Structural and Functional Characterization of Phosphatidylinositol-Phosphate Biosynthesis in Mycobacteria.

In mycobacteria, phosphatidylinositol (PI) acts as a common lipid anchor for key components of the cell wall, including the glycolipids phosphatidylinositol mannoside, lipomannan, and lipoarabinomannan. Glycolipids in Mycobacterium tuberculosis, the causative agent of tuberculosis, are important virulence factors that modulate the host immune response. The identity-defining step in PI biosynthesis in prokaryotes, unique to mycobacteria and few other bacterial species, is the reaction between cytidine diphosphate-diacylglycerol and inositol-phosphate to yield phosphatidylinositol-phosphate, the immediate precursor to PI. This reaction is catalyzed by the cytidine diphosphate-alcohol phosphotransferase phosphatidylinositol-phosphate synthase (PIPS), an essential enzyme for mycobacterial viability. Here we present structures of PIPS from Mycobacterium kansasii with and without evidence of donor and acceptor substrate binding obtained using a crystal engineering approach. PIPS from Mycobacterium kansasii is 86% identical to the ortholog from M. tuberculosis and catalytically active. Functional experiments guided by our structural results allowed us to further characterize the molecular determinants of substrate specificity and catalysis in a new mycobacterial species. This work provides a framework to strengthen our understanding of phosphatidylinositol-phosphate biosynthesis in the context of mycobacterial pathogens.

Structure of Mycobacterium tuberculosis phosphatidylinositol phosphate synthase reveals mechanism of substrate binding and metal catalysis.

Tuberculosis causes over one million yearly deaths, and drug resistance is rapidly developing. Mycobacterium tuberculosis phosphatidylinositol phosphate synthase (PgsA1) is an integral membrane enzyme involved in biosynthesis of inositol-derived phospholipids required for formation of the mycobacterial cell wall, and a potential drug target. Here we present three crystal structures of M. tuberculosis PgsA1: in absence of substrates (2.9 Å), in complex with Mn2+ and citrate (1.9 Å), and with the CDP-DAG substrate (1.8 Å). The structures reveal atomic details of substrate binding as well as coordination and dynamics of the catalytic metal site. In addition, molecular docking supported by mutagenesis indicate a binding mode for the second substrate, D-myo-inositol-3-phosphate. Together, the data describe the structural basis for M. tuberculosis phosphatidylinositol phosphate synthesis and suggest a refined general catalytic mechanism-including a substrate-induced carboxylate shift-for Class I CDP-alcohol phosphotransferases, enzymes essential for phospholipid biosynthesis in all domains of life.

Structural basis for phosphatidylinositol-phosphate biosynthesis.

Phosphatidylinositol is critical for intracellular signalling and anchoring of carbohydrates and proteins to outer cellular membranes. The defining step in phosphatidylinositol biosynthesis is catalysed by CDP-alcohol phosphotransferases, transmembrane enzymes that use CDP-diacylglycerol as donor substrate for this reaction, and either inositol in eukaryotes or inositol phosphate in prokaryotes as the acceptor alcohol. Here we report the structures of a related enzyme, the phosphatidylinositol-phosphate synthase from Renibacterium salmoninarum, with and without bound CDP-diacylglycerol to 3.6 and 2.5 Å resolution, respectively. These structures reveal the location of the acceptor site, and the molecular determinants of substrate specificity and catalysis. Functional characterization of the 40%-identical ortholog from Mycobacterium tuberculosis, a potential target for the development of novel anti-tuberculosis drugs, supports the proposed mechanism of substrate binding and catalysis. This work therefore provides a structural and functional framework to understand the mechanism of phosphatidylinositol-phosphate biosynthesis.

The active site of yeast phosphatidylinositol synthase Pis1 is facing the cytosol.

