Lysosomal glycosphingolipid recognition by NKT cells

NKT cells represent a distinct lineage of T cells that coexpress a conserved alphabeta T cell receptor (TCR) and natural killer (NK) receptors. Although the TCR of NKT cells is characteristically autoreactive to CD1d, a lipid-presenting molecule, endogenous ligands for these cells have not been identified. We show that a lysosomal glycosphingolipid of previously unknown function, isoglobotrihexosylceramide (iGb3), is recognized both by mouse and human NKT cells. Impaired generation of lysosomal iGb3 in mice lacking beta-hexosaminidase b results in severe NKT cell deficiency, suggesting that this lipid also mediates development of NKT cells in the mouse. We suggest that expression of iGb3 in peripheral tissues may be involved in controlling NKT cell responses to infections and malignancy and in autoimmunity.

Erythrocyte P antigen: cellular receptor for B19 parvovirus

The pathogenic human parvovirus B19 replicates only in erythroid progenitor cells. This virus was shown to bind to blood-group P antigen, as measured by hemagglutination. Erythrocytes lacking P antigen were not agglutinated with B19. Purified P antigen (globoside) blocked the binding of the virus to erythroid cells and the infectivity of the virus in a hematopoietic colony assay. Target cells were protected from infection by preincubation with monoclonal antibody to globoside. Knowledge of a parvovirus receptor has implications for understanding the pathogenesis of parvovirus infections and for the use of parvoviruses in gene therapy.

Pathogenesis of shigella diarrhea. XI. Isolation of a shigella toxin-binding glycolipid from rabbit jejunum and HeLa cells and its identification as globotriaosylceramide

A glycolipid that specifically binds shigella toxin was isolated from both HeLa cells and rabbit jejunal mucosa and identified as globotriaosylceramide (Gb3) by its identical mobility on HPTLC to authentic erythrocyte Gb3. Toxin also bound to a band tentatively identified as alpha-hydroxylated Gb3. In addition, toxin bound to P1 antigen present in group B human erythrocyte glycolipid extracts. The common feature of the three binding glycolipids is a terminal Gal alpha 1----4Gal disaccharide linked beta 1----4 to either Glc or GlcNAc. Globoisotriaosylceramide, which differs from Gb3 only in possessing a Gal alpha 1----3Gal terminal disaccharide, and LacCer, which lacks the terminal Gal residue of Gb3, were incapable of binding the toxin. Binding was shown to be mediated by the B subunit by the use of isolated toxin A and B subunits and monoclonal subunit-specific antibodies. Gb3-containing liposomes competitively inhibited the binding of toxin to HeLa cell monolayers but did not inhibit toxin-induced cytotoxicity. These studies show an identical carbohydrate-specific glycolipid receptor for shigella toxin in gut and in HeLa cells. The toxin B subunit that mediates this binding has also been shown to recognize a glycoprotein receptor with different sugar specificity. Thus, we have demonstrated that the same small (Mr 6,500) B subunit polypeptide has two distinctive carbohydrate-specific binding sites. The Gal alpha 1----4Gal disaccharide of the glycolipid toxin receptor is also recognized by the Gal-Gal pilus of uropathogenic E. coli. This suggests the possibility that the pilus and toxin B subunit contain homologous sequences. If this is true, it may be possible to use the purified Gal-Gal pilus to produce toxin-neutralizing antibodies.

Glycolipid receptors for uropathogenic Escherichia coli on human erythrocytes and uroepithelial cells

A specific family of glycolipids, the globoseries, was shown to act as receptors on human uroepithelial cells and erythrocytes for the majority of uropathogenic Escherichia coli strains attaching to or hemagglutinating those cells. This was demonstrated in three different ways: (i) correlation between the natural presence of glycolipid in the target cell (erythrocytes of different species) and binding of bacteria; (ii) inhibition of attachment to human uroepithelial cells by preincubation of bacteria and glycolipid; and (iii) induction of binding to unreactive cells by coating of these cells with glycolipid. Strains reacting with the receptor agglutinated guinea pig erythrocytes in a mannose-resistant way after, but not before, coating of the cells with globotetraosylceramide. Unrelated glycolipids were not recognized. The reaction was made independent of simultaneous occurrence of mannose-sensitive adhesions on the strains by addition of D-mannose. The receptor-coated cells were used as a tool to screen for prevalence of receptor recognition in a collection of 453 E. coli strains isolated from patients with urinary tract infection or from the stools of healthy children. Of 150 strains attaching to human uroepithelial cells and agglutinating human erythrocytes, 121 bound to globotetraosylceramide (81%). Globoside recognition was especially frequent among pyelonephritis strains (74/81). The glycolipid composition of the urogenital epithelium and kidney tissue and the ability of uropathogenic E. coli to bind to these glycolipids may be a determinant in host-parasite interaction leading to urinary tract infection.

