Cold shock and regulation of surface protein trafficking convey sensitization to inducers of stage differentiation in Trypanosoma brucei

Abstract

Transmission of a protozoan parasite from a vertebrate to invertebrate host is accompanied by cellular differentiation. The signals from the environment that trigger the process are poorly understood. The model parasite Trypanosoma brucei proliferates in the mammalian bloodstream and in the tsetse fly. On ingestion by the tsetse, the trypanosome undergoes a rapid differentiation that is marked by replacement of the variant surface glycoprotein (VSG) coat with GPI-anchored EP and GPEET procyclins. Here we show that a cold shock of DeltaT > 15 degrees C is sufficient to reversibly induce high-level expression of the insect stage-specific EP gene in the mammalian bloodstream stages of T. brucei. The 3'-UTR of the EP mRNA is necessary and sufficient for the increased expression. During cold shock, EP protein accumulates in the endosomal compartment in the proliferating, slender, bloodstream stage, whereas the EP is present on the plasma membrane in the quiescent, stumpy, bloodstream stage. Thus, there is a novel developmentally regulated cell surface access control mechanism for a GPI-anchored protein. In addition to inducing EP expression, cold shock results in the acquisition of sensitivity to micromolar concentrations of cis-aconitate and citrate by stumpy but not slender bloodstream forms. The cis-aconitate and citrate commit stumpy bloodstream cells to differentiation to the procyclic stage along with rapid initial proliferation. We propose a hierarchical model of three events that regulate differentiation after transmission to the tsetse: sensing the temperature change, surface access of a putative receptor, and sensing of a chemical cue.

Figure 1.

Kinetics and properties of cold shock induction of EP expression. (A) Detection of EP on the cell surface of trypanosomes by flow cytometry (mAb TRBP/247). Stumpy bloodstream cells were incubated at 27°C in the absence (left panel) or presence (middle panel) of 6 mM cis-aconitate or at 20°C (right panel). (B) Cold shock induces reporter gene expression from the EP1 expression site. Transgenic trypanosomes carrying a luciferase gene (construct pG-ΔLII.Luc) or an EP1:GFP fusion gene (construct pG-ΔLII.EG) within the EP1 expression site were incubated at 18°C. At the indicated time points, luciferase activity was measured by luminometry (•), and EP1:GFP expression was quantified by fluorescence microscopy (○). (C) Temperature profile of reporter gene induction. Cell lines expessing luciferase (•) or EP1:GFP (○) from the EP1 locus (see B) were incubated for 16 h at the indicated temperatures.

Figure 2.

(A) Structure of the basic constructs used in this work: pL20 (Wirtz et al. 1998), pL82 (Wirtz et al. 1999), pTR (Xong et al. 1998), pG (Furger et al. 1997), pTub (Wirtz et al. 1994), pES221 (Muñoz-Jordán et al. 1996; M. Engstler and G.A.M. Cross, unpubl.). (Ti) Tetracycline-inducible; (RDS) rDNA spacer; (RDL) rDNA locus; (Ω) T7 terminator; (PAG) procyclin-associated gene; (Tub) tubulin array; (221P1) VSG221-pseudogene I. (B) Predicted secondary structure of the EP1 3′-UTR. For calculation the M-fold software (http://www.bioinfo.rpi.edu/applications/mfold) was used with constraints derived from Drozdz and Clayton (1999). Independent loops are labeled I–III, and the endpoints of relevant deletion mutants are marked by arrows and by large font size. Deletions ΔLII and Δ164 are shaded. (C) RT–PCR of EP1:GFP mRNAs in recombinant T. brucei cell lines. Sequences between the GFP reporter and the cognate EP1 polyadenylation site were amplified (for details see Supplemental Material). The expected sizes of the PCR products are 850 bp (L20-wt.EG), 777 bp (L20-ΔLII.EG), and 688 bp (L20-Δ164.EG), respectively (M, size marker; 100-bp ladder). The cDNA samples were shown free of contaminating genomic DNA by an independent PCR reaction.

Figure 3.

Developmentally regulated routing of EP1 and EP1:GFP in bloodstream-stage trypanosomes. Pure AnTat1.1 slender (sl) and stumpy (st) bloodstream populations and established isogenic procyclic cultures (pcf) were compared by Northern and Western blot analyses. (A) Wild type. (B) Transgenic line with EP1:GFP replacing the EP1 gene (construct pG-ΔLII.EG). (C) Transgenic line expressing EP1:GFP from the ribosomal promoter region (construct pTR.EG). Cells were incubated for 14 h at 37°C or 20°C as indicated or at 27°C for procyclic cultures. Normalized quantitative expression data for EP and GFP are given below the corresponding bands. A β-tubulin hybridization probe and an anti-PFR antibody were used for normalization of Northern and Western blots, respectively. The asterisk indicates that fivefold less total RNA or cell equivalents was loaded to the PCF lane. Three-channel fluorescence imaging of selected samples (as indicated) of panels A–C is shown in D–F, respectively. (Blue) DAPI staining of the nuclear and mitochondrial (kinetoplast) DNA; (red) AnTat1.1 VSG immunofluorescence, outlining the cell surface; (green) EP immunofluorescence in wild-type cells (D) or EP1:GFP autofluorescence (E,F). Yellow/orange color indicates colocalization of red and green fluorescence emission of nonequalized channels. Maximum intensity projections of 3D stacks are shown, and the red channel of slender cells was processed using a morphological gradient filter to facilitate the visualization of intracellular staining in 3D images. The right-side cell is a representation without red channel information. The flagellar pocket is indicated by arrowheads. All images are representative subregions of larger wide-field data sets displaying >20 cells each. The phenotypic homogeneity of slender and stumpy populations with respect to surface routing of EP or EP1:GFP was 99% ± 1% (n > 1000 cells analyzed).

