Fig 1.
Schematic representation of fatty acid and sterol de novo biosynthesis in procyclic trypanosomes.
Black arrows indicate enzymatic steps of leucine, glucose, threonine and acetate metabolism, with dashed arrows symbolizing several steps, to feed fatty acid and ergosterol biosynthesis. Acetyl-CoA and HMG-CoA are boxed to highlight their branching point position. For simplification and clarity only the mitochondrial subcellular compartment is represented. The microsomal elongase system and mitochondrial fatty acid synthesis are represented by a dashed circle labelled ELO and FASII, respectively. The boxed enzymes have been investigated by reverse genetics approaches in this manuscript. Abbreviations: AOB, amino oxobutyrate; Ac-CoA, acetyl-CoA; AcAc-CoA, acetoacetyl-CoA; HMG-CoA, 3-hydroxy-3-methylglutaryl-CoA. Indicated enzymes are: ACC, acetyl-CoA carboxylase; ACH, acetyl-CoA thioesterase; AceCS, AMP-dependent acetyl-CoA synthetase; AKCT, 2-amino-3-ketobutyrate CoA transferase; ASCT, acetate:succinate CoA-transferase; BCAT, branched-chain aminotransferase; BCKDH, branched-chain α-keto acid dehydrogenase complex; HMGR, HMG-CoA reductase; HMGS, HMG-CoA synthase; IVDH, isovaleryl-CoA dehydrogenase; MCC, 3-methylcrotonoyl-CoA decarboxylase; MGH, 3-methylglutaconyl-CoA hydratase; PDH, pyruvate dehydrogenase complex; SCP2, SCP2-thiolase; TDH, threonine 3-dehydrogenase.
Fig 2.
Incorporation of radio-labelled or 13C-labelled carbon sources into lipids of PCF trypanosomes.
In panels A and B, [14C]-labelled fatty acid methyl esters and sterols were separated by HPTLC and analyzed as described in the Materials and methods section. Prior to lipid extraction, the EATRO1125.T7T PCF were incubated 16 h in SDM79 containing 4 mM glucose, 4 mM threonine, 4 mM proline, 4 mM leucine, 1 mM isoleucine, 1 mM valine, 4 mM acetate and one radio-labelled tracer (A, [1-14C]-acetate; T, L-[U-14C]-threonine; G, D-[U-14C]-glucose; P, L-[U-14C]-proline; L, L-[U-14C]-leucine; I, L-[U-14C]-isoleucine; V, L-[U-14C]-valine). The cell has also been incubated with L-[U-14C]-proline, 0.15 mM threonine, 1 mM isoleucine and 1 mM valine, without glucose and acetate (P*). The data are expressed as nmol of precursor (radioactive and non-radioactive molecules) incorporated into fatty acids (A) and/or sterols (B) in 108 cells per hour. Error bars indicate mean ± SD of 3 biological replicates. nd: not detectable. In panel C, the EATRO1125.T7T PCF were incubated in the same conditions as described above with non-enriched (control) or uniformly [13C]-enriched proline, leucine, glucose, threonine or acetate, before processing the sample for sterol analysis by GC-MS. The isotope enrichment factor corresponds to the percentage of molecules containing 13C-enriched carbons. The amount of each of the five sterols identified is similar in the six experimental conditions; cholesta-5,7,24-trienol, 0.5 ± 0.1 μg; prothothecasterol, 2.2 ± 0.6 μg; ergosta-5,7,24(25)-trienol, 0.7 ± 0.1 μg; ergosta-5,7,25(27)-trienol, 2.2 ± 0.4 μg; cholesterol, 7.9 ± 1.7 μg.
Fig 3.
Characterization of the AKCT gene involved in threonine degradation.
(A) PCR analysis of genomic DNA isolated from the parental EATRO1125.T7T (PCL) and Δakct cells. Amplifications were performed with primers based on sequences flanking the 5’UTR and 3’UTR fragments used to target the AKCT gene depletion (black boxes) and primers binding the ORF of the AKCT gene (PCR products 1 and 2), or the blasticidin (BSD, PCR products 3 and 4) and the puromycin (PAC, PCR products 5 and 6) resistance genes. (B) NMR analysis of the EATRO1125.T7T (PCL) and Δakct cell lines using D-[U-13C]-glucose and threonine. Excretion of 13C-enriched succinate (S13) and acetate (A13) and non-enriched succinate (S12) and acetate (A12) is displayed. (C) L-[U-14C]-threonine and [1-14C]-acetate incorporation into sterols and fatty acids, of parental (PCL) and Δakct cell lines. [14C]-labelled fatty acid methyl esters and sterols were separated by HPTLC after transesterification and analyzed as described in the Materials and methods section. Error bars indicate mean ± SD of 3 biological replicates. Data are normalized with the parental cell line (PCL) values with an arbitrary value of 100 for the PCL samples. For more detail see Fig 2 legend. (D) Localization of the TY1-tagged AKCT by immunofluorescence microscopy, with the anti-ASCT immune serum used as mitochondrial marker. The right part of the panel shows a western blot of the parental (PCL) and TY1-tagged AKCT (AKCT-TY1) cell lines with the anti-TY1 immune serum. Differential interference contrast (DIC) of cells is shown to the left of the panel. Scale bar, 5 μm.
