A TAT-frataxin fusion protein increases lifespan and cardiac function in a conditional Friedreich's ataxia mouse model

Abstract

Friedreich's ataxia (FRDA) is the most common inherited human ataxia and results from a deficiency of the mitochondrial protein, frataxin (FXN), which is encoded in the nucleus. This deficiency is associated with an iron-sulfur (Fe-S) cluster enzyme deficit leading to progressive ataxia and a frequently fatal cardiomyopathy. There is no cure. To determine whether exogenous replacement of the missing FXN protein in mitochondria would repair the defect, we used the transactivator of transcription (TAT) protein transduction domain to deliver human FXN protein to mitochondria in both cultured patient cells and a severe mouse model of FRDA. A TAT-FXN fusion protein bound iron in vitro, transduced into mitochondria of FRDA deficient fibroblasts and reduced caspase-3 activation in response to an exogenous iron-oxidant stress. Injection of TAT-FXN protein into mice with a conditional loss of FXN increased their growth velocity and mean lifespan by 53% increased their mean heart rate and cardiac output, increased activity of aconitase and reversed abnormal mitochondrial proliferation and ultrastructure in heart. These results show that a cell-penetrant peptide is capable of delivering a functional mitochondrial protein in vivo to rescue a very severe disease phenotype, and present the possibility of TAT-FXN as a protein replacement therapy.

Figure 1.

Expression and application of TAT-human frataxin (TAT–FXN). (A) Domains of TAT–FXN fusion protein. Expression is driven by the T7 promoter, and purification is based on a 6X-His tag. The TAT peptide sequence is expanded and placed at the N terminus of the human precursor FXN cDNA. (B) 5-IAF-labeled TAT–FXN incubated with FRDA human fibroblasts and imaged live by confocal microscopy. 5-IAF primarily reacts with sulfhydryl groups, but can also react with methionine, histidine and potentially tyrosine. Mitochondria are stained with CMX-rosamine. TAT–FXN is green (panel 1), mitochondria are red (panel 2) and signal co-localization is yellow (panel 3). Panels 1–3 are 120 h after a 3 h exposure to TAT–FXN.

Figure 2.

Processing of TAT–FXN by mitochondria processing peptidase (MPP). (A) Western blot showing progressive cleavage of TAT–mMDH–eGFP and probed with anti-GFP. Lanes 0 (starting condition), 1, 3 h and overnight (o/n) incubation with MPP at 37°C. Upper arrow is precursor band, and lower arrow is processed band. (B) Western blot of TAT–FXN cleavage by MPP and probed with anti-human FXN monoclonal antibody. Lanes as in (A). Upper arrow is precursor band and lower arrows show intermediate and mature processed FXN. Lanes include MPP (M), which was run on the same gel but is not contiguous, and precursor TAT–FXN (F) not exposed to MPP. (C) Reaction products of (A) were separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) and stained by Coomassie. Lanes are as marked in (A). Cartoon of precursor TAT–mMDH–eGFP is shown beside precursor band (upper arrow), and cartoon of processed band is shown next to lower arrow. Light grey is the His–TAT moiety, dark grey is the mMDH mitochondrial targeting sequence (MTS) and white is eGFP (drawn to scale). Boxed bands were sequenced by Edman degradation and the first 5 sequenced amino acids are shown as underlined single letter codes (Panels C and D). (D) Reaction products of (B) were separated by 12% SDS–PAGE, and boxed bands were sequenced. Upper cartoon shows precursor TAT–FXN (drawn to scale) with shaded regions as in (C) except that dark grey is the native FXN MTS. Cartoon of processed intermediate band is shown with the first 5 amino acids identified by Edman degradation. Numbers next to first amino acid in (C) and (D) refer to the position of the amino acid in the native mMDH and FXN precursor proteins, respectively.

Figure 3.

TAT–FXN binds iron and rescues FRDA cells from iron-oxidant stress. (A) Basis for an iron-binding assay for TAT–FXN. (B) TAT–FXN (20 or 40 μg/ml) was incubated in PBS in the presence of 5 μm each of ferrous sulfate (Fe) and HQ. Superoxide generation was measured by DCF fluorescence and expressed as arbitrary fluorescence units (y-axis). BSA (20 μg/ml) was used as a negative control and EDTA was used as a positive control. Data represents mean (±SD) of nine replicates of each condition. Significant statistics marked above columns. (C) FRDA fibroblasts (cell line GM04087, Coriell Institute) were grown to 80% confluence and treated with either 20 or 40 μg/ml of TAT–FXN, or with PBS vehicle (–TAT–FXN), after which the culture media was changed to control conditions (TAT–FXN removed from media). Cells were exposed to Fe/HQ for 5 h, or cultured in standard media (Control). (D) Fibroblasts from a healthy age- and sex-matched control (clear bar) (cell line GM01661, Coriell Institute) or from FRDA fibroblasts (black bars) were treated as in (C), washed with PBS and assayed for caspase-3 activation after normalizing for total protein loading. The data represents the mean (±SD) of three replicates of each condition analyzed by Student's t-test. * = P < 0.001 versus healthy control. ^#P < 0.005 vs FRDA and ^ΔP < 0.001 versus FRDA + Fe/HQ.

