A hepatocyte-specific basolateral membrane protein is targeted to the same domain when expressed in Madin-Darby canine kidney cells

Molecular Evolution of the Transferrin Receptor/Glutamate Carboxypeptidase II Family

The transferrin receptor family is represented by at least seven different homologous proteins in primates. Transferrin receptor (TfR1) is a type II membrane glycoprotein that, as a cell surface homodimer, binds iron-loaded transferrin as part of the process of iron transfer and uptake. Other family members include transferrin receptor 2 (TfR2), glutamate carboxypeptidase II (GCP2 or PSMA), N-acetylated alpha-linked acidic dipeptidase-like protein (NLDL), N-acetylated alpha-linked acidic dipeptidase 2 (NAALAD2), and prostate-specific membrane antigen-like protein (PMSAL/GCPIII). We compared 86 different sequences from 24 different species, from mammals to fungi. Through this comparison, we have identified several highly conserved residues specific to each family not previously associated with clinical mutations. The evolutionary history of the TfR/GCP2 family shows repeated episodes of duplications consistent with recent theories that nondispensable, slowly evolving genes are more likely to form multiple gene families.

Conversion of proteins from a non-polarized to an apical secretory pattern in MDCK cells

Previously it was shown that fusion proteins containing the amino terminus of an apical targeted member of the serpin family fused to the corresponding carboxyl terminus of the non-polarized secreted serpin, antithrombin, are secreted mainly to the apical side of MDCK cells. The present study shows that this is neither due to the transfer of an apical sorting signal from the apically expressed proteins, since a sequence of random amino acids acts the same, nor is it due to the deletion of a conserved signal for correct targeting from the non-polarized secreted protein. Our results suggest that the polarity of secretion is determined by conformational sensitive sorting signals.

Suppression of distant pulmonary metastasis of MDA-MB 435 human breast carcinoma established in mammary fat pads of nude mice by retroviral-mediated TIMP-2 gene transfer

BACKGROUND

Previous studies have shown that TIMP-2 overexpression is a useful therapeutic tool for inhibiting tumor growth and invasion in animals. However, it has not been reported whether genetic manipulation for TIMP-2 overexpression can induce an inhibitory effect on spontaneous metastasis from the primary tumor site to other organs such as lungs or lymph nodes in an animal model.

METHODS

The present studies describe the effects of retrovirus-mediated TIMP-2 gene transfer into human breast cancer cell lines on the in vitro invasion of the tumor cells or the in vivo growth in nude mouse. Here we also used retroviral-mediated TIMP-2 overexpression by intratumoral injection for suppression of metastasis in human breast carcinoma established in the mammary fat pad of nude mice.

RESULTS

As expected, overexpression of TIMP-2 inhibited matrix metalloprotenase (MMP) activity and invasion of the tumor cells. Also, the growth rate of tumors grafted with the breast cancer cells transduced with the retrovirus vector encoding TIMP-2 cDNA was significantly slower than the growth rate of tumors grafted with the breast cancer cells transduced with a control retrovirus vector. Furthermore, single intratumoral injection of the TIMP-2 retrovirus-producing cells into human breast tumor tissue established in mammary fat pads of nude mice showed a dramatic decrease in size and number of lung metastatic tumors.

CONCLUSIONS

Retrovirus-mediated TIMP-2 gene transfer into human breast cancer cells is able to down-regulate invasion and show that tumor-derived angiogenesis is reduced. In this model, retroviral-mediated transduction of TIMP-2 cDNA into a limited population of human tumor cells inhibits tumor growth and prevents distant pulmonary metastasis. These results indicate that it may not be necessary to deliver and express these genes in every single tumor cell as long as the level of expression in a limited number of transduced cells is sufficient to prevent the excessive breakdown of the extracellular matrix.

