Terminal Differentiation of Epithelia from Trophectoderm to the Intercalated Cell: The Role of Hensin
Qais Al-Awqati,
Soundarapandian Vijayakumar and
Jiro Takito
Departments of Medicine and Physiology & Cellular Biophysics, College of Physicians & Surgeons of Columbia University, New York, New York.
Correspondence to Dr. Qais Al-Awqati, Departments of Medicine and Physiology & Cellular Biophysics, College of Physicians & Surgeons of Columbia University, 630 W 68th Street, New York, NY 10032; Phone: 212-305-3512; Fax: 212-305-3475;
ABSTRACT. The intercalated cells of the collecting tubules ofmammalian kidneys were discovered by Haggege and Richet to changetheir morphology in response to a variety of physiologic stimulirelated to changes in acid base status. Recent studies showedthat the conversion of to intercalated cell under the influenceof acidification of the medium is due to the deposition of hensinin the extracellular matrix of these cells and activation ofa novel inductive signal transduction pathway. The conversionof to cells is shown to be a process of terminal differentiation.Hensin is secreted as a monomer, and activation of the cellinduces two activities that convert it to a dimer by foldingand into a fiber by bundling of the folded dimers by galectin3. Only the fiber is functional. Hensin is expressed in mostepithelial cells, and its staining pattern suggests that itmight be involved in the terminal differentiation of most epithelia.There is loss of heterozygosity of hensin in a large numberof epithelial and neural tumors, making it likely that it isa tumor suppressor gene. E-mail: qa1@columbia.edu
It has been known since the late 19th century that the distalnephron, really the collecting tubule of the mammalian kidney,contains two cell types, originally called light and dark cells.Dark cells, the minority cells, were dark because they wereenriched in mitochondria. Similar cells were discovered in thefrog skin and urinary bladder of amphibian and reptiles. However,it was only with the work of Haggege and Richet that it becameclear that these cells respond to changes in acid-base statuswith changes in the structure, suggesting that they were responsiblefor the function of acid and base transport. Later it was discoveredthat he dark cells were also enriched in carbonic anhydrase,solidifying the morphologic evidence that these cells were involvedin acid transport. These cells, now called intercalated cells,have a well-established role in H+/HCO3 transport in the corticalcollecting tubule.
In the seminal studies of Haggege and Richet, the use of scanningelectron microscopy revealed that the surface of these cellshad a morphologic specialization that was heterogeneous (1).They found that the apical surface of these cells had eithershort or stubby and rare microvilli or exuberant ruffles andabundant microvilli, depending on the physiologic state of theanimal. In the "resting" control state, there were few microvilli,but when animals were treated with sodium bicarbonate loadingor exposed to a high pCO2, there was a dramatic change in theapical surface. They concluded that these maneuvers led to achange in the phenotype of these cells. Although they were notquite sure whether the dark cells increased because of conversionof some light cells to dark cells, there was no doubt that therewas an increase in the dark cells, especially those with exuberantapical surface.
It became clear from the work of Steinmetz et al. (2) that theturtle urinary bladder was capable of secreting HCO3 as wellas acid, a phenomenon later established in the rabbit corticalcollecting tubule. Steinmetz later found that the intercalatedcells of the urinary bladder had two types of intercalated cellsbased on the surface morphology, one had exuberant microvilliand microplicae, similar to what Haggege and Richet found, whichthey termed intercalated cells. The other type of cell foundwas one that had very few microvilli on the surface, and theycalled those intercalated cells. Examination of the originalfigures of Haggege and Richet shows the presence of these twocell types in the rat collecting tubules, and these findingswere confirmed by the studies of Madsen and Tisher (3).
Whereas rabbit corticol collecting tubule (CCT) usually secretedHCO3, those of other mammals usually secreted acid (i.e., theyabsorbed HCO3), but they could be induced to secrete HCO3 byprolonged treatment with corticosterone and sodium bicarbonate.It was also found that the isolated perfused CCT taken fromrabbits that had become acidotic after acid ingestion now secretedacid (4). We began our studies to identify the mechanism ofthe conversion of these fluxes from HCO3 secretion to acid secretion.
