TAT Protein Transduction Into Isolated Perfused Hearts
TAT–Apoptosis Repressor With Caspase Recruitment Domain Is Cardioprotective
Background— Linkage of the 11–amino-acid transduction domain of HIV TAT to a heterologous protein allows the protein to be transduced readily into cells.
Methods and Results— In this study, we inserted the apoptosis repressor with caspase recruitment domain (ARC) or β-galactosidase (β-gal) cDNA into the pTAT-hemagglutinin bacterial expression vector to produce genetic in-frame TAT-ARC or TAT-β-gal fusion proteins for use in cell culture and in Langendorff perfusion of adult rat hearts. TAT-β-gal and TAT-ARC were conjugated with Texas Red and could be detected in >95% of cells. TAT-ARC was able to protect H9c2 cells against cell death mediated by hydrogen peroxide, as measured by protection against the loss of mitochondrial membrane potential and preservation of nuclear morphology. Isolated adult hearts were perfused with recombinant TAT-β-gal or TAT-ARC (20 nmol/L) for 15 minutes and then subjected to 30 minutes of global no-flow ischemia, followed by 2 hours of reperfusion. Protein transduction was assessed by Western blotting of cell lysates and cytosolic and mitochondrial fractions and by fluorescence microscopy of Texas Red–conjugated TAT proteins. TAT-β-gal and TAT-ARC readily transduced into perfused hearts and were homogeneously distributed. Infarct size was determined by 2,3,5-triphenyltetrazolium chloride staining, and creatine kinase release was measured. Transduction of TAT-ARC was cardioprotective when administered before global ischemia and reperfusion.
Conclusions— Our results demonstrate that TAT-linked fusion protein transduction into the myocardium is feasible and that transduction of TAT-ARC is protective in cell culture and in the perfused heart.
Received December 7, 2001; revision received May 1, 2002; accepted May 10, 2002.
The HIV TAT protein contains a domain that facilitates protein transduction across cellular membranes.1 A 36–amino-acid domain of TAT was originally chemically cross-linked to heterologous proteins and transduced into cells.2 Other protein transduction domains have been identified in the Antennapedia protein from Drosophila and in herpes simplex virus VP22 protein.3,4⇓ A minimal sequence of 11 amino acids is sufficient for transduction and is characterized by an amphipathic α-helix.5–7⇓⇓ The exact mechanisms of protein transduction across cellular membranes remains unknown; however, TAT-mediated protein transduction has been shown to occur even at 4°C and is receptor independent.3 We used the TAT undecapeptide fused to the β-galactosidase protein and to the apoptosis repressor with caspase recruitment domain (ARC) to assess the potential efficacy of protein transduction in the isolated perfused rat heart model.
ARC was identified by homology cloning on the basis of the caspase recruitment domain.8 ARC is expressed at high levels almost exclusively in heart and skeletal muscle.8 ARC has been reported to selectively interact with caspases 2 and 8 and to inhibit receptor-induced apoptosis.8 ARC has been shown to prevent cytochrome c release in models of hypoxia9 and to inhibit hydrogen peroxide–induced cell death in H9c2 cells.10 In the present study, we used TAT-ARC to assess the potential efficacy of this protein transduction method in H9c2 cells exposed to hydrogen peroxide and in isolated perfused hearts subjected to ischemia and reperfusion (I/R).
TAT-β-galactosidase (TAT-β-gal) and the 6xHis-TAT-HA cloning vector (pTAT-HA, where HA is hemagglutinin) were kindly provided by Dr Steven Dowdy, Washington University, St Louis, Mo. The ARC construct, pcDNA3.1(+)neo-hARC, was generated as previously reported.10 DH5α and BL21(DE3)pLysS bacteria were purchased from Invitrogen. The restriction enzymes KpnI and EcoRI and DNA-related chemicals were obtained from Life Technologies. Ni-NTA resin and PD-10 desalting columns were purchased from Qiagen and Amersham Pharmacia, respectively. Sprague-Dawley rats (250 g, male) were obtained from Harlan Sprague Dawley (San Diego, Calif). Anti-ARC and anti-HA were obtained from Alexis and Santa Cruz Biotechnology, respectively. Goat anti-rabbit horseradish peroxidase and goat anti-mouse horseradish peroxidase antibodies were purchased from Caltag Laboratories. Electrophoresis-grade reagents for polyacrylamide gel electrophoresis and prepackaged gels were purchased from Sigma Chemical Co and Invitrogen, respectively. 2,3,5-Triphenyltetrazolium chloride (TTC) and the creatine kinase (CK) test kit were purchased from Sigma.