N-Glycans, not the GPI anchor, mediate the apical targeting of a naturally glycosylated, GPI-anchored protein in polarised epithelial cells

The glycosyl-phosphatidylinositol (GPI) anchor mediates the apical sorting of proteins in polarised epithelial cells through its interaction with lipid rafts. Here we investigated the signals required for the apical targeting of the naturally N-glycosylated and GPI-anchored membrane dipeptidase by selective point mutation to remove the GPI anchor addition signal or the sites for N-linked glycosylation, or both. Activity assays, immunoblotting and immunofluorescence microscopy revealed that the constructs lacking the GPI anchor were secreted from Madin-Darby canine kidney (MDCK) cells, whereas those retaining the GPI anchor were attached at the cell surface, irrespective of the glycosylation status. Wild-type membrane dipeptidase was expressed preferentially on the apical surface of both MDCK and CaCo-2 cells. By contrast, the GPI-anchored construct lacking the N-glycans was targeted preferentially to the basolateral surface of both cell types. In constructs lacking the GPI anchor, the N-glycans also targeted the protein to the apical surface. Both the apically targeted, glycosylated and the basolaterally targeted, unglycosylated GPI-anchored forms of the protein were located in detergent-insoluble lipid rafts. These data indicate that it is the N-glycans, not the association of the GPI anchor with lipid rafts, which determine apical targeting of an endogenously N-glycosylated, GPI-anchored protein in polarised epithelial cells.

Serpins are apically secreted from MDCK cells independently of their raft association

It has been suggested that detergent-resistant membranes (DRMs), also known as lipid rafts, are involved in vectorial transport of proteins to the apical surface. In this report we use Madin-Darby canine kidney (MDCK) cells expressing the apically secreted C1-esterase inhibitor, the non-sorted antithrombin or chimeras of serpins to study the possible connection between DRM association and apical targeting of secretory proteins. We found newly synthesised C1-esterase inhibitor associated with DRMs in MDCK cells, whereas antithrombin was not. However, two chimeric proteins, secreted mainly from the apical membrane, do not associate with DRMs. Based on these observations we suggest that apical targeting and association with DRMs are two independent events for secretory serpins.

Secretion of antithrombin is converted from nonpolarized to apical by exchanging its amino terminus for that of apically secreted family members

The three members of the serpin family, corticosteroid binding globulin, alpha1-antitrypsin, and C1 inhibitor are secreted apically from Madin-Darby canine kidney (MDCK) cells, whereas two homologous family members, antithrombin and plasminogen activator inhibitor-1, are secreted in a nonpolarized fashion. cDNAs coding for chimeras composed of complementary portions of an apically targeted serpin and a nonsorted serpin were generated, expressed in MDCK cells, and the ratio between apical and basolateral secretion was analyzed. These experiments identified an amino-terminal sequence of corticosteroid binding globulin (residues 1-19) that is sufficient to direct a chimera with antithrombin mainly to the apical side. A deletion/mutagenesis analysis showed that no individual amino acid is absolutely required for the apical targeting ability of amino acids 1-30 of corticosteroid binding globulin. The corresponding amino-terminal sequences of alpha1-antitrypsin and C1 inhibitor were also sufficient to confer apical sorting. Based on our results we suggest that the apical targeting ability is encoded in the conformation of the protein.

Chemical synthesis of lymphotactin: a glycosylated chemokine with a C-terminal mucin-like domain

The synthesis of a 93-residue chemokine, lymphotactin, containing eight sites of O-linked glycosylation, was achieved using the technique of native chemical ligation. A single GalNAc residue was incorporated at each glycosylation site using standard Fmoc-chemistry to achieve the first total synthesis of a mucin-type glycoprotein. Using this approach quantities of homogeneous material were obtained for structural and functional analysis.

