cAMP protects endothelial barrier functions by preventing Rac-1 inhibition

cAMP enhances endothelial barrier properties and is protective against various inflammatory mediators both in vivo and in vitro. However, the mechanisms whereby cAMP stabilizes the endothelial barrier are largely unknown. Recently we demonstrated that the Rho family GTPase Rac-1 is required for maintenance of endothelial barrier functions in vivo and in vitro. Therefore, in the present study we investigated the effect of forskolin (5 microM)- and rolipram (10 microM)-induced cAMP increase on reduction of barrier functions in response to Rac-1 inhibition by Clostridium sordellii lethal toxin (LT). Forskolin and rolipram treatment blocked LT (200 ng/ml)-induced hydraulic conductivity (Lp) increase in mesenteric microvessels in vivo. Likewise, LT-induced intercellular gap formation in monolayers of cultured microvascular myocardial endothelial (MyEnd) cells and LT-induced loss of adhesion of vascular endothelial cadherin-coated microbeads were abolished. Inhibition of PKA by myristoylated inhibitor peptide (14-22) of PKA (100 microM) reduced the protective effect of cAMP on LT-induced Lp increase in vivo and gap formation in vitro, indicating that the effect of cAMP on Rac-1 inhibition was PKA dependent. Glucosylation assays demonstrated that cAMP prevents inhibitory Rac-1 glucosylation by LT, indicating that one way that cAMP enhances endothelial barrier functions may be by regulating Rac-1 signaling. Our study suggests that cAMP may provide its well-established protective effects at least in part by regulation of Rho proteins.     

Evidence for the concentration of F-actin and myosin in synapses and in the plasmalemmal zone of axons

Fluorescent staining with phalloidin, a specific probe for F-actin, and antibodies to non-muscle myosin from thymus was used to localize actin and myosin in brain neurons of the rat. Phalloidin and anti-myosin displayed a preferential affinity for synaptic formations in the cerebellum, the brain stem, the spinal cord and the retina. The conclusion that F-actin and myosin are concentrated in synaptic terminals was further established by simultaneous staining of isolated rat brain synaptosomes with phalloidin and anti-thymus myosin as well as by the demonstration of a selective affinity of anti-thymus myosin for a 200 000-Mr protein band in gel electrophoretograms of synaptic fractions. Apart from synaptic areas, phalloidin and anti-thymus myosin reacted also, albeit rather weakly, with a narrow circumferential layer located in the area of the plasma membrane of virtually all axons in the white matter and the spinal roots. The spatial coexistence of myosin and actin in brain synapses and axons is of particular interest in view of various dynamic functions that have been proposed for axonal and synaptic actin.     

State-of-the-art technologies, current opinions and developments, and novel findings: news from the field of histochemistry and cell biology

Investigations of cell and tissue structure and function using innovative methods and approaches have again yielded numerous exciting findings in recent months and have added important data to current knowledge, inspiring new ideas and hypotheses in various fields of modern life sciences. Topics and contents of comprehensive expert reviews covering different aspects in methodological advances, cell biology, tissue function and morphology, and novel findings reported in original papers are summarized in the present review.     

Distribution of actin and the actin-associated proteins myosin, tropomyosin, alpha-actinin, vinculin, and villin in rat and bovine exocrine glands

Actin, myosin, and the actin-associated proteins tropomyosin, alpha-actinin, vinculin, and villin were localized in acinar cells of rat and bovine pancreas, parotid, and prostate glands by means of immunofluorescent staining of both frozen tissue sections and semithin sections of quick-frozen, freeze-dried, and plastic-embedded tissues. Antibodies to actin, myosin, tropomyosin, alpha-actinin, and villin reacted strongly with a narrow cytoplasmic band extending beneath the luminal border of acinar cells. The presence of villin, which has so far been demonstrated only in intestinal and kidney brush border, was further confirmed by antibody staining of blotted electrophoresis gels of whole acinar cell extracts. Fluorescently labelled phalloidin, which reacts specifically with F-actin, gave similar staining, within the cell apex to that obtained with antibodies to actin, myosin, tropomyosin, alpha-actinin, and villin. In contrast, immunostaining with antibodies to vinculin was restricted to the area of the junctional complex. Ultrastructurally, the apical immunoreactive band corresponded to a dense web composed of interwoven microfilaments, which could be decorated with heavy meromyosin. Outside this apical terminal web, antibodies to myosin and tropomyosin gave only a weak immunostaining (confined to the lateral cell borders) whereas antibodies to actin and alpha-actinin led to a rather strong bead-like staining along the lateral and basal cell membrane most probably marking microfilament-associated desmosomes. Anti-villin immunofluorescence was confined to the apical terminal web. It is suggested that the apical terminal web is important for the control of transport and access of secretory granules to the luminal plasma membrane and that villin, which is known to bundle or sever actin filaments in a Ca(++)-dependent manner, might participate in the regulation of actin polymerization within this strategically located network of contractile proteins.     