Cloning Strategy and Recombinant Protein Expression and Purification
Genetic TAT-ARC fusions were generated by insertion of the ARC open reading frame DNA into the pTAT-HA plasmid. The human ARC cDNA was removed from the pcDNA3.1(+)neo-hARC plasmid by double digestion with KpnI and EcoRI. The hARC open reading frame DNA was then ligated into the pTAT-HA vector by double digestion of the multiple cloning site within pTAT-HA with KpnI and EcoRI. After ligation, plasmids were transformed into DH5α bacteria. The pTAT-HA vector contains an ampicillin resistance marker for selection after transformation, a T7 polymerase promoter, an N-terminal 6-histidine leader before the TAT domain, and an HA tag.7 Individual clones were isolated and analyzed for the correct hARC insert size. The pTAT-HA-hARC plasmid was then transformed into BL21(DE3)pLysS bacteria.
Recombinant TAT–fusion protein expression and purification were performed exactly as described by Becker-Hapak et al.7 In brief, a 500 mL LB ampicillin overnight culture of TAT-ARC of TAT-β-gal was grown in the presence of 100 μmol/L isopropylthiogalactoside (Sigma) at 37°C with shaking. The bacterial pellet was isolated by centrifugation, washed with PBS, resuspended in 10 mL buffer Z (8 mol/L urea, 100 mmol/L NaCl, and 20 mmol/L HEPES, pH 8.0), and sonicated on ice 3 times with 15-second pulses. The sonicate was then clarified by centrifugation at 20 000g at 4°C for 20 minutes. The clarified lysate was then equilibrated in 20 mmol/L imidazole and applied at room temperature to a preequilibrated 25-mL column packed with 5 mL Ni-NTA resin in buffer Z, including 20 mmol/L imidazole. The column was allowed to proceed by gravity flow, and the flow-through was then reapplied. The column was washed with 50 mL of 20 mmol/L imidazole in buffer Z, and the fusion protein was eluted from the Ni-NTA column at concentrations of imidazole of 100 and 250 mmol/L in buffer Z; however, the 100- and 250-mmol/L fractions were pooled and desalted into 1× PBS on PD-10 columns. The fusion proteins were applied in 2.5-mL aliquots and eluted with 3.5 mL PBS. TAT fusion proteins were stored at 4°C and used within 1 week. For visual confirmation of the transduction of TAT-β-gal and TAT-ARC, the fusion proteins were directly labeled with Texas Red-X-succinimidyl ester according to the manufacturer’s protocol (Molecular Probes). The unconjugated dye was removed by a PD-10 desalting column.
Protein Transduction in Cell Culture and Analysis of Mitochondrial Membrane Potential and Nuclear Morphology
The rat heart–derived cell line, H9c2, was seeded in DMEM supplemented with 10% FBS at 3600 cells/cm2 on microscope slides 2 days before transduction with TAT-linked proteins. TAT fusion proteins were added directly to cells in 1× PBS, followed by dilution with serum-free low-glucose DMEM to a final concentration of 100 nmol/L for 1 hour, followed by treatment with hydrogen peroxide. Hydrogen peroxide was diluted from 30% stock immediately before use, applied at a final concentration of 400 to 800 μmol/L, and left on for 8 hours. The TAT proteins were included in the media at 50 nmol/L during hydrogen peroxide treatment. For visual confirmation of the transduction of TAT-hARC or TAT-β-gal, the fusion proteins were directly labeled with Texas Red and transduced in the absence of hydrogen peroxide. Images were captured with a Nikon TE300 fluorescence microscope (Nikon) and digitally collected with a Spot2 digital camera (Diagnostic Instruments).