The cytoplasmic domain of the interleukin-6 receptor gp80 mediates its basolateral sorting in polarized madin-darby canine kidney cells

The IL-6 receptor complex is expressed in different polarized epithelial cells such as liver hepatocytes and intestinal cells. It consists of two subunits: gp80, which binds the ligand, and gp130, which is responsible for signal transduction. In stably transfected Madin-Darby canine kidney (MDCK) cells we have studied the localization of the human IL-6 receptor subunits and found that gp80 and gp130 are predominantly expressed at the basolateral membrane. Analysis of MDCK cells expressing truncated forms of gp80 or gp130 showed that loss of the cytoplasmic domains results in apical delivery. Expression of deletion mutants of gp80 in MDCK cells led to the identification of two discontinous motifs responsible for basolateral sorting: a membrane-proximal tyrosine-based motif (YSLG) and a more membrane-distal dileucine-type motif (LI). Activation of signal transducer and activator of transcription-3 (STAT-3) only occurred via basolaterally located gp80, suggesting that endogenous gp130 is also constrained to the basolateral plasma membrane. Our identification of a basolateral sorting signal within the cytoplasmic region of gp80 for the first time attributes a function to this domain.

Apical and non-polarized secretion of serpins from MDCK cells

Corticosteroid binding globulin, a member of the serpin family, was previously shown to be secreted mainly apically from MDCK cells in an N-glycan independent manner [Larsen et al. (1999) FEBS Lett. 451, 19-22]. Apart from N-glycosylation, serpins are not known to carry any other posttranslational modifications, suggesting the presence of a proteinaceous apical sorting signal. In the present study we have expressed four other members of the serpin family: alpha1-antitrypsin, C1 inhibitor, plasminogen activator inhibitor-1 and antithrombin in MDCK cells. Tight monolayers of transfected cells were grown on filters and the amounts of recombinantly expressed serpins in the apical and the basolateral media were determined. alpha1-Antitrypsin and C1 inhibitor were found mainly in the apical medium whereas plasminogen activator inhibitor-1 and antithrombin were found in roughly equal amounts in the apical and basolateral media. Control experiments showed that all four serpins are transported along the exocytotic pathway in an uncomplicated way that does not involve transcytosis or differences in stability on the two sides of the cells. We conclude that some members of the serpin family including corticosteroid binding globulin, alpha1-antitrypsin and C1 inhibitor are secreted mainly apically from MDCK cells whereas plasminogen activator inhibitor-1 and antithrombin are secreted in a non-polarized manner.

Basolateral sorting signals differ in their ability to redirect apical proteins to the basolateral cell surface

Polarized sorting of membrane proteins in epithelial cells is mediated by cytoplasmic basolateral signals or by apical signals in the transmembrane or exoplasmic domains. Basolateral signals were generally found to be dominant over apical determinants. We have generated chimeric proteins with the cytoplasmic domain of either the asialoglycoprotein receptor H1 or the transferrin receptor, two basolateral proteins, fused to the transmembrane and exoplasmic segments of aminopeptidase N, an apical protein, and analyzed them in Madin-Darby canine kidney cells. Whereas both cytoplasmic sequences induced endocytosis of the chimeras, only that of the transferrin receptor mediated basolateral expression in steady state. The H1 fusion protein, although still largely sorted to the basolateral side in biosynthetic surface transport, was subsequently resorted to the apical cell surface. We tested whether the difference in sorting between trimeric wild-type H1 and the dimeric aminopeptidase chimera was caused by the number of sorting signals presented in the oligomers. Consistent with this hypothesis, the H1 signal was fully functional in a tetrameric fusion protein with the transmembrane and exoplasmic domains of influenza neuraminidase. The results suggest that basolateral signals per se need not be dominant over apical determinants for steady-state polarity and emphasize an important contribution of the valence of signals in polarized sorting.