Molecular basis of Escherichia coli colonization of the upper urinary tract in BALB/c mice. Gal-Gal pili immunization prevents Escherichia coli pyelonephritis in the BALB/c mouse model of human pyelonephritis

Most human pyelonephritis Escherichia coli isolates express both mannose (MS)- and globoside (Gal-Gal)-binding pili. An ascending E. coli urinary tract infection model was established in the 16-wk-old female BALB/c mouse to compare the pathogenic significance of MS and Gal-Gal pili and their efficacy as vaccines for the prevention of pyelonephritis. The distribution and density of pilus receptor compounds in urogenital tissues and as soluble compounds in urine were determined with antibodies to the synthetic receptor analogues, alpha D-Gal(1----4) beta D-Gal and alpha D-Man(1----2) alpha D-Man. Both carbohydrates were detected in vagina, bladder, ureter, and renal pelvis epithelium and in collecting duct and tubular cells. A pilus receptor compound also was detected in urine. It competitively inhibited the binding capacity of MS pili and was found to be physically, chemically, and immunologically related to Tamm-Horsfall uromucoid. Infectivity and invasiveness were quantitatively and histologically characterized for four E. coli strains: J96, a human pyelonephritis strain that expresses both MS and Gal-Gal pili; two recombinant strains prepared from J96 chromosomal DNA encoding MS pili or Gal-Gal pili; and the nonpiliated K12 recipient. Intravesicular administration of J96 (10(6) colony-forming units [CFU]) resulted in renal colonization and invasion in each of nine mice. The Gal-Gal clone (10(6) CFU) colonized the kidneys in each of 10 mice but did not invade. In contrast, the MS clone (10(6) CFU) did not colonize renal epithelium or invade. This effect was superceded when larger doses (greater than or equal to 10(10) CFU) of the MS clone were administered in volumes that cause acute vesicoureteric reflux. The efficacy was determined of vaccines composed of pure MS or Gal-Gal pili or the lipopolysaccharide containing O somatic antigen of the challenge strain, J96. The Gal-Gal pilus vaccine blocked renal colonization in 19 of 22 mice and renal invasion in 10 of 11 mice. Gal-Gal pili may be useful immunogens for the prevention of pyelonephritis in anatomically normal urinary tracts.

Structural basis of the interaction of the pyelonephritic E. coli adhesin to its human kidney receptor

PapG is the adhesin at the tip of the P pilus that mediates attachment of uropathogenic Escherichia coli to the uroepithelium of the human kidney. The human specific allele of PapG binds to globoside (GbO4), which consists of the tetrasaccharide GalNAc beta 1-3Gal alpha 1-4Gal beta 1-4Glc linked to ceramide. Here, we present the crystal structures of a binary complex of the PapG receptor binding domain bound to GbO4 as well as the unbound form of the adhesin. The biological importance of each of the residues involved in binding was investigated by site-directed mutagenesis. These studies provide a molecular snapshot of a host-pathogen interaction that determines the tropism of uropathogenic E. coli for the human kidney and is critical to the pathogenesis of pyelonephritis.

Verotoxin receptor glycolipid in human renal tissue

Infection with verotoxin producing Escherichia coli has been strongly implicated in the etiology of the hemolytic uremic syndrome (HUS). We have previously shown that this toxin specifically binds to a glycolipid receptor-globotriosyl ceramide (Gb3). We have therefore quantitated the level of this glycolipid by HPLC in human renal cortex and medulla as a function of age. We have also measured the binding of verotoxin to Gb3 isolated from each renal tissue sample. Gb3 was a major component of the glycolipid fraction of all renal samples analyzed. The levels were found to be higher in the cortex than medulla, correlating with the clinical incidence of renal lesions in HUS, but reduced in the kidneys of infants as compared to adults. Verotoxin binding was directly proportional to the renal Gb3 content. Thus, human renal tissue is a rich source of the verotoxin receptor glycolipid. However, changes in receptor concentration cannot explain the age-related incidence of HUS.

Glycolipid core structure switching from globo- to lacto- and ganglio-series during retinoic acid-induced differentiation of TERA-2-derived human embryonal carcinoma cells

We have analyzed the glycolipid markers of a recently cloned human embryonal carcinoma (EC) cell line, NTERA-2, which differentiates extensively into a variety of somatic cell types when exposed to retinoic acid. These tumor cells provide a model system that can be used to study the ontogeny of glycolipid diversity during human embryonic development. Glycolipid antigens were identified by cell surface immunofluorescence and thin-layer chromatography immunostaining using a comprehensive set of anticarbohydrate monoclonal antibodies. Undifferentiated NTERA-2 cells were found to express predominantly globo-series glycolipids, including Gb3, Gb5 (IV3GalGb4), globo-ganglioside (IV3NeuAc alpha 2----3GalGb4), globo-H (IV3Fuc alpha 1----2GalGb4), and globo-A (IV3GalNAc alpha 1----3[Fuc alpha 1----2]GalGb4). When NTERA-2 cells were induced to differentiate by culturing in the presence of 10(-5) M retinoic acid, a remarkable shift of cellular glycolipids from globo-series to lacto- and ganglio-series was observed: Globo-series structures declined, particularly during the period 7-20 days after first exposure to retinoic acid, while lacto-series structures, including fucosyl alpha 1----3 type 2 chain (Lex) and sialosyl type 2 chain, and ganglio-series structures, including GM3, GD3, 9-O-acetyl-GD3, GM2, GD2, and GT3, increased. The presence of globo-A and globo-H as the major ABH blood group antigens in undifferentiated NTERA-2 cells suggests that globo-series blood group antigens are embryonic antigens, synthesis of which switches to lacto-series during human development. Two-color immunofluorescence analysis indicated preferential expression of several ganglio- and lacto-series antigens on different subsets of differentiated cells and permitted the relationship of these subsets to the development of neurons in NTERA-2 cultures to be determined. The results suggest that glycosyltransferase, particularly those involved in controlling glycoconjugate core structure assembly, are key enzymes regulated during the differentiation of human EC cells and, by implication, during human embryogenesis.