For analysis of mitochondrial membrane potential and nuclear morphology, cells were loaded with rhodamine 123 (Molecular Probes) at 1 μg/mL for 30 minutes at the end of the treatment period, rinsed once with media, and costained for nuclear detection with Hoechst 33342 at 30 μg/mL for 5 minutes, followed by a rinse of unincorporated dye with 1× PBS. Images of random fields of live cells were captured and digitally collected as described above by using identical parameters. Mitochondrial membrane potential and nuclear condensation were assessed with the use of MetaMorph imaging software (Universal Imaging). Membrane potential was measured as the ratio of total fluorescence intensity of rhodamine 123 to the number of nuclei in each field. Percent nuclear condensation was measured as the number of nuclei that gave images with saturating fluorescence divided by the total number of nuclei in each field. Ten fields of 500 to 1000 cells were scored from each condition. Experiments were repeated 3 times. ANOVA was performed with the Bonferroni post hoc test. A value of P<0.05 was considered significant.
Langendorff Perfusion and Global I/R
The global ischemia protocol was adapted from that of Tsuchida et al.12 All procedures were approved by the Animal Care and Use Committee at The Scripps Research Institute. In brief, the heart was excised from the anesthetized rat and quickly cannulated onto the Langendorff perfusion apparatus. The heart was perfused with Krebs-Ringer buffer (with or without 20 nmol/L TAT protein recirculation) for 15 minutes before I/R episodes. No-flow ischemia was maintained for 30 minutes, and reperfusion was accomplished by restoring flow for 2 hours (unless otherwise indicated). The efficacy of these interventions was verified by measurement of CK release and infarct size measurement.13 Ventricles were processed for the preparation of whole-cell lysates and cytosolic and mitochondrial fractions.
For histological analysis, hearts were perfusion-fixed with 4% formaldehyde, embedded, and thin-sectioned, followed by deparaffinization and nuclear staining with Hoechst 33342 before fluorescence microscopy. Images were captured with a Nikon TE300 fluorescence microscope (Nikon) and digitally collected with a Spot2 digital camera (Diagnostic Instruments).
Isolation of Mitochondria and Cytosol and Western Blot Analysis
Whole-cell lysates were prepared by Polytron (Brinkmann) homogenization in buffer containing 225 mmol/L mannitol, 75 mmol/L sucrose, 1 mmol/L EGTA, 20 mmol/L HEPES-KOH (pH 7.4), 1% Triton X-100, and complete protease inhibitor cocktail (Roche). Lysates were cleared by centrifugation at 20 000g for 20 minutes at 4°C.
Mitochondria and cytosol were prepared as described previously,14 and all procedures were performed at 4°C. Ventricles were minced in MSE buffer (225 mmol/L mannitol, 75 mmol/L sucrose, 1 mmol/L EGTA, and 20 mmol/L HEPES-KOH, pH 7.4) and further homogenized with a Polytron for 5 seconds at maximum power. The homogenate was centrifuged for 10 minutes at 600g, nuclear and cytoskeletal fractions were discarded, and the centrifugation was repeated. The supernatant was centrifuged for 10 minutes at 10 000g to pellet mitochondria. The supernatant (crude cytosol) was further centrifuged for 30 minutes at 100 000g to obtain cytosol.
Protein concentration was determined by the Coomassie blue binding assay (Pierce Chemical Co) with BSA standards. Samples were adjusted to equal protein concentrations and solubilized in SDS-PAGE sample buffer before electrophoresis and transfer to polyvinylidine difluoride nylon membranes for Western blot analysis.
CK Release and Infarct Size Measurements
Coronary effluent was collected from the Langendorff-perfused hearts during the 15-minute stabilization period before ischemia and during the first 15 minutes of reperfusion after ischemia. CK activity was measured according to the manufacturer’s instructions (Sigma), and total units of CK activity released over 15 minutes were reported.
The measurement of infarct size was essentially identical to that detailed by Downey,13 except the method of quantification.15,16⇓ After reperfusion, the heart was frozen, cut into 2-mm-thick rings, and incubated in 1% TTC in PBS (pH 7.0) at 37°C for 20 minutes. After the TTC reaction, tissue slices were fixed with 3.7% formaldehyde in PBS for 20 minutes and washed twice with PBS. Heart slices were digitized and converted to TIFF files and analyzed by Adobe PhotoShop 5.5. The unstained area in each slice was traced manually, and the infarct was calculated as a percentage of the total heart surface area.