Fig 4.
IVDH is involved in sterol biosynthesis from leucine.
(A) PCR analysis of genomic DNA isolated from the parental EATRO1125.T7T (PCL), Δivdh-C3 and Δivdh-C8 cell lines. Amplifications were performed with primers based on sequences flanking the 5’UTR and 3’UTR fragments used to target the IVDH gene depletion (black boxes) and the 5’UTR 3’-extremity (PCR product 2), 3’UTR 5’-extremity (PCR product 3) or internal sequences of the puromycin (PAC, PCR products 4 and 5) and blasticidin (BSD, PCR products 6 and 7) resistance genes. The full-length IVDH coding sequence was also PCR-amplified with specific primers sequences (PCR product 1). As expected, PCR amplification using primers derived from the IVDH gene and drug resistant genes were only observed for the parental EATRO1125.T7T and both Δivdh cell lines, respectively. (B) Western blot analysis of the EATRO1125.T7T, Δivdh-C3 and Δivdh-C8 cell lines using anti-IVDH and anti-PFR immune sera. (C) Incorporation of [14C]-labelled leucine (L-[U-14C]-leucine) and threonine (L-[U-14C]-threonine) into sterols and fatty acids of parental (PCL), Δivdh-C3 and Δivdh-C8 cell lines. [14C]-labelled fatty acid methyl esters and sterols were separated by HPTLC after transesterification and analyzed as described in the Materials and methods section. Error bars indicate mean ± SD of 3 biological replicates. Data are normalized with the parental cell (PCL) values with an arbitrary value of 100 for the PCL samples. For more detail see Fig 2 legend. Panel D shows the mitochondrial localization of IVDH by immunofluorescence. IVDH with or without the first 33 amino acids, corresponding to the putative mitochondrial-targeting motif, was fused to the N-terminus extremity of EGFP and expressed in PCF. Fluorescence associated with EGFP is shown on the left panel and ASCT was detected with an anti-ASCT immune serum (central panel). Differential interference contrast (DIC) of cells is shown to the right of each panel. Scale bar, 5 μm.
Fig 5.
Sterol biosynthesis from glucose, threonine and acetate requires SCP2-thiolase, and fatty acid biosynthesis from acetate requires ASCT.
(A) Western blot analyses of the parental (WT), knock-out and tetracycline-induced (.i) or non-induced (.ni) mutant cell lines with the immune sera indicated in the right margin. (B) The top and lower parts represent the relative incorporation of radio-labelled carbon sources (D-[U-14C]-glucose, L-[U-14C]-threonine, [1-14C]-acetate and L-[U-14C]-leucine) into fatty acids and sterols, respectively, of tetracycline-induced (.i) and non-induced (.ni) RNAiSCP2, Δach and Δasct cell lines, compared to the parental cell line (PCL). [14C]-labelled fatty acid methyl esters and sterols were separated by HPTLC and analyzed as described in the Materials and methods section. Data are normalized with the parental cell (PCL) values with an arbitrary value of 100 for the PCL samples, which is represented by a horizontal dashed lane. Error bars indicate mean ± SD of 3 biological replicates. nd: not detectable.
Fig 6.
Metabolic flux redistribution between the sterol and fatty acid biosynthetic pathways.
(A) Western blot analyses of the parental (WT), knock-out and tetracycline-induced (.i) or non-induced (.ni) mutant cell lines with the immune sera indicated in the right margin. (B) The top and lower parts represent the relative incorporation of radio-labelled carbon sources (D-[U-14C]-glucose, L-[U-14C]-threonine, [1-14C]-acetate and L-[U-14C]-leucine) into fatty acids and sterols, respectively, of tetracycline-induced (.i) and non-induced (.ni) RNAiTDH, RNAiTDH/RNAiPDH, and Δpdh cell lines, compared to the parental cell line (PCL). [14C]-labelled fatty acid methyl esters and sterols were separated by HPTLC and analyzed as described in the Materials and mehods section. Data are normalized with the parental cell (PCL) values with an arbitrary value of 100 for the PCL samples, which is represented by a horizontal dashed lane. The arrows highlight reduction of radio-label incorporation into lipids expected from the metabolic map in Fig 1. Error bars indicate mean ± SD of 3 biological replicates. nd: not detectable.
