Purification and reconstitution of a 75-kilodalton protein identified as a component of the renal Na+/glucose symporter

A 75-kilodalton (kDa) protein was purified from solubilized renal brush border membranes by using high-pressure liquid chromatography (HPLC) and preparative sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Functional and immunological properties identified the 75-kDa protein as a component of the Na+/glucose symport system. The purified protein was specifically recognized by a monoclonal antibody that functionally interacts with the Na+/glucose symporter. Na+-dependent phlorizin binding activity was associated with fractions containing the 75-kDa protein during HPLC fractionation on the anion exchanger Mono-Q and was greatly increased after reconstitution into egg yolk phosphatidylcholine vesicles. The final purified preparation contained glucosamine and a blocked N-terminus.     

Clonal analysis of immortalized renal proximal tubule cells: Na(+)/glucose cotransport system levels are maintained despite a decline in transport function

Primary rabbit kidney proximal tubule (RPT) cells (S.D. Chung et al., 1982, J. Cell Biol. 95, 118-126) were transfected with the vector pRSV-T, which contains SV40 early region genes. After the third passage (when normal cells had stopped dividing), individual colonies formed in cultures transfected with pRSV-T. Clonal isolates (RPT-I cells) could be obtained in a simple and reproducible manner. Southern analysis of clone RPT-I8 indicated the presence of SV40 early region genes. Nuclear SV40 T was detected. After 23 passages, and subcloning, RPT-I8 (and subclones) was found to express renal proximal tubule markers to a similar extent, indicating that the phenotype was stable. Nevertheless, the activities of the Na(+)/glucose cotransport system, gamma-glutamyl transpeptidase and alkaline phosphatase, were reduced as compared with primary cultures. Western analysis indicated that the level of Na(+)/glucose cotransporters was maintained in RPT-I8 cells, when compared with intact proximal tubules and primary cultures. Thus, the reduction in alpha-MG uptake in RPT-I8 cells may be attributed to other types of cellular alterations, including changes in energy metabolism. Indeed, growth in glucose-free medium was not observed in RPT-I8 cell cultures, suggesting that unlike primary RPT cells (J. C. Chung et al., 1992, J. Cell. Physiol. 150, 243-250), the gluconeogenic pathway was not intact.     

Regulation of intracellular pH in LLC-PK1 cells by Na+/H+ exchange

Summary Suspensions of LLC-PK₁ cells (a continuous epitheliod cell line with renal characteristics) are examined for mechanisms of intracellular pH regulation using the fluorescent probe BCECF. Initial experiments determine suitable calibration procedures for use of the BCECF fluorescent signal. They also determine that the cell suspension contains cells which (after 4 hr in suspension) have Na⁺ and K⁺ gradients comparable to those of cells in monolayer culture. The steady-state intracellular pH (7.05±0.01,n=5) of cells which have recovered in (pH 7.4) Na⁺-containing medium is not affected over several minutes by addition of 100 M amiloride or removal of extracellular Na⁺ (Na _o ⁺ /H _i ⁺ and Na _i ⁺ /H _o ⁺ exchange reactions are functionally inactive (compared to cellular buffering capacity). In contrast, Na _o ⁺ /H _i ⁺ exchange is activated by an increased cellular acid load. This activation may be observed directly either as a stimulation of net H⁺ efflux or net Na⁺ influx with decreasing intracellular pH. The extrapolation of this latter data suggests a set point of Na⁺/H⁺ exchange of approximately pH 7.0, consistent with the observed resting intracellular pH of approximately 7.05.     

Antibodies to a renal Na+/glucose cotransport system localize to the apical plasma membrane domain of polar mouse embryo blastomeres.