One-way ANOVA with Bonferroni multiple comparisons test was applied for the analysis of cell culture results, CK release, and infarct size determinations. All statistical analyses used GraphPad InStat version 4.10 for Windows 98 (GraphPad Software, available at www.graphpad.com). A value of P<0.05 was considered significant.
TAT-β-Gal and TAT-ARC Transduction Into Cultured Cells and Perfused Hearts
Before proceeding with perfusion of the TAT fusion proteins, the efficiency of transduction of TAT-β-gal and TAT-ARC was assessed in cultured cells. H9c2 cells were incubated with TAT-β-gal or TAT-ARC in cell culture media for 1 hour, followed by fresh medium for 30 minutes. TAT proteins were conjugated with Texas Red, and efficient transduction was confirmed by fluorescence microscopy (Figure 1A).
To document the distribution of the TAT proteins in the perfused heart model, they were conjugated with Texas Red and perfused into isolated rat hearts for 15 minutes with recirculation. These procedures were followed by a 15-minute washout period, which was followed by perfusion-fixation. Fluorescence microscopy revealed homogeneous distribution of TAT-β-gal and TAT-ARC across the tissue (Figure 1B).
Transduced TAT-ARC Is Functional in Cell Culture
To verify that the transduced TAT-ARC was functional, we assessed its ability to protect H9c2 cells against cell death mediated by hydrogen peroxide. Adenoviral delivery of ARC to H9c2 cells preserves mitochondrial membrane potential (ΔΨm) and viability of cells exposed to hydrogen peroxide.10 We used TAT proteins (not conjugated with Texas Red) to assess functionality. TAT-ARC, but not TAT-β-gal, efficiently protected H9c2 cells against cell death induced by hydrogen peroxide, measured by loss of ΔΨm and nuclear condensation over a range of 400 to 800 μmol/L (Figure 2).
TAT Protein Transduction in Isolated Perfused Hearts
Western blot analysis of cardiac lysates from hearts perfused with TAT proteins for 15 minutes demonstrated that TAT-β-gal and TAT-ARC were taken up with similar efficiency (Figure 3A). TAT-ARC was noted to be present in whole-cell lysates at a level ≈1.5-fold over the endogenous protein (Figure 3B). TAT-ARC was detected in the cytosolic and mitochondrial fractions (Figure 3C). The abundance of TAT-ARC in these fractions remained unchanged after 30 minutes of ischemia and 15 minutes of reperfusion, suggesting that there was no degradation of the TAT proteins or reverse transduction (out of the cell). Rho GTP dissociation inhibitor (RhoGDI) and voltage dependent anion channel (VDAC) were used in these experiments as markers for cytosol and mitochondria, respectively, and equal protein loading was observed in fractions from all hearts.
Effect of TAT-ARC on Infarct Size and CK Release
CK release was measured in the rat hearts treated as described above, and TAT-ARC, but not TAT-β-gal, reduced CK release by >50% after I/R (Figure 4A). Infarct size was determined after 2 hours of reperfusion. Infarcts in ischemic and reperfused hearts not perfused with any TAT protein were subendocardial and circumferential and averaged 53.6% of the total area (Figure 4B). Hearts perfused with TAT-β-gal and subjected to I/R showed a similar infarct size of 48.2% (Figure 4B). In contrast, hearts perfused with TAT-ARC and subjected to I/R demonstrated smaller infarcts (19.6%, P<0.001). Representative sections are shown (Figure 4C).
The present study illustrates the potential for TAT protein transduction to be applied to studies in the myocardium. Because the protein can be introduced rapidly, it is applicable in the Langendorff model and avoids the need for genetic manipulation either by adenovirus or transgenesis. This overcomes one of the hidden problems of transgenic models, which can be confounded by compensatory responses that extend far beyond the acute effects of the modified gene. However, this approach poses its own set of technical challenges. A substantial protein gradient across the membrane is necessary, although protein renaturation will help to maintain a favorable gradient. Therefore, a large amount of recombinant protein is required for these experiments.
