Inhibiting GPI anchor biosynthesis in fungi stresses the endoplasmic reticulum and enhances immunogenicity

In fungi, the anchoring of proteins to the plasma membrane via their covalent attachment to glycosylphosphatidylinositol (GPI) is essential and thus provides a valuable point of attack for the development of antifungal therapeutics. Unfortunately, studying the underlying biology of GPI-anchor synthesis is difficult, especially in medically relevant fungal pathogens because they are not genetically tractable. Compounding difficulties, many of the genes in this pathway are essential in Saccharomyces cerevisiae. Here, we report the discovery of a new small molecule christened gepinacin (for GPI acylation inhibitor) which selectively inhibits Gwt1, a critical acyltransferase required for the biosynthesis of fungal GPI anchors. After delineating the target specificity of gepinacin using genetic and biochemical techniques, we used it to probe key, therapeutically relevant consequences of disrupting GPI anchor metabolism in fungi. We found that, unlike all three major classes of antifungals in current use, the direct antimicrobial activity of this compound results predominantly from its ability to induce overwhelming stress to the endoplasmic reticulum. Gepinacin did not affect the viability of mammalian cells nor did it inhibit their orthologous acyltransferase. This enabled its use in co-culture experiments to examine Gwt1's effects on host-pathogen interactions. In isolates of Candida albicans, the most common fungal pathogen in humans, exposure to gepinacin at sublethal concentrations impaired filamentation and unmasked cell wall β-glucan to stimulate a pro-inflammatory cytokine response in macrophages. Gwt1 is a promising antifungal drug target, and gepanacin is a useful probe for studying how disrupting GPI-anchor synthesis impairs viability and alters host-pathogen interactions in genetically intractable fungi.     

Inhibition of the GlcNAc transferase of the glycosylphosphatidylinositol anchor biosynthesis in African trypanosomes.

A wide variety of eukaryotic membrane proteins are anchored to the cell surface by a covalent linkage to glycosylphosphatidylinositol. One of the best characterised examples is the variant surface glycoprotein of the protozoan parasite, Trypanosoma brucei. The pathway for the formation of the glycosylphosphatidylinositol precursor has been previously described, with the first step being the transfer of GlcNAc, from UDP-GlcNAc to endogenous phosphatidylinositol to form N-acetyl-glucosaminylphosphatidylinositol [Doering, T. L., Masterson, W. J., Hart, G. W. & Englund, P. T. (1989) J. Biol. Chem. 264, 11,168-11,173]. Here we report that low concentrations of sulphydryl alkylating reagents irreversibly inhibit this transferase in a trypanosome-derived cell-free system. The site of inactivation by N-ethylmaleimide appears to be at, or close to, the enzyme active site, since incubation of the enzyme preparation with the donor molecule UDP-GlcNAc substantially protects the enzyme from inactivation. The protection appears to be primarily dependent on the nucleotide portion of the molecule, since UMP and UDP can mimic the protection seen with UDP-GlcNAc.     

Enzymes and auxiliary factors for GPI lipid anchor biosynthesis and post-translational transfer to proteins

GPI lipid anchoring is an important post-translational modification of eukaryote proteins in the endoplasmic reticulum. In total, 19 genes have been directly implicated in the anchor synthesis and the substrate protein modification pathway. Here, the molecular functions of the respective proteins and their evolution are analyzed in the context of reported literature data and sequence analysis studies for the complete pathway (http://mendel.imp.univie.ac.at/SEQUENCES/gpi-biosynthesis/) and questions for future experimental investigation are discussed. Studies of two of these proteins have provided new mechanistic insights. The cytosolic part of PIG-A/GPI3 has a two-domain alpha/beta/alpha-layered structure; it is suggested that its C-terminal subsegment binds UDP-GlcNAc whereas the N-terminal domain interacts with the phosphatidylinositol moiety. The lumenal part of PIG-T/GPI16 apparently consists of a beta-propeller with a central hole that regulates the access of substrate protein C termini to the active site of the cysteine protease PIG-K/GPI8 (gating mechanism) as well as of a polypeptide hook that embraces PIG-K/GPI8. This structural proposal would explain the paradoxical properties of the GPI lipid anchor signal motif and of PIG-K/GPI8 orthologs without membrane insertion regions in some species.     

