Decreasing cytosolic translation is beneficial to yeast and human Tafazzin-deficient cells

Abstract

Cardiolipin (CL) optimizes diverse mitochondrial processes, including oxidative phosphorylation (OXPHOS). To function properly, CL needs to be unsaturated, which requires the acyltransferase Tafazzin (TAZ). Loss-of-function mutations in the TAZ gene are responsible for the Barth syndrome (BTHS), a rare X-linked cardiomyopathy, presumably because of a diminished OXPHOS capacity. Herein we show that a partial inhibition of cytosolic protein synthesis, either chemically with the use of cycloheximide or by specific genetic mutations, fully restores biogenesis and the activity of the oxidative phosphorylation system in a yeast BTHS model (taz1Δ). Interestingly, the defaults in CL were not suppressed, indicating that they are not primarily responsible for the OXPHOS deficiency in taz1Δ yeast. Low concentrations of cycloheximide in the picomolar range were beneficial to TAZ-deficient HeLa cells, as evidenced by the recovery of a good proliferative capacity. These findings reveal that a diminished capacity of CL remodeling deficient cells to preserve protein homeostasis is likely an important factor contributing to the pathogenesis of BTHS. This in turn, identifies cytosolic translation as a potential therapeutic target for the treatment of this disease.

Keywords: Barth syndrome; cycloheximide; cardiolipin remodeling; cytosolic protein synthesis; mitochondrial disease; oxidative phosphorylation.

Figure 1. FIGURE 1: Partially decreasing cytosolic translation in Taffazin-deficient (taz1Δ) yeast improves respiration-dependent growth and mtDNA maintenance.

(A) taz1Δ yeast cells were spread as dense layers onto rich ethanol solid media and then exposed to sterile filters spotted with cycloheximide, anisomycin or emetine (dissolved in DMSO). The plates were scanned after 5 days of incubation at 36°C. The filter at the top left was spotted with DMSO alone to provide a negative control. (B) Determination in liquid cultures of CHX concentrations that optimally rescue taz1Δ yeast. Complete synthetic media (CSM) containing 0.5% galactose + 2% ethanol supplemented or not with CHX at the indicated concentrations were inoculated with WT and taz1Δ cells pre-grown in CSM containing 2% glucose at 28°C. The cultures were performed at 36°C and cells densities (OD_600nm) taken over a period of 36 hours. (C) Rate of cytosolic protein synthesis. Total proteins and mitochondrial proteins were labeled with a mixture of [³⁵S]-methionine and [³⁵S]-cysteine for 20 min in whole cells from wild type, taz1Δ rei1Δ, taz1Δ rpl6bΔ and taz1Δ yeast grown for 24 hours in rich 0.5% galactose + 2% ethanol at 36°C, and taz1Δ cells grown in the same conditions in presence of 10 nM cycloheximide (CHX). After the labeling reactions, total protein extracts were prepared and separated by SDS-PAGE on a 12% polyacrylamide gel (75 µg per lane). The gels were dried and analyzed with a PhosphorImager. Quantification was performed using Image J. Data are expressed in % relative to the WT (n=3). The shown data are cytosolic protein synthesis rates (total minus mitochondrial protein synthesis rates). Statistical analysis was done with Tukey’s test (*P<0.05; **P<0.01; ***P<0.001; ****P<0.0001). (D) Genetic ablation of REI1 (rei1Δ) or RPL6B (rpl6bΔ) improves respiratory growth of taz1Δ yeast. WT, taz1Δ, taz1Δ rei1Δ and taz1Δ rpl6bΔ cells freshly grown at 28°C in rich glucose were serially diluted and spotted onto rich ethanol and glucose plates. The plates were scanned after 4 days of incubation at the indicated temperature. (E) Growth of WT, taz1Δ, taz1Δ rei1Δ and taz1Δ rpl6bΔ strains in liquid complete synthetic media containing 0.5% galactose + 2% ethanol at 36°C. The cultures were inoculated with cells grown in CSM containing 2% glucose at 28°C. The cultures were performed at 36°C and cell densities (OD_600nm) taken over a period of 60 hours. (F) Genetic ablation of REI1 (rei1Δ) or RPL6B (rpl6bΔ) in taz1Δ yeast preserves mtDNA maintenance. Proportions of ρ^-/ρ⁰ cells produced in glucose cultures at 28°C of strains WT, taz1Δ, taz1Δ reiΔ, and taz1Δ rpl6bΔ were determined using the procedure described in (n=3). Data are expressed in % relative to the WT and were statistically analyzed using Tukey’s test (*P<0.05; **P<0.01; ***P<0.001).

Figure 2. FIGURE 2: Genetic ablation of REI1 or RPL6B in taz1Δ yeast does not restore cardiolipin remodeling.

Lipids were extracted from mitochondria isolated from WT (black bars), taz1Δ (open bars), taz1Δ rei1Δ (grey bars) and taz1Δ rpl6bΔ (striped bars) cells grown in CSM 0.5% galactose + 2% ethanol at 36°C until a density of 2-3 OD_600nm. (A) Relative contents of PE (phosphatidylethanolamine), CL (cardiolipin), PI (phosphatidylinositol) and PC (phosphatidylcholine) within each strain. (B) Relative fatty acid chain composition of CL within each strain (16:0, palmitic acid; 16:1, palmitoleic acid; 18:0, stearic acid; 18:1: oleic acid). Statistical analysis was done with Kruskal-Wallis test followed by Dunn’s multiple comparison test (*P<0.05; **P<0.01; ***P<0.001). Data are expressed as mean ± s.d. (n=4). The data for WT and taz1Δ strains were reported previously .

