Characterization and localization of myosin in the brush border of intestinal epithelial cells

The brush border of intestinal epithelial cells consists of a tightly packed array of microvilli, each of which contains a core of actin filaments. It has been postulated that microvillar movements are mediated by myosin interactions in the terminal web with the basal ends of these actin cores (Mooseker, M.S. 1976. J. Cell. Biol. 71:417-433). We report here that two predictions of this model are correct: (a) The brush border contains myosin, and (b) myosin is located in the terminal web. Myosin is isolated in 70 percent purity by solubilization of Triton-treated brush borders in 0.6 M KI, and separation of the components by gel filtration. Most of the remaining contaminants can be removed by precipitation of the myosin at low ionic strength. This yield is approximately 1 mg of myosin/30 mg of solubilized brush border protein. The molecule consists of three subunits with molecular weights of 200,000, 19,000, and 17,000 daltons in a 1:1:1 M ratio. At low ionic strength, the myosin forms small, bipolar filaments with dimensions of 300 X 11nm, that are similar to filaments seen previously in the terminal web of isolated brush borders. Like that of other vertebrate, nonmuscle myosins, the ATPase activity of isolated brush border myosin in 0.6 M KCI is highest with EDTA (1 μmol P(i)/mg-min; 37 degrees C), intermediate with Ca++ (0.4 μmol P(i)/mg-min), and low with Mg++ (0.01 μmol P(i)/mg-min). Actin does not stimulate the Mg-ATPase activity of the isolated enzyme. Antibodies against the rod fragment of human platelet myosin cross-react by immunodiffusion with brush border myosin. Staining of isolated mouse or chicken brush borders with rhodamine-antimyosin demonstrates that myosin is localized exclusively in the terminal web.     

Cytoskeleton functions in membrane traffic in polarized epithelial cells.

The complexity of membrane traffic in polarized epithelial cells between the Golgi complex and either the apical or basal-lateral membrane domain, and between different membrane domains (transcytosis) requires that vesicles leaving one membrane compartment efficiently and rapidly reach their (correct) destination. There is increasing evidence that microtubules, actin microfilaments and the membrane-cytoskeleton are involved in several aspects of vesicle transport and in the regulation of protein distributions in polarized epithelial cells. These possible functions are discussed in the context of the development and maintenance of cell polarity.     

Trafficking of Shigella lipopolysaccharide in polarized intestinal epithelial cells.

Bacterial lipopolysaccharide (LPS) at the apical surface of polarized intestinal epithelial cells was previously shown to be transported from the apical to the basolateral pole of the epithelium (Beatty, W.L., and P.J. Sansonetti. 1997. Infect. Immun. 65:4395-4404). The present study was designed to elucidate the transcytotic pathway of LPS and to characterize the endocytic compartments involved in this process. Confocal and electron microscopic analyses revealed that LPS internalized at the apical surface became rapidly distributed within endosomal compartments accessible to basolaterally internalized transferrin. This compartment largely excluded fluid-phase markers added at either pole. Access to the basolateral side of the epithelium subsequent to trafficking to basolateral endosomes occurred via exocytosis into the paracellular space beneath the intercellular tight junctions. LPS appeared to exploit other endocytic routes with much of the internalized LPS recycled to the original apical membrane. In addition, analysis of LPS in association with markers of the endocytic network revealed that some LPS was sent to late endosomal and lysosomal compartments.     