Figure 4.

Subcellular localization of EP1:GFP in bloodstream-stage T. brucei MITat1.2. All images represent a single confocal plane extracted from digitally deconvolved multichannel 3D image data sets that were acquired using the same batch of fixed and permeabilized cells expressing EP1:GFP from the VSG221 expression site (pES221.EG). The left column images show an intensity merge of green EP1:GFP and blue DAPI fluorescence. The middle column shows the intracellular localization of markers, and the right column images show a merge of the three channels. (A) Cell surface staining by rabbit polyclonal VSG221 antiserum (1:1000) (red). (B) Endogenous EP (and EP1:GFP) detected by a mouse monoclonal EP antibody (1:500) (red). (C) The ER chaperone BiP detected by polyclonal rabbit anti-BiP (antibody courtesy of Jay Bangs, Madison; 1:500). (D) The Golgi marker TbRAB31 detected by rabbit anti-TbRAB31 (antibody courtesy of Mark Field, London; 1:200). (E) Endocytosed Alexa Fluor 594 conjugated dextran (see Materials and Methods). (F) Endocytosed biotinylated VSG (see Materials and Methods) detected with Alexa Fluor 594-conjugated streptavidin (Molecular Probes). (G) The lysosome visualized with a mouse monoclonal antibody directed against the lysosomal membrane protein p67 (antibody courtesy of David Alexander and Jay Bangs, Wisconsin; 1:1000). Alexa Fluor 594-conjugated goat anti-rabbit second antibody (1:2000) (A,C,D) or CY3-conjugated sheep anti-mouse second antibody (1:1000) (B,G) was used.

Figure 5.

Cold shock alone is not sufficient to promote developmental progression. (A) Analysis of the initial EP1:GFP surface expression kinetics by flow cytometry. Aliquots of a transgenic AnTat1.1 stumpy population carrying the EP1:GFP gene within the EP1 expression site (construct pG-ΔLII. EG) were incubated at 27°C in the absence (left panel) or presence (middle panel) of 6 mM cis-aconitate, and at 20°C (right panel). (B) The resuming cellular proliferation and differentiation of the cold-shocked EP1:GFP-expressing cells was further monitored for 100 h in the presence and absence of 6 mM cis-aconitate at 20°C (note arrow from A to B). Total fluorescence per culture volume was measured in intervals of 1 h (mean fluorescence intensity of six parallels ± standard deviation; ○, •) using an automated fluorescence reader. In parallel, cell numbers were counted (□, ▪). Cis-aconitate induces cellular proliferation and differentiation to the procyclic insect stage (•, ▪). In the absence of cis-aconitate (○, □), only a minor fluorescence gain and almost no cellular proliferation can be observed.

Figure 6.

Cold shock induction of EP procyclin synthesis is reversible in slender and stumpy bloodstream stages and is not accompanied by VSG coat release. (A) Slender AnTat1.1 cells, carrying EP1:GFP within the EP1 expression site (construct pG-ΔLII.EG), were cultivated at 20°C with a daily change of cell culture medium. After 72 h, cultivation was continued at 37°Cand cells were diluted fivefold with fresh medium. (B) Stumpy populations derived from the same transgenic clone as in A were incubated for 24 h at 20°C and then transferred to 37°C. The EP1:GFP expression (•) was measured by quantitative fluorescence microscopy, and cell proliferation (○) was monitored by cell counting. (C,D) Induction of EP1:GFP synthesis (•) and loss of VSG (○) were monitored in transgenic stumpy cells, cultivated for 24 h in the presence (C) or absence (D) of 6 mM cis-aconitate at 20°C. EP1:GFP-specific and VSG-specific fluorescence were quantified with a CCD camera.

Figure 7.

Cold shock renders T. brucei hypersensitive to cis-aconitate. (A) Wild-type stumpy cells of strain AnTat1.1 were cultivated for 7 d at 27°C in DTM medium in the presence of various concentrations of cis-aconitate (0–6 mM; see numbers in graph). To provide logarithmic growth conditions, the cell density was kept below 7 × 10⁶ per milliter by dilution with fresh culture medium. The population growth was calculated as cell density multiplied by the cumulative dilution factors. At cis-aconitate concentrations below 1 mM, virtually no cellular proliferation was seen. (B) Cells were cold-shocked at 20°C for 16 h, and then treated exactly as in A. Growth data for cis-aconitate concentrations ≥0.8 mM were identical to the 6 mM data points (data not shown). (C) Transgenic AnTat1.1 stumpy cells, constitutively expressing EP1 from the rDNA spacer (construct pL82.E) were treated as in A. Expression of EP was monitored after 1, 3, and 7 d by immunofluorescence microscopy, revealing that irrespective of the amount of cis-aconitate added, all cells were expressing EP on the cell surface.

Figure 8.

Model for the mechanisms regulating differentiation of bloodstream form trypanosomes to the procyclic stage. Large gray arrows indicate transitions between defined developmental or functional states. Additional dashed arrow lines indicate reversibility of the respective transitions. Small black arrows indicate the action of environmental cues , , or checkpoint-like control . Crosses symbolize surface proteins (including EP and the putative citrate receptor) up-regulated by cold shock. The model does not exclude the participation of intracellular cold-induced proteins.