Chemical validation of GPI biosynthesis as a drug target against African sleeping sickness

It has been suggested that compounds affecting glycosylphosphatidylinositol (GPI) biosynthesis in bloodstream form Trypanosoma brucei should be trypanocidal. We describe cell-permeable analogues of a GPI intermediate that are toxic to this parasite but not to human cells. These analogues are metabolized by the T. brucei GPI pathway, but not by the human pathway. Closely related nonmetabolizable analogues have no trypanocidal activity. This represents the first direct chemical validation of the GPI biosynthetic pathway as a drug target against African human sleeping sickness. The results should stimulate further inhibitor design and synthesis and encourage the search for inhibitors in natural product and synthetic compound libraries.     

A novel pathway for glycan assembly: biosynthesis of the glycosyl-phosphatidylinositol anchor of the trypanosome variant surface glycoprotein

The trypanosome variant surface glycoprotein (VSG), like many other eukaryotic cell surface proteins, is anchored to the plasma membrane by a glycosyl-phosphatidylinositol (GPI) moiety. This glycolipid is assembled first as a precursor (glycolipid A) that is then covalently attached to the newly synthesized polypeptide. We have developed a trypanosome cell-free system capable of performing all of the steps in the biosynthesis of the glycan portion of glycolipid A. Using [3H]sugar nucleotides as substrates, several biosynthetic intermediates have been identified. From structural analyses of these intermediates, we propose a pathway for GPI biosynthesis. Based on comparisons between the VSG GPI anchor and similar structures in other cells, we believe that this same pathway will apply to the GPI anchors, and the related insulin-mediator compound, of higher eukaryotes.     

GPI anchor biosynthesis in yeast: phosphoethanolamine is attached to the alpha1,4-linked mannose of the complete precursor glycophospholipid

Cells synthesize the GPI anchor carbohydrate core by successively adding N-acetylglucosamine, three mannoses, and phosphoethanolamine (EtN-P) onto phosphatidylinositol, thus forming the complete GPI precursor lipid which is then added to proteins. Previously, we isolated a GPI deficient yeast mutant accumulating a GPI intermediate containing only two mannoses, suggesting that it has difficulty in adding the third, alpha1,2-linked Man of GPI anchors. The mutant thus displays a similar phenotype as the mammalian mutant cell line S1A-b having a mutation in the PIG-B gene. The yeast mutant, herein named gpi10-1 , contains a mutation in YGL142C, a yeast homolog of the human PIG-B. YGL142C predicts a highly hydrophobic integral membrane protein which by sequence is related to ALG9, a yeast gene required for adding Man in alpha1,2 linkage to N-glycans. Whereas gpi10-1 cells grow at a normal rate and make normal amounts of GPI proteins, the microsomes of gpi10-1 are completely unable to add the third Man in an in vitro assay. Further analysis of the GPI intermediate accumulating in gpi10 shows it to have the structure Manalpha1-6(EtN-P-)Manalpha1-4GlcNalpha1- 6(acyl) Inositol-P-lipid. The presence of EtN-P on the alpha1,4-linked Man of GPI anchors is typical of mammalian and a few other organisms but had not been observed in yeast GPI proteins. This additional EtN-P is not only found in the abnormal GPI intermediate of gpi10-1 but is equally present on the complete GPI precursor lipid of wild type cells. Thus, GPI biosynthesis in yeast and mammals proceeds similarly and differs from the pathway described for Trypanosoma brucei in several aspects.      

Developmental variation of glycosylphosphatidylinositol membrane anchors in Trypanosoma brucei. In vitro biosynthesis of intermediates in the construction of the GPI anchor of the major procyclic surface glycoprotein.

The African trypanosome, Trypanosoma brucei, expresses two abundant stage-specific glycosylphosphatidylinositol (GPI)-anchored glycoproteins, the procyclic acidic repetitive protein (PARP or procyclin) in the procyclic form, and the variant surface glycoprotein (VSG) in the mammalian bloodstream form. The GPI anchor of VSG can be readily cleaved by phosphatidylinositol (PI)-specific phospholipase C (PI-PLC), whereas that of PARP cannot, due to the presence of a fatty acid esterified to the inositol. In the bloodstream form trypanosome, a number of GPIs which are structurally related to the VSG GPI anchor have been identified. In addition, several structurally homologous GPIs have been described, both in vivo and in vitro, that contain acyl-inositol. In vivo the procyclic stage trypanosome synthesizes a GPI that is structurally homologous to the PARP GPI anchor, i.e. contains acyl-inositol. No PI-PLC-sensitive GPIs have been detected in the procyclic form. Using a membrane preparation from procyclic trypanosomes which is capable of synthesizing GPI lipids upon the addition of nucleotide sugars we find that intermediate glycolipids are predominantly of the acyl-inositol type, and the mature ethanolamine-phosphate-containing precursors are exclusively acylated. We suggest that the differences between the bloodstream and procyclic form GPI biosynthetic intermediates can be accounted for by the developmental regulation of an inositol acylhydrolase, which is active only in the bloodstream form, and a glyceride fatty acid remodeling system, which is only partially functional in the procyclic form.     

