Cap-independent translation directs stress-induced differentiation of the protozoan parasite Toxoplasma gondii

Abstract

Translational control mechanisms modulate the microbial latency of eukaryotic pathogens, enabling them to evade immunity and drug treatments. The protozoan parasite Toxoplasma gondii persists in hosts by differentiating from proliferative tachyzoites to latent bradyzoites, which are housed inside tissue cysts. Transcriptional changes facilitating bradyzoite conversion are mediated by a Myb domain transcription factor called BFD1, whose mRNA is present in tachyzoites but not translated into protein until stress is applied to induce differentiation. We addressed the mechanisms by which translational control drives BFD1 synthesis in response to stress-induced parasite differentiation. Using biochemical and molecular approaches, we show that the 5'-leader of BFD1 mRNA is sufficient for preferential translation upon stress. The translational control of BFD1 mRNA is maintained when ribosome assembly near its 5'-cap is impaired by insertion of a 5'-proximal stem-loop and upon knockdown of the Toxoplasma cap-binding protein, eIF4E1. Moreover, we determined that a trans-acting RNA-binding protein called BFD2/ROCY1 is necessary for the cap-independent translation of BFD1 through its binding to the 5'-leader. Translation of BFD2 mRNA is also suggested to be preferentially induced under stress but by a cap-dependent mechanism. These results show that translational control and differentiation in Toxoplasma proceed through cap-independent mechanisms in addition to canonical cap-dependent translation. Our identification of cap-independent translation in protozoa underscores the antiquity of this mode of gene regulation in cellular evolution and its central role in stress-induced life-cycle events.

Keywords: Toxoplasma; gene expression; parasite; stress response; translation.

Figure 1

BFD1 mRNA is preferentially translated under alkaline stress. BFD1^HA parasites were cultured in tachyzoite conditions (No Stress) or under alkaline stress conditions (Stress) for 24 h or 5 days (A) cDNA was synthesized from total RNA collected from the parasites and mean transcript abundance ± standard deviation from 3 biological replicates is plotted relative to constitutive expressed β-Tubulin. Values are normalized to no stress tachyzoites for each transcript. ns: p > 0.05; ∗: p ≤ 0.05, ∗∗∗∗: p ≤ 0.0001 by a one-way ANOVA and Tukey’s multiple comparisons test. (B) Immunoblot of lysates from unstressed and stressed BFD1^HA parasites were probed with the designated antibody. Molecular weight markers are indicated to the left of each panel in kDa. (C) Tub-FLuc or BFD1-FLuc reporters are highlighted in the illustration including the 5′-leaders of Tubulin and BFD1 (not to scale) fused to firefly luciferase (FLuc). Reporters were co-transfected with Tub-NLuc into WT ME49 parasites and cultured under normal no stress or alkaline stress conditions for 24 h. Luciferase measurements were obtained by Dual Glo luciferase assay in alkaline stress (Stress, blue bars) or tachyzoite conditions (No Stress, grey bars). All luciferase measurements were adjusted into relative luciferase units (RLUs) by normalizing against Tub-FLuc reporter in unstressed culture conditions ± standard deviation from 3 biological replicates, which is plotted on right. Luciferase reporter mRNAs shown on the left of the panels were measured by RT-qPCR. Mean transcript abundance ± standard deviation from 3 biological replicates is plotted relative to β-Tubulin, with normalization to tachyzoites transfected with Tub-FLuc reporter. ns: p > 0.05; ∗∗∗∗: p ≤ 0.0001 by Student’s two-tailed t test. Fold changes are shown in parentheses.

Figure 2

Insertion of stem-loop blocks preferential translation of BFD2 but not BFD1. The Tub-FLuc (A), BFD1-FLuc (B) and BFD2-Fluc (C) reporters containing the respective 5′-leaders adjoined to the FLuc coding sequence with or without the indicated 5′-stem-loop (SL). These reporters were individually co-transfected with Tub-NLuc into WT ME49 parasites. RLU was measured for transfected parasites cultured under alkaline stress (Stress, blue bars) or tachyzoite conditions (No Stress, grey bars). Mean fold change ± standard deviation from 3 biological replicates is plotted with normalization to the RLU of transfected parasites without the stem-loop with no stress. Luciferase reporter mRNAs measured by RT-qPCR. Mean transcript abundance ± standard deviation from 3 biological replicates is plotted relative to transfected tachyzoites without the stem-loop with no stress. ns: p > 0.05; ∗: p ≤ 0.05, ∗∗∗∗: p ≤ 0.0001 by a two-way ANOVA and Tukey’s multiple comparisons test. Fold changes are shown in parentheses.

Figure 3

Translation of BFD1 persists despite depletion of eIF4E1. (A), Tub-FLuc reporter with or without the stem-loop co-transfected with Tub-NLuc into eIF4E1^{mAID-HAΔFLuc} parasites. Transfected parasites were cultured under alkaline stress (Stress, blue bars) or tachyzoite conditions (No Stress, grey bars) for 24 h. Parasites were treated with vehicle (Veh) or IAA for 4 h for targeted depletion of eIF4E1. FLuc activity was measured and mean raw luminescence units ± standard deviation from 3 biological replicates is presented normalized to Tub-FLuc without stress (Vehicle). ns: p > 0.05; ∗∗: p ≤ 0.01, ∗∗∗∗: p ≤ 0.0001 by a two-way ANOVA and Tukey’s multiple comparisons test. Luciferase reporter mRNAs were also measured by RT-qPCR as described in Fig. 2. ns = p > 0.05 by two-way ANOVA and Tukey’s multiple comparisons test. Fold changes relative to unstressed parasites are shown in parentheses. BFD1-FLuc (B) and BFD2-FLuc (C) reporters with or without the stem-loop co-transfected with Tub-NLuc into eIF4E1^{mAID-HAΔFLuc} parasites and luciferase activity and mRNA measurements and results are presented as in panel A.