Globotriosyl ceramide is specifically recognized by the Escherichia coli verocytotoxin 2

Two Escherichia coli cytotoxins (verotoxins 1 and 2) have been previously implicated in the cytopathology of the Hemolytic Uremic Syndrome. We have examined the glycolipid binding specificity of verotoxin (VT)2. This toxin specifically binds to globotriosyl ceramide (galactose alpha 1-4 galactose beta 1-4 glucosyl ceramide). Removal, or substitution of the terminal a galactose residue with N-acetyl galactosamine in beta 1-3 linkage, deletes binding activity. The toxin does not recognize similar terminal a galactose residues on a glycoglycerolipid. Thus the binding specificity of VT2 is the same as previously reported for VT1. Liposomes containing globotriosyl ceramide are able to specifically remove VT1 and VT2 cytotoxicity and cell lines selected in vitro for resistance to VT1 are cross resistant to VT2.

Tumor-associated glycolipid antigens, their metabolism and organization

A number of experimental animal tumors as well as human cancers have been characterized by dramatic changes of glycolipid composition and metabolism. This review focuses on the chemical and enzymatic basis of the appearance of tumor-associated glycolipid antigens belonging to four major structural classes, i.e., globo, ganglio, lacto type 1, and lacto type 2 series. Some antigens represent the accumulation of precursors with deletion of more complex glycolipids, and others are the result of enhanced synthesis of new structures, most of which are aberrant fucosylation or sialylation or their combination; thus, novel structures such as di- or trimeric Le chi, trifucosyl Le gamma, sialyl Le chi, sialyl dimeric Le chi and disialyl Le alpha A have been isolated and characterized. Many monoclonal antibodies are capable of recognizing antigens in high density but are not capable of reacting with the same antigen in low density. Therefore, the expression of novel structures in high densities at the cell surface is important for recognition of tumor-association antigens. Molecular models of a typical tumor-associated antigen and its organization in membranes are also presented.

Dynamic measurement of the pH of the Golgi complex in living cells using retrograde transport of the verotoxin receptor

The B subunit of verotoxin (VT1B) from enterohemorrhagic Escherichia coli is responsible for the attachment of the holotoxin to the cell surface, by binding to the glycolipid, globotriaosyl ceramide. After receptor-mediated endocytosis, the toxin is targeted to the Golgi complex by a process of retrograde transport. We took advantage of this unique property of VT1B to measure the pH of the Golgi complex in intact live cells. Purified recombinant VT1B was labeled with either rhodamine or fluorescein for subcellular localization by confocal microscopy. After 1 h at 37 degrees C, VT1B accumulated in a juxtanuclear structure that colocalized with several Golgi markers, including alpha-mannosidase II, beta-COP, and NBD-ceramide. Moreover, colchicine and brefeldin A induced dispersal of the juxtanuclear staining, consistent with accumulation of VT1B in the Golgi complex. Imaging of the emission of fluorescein-labeled VT1B was used to measure intra-Golgi pH (pHG), which was calibrated in situ with ionophores. In intact Vero cells, pHG averaged 6.45 +/- 0.03 (standard error). The acidity of the Golgi lumen dissipated rapidly upon addition of bafilomycin A1, a blocker of vacuolar-type ATPases, pHG remained constant despite acidification of the cytosol by reversal of the plasmalemmal Na+/H+ antiport. Similarly, pHG was unaffected by acute changes in cytosolic calcium. Furthermore, pHG recovered quickly toward the basal level after departures imposed with weak bases. These findings suggest that pHG is actively regulated, despite the presence of a sizable H+ "leak" pathway. The ability of VT1B to target the Golgi complex should facilitate not only studies of acid-base regulation, but also analysis of other ionic species.