Mouse preimplantation embryos were examined for the cell surface expression of epitopes that cross-react with antibodies to a 75-kDa subunit of a purified porcine renal brush border Na+/glucose cotransport system. A Na+ cotransport system is hypothesized to reside in the apical plasma membrane domain of mouse polar blastomeres and to be associated with the induction of their apical-basal polarity. Western blot analysis showed that unfertilized oocytes as well as preimplantation embryos contain a cross-reacting antigen with an apparent molecular weight of about 75,000. Embryos and their isolated blastomeres were double-labeled and assayed by indirect immunofluorescence (IIF) for the expression of epitopes (visualized by labeling with rabbit antiserum or mouse monoclonal IgG to cotransporter followed by the appropriate rhodamine-conjugated second antibodies) and for the development of cell surface polarity (visualized by the apical restriction of fluoresceinated succinylated concanavalin A binding; FS Con A). IIF did not detect these epitopes until after the second cleavage when 4-cell embryos expressed low-to-moderate levels. Although epitopes were expressed on all surfaces of 4-cell blastomeres, some blastomeres expressed more epitopes on their apical surfaces than on their basolateral ones. All precompaction 8-cell embryos expressed epitopes, with expression being greater apically on some blastomeres. The level of expression appeared to reach a maximum on morulae and to decline on cavitating embryos. Assays performed on isolated blastomeres from postcompaction embryos showed that by the 16-cell stage epitope expression appeared to become restricted to FS Con A-labeled apical plasma membrane domains and was no longer evident on basolateral domains. This apparent apical restriction of epitope expression was confirmed by electron microscopic examination of immunogold-labeled isolated polar 16-cell blastomeres. These results demonstrate that preimplantation mouse embryos contain an antigen(s) that is immunologically and structurally similar to a 75-kDa renal Na+/glucose cotransporter. The onset of cell surface expression of this antigen precedes development of the stable polar phenotype.     

Developmentally regulated expression of a cell surface protein,neuropilin, in the mouse nervous system

Neuropilin (previously A5) is a cell surface glycoprotein that was originally identified in Xenopus tadpole nervous tissues. In Xenopus, neuropilin is expressed on both the presynaptic and postsynaptic elements in the visual and general somatic sensory systems, suggesting a role in neuronal cell recognition. In this study, we identified a mouse homologue of neuropilin and examined its expression in developing mouse nervous tissues. cDNA cloning and sequencing revealed that the primary structure of the mouse neuropilin was highly similar to that of Xenopus and that the extracellular segment of the molecule possessed several motifs that were expected to be involved in cell-cell interaction. Immunohistochemistry and in situ hybridization analyses in mice indicated that the expression of neuropilin was restricted to particular neuron circuits. Neuropilin protein was localized on axons but not on the somata of neurons. The expression of neuropilin persisted through the time when axons were actively growing to form neuronal connections. These observations suggest that neuropilin is involved in growth, fasciculation, and targeting for a particular groups of axons. © 1996 John Wiley & Sons, Inc.     

Kinetics and specificity of the renal Na+/myo-inositol cotransporter expressed in Xenopus Oocytes

The two-microelectrode voltage clamp technique was used to examine the kinetics and substrate specificity of the cloned renal Na⁺/myo-inositol cotransporter (SMIT) expressed in Xenopus oocytes. The steady-state myo-inositol-induced current was measured as a function of the applied membrane potential (V _m ), the external myo-inositol concentration and the external Na⁺ concentration, yielding the kinetic parameters: K _0.5 ^MI , K _0.5 ^Na , and the Hill coefficient n. At 100 mM NaCl, K _0.5 ^MI was about 50 m and was independent of V _m . At 0.5 mm myo-inositol, K _0.5 ^Na ranged from 76 mm at V _m =–50 mV to 40 mm at V _m =–150 mV. n was voltage independent with a value of 1.9±0.2, suggesting that two Na⁺ ions are transported per molecule of myo-inositol. Phlorizin was an inhibitor with a voltage-dependent apparent K _I of 64 m at V _m =–50 mV and 130 m at V _m = –150 mV. To examine sugar specificity, sugar-induced steady-state currents (at V _m =–150 mV) were recorded for a series of sugars, each at an external concentration of 50 mm. The substrate selectivity series was myo-inositol, scyllo-inositol > l-fucose > l-xylose > l-glucose, d-glucose, -methyl-d-glucopyranoside > d-galactose, d-fucose, 3-O-methyl-d-glucose, 2-deoxy-d-glucose > d-xylose. For comparison, oocytes were injected with cRNA for the rabbit intestinal Na⁺/glucose cotransporter (SGLT1) and sugar-induced steady-state currents (at V _m =–150 mV) were measured. For oocytes expressing SGLT1, the sugar selectivity was: d-glucose, -methyl-d-glucopyranoside, d-galactose, d-fucose, 3-O-methyl-d-glucose > d-xylose, l-xylose, 2-deoxy-d-glucose > myo-inositol, l-glucose, l-fucose. The ability of SMIT to transport glucose and SGLT1 to transport myo-inositol was independently confirmed by monitoring the Na⁺-dependent uptake of ³H-d-glucose and ³H-myo-inositol, respectively. In common with SGLT1, SMIT gave a relaxation current in the presence of 100 mm Na⁺ that was abolished by phlorizin (0.5 mm). This transient current decayed with a voltage-sensitive time constant between 10 and 14 msec. The presteady-state current is apparently due to the reorientation of the cotransporter protein in the membrane in response to a change in V _m . The kinetics of SMIT is accounted for by an ordered six-state nonrapid equilibrium model. Present address: W.M. Keck Biotechnology Resource Laboratory, Boyer Center for Molecular Medicine, Rm, 305A, Yale University, 295 Congress Ave., New Haven, Connecticut 06536-0812 Present address: National Institute for Physiological Sciences, Department of Cell Physiology, Okazaka, 444, JapanContributed equally to this workWe thank John Welborn for the HPLC analysis of the sugar substrates. This work was supported by grants from the National Institutes of Health DK19567, DK42479 and NS25554.     

