Determination of the upper size limit for uptake and processing of ligands by the asialoglycoprotein receptor on hepatocytes in vitro and in vivo

The asialoglycoprotein receptor (ASGPr) on hepatocytes plays a role in the clearance of desialylated proteins from the serum. Although its sugar preference (N-acetylgalactosamine (GalNAc) > galactose) and the effects of ligand valency (tetraantennary > triantennary > diantennary > monoantennary) and sugar spacing (20 A 10 A 4 A) are well documented, the effect of particle size on recognition and uptake of ligands by the receptor is poorly defined. In the present study, we assessed the maximum ligand size that still allows effective processing by the ASGPr of mouse hepatocytes in vivo and in vitro. Here too, we synthesized a novel glycolipid, which possesses a highly hydrophobic steroid moiety for stable incorporation into liposomes, and a triantennary GalNAc(3)-terminated cluster glycoside with a high nanomolar affinity (2 nm) for the ASGPr. Incorporation of the glycolipid into small (30 nm) [(3)H]cholesteryl oleate-labeled long circulating liposomes (1-50%, w/w) caused a concentration-dependent increase in particle clearance that was liver-specific (reaching 85 +/- 7% of the injected dose at 30 min after injection) and mediated by the ASGPr on hepatocytes, as shown by competition studies with asialoorosomucoid in vivo. By using glycolipid-laden liposomes of various sizes between 30 and 90 nm, it was demonstrated that particles with a diameter of >70 nm could no longer be recognized and processed by the ASGPr in vivo. This threshold size for effective uptake was not related to the physical barrier raised by the fenestrated sinusoidal endothelium, which shields hepatocytes from the circulation, because similar results were obtained by studying the uptake of liposomes on isolated mouse hepatocytes in vitro. From these data we conclude that in addition to the species, valency, and orientation of sugar residues, size is also an important determinant for effective recognition and processing of substrates by the ASGPr. Therefore, these data have important implications for the design of ASGPr-specific carriers that are aimed at hepatocyte-directed delivery of drugs and genes.     

Influence of type of linkage and spacer on the interaction of beta-galactoside-binding proteins with immobilized affinity ligands

Affinity chromatography provides a powerful tool for isolation of carbohydrate-binding proteins. However, the choice of the ligand and spacer has an important impact on effectiveness. The influence of several different ligands on qualitative and quantitative aspects of the purification of two beta-galactoside-specific lectins has been evaluated. Sepharose was modified by coupling four types of neoglycoproteins (galactosylated or lactosylated bovine serum albumin with increasing sugar content) and two naturally occurring asialoglycoproteins at similar densities. Carbohydrate ligands at essentially equal density were made accessible to the lectins by seven commonly used methods. The yield of mistletoe lectin was high when lactosylated neoglycoproteins were used for separation. For these resins the sugar incorporation exceeded 10 sugar groups per protein carrier molecule. The yield was similarly high with the asialoglycoproteins and with lactose; the sugar was coupled to the resin as a p-aminophenyl derivative or by means of divinyl sulfone activation. An epoxy group in linkages of galactose or lactose decreased the binding capacity. A quantitatively similar degree of protein yields was obtained for the beta-galactoside-binding protein of bovine heart, although different proteins were obtained when neoglycoproteins were used as ligand. The nature of the affinity ligand in lectin purification can increase the yield and may also influence the profile of the carbohydrate-binding proteins.     