We initially found that some intercalated cells that endocytoseapical material into acid vesicles responded to increases inambient pCO2 by stimulation of exocytosis of the apical vesicles(5,6). Because these vesicles were acidified by a proton translocatingATPase, these studies suggested that the apically endocytosingcells were intercalated cells. This notion was confirmed usingsimultaneous measurements of apical endocytosis and cell pHin the presence and absence of basolateral chloride; we foundthat these cells had basolateral Cl:HCO3 exchange, indicatingthat they secreted acid into the lumen (7). Previous studiesby others demonstrated that cells in the CCT of rabbits boundpeanut lectin on their apical surface (8). Using cell pH measurementsin the presence and absence of luminal chloride, we establishedthat these cells had apical Cl:HCO3 exchange, suggesting thatthey secreted HCO3 into the lumen; i.e., they were -intercalatedcells (7). When rabbits were fed an acid diet, we discoveredthat whereas the total number of intercalated cells remainedthe same, the number of cells with apical endocytosis increasedby a factor of 10 whereas cells with apical peanut lectin capsdecreased by an equal number (7). We concluded that there wasa conversion of to phenotypes of these cells, a process thatwe called plasticity of epithelial polarity.
Does the same cell with apical Cl:HCO3 exchange but no basolateralanion exchange change to another with the oppositely polarizedtransporters? Using cell pH measurements in identified cells,we examined the same -intercalated cells before and after 3h of incubation in acid media and found that apical Cl-/HCO3-exchange was abolished in all of these cells (9). Even moreimportant, this short exposure to acid treatment caused theinsertion of basolateral Cl-/HCO3- exchangers in one third ofthe identified -intercalated cells. The functional consequenceof this -intercalated cell remodeling is to reverse the polarityof HCO3- flux during acid treatment. Indeed, measurement oftransepithelial HCO3 flux demonstrated that acid incubationfor 3 h resulted in conversion of HCO3 secretion to that ofabsorption (9). These effects on individually identified -intercalatedcells required intact RNA and protein synthesis and an intactactin and microtubule cytoskeleton. Acidifying the basolateralmedium increased intracellular calcium in -intercalated cells,which caused exocytosis of H+ ATPase containing vesicles, yetif an increase in intracellular calcium occurred in -intercalatedcells, it did not seem to be responsible for the conversionof to cell types. We found that buffering changes in intracellularcalcium with BAPTA had no effect of the conversion of individuallyidentified -intercalated cells after acid treatment.
What the signal transduction mechanism after acid treatmentis remains to be identified. Alpern et al. (10) recently studiedthe role of tyrosine phosphorylation in the mechanism of regulationof the NaH exchanger (NHE3) in the proximal tubule after acidtreatment. Our present studies suggest that such a mechanismmight well account for the conversion in the CCT as well. However,much remains to be done in identification of the specific pathwayfor this effect.
Studies by others using a variety of methods showed that therewere many types of intercalated cells (reviewed by Schuster(11)). In the most comprehensive of these, Bastani et al. (12),using localization of the proton ATPase, found that there wereas many as six or seven identifiable subtypes of the intercalatedcells. One, the rim cell in their terminology, has strict localizationof the proton ATPase to the apical membrane, whereas anotherhad basolateral localization. There were many intermediate forms.Treatment of the animals with acid or base shifted the populationdensity to one or another extreme. Others using staining forthe ATPase and the anion exchanger AE1 had found other subtypes(11). Although there was much interest in our hypothesis, itcertainly was not considered the standard model, but my readingof the literature did not identify a specific competing hypothesisthat could be excluded with a decisive experiment. We believethat our recent studies on individually identified -intercalatedcells should provide the definitive study that proves our modelof conversion of - to -intercalated cells (7). Many studiesby others attempted to identify the molecular basis of the apicalCl:HCO3 exchanger in the -intercalated cell. We had isolatedpeanut lectin binding membranes from rabbit kidneys and probedthem with specific antibodies against the red cell band 3 anionexchanger AE1 (13). We found this protein in these fractionseven though we and others were unable to find apical stainingfor AE1. Its expression was low; perhaps that was the reasonfor not identifying it by immunocytochemistry. Alternatively,the protein might have been located in intracellular vesiclesthat were purified with the apical membrane. However, otherssimply concluded that we were wrong in this finding and thatthe results, as they claim, were not reproducible. Since then,two other candidate proteins have been implicated, AE4 and anew protein termed pendrin. AE4 belongs to the NBC family ofNaHCO3 cotransporters but was capable of Cl:HCO3 exchange whenexpressed in Xenopus oocytes. Antibodies to this protein stainedthe apical membrane of some -intercalated cells but not others(14). However, other studies showed that this protein mightbe present in the basolateral membrane. Pendrin, the proteinmutated in Pendred’s syndrome (an inherited syndrome ofgoiter and deafness), is located in the apical membranes of-intercalated cells. Expression of pendrin in Xenopus oocytesleads to Cl:I exchange, which is presumably its "natural" behaviorin thyroid epithelial cells, but there have been studies thatdemonstrate that pendrin can also mediate Cl:HCO3 exchange underthe right conditions of HCO3 availability. Mice normally secreteacid in the CCT but are capable of secreting HCO3 after prolongedtreatment with corticosterone and ingestion of NaHCO3. Underthese conditions, deletion of pendrin led inhibition of CCTHCO3 secretion, rather the isolated perfused tubule absorbedHCO3 (15). These results definitively demonstrate that pendrinmediates Cl:HCO3 exchange in the mouse CCT. However, the mutantmice did not develop alkalosis as would have been expected.