MYO1A (brush border myosin I) dynamics in the brush border of LLC-PK1-CL4 cells

The kidney epithelial cell line, LLC-PK1-CL4 (CL4), forms a well ordered brush border (BB) on its apical surface. CL4 cells were used to examine the dynamics of MYO1A (M1A; formerly BB myosin I) within the BB using GFP-tagged MIA (GFP-M1A), MIA motor domain (GFP-MDIQ), and tail domain (GFP-Tail). GFP-beta-actin (GFP-Actin) was used to assess actin dynamics within the BB. GFP-M1A, GFP-Tail, but not GFP-MDIQ localized to the BB, indicating that the tail is sufficient for apical targeting of M1A. GFP-Actin targeted to all the actin domains of the cell including the BB. Fluorescence recovery after photobleaching analysis revealed that GFP-M1A and GFP-Tail turnover in the BB is rapid, approximately 80% complete in <1 min. As expected for an actin-based motor, ATP depletion resulted in significant inhibition of GFP-M1A turnover yet had little effect on GFP-Tail exchange. Rapid turnover of GFP-M1A and GFP-Tail was not due to actin turnover as GFP-Actin turnover in the BB was much slower. These results indicate that the BB population of M1A turns over rapidly, while its head and tail domains interact transiently with the core actin and plasma membrane, respectively. This rapidly exchanging pool of M1A envelops an actin core bundle that, by comparison, is static in structure.     

Involvement of RhoA, ROCK I and myosin II in inverted orientation of epithelial polarity

In multicellular epithelial tissues, the orientation of polarity of each cell must be coordinated. Previously, we reported that for Madin-Darby canine kidney cells in three-dimensional collagen gel culture, blockade of beta1-integrin by the AIIB2 antibody or expression of dominant-negative Rac1N17 led to an inversion of polarity, such that the apical surfaces of the cells were misorientated towards the extracellular matrix. Here, we show that this process results from the activation of RhoA. Knockdown of RhoA by short hairpin RNA reverses the inverted orientation of polarity, resulting in normal cysts. Inhibition of RhoA downstream effectors, Rho kinase (ROCK I) and myosin II, has similar effects. We conclude that the RhoA-ROCK I-myosin II pathway controls the inversion of orientation of epithelial polarity caused by AIIB2 or Rac1N17. These results might be relevant to the hyperactivation of RhoA and disruption of normal polarity frequently observed in human epithelial cancers.     

Suppression of villin expression by antisense RNA impairs brush border assembly in polarized epithelial intestinal cells.

We have used an antisense RNA strategy to investigate the role of the actin-associated protein, villin, in the brush-border morphogenesis of human intestinal CaCO2 cells. Stable expression of a cDNA encoding antisense villin RNA resulted in the permanent down-regulation of the endogenous villin message and dramatically affected brush-border assembly. Ultrastructural and immunolocalization studies revealed that epithelial cell polarity was largely maintained. However, in contrast to brush-border markers such as dipeptidyl-peptidase IV, the apical localization of sucrase-isomaltase was specifically impaired. Retransfection of the villin antisense-expressing cell line with a cDNA encoding a partial sense villin RNA restored both brush-border assembly and sucrase-isomaltase apical expression. The suggestion that brush-border morphogenesis may be important for the trafficking of certain proteins is discussed.     

Organization of actin, myosin, and intermediate filaments in the brush border of intestinal epithelial cells

Terminal webs prepared from mouse intestinal epithelial cells were examined by the quick-freeze, deep-etch, and rotary-replication method. The microvilli of these cells contain actin filaments that extend into the terminal web in compact bundles. Within the terminal web these bundles remain compact; few filaments are separated from the bundles and fewer still bend towards the lateral margins of the cell. Decoration with subfragment 1 (S1) of myosin confirmed that relatively few actin filaments travel horizontally in the web. Instead, between actin bundles there are complicated networks of the fibrils. Here we present two lines of evidence which suggest that myosin is one of the major cross-linkers in the terminal web. First, when brush borders are exposed to 1 mM ATP in 0.3 M KCl, they lose their normal ability to bind antimyosin antibodies as judged by immunofluorescence, and they lose the thin fibrils normally found in deep-etch replicas. Correspondingly, myosin is released into the supernatant as judged by SDS gel electrophoresis. Second, electron microscope immunocytochemistry with antimyosin antibodies followed by ferritin- conjugated second antibodies leads to ferritin deposition mainly on the fibrils at the basal part of rootlets. Deep-etching also reveals that the actin filament bundles are connected to intermediate filaments by another population of cross-linkers that are not extracted by ATP in 0.3 M KCl. From these results we conclude that myosin in the intestinal cell may not only be involved in a short range sliding-filament type of motility, but may also play a purely structural role as a long range cross-linker between microvillar rootlets.     