Evidence for transmembrane modulation of the ligand-binding site of the hepatocyte galactose/N-acetylgalactosamine-specific receptor

The ligand-binding activity of the galactose/N-acetylgalactosamine-specific receptor (Gal/GalNAc receptor) present on the surface of hepatocytes can be modulated under a number of conditions in the intact cell. The carboxylic acid ionophores monensin and nigericin inhibit endocytosis by the Gal/GalNAc receptor in a concentration-dependent manner. Monensin at a concentration of 100 microM reduces the number of binding sites for asialo-orosomucoid and a tri-branched glycopeptide (F2) 5-10-fold; however, the number of Gal/GalNAc receptor subunits detected at the cell surface by a competitive radioimmunoassay and by immunoprecipitation of surface labeled receptor is not significantly altered. Replacement of NaCl in the medium with either N-methylglucamine or sorbitol to isotonicity also inhibits binding and endocytosis. The monensin, nigericin, N-methylglucamine, and sorbitol treatments have in common the ability to alkalinize the cytosol of the hepatocyte. None of these agents has any effect on binding by the isolated Gal/GalNAc receptor nor is the intracellular pH shift of such a magnitude that it would alter binding by the isolated Gal/GalNAc receptor. This has led us to conclude that the ligand-binding properties of the Gal/GalNAc receptor at the cell surface can be modulated in a transmembrane fashion by events other than those involving pH or Ca2+ regulation at the ligand-binding site itself. Such transmembrane modulation of ligand binding by the Gal/GalNAc receptor may provide a rapid and efficient mechanism for mediating ligand release and immediate return of the receptor to the cell surface.     

Endocytosis via galactose receptors in vivo : Ligand size directs uptake by hepatocytes and/or liver macrophages

We followed the intrahepatic binding and uptake of variously sized ligands with terminal galactosyl residues in rat livers. The ligands were administered to prefixed livers in binding studies and in vivo and in situ (serum-free perfused livers) in uptake studies. Gold sols with different particle diameters were prepared: 5 nm (Au5), 17 nm (Au17), 50 nm (Au50) and coated with galactose exposing glycoproteins (asialofetuin (ASF) or lactosylated BSA (LacBSA)). Electron microscopy of mildly prefixed livers perfused with LacBSA-Au5 in serum-free medium showed ligand binding to liver macrophages, hepatocytes and endothelial cells. Ligands bound to prefixed cell surfaces reflect the initial distribution of receptor activity: pre-aggregated clusters of ligands are found on liver macrophages, single particles statistically distributed on hepatocytes and pre-aggregated clusters of particles restricted to coated pits on endothelial cells. Ligand binding is prevented in the presence of 80 mM N-acetylgalactosamine (GalNAc), while N-acetylglucosamine (GlcNAc) is without effect. Electron microscopy of livers after ligand injection into the tail vein shows that in vivo uptake of electron-dense galactose particles by liver cells is size-dependent. Using a LacBSA-Au preparation with heterogeneous particle diameter (2.2-11.7 nm) we found that hepatocytes take up only ligands up to the size of 7.8 nm, whereas particles of all sizes available in this experiment are found in liver macrophages and endothelial cells. ASF-Au17 and LacBSA-Au17 are endocytosed by liver macrophages and endothelial cells, but not by hepatocytes. ASF-Au50 is taken up by liver macrophages only. In vivo uptake by liver macrophages is mediated by galactose-specific recognition as shown by inhibition with GalNAc. Some 52-65% inhibition was measured in in vivo experiments and 78% inhibition in in situ experiments. GlNAc showed no inhibitory effect. Furthermore, we measured uptake of [125J]ASF and of [125J]ASF adsorbed to Au17 by the different cell populations of rat livers in vivo. While the bulk of the molecular ligand is found in the hepatocyte fraction, the particulate ligand is located in the sinusoidal fraction.     

Use of hydrophilic diamines for bridging of two opioid peptide pharmacophores. Synthesis and receptor binding of two new analogues.

