Incorporating a Polyethyleneglycol Linker to Enhance the Hydrophilicity of Mitochondria-Targeted Triphenylphosphonium Constructs

Abstract

The targeting of bioactive molecules and probes to mitochondria can be achieved by coupling to the lipophilic triphenyl phosphonium (TPP) cation, which accumulates several hundred-fold within mitochondria in response to the mitochondrial membrane potential (Δψ_m ). Typically, a simple alkane links the TPP to its "cargo", increasing overall hydrophobicity. As it would be beneficial to enhance the water solubility of mitochondria-targeted compounds we explored the effects of replacing the alkyl linker with a polyethylene glycol (PEG). We found that the use of PEG led to compounds that were readily taken up by isolated mitochondria and by mitochondria inside cells. Within mitochondria the PEG linker greatly decreased adsorption of the TPP constructs to the matrix-facing face of the mitochondrial inner membrane. These findings will allow the distribution of mitochondria-targeted TPP compounds within mitochondria to be fine-tuned.

Keywords: biological membrane; lipophilic cation; mitochondria-targeting; polyethylene glycol.

Figure 1

Mitochondria‐targeted delivery of cargoes conjugated to TPP compounds. (A) Uptake of a typical mitochondria‐targeted TPP compound. After accumulation into the cytoplasm driven by the plasma membrane potential (Δψ_p) the compound is further taken up into the mitochondrial matrix in response to the mitochondrial membrane potential (Δψ_m). Within the mitochondria matrix these compounds are largely adsorbed on to the matrix‐facing surface of mitochondrial inner membrane (Inset). Compounds can be designed to release bioactive cargo inside the mitochondria, for example by linkage via a cleavable ester bond. (B, C) The structures of the TPP compounds are shown. (D) Relative hydrophobicity of TPP compounds conjugated to a hydroxyl. Each compound (10 nmol) was analyzed by RP‐HPLC.

Figure 2

Uptake of TPP‐conjugated compounds by energized mitochondria. Compounds (10 μM) were incubated with energized rat liver mitochondria (2 mg protein/mL)±FCCP (0.5 μM) for 5 min before pelleting the mitochondria by centrifugation, extracting the pellet, and quantifying the amount of compound present by RP‐HPLC. (A) Representative RP‐HPLC traces from incubations with TPP₁₁‐OH or TPP_4EG‐OH±FCCP (B, C) Quantification of peak areas of hydroxyl derivatives (B) and benzyl derivatives (C). (D) Accumulation ratios (ACRs) of TPP₈‐OH and TPP_3EG‐OH relative to that of TPMP. Mitochondria were incubated with TPP₈‐OH or TPP_3EG‐OH (10 μM) along with TPMP (3 μM) and different concentrations of FCCP (0–0.5 μM) to establish a range of Δψ_m values. The mitochondria and supernatants were then isolated and analyzed by RP‐HPLC to indicate the relative concentrations of the compounds in the two compartments, and from the ACR values were calculated. Data are means±SEM of three independent experiments. Statistical significance was assessed by two‐way ANOVA with Dunnett's correction for multiple comparisons; ***p<0.01, ****p<0.001.

Figure 3

Real‐time measurement of uptake of TPP‐conjugated compounds by energized mitochondria. After addition of rat liver mitochondria (2 mg protein/mL) to the electrode chamber, electrode response was calibrated by 5×1 μM additions of the TPP compound. Mitochondria were then energized by addition of succinate (10 mM) and where indicated FCCP (0.5 μM) was added. (A) TPP₁₁‐OH, (B) TPP₈‐OH, (C) TPP_4EG‐OH, and (D) TPP_3EG‐OH. Data are typical traces repeated at least 3 times. (E) Normalized compound uptake. The Ln(electrode response) was divided by the difference between the Ln response at t=0 (0 % uptake) and that at t=60 (100 % uptake) to generate plots of uptake as % of maximum over time. Data are shown for 60 seconds after the addition of succinate (t=0). Data are means±SEM (shading) of three independent experiments and traces are offset for clarity. Pseudo first order rate constants (k) for compound uptake were estimated from the slopes of plots of the Ln values of the normalized uptake data against time over the first 10s of uptake.

Figure 4

Enzymatic hydrolysis of TPP‐conjugated acetyl esters. (A) TPP acetyl esters (200 μM) were incubated with rat liver cytosol (1 mg protein/mL) for various times before extraction and analysis by RP‐HPLC. (B) TPP acetyl esters (10 μM) and internal standard TPMP (3 μM) were incubated with rat heart mitochondria (0.5 mg protein/mL) before pelleting mitochondria, extracting the pellets and analyzing by RP‐HPLC. The data show the peak areas of the ester and hydroxyl compounds within mitochondria. (C) Hydrolysis rates of the TPP acetyl esters within mitochondria. Data from the experiments described in Panel B were analyzed at each time point by dividing the peak areas of residual acetyl derivatives of the acetyl esters by that of the sum of the peak areas of the acetyl esters and its hydroxyl derivatives. These values are expressed as % of amount present at 2 min. All data are means±SEM of three independent experiments.

Figure 5

The uptake of LuciferTPPAlkyl and LuciferTPPGlycol by mitochondria within cells. HeLa cells stably expressing the mitochondrial outer membrane protein TOMM20 tagged with the fluorescent protein mCherry (HeLa‐TOMM20‐mCh) (white) were used to visualize mitochondria. The fluorescence of Lucifer Yellow (green) was observed here 500s after the addition of 250 nM of either compound. Scale bars=15 μm.

Figure 6

Distribution of LuciferTPPAlkyl and LuciferTPPGlycol. HeLa cells stably expressing the mitochondrial outer membrane protein TOMM20 tagged with the fluorescent protein mCherry (HeLa‐mCherry‐TOMM20) were used to visualize mitochondria (magenta). Fluorescence (green) from LuciferTPPAlkyl (A) and LuciferTPPGlycol (B) was observed within mitochondrial matrix after incubating the cells for 500s with either compound (250 nM). Fluorescence intensity values (arbitrary unit) for mCherry‐TOMM20 and the respective dyes were measured along the bisecting line (2 μm; yellow) and plotted against distance (μm) in panels (C) and (D) for LuciferTPPAlkyl and LuciferTPPGlycol, respectively. Scale bar=15 μm; Scale bar for zoom=5 μm.

Figure 7

(A, B) The mitochondrial fluorescent intensity of Lucifer Yellow, normalized to that of HeLa‐TOMM20‐mCh, over time at various concentrations for LuciferTPPAlkyl (A) and for LuciferTPPGlycol (B). LuciferTPP was added at t=150. (C) Data for the mitochondrial accumulation of 500 nM of LuciferTPPAlkyl and LuciferTPPGlycol from Panels (A) & (B) are replotted on the same scale. (D) Loss of LuciferTPPAlkyl and LuciferTPPGlycol from mitochondria within cells by the addition of uncoupler FCCP (500 nM). Cells were incubated with the compounds at 10 nM of LuciferTPPAlkyl or 250 nM of LuciferTPPGlycol for 500 s, media was then removed, cells washed and reincubated with new medium before addition of FCCP (500 nM). Data in A–D are means±SEM of three independent experiments.