Amino acid conditions near the GPI anchor attachment site of prion protein for the conversion and the GPI anchoring

Prion protein (PrP) is a glycosylphosphatidylinositol (GPI)-anchored protein, and the C-terminal GPI anchor signal sequence (GPI-SS) of PrP is cleaved before GPI anchoring. However, mutations near the GPI anchor attachment site (the ω site) in the GPI-SS have been recognized in human genetic prion diseases. Moreover, the ω site of PrP has not been identified except hamster, though it is known that amino acid restrictions are very severe at the ω and ω + 2 sites in other GPI-anchored proteins. To investigate the effect of mutations near the ω site of PrP on the conversion and the GPI anchoring, and to discover the ω site of murine PrP, we systematically created mutant murine PrP with all possible single amino acid substitutions at every amino acid residue from codon 228 to 240. We transfected them into scrapie-infected mouse neuroblastoma cells and examined the conversion efficiencies and the GPI anchoring of each mutant PrP. Mutations near the ω site altered the conversion efficiencies and the GPI anchoring efficiencies. Especially, amino acid restrictions for the conversion and the GPI anchoring were severe at codons 230 and 232 in murine PrP, though they were less severe than in other GPI-anchored proteins. Only the mutant PrPs presented on a cell surface via a GPI anchor were conversion competent. The present study shows that mutations in the GPI-SS can affect the GPI anchoring and the conversion efficiency of PrP. We clarified for the first time the ω site of murine PrP and the amino acid conditions near the ω site for the conversion as well as GPI anchoring.     

Yeast Gpi8p is essential for GPI anchor attachment onto proteins.

Glycosylphosphatidylinositol (GPI) anchors are added onto newly synthesized proteins in the ER. Thereby a putative transamidase removes a C-terminal peptide and attaches the truncated protein to the free amino group of the preformed GPI. The yeast mutant gpi8-1 is deficient in this addition of GPIs to proteins. GPI8 encodes for an essential 47 kDa type I membrane glycoprotein residing on the luminal side of the ER membrane. GPI8 shows significant homology to a novel family of vacuolar plant endopeptidases one of which is supposed to catalyse a transamidation step in the maturation of concanavalin A and acts as a transamidase in vitro. Humans have a gene which is highly homologous to GPI8 and can functionally replace it.     

GPI1 stabilizes an enzyme essential in the first step of glycosylphosphatidylinositol biosynthesis.

Attachment of glycosylphosphatidylinositol (GPI) is essential for the surface expression of many proteins. Biosynthesis of glycosylphosphatidylinositol is initiated by the transfer of N-acetylglucosamine from UDP-N-acetylglucosamine to phosphatidylinositol. In mammalian cells, this reaction is mediated by a complex of PIG-A, PIG-H, PIG-C, and GPI1. This complexity may be relevant for regulation and for usage of a particular phosphatidylinositol. However, the functions of the respective components have been unclear. Here we cloned the mouse GPI1 gene and disrupted it in F9 embryonal carcinoma cells. Disruption of the GPI1 gene caused a severe but not complete defect in the generation of glycosylphosphatidylinositol-anchored proteins, indicating some residual biosynthetic activity. A complex of PIG-A, PIG-H, and PIG-C decreased to a nearly undetectable level, whereas a complex of PIG-A and PIG-H was easily detected. A lack of GPI1 also caused partial decreases of PIG-C and PIG-H. Therefore, GPI1 stabilizes the enzyme by tying up PIG-C with a complex of PIG-A and PIG-H.     