Structural determination of gangliosides that bind to influenza A, B, and C viruses by an improved binding assay: strain-specific receptor epitopes in sialo-sugar chains

An improved binding assay for detection of ganglioside receptors for influenza A, B, and C viruses was developed. In this system, the virions bound to gangliosides that were developed on a silica gel thin-layer plate were detected by mouse monoclonal antibody against viral hemagglutinin and peroxidase-conjugated anti-mouse immunoglobin. No hydrolysis of the gangliosides by viral receptor-destroying enzyme was detected in the present condition. The reactivity of the viruses to gangliosides depended on the amount of developed gangliosides (10 pmols-10 nmols), the molecular species of sialic acid, and their sugar sequences. Human influenza A (PR/8/34), B (Lee/40), and C (Ann Arbor/1/50) viruses bound different receptor epitopes of sialo-sugar chains of gangliosides. The A/PR/8 virus bound most effectively to Neu5Ac-containing lacto-series gangliosides carrying type I and type II sugar chains, followed by ganglio-series and hematoside-series gangliosides. The A/PR/8 virus weakly bound to Neu5Ac alpha 2,6lactotetraosylceramide [IV6(Neu5Ac)Lc4Cer] and Neu5Ac alpha 2,6paragloboside [IV6(Neu5Ac)nLc4Cer] carrying Neu5Ac alpha 2,6Gal sequence, although their Neu5Ac alpha 2,3Gal derivatives were the most potent gangliosides tested. B/Lee/40 bound restrictively to IV6(Neu5Ac)Lc4Cer and IV6(Neu5Ac)nLc4Cer, which carry Neu5Ac alpha 2,6Gal sequence, and type I and type II lacto-series sugar chain, respectively. C/Ann Arbor/1/50 reacted only with 9-O-Ac-Neu5Ac-carrying sugar chains in all the gangliosides tested. This method also allowed the microanalysis of receptor gangliosides of unknown samples. ESK cells, sensitive to the influenza A viruses infection, expressed several kinds of receptor active gangliosides, while those from ESK-R cells, resistant to the virus infection, were undetectable.

Expression cloning of a new member of the ABO blood group glycosyltransferases, iGb3 synthase, that directs the synthesis of isoglobo-glycosphingolipids

The large array of different glycolipids described in mammalian tissues is a reflection, in part, of diverse glycosyltransferase expression. Herein, we describe the cloning of a UDP-galactose: beta-d-galactosyl-1,4-glucosylceramide alpha-1, 3-galactosyltransferase (iGb(3) synthase) from a rat placental cDNA expression library. iGb(3) synthase acts on lactosylceramide, LacCer (Galbeta1,4Glcbeta1Cer) to form iGb(3) (Galalpha1,3Galbeta1, 4Glcbeta1Cer) initiating the synthesis of the isoglobo-series of glycosphingolipids. The isolated cDNA encoded a predicted protein of 339 amino acids, which shows extensive homology (40-50% identity) to members of the ABO gene family that includes: murine alpha1, 3-galactosyltransferase, Forssman (Gb(5)) synthase, and the ABO glycosyltransferases. In contrast to the murine alpha1, 3-galactosyltransferase, iGb(3) synthase preferentially modifies glycolipids over glycoprotein substrates. Reverse transcriptase-polymerase chain reaction revealed a widespread tissue distribution of iGb(3) synthase RNA expression, with high levels observed in spleen, thymus, and skeletal muscle. As an indirect consequence of the expression cloning strategy used, we have been able to identify several potential glycolipid biosynthetic pathways where iGb(3) functions, including the globo- and isoglobo-series of glycolipids.

Humans lack iGb3 due to the absence of functional iGb3-synthase: implications for NKT cell development and transplantation

The glycosphingolipid isoglobotrihexosylceramide, or isogloboside 3 (iGb3), is believed to be critical for natural killer T (NKT) cell development and self-recognition in mice and humans. Furthermore, iGb3 may represent an important obstacle in xenotransplantation, in which this lipid represents the only other form of the major xenoepitope Galα(1,3)Gal. The role of iGb3 in NKT cell development is controversial, particularly with one study that suggested that NKT cell development is normal in mice that were rendered deficient for the enzyme iGb3 synthase (iGb3S). We demonstrate that spliced iGb3S mRNA was not detected after extensive analysis of human tissues, and furthermore, the iGb3S gene contains several mutations that render this product nonfunctional. We directly tested the potential functional activity of human iGb3S by expressing chimeric molecules containing the catalytic domain of human iGb3S. These hybrid molecules were unable to synthesize iGb3, due to at least one amino acid substitution. We also demonstrate that purified normal human anti-Gal immunoglobulin G can bind iGb3 lipid and mediate complement lysis of transfected human cells expressing iGb3. Collectively, our data suggest that iGb3S is not expressed in humans, and even if it were expressed, this enzyme would be inactive. Consequently, iGb3 is unlikely to represent a primary natural ligand for NKT cells in humans. Furthermore, the absence of iGb3 in humans implies that it is another source of foreign Galα(1,3)Gal xenoantigen, with obvious significance in the field of xenotransplantation.