N-glycans are not the signal for apical sorting of corticosteroid binding globulin in MDCK cells

It has been suggested that N-glycans act as a general sorting signal for secretory proteins in MDCK cells [Scheiffele et al. (1995) Nature 378, 96-98]. Human corticosteroid binding globulin contains six consensus sites for N-glycosylation and is known to be secreted to the apical side of MDCK cells. Our results show that wild-type corticosteroid binding globulin is N-glycosylated when it is recombinantly expressed in MDCK cells. Six mutants, each lacking one of the N-glycosylation sites, and a mutant lacking all six N-glycosylation sites were also secreted to the apical side of MDCK cells in a polarized manner. Thus, the N-glycans on corticosteroid binding globulin do not act as an apical sorting signal in MDCK cells.

Predominant basolateral proteolytic processing of prosomatostatin into somatostatin-28 in polarized LLC-PK1 cells

Polarized epithelial cells secrete specific proteins through their apical or basolateral membrane. In the present study, we have expressed the human prosomatostatin cDNA in the pig kidney epithelial cell line (LLC-PK1) and monitored the processing and release of the somatostatin-related peptides. Analysis by high-performance liquid chromatography and radioimmunoassay of the somatostatin-related peptides synthesized by the transfected cells showed that the LLC-PK1 cells released prosomatostatin and somatostatin-28 (S-28) in the culture medium. Furthermore, when the cells were polarized, we observed release of prosomatostatin from both membrane domains (apical and basolateral), while liberation of S-28 was mostly from the basolateral side. This observation suggests that, in these cells, the proprotein convertase(s) responsible for prosomatostatin processing is(are) associated with the basolateral secretory pathway.

Pituitary Folliculo-Stellate-Like Cells Stimulate Somatotroic Pituitary Tumor Growth in Nude Mice

A new cell line (TtT/GF) established from a murine pituitary thyrotropic tumor having characteristics similar to those of pituitary folliculo-stellate cell (FS cell) was implanted into nude mice together with cells from a rat pituitary somatotrophic tumor cell line (MtT/S) to determine whether the former enhances pituitary tumor growth. For as long as 2-3 mo after implantation, MtT/S cells implanted either alone or together with fibroblasts formed either no tumors or only very small tumors in the nude mice. In contrast, all of the nude mice that had received MtT/S cells implanted together with TtT/GF cells developed large tumors. Furthermore, the mice bearing the MtT/S and TtT/GF implants showed a significantly higher body weight and serum growth hormone level than those bearing only MtT/S cells or a combination of MtT/S cells and fibroblasts. The TtT/GF cell line itself had no tumorigenicity during the experimental period. Therefore, the TtT/GF cell line as a model of FS cells enhanced pituitary endocrine cell tumor formation. Additionally, immunocytochemistry showed that TtT/GF cells positive for glial fibrillary acidic protein (GFAP) or S-100 protein were present in the parenchymatous tissue elements or connective tissue surrounding the tumor nests. In the parenchymatous tissue, the TtT/GF cells exhibited a stellate appearance and surrounded neighboring tumor cells with their long cell processes. These results suggest that TtT/GF cells can serve as a model for pituitary FS cells, and are capable of stimulating pituitary tumor growth either by modifying the microenvironment or producing growth factors.

Expression of an endogenous asialoglycoprotein receptor in a human intestinal epithelial cell line, Caco-2

