Figure 1.
Structure and the biosynthesis of T. cruzi GPI anchors.
(A) Structure of a T. cruzi GPI anchor, according to Previato et al. [3]. (B) Proposed biosynthetic pathway of GPI anchor in the endoplasmic reticulum of T. cruzi. N-acetylglucosamine (GlcNAc) is added to phosphatidylinositol (PI) in step 1 and, during the following steps, deacetylation and addition of four mannose residues occur. The addition of ethanolamine-phosphate on the third mannose (step 7) enables the transferring of the completed GPI anchor to the C-terminal of a protein (step 8). Dolichol-P-mannose acts as a mannose donor for all mannosylation reactions that are part of the GPI biosynthesis. This pathway was based on the structure of the T. cruzi GPI and sequence homology of T. cruzi genes with genes known to encode components of this pathway in Saccharomyces cerevisiae, Homo sapiens, Trypanosoma brucei and Plasmodium falciparum. Not shown in the figure, free glycoinositolphospholipids (GIPLs), also present in the T. cruzi membrane, are likely to be by-products of the same GPI biosynthetic pathway.
Table 1.
T. cruzi genes encoding enzymes of the GPI biosynthetic pathway.
Figure 2.
mRNA expression of T. cruzi genes encoding enzymes of the GPI biosynthetic pathway.
Total RNA extracted from epimastigotes (E), trypomastigotes (T) and amastigotes (A) were separated in agarose gels, transferred to nylon membranes and hybridized with [α-32P]-labeled probes specific for TcGPI8 and TcGPI10 genes. The bottom panel shows hybridization with a probe for 24Sα rRNA, used as loading control. The size of ribosomal RNA bands are indicated on the left.
Figure 3.
Cellular localization of T. cruzi enzymes of the GPI biosynthetic pathway.
Epimastigotes were transiently transfected with the plasmids pTREX-TcDPM1-GFP (A), pTREX-TcGPI3-GFP (B), pTREX-TcGPI12-GFP (C) or pTREXnGFP as a control plasmid (D) and (E). Transfected parasites were fixed with 4% paraformaldehyde, incubated with the ER marker anti-BiP (1∶1000) and the secondary antibody conjugated to Alexa 555 (1∶1000). Cells were also stained with DAPI showing the nuclear and kinetoplast DNA. In panel E, parasites that were not incubated with the primary, anti-BiP antibody are shown as negative controls. Images were captured with the Nikon Eclipse Ti fluorescence microscope. Scale bars: 5 µm.
Figure 4.
Yeast complementation with T. cruzi genes encoding enzymes of the GPI biosynthetic pathway.
(A) DPM1, GPI10 and GPI12 yeast conditional lethal mutants (YPH499-HIS-GAL-DPM1, YPH499-HIS-GAL-GPI10 and YPH499-HIS-GAL-GPI12, respectively) were transformed with pRS426Met plasmids carrying either T. cruzi or S. cerevisiae genes encoding DPM1, GPI10 and GPI12 (TcDPM1 or ScDPM1, TcGPI10 or ScGPI10, and TcGPI12 or ScGPI12, respectively). Wild-type (WT), non-transformed mutants and transformed yeast mutants were streaked onto plates with nonpermissive, glucose-containing SD medium lacking histidine, with or without uracil or in galactose-containing medium (with uracil) and incubated at 30°C for 3 days. In the bottom panel, yeast mutants (YPH499-HIS-GAL-GPI14) transformed with pRS426Met plasmid carrying T. cruzi gene (TcGPI14), which could not restore cell growth of GPI14 deficient yeast are shown. (B) GPI-anchored proteins synthesized by the conditional lethal yeast mutants expressing T. cruzi genes were separated by SDS-PAGE and analyzed after fluorography. Wild-type (WT), non-transformed yeast mutants and yeast mutants that were transformed with plasmids containing the corresponding yeast genes (ScDPM1 or ScGPI12) or with the T. cruzi genes (TcDPM1 or TcGPI12), were cultivated in medium glucose-containing in the presence of [2-3H]myo-inositol for 1 hour. Total protein extract corresponding to 1×108 cells were loaded on each lane of a 10% SDS-PAGE and the labeled proteins were visualized by fluorography (top panels). As a loading control, Coomassie Blue stained gels prepared with equivalents amounts of total proteins are shown in the bottom panels. Untransfected DPM1 and GPI12 mutants were grown in the presence of galactose for 2 days and then switched to glucose-containing medium for 16 hours before addition of [2-3H]myo-inositol. Molecular weight markers (M) are shown on the left.
Table 2.
Functional complementation of yeast mutants by T. cruzi genes.
Figure 5.
Generation of TcGPI8 heterozygous mutants.