Identification of endogenous antigens that regulate natural killer T (NKT) cell development and function is a major goal in immunology. Originally the glycosphingolipid, iGb3, was suggested to be the main endogenous ligand in both mice and humans. However, recent studies have challenged this hypothesis. From a xenotransplantation (animal to human transplants) perspective, iGb3 expression is also important as it represents another form of the major xenoantigen Galα(1,3)Gal. In this study, we assessed whether humans expressed a functional iGb3 synthase (iGb3S), the enzyme responsible for lipid synthesis. We showed that spliced iGb3S mRNA was not detected in any human tissue analysed. Furthermore, chimeric molecules composed of the catalytic domain of human iGb3S were unable to synthesize iGb3 lipid, due to at least one amino acid substitution. We also demonstrated that purified human anti-Gal antibodies bound iGb3 lipid and mediated destruction of cells transfected to express iGb3. A nonfunctional iGb3S in humans has two major consequences: (1) iGb3 is unlikely to be a natural human NKT ligand and (2) natural human anti-Gal antibodies in human serum could react with iGb3 on the surface of organs from pigs, marking these tissues for immunological destruction.

Controversy surrounds the glycolipid iGb3. Our data show that humans do not express this lipid. This has important implications in natural killer T cell development, self-recognition, and transplantation.

The receptor repertoire defines the host range for attaching Escherichia coli strains that recognize globo-A

Escherichia coli strains which colonize the human urinary tract express lectins specific for different members of the globoseries of glycolipids, e.g., globotetraosylceramide and globo-A. This study investigated the importance of globo-A expression for attachment to human uroepithelial cells, colonization of the urinary tract, and severity of urinary tract infection. The expression of receptor-active glycolipids by erythrocytes and epithelial cells was analyzed by thin-layer chromatography and bacterial overlay as well as by bacterial binding to those cells. The epithelial expression of the globo-A receptor was restricted to individuals of blood group A with a positive secretor state. Consequently, globo-A binding E. coli strains attached only to epithelial cells from these individuals. In contrast, globoside-recognizing strains attached in similar numbers to uroepithelial cells regardless of the ABH blood group and secretor state of the donor. The role of host receptor expression for infection with globo-A-specific E. coli was analyzed in 1,473 children with urinary tract infections. All those infected with strains exclusively expressing globo-A-specific adhesins were found to be of blood group A, compared with 45% in the population at large (P less than 0.006). The inflammatory response (fever, C-reactive protein, erythrocyte sedimentation rate) of individuals infected with these strains was lower than that in individuals with infections caused by globoside binding strains. The results demonstrate the importance of fitness between host receptors and bacterial adhesins for infection and suggest that minor receptor epitope differences have profound effects on the disease process.

The Niemann-Pick type C2 protein loads isoglobotrihexosylceramide onto CD1d molecules and contributes to the thymic selection of NKT cells

The Niemann-Pick type C2 (NPC2) protein is a small, soluble, lysosomal protein important for cholesterol and sphingolipid transport in the lysosome. The immunological phenotype of NPC2-deficient mice was limited to an impaired thymic selection of Vα14 natural killer T cells (NKT cells) and a subsequent reduction of NKT cells in the periphery. The remaining NKT cells failed to produce measurable quantities of interferon-γ in vivo and in vitro after activation with α-galactosylceramide. In addition, thymocytes and splenocytes from NPC2-deficient mice were poor presenters of endogenous and exogenous lipids to CD1d-restricted Vα14 hybridoma cells.

Importantly, we determined that similar to saposins, recombinant NPC2 was able to unload lipids from and load lipids into CD1d. This transfer activity was associated with a dimeric form of NPC2, suggesting a unique mechanism of glycosphingolipid transfer by NPC2. Similar to saposin B, NPC2 dimers were able to load isoglobotrihexosylceramide (iGb3), the natural selecting ligand of NKT cells in the thymus, into CD1d. These observations strongly suggested that the phenotype observed in NPC2-deficient animals was directly linked to the efficiency of the loading of iGb3 into CD1d molecules expressed by thymocytes. This conclusion was supported by the rescue of endogenous and exogenous iGb3 presentation by recombinant NPC2. Thus, the loading of endogenous and exogenous lipids and glycolipids onto CD1d is dependent on various small, soluble lipid transfer proteins present in the lysosome.

Crystal structures of mouse CD1d-iGb3 complex and its cognate Valpha14 T cell receptor suggest a model for dual recognition of foreign and self glycolipids

The semi-invariant Valpha14Jalpha18 T cell receptor (TCR) is expressed by regulatory NKT cells and has the unique ability to recognize chemically diverse ligands presented by CD1d. The crystal structure of CD1d complexed to a natural, endogenous ligand, isoglobotrihexosylceramide (iGb3), illustrates the extent of this diversity when compared to the binding of potent, exogenous ligands, such as alpha-galactosylceramide (alpha-GalCer). A single mode of recognition for these two classes of ligands would then appear problematic for a single T cell receptor. However, the Valpha14 TCR adopts two different conformations in the crystal where, in one configuration, the presence of a larger cavity between the two CDR3 regions could accommodate iGb3 and, in the other, a smaller cavity fits alpha-GalCer more snugly. Alternatively, the extended iGb3 headgroup could be "squashed" upon docking of the TCR and accommodated between the CD1 and TCR surfaces. Thus, the same TCR may adopt alternative modes of recognition for these foreign and self-ligands for NKT cell activation.