The bivalent ligand approach, which assumes that two pharmacophores are connected by a spacer, was used to design receptor type-selective ligands for opioid receptors. The first two opioid peptide bivalent ligands with different spacer lengths containing different numbers of hydroxyl groups, (Tyr-D-Ala-Gly-Phe-NH-CH2-CHOH-)2 (Tyr-D-Ala-Gly-Phe-NH-CH2-CHOH-CHOH-)2, were synthesized and their binding to mu, delta, and kappa opioid receptors was characterized. Both analogues were found to possess high opioid in vitro activities. The length of the hydrophilic spacer does not affect the affinity for delta receptors, whereas shorter spacer length increases affinity for mu and even more so for kappa receptors. Thus receptor type-selective peptides for opioid receptors can be designed using the bivalent approach.     

Cooperation between lateral ligand mobility and accessibility for receptor recognition in selectin-induced cell rolling

Selectin-induced leukocyte rolling along the endothelial surface is an essential step in the immune response. Several in vitro studies showed that this cell rolling is a highly regulated adhesion phenomenon, controlled by the kinetics and forces of selectin-ligand interactions. In the flow chamber study presented here, we focused on the requirements on the ligand structure in this context. A series of neoglycolipids bearing the binding epitope Sialyl Lewis X was synthesized and used as artificial ligands. These lipids differed in their spacer structures between headgroup and membrane anchor, resulting in a gradual variation in accessibility and mobility of the binding epitope when immobilized in model membranes. Consequently, analysis of cell rolling along such membranes allowed correlation of ligand structures and functionality. All model membranes containing such ligands were further characterized by film balance measurements, epifluorescence, and atomic force microscopy. Generally, the glycolipids exhibited a high tendency for lateral aggregation, but the resulting clusters were of different morphology. This was also reflected by strong differences in the rolling experiments. Our results confirm that, in addition to a sufficient headgroup accessibility, the cell rolling process is governed by two further interdependent factors: (i) the headgroup flexibility caused by the intramolecular uncoupling between the headgroup and the hydrophobic moiety due to introduction of a spacer, and (ii) the stiffness of the molecules resulting from their supramolecular arrangement in clustered assemblies. Since both factors are influenced simultaneously by the spacer modification, we present for the first time a clear correlation between structural aspects of selectin ligands and their ability to mediate cell rolling. This might help to develop a better understanding for the function of the natural selectin ligands.     

Role of N-acetyl-D-galactosamine residue on B151K12-derived T cell-replacing factor (B151-TRF) molecule in B cell-receptor binding and -stimulating activity

The role of sugar moiety on T cell-replacing factor molecule derived from a monoclonal T cell hybridoma B151K12 (B151-TRF) was analyzed with respect to the interaction with receptor on B cells. The induction of B cell differentiation into Ig-secreting cells by B151-TRF was specifically inhibited by addition of N-acetyl-D-galactosamine (GalNAc) to culture. Such inhibition appeared to be attributed to the interference of GalNAc in the interaction of TRF with its receptor, because absorption of TRF activity with B cells was notably inhibited by the presence of GalNAc. To substantiate this point further, we established binding assay of B151-TRF molecule to the receptor on B cells by using 125I-labeled TRF fraction enriched by reversed-phase high-performance liquid chromatography and gel filtration. The results revealed that the binding of 125I-TRF molecule to the B cells was almost completely blocked by GalNAc. Moreover, the existence of GalNAc residue(s) on B151-TRF molecule was evidenced by the facts that 1) the TRF activity was eluted from lectin gels with specificity for GalNAc as revealed by the functional assay, and 2) the 125I-TRF molecule specifically bound to such lectin gels. Thus, the GalNAc residue(s) on B151-TRF molecule plays an important role in binding of TRF molecule to the receptor and in the stimulation of B cells. The molecular properties of B cell-stimulatory B151-TRF and its mode of interaction with corresponding receptor on B cells were discussed in the context of B151-TRF as a glycosylated lymphokine molecule and B151-TRF receptor as a carbohydrate-binding protein (animal lectin).     