Polarized distribution of the Na+/H+ exchange system in a renal cell line (LLC-PK1) with characteristics of proximal tubular cells

Studies of Na+ and H+ transport by confluent monolayers of the epithelial cell line LLC-PK1 were performed to verify the presence of a Na+/H+ exchange system. The presence of an outwardly directed H+ gradient produced a large stimulation of Na+ influx measured under net flux conditions. Amiloride (10(-3) M) completely inhibited Na+ influx stimulated by the H+ gradient and part of the Na+ influx measured in the absence of a pH gradient. Half-maximal inhibition of the Na+ influx stimulated by a pH gradient at 143 mM Na was observed at 5 microM amiloride. The presence of an inwardly oriented proton gradient also stimulated Na+ efflux from Na+-loaded cells. The stimulation was completely inhibited by the presence of 10(-3) M amiloride in the washout medium. These results indicate that this system could operate in the opposite direction depending on the orientation of the Na+ and H+ gradient. Incubation in Na+-free medium or in the presence of 10(-3) M ouabain resulted in a dramatic decrease of H+ release from LLC-PK1 cells. This H+ release was largely, although not completely, inhibited by 10(-4) M amiloride. Neither chloride substitution by the impermeable anion isethionate nor incubation in the presence of the ionophore valinomycin in high K+ medium affected Na+ influx by stimulated by a pH gradient. Inhibition of the Na+ influx by amiloride occurred only from the apical side of the monolayer. These results indicate that the Na+/H+ exchange system in LLC-PK1 monolayers is specifically localized in the apical membrane of the epithelial cells.     

Aquaporin-2 and Na+/H+ exchanger isoform 1 modulate the efficiency of renal cell migration

Aquaporin-2 (AQP2) promotes renal cell migration by the modulation of integrin β1 trafficking and the turnover of focal adhesions. The aim of this study was to investigate whether AQP2 also works in cooperation with Na⁺/H⁺ exchanger isoform 1 (NHE1), another well-known protein involved in the regulation of cell migration. Our results showed that the lamellipodia of AQP2-expressing cells exhibit significantly smaller volumes and areas of focal adhesions and more alkaline intracellular pH due to increased NHE1 activity than AQP2-null cells. The blockage of AQP2, or its physically-associated calcium channel TRPV4, significantly reduced lamellipodia NHE1 activity. NHE1 blockage significantly reduced the rate of cell migration, the number of lamellipodia, and the assembly of F-actin only in AQP2-expressing cells. Our data suggest that AQP2 modulates the activity of NHE1 through its calcium channel partner TRPV4, thereby determining pH-dependent actin polymerization, providing mechanical stability to delineate lamellipodia structure and defining the efficiency of cell migration.     

Different molecular capacity in the induction of apoptosis by polychlorinated biphenyl congeners in rat renal tubular cell cultures

The effects of a commercial polychlorinated biphenyl (PCB) mixture (Aroclor 1248) and two individual PCB congeners were evaluated on rat renal proximal tubule culture cell viability and internucleosomal DNA fragmentation (DNA ladder) characteristic of apoptosis. Treatment with Aroclor 1248 caused the loss of cell viability and promoted apoptosis in a concentration- and time-dependent manner. The two PCB congeners assessed can also induce apoptosis. However, the extent of apoptosis generated was greater for the non-ortho-substituted planar congener (3,3,4,4-tetrachlorobiphenyl) than for the di-ortho-substituted nonplanar congener (2,2,4,4,5,5-hexachlorobiphenyl). This correlated with the loss of cell viability since the planar compound is much more cytotoxic. The results suggest a different molecular mechanism in the induction of apoptosis by planar or nonplanar PCB congeners.     