Unique isoactins in the brush border of rat intestinal epithelial cells

The mammalian genome contains 20-30 genes encoding a family of actins. To date, however, only six proteins (four muscle and two nonmuscle isoforms) encoded by this multigene complex have been identified. We have isolated two actins from the brush border of rat intestinal epithelial cells that have isoelectric points and N-terminal peptides characteristic of the cytoplasmic beta- and gamma-actins. However, using a panel of actin-specific monoclonal antibodies, we show that these actins contain a set of epitopes that distinguishes them from any of the known cytoplasmic or muscle isoforms. These unique actins share features of both the nonmuscle and muscle isoforms, suggesting that they represent an intermediate in the evolution of the specialized muscle actins.     

Characterization of intestinal microvillar membrane disks: detergent- resistant membrane sheets enriched in associated brush border myosin I (110K-calmodulin)

The actin bundle within each microvillus of the intestinal brush border (BB) is tethered laterally to the membrane by bridges composed of BB myosin I. Avian BB myosin I, formerly termed 110K-calmodulin, consists of a heavy chain with an apparent Mr of 110 kD and three to four molecules of calmodulin "light chains." Recent studies have shown that this complex shares many properties with myosin including mechanochemical activity. In this report, the isolation and characterization of a membrane fraction enriched in bound BB myosin I is described. This membrane fraction, termed microvillar membrane disks, was purified from ATP extracts of nonionic detergent-treated microvilli prepared from avian intestinal BBs. Ultrastructural analysis revealed that these membranes are flat, disk-shaped sheets with protrusions which are identical in morphology to purified BB myosin I. The disks exhibit actin-activated Mg-ATPase activity and bind and cross-link actin filaments in an ATP-dependent fashion. The mechanochemical activity of the membrane disks was assessed using the Nitella bead movement assay (Sheetz, M. P., and J. A. Spudich. 1983. Nature [Lond.]. 303:31-35). These preparations were shown to be free of significant contamination by conventional BB myosin. Latex beads coated with microvillar membrane disks move in a myosin-like fashion along Nitella actin cables at rates of 12-60 nm/s (average rate of 33 nm/s); unlike purified BB myosin I, the movement of membrane disk-coated beads was most reproducibly observed in buffers containing low Ca2+.     

Distribution of actin and myosin in muscle and non-muscle cells

Summary Specific anti-actin and anti-myosin antibodies were shown to react in single and double immunofluorescence sandwich tests with identical sites in non-muscle cells in frozen sections of tissues and in cultured cells. In tissues, both antibodies reacted with liver cell membranes, parts of renal glomeruli, brush borders and peritubular fibrils of renal tubules, brain synaptic junctions, and membranes of lymphoid cells in thymic medulla, lymph nodes and spleen. Both antibodies reacted strongly with long parallel cytoplasmic fibrils in cultured fibroblasts, and with disrupted fibrils in cytochalasin-B treated cells. In neuroblastoma cells both antibodies gave prominent staining of growth cones and microspikes. The observation that the distribution of myosin parallels that of actin in non-muscle cells argues strongly in favour of a functional interaction between the two molecules in the generation of contractile activity in nonmuscle cells.The authors thank Dr. M. Owen, National Institute of Medical Research, Mill Hill, for the gift of rabbit anti-actin antibodyOn sabbatical leave from Monash University, and supported by a Commonwealth Medical FellowshipThe Brompton Hospital, London     

Effect of monensin on ricin and fluid phase transport in polarized MDCK cells.