Structural changes of the Cry1Ac oligomeric pre-pore from bacillus thuringiensis induced by N-acetylgalactosamine facilitates toxin membrane insertion

The primary action of Cry toxins produced by Bacillus thuringiensis is to lyse midgut epithelial cells in their target insect by forming lytic pores. The toxin-receptor interaction is a complex process, involving multiple interactions with different receptor and carbohydrate molecules. It has been proposed that Cry1A toxins sequentially interact with a cadherin receptor, leading to the formation of a pre-pore oligomer structure, and that the oligomeric structure binds to glycosylphosphatidyl-inositol-anchored aminopeptidase-N (APN) receptor. The Cry1Ac toxin specifically recognizes the N-acetylgalactosamine (GalNAc) carbohydrate present in the APN receptor from Manduca sexta larvae. In this work, we show that the Cry1Ac pre-pore oligomer has a higher binding affinity with APN than the monomeric toxin. The effects of GalNAc binding on the toxin structure were studied in the monomeric Cry1Ac, in the soluble pre-pore oligomeric structure, and in its membrane inserted state by recording the fluorescence status of the tryptophan (W) residues. Our results indicate that the W residues of Cry1Ac have a different exposure to the solvent when compared with that of the closely related Cry1Ab toxin. GalNAc binding specifically affects the exposure of W545 in the pre-pore oligomer in contrast to the monomer where GalNAc binding did not affect the fluorescence of the toxin. These results indicate a subtle conformational change in the GalNAc binding pocket in the pre-pore oligomer that could explain the increased binding affinity of the Cry1Ac pre-pore to APN. Although our analysis did not reveal major structural changes in the pore-forming domain I upon GalNAc binding, it showed that sugar interaction enhanced membrane insertion of soluble pre-pore oligomeric structure. Therefore, the data presented here permits to propose a model in which the interaction of Cry1Ac pre-pore oligomer with APN receptor facilitates membrane insertion and pore formation.     

Mechanism of pH-dependent N-acetylgalactosamine binding by a functional mimic of the hepatocyte asialoglycoprotein receptor

Efficient release of ligands from the Ca(2+)-dependent carbohydrate-recognition domain (CRD) of the hepatic asialoglycoprotein receptor at endosomal pH requires a small set of conserved amino acids that includes a critical histidine residue. When these residues are incorporated at corresponding positions in an homologous galactose-binding derivative of serum mannose-binding protein, the pH dependence of ligand binding becomes more like that of the receptor. The modified CRD displays 40-fold preferential binding to N-acetylgalactosamine compared with galactose, making it a good functional mimic of the asialoglycoprotein receptor. In the crystal structure of the modified CRD bound to N-acetylgalactosamine, the histidine (His(202)) contacts the 2-acetamido methyl group and also participates in a network of interactions involving Asp(212), Arg(216), and Tyr(218) that positions a water molecule in a hydrogen bond with the sugar amide group. These interactions appear to produce the preference for N-acetylgalactosamine over galactose and are also likely to influence the pK(a) of His(202). Protonation of His(202) would disrupt its interaction with an asparagine that serves as a ligand for Ca(2+) and sugar. The structure of the modified CRD without sugar displays several different conformations that may represent structures of intermediates in the release of Ca(2+) and sugar ligands caused by protonation of His(202).     

Studies on the mechanism of enkephalin receptor down regulation

The association of [3H] [D-Ala2, D-Leu5] enkephalin ([3H]DADLE]) with mouse neuroblastoma cells (N4TG1) was investigated. Under identical conditions the time course, dose response curve and temperature dependence for ligand uptake were similar to those for ligand-induced receptor loss (down regulation). Uptake of [3H]DADLE was inhibited by opiate ligands as well as by the metabolic inhibitors sodium azide and 2,4 dinitrophenol. Comparison of the effects of these inhibitors on receptor binding, ligand uptake and receptor loss indicated that these cells accumulate [3H]DADLE in excess of their surface receptor number. The data suggest that receptor recycling occurs and that ligand is internalized via receptor mediated endocytosis.     