Glycosylphosphatidylinositol-specific phospholipase C regulates transferrin endocytosis in the African trypanosome

GPI-PLC (glycosylphosphatidylinositol-specific phospholipase C) is expressed in bloodstream-form Trypanosoma brucei, a protozoan that causes human African trypanosomiasis. Loss of genes encoding GPI-PLC reduces the virulence of a pleomorphic strain of the parasite, for reasons that are not clear. In the present paper, we report that GPI-PLC stimulates endocytosis of transferrin by 300-500%. Surprisingly, GPI-PLC is not detected at endosomes, suggesting that the enzyme does not interact directly with the endosomal machinery. We therefore hypothesized that a diffusible product of the GPI-PLC enzyme reaction [possibly DAG (diacylglycerol)] mediated the biological effects of the protein. Two sets of data support this assertion. First, a catalytically inactive Q81L mutant of GPI-PLC, expressed in a GPI-PLC-null background, had no effect on endocytosis, indicating that enzyme activity is essential for the protein to stimulate endocytosis. Secondly, the exogenous DAGs OAG (1-oleyl-2-acetyl-sn-glycerol) and DMG (dimyristoylglycerol) independently stimulated endocytosis of transferrin. Furthermore, the DAG mimic PMA, a phorbol ester, also activated endocytosis in T. brucei. DAG-stimulated endocytosis is a novel pathway in the trypanosome. We surmise that (i) GPI-PLC regulates transferrin endocytosis in T. brucei, (ii) GPI-PLC is a signalling enzyme, and (iii) DAG is a second messenger for GPI-PLC. We propose that regulation of endocytosis is a physiological function of GPI-PLC in bloodstream T. brucei.     

A model for the sequential dominance of antigenic variants in African trypanosome infections.

Trypanosoma brucei infects various domestic and wild mammals in equatorial Africa. The parasite's genome contains several hundred alternative and highly diverged surface antigens, of which only a single one is expressed in any cell. Individual cells occasionally change expression of their surface antigen, allowing them to escape immune surveillance. These switches appear to occur in a partly random way, creating a diverse set of antigenic variants. In spite of this diversity, the parasitaemia develops as a series of outbreaks, each outbreak dominated by relatively few antigenic types. Host-specific immunity eventually clears the dominant antigenic types and a new outbreak follows from antigenic types that have apparently been present all along at low frequency. This pattern of sequential dominance by different antigenic types remains unexplained. I use a mathematical model of parasitaemia and host immunity to show that small variations in the rate at which each type switches to other types can explain the observations. My model shows that randomly chosen switch rates do not provide sufficiently ordered parasitaemias to match the observations. Instead, minor modifications of switch rates by natural selection are required to develop a sequence of ordered parasitaemias.     

Synthesis of two oligosaccharides, the GPI anchor glycans from S. cerevesiae and A. fumigatus

Two oligosaccharides, alpha-D-Manp-(1-->2)-alpha-D-Manp-(1-->2)-alpha-D-Manp-(1-->6)-alpha-D-Manp-(1-->4)-alpha-D-GlcpNAc (I) and alpha-D-Manp-(1-->3)-alpha-D-Manp-(1-->2)-alpha-D-Manp-(1-->2)-alpha-D-Manp-(1-->6)-alpha-D-Manp-(1-->4)-alpha-D-GlcpNAc (II), the glycosylphosphatidylinositol (GPI) anchor glycans from S. cerevesiae and A. fumigatus were synthesized as their methyl glycosides in a regio- and stereoselective manner. The pentasaccharide I was obtained from 6-O-selective glycosylation of methyl 2,3-di-O-benzoyl-alpha-D-mannopyranosyl-(1-->4)-2-acetamido-3,6-di-O-benzoyl-2-deoxy-alpha-D-glucopyranoside (8) with 2-O-acetyl-3,4,6-tri-O-benzoyl-alpha-D-mannopyranosyl-(1-->2)-3,4,6-tri-O-benzoyl-alpha-D-mannopyranosyl trichloroacetimidate (9), followed by benzoylation, deacetylation, and mannosylation, and then by deprotection. The hexasaccharide (II) was obtained via condensation of allyl 3,4,6-tri-O-benzoyl-alpha-D-mannopyranosyl-(1-->2)-3,4,6-tri-O-benzoyl-alpha-D-mannopyranoside (17) with 2,3,4,6-tetra-O-benzoyl-alpha-D-mannopyranosyl-(1-->3)-2,4,6-tri-O-acetyl-alpha-D-mannopyranosyl trichloroacetimidate (16), followed by deallylation, trichloroacetimidation, and coupling with acceptor (8), and finally by deprotection.     