The effect of monensin on endocytosis, transcytosis, recycling and transport to the Golgi apparatus in filter-grown Madin-Darby canine kidney (MDCK) cells was investigated using 125I-labeled ricin as a marker for membrane transport, and horseradish peroxidase (HRP) as a marker for fluid phase transport. Monensin (10 microM) stimulated transcytosis of both markers about 3-fold in the basolateral to apical direction. Transcytosis of HRP in the opposite direction, apical to basolateral, was reduced to approximately 50% of the control by monensin, whereas that of ricin was slightly increased. Recycling of markers endocytosed from the apical surface was reduced in the presence of monensin and there was an increased accumulation of both ricin and HRP in the cells. Transport of ricin to the Golgi apparatus increased to the same extent as the increase in intracellular accumulation. No change in recycling or accumulation was observed with monensin when the markers were added basolaterally, but transport of ricin to the Golgi apparatus increased almost 3-fold. Our results indicate that basolateral to apical transcytosis is increased in the absence of low endosomal pH, and they suggest that apical to basolateral transcytosis of a membrane-bound marker (ricin) is affected by monensin differently from that of a fluid phase marker (HRP).     

Localization of brush border cytoskeletal proteins in gastric oxynticopeptic cells from the bullfrog Rana catesbeiana

The contribution of brush border cytoskeletal proteins (actin, villin, fimbrin, and brush border myosin-1) to organization of the cytoskeletal network underlying apical plications of oxynticopeptic cells was examined by immunohistochemical techniques in frozen sections of gastric mucosa from the bullfrog, Rana catesbeiana. Apical localization of F-actin with phalloidin in oxynticopeptic cells inhibited with cimetidine revealed small, punctate domains within the apical cytoplasm that were consistent with the presence of short microvilli revealed by electron microscopy. Localization of F-actin in cells stimulated with forskolin was limited to a wide continuous band of cytoplasm corresponding to the location of numerous long surface folds. Inhibition of protein synthesis with cycloheximide did not prevent acid secretion or formation of actin filaments within surface folds in stimulated oxynticopeptic cells, suggesting that the formation of filaments does not require actin synthesis. Staining of gastric mucosae with fluorescent DNase-1 demonstrated that oxynticopeptic cells possess an unusually large pool of non-filamentous actin. Taken together, these results suggest that actin-filament formation in stimulated cells occurs by polymerization of an existing pool of non-filamentous actin. Localization of antibodies specific for villin and fimbrin revealed that these proteins were present within intestinal absorptive cells and gastric surface and neck cells but were not present within inhibited or stimulated oxynticopeptic cells. Brush border myosin-1, present in intestinal absorptive cells, was not present in gastric epithelium. Thus, we propose that actin-containing projections in oxynticopeptic cells are not organized like intestinal microvilli and that filament formation occurs after stimulation by modulating intracellular pools of filamentous and non-filamentous actin.     

Functional coupling to brush border creatine kinase imparts a selective energetic advantage to contractile ring myosin in intestinal epithelial cells.

The B-CK isozyme of cytoplasmic creatine kinase is localized distinctly in the terminal web region of the intestinal epithelial cell brush border (Keller and Gordon: Cell Motil. Cytoskeleton 19:169-179, 1991). Experiments were performed to determine whether this CK is energetically coupled to the myosin II that is present in the circumferential ring and interrootlet structural domains of the brush border terminal web. In isolated brush borders, ATP-dependent circumferential ring contraction and interrootlet myosin solubilization were supported either by an exogenous PEP-pyruvate kinase-based ATP-regeneration system (PEP-PK) or by the addition of phosphocreatine to the endogenous B-CK-based ATP-regeneration system (PCr-B-CK). Addition of an exogenous hexokinase-glucose ATP-hydrolysis system (HK-G) effectively blocked both contraction and myosin solubilization in the PEP-PK assay. In contrast, HK-G had no significant effect on PCr-B-CK-supported brush border contraction, although it did inhibit interrootlet myosin solubilization. Thus, when high-energy phosphate is supplied as phosphocreatine, brush border B-CK imparts to the circumferential ring myosin a selective energetic advantage over other ATPases. These results suggest that myosin and B-CK are functionally coupled in the brush border circumferential ring, where they might comprise one end of an energy circuit that supplies energy for contraction, but that colocalization of CK with myosin in the brush border interrootlet domain is insufficient to establish functional coupling.     