Mechanism of Conversion of - to -Intercalated Cells
To identify the molecular mechanism of the plasticity, we generatedan immortalized -intercalated cell line using a temperaturesensitive mutation of the SV40 large T antigen (16). These clonalcells had apical Cl:HCO3 exchange, apical AE1, and basolateralproton ATPase. They also had apical peanut lectin staining butno apical endocytosis. All of these characteristics reproducedthose of authentic -intercalated cells in situ. We discovered,by chance, that when cells were seeded at superconfluent densityand examined a few days later, they secreted acid and had apicalendocytosis, apical ATPase, and basolateral AE1. When seededat subconfluent density and allowed to form a tight epithelium,a few days later, they had the phenotype of HCO3 secretion,apical ClHCO3 exchange, and basolateral ATPase. Clearly, thisclonal cell line exhibited the phenomenon of plasticity of epithelialpolarity that we had been studying in the isolated perfusedCCT (17).
No cell culture model of any process reproduces all of the manifestationsof that phenomenon. Our cell line did not respond to changesin basolateral pH, for instance. Furthermore, the process ofrandom insertion of an immortalizing gene could disrupt someimportant function needed for the process under study. In particular,our cells in culture were composed of pure intercalated cells,a condition that does not exist in vivo, and we do not the roleof principal cells or any interstitial cells in the processof plasticity. Despite all of these caveats, we hoped that whateverpathway produced the density-induced plasticity would intersectwith that induced by acid treatment at some point. Furthermore,it would be impossible to identify the biochemical basis ofsuch a complex phenomenon using isolated perfused tubules composedof only a few hundred intercalated cells among a thousand ormore principal cells. Hence, we forged ahead anyway.
When cells were seeded at high density, they developed apicalendocytosis within a few hours of plating. Cells seeded at lowdensity never developed apical endocytosis even after weeksof observation, at which point their density was similar tothat at high density. Hence, we conclude that high-density seedingwas acting as a molecular switch forcing the cells into a newpathway during the time of seeding. To identify the moleculethat was acting as a switch, we reasoned that it could be asecreted factor induced by some characteristic of high-densityculture. Extensive attempts using media conditioned by high-densitycells failed to induce conversion of low-density cells to thehigh-density phenotype, but the extracellular matrix (ECM) ofhigh-density cells contained the factor that was responsiblefor these effects. We seeded cells at high density on filtersfor 2 d and then treated them with detergents mild enough toremove the cells but keep the ECM as intact as possible. Wethen seeded cells at low density on this high-density ECM. Nowthe cells acquired all of the characteristics of the high-densityphenotype, including apical endocytosis and basolateral AE1(17).
We then prepared 1000 of the filters conditioned by high-densityseeding and developed a miniature apical endocytosis assay.Using these assays, we purified a 250-kD protein that was capableof producing the effect. We termed this protein hensin, forchange in shape in Japanese (18). Antibodies against hensinblocked the development of apical endocytosis in high-densitycells. Hence, we were convinced that hensin was actually themediator of the effect, but purification of hensin to homogeneityrequired harsh treatment by SDS, which resulted in loss of activity.Hence, we could not tell whether an additional protein was required(18).