Neurexin/Neuroligin Interaction Kinetics Characterized by Counting Single Cell-Surface Attached Quantum Dots

We report what to our knowledge is a new method to characterize kinetic rates between cell-surface-attached adhesion molecules. Cells expressing specific membrane receptors are surface-labeled with quantum dots coated with their respective ligands. The progressive diminution in the total number of surface-diffusing quantum dots tracked over time collectively reflects intrinsic ligand/receptor interaction kinetics. The probability of quantum dot detachment is modeled using a stochastic analysis of bond formation and dissociation, with a small number of ligand/receptor pairs, resulting in a set of coupled differential equations that are solved numerically. Comparison with the experimental data provides an estimation of the kinetic rates, together with the mean number of ligands per quantum dot, as three adjustable parameters. We validate this approach by studying the calcium-dependent neurexin/neuroligin interaction, which plays an important role in synapse formation. Using primary neurons expressing neuroligin-1 and quantum dots coated with purified neurexin-1β, we determine the kinetic rates between these two binding partners and compare them with data obtained using other techniques. Using specific molecular constructs, we also provide interesting information about the effects of neurexin and neuroligin dimerization on the kinetic rates. As it stands, this simple technique should be applicable to many types of biological ligand/receptor pairs.     

O-Linked Glycosylation of the Mucin Domain of the Herpes Simplex Virus Type 1-specific Glycoprotein gC-1 Is Temporally Regulated in a Seed-and-spread Manner

The herpes simplex virus type 1 (HSV-1) glycoprotein gC-1, participating in viral receptor interactions and immunity interference, harbors a mucin-like domain with multiple clustered O-linked glycans. Using HSV-1-infected diploid human fibroblasts, an authentic target for HSV-1 infection, and a protein immunoaffinity procedure, we enriched fully glycosylated gC-1 and a series of its biosynthetic intermediates. This fraction was subjected to trypsin digestion and a LC-MS/MS glycoproteomics approach. In parallel, we characterized the expression patterns of the 20 isoforms of human GalNAc transferases responsible for initiation of O-linked glycosylation. The gC-1 O-glycosylation was regulated in an orderly manner initiated by synchronous addition of one GalNAc unit each to Thr-87 and Thr-91 and one GalNAc unit to either Thr-99 or Thr-101, forming a core glycopeptide for subsequent additions of in all 11 GalNAc residues to selected Ser and Thr residues of the Thr-76–Lys-107 stretch of the mucin domain. The expression patterns of GalNAc transferases in the infected cells suggested that initial additions of GalNAc were carried out by initiating GalNAc transferases, in particular GalNAc-T2, whereas subsequent GalNAc additions were carried out by followup transferases, in particular GalNAc-T10. Essentially all of the susceptible Ser or Thr residues had to acquire their GalNAc units before any elongation to longer O-linked glycans of the gC-1-associated GalNAc units was permitted. Because the GalNAc occupancy pattern is of relevance for receptor binding of gC-1, the data provide a model to delineate biosynthetic steps of O-linked glycosylation of the gC-1 mucin domain in HSV-1-infected target cells.     

Design,synthesis and in vitro evaluation of novel uni- and bivalent ligands for the cannabinoid receptor type 1 with variation of spacer length and structure

Using rimonabant, a potent inverse agonist for cannabinoid receptor type 1 (CB₁R), as parent ligand, a series of novel univalent and bivalent ligands were designed by variation of spacer length and its chemical structure. The ligands synthesized were evaluated for affinity and selectivity by radioligand displacement and a functional steady-state GTPase assay. The results showed the nature of the spacer influences the biological readout. Albeit all compounds show significantly lower affinities than rimonabant, this fact could be used to demonstrate that affinities and selectivity are influenced by the chemical structure and length of the spacer and might be helpful for designing bivalent probes for other GPCR receptors.     

