The de novo synthesis of myo-inositol occurs via a two-step process: first, glucose 6-phosphate is converted into inositol 1-phosphate by an INO1 (myo-inositol-1-phosphate synthase; EC 188.8.131.52); then, it is dephosphorylated by an inositol monophosphatase. The myo-inositol can then be incorporated into PI (phosphatidylinositol), which is utilized in a variety of cellular functions, including the biosynthesis of GPI (glycosylphosphatidylinositol) anchors. A putative INO1 was identified in the Trypanosoma brucei genome database and, by recombinant expression in Escherichia coli, was shown to be a catalytically active INO1. To investigate the importance of INO1, we created a conditional knockout, which, under non-permissive conditions, showed that INO1 is an essential gene in bloodstream form T. brucei and that the de novo synthesized myo-inositol is used for the formation of PI and GPI anchors.
- inositol 1-phosphate
- inositol-1-phosphate synthase (INO1)
- myo-inositol synthesis
- Trypanosoma brucei
myo-Inositol is an essential metabolite in all eukaryotes. It plays a vital role in growth regulation, signal transduction, membrane biogenesis, osmotolerance and other essential biochemical processes, as well as the formation of GPI (glycosylphosphatidylinositol) anchors. myo-Inositol is de novo synthesized via the concerted action of two enzymes: INO1 (myo-inositol-1-phosphate synthase; EC 184.108.40.206) and L-myo-inositol-1-phosphatase (EC 220.127.116.11) (for a review, see ).
In this study, we begin to investigate the de novo synthesis and metabolism of myo-inositol in Trypanosoma brucei, including the characterization of the recombinantly expressed enzyme and the creation of a T. brucei conditional null mutant, which showed that de novo synthesis of myo-inositol is essential for the bloodstream form of this parasite.
Results and discussion
Using the yeast homologue, a single putative INO1 was identified in the T. brucei genome (Sanger Centre, Hinxton, Cambridgeshire, U.K.). This TbINO1 was recombinantly expressed in Escherichia coli using the expression vector TOPO pBAD (Invitrogen), which also encodes a C-terminal His tag. After the soluble recombinant protein was purified by affinity chromatography using Ni2+-NTA (Ni2+-nitrilotriacetate) beads, it was shown to have INO1 activity by its ability to catalyse the isomerization of glucose 6-phosphate to inositol 1-phosphate. This activity was dependent on NAD+ and was stimulated by the presence of NH4+, consistent with INO1s from other organisms . Saturation kinetics was shown by the enzyme with an apparent Km of 0.58 mM, similar to the human INO1 with a Km of 0.5 mM . Interestingly, the specific activity of the recombinant TbINO1 (756 units/mg) is approx. 10-fold higher than that of the human recombinant INO1 (70 units/mg) , which may be an indication of the importance of TbINO1 to T. brucei. The TbINO1 sequence has been deposited in GenBank® under accession no. AJ866770.
Essentiality of TbINO1 in bloodstream form T. brucei
Southern-blot analysis of TbINO1 showed it to be a single copy gene per haploid T. brucei genome, making it amenable for gene disruption studies. One allele of TbINO1 was replaced by homologous recombination with the puromycin resistance gene. Before deleting the second allele, an ectopic copy of the TbINO1 gene under tight tetracycline regulation was introduced into the rDNA locus using the expression vector pLew100 . Transcription of the ectopic TbINO1 was induced with tetracycline prior to replacement of the remaining allele of TbINO1 with the hygromycin resistance gene and the genotype confirmed by Southern blotting.
To establish if TbINO1 is an essential gene in bloodstream form T. brucei, the TbINO1 cells were grown in either tetracycline-free or tetracycline-containing media. When conditional knockout cells were grown in the presence of tetracycline, they exhibited normal growth rates compared with wild-type cells (Figures 1A and 1B). However, when grown in the absence of tetracycline, the cells grew normally for the first 2 days, but after day 3 the cells either died or failed to divide (Figure 1C), showing that TbINO1 is an essential gene in bloodstream form T. brucei. Surprisingly, even when the medium was supplemented with an additional 0.6 mM myo-inositol, the cells were unable to overcome the deletion of INO1 (Figure 1D), which is in contrast with INO1 null mutants in other organisms [4–6]. After day 8, the conditional knockout cells resumed normal growth kinetics in the absence of tetracycline, which was due to the cells overcoming the tetracycline control and constitutively expressing the ectopic copy (Figure 1E), a phenomenon previously observed for other essential genes in T. brucei [7,8].