The use of experimental artefacts in African trypanosome research

When trypanosomes are removed from the field and maintained in laboratory conditions, phenotypic changes commonly occur such that the lines used by many investigators in routine work show several differences from the populations that affect humans and cattle in Africa. Whether these differences are important or irrelevant of course depends on the purpose of each particular experiment, but an awareness of what the differences are can be a useful aid in the interpretation of results. Furthermore, trypanosomes can be manipulated in the laboratory to possess particular characteristics that aid in the testing of hypotheses that are difficult to test using 'wild-type' trypanosomes. In this article, Mike Turner describes how some defined trypanosome lines have been created, how they differ from one another and several of their uses.     

Synthesis of the essential core of the human glycosylphosphatidylinositol (GPI) anchor

The biological role of GPI anchors is of paramount importance; however, we are still far from fully understanding the structure-function relationship of these molecules. One major limiting factor has been the tiny quantities available from natural sources; obtaining homogeneous and well-defined GPI structures by synthesis, is both a challenge and an attractive goal. We report here the convergent synthesis of the essential core of the human GPI anchor 1, exploiting a common precursor to obtain the trisaccharidic donor 2 and a novel protecting groups sequence. The final product, prepared for the first time, is biologically active.     

The human GPI1 gene is required for efficient glycosylphosphatidylinositol biosynthesis.

The first step in glycosylphosphatidylinositol (GPI) membrane anchor biosynthesis that is defective in paroxysmal nocturnal haemoglobinuria is mediated by an N-acetylglucosaminyl transferase expressed in the endoplasmic reticulum. Six human genes encode subunits of this enzyme, namely PIG-A, PIG-C, PIG-H, PIG-P, GPI1, and DPM2. Here, the human GPI1 gene is characterised. This gene is organised into eleven exons. The locus was mapped to chromosome 16p13.3 near the haemoglobin alpha chain locus. GPI1 is expressed ubiquitously in human cells and tissues. Expression levels are markedly elevated in haematopoietic tissues (bone marrow, foetal liver). To determine whether human GPI1 is essential for human GPI biosynthesis, antisense RNA was expressed in HEK293 cells. Transfectants exhibited a marked but incomplete decrease in the expression of a GPI-linked reporter protein, confirming that GPI1 is required for efficient GPI biosynthesis. In contrast, expression of GPI-linked proteins is normal in lymphatic cell lines from individuals with the alpha thalassaemia/mental retardation syndrome, which is characterised by large deletions from chromosome 16p removing one of the two GPI1 alleles along with the haemoglobin alpha locus. In conclusion, GPI1 plays an important role in the biosynthesis of GPI intermediates. Due to its autosomal localisation, the heterozygous deletion of GPI1 does not lead to an overt defect in the expression of GPI-linked proteins.     

Glycosylphosphatidylinositol (GPI) proteins of Saccharomyces cerevisiae contain ethanolamine phosphate groups on the alpha1,4-linked mannose of the GPI anchor

In humans and Saccharomyces cerevisiae the free glycosylphosphatidylinositol (GPI) lipid precursor contains several ethanolamine phosphate side chains, but these side chains had been found on the protein-bound GPI anchors only in humans, not yeast. Here we confirm that the ethanolamine phosphate side chain added by Mcd4p to the first mannose is a prerequisite for the addition of the third mannose to the GPI precursor lipid and demonstrate that, contrary to an earlier report, an ethanolamine phosphate can equally be found on the majority of yeast GPI protein anchors. Curiously, the stability of this substituent during preparation of anchors is much greater in gpi7Delta sec18 double mutants than in either single mutant or wild type cells, indicating that the lack of a substituent on the second mannose (caused by the deletion of GPI7) influences the stability of the one on the first mannose. The phosphodiester-linked substituent on the second mannose, probably a further ethanolamine phosphate, is added to GPI lipids by endoplasmic reticulum-derived microsomes in vitro but cannot be detected on GPI proteins of wild type cells and undergoes spontaneous hydrolysis in saline. Genetic manipulations to increase phosphatidylethanolamine levels in gpi7Delta cells by overexpression of PSD1 restore cell growth at 37 degrees C without restoring the addition of a substituent to Man2. The three putative ethanolamine-phosphate transferases Gpi13p, Gpi7p, and Mcd4p cannot replace each other even when overexpressed. Various models trying to explain how Gpi7p, a plasma membrane protein, directs the addition of ethanolamine phosphate to mannose 2 of the GPI core have been formulated and put to the test.