Targeted lysosome disruptive elements for improvement of parenchymal liver cell-specific gene delivery

The transfection ability of nonviral gene therapy vehicles is generally hampered by untimely lysosomal degradation of internalized DNA. In this study we describe the development of a targeted lysosome disruptive element to facilitate the escape of DNA from the lysosomal compartment, thus enhancing the transfection efficacy, in a cell-specific fashion. Two peptides (INF7 and JTS-1) were tested for their capacity to disrupt liposomes. In contrast to JTS-1, INF7 induced rapid cholesterol-independent leakage (EC(50), 1.3 microm). INF7 was therefore selected for coupling to a high affinity ligand for the asialoglycoprotein receptor (ASGPr), K(GalNAc)(2), to im- prove its uptake by parenchymal liver cells. Although the parent peptide disrupted both cholesterol-rich and -poor liposomes, the conjugate, INF7-K(GalNAc)(2), only induced leakage of cholesterol-poor liposomes. Given that endosomal membranes of eukaryotic cells contain <5% cholesterol, this implies that the conjugate will display a higher selectivity toward endosomal membranes. Although both INF7 and INF7-K(GalNAc)(2) were found to increase the transfection efficiency on polyplex-mediated gene transfer to parenchymal liver cells by 30-fold, only INF7-K(GalNAc)(2) appeared to do so in an ASGPr-specific manner. In mice, INF7-K(GalNAc)(2) was specifically targeted to the liver, whereas INF7 was distributed evenly over various organs. In summary, we have prepared a nontoxic cell-specific lysosome disruptive element that improves gene delivery to parenchymal liver cells via the ASGPr. Its high cell specificity and preference to lyse intracellular membranes make this conjugate a promising lead in hepatocyte-specific drug/gene delivery protocols.     

Short and long spacer sequences and other structural features of zinc binding sites in zinc enzymes

The crystal structures of eleven zinc enzymes have served to identify common features of their Zn binding sites. Two of them have non-catalytic Zn sites, both of which contain four cysteine ligands closely spaced in the linear sequence of the protein with no bound water. In contrast, all the catalytic Zn sites have three protein ligands and, in addition, one coordinated, 'activated' water. Histidine is the predominant ligand. The spacing between the first two ligands (1-3 amino acids), the short spacer, ensures a nucleus for Zn binding. The third ligand, separated by from approximately 20 to approximately 120 amino acids, the long spacer, not only completes the coordination but also aligns protein residues for interaction with the substrate. The short and long spacing observed for catalytic zinc sites may also pertain to Fe and Cu proteins.     

Carbohydrate specificity of a lectin isolated from the fungus Sclerotium rolfsii

In order to investigate the functional roles of a phytopathogenic fungal lectin (SRL) isolated from the bodies of Sclerotium rolfsii, the binding properties of SRL were studied by enzyme linked lectinosorbent assay and by inhibition of SRL-glycan interaction. Among glycoproteins (gp) tested for binding, SRL reacted strongly with GalNAc alpha1-->4Ser/Thr (Tn) and/or Gal beta1-->3GalNAc alpha1-->(T(alpha)) containing gps: human T(alpha) and Tn glycophorin, asialo salivary gps, and asialofetuin, but its reactivity toward sialylated glycoproteins was reduced significantly. Of the sugar ligands tested for inhibition of SRL-asialofetuin binding, Thomsen-Friedenreich residue (T(alpha)) was the best, being 22.4 and 2.24 x 10(3) more active than GalNAc and Gal beta1--> residues, respectively. Other ligands tested were inactive. When the glycans used as inhibitors, T(alpha), and/or Tn containing gps, especially asialo PSM, asialo BSM, asialo OSM, active antifreeze gp, asialo glycophorin and Tn-glycophorin were very active, and 1.0 x 10(4) times more potent than GalNAc. From these results, it is clear that the combining site of SRL should be of a cavity type and recognizes only Tn and T(alpha) residues of glycans; it is suggested that T(alpha) and Tn glycotopes, which are present only in abnormal carbohydrate sequences of higher orders of mammal, are the most likely sites for phytopathogenic fungal attachment as an initial step of infection. The affinity of SRL for ligands can be ranked in decreasing order as follows: multivalent T(alpha) and Tn > monomeric T(alpha) and Tn > GalNAc > II (Gal beta1-->4GlcNAc), L (Gal beta1-->4Glc), and Gal.     