In vivo labelling was used to investigate the biochemical phenotype of these conditional knockout cells. The cells were grown either in the presence or absence of tetracycline for 2 days prior to all metabolic labelling experiments. Mid-exponential growth phase cells were metabolically labelled with either [3H]glucose or [3H]mannose for 1 h, the cells were collected by centrifugation and the lipids were extracted using chloroform and methanol. After desalting by butanol/water partitioning, the lipids were separated by TLC and 3H-labelled lipids were detected by autoradiography.
Wild-type and INO1 conditional knockout cells grown in the presence of tetracycline showed identical incorporation into the [3H]lipids as expected. [3H]Mannose labelling yielded the mature GPI glycolipids A and C, as well as low levels of galactosylated versions of glycolipid A (Figure 2, lanes 5 and 6), all of which have been described previously [9,10]. Interestingly, we observed a significant and reproducible decrease in the amount of glycolipids A and C when the INO1 conditional knockout cells were grown in the absence of tetracycline (Figure 2, lane 7), suggesting that the GPI biosynthetic pathway had slowed down significantly.
When labelled with [3H]glucose (Figure 2, lanes 1 and 2), both wild-type and the conditional knockout produced two predominant [3H]lipid species (Figure 2, 1 and 2). Lipid species (1) has an identical RF with that of PI (phosphatidylcholine); further characterization showed that (a) it was a PI-containing phospholipid, (b) the 3H label was on the head group, and (c) the head group was inositol. Lipid species 2 from [3H]glucose labelling had an identical RF with that of glycolipid C; characterization showed that the 3H label was on the head group and that it was glycolipid C. This 3H label was also observed in GPI-anchored variant surface glycoprotein. These results clearly show that (a) [3H]-glucose was converted into [3H]inositol, demonstrating the presence of a functional INO1, and (b) this de novo synthesized [3H]inositol was used by the trypanosome to form PI and mature GPIs. Labelling of the INO1 conditional knockout cells grown in the absence of tetracycline with [3H]glucose showed a decrease in the amount of PI formed (Figure 2, lane 3, lipid species 1) and a concomitant decrease in the amount of glycolipid C (lipid species 2), suggesting a direct knock-on effect to the GPI biosynthetic pathway due to the lack of PI in which [3H]inositol was derived from [3H]glucose.
Bloodstream form T. brucei has a clear dependence on myo-inositol derived from de novo synthesis; in this study, we have shown it to be essential for their survival in culture. Surprisingly, the deletion of TbINO1 cannot be overcome by the addition of extra myo-inositol to the growth medium, suggesting that de novo synthesized myo-inositol has a distinct function in the cell. Metabolic labelling experiments suggested that this was to supply the GPI pathway with PI synthesized from de novo-derived myo-inositol. The reasons for this preference for glucose-derived myo-inositol for GPI biosynthesis are unclear. We are currently investigating this further.
This work was supported by a Wellcome Trust Senior Research Fellowship 067441.
Mechanistic and Functional Studies of Proteins: A Focus Topic at BioScience2005, held at SECC Glasgow, U.K., 17–21 July 2005. Edited by S. Crosthwaite (Manchester, U.K.), M. Ginger (Oxford, U.K.), K. Gull (Oxford, U.K.), A. Lee (Southampton, U.K.), H. McWatters (Oxford, U.K.), J. Mottram (Glasgow, U.K.), P. Rich (University College London, U.K.), C. Robinson (Warwick, U.K.) and H. van Veen (Cambridge, U.K.).
Abbreviations: GPI, glycosylphosphatidylinositol; INO1, inositol-1-phosphate synthase; PI, phosphatidylinositol
- © 2005 The Biochemical Society