Histidyl residues at the active site of the Na/succinate cotransporter in rabbit renal brush borders

Summary Mono-, dicarboxylic acid-, andd-glucose transport were measured in brush border vesicles from renal cortex after treatment with reagents known to modify terminal amino, lysyl, -amino, guanidino, serine/threonine, histidyl, tyrosyl, tryptophanyl and carboxylic residues. All three sodium-coupled cotransport systems proved to possess sulfhydryl (and maybe tryptophanyl sulfhydryl, disulfide, thioether and tyrosyl) residues but not at the substrate site or at the allosteric cavity for the Na coion. Histidyl groups seem to be located in the active site of the dicarboxylic transporter in that the simultaneous presence of Na and succinate protects the transporter against the histidyl specific reagent diethylpyrocarbonate. Lithium, which specifically competes for sodium sites in the dicarboxylic acid transporter, substantially blocked the protective effect of Na and succinate. Hydroxylamine specifically reversed the covalent binding of diethylpyrocarbonate to the succinate binding site. The pH dependence of the Na/succinate cotransport is consistent with an involvement of histidyl and sulfhydryl residues. We conclude that a histidyl residue is at, or is close to, the active site of the dicarboxylate transporter in renal brush border membranes.     

Peculiarities of the Na+/d-glucose cotransport system in necturus renal tubules

The effects of d-glucose addition to a glucose-free luminal perfusate were investigated in the proximal tubule of Necturus kidney, by electrophysiological techniques. The main findings are: (1) In the presence of sodium, d-glucose produces $\text{10.5}\text{mV}\text{± 1.1}$ (S.E.) depolarization. (2) Phlorizin reduces the magnitude of this response to $\text{2.1 ± 0.1}\text{mV}$. (3) The glucose-evoked depolarization, $\text{ΔV}{}_{\mathrm{<mtext>G</mtext>}}$, does not alter the intracellular K⁺ activity nor is it affected by peritubular addition of ouabain. (4) Isosmotic reduction of Na⁺ concentration in luminal perfusate from 95 to 2 mmol/l (choline or Li⁺ substituting for Na⁺) does not change the magnitude of $\text{ΔV}{}_{\mathrm{<mtext>G</mtext>}}$; complete removal of sodium from the lumen lowers the value of $\text{ΔV}{}_{\mathrm{<mtext>G</mtext>}}$ ($\text{3.2 ± 0.2}\text{mV}$) but the response is not abolished. This observation suggests that the d-glucose carrier of renal tubules in Necturus is poorly specific with regard to the cotransported cation species.     

Measurements of intracellular pH in single LLC-PK1 cells: Recovery from an acid load via basolateral Na+/H+ exchange

Summary LLC-PK₁ cells (a continuous epithelioid cell line with renal characteristics) are examined by microspectrofluorometry as single cells, in order to determine the mechanism of intracellular pH (pH _i ) recovery from an acid load imposed by ammonium preincubation and removal (NH₄ prepulse). Initial experiments evaluate the intracellular K⁺ levels through a null point analysis of total cellular K⁺ with flame photometry. The response of BCECF (a pH-sensitive fluorescent dye) is then calibrated, using saturating concentrations of nigericin to cause defined changes in pH _i . For experiments with the microspectrofluorometer, LLC-PK₁ cells were grown on either glass coverslips or filters (the latter attached to plastic coverslips with a hole under the filter). The cells on glass coverslips demonstrate a Na⁺-dependent recovery from an (NH₄ prepulse) acid load which is sensitive to 1 M ethylisopropylamiloride. They also demonstrate a set point of activation of Na⁺/H⁺ exchange. When examined for changes in pH _i due to changes in membrane potential, plasma membrane proton conductance could not be detected at resting pH _i . Cells grown on filters also demonstrate a pH _i recovery from an acid load which is Na⁺ dependent and ethylisopropylamiloride sensitive, but in this configuration, the majority of cells (22/23 preparations) require Na⁺ at the basolateral membrane for rapid pH _i recovery. The morphology and polarity of the cells grown on permeable supports appears normal at the electron-microscopic level. The results are not affected by changes in cell seeding density or collagen treatment of the filters.     