Distribution of myosin and the glial fibrillary acidic protein (GFA Protein) in rat spinal cord and in the human frontal cortex as revealed by immunofluorescence microscopy

Summary The glial fibrillary acidic (GFA) protein and myosin were localized in rat spinal cord and human frontal cortex using specific antibodies against GFA protein from human spinal cord and highly purified smooth myosin from chicken gizzard by means of an indirect immunofluorescence microscopical approach. A strong GFA protein and myosin immunoreactivity was found in astrocytes of the white and grey matter and in the external glial limitans membrane. The very fine branches of astrocytic processes stained with antiGFA protein, but not with anti-myosin. Similar results were obtained with the human frontal cortex, where myosin antibodies failed to reveal the very fine branches of protoplasmic astrocytes.As a whole, staining with the GFA protein antiserum was more crisp than with the myosin antibody.Thanks are due to Professor J.R. Wolff, Max-Planck Institute for Biophysical Chemistry, Göttingen, for stimulating discussions, to Ursula König, Christa Mahlmeister and Renate Steffens for skilful technical assistance, and to Heidi Waluk for the photographic workSupported by grants from Deutsche Forschungsgemeinschaft (Br 634/1, Dr 91/1, Un 34/4, Ste 105/19)Dedicated to Prof. Dr. med. H. Leonhardt on the occasion of his 60. birthday     

Nano-scale dynamic recognition imaging on vascular endothelial cells

Combination of high-resolution atomic force microscope topography imaging with single molecule force spectroscopy provides a unique possibility for the detection of specific molecular recognition events. The identification and localization of specific receptor binding sites on complex heterogeneous biosurfaces such as cells and membranes are of particular interest in this context. Here simultaneous topography and recognition imaging (TREC) was applied to gently fixed microvascular endothelial cells from mouse myocardium (MyEnd) to identify binding sites of vascular endothelial (VE)-cadherin, known to play a crucial role in calcium-dependent, homophilic cell-to-cell adhesion. TREC images were acquired with magnetically oscillating atomic-force microscope tips functionalized with a recombinant VE-cadherin-Fc cis-dimer. The recognition images revealed single molecular binding sites and prominent, irregularly shaped dark spots (domains) with sizes ranging from 10 to 100 nm. These domains arose from a decrease of the oscillation amplitude during specific binding between active VE-cadherin cis-dimers. The VE-cadherin clusters were subsequently assigned to topography features. TREC represents an exquisite method to quickly obtain the local distribution of receptors on cellular surface with an unprecedented lateral resolution of 5 nm.     

Identification of cytoskeletal markers for the different microvilli and cell types of the rat vomeronasal sensory epithelium

The vomeronasal organ (VNO) of the mammal nose is specialized to detect pheromones. The presumed site of the chemosensory signal transduction of pheromones is the vomeronasal brush border of the VNO sensory epithelium, which has been shown to contain two different sets of microvilli: (i) the tall microvilli of supporting cells and (ii) the short microvilli of the chemoreceptive VNO neurons that branch and intermingle with the basal portions of the longer supporting cell microvilli. A key problem when studying the subcellular distribution of possible VNO signal transduction molecules at the light microscope level is the clear discrimination of immunosignals derived from dendritic microvilli of the VNO neurons and surrounding supporting cell structures. In the present study we therefore looked for cytoskeletal marker proteins, that might help to distinguish at the light microscope level between the two sets of microvilli. By immunostaining we found that the VNO dendritic microvilli can be selectively labelled with antibodies to the calcium-sensitive actin filament-bundling protein villin, whereas supporting cell microvilli contain the actin filament cross-linking protein fimbrin, but not villin. Useful cytoplasmic marker molecules for cellular discrimination were cytokeratin 18 for supporting cells and β-tubulin for dendrites of VNO neurons. A further finding was that the non-sensory epithelium of the rat VNO contains brush cells, a cell type that appears to be involved in certain aspects of chemoreception in the gut. Brush cells or other structures of the vomeronasal brush border did not contain α-gustducin.     

Freeze-fracture studies of cytoplasmic inclusions occurring in experimental lipidosis as induced by amphiphilic cationic drugs.

The ultrastructure of cytoplasmic inclusions, which characterize experimental lipidosis as induced by several amphiphilic cationic drugs, was studied by means of freeze-fracturing and thin-sectioning. Retinal and adrenal tissues of rats chronically treated with high oral doses of chlorphentermine were used. In thin sections the cytoplasmic inclusions, which were previously shown to represent lysosomes overloaded with polar lipids, exhibit lamellated or lattice-like internal patterns. The present freeze-fracture observations are interpreted as to indicate that the lamellated inclusions contain polar lipids in the lamellar phase, whereas those with lattice-like patterns contain polar lipids in a hexagonal phase.     

Axonal and cellular alterations in the inner ear of rats treated with chlorphentermine or iprindole

An electron-microscopic study was carried out on the inner ear of rats, which had been treated with the anorectic drug chlorphentermine and the antidepressant drug iprindole, two cationic amphiphilic compounds known to induce a generalized lipidosis. After chronic drug treatment the following vestibular and cochlear alterations were observed: a) numerous lamellated and crystalloid cytoplasmic inclusion bodies in various cell types, typical of drug-induced lipidosis; b) axonal balloonings predominantly affecting preterminal sensory endings which were filled with masses of coarse osmiophilic inclusions and autophagic vacuoles. With prolonged treatment degeneration of nerve fibers below the sensory epithelium was observed in increased numbers. Axonal changes are tentatively interpreted to result from drug-induced interference with certain catabolic processes involved in the normal degradation of axoplasmic constituents.