Five yeast enzymes synthesizing various glycerophospholipids belong to the CDP-alcohol phosphatidyltransferase (CAPT) superfamily. They only share the so-called CAPT motif, which forms the active site of all these enzymes. Bioinformatic tools predict the CAPT motif of phosphatidylinositol synthase Pis1 as either ER luminal or cytosolic. To investigate the membrane topology of Pis1, unique cysteine residues were introduced into either native or a Cys-free form of Pis1 and their accessibility to the small, membrane permeating alkylating reagent N-ethylmaleimide (NEM) and mass tagged, non-permeating maleimides, in the presence and absence of non-denaturing detergents, was monitored. The results clearly point to a cytosolic location of the CAPT motif. Pis1 is highly sensitive to non-denaturing detergent, and low concentrations (0.05%) of dodecylmaltoside change the accessibility of single substituted Cys in the active site of an otherwise cysteine free version of Pis1. Slightly higher detergent concentrations inactivate the enzyme. Removal of the ER retrieval sequence from (wt) Pis1 enhances its activity, again suggesting an influence of the lipid environment. The central 84% of the Pis1 sequence can be aligned and fitted onto the 6 transmembrane helices of two recently crystallized archaeal members of the CAPT family. Results delineate the accessibility of different parts of Pis1 in their natural context and allow to critically evaluate the performance of different cysteine accessibility methods. Overall the results show that cytosolically made inositol and CDP-diacylglycerol can access the active site of the yeast PI synthase Pis1 from the cytosolic side and that Pis1 structure is strongly affected by mild detergents.

Ubiquitous distribution of phosphatidylinositol phosphate synthase and archaetidylinositol phosphate synthase in Bacteria and Archaea, which contain inositol phospholipid.

In Eukarya, phosphatidylinositol (PI) is biosynthesized from CDP-diacylglycerol (CDP-DAG) and inositol. In Archaea and Bacteria, on the other hand, we found a novel inositol phospholipid biosynthetic pathway. The precursors, inositol 1-phosphate, CDP-archaeol (CDP-ArOH), and CDP-DAG, form archaetidylinositol phosphate (AIP) and phosphatidylinositol phosphate (PIP) as intermediates. These intermediates are dephosphorylated to synthesize archaetidylinositol (AI) and PI. To date, the activities of the key enzymes (AIP synthase, PIP synthase) have been confirmed in only three genera (two archaeal genera, Methanothermobacter and Pyrococcus, and one bacterial genus, Mycobacterium). In the present study, we demonstrated that this novel biosynthetic pathway is universal in both Archaea and Bacteria, which contain inositol phospholipid, and elucidate the specificity of PIP synthase and AIP synthase for lipid substrates. PIP and AIP synthase activity were confirmed in all recombinant cells transformed with the respective gene constructs for four bacterial species (Streptomyces avermitilis, Propionibacterium acnes, Corynebacterium glutamicum, and Rhodococcus equi) and two archaeal species (Aeropyrum pernix and Sulfolobus solfataricus). Inositol was not incorporated. CDP-ArOH was used as the substrate for PIP synthase in Bacteria, and CDP-DAG was used as the substrate for AIP synthase in Archaea, despite their fundamentally different structures. PI synthase activity was observed in two eukaryotic species, Saccharomyces cerevisiae and Homo sapiens; however, inositol 1-phosphate was not incorporated. In Eukarya, the only pathway converts free inositol and CDP-DAG directly into PI. Phylogenic analysis of PIP synthase, AIP synthase, and PI synthase revealed that they are closely related enzymes.

Alternative metabolic fates of phosphatidylinositol produced by phosphatidylinositol synthase isoforms in Arabidopsis thaliana.