The cDNA of hensin showed that it was a modular protein composedof several known domains (19). These included eight scavengerreceptor cysteine-rich domains (20), two CUB domains (21), andone Zp domain (22). Hensin was expressed in almost all epitheliatested, with the exception of the proximal nephron segments.Only the collecting tubule expressed it in the kidney. The intestineshad the highest level of expression. In addition, some neuronsand macrophages, including osteoclasts, also expressed it. Fiveother proteins have now been sequenced, all of which were composedof these three domains but in different combinations. Theseother similar proteins were identified by cDNA cloning froma variety of epithelial and nonepithelial tissues and givennames that reflected their method or locus of identification.The first, Ebnerin, was identified in a rat cDNA library obtainedfrom von Ebner’s gland in taste buds (23). CRP-ductinwas found as an overexpressed cDNA in a mouse intestinal cryptcell library (24). DMBT1, a human gene located in chromosome10q25.3, is expressed in brain and was found to be deleted inmany malignant brain tumors (25). Vomeroglandin was identifiedfrom a cDNA library of olfactory neurons (26), whereas gp340was found in lung macrophages (27). We recently discovered thatall of these transcripts were derived from a single gene byalternative splicing (19). Hensin is expressed in all earlyembryonic epithelia, such as trophectoderm and primitive endoderm.The gene is deleted in a large fraction of epithelial (lung,skin, esophagus, stomach, and colon) and neural tumors (glioblastomaand gliomas) (28–30). As is discussed below, we thinkthat the change in polarity is a manifestation of terminal differentiation.Given that interruption of terminal differentiation pathwaysis often invoked as a cause of malignant transformation, thesefindings raise the possibility that hensin or its alternatelyspliced products act as tissue-specific tumor suppressors.
Antibodies against the scavenger receptor cysteine-rich domainshowed that hensin was expressed in low-density cells, whereit was secreted to the basolateral medium but did not localizeto the ECM. Hence, it seemed that the molecular switch was notreally induction of hensin, as we had originally thought; rather,it was its localization to the ECM. The mechanism of hensinlocalization to the ECM turned out to be a complicated processrequiring two additional proteins. Hensin is secreted as a monomerthat is soluble. The high-density state results in the appearanceof a folding enzyme on the surface of the high-density cellsthat converts hensin into dimers and tetramers that are stillsoluble (31). Neither monomer nor soluble oligomers are capableof inducing the conversion of phenotype. A third protein, galectin3, is needed to convert the soluble oligomers to a fiber thatis localized to the ECM. It is only the fiber that is the activemoiety (32).
Hensin Mediates Terminal Differentiation of Epithelial Cells
The most surprising finding of our work in the cell culturemodel is the that the conversion of the low-density phenotype(i.e., cell) to the high-density form represents terminal differentiation.As is described below, the high-density cells had all of thecharacteristics of terminally differentiated epithelia. We thereforeasserted that the low-density cell represents an earlier formof an epithelial cell, the proto-epithelial cell, which is similarin all organs in that it has all of the characteristics of epitheliasuch as tight junctions, polarized membrane proteins and lipids,and transepithelial transport; i.e., they are "generic" epithelia.However, another step is needed to convert these epithelia tothose that are "type-specific," i.e., "brand name," such assmall or large intestine, pancreas, or collecting tubule.
One of the most type-specific characteristics of mature epitheliais the presence of specialized apical membrane structures. Forinstance, some cells have exuberant microvilli that form a "brush-border,"such as the proximal tubule and small intestines. In others,there is a single flagellum, whereas in many others, there arecilia. These apical structures are produced by specialized cytoskeletalproteins largely composed of actin cables stabilized by a varietyof actin-binding proteins such as villin, in the case of brush-bordermicrovilli. These structures are anchored in the subapical regionby a specialized part of the cortical cytoskeleton, which isalso composed of actin but with the inclusion of cytokeratinfilaments of the mature type such as cytokeratin 18 or 19.
All terminally differentiated epithelia have specialized organelles,some of which participate in regulated exocytosis and endocytosis.For instance, storage granules in the pancreas develop onlyafter the cells have terminally differentiated; the embryonicpancreas has no storage granules even though its morphologyis clearly that of an epithelium with tight junction and someevidence of polarity. Hence, the development of regulated exocytosisof storage granules is clear evidence of terminal differentiation.There is now increasing evidence that endocytosis in epitheliais also polarized. There are recycling pathways of apical endocytosisas well as those of basolateral endocytosis. Although theremay be some communications between the two recycling pathwaysin some cells, it is clear that they are separate. Some of theseendocytic pathways are also regulated. For instance, in theprincipal cells of the cortical collecting tubule of the kidney,there is little apical endocytosis in the absence of vasopressin.Vasopressin causes exocytosis of aquaporin 2–containingvesicles with the apical membranes. Removal of vasopressin resultsin apical endocytosis of the aquaporin-containing membranes.Terminally differentiated epithelia have a characteristic cellshape. Some are tall, i.e., they are columnar, whereas othersare cuboidal. In some tissues, epithelia are multilayered, suchas those of transitional epithelia of the urinary bladder orsquamous epithelia of the skin.