Fig 7.
D-[U-14C]-glucose incorporation into lipids of the EATRO1125.T7T (PCL) and Δach/RNAiASCT cell lines.
In this experiment, trans-esterification of fatty acids was not performed before HPTLC separation and the different fatty acid-containing lipids were combined as performed before [10]. Data are normalized with the parental cell (PCL) values with an arbitrary value of 100 for the PCL samples, which is represented by a horizontal dashed lane. Black columns correspond to the EATRO1125.T7T parental cell line, grey columns to the non-induced (.ni) mutant Δach/RNAiASCT cells, and the light grey and white columns to the mutant cells tetracycline-induced for 8 days (.i 8d) or 15 days (.i 15d), respectively. It is to note that the SDM79 used for this labelling experiment does not contain acetate. Error bars indicate mean ± SD of 3 biological replicates. The inset shows a western blot control using the anti-ASCT and anti-hsp60 immune sera.
Fig 8.
Tsetse fly midgut infection rate with the Δakct, Δivdh-C3 and Δivdh-C8 cell lines.
In this experiment, 50 to 100 Glossina morsitans morsitans teneral males were artificially fed with either the parental (PCL), Δakct, Δivdh-C3 or Δivdh-C8 PCF cell lines in culture medium as previously described (n = 850) [41]. Two weeks after the infective meal, all living flies were dissected to assess the presence of parasites in their entire midgut by microscope examination. Midgut infection rates (in % ±SD) are presented for each strain as the mean of three (mutant cell line) or six (PCL) independent biological replicates (n = 243 for PCL, n = 169 for Δakct, n = 78 for Δivdh-C3, n = 65 for Δivdh-C8).
Fig 9.
IVDH is needed for growth of PCF in medium depleted in ketogenic carbon sources.
The figure shows growth curve of the parental cell line (PCL, black curve), the Δivdh-C3 (ΔC3, blue curve) and Δivdh-C8 (ΔC8, red curve) null mutants, as well as the Δivdh-C3/IVDH rescue cell line (R, grey curve) incubated in SDM79 medium containing or not 4 mM of glucose (G), threonine (T) and acetate (A), and supplemented with normal FCS (solid lines) or delipidated FCS (dashed lines). Since threonine is required for protein biosynthesis, 150 μM of this amino acid is present in the glucose/threonine/acetate-depleted conditions. Cells were maintained in the exponential growth phase (between 106 and 107 cells/ml) and cumulative cell numbers reflect normalization for dilution during cultivation. The inset shows western blot analysis of these cell lines with the anti-IVDH and anti-PFR immune sera.
Fig 10.
Metabolic flux distributions in parental and mutant cell lines.
Schemes in panel A compare metabolic flux distribution between the different branches of fatty acid and sterol biosynthesis of the T. brucei procyclic parental and mutant cell lines grown in the carbon source-rich SDM79 medium (in vitro). The carbon sources included in the model are leucine, acetate, glucose and threonine, but not fatty acids, since their incorporation into lipids through de novo biosynthetic pathways has not been demonstrated in rich medium yet. The arrow thickness reflects the strength of metabolic flux redistributions, such as upregulation of leucine metabolism and fatty acid preference, observed in the Δivdh, Δach/RNAiASCT, RNAiAceCS, RNAiSCP2 and/or RNAiTDH/RNAiPDH mutants compared to the parental PCF cell line. The estimated flux distribution in PCF trypanosomes developing in the tsetse fly midgut is presented in the right box chart. The question mark indicates that the in vivo ketogenic carbon source(s) supplementing threonine, as well as the flux through the acetyl-CoA/HMG-CoA bridge are unknown; this diagram assumes a limited availability of ketogenic carbon sources. Panel B describes metabolic adaptations using as reference the parental PCF grown in rich in vitro conditions. The question mark means that the possible metabolic adaptation in vivo is still unknown, since the carbon source contents in the tsetse's organs, including the gut and salivary glands, remain unknown. In Panel C, these metabolic adaptations are re-interpreted considering the probable physiological conditions that PCF have to face in vivo as reference, with the assumption that ketogenic carbon sources are limited in the tsetse midgut and/or in the salivary glands. Abbreviations: A, acetate; AcCoA, acetyl-CoA; FA, fatty acids; G, glucose; HMGCoA, 3-hydroxy-3-methylglutaryl-CoA; L, leucine; T, threonine; Ste, sterols.