Figure 4

Preferential translation of BFD1 requires BFD2-binding elements in its 5′ leader. (A) Schematic of BFD2 CLIP-seq highlighting BFD2-binding regions (designated BR 1, 2, 3) within the 5′-leader of the BFD1 transcript. Illustrations in subsequent panels depict each BFD2-binding site as a different colored box with “X” denoting the deletion of that BFD2-binding site. (B) Unmodified and mutant BFD1 5′-leader versions with all BFD2-binding sites deleted were transfected into WT ME49 parasites. Transfected parasites were cultured under alkaline stress (Stress, blue bars) or tachyzoite culture conditions (No Stress, grey bars) for 24 h and relative luciferase activity was measured. Mean fold change ± standard deviation from 3 biological replicates is plotted with normalization to the RLU of the WT BFD1-FLuc reporter with no stress. Luciferase reporter mRNAs were measured by RT-qPCR. Mean transcript abundance ± standard deviation from 3 biological replicates is plotted relative to β-Tubulin with normalization to no stress transfected with BFD1-FLuc reporter. ns: p > 0.05; ∗: p ≤ 0.05, ∗∗∗: p ≤ 0.001, ∗∗∗∗: p ≤ 0.0001 by a two-way ANOVA and Tukey’s multiple comparisons test. Fold changes are shown in parentheses. (C-D) Intact BFD1-FLuc reporter or with individual deletions (C) of BFD2-binding sites or combinations of these deletions (D) were co-transfected with Tub-NLuc into WT ME49 parasites. Luciferase activity and mRNA measurements and results are presented as in panel B.

Figure 5

Cap-independent translation of BFD1 requires BFD2-binding sites. (A-C) 5′-stem-loops were inserted in the BFD1-Fluc reporters with individual deletions (B) or combinations of BFD2 binding sites (A, C), as designed by the “X”. The reporter plasmids reporters were transfected into WT ME49 parasites cultured under alkaline stress (Stress, blue bars) or tachyzoite culture conditions (No Stress, grey bars) for 24 h and relative luciferase activity was measured. Mean fold change ± standard deviation from 3 biological replicates is plotted with normalization to the relative luciferase activity of the WT BFD1-FLuc reporter with no stress. Luciferase reporter mRNAs were measured by RT-qPCR. Mean transcript abundance ± standard deviation from 3 biological replicates is plotted relative to β-Tubulin with normalization to no stress transfected with BFD1-FLuc reporter. ns: p > 0.05; ∗: p ≤ 0.05, ∗∗: p ≤ 0.01, ∗∗∗: p ≤ 0.001, ∗∗∗∗: p ≤ 0.0001 by a two-way ANOVA and Tukey’s multiple comparisons test. Fold changes are shown in parentheses.

Figure 6

Preferential translation of BFD1 is blunted in BFD2 knockout parasites. (A) BFD1-FLuc reporters with or without an inserted stem-loop were co-transfected with Tub-NLuc into WT ME49 or ΔBFD2 parasites. Transfected parasites were cultured under alkaline stress (Stress, blue bars) or tachyzoite culture conditions (No Stress, grey bars) for 24 h and RLU were measured. Mean fold change ± standard deviation from 3 biological replicates is plotted with normalization to the RLU of the BFD1-FLuc reporter in ME49 devoid of stress. Luciferase reporter mRNAs were measured by RT-qPCR. Mean transcript abundance ± standard deviation from 3 biological replicates is plotted relative to β-Tubulin with normalization to ME49 tachyzoites transfected with the BFD1-FLuc reporter. ns: p > 0.05; ∗∗∗∗: p ≤ 0.0001 by a two-way ANOVA and Tukey’s multiple comparisons test. Fold changes are shown in parentheses. (B) BFD1-FLuc reporters lacking BFD2-binding sites BR2+BR3 with or without a stem-loop version insertion were co-transfected with Tub-NLuc into WT ME49 or ΔBFD2 parasites and measured and analyzed as described for panel A.

Figure 7

Model for BFD1 preferential translation during stress-induced differentiation. Cellular stress promotes the cap-dependent preferential translation of BFD2 mRNA. Increased levels of BFD2 protein contribute to its binding on BR2 and BR3 regions of the BFD1 5′-leader. BFD2 PTM and interaction with additional proteins may contribute to its association with the BFD1 mRNA and preferential translation. BFD2 protein facilitates the translation of the BFD1 CDS likely by recruiting the preinitiation complex downstream of the 5′-cap and independently of eIF4E1 abundance. BFD1 protein directly binds its target promoters, thus driving the transcriptional response that promotes bradyzoite formation, including increased BFD2 transcription by positive feedback. Enhanced BFD2 mRNA contributes to further translation of BFD2, amplifying BFD1 expression in response to stress.