Globoside-dependent adhesion of human embryonal carcinoma cells, based on carbohydrate-carbohydrate interaction, initiates signal transduction and induces enhanced activity of transcription factors AP1 and CREB

Undifferentiated human embryonal carcinoma cells are characterized by high expression of lactoneotetraosylceramide (nLc4), globoside (Gb4), and extended globo-series glycosphingolipids (GSLs) termed "stage-specific embryonic antigens 3 and 4" (SSEA-3 and -4). Expression of these GSLs declines in association with a decline of homotypic adhesion during the differentiation process. Therefore, these GSLs may play an essential role in adhesion among these cells. As an example, human embryonal carcinoma 2102 cells display strong adhesion to plates coated with Gb4 ("Gb4-dependent cell adhesion"). This adhesion, which simulates homotypic 2102 cell aggregation, is based on interaction between Gb4 and nLc4, or between Gb4 and GalGb4 (IV3GalGb4; the major SSEA-3 epitope), as indicated by the following observations: (i) adhesion of 2102 cells or GSL-liposomes to GSL-coated plates in various combinations; (ii) inhibition of Gb4-dependent 2102 cell adhesion by preincubation of cells with anti-SSEA-3 or anti-nLc4 antibodies, or by pretreatment of Gb4-coated plates with aqueous micellar solution of nLc4 or GalGb4; (iii) decline of the cell adhesion in association with retinoic acid-induced differentiation, whereby SSEA-3 and nLc4 levels are reduced. Since cell adhesion is an essential prerequisite for induction of differentiation, as observed at each step of embryogenesis, expression of seven transcription factors following adhesion of 2102 cells to Gb4-coated plates, and to detergent-insoluble substrate adhesion matrix prepared from 2102 cells, were studied. In both types of adhesion, a strong enhancement of AP1 and CREB site binding activity was observed during the early stage (15-60 min following initial adhesion). Although 2102 cells showed strong adhesion to Gg3-coated plates, based on interaction between Gg3 and Gb4, adhesion of the cells to Gg3 did not cause changes of AP1 and CREB activity. No other transcription factors showed changes induced by Gg3- or Gb4-dependent adhesion.

Escherichia coli K99 binds to N-glycolylsialoparagloboside and N-glycolyl-GM3 found in piglet small intestine

Escherichia coli K12, which possess the K99 plasmid and synthesize K99 fimbriae (E. coli K99), cause severe neonatal diarrhea in piglets, calves, and lambs but not in humans. The organism binds specifically and with high affinity to only two glycolipids in piglet intestinal mucosa as demonstrated by overlaying glycolipid chromatograms with 125I-labeled bacteria. These glycolipids, which are N-glycolyl-GM3 (NeuGc alpha 2-3Gal beta 1-4Glc beta 1-1Cer) and N-glycolylsialoparagloboside (NeuGc alpha 2-3Gal beta 1-4GlcNAc beta 1-3Gal beta 1-4Glc beta 1-1Cer), occur at about 13 and 0.3 micrograms per gram wet weight of mucosa, respectively. E. coli K99 grown at 18 degrees C, a temperature at which the K99 fimbriae are not expressed, do not bind to these glycolipids. Of the standard glycolipids tested in solid phase binding assays, E. coli K99 binds with highest affinity to N-glycolylsialoparagloboside, with less affinity to N-glycolyl-GM3, and with very low affinity to N-acetylsialoparagloboside. The bacteria do not bind to GM3 (NeuAc alpha 2-3Gal beta 1-4Glc beta 1-1Cer), GM2 (GalNAc beta 1-4[Neu-Ac alpha 2-3]Gal beta 1-4Glc beta 1-1Cer), GM1 (Gal beta 1-3GalNAc beta 1-4[NeuAc alpha 2-3]Gal beta 1-4Glc beta 1-1Cer), or several other N-acetylsialic acid-containing gangliosides and neutral glycolipids at the levels tested. N-Glycolylsialyl residues are found in the glycoproteins and glycolipids of piglets, calves, and lambs but not in the glycoproteins and glycolipids of humans. Possibly this distribution of sialyl derivatives explains the host range of infection by the organism.

It's a lipid's world: bioactive lipid metabolism and signaling in neural stem cell differentiation

Lipids are often considered membrane components whose function is to embed proteins into cell membranes. In the last two decades, studies on brain lipids have unequivocally demonstrated that many lipids have critical cell signaling functions; they are called "bioactive lipids". Pioneering work in Dr. Robert Ledeen's laboratory has shown that two bioactive brain sphingolipids, sphingomyelin and the ganglioside GM1 are major signaling lipids in the nuclear envelope. In addition to derivatives of the sphingolipid ceramide, the bioactive lipids discussed here belong to the classes of terpenoids and steroids, eicosanoids, and lysophospholipids. These lipids act mainly through two mechanisms: (1) direct interaction between the bioactive lipid and a specific protein binding partner such as a lipid receptor, protein kinase or phosphatase, ion exchanger, or other cell signaling protein; and (2) formation of lipid microdomains or rafts that regulate the activity of a group of raft-associated cell signaling proteins. In recent years, a third mechanism has emerged, which invokes lipid second messengers as a regulator for the energy and redox balance of differentiating neural stem cells (NSCs). Interestingly, developmental niches such as the stem cell niche for adult NSC differentiation may also be metabolic compartments that respond to a distinct combination of bioactive lipids. The biological function of these lipids as regulators of NSC differentiation will be reviewed and their application in stem cell therapy discussed.

