesion assays. Transformation of chick fibroblasts results in decreased quantities of CSP due primarily to a fivefold reduction in CSP biosynthesis, although increased proteolytic degradation and shedding from the cell surface also contribute. The decreased biosynthesis is apparently due to a fivefold reduction in translatable mRNA for CSP. Reconstitution of isolated purified CSP on 14 transformed cell lines from several species results in reversion to a more normal fibroblastic morphology, adhesiveness, cell surface architecture, microfilament bundle organization, motility, and alignment at confluence. Cell surface protein does not restore growth control. The effects of CSP appear to be due to at least two actions, increased cell-substratum adhesion plus altered cell-cell interactions. Untransformed chick cells treated with affinity-purified antibodies to CSP develop the rounded morphology characteristic of many transformed cells that are deficient in CSP (LETS protein). Cell surface protein is found primarily in fibrillar aggregates on the cell surface. These CSP fibrils are relatively immobile and do not affect the mobility of other cell surface components. However, CSP can be eventually redistributed to caplike structures with anti-CSP. Isolated CSP consists of highly asymmetric disulfide-linked dimers and multimers. The interchain disulfide bridges are confined to a short terminal fragment that is readily removed by trypsin. Cell surface protein and cold-insoluble globulin have similar compositions but differ in solubility and amino termini. Cell surface protein contains primarily asparagine-linked oligosaccharides that appear to be responsible for CSP's concanavalin A receptor activity. Inhibition of CSP's glycosylation by treatment with tunicamycin results in decreased CSP due to marked increases in its degradation rate, without inhibition of synthesis or secretion. Studies of this major cell surface glycoprotein have provided insight into the biochemical mechanisms of cellular adhesion, morphology, and social interaction and provide an approach to analyze the dynamics and regulation of protein synthesis, glycosylation, secretion, and turnover.

Contact-inhibited revertant cell lines isolated from SV40-transformed cells VIII. Membrane protein and glycoprotein composition

The composition of enriched surface membrane preparations from BALB/c 3T3, SV40-transformed 3T3 (SVT2), and concanavalin A-selected revertant variant cells were investigated by polyacrylamide slab gel electrophoresis in the presence of sodium dodecyl sulfate. The electrophoretic profiles of the membrane proteins from the three cell lines were remarkably similar. However, [ 14C]leucine-radiolabeled or Coomassie blue-stained preparations from revertant cells lacked a protein band of apparent molecular weight 110,000 (band 15), which was present in 3T3 and SVT2 membranes and which was replaced in revertant membranes by two new bands (15a and 15b) migrating on either side of band 15. Lactoperoxidase-catalyzed iodination studies indicated that the relative proportion of band 5 (a LETS-like glycoprotein) was decreased, and band 24 (MW, 65,000) was increased in 3T3 cells compared to the other cell lines. Revertant membranes had increased proportions of sialic acid residues on several glycoproteins. The LETS glycoprotein from all three cell types was found to be sialylated. Two different compartmentalized pools of cell surface-associated LETS have been identified. These results are discussed in relation to the method of selection of revertant cells and to the possible changes induced in cell surface architecture by transformation and reversion.

Differences in agglutinability of adult and fetal human fibroblasts using phytohemagglutinin

Whereas Concanavalin A (Con A) and Wheat Germ Agglutinin (WGA) detect differences in the agglutinability of transformed, established and secondary cultures, Phytohemagglutinin (PHA) detects differences between cultured adult and fetal human fibroblasts. Adult cells agglutinate with PHA to the same extent as transformed cells, whereas fetal cells show significant agglutination only after trypsinization. Differences in cell size, growth rate, surface architecture or binding of fluorescent PHA could not be demonstrated between adult and fetal cells. Although the basis for this apparent difference in agglutinability remains unknown, it is the first demonstration that fetal cells (even after prolonged in vitro culture) retain at least some surface properties not shared by adult or transformed cells.

Isolation of a lectin from liver plasmamembrane its binding to cellular membrane receptors in vitro

1. Introduction Recently, much interest has been stimulated by the fact that the surface membrane of mammalian cells possess receptor sites for the binding of lectins from plant seeds [ 1,2] . Such data has led to com- pletely new perspectives regarding cell surface architecture and membrane fluidity of normal and virus transformed cells [3,4]. In addition, the binding of plant lectins to various types of animal cells has been shown to induce changes in many diverse physiological phenomena, such as mitosis

Some approaches to mapping the redistribution of parent-cell H-2 antigens on the membrane of cell hybrids

Cell hybrids were created by Sendai virus-induced fusion of A 9 (H- 2 k) cells and peritoneal exudate cells (presumably macrophages, H- 2 d), and the distribution of parent-derived H- 2 specificities was followed using two indirect methods. Competition within pairs of non-crossreacting H- 2 specificities for the attachment of specific antibodies was measured by the blocking index. In cell hybrids, no blocking was detectable within the pairs H- 2·4–H- 2·11 and H- 2·4–H- 2·5 where the members of each pair were derived from different parents. When, however, both specificities of a pair (H- 2·5, H- 2·11) were from the same parent, their blocking relationship was similar as in the particular parent. In different experiments and with different types of cells, the presence of blocking between two H- 2 specificities was found to be confined to the “double cis” configuration of their genetic determinants. No blocking was found in “single trans”, “single cis”, and “double trans” configurations. The difference in blocking could not be simply traced back to a different cell surface density of the antigens under test. Trans specificities that do not compete (H- 2·4, H- 2·11) were shown to cooperate in a complement-dependent reaction with a specific antibody. The competition clearly indicates an intermixing of parent derived H- 2 antigens on the cell hybrid membrane. The possible bearing of the data on the understanding of the control of the specificity of the cell-surface architecture is discussed.

Alterations of Cell Surfaces as a Pathogenetic Factor in Psoriasis

The plasma membrane, surface and glycoprotein-rich surface coat of psoriatic keratinocytes are significantly altered, showing: (1) microvillous transformation of plasma membranes, (2) decreased adhesiveness of extradesmosomal membrane regions, (3) nearly complete absence of glycoprotein-rich surface coats in the malpighian layer and thickening of these coats in the horny layer, and (4) incomplete marginal bands and irregular spatial architecture of horny cell surfaces. These findings indicate decreased capability of psoriatic keratinocytes to participate in an intracellular adhesive system, and suggest loss of "contact inhibition of growth" in psoriatic epidermis in vivo, as known from in vitro models. This loss may be responsible for the increased mitotic rate in psoriasis. Changes are limited to the involved skin in psoriasis and show good response to treatment with corticosteroids locally or methotrexate systemically.