Seeding cells at low density on hensin ECM induced the developmentof apical microvilli, the expression and assembly of a subapicalactin network containing cytokeratin 19 and villin (33). Italso induced columnarization of cells doubling the height ofthe cells. Antibodies to hensin prevented the development ofthese characteristics in high-density cells. Furthermore, antibodiesto hensin prevented the conversion of - to -intercalated cellsin response to acidification of the bathing media of isolatedperfused CCTs. This raised the question of whether cells wereterminally differentiated whereas cells were proto-epithelialcells. Examination of several morphologic studies shows thatindeed cells projected into the lumen of the CCT, whereas cells were flat, clearly an indication of cell shape change.The development of apical endocytosis and regulated exocytosis(in response to CO2) in the cells was also consonant with terminaldifferentiation.
Does Hensin Play a Role in the Terminal Differentiation of Other Epithelial Cells?
Some epithelia continue to differentiate terminally during adultlife. These include the cells of the small intestine, skin,and prostate, among others. Using immunocytochemistry, we foundthat hensin was present in intracellular vesicles but not inECM of low-density cells, whereas in high-density intercalatedcells in vitro, it was largely localized in the ECM. Remarkably,hensin was present largely in intracellular organelles in thecrypt cells of the small and large intestine but only in theECM of the villus cells of the small intestine and surface cellsof the colon (31,33). Similarly, it was largely intracellularin the basal cells of the prostate but only in the ECM of theluminal cells. That loss of heterozygosity of hensin/DMBT1 wasfound in many tumors has raised the suspicion that hensin isa tumor suppressor. Many oncologists believe that carcinogenesisis produced by interruption of the pathways for terminal differentiation.Hence, it is possible that hensin might play a significant rolein the terminal differentiation of many, if not all, epithelialcells. We have to await the results of tissue-specific deletionof hensin to test this hypothesis.
Although we were "grateful" to discover the effect of high densityon terminal conversion of phenotype of intercalated cells invitro, we are under no illusion that this is the mechanism thatcauses deposition of hensin in the ECM in vivo. We found thatacid treatment of the isolated rabbit CCT induced depositionof hensin in the ECM of the interconverting intercalated cell.Presumably, hensin is being polymerized with galectin 3 in thesecells. We point out that the -intercalated cell is the onlycell that expresses galectin 3 in the kidney (34). What couldbe the equivalent of high-density seeding in this setting aswell as during conversion of crypt to villus cells and the otherepithelia where hensin might be involved? This process mustbe complex because it would involve not only polymerizationof hensin by the extracellular folding enzyme but also the secretionof galectin 3. One possibility suggested itself from the studyof inside-out signaling in the integrin ECM receptor system.Resting platelets are unable to aggregate because their integrinreceptors have a low affinity for fibrinogen, but after activationby specific ligands such as ADP, the affinity of their receptorsfor fibrinogen increases dramatically and aggregation occurs(35). This affinity modulation apparently results from bindingof signaling proteins to the cytoplasmic domains of the receptorcausing a conformational change in the receptor and is termedinside-out signaling (35). One consequence of the increasedaffinity is a change in the solubility of fibrinogen. Similarly,affinity modulation of 51 integrin induces a conformationalchange in fibronectin that will allow its assembly into fibrils(36). Hence, it is possible that activation of the hensin receptor,perhaps by tyrosine phosphorylation induced by acidosis, couldplay a central role in the polymerization of hensin. This effectmust be in addition to the one that causes the secretion ofgalectin 3 and the folding enzyme. Many receptors for ECM proteinssuch as integrins or receptor tyrosine kinases require thatthey aggregate before signal transduction. The interaction ofthe receptor subunit, in the case of receptor tyrosine kinases,causes autophosphorylation of the receptors and is necessaryfor their activation (37). One speculation that we have is thatthe high-density seeding might simulate receptor aggregationin which a receptor from one cell could touch and interact withthe receptor from another cell, causing spurious activation.Once a receptor is activated, it leads to further aggregationwith other receptors.