The globoseries glycosphingolipid sialosyl galactosyl globoside is found in urinary tract tissues and is a preferred binding receptor In vitro for uropathogenic Escherichia coli expressing pap-encoded adhesins

Women with a history of recurrent Escherichia coli urinary tract infections (UTIs) are significantly more likely to be nonsecretors of blood group antigens than are women without such a history, and vaginal epithelial cells (VEC) from women who are nonsecretors show enhanced adherence of uropathogenic E. coli isolates compared with cells from secretors. We previously extracted glycosphingolipids (GSLs) from native VEC and determined that nonsecretors (but not secretors) selectively express two extended globoseries GSLs, sialosyl galactosyl globoside (SGG) and disialosyl galactosyl globoside (DSGG), which specifically bound uropathogenic E. coli R45 expressing a P adhesin. In this study, we demonstrated, by purifying the compounds from this source, that SGG and DSGG are expressed in human kidney tissue. We also demonstrated that SGG and DSGG isolated from human kidneys bind uropathogenic E. coli isolates expressing each of the three classes of pap-encoded adhesins, including cloned isolates expressing PapG from J96, PrsG from J96, and PapG from IA2, and the wild-type isolates IA2 and R45. We metabolically 35S labeled these five E. coli isolates and measured their relative binding affinities to serial dilutions of SGG and DSGG as well as to globotriaosylceramide (Gb3) and globotetraosylceramide (Gb4), two other globoseries GSLs present in urogenital tissues. Each of the five E. coli isolates bound to SGG with the highest apparent avidity compared with their binding to DSGG, Gb3, and Gb4, and each isolate had a unique pattern of GSL binding affinity. These studies further suggest that SGG likely plays an important role in the pathogenesis of UTI and that its presence may account for the increased binding of E. coli to uroepithelial cells from nonsecretors and for the increased susceptibility of nonsecretors to recurrent UTI.

Glycolipid-lectin interactions: reactivity of lectins from Helix pomatia, Wisteria floribunda, and Dolichos biflorus with glycolipids containing N-acetylgalactosamine

The autoradiographic detection of 125I-labeled lectins binding to glycolipids on thin-layer chromatograms can be used to rapidly analyze total glycolipid extracts of cells or tissues for specific oligosaccharide structures. The Helix pomatia lectin which binds with high affinity to terminal alpha-linked GalNAc residues did not bind to globoside (terminal beta 1-3GalNAc) but did bind the ganglioside GM2 and its asialo derivative which have terminal beta 1-4GalNAc residues. The lectin from Dolichos biflorus bound specifically to the Forssman glycolipid with relatively low affinity. The lectin from Wisteria floribunda was bound to Forssman glycolipid, globoside, and the asialo derivative of the ganglioside GM2. The interactions of these lectins with the glycolipid-derived, 3H-labeled oligosaccharides was also analyzed by affinity chromatography. The results indicated that the reactivity of multivalent carbohydrate-binding proteins with polyvalent surfaces of glycolipids is strong enough to permit detection of low-affinity interactions that may not be observed in binding assays that are based on carbohydrate-protein interactions in solution. The autoradiographic analysis of 125I-Helix pomatia lectin binding to thin-layer chromatograms of total lipid extracts from human erythrocyte membranes detected the quantitative differences in the A-active glycolipids from type A1 and A2 cells.

Variability in the binding of anti-MAG and anti-SGPG antibodies to target antigens in demyelinating neuropathy and IgM paraproteinemia

Densitometry of immunostained Western blots or thin layer chromatograms and enzyme-linked immunosorbent assays (ELISAs) were used to compare the relative strengths of IgM binding to myelin-associated glycoprotein (MAG), P0 glycoprotein, peripheral myelin protein-22 (PMP-22), sulfate-3-glucuronyl paragloboside (SGPG), and other potential target antigens in a series of eleven patients with sensory or sensorimotor demyelinating neuropathy and IgM paraproteinemia. The IgM from all patients exhibited reactivity with both MAG and SGPG, and there was a statistically significant correlation between the overlay assays and ELISAs for measuring the strength of IgM binding to MAG and to SGPG. However, the data revealed variations in the relative strengths with which the antibodies bound to the potential target antigens and heterogeneity in their fine specificities. First, there was a poor correlation between the strength of binding to MAG and to SGPG, respectively. Second, reactivity with MAG or SGPG in a few of the patients was only detected by one of the two assay systems. Third, about one-third of the patients' IgM absolutely required the sulfate on SGPG for reactivity, whereas the others retained some reactivity after removal of the sulfate. Fourth, IgM from two of the patients exhibited unusually strong reactivity with the proteins of compact myelin, P0 and PMP22. These relative differences in strengths of antibody binding to the potential antigens were compared with the patients' clinical presentations and with their responses to intravenous immunoglobulin (IVIg) therapy in a clinical trial in which they participated. For the most part, these variations did not correlate with clinical presentation, which was relatively homogeneous in this series of patients. However, an inverse relationship was noted between degree of reactivity to MAG by ELISA and response to IVIg. Two of the patients who responded had only mild elevations of IgM antibodies to nerve glycoconjugates and exhibited some unusual immunochemical and clinical characteristics in comparison to the other patients. The results demonstrate differences in the relative strengths with which anti-MAG and anti-SGPG IgM antibodies from different patients bind to potential neural target antigens which may affect pathogenic mechanisms and response to therapy.

