During terminal differentiation, epithelia become columnar and develop specialized apical membrane structures (microvilli) and functions (regulated endocytosis and exocytosis). Using a clonal intercalated epithelial cell line, we found that high seeding density induced these characteristics, whereas low density seeding maintained a protoepithelial state. When cells were plated at low density, but on the extracellular matrix of high density cells, they converted to the more differentiated phenotype. The extracellular matrix (ECM) protein responsible for this activity was purified and found to be a large 230-kD protein, which we termed hensin. High density seeding caused hensin to be polymerized and deposited in the extracellular matrix, and only this form of hensin was able to induce terminal differentiation. Antibodies to hensin blocked the change in phenotype. However, its purification to homogeneity resulted in loss of activity, suggesting that an additional protein might be necessary for induction of terminal differentiation. Here, we found that a 29-kD protein specifically associates with hensin in the ECM. Addition of purified p29 restored the activity of homogenously purified hensin. Mass fingerprinting identified p29 as galectin 3. Purified recombinant galectin 3 was able to bind to hensin and to polymerize it in vitro. Seeding cells at high density induced secretion of galectin 3 into the ECM where it bundled hensin. Hence, the high density state causes a secretion of a protein that acts on another ECM protein to allow the new complex to signal the cell to change its phenotype. This is a new mechanism of inside-out signaling.

During skeletal development, osteoblasts produce large amounts of extracellular matrix proteins and must therefore increase their secretory machinery to handle the deposition. The accumulation of unfolded protein in the endoplasmic reticulum induces an adoptive mechanism called the unfolded protein response (UPR). We show that one of the most crucial UPR mediators, inositolrequiring protein 1a (IRE1a), and its target transcription factor X-box binding protein 1 (XBP1), are essential for bone morphogenic protein 2-induced osteoblast differentiation. Furthermore, we identify Osterix (Osx, a transcription factor that is indispensible for bone formation) as a target gene of XBP1. The promoter region of the Osx gene encodes two potential binding motifs for XBP1, and we show that XBP1 binds to these regions. Thus, the IRE1a-XBP1 pathway is involved in osteoblast differentiation through promoting Osx transcription.

Intercalated epithelial cells exist in a spectrum of phenotypes; at one extreme, β cells secrete HCO3 by an apical Cl/HCO3 exchanger and a basolateral H+ ATPase. When an immortalized β cell line is seeded at high density it deposits in its extracellular matrix (ECM) a new protein, hensin, which can reverse the polarity of several proteins including the Cl/HCO3 exchanger (an alternately spliced form of band 3) and the proton translocating ATPase. When seeded at low density and allowed to form monolayers these polarized epithelial cells maintain the original distribution of these two proteins. Although these cells synthesize and secrete hensin, it is not retained in the ECM, but rather, hensin is present in a large number of intracellular vesicles. The apical cytoplasm of low density cells is devoid of actin, villin, and cytokeratin19. Scanning electron microscopy shows that these cells have sparse microvilli, whereas high density cells have exuberant apical surface infolding and microvilli. The apical cytoplasm of high density cells contains high levels of actin, cytokeratin19, and villin. The cell shape of these two phenotypes is different with high density cells being tall with a small cross-sectional area, whereas low density cells are low and flat. This columnarization and the remodeling of the apical cytoplasm is hensin-dependent; it can be induced by seeding low density cells on filters conditioned by high density cells and prevented by an antibody to hensin. The changes in cell shape and apical cytoskeleton are reminiscent of the processes that occur in terminal differentiation of the intestine and other epithelia. Hensin is highly expressed in the intestine and prostate (two organs where there is a continuous process of differentiation). The expression of hensin in the less differentiated crypt cells of the intestine and the basal cells of the prostate is similar to that of low density cells; i.e., abundant intracellular vesicles but no localization in the ECM. On the other hand, as in high density cells hensin is located exclusively in the ECM of the terminally differentiated absorptive villus cells and the prostatic luminal cell. These studies suggest that hensin is a critical new molecule in the terminal differentiation of intercalated cell and perhaps other epithelial cells.

Two forms of intercalated cells are present in kidney collecting tubules, the ␣ cell has apical endocytosis, apical H ϩ -ATPase and basolateral band 3, while ␤ cells have reversed polarity of these proteins and no apical endocytosis. When a ␤ cell line was seeded at high density, it changed into the ␣ form. We previously showed that a partially purified 230 kD extracellular matrix protein of high density cells was able to retarget band 3 from apical to basolateral domains and stimulated apical endocytosis in vitro (Van Adelsberg, J., J.C. Edwards, J. Takito, B. Kiss, and Q. Al-Awqati. 1994. Cell. 76:1053-1061). We now purify this protein, which was named hensin, to near homogeneity and find that it belongs to the macrophage scavenger receptor cysteine rich (SRCR) family. An antibody, generated against a fusion protein made from a partial cDNA recognized a 230-kD protein in rabbit kidney and in the intercalated cell line.

No abstract