Figure 4.

Survival analysis. (A) Treatment begun at 3 days of age. Groups: KO PBS (3d) = Fxn-KO mouse with PBS beginning at 3d of age. KO TAT–FXN (3d) = Fxn-KO mouse with TAT–FXN beginning at 3d of age. Ctl PBS = heterozygous age-matched littermates with PBS injections beginning at 3d. Study terminated at 60 days (=Δ). (B) Treatment begun at 12 days of age. Groups are as follows: KO no PBS (12d) = Fxn-KO mouse without PBS treatment starting at 12d; KO TAT–FXN (12d) = Fxn-KO mouse with TAT–FXN treatment starting at 12d; Ctl no PBS = heterozygous age-matched littermates without PBS injections. Filled circles represent censored event. Study continued to end of life. (C) Survival of groups to 31 days of age and beyond with significance (P ≤ 0.05) for relevant groups using Student's t-test. The 12d groups are adjusted for censoring prior to 31 days, which lowers their total numbers when compared with (B). (D) Two-month-old heterozygous female mice were injected with PBS (100 μl/injection), TAT–FXN or TAT-mMDH-eGFP, each at 2 mg/kg/week (total protein) for 2 months per IP route. Tissues were fixed in 10% formalin and stained by hematoxylin and eosin.

Figure 5.

Growth velocity was calculated for the three groups as follows: . The growth velocity for each animal was calculated and averaged for each group, and then plotted as median, 75 and 25% (upper and lower box limits). Groups are as follows: Ctl no PBS = heterozygous age-matched littermates without PBS injections. KO TAT–FXN (12d) = FXN-KO mouse with TAT–FXN treatment starting at 12d; KO no PBS (12d) = FXN-KO mouse without PBS treatment starting at 12d.

Figure 6.

(A) Aconitase-specific activity in heart. Ctl PBS (3d) = heterozygous littermates receiving PBS from 3 days of life. KO TAT–FXN = FXN-KO (3d) mice treated with TAT–FXN beginning at 3 days of age. KO PBS = FXN-KO (3d) mice treated with PBS beginning at 3 days of age. Aconitase-specific activity on y-axis. N = 3 per group with triplicate assays, and analyzed by Student's t-test. (B) Western blot of mitochondrial aconitase protein from the mouse hearts in ‘C’ containing 25 μg of total protein per lane. Lanes 1–3 = Ctl, lanes 4–6 = KO TAT–FXN, lanes 7–9 = KO PBS. The lanes were scanned and signal quantified using MetaMorph software (reported in text). M_r in kDa for (B) and (C). (C) Recovery of processed TAT–FXN from heart (small arrow). Lane 0 = injected TAT–FXN (large arrow). Lanes 1–3 are FXN-KO animals receiving TAT–FXN (12d). Lanes 4–6 are FXN-KO no PBS (12d) animals. Lanes 7–9 are heterozygous littermates [Ctl no PBS (12d)] and were run on the same gel but are not contiguous.

Figure 7.

Echocardiography of FRDA animals. (A) Ctl = heterozygous littermates, no PBS (12d). KO TAT–FXN = FXN-KO-treated with TAT–FXN (12d). KO no PBS = FXN-KO without PBS (12d). The M-mode column shows the internal dimension of the left ventricle between the interventricular septum (IVS) and left ventricular posterior walls (LVPW). Scale in millimeters is on the right, and EKG is on the bottom. The aorta column shows the Doppler velocity of blood flow in the ascending aorta for all three groups. Note that the velocity scale for the Ctl animal is 1 m/s, whereas it is cm/s for the other two groups. The mitral column shows the Doppler mitral inflow pattern and velocity for all three groups. E-wave and A-wave are marked. The velocity scale on the right is cm/s for the Ctl and KO no PBS animals, and 1 m/s for the KO TAT–FXN animal. (B) The EKG from a representative animal in each of the three groups is shown. The heart rate was determined from the EKG of each mouse at the same time point during echocardiography. Note that the time scale (bottom) is in milliseconds. (C) The mean heart rate of all three groups analyzed by one-way ANOVA with pairwise multiple comparison procedures (Holm–Sidak method).

Figure 8.

(A) EM of heart. Bar = 1 μm in all panels. 1 = Ctl PBS (3d). 2 = KO TAT–FXN (3d). 3 = KO PBS (3d). (B) Ratio of planimetry of mitochondria (Mito) to sarcomere (Sarc) area was determined using Photoshop to calculate area from EM. (C) Ratio of mitochondrial complex I protein (NDUFA9) to sarcomere α-actinin protein. All mice in (A)–(C) received injections beginning at 3 days of age (3d group) and all mice were assayed at 28 days of age. (D) Apoptosis counts of cardiomyocytes (CMs) in the heart from Ctl PBS (3d) at 28 days, KO TAT–FXN (3d) and KO PBS (3d) both at 26 ± 2 days. (E) Micrograph of caspase-3 activation in a 28d KO PBS (3d) heart. Arrow points to caspase-3-positive cardiomyocyte.