Hagege J, Richet G: [Study with scanning electron microscopy of apical surface of kidney distal convoluted tubule cells in rats]. C R Acad Sci Hebd Seances Acad Sci D 271: 331–332, 1970
Steinmetz PR: Cellular organization of urinary acidification. Am J Physiol 251: F173–187, 1986
Madsen KM, Tisher CC: Structural-functional relationships along the distal nephron. Am J Physiol 250: F1–F15, 1986
McKinney TD, Burg MB: Bicarbonate secretion by rabbit cortical collecting tubules in vitro. J Clin Invest 61: 1421–1427, 1978
Gluck S, Cannon C, Al-Awqati Q: Exocytosis regulates urinary acidification in turtle bladder by rapid insertion of H+ pumps into the luminal membrane. Proc Natl Acad Sci U S A 79: 4327–4331, 1982[Abstract/Free Full Text]
Schwartz GJ, Al-Awqati Q: CO2 causes exocytosis of vesicles containing H+ pumps in isolated perfused proximal and collecting tubules. J Clin Invest 75: 1638–1644, 1985
LeHir M, Kaissling B, Koeppen BM, Wade JB: Binding of peanut lectin to specific epithelial cell types in kidney. Am J Physiol 242: C117–C122, 1982
Schwartz GJ, Tsurouka S, Vijayakumar S, Petrovic S, Mian A, Al-Awqati Q: Acid incubation reverses the polarity of intercalated cell transporters: an effect mediated by hensin. J Clin Invest 109: 89–99, 2002[CrossRef][Medline]
Yamaji Y, Tsuganezawa H, Moe OW, Alpern RJ: Intracellular acidosis activates c-Src. Am J Physiol 272(3 Pt 1): C886–C893, 1997
Schuster VL. Function and regulation of collecting duct intercalated cells: Annu Rev Physiol 55: 267–288, 1993[CrossRef][Medline]
Bastani B, Purcell H, Hemken P, Trigg D, Gluck S: Expression and distribution of renal vacuolar proton-translocating adenosine triphosphatase in response to chronic acid and alkali loads in the rat. J Clin Invest 88: 126–136, 1991
van Adelsberg JS, Edwards JC, Al-Awqati Q: The apical Cl/HCO3-exchanger of ’24 intercalated cells. J Biol Chem 268: 11283–11289, 1993[Abstract/Free Full Text]
Tsuganezawa H, Kobayashi K, Iyori M, Araki T, Koizumi A, Watanabe S, Kaneko A, Fukao T, Monkawa T, Yoshida T, Kim DK, Kanai Y, Endou H, Hayashi M, Saruta T: A new member of the HCO3(-) transporter superfamily is an apical anion exchanger of beta-intercalated cells in the kidney. J Biol Chem 276: 8180–8189, 2001[Abstract/Free Full Text]
Royaux IE, Wall SM, Karniski LP, Everett LA, Suzuki K, Knepper MA, Green ED: Pendrin, encoded by the Pendred syndrome gene, resides in the apical region of renal intercalated cells and mediates bicarbonate secretion. Proc Natl Acad Sci U S A 98: 4221–4226, 2001[Abstract/Free Full Text]
Edwards JC, van Adelsberg J, Rater M, Herzlinger D, Lebowitz J, Al-Awqati Q: Conditional immortalization of bicarbonate-secreting intercalated cells from rabbit. Am J Physiol 263: C521–C529, 1992
van Adelsberg J, Edwards JC, Takito J, Kiss B, Al-Awqati Q: An induced extracellular matrix protein reverses the polarity of band 3 in intercalated epithelial cells. Cell 76: 1053–1061, 1994[CrossRef][Medline]
Takito J, Hikita C, Al-Awqati Q: Hensin, a new collecting duct protein involved in the in vitro plasticity of intercalated cell polarity. J Clin Invest 98: 2325–2331, 1996
Takito J, Yan L, Ma J, Hikita C, Vijayakumar S, Warburton D, Al-Awqati Q: Hensin, the polarity reversal protein, is encoded by DMBT1, a gene frequently deleted in malignant gliomas. Am J Physiol 277: F277–F289, 1999
Resnick D, Pearson A, Krieger M: The SRCR Superfamily: A family reminiscent of the Ig superfamily. Trends Biochem Sci 19: 5–8, 1994[CrossRef][Medline]