Isolation and function of the amino acid transporter PAT1 (slc36a1) from rabbit and discrimination between transport via PAT1 and system IMINO in renal brush-border membrane vesicles

Reabsorption of amino acids is an important function of the renal proximal tubule. pH-dependent amino acid transport has been measured previously using rabbit renal brush-border membrane vesicles (BBMV). The purpose of this investigation was to determine whether this pH-dependent uptake represents H⁺/amino acid cotransport via a PAT1-like transport system. The rabbit PAT1 cDNA was isolated (2296bp including both 5′ and 3′ untranslated regions and poly(A) tail) and the open reading frame codes for a protein of 475 amino acids (92% identity to human PAT1). Rabbit PAT1 mRNA was found in all tissues investigated including kidney. When expressed heterologously in a mammalian cell line, rabbit PAT1 mediates pH-dependent, Na⁺-independent uptake of proline, glycine, l-alanine and α-(methylamino)isobutyric acid. Proline uptake was maximal at pH?5.0 (K_m?2.2±0.7?mM). A transport system with identical characteristics (ion dependency, substrate specificity) was detected in rabbit renal BBMV where an overshoot was observed in the absence of Na⁺ but in the presence of an inwardly directed H⁺ gradient. In the presence of Na⁺ and under conditions in which PAT1 transport function was suppressed, a second proline uptake system was detected that exhibited functional characteristics similar to those of the IMINO system. The functional characteristics of rabbit PAT1 in either mammalian cells or renal BBMV suggest that PAT1 is the low-affinity transporter of proline, glycine and hydroxyproline believed to be defective in patients with iminoglycinuria.     

Use of a renal tubule cell line (LLC-PK1) to study the nephrotoxic potential of a kappa-type bence-jones protein

Summary The cytotoxicity of a Bence-Jones protein was assessed using a porcine renal tubule cell line (LLC-PK1), with the aim of developing a model for studying the potential nephrotoxicity of these proteins. The effects of a kappa Bence-Jones protein on cell viability were studied by means of biochemical methods (supravital dye uptake and measurement of cellular enzyme activities) and morphological electron microscopy. After a 24-h-treatment with Bence-Jones protein, a moderate cytotoxicity (about 15%) was noted but only a minor difference compared to treatment with bovine albumin in the same conditions. The morphological study showed a few cells in the process of lysis, but their numbers were insufficient for the demonstration of a clear cytotoxic effect. Immunocytochemical studies showed Bence-Jones protein fixation on some cells, especially on the outer membrane. Labeling of the hyaloplasm and basal pole of a few cells pointed to internalization of protein by LLC-PK1 cells. Although the cytotoxicity of the Bence-Jones protein tested here was only moderate, the use of this model enabled its cytotoxic effect to be distinguished from that ofβ-lactoglobulin. This isolate could serve as a “moderate control” for a later study with a BJP having caused acute renal failure.     

Localization of the Na+-sugar cotransport system in a kidney epithelial cell line (LLC PK1)

Studies of the localization of the Na⁺-dependent sugar transport in monolayers of LLC PK₁ cells show that the uptake of a methyl α-d-glucoside, a nonmetabolizable sugar which shares the glucose-galactose transport system, occurs mainly from the apical side of the monolayer. Kinetics of [³H]phlorizin binding to monolayers of LLC PK₁ cells were also measured. These studies demonstrate the presence of two distinct classes of receptor sites. The class comprising high affinity binding sites had a dissociation constant ($\text{K}{}_{\mathrm{<mtext>d</mtext>}}$) of 1.2 μM and a concentration of high affinity receptors of 0.30 μmol binding sites per g DNA. The other class involving low affinity sites had a $\text{K}{}_{\mathrm{<mtext>d</mtext>}}$ of 240 μM with the number of binding sites equal to 12 μmol/g DNA. Phlorizin binding at high affinity binding sites is a Na⁺-dependent process. Binding at the low affinity sites on the contrary is Na⁺-independent. The mode of action of Na⁺ on the high affinity binding sites was to increase the dissociation constant without modifying the number of binding sites. The Na⁺ dependence and the matching of $\text{K}{}_{\mathrm{<mtext>d</mtext>}}$ for high affinity binding sites with the $\text{K}{}_{\mathrm{<mtext>i</mtext>}}$ of phlorizin for the inhibition of methyl α-d-glucoside strongly suggest that the high affinity phlorizin binding site is, or is part of the methyl α-d-glucoside transport system. Binding studies from either side of the monolayer also show that the binding of phlorizin at the Na⁺ dependent high affinity binding sites occurs mainly from the apical rather than the basolateral side. The specific location of the Na⁺-dependent sugar transport system in the apical membrane of LLC PK₁ cells is, therefore, another expression of the functional polarization of epithelial cells that is retained under tissue culture condition. In addition, since this sugar transport almost disappears after the cells are brought into suspension, it can be used as a marker to study the development of the apical membrane in this cell line.     