Calmodulin dissociation regulates brush border myosin I (110-kD- calmodulin) mechanochemical activity in vitro

110-kD-calmodulin, when immobilized on nitrocellulose-coated coverslips, translocates actin filaments at a maximal rate of 0.07-0.1 micron/s at 37 degrees C. Actin activates MgATPase activity greater than 40-fold, with a Km of 40 microM and Vmax of 0.86 s-1 (323 nmol/min/mg). The rate of motility mediated by 110-kD-calmodulin is dependent on temperature and concentration of ATP, but independent of time, actin filament length, amount of enzyme, or ionic strength. Tropomyosin inhibits actin binding by 110-kD-calmodulin in MgATP and inhibits motility. Micromolar calcium slightly increases the rate of motility and increases the actin-activated MgATP hydrolysis of the intact complex. In 0.1 mM or higher calcium, motility ceases and actin-dependent MgATPase activity remains at a low rate not activated by increasing actin concentration. Correlated with these inhibitions of activity, a subset of calmodulin is dissociated from the complex. To determine if calmodulin loss is the cause of calcium inhibition, we assayed the ability of calmodulin to rescue the calcium-inactivated enzyme. Readdition of calmodulin to the nitrocellulose-bound, calcium-inactivated enzyme completely restores motility. Addition of calmodulin also restores actin activation to MgATPase activity in high calcium, but does not affect the activity of the enzyme in EGTA. These results demonstrate that in vitro 110-kD-calmodulin functions as a calcium-sensitive mechanoenzyme, a vertebrate myosin I. The properties of this enzyme suggest that despite unique structure and regulation, myosins I and II share a molecular mechanism of motility.     

Differences in neutral amino acid and glucose transport between brush border and basolateral plasma membrane of intestinal epithelial cells.

A comparison of L-valine and D-glucose transport was carried out with vesicles of plasma membrane isolated either from the luminal (brush border) or from the contra-luminal (basolateral) region of small intestinal epithelial cells. The existence of transport systems for both non-electrolytes was demonstrated by stereospecificity and saturability of uptake, as well as tracer coupling. Transport of L-valine and D-glucose differs markedly in the two types of plasma membrane with respect to stimulation by Na+. The presence of Na+ stimulated initial L-valine and D-glucose uptake in brush border, but not in basolateral membrane. Moreover, an electro-chemical Na+ gradient, oriented with the lower potential on the inside, supported accumulation of the non-electrolytes above medium concentration only in the brush border membrane. L-Valine and D-glucose transport also were saturated at lower concentrations in brush border (10-20 mM) than in basolateral plasma membranes (30-50 mM). A third difference between the two membranes was found in the effectiveness of known inhibitors of D-glucose transport. In brush border membranes phlorizin was more potent than phloretin and 2', 3', 4'-trihydroxy-4-methoxy chalcone and cytochalasin B did not inhibit at all. In contrast, with the basolateral plasma membranes the order of potency was changed to phloretin = 2',3',4'-trihydroxy-4-methoxy chalcone greater than cytochalasin B greater than phlorizin. These results indicate the presence of different types of transport systems for monosaccharides and neutral amino acids in the luminal and contra-luminal region of the plasma membrane. Active transepithelial transport can be explained on the basis of the different properties of the non-electrolyte transport systems in the two cellular regions and an electro-chemical Na+ gradient that is dependent on cellular metabolism.