(A) DNA constructs generated to delete both TcGPI8 alleles by homologous recombination are shown with the NeoR or HygR genes flanked by 5′ and 3′ sequences of the TcGPI8 gene and the SacI/SpeI and XhoI/XbaI cloning sites from the pCR2.1TOPO vector. After transfecting epimastigotes with the purified DNA fragments, parasites were selected in LIT medium containing 200 µg/ml of G418 or hygromycin. Total DNA, isolated from G418 or hygromycin resistant parasites was analyzed by PCR amplifications, using the primers indicated by arrows. Below the schemes of DNA constructs, the sizes of the NeoR or HygR genes and the 5′ and 3′ sequences of the TcGPI8 gene are shown. (B) PCR amplification products analyzed on 1% agarose gel electrophoresis were obtained from DNA isolated from epimastigotes transfected with the GPI8-Neo (top panel) or GPI8-Hyg construct (bottom panel) and using pairs of primers showed in A. Amplicons derived from PCR using the primer pair 1F/7R that amplify a T. cruzi GPI8 allele which was not deleted are shown. On lanes indicated by (-), loaded samples were from PCR in which no template DNA was added. (C) Expression levels of TcGPI8 mRNA in WT and TcGPI8 single knockout of each allele interrupted by NeoR or HygR genes (+/− N or +/− H, respectively). Total RNAs purified from epimastigotes were hybridized to [α-32P]-labeled probes specific for the TcGPI8 gene (top panel) or for the 24Sα rRNA (bottom panel) used as loading control. The size of ribosomal RNA bands are indicated on the left and a graph with the quantification of the signals from the TcGPI8 probe after normalization using the 24Sα rRNA probe is shown below.
Figure 6.
Translocation of the TcGPI8 gene in T. cruzi mutants.
(A) DNA constructs generated to delete both TcGPI8 alleles are shown with the NeoR or HygR genes flanked by 5′ and 3′ sequences of the TcGPI8 and spliced leader (SL) addition and polyadenylation signals from TcP2β (HX1) and gapdh genes. (B) DNA isolated from two cloned epimastigote cell lines that have been sequentially transfected with NeoR and HygR constructs and selected with G418 and hygromycin (double resistant mutants, N/H) were PCR amplified with primers shown in (A). Amplicons generated using primers 2F/2R indicate the integration of the NeoR in one of the TcGPI8 alleles whereas amplicons generated with primers 1F/4R and 5F/2R indicate the integration of the HygR sequences in the second TcGPI8 allele. PCR amplification using primers 1F/8R shows that double resistant parasite cell lines still maintained at least one intact copy of the TcGPI8 gene. As positive controls, DNA from NeoR (for primers 2F/2R), HygR (for primers 1F/4R and 5F/2R) single knockout and wild-type parasites (for primers 1F/8R) were used. (C) Expression levels of TcGPI8 mRNA in WT, TcGPI8 single knockout NeoR (+/− N1) and two double resistant clones (N/H1 and N/H2). RNA purified from epimastigotes were hybridized to [α-32P]-labeled TcGPI8 (top panel) or 24Sα rRNA (middle panel) probes. (D) RT-PCR amplification of TcGPI8 sequences. Reverse transcribed TcGPI8 mRNA obtained from WT, single knockout NeoR (+/− N1) and two double resistant clones (N/H1 and N/H2) were PCR amplified with primers annealing with TcGPI8 sequences and the T. cruzi SL sequence. First strand cDNA synthesis reactions were done with primers complementary to TcGPI8 in the presence (+) or absence (−) of reverse transcriptase. PCR products, separated on 1% agarose gels were stained with ethidium bromide. Molecular weight DNA markers are shown on the left.
Figure 7.
Cell membrane morphology of T. cruzi GPI8 mutants.
Transmission electron microscopy showing cellular membranes of wild type T. cruzi epimastigotes (WT), TcGPI8 single allele knockout, neomycin resistant (+/− N1) and double resistant TcGPI8 epimastigote mutants (N/H1). Although displaying similar morphologies, representative images show that single allele TcGPI8 mutants present a thinner layer of parasite glycocalyx, when compared to wild type cells, whereas cell membranes of double resistant parasites present a glycocalyx layer that is slightly thicker than the glycocalyx of wild type parasite membranes (indicated by the arrows).
Figure 8.
Cell membrane mucins in T. cruzi GPI8 mutants.
Immunoblot of total (T), cytoplasmic (C) and membrane (M) fractions of WT epimastigotes, TcGPI8 single allele knockout, neomycin resistant (+/−N2) and double resistant TcGPI8 (N/H2) mutant cell lines. Equivalent amounts of protein from each fraction, as showed by the Coomassie blue stained bands (bottom panel), were transferred to nitrocellulose membranes and incubated with anti-mucin antibodies and revealed with horseradish peroxidase conjugated secondary antibodies.