A member of the mitogen-activated protein 3-kinase family is involved in the regulation of plant vacuolar glucose uptake

Vacuolar solute accumulation is an important process during plant development, growth and stress responses. Although several vacuolar carriers have been identified recently, knowledge regarding the regulation of transport is still limited. Solute accumulation may be controlled by various factors, such as alterations in carrier abundance or activity. Phosphorylation via kinases is a well-known principle for activation or deactivation of proteins. Several phosphorylated proteins have been identified in the tonoplast proteome; however, kinases that catalyse the phosphorylation of tonoplast proteins are currently unknown. The tonoplast monosaccaride transporter from Arabidopsis (AtTMT1) and its homologue from barley have multiple phosphorylation sites in their extremely large loops. Here we demonstrate that the loop of AtTMT1 interacts with a mitogen-activated triple kinase-like protein kinase (VIK), that an aspartate-rich loop domain is required for effective interaction, and that the presence of VIK stimulates glucose import into isolated vacuoles. Furthermore, the phenotype of VIK loss-of-function plants strikingly resembles that of plants lacking AtTMT1/2. These data suggest that VIK-mediated phosphorylation of the AtTMT1 loop enhances carrier activity and consequently vacuolar sugar accumulation. As many phosphorylated proteins have been identified in the tonoplast, differential phosphorylation may be a general mechanism regulating vacuolar solute import.     

Role of hydration in the conformational transitions between unliganded and liganded forms of loop 13 of the Na+/glucose cotransporter 1

SGLT1 as a Na+/glucose cotransporter is inhibited by phlorizin, a phloretin 2'-glucoside that has strong interactions with the C-terminal loop 13 (residues 541-638). Here we investigated the effect of a partial substitution of glycerol for water in the medium on the stability and phlorizin-binding function of loop 13 using fluorescence spectroscopy. Increasing the glycerol concentration promoted an increase in the stability of the protein to urea. The ability of loop 13 to expose hydrophobic surface promoted by phlorizin binding was partially lost in the presence of glycerol (20%). Glycerol also led to a decrease in the phlorizin affinity of loop 13 in solution. Approximately 15 molecules of water were taken up to cover additional surface area (137.7+/-27.9A(2)) upon formation of the loop 13-phlorizin complex. Together these results demonstrate quantitatively that the stability and phlorizin affinity of loop 13 are critically dependent on protein hydration.     

Expression of the Wnt inhibitor Dickkopf-1 is required for the induction of neural markers in mouse embryonic stem cells differentiating in response to retinoic acid

Cultured mouse D3 embryonic stem (ES) cells differentiating into embryoid bodies (EBs) expressed several Wnt isoforms, nearly all isotypes of the Wnt receptor Frizzled and the Wnt/Dickkopf (Dkk) co-receptor low-density lipoprotein receptor-related protein (LRP) type 5. A 4-day treatment with retinoic acid (RA), which promoted neural differentiation of EBs, substantially increased the expression of the Wnt antagonist Dkk-1, and induced the synthesis of the Wnt/Dkk-1 co-receptor LRP6. Recombinant Dkk-1 applied to EBs behaved like RA in inducing the expression of the neural markers nestin and distal-less homeobox gene (Dlx-2). Recombinant Dkk-1 was able to inhibit the Wnt pathway, as shown by a reduction in nuclear beta-catenin levels. Remarkably, the antisense- or small interfering RNA-induced knockdown of Dkk-1 largely reduced the expression of Dlx-2, and the neuronal marker beta-III tubulin in EBs exposed to RA. These data suggest that induction of Dkk-1 and the ensuing inhibition of the canonical Wnt pathway is required for neural differentiation of ES cells.