We have previously shown that rat asialoglycoprotein receptor expressed in the intestine and liver differ in mRNA size, cell surface distribution, and ratio of compositional protein subunits. In this study, we examined a well characterized intestinal epithelial cell line, Caco-2, as a potential model for studying endogenous receptor in a polarized cell line. Both subunits H1 and H2 of human asialoglycoprotein receptor were detected in Caco-2 cells by Western blots using subunit-specific antisera raised against the hepatic receptor. Antigenic receptor level in fully differentiated Caco-2 cells was approx. 1/3 to 1/2 the level of hepatic HepG2 cells H1 was the dominant subunit in both cell lines. The apparent size of H1 and H2 in Caco-2 cells was not the same as that in HepG2 cells, due to differences in N-linked glycosylation. Consistent with this finding, Northern blot analysis showed that receptor mRNA in the two cell types was of identical size. In pulse-chase experiments H1 was first detected as a 'high-mannose' precursor (40 kDa) in Caco-2 cells that was converted to mature H1 (43 kDa) with a half-life of approx. 60 min. Antigenic levels of H1 and H2 in undifferentiated Caco-2 cells were low, but increased rapidly during cell differentiation, reaching a peak level at 7 days after confluence. Immunocytochemical staining and domain-selective cell surface biotinylation assays showed that the ASGP-R was predominantly localized in the basolateral domain. The receptor in Caco-2 cells was capable of mediating specific uptake and degradation of [125I]asialoorosomucoid. The ligand uptake capacity of the basolateral surface of was approx. 10-fold higher than the apical. These characteristics (H1 subunit and basolateral predominance) of the receptor in Caco-2 cells, resembles the hepatic receptor. We conclude that Caco-2 cells endogenously express in ectopic hepatic-type functional asialoglycoprotein receptor.

Uteroglobin, an apically secreted protein of the uterine epithelium, is secreted non-polarized form MDCK cells and mainly basolaterally from Caco-2 cells

A complete cDNA encoding rabbit uteroglobin was constructed and expressed in MDCK and Caco-2 cells. The MDCK cells secrete uteroglobin in approximately equal amounts to the apical and the basolateral side, whereas the Caco-2 cells secrete uteroglobin mainly to the basolateral side. Both MDCK and Caco-2 cells thus secrete uteroglobin in a non-sorted manner. It has, however, previously been shown that uteroglobin is secreted exclusively at the apical membrane in primary cell culture of endometrial epithelial cells [S.K. Mani et al. (1991) Endocrinology 128, 1563-1573]. This suggests that either the endometrial epithelium has an apical default pathway or recognises a sorting signal not recognised by MDCK cells and Caco-2 cells. Our data thus show that a soluble molecule can be secreted at the apical, the basolateral or both membranes depending on the cell type.

The apical sorting signal on human aminopeptidase N is not located in the stalk but in the catalytic head group

Human aminopeptidase N carries an apical sorting signal on its ectodomain necessary for its correct transport to the apical membrane in Madin-Darby canine kidney cells. To determine whether the apical sorting signal is localized in the serine/threonine rich stalk or in the catalytic head group, anchor/stalk-minus aminopeptidase N, consisting of the hemagglutinin signal peptide and the catalytic head group of human aminopeptidase N, was expressed in MDCK cells. Anchor/stalk-minus aminopeptidase N was secreted mainly to the apical side. The catalytic head group of human aminopeptidase N thus carries an apical sorting signal.

Differential expression of asialoglycoprotein receptor subunits in the endocytic compartment during liver regeneration

Asialoglycoprotein receptors, responsible for the removal of circulating asialoglycoproteins by the liver, are located in at least two different membrane locations in hepatocytes. Receptors on the cell surface account only for a minor proportion (20-36%), for the majority of receptors in the liver are located intracellularly, mainly in the endocytic membrane networks. An understanding of the basis of receptor distribution and the underlying trafficking of receptors between the hepatocyte's polarised cell surface and the endocytic compartment would be aided if biochemical differences between the receptors in these pools were established. We now show, using three antibodies that recognise the receptor subunits in rat liver (RHL-1, RHL-2 and RHL-3), that the asialoglycoprotein receptors located in the plasma membrane domains and the endocytic compartment differ in oligomeric composition, sialic acid content, and solubility in Triton X-114 using two-phase systems. It is well established that the expression of the asialoglycoprotein receptor is down-regulated in livers regenerating after a partial hepatectomy. We demonstrate that the levels of the receptor subtype that is located mainly in the endocytic compartment (RHL-1, 42 kDa) was elevated in regenerating liver by agents that regulate cAMP production, whereas the levels of the other receptor subtypes remained unchanged. The asialoglycoprotein receptor subtypes that are present in different subcellular locations are thus regulated independently.