Role of the Cldn6 cytoplasmic tail domain in membrane targeting and epidermal differentiation in vivo

It is widely recognized that the claudin (Cldn) family of four tetraspan transmembrane proteins is crucial for tight junction assembly and permeability barrier function; however, the precise role of the tail and loop domains in Cldn function is not understood. We hypothesized that the cytoplasmic tail domain of Cldn6 is crucial for membrane targeting and hence epidermal permeability barrier (EPB) formation. To test this hypothesis via a structure-function approach, we generated a tail deletion of Cldn6 (CDelta187) and evaluated its role in epidermal differentiation and EPB formation through its forced expression via the involucrin (Inv) promoter in the suprabasal compartment of the transgenic mouse epidermis. Even though a functional barrier formed, Inv-CDelta187 mice displayed histological and biochemical abnormalities in the epidermal differentiation program and stimulation of epidermal cell proliferation in both the basal and suprabasal compartments of the interfolliclar epidermis, leading to a thickening of the epidermis after 1 week of age that persisted throughout life. Although some membrane localization was evident, our studies also revealed a significant amount of not only Cldn6 but also Cldn10, Cldn11, and Cldn18 in the cytoplasm of transgenic epidermal cells as well as the activation of a protein-unfolding pathway. These findings demonstrate that the overexpression of a tail truncation mutant of Cldn6 mislocalizes Cldn6 and other Cldn proteins to the cytoplasm and triggers a postnatal increase in proliferation and aberrant differentiation of the epidermis, emphasizing the importance of the Cldn tail domain in membrane targeting and function in vivo.     

Effects of polyelectrolyte chain stiffness, charge mobility, and charge sequences on binding to proteins and micelles

The binding affinities of polyanions for bovine serum albumin in NaCl solutions from I = 0.01-0.6 M, were evaluated on the basis of the pH at the point of incipient binding, converting each such pH(c) value into a critical protein charge Zc. Analogous values of critical charge for mixed micelles were obtained as the cationic surfactant mole fraction Yc. The data were well fitted as Yc or Zc = KI a, and values of K and a were considered as a function of normalized polymer charge densities (tau), charge mobility, and chain stiffness. Binding increased with chain flexibility and charge mobility, as expected from simulations and theory. Complex effects of tau were related to intrapolyanion repulsions within micelle-bound loops (seen in the simulations) or negative protein domain-polyanion repulsions. The linearity of Zc with radicalI at I < 0.3 M was explained by using protein electrostatic images, showing that Zc at I < 0.3 M depends on a single positive "patch"; the appearance of multiple positive domains I > 0.3 M (lower pH(c)) disrupts this simple behavior.     

Isolation of monoclonal antibodies recognizing rat bone-associated molecules in vitro and in vivo.

Knowledge of the number and kinds of differentiation steps that characterize cells of the osteoblast lineage is inadequate. To further analyze osteoblast differentiation, we generated a series of monoclonal antibodies (MAb) to osteogenic cells. Spleen cells from mice immunized with whole-cell populations enriched for expression of osteoblast-associated properties or bone formation in vitro were fused with the SP2/0 myeloma cell line. Supernatants from growing hybridomas were screened by indirect immunofluorescence on frozen sections of a portion of 21-day fetal rat heads that included the calvaria bone, periosteum, muscle, fibrous connective tissue, and skin. Six MAb were selected with bone-associated staining and limited ability to label other tissues. Either cell surface or cytoplasmic molecules were recognized by five of the MAb; one recognized a molecule detectable both in the cytoplasm, on the cell surface, and in the extracellular matrix. Of the antibodies selected, one identified both preosteoblasts and osteoblasts and has been found to be against alkaline phosphatase. The others recognized the mature osteoblasts, osteocytes, and chondrocytic cells. The pattern and distribution of the labeling in vivo extended to primary cells and cell lines in vivo. These results support earlier observations on molecules differentially expressed by cells at different stages of the osteoblast lineage and extend the available cell surface and cytoplasmic epitopes identifiable as marker molecules.     

Positive and negative immunoselection for enrichment of two classes of osteoprogenitor cells.

The number of identifiable stages and expression of differentiation markers in cells of the osteoblast lineage are not well understood. In the present study, a mAb, designated rat bone marrow (RBM) 211.13, was prepared that stained selectively the osteogenic and preosteoblastic cells along the surfaces of bone in calvariae, femurs, and metatarsals. The staining was cell surface associated and coincided with that for alkaline phosphatase (APase) detected histochemically. Only cells positive for APase activity by biochemical assay and not those without APase activity (e.g., fetal rat skin) stained with RBM 211.13. By immunoblotting, RBM 211.13 recognized a band coinciding with APase activity on nonreducing/nondenaturing gels, and RBM 211.13 precipitated a protein which on reduced gels migrated with an apparent molecular mass of approximately 80 kD. RBM 211.13 labeling was abolished by phosphatidylinosital-specific phospholipase C, known to release APase from the cell surface. All of these data support the concept that RBM 211.13 recognizes the bone isoenzyme of APase. RBM 211.13 was used to sort by flow cytometry the APase-positive and APase-negative cells from mixed fetal rat calvaria (RC) cell populations. The osteoprogenitors we identified earlier that form bone nodules in vitro (Bellows, C. G., J. E. Aubin, J. N. M. Heersche, and M. E. Antosz. 1986. Calcif. Tissue Int. 36:143-154; Bellows, C. J., J. N. M. Heersche, and J. E. Aubin. 1990. Dev. Biol. 140:132-138) were found within the APase-positive pool. By immunopanning, RC cells were separated into APase-enriched (APase-positive, adherent) and APase-depleted (APase-negative, nonadherent) populations. The APase-positive fraction was enriched two-to-threefold for bone-forming osteoprogenitors compared to unfractionated cells, while the APase-negative population formed very few nodules under the same conditions. Both populations responded to the glucocorticoid dexamethasone (DEX) with an increase in bone nodule formation. However, the fold stimulation in bone formation in the APase-negative population was approximately 30-fold, while the fold stimulation in the APase-positive population was only approximately 5-fold. These data suggest that APase expression can be used for immunoselection to fractionate osteoblastic populations into an APase-positive population and a population initially APase-negative, that virtually all osteoprogenitors forming bone in vitro in the absence of added glucocorticoids reside in the APase-positive pool, and that the only osteoprogenitors present in the APase-negative pool are those requiring DEX to differentiate.