Morphological analysis of ligand uptake and processing: the role of multivesicular endosomes and CURL in receptor-ligand processing

The receptor-mediated endocytosis and intracellular processing of transferrin and mannose receptor ligands were investigated in bone marrow-derived macrophages, fibroblasts and reticulocytes. Mannosylated bovine serum albumin (BSA) conjugated to colloidal gold (Au-man-BSA) or colloidal gold-transferrin (AuTf) were used to trace ligand processing in these cells. These ligands appeared to be processed by mechanisms similar to those observed previously with other mannose receptor and galactose receptor ligand probes. After uptake via coated pits and coated vesicles, Au-man-BSA appeared in small uncoated vesicles and tubular structures and was transferred to large, sometimes multivesicular endosomes (MVEs), which sometimes had arm-like protrusions reminiscent of CURL (compartment of uncoupling of receptor and ligand) [10, 11]. Initially these structures became increasingly multivesicular, but during longer incubations the inclusion vesicles appeared to disintegrate to leave a denser, amorphous lumen. Inclusion vesicle disintegration may result from the introduction of lysosomal enzymes into these structures. These results suggest a model for differential receptor-ligand and ligand-ligand sorting. As suggested [10, 11] membrane constituents may be recycled to the plasma membrane from the arms of CURL. Receptor-bound ligands, such as transferrin, would also recycle. The luminal contents, including dissociated ligands, other soluble proteins and inclusion vesicles (containing some membrane proteins), would target to lysosomes. This would result in the lysosomal degradation of any membrane proteins that were incorporated in the inclusion vesicle membranes.     

Transport of N-acetyl-D-galactosamine in Escherichia coli K92: effect on acetyl-amino sugar metabolism and polysialic acid production

The N-acetyl-D-galactosamine (GalNAc) transport system of Escherichia coli K92 was studied when the bacterium was grown in a chemically defined medium containing GalNAc as a carbon source. Kinetic measurements were carried out in vivo at 37 degrees C in 25 mM phosphate buffer, pH 7.0. Under these conditions, the uptake rate was linear for at least 3 min and the calculated Km for GalNAc was 3 microM. The transport system was strongly inhibited by sodium arsenate (70%), potassium cyanide (62%) and 2,4-dinitrophenol (75%). Analysis of bacterial GalNAc phosphotransferase activity revealed in vitro GalNAc phosphorylation activity only when phosphoenolpyruvate was present. These results strongly support the notion that GalNAc uptake depends on a specific phosphotransferase system. Study of activity regulation showed that N-acetylglucosamine and mannosamine specifically inhibit the transport of GalNAc in this bacterium. Analysis of expression revealed that the GalNAc transport system is specifically induced by GalNAc but not by N-acetylglucosamine (GlcNAc) or N-acetylmannosamine (ManNAc), two intimately related sugars. Moreover, full induction of GalNAc transport required the presence of both cAMP and GalNAc. Comparative studies revealed that E. coli K92 has developed a regulation mechanism that specifically induces the appropriate permease based on the presence of each respective phospho-amino sugar (GlcNAc, ManNAc and GalNAc). In this regulation system, GlcNAc is the preferred amino sugar as the carbon source. Finally, when E. coli K92 was grown using GalNAc, capsular polysialic acid production was strongly affected. The presence of intracellular phosphoderivative acetylamino sugars, generated by the action of the phosphotransferase transport system, can be responsible for this effect.