We were impressed with the homogeneity of the uptake of the Texas Red–labeled protein across the myocardium, because we expected most of the protein to accumulate in the endothelial cells. This is likely due to the fact that the protein can just as easily transduce out of the cell if it has not been captured and refolded by cellular chaperonins.7 The amphipathic α-helix may interact with intracellular membranes, raising the concern that this might result in mislocalization of proteins to a variety of organelles. Once the TAT protein enters the cell, it must be renatured by cellular proteins to be retained and to regain function. Therefore, it was essential to demonstrate an effect of the transduced protein. The protection against hydrogen peroxide–induced cell death and the reduction in CK release and in infarct size by TAT-ARC are consistent with its ability to prevent mitochondrial dysfunction after hypoxia or oxidative stress, as recently reported.10 This is not likely to be due to a nonspecific effect of protein loading, inasmuch as TAT-β-gal was not protective in either of these models. The relatively modest protective effect of TAT-ARC may be due to inefficient uptake of the protein, incomplete renaturation within the cell, or mislocalization of the protein. Although adult heart and H9c2 cells express substantial amounts of ARC, overexpression by transfection or adenoviral gene delivery in cell culture models confers additional protection against hypoxia or oxidative stress.9,10⇓ Our observations with TAT-ARC are consistent with this.
TAT protein transduction represents a potentially valuable tool for exploring molecular mechanisms in the heart. There are some limitations: proteins with membrane-spanning domains may not be renatured or localized properly or efficiently, and structural proteins may not accumulate sufficiently to demonstrate an effect. However, the potential for exploring signal transduction pathways and for probing protein-protein interactions makes this an exciting technology. Schwarze et al6 have shown that intraperitoneal injection of TAT-β-gal results in uptake in a wide variety of tissues (including heart and brain); thus, it has already been shown to be applicable to the whole animal. Therefore, it may be possible to compare the effects of a protein introduced into an adult mouse with the effects of a chronically expressed transgene to unravel the confounding effects of compensation for transgene expression. It is of great value in models such as the isolated perfused heart, in which transgenesis is not an option and in which adenoviral gene delivery is not feasible because of the extended time needed for gene expression after infection. Moreover, the therapeutic potential of this approach should not be overlooked.
This work was supported by National Institutes of Health grants AG-13501 (to Dr Gottlieb) and HL-60590 (to Dr Gottlieb) and postdoctoral training grant DK-07022 (to Drs Gustafsson and Williams). DNA sequencing was performed in the Molecular and Experimental Medicine DNA Core Facility, supported by the Sam and Rose Stein Endowment Fund. This is manuscript No. 14613-MEM of The Scripps Research Institute.
Derossi D, Joliot AH, Chassaing G, et al. The third helix of the Antennapedia homeodomain translocates through biological membranes. J Biol Chem. 1994; 269: 10444–10450.
Schwarze SR, Ho A, Vocero-Akbani A, et al. In vivo protein transduction: delivery of a biologically active protein into the mouse. Science. 1999; 285: 1569–1572.
Koseki T, Inohara N, Chen S, et al. ARC, an inhibitor of apoptosis expressed in skeletal muscle and heart that interacts selectively with caspases. Proc Natl Acad Sci U S A. 1998; 95: 5156–5160.
Neuss M, Monticone R, Lundberg MS, et al. The apoptotic regulatory protein ARC (apoptosis repressor with caspase recruitment domain) prevents oxidant stress-mediated cell death by preserving mitochondrial function. J Biol Chem. 2001; 276: 33915–33922.
Deleted in proof.
Tsuchida A, Liu Y, Liu GS, et al. Alpha 1-adrenergic agonists precondition rabbit ischemic myocardium independent of adenosine by direct activation of protein kinase C. Circ Res. 1994; 75: 576–585.
Downey JM. Measuring infarct size by the tetrazolium method. Available at: http://www.usouthal.edu/ishr/help/ttc./ Accessed in 2000.
Klein HH, Puschmann S, Schaper J, et al. The mechanism of the tetrazolium reaction in identifying experimental myocardial infarction. Virchows Arch. 2001; 393: 287–297.