PtdIns is an important precursor for inositol-containing lipids, including polyphosphoinositides, which have multiple essential functions in eukaryotic cells. It was previously proposed that different regulatory functions of inositol-containing lipids may be performed by independent lipid pools; however, it remains unclear how such subcellular pools are established and maintained. In the present paper, a previously uncharacterized Arabidopsis gene product with similarity to the known Arabidopsis PIS (PtdIns synthase), PIS1, is shown to be an active enzyme, PIS2, capable of producing PtdIns in vitro. PIS1 and PIS2 diverged slightly in substrate preferences for CDP-DAG [cytidinediphospho-DAG (diacylglycerol)] species differing in fatty acid composition, PIS2 preferring unsaturated substrates in vitro. Transient expression of fluorescently tagged PIS1 or PIS2 in onion epidermal cells indicates localization of both enzymes in the ER (endoplasmic reticulum) and, possibly, Golgi, as was reported previously for fungal and mammalian homologues. Constitutive ectopic overexpression of PIS1 or PIS2 in Arabidopsis plants resulted in elevated levels of PtdIns in leaves. PIS2-overexpressors additionally exhibited significantly elevated levels of PtdIns(4)P and PtdIns(4,5)P(2), whereas polyphosphoinositides were not elevated in plants overexpressing PIS1. In contrast, PIS1-overexpressors contained significantly elevated levels of DAG and PtdEtn (phosphatidylethanolamine), an effect not observed in plants overexpressing PIS2. Biochemical analysis of transgenic plants with regards to fatty acids associated with relevant lipids indicates that lipids increasing with PIS1 overexpression were enriched in saturated or monounsaturated fatty acids, whereas lipids increasing with PIS2 overexpression, including polyphosphoinositides, contained more unsaturated fatty acids. The results indicate that PtdIns populations originating from different PIS isoforms may enter alternative routes of metabolic conversion, possibly based on specificity and immediate metabolic context of the biosynthetic enzymes.

Genomewide analysis of phospholipid signaling genes in Phytophthora spp.: novelties and a missing link.

Phospholipids are cellular membrane components in eukaryotic cells that execute many important roles in signaling. Genes encoding enzymes required for phospholipid signaling and metabolism have been characterized in several organisms, but only a few have been described for oomycetes. In this study, the genome sequences of Phytophthora sojae and P. ramorum were explored to construct a comprehensive genomewide inventory of genes involved in the most universal phospholipid signaling pathways. Several genes and gene families were annotated, including those encoding phosphatidylinositol synthase (PIS), phosphatidylinositol (phosphate) kinase (PI[P]K), diacylglycerol kinase (DAG), and phospholipase D (PLD). The most obvious missing link is a gene encoding phospholipase C (PLC). In all eukaryotic genomes sequenced to date, PLC genes are annotated based on certain conserved features; however, these genes seem to be absent in Phytophthora spp. Analysis of the structural and regulatory domains and domain organization of the predicted isoforms of PIS, PIK, PIPK, DAG, and PLD revealed many novel features compared with characterized representatives in other eukaryotes. Examples are transmembrane proteins with a C-terminal catalytic PLD domain, secreted PLD-like proteins, and PIPKs that have an N-terminal G-protein-coupled receptor-transmembrane signature. Compared with other sequenced eukaryotes, the genus Phytophthora clearly has several exceptional features in its phospholipid-modifying enzymes.

Phosphatidylinositol synthesis is essential in bloodstream form Trypanosoma brucei.

PI (phosphatidylinositol) is a ubiquitous eukaryotic phospholipid which serves as a precursor for messenger molecules and GPI (glycosylphosphatidylinositol) anchors. PI is synthesized either de novo or by head group exchange by a PIS (PI synthase). The synthesis of GPI anchors has previously been validated both genetically and chemically as a drug target in Trypanosoma brucei, the causative parasite of African sleeping sickness. However, nothing is known about the synthesis of PI in this organism. Database mining revealed a putative TbPIS gene in the T. brucei genome and by recombinant expression and characterization it was shown to encode a catalytically active PIS, with a high specificity for myo-inositol. Immunofluorescence revealed that in T. brucei, PIS is found in both the endoplasmic reticulum and Golgi. We created a conditional double knockout of TbPIS in the bloodstream form of T. brucei, which when grown under non-permissive conditions, clearly showed that TbPIS is an essential gene. In vivo labelling of these conditional double knockout cells confirmed this result, showing a decrease in the amount of PI formed by the cells when grown under non-permissive conditions. Furthermore, quantitative and qualitative analysis by GLC-MS and ESI-MS/MS (electrospray ionization MS/MS) respectively showed a significant decrease (70%) in cellular PI, which appears to affect all major PI species equally. A consequence of this fall in PI level is a knock-on reduction in GPI biosynthesis which is essential for the parasite's survival. The results presented here show that PI synthesis is essential for bloodstream form T. brucei, and to our knowledge this is the first report of the dependence on PI synthesis of a protozoan parasite by genetic validation.