Figure 3. FIGURE 3: Partially decreasing cytosolic translation preserves oxidative phosphorylation in taz1Δ yeast.

The experiments here described were performed using mitochondria isolated from cells grown for 24 hours at 36°C in CSM containing 0.5% galactose + 2% ethanol, supplemented or not as indicated with 10 nM CHX, until a density of 2-3 OD_600nm. (A, B) Steady-state levels of proteins involved in the transfer of electrons to oxygen. (A) Proteins were extracted from the mitochondrial samples using 2 g digitonin per g of proteins. The supercomplexes III₂-IV₂ and III₂-IV₁ were revealed by the complex IV activity after separation by CN-PAGE or by western blot with antibodies against Cox2 in BN-PAGE gels. (B) Left panel. Total mitochondrial protein samples were resolved by SDS-PAGE (50 μg per lane) and probed with antibodies against the indicated proteins. The shown gels are representative of at least 3 experiments. Right panel. Quantification using ImageJ software. Levels of Cytc, Cox2, Atp1 and Sdh2 are normalized to Por1p and expressed relative to WT. (C) Genetic ablation of REI1 or RPL6B in taz1Δ yeast preserves mitochondrial respiration. On the left are the rates of oxygen consumption from NADH (4 mM) alone (state 4), after further addition (150 μM) of ADP (state 3) or CCCP (4 μM) (uncoupled respiration). The data are expressed in % of WT state 4 respiration (mean ± s.d, n=4). On the right are the oxygen consumption rates from electrons delivered directly to complex IV by ascorbate 12.5 mM/TMPD 1.4 mM in the presence of CCCP. Data are expressed relative to the WT (mean ± s.d, n=4). (D) Mitochondrial respiration is preserved in taz1Δ yeast grown in the presence of 10 nM CHX. NADH was used as the electron donor, as described in panel C (n=4). (E) ATP synthesis was measured using NADH as a respiratory substrate in the presence of 1 mM ADP. Data are expressed as mean ± s.d. (n=4) relative to the WT. The data for WT and taz1Δ strains were reported previously .

Figure 4. FIGURE 4: Mitochondrial membrane potential.

Variations in mitochondrial ΔΨ were monitored by fluorescence quenching of Rhodamine 123, using intact, osmotically-protected, mitochondria isolated from WT, taz1Δ, taz1Δ rei1Δ and taz1Δ rpl6bΔ cells grown in CSM containing 0.5% galactose + 2% ethanol at 36°C until a density of 2-3 OD_600nm. The additions were 75 μM ADP, 0.5 μg/ml Rhodamine 123, 75 μg/ml mitochondrial proteins (Mito), 10 μl ethanol (EtOH), 2 mM potassium cyanide (KCN), 4 μM CCCP (carbonyl cyanide-m-chlorophenyl hydrazone) and 4 μg/ml oligomycin (oligo). The shown tracings are representative of four experimental trials. The data for WT and taz1Δ strains were reported previously .

Figure 5. FIGURE 5: Partially decreasing cytosolic translation in taz1Δ yeast preserves a normal production of ROS.

The cells were grown in CSM containing 0.5% galactose + 2% ethanol at 36°C for 48 hours. At the indicated times, ROS levels were measured by flow cytometry using dihydroethidium as a probe. The data are expressed in % relative to the WT at T0 (n=3). Statistical analysis was done with Tukey’s test (*P<0.05; **P<0.01; ***P<0.001; ****P<0.0001). The data for WT and taz1Δ strains were reported previously .

Figure 6. FIGURE 6: CHX improves cell proliferation and viability of human Tafazzin-deficient cells.

These experiments used our previously described HeLa cells in which TAZ gene has been knocked down by RNA interference (shTaz1) and two control cell lines, shWT1 and shTaz1R, in which expression of TAZ1 is not inhibited . (A) Growth curves in 200 µL wells inoculated with 5 000 cells. After reaching a plateau, the cells die and detach from their support. S1, S2 and S3 are the slopes of the proliferation state for each cell lines. (B) Relative slopes (1/h) deduced from the proliferation curves shown in panel A. (C) xCELLigence recording of ShTaz1 proliferation in absence or presence of CHX at a concentration of 50 pM. CHX was added (black arrow) after a 24-hour adhesion step. The cultures were inoculated with 5 000 cells in 200 μL wells. The optimal Cell Index values are indicated along the curves as well as the time (in hours) taken to reach the inflexion point. The horizontal red bar stands for the 48 hours lag phase induced by CHX. Four independent experiments with 4 wells for each growth condition have been done (16 wells in total for establishing mean values). the ±SD is directly drawn on the top of the curves (in black for ShTaz1 and in red for SgTaz1 + 50 pM CHX). (D) xCELLigence recording of ShWT1 proliferation in absence or presence of CHX at a concentration of 50 pM. The cultures were inoculated with 5 000 cells in 200 μL wells. CHX was added before the binding of the cell to the substrate in this case in order to avoid the perturbations induced by the injection of CHX along the trace; the ±SD is directly drawn on the top of the curves (in black for ShWT1 and in red for SgTaz1, there is no significant variation).