Bork P, Beckmann G: The CUB domain. A widespread module in developmentally regulated proteins. J Mol Biol 231: 539–545, 1993[CrossRef][Medline]
Bork P, Sander C: A large domain common to sperm receptors (Zp2 and Zp3) and TGF-beta type III receptor. FEBS Lett 300: 237–240, 1992[CrossRef][Medline]
Li XJ, Snyder SH: Molecular cloning of Ebnerin, a von Ebner’s gland protein associated with taste buds. J Biol Chem 270: 17674–17679, 1995[Abstract/Free Full Text]
Cheng H, Bjerknes M, Chen H: CRP-ductin: A gene expressed in intestinal crypts and in pancreatic and hepatic ducts. Anat Rec 244: 327–343, 1996[CrossRef][Medline]
Mollenhauer J, Wiemann S, Schurlen W, Korn B, Hayashi Y, Wilgenbus KK, Deimling AV, Poustka A: DMBT1, a new member of the SRCR superfamily, on chromosome 10q25.3–26.1 is deleted in malignant brain tumors. Nat Genet 17: 32–39, 1997[CrossRef][Medline]
Matsushita F, Miyawaki A, Mikoshiba K: Vomeroglandin/CRP-ductin is strongly expressed in the glands associated with the mouse vomeronasal organ: Identification and characterization of mouse vomeroglandin. Biochem Biophys Res Commun 268: 275–281, 2000[CrossRef][Medline]
Holmskov U, Mollenhauer J, Madsen J, Vitved L, Gronlund J, Tornoe I, Kliem A, Reid KB, Poustka A, Skjodt K: Cloning of gp-340, a putative opsonin receptor for lung surfactant protein D. Proc Natl Acad Sci U S A 96: 10794–10799, 1999[Abstract/Free Full Text]
Takeshita H, Sato M, Shiwaku H, Tagawa Y, Ayabe H, Horii A: Expression of the DMBT1 gene is frequently suppressed in human lung cancer. Jpn J Cancer Res 90: 903–908, 1999[CrossRef][Medline]
Wu W, Kemp BL, Proctor ML, Gazdar AF, Minna JD, Hong WK, Mao L: Expression of DMBT1, a candidate tumor suppressor gene, is frequently lost in lung cancer. Cancer Res 59: 1846–1851, 1999[Abstract/Free Full Text]
Mori M, Shiraishi T, Tanaka S, Yamagata M, Mafune K, Tanaka Y, Ueo H, Barnard GF, Sugimachi K: Lack of DMBT1 expression in esophageal, gastric and colon cancers. Br J Cancer 79: 211–213, 1999[Medline]
Hikita C, Takito J, Vijayakumar S, Al-Awqati Q: Only multimeric hensin located in the extracellular matrix can induce apical endocytosis and reverse the polarity of intercalated cells. J Biol Chem 274: 17671–17676, 1999[Abstract/Free Full Text]
Hikita C, Vijayakumar S, Takito J, Erdjument-Bromage H, Tempst P, Al-Awqati Q: Induction of terminal differentiation in epithelial cells requires polymerization of hensin by galectin 3. J Cell Biol 151: 1235–1246, 2000[Abstract/Free Full Text]
Vijayakumar S, Takito J, Hikita C, Al-Awqati Q: Hensin remodels the apical cytoskeleton and induces columnarization of intercalated epithelial cells: Processes that resemble terminal differentiation. J Cell Biol 144: 1057–1067, 1999[Abstract/Free Full Text]
Winyard PJD, Bao Q, Hughes RC, Woolf AS: Epithelial galectin-3 during human nephrogenesis and childhood cystic diseases. J Am Soc Nephrol 8: 1647–1657, 1997[Abstract]
Sechler JL, Takada Y, Schwarzbauer JE: Altered rate of fibronectin matrix assembly by deletion of the first type III repeats. J Cell Biol 134: 573–583, 1996[Abstract/Free Full Text]
Trusolino L, Bertotti A, Comoglio PM: A signaling adapter function for alpha6beta4 integrin in the control of HGF-dependent invasive growth. Cell 107: 643–654, 2001[CrossRef][Medline]
Top
Abstract
Introduction
to 
intercalated cell under the influence of acidification of the medium is due to the deposition of hensin in the extracellular matrix of these cells and activation of a novel inductive signal transduction pathway. The conversion of 