Alteration of the glycolipid binding specificity of the pig edema toxin from globotetraosyl to globotriaosyl ceramide alters in vivo tissue targetting and results in a verotoxin 1-like disease in pigs

All members of the verotoxin (VT) family specifically recognize globo-series glycolipids on the surface of susceptible cells. Those toxins that are associated with human disease, VT1, VT2, and VT2c, bind to globotriaosyl ceramide (Gb3) while VT2e, which is associated with edema disease of swine, binds preferentially to globotetraosyl ceramide (Gb4). We were recently able to identify, using site-directed mutagenesis, amino acids in the binding subunit of these toxins that are important in defining their glycosphingolipid (GSL) binding specificity (Tyrrell, G. J., K. Ramotar, B. Boyd, B. W. Toye, C. A. Lingwood, and J. L. Brunton. 1992. Proc. Natl. Acad. Sci. USA. 89:524). The concomitant mutation of Gln64 and Lys66 in the VT2e binding subunit to the corresponding residues (Glu and Gln, respectively) found in VT2 effectively converted the GSL binding specificity of the mutant toxin from Gb4 to Gb3 in vitro. We now report that the altered carbohydrate recognition of the mutant toxin (termed GT3) has biological significance, resulting in a unique disease after intravascular injection into pigs as compared with classical VT2e-induced edema disease. The tissue localization of radiolabeled GT3 after intravascular injection was elevated in neural tissues compared with VT2e accumulation, while localization of GT3 to the gastrointestinal tract was relatively reduced. Accordingly, the pathological lesions after challenge with GT3 involved gross edema of the cerebrum, cerebellum, and brain stem, while purified VT2e caused hemorrhage and edema of the cerebellum, and submucosa of the stomach and large intestine. In addition, both radiolabeled toxins bound extensively to tissues not directly involved in the pathology of disease. VT2e, unlike GT3 or VT1, bound extensively to red cells, which have high levels of Gb4. The overall tissue distribution of VT2e was thus found to be influenced by regional blood flow to each organ and not solely by the Gb4 levels of these tissues. Conversely, the distribution of GT3 (and VT1), which cleared more rapidly from the circulation, correlated with respective tissue Gb3 levels rather than blood flow. These studies indicate the primary role of carbohydrate binding specificity in determining systemic pathology, suggest that the red cells act as a toxin carrier in edema disease, and indicate that red cell binding does not protect against the pathology of systemic verotoxemia.

Receptor-specific agglutination tests for detection of bacteria that bind globoseries glycolipids

Specific binding to the globoseries of glycolipid receptors explains the adherence of uropathogenic Escherichia coli to host cells. The minimal receptor disaccharide Gal alpha 1----4Gal beta [galactose alpha (1----4)galactose beta] is recognized by most attaching clinical isolates. However, wild-type isolates can express adhesins with several different receptor specificities. Bioassays do not permit separate analysis of each receptor specificity, since the target cells contain multiple potentially receptor-active molecules. In this study, bacterial adhesins were analyzed by using receptors immobilized into latex beads in one of two ways. In one way, di- and trisaccharides were covalently linked via a spacer arm to latex beads coupled with bovine serum albumin. In the other way, receptor-active glycolipids were coated onto the bovine serum albumin-latex beads. The latex beads were subsequently used for agglutination by using type strains with known receptor specificity. The composition was optimized regarding receptor structure and size of latex beads. Gal alpha 1----4Gal beta was as active as the trisaccharide derivative Gal alpha 1----4Gal beta 1----3glucose or Gal alpha 1----4Gal beta 1----3glucosamine. Among the natural glycolipids tested, globotetraosylceramide was the most active. Subsequently, the sensitivity and specificity of the Gal alpha 1----4Gal beta-latex and globotetraosylceramide-latex reagents were compared for 733 E. coli urinary isolates. Hemagglutination of human erythrocytes was used as the positive standard. No significant difference in the specificity or sensitivity of the latex reagents was found; the sensitivity ranged from 86%, when isolates agglutinating human erythrocytes of blood groups P1 and p were included, to 93%, when those isolates agglutinating erythrocytes of blood group p were excluded. These reagents provide tools for bacterial identification in patients with urinary tract infection.