Bioconjugate Chem. 2004, 15, 475−481
475
Arginine Carrier Peptide Bearing Ni(II) Chelator to Promote Cellular Uptake of Histidine-Tagged Proteins Shiroh Futaki,*,†,‡ Miki Niwa,† Ikuhiko Nakase,† Akiko Tadokoro,† Youjun Zhang,† Makoto Nagaoka,† Naoya Wakako,† and Yukio Sugiura† Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan, and PRESTO, Japan Science and Technology Agency, Kawaguchi, Saitama 332-0012, Japan. Received October 1, 2003; Revised Manuscript Received February 7, 2004
Arginine-rich peptide-mediated protein delivery into living cells is a novel technology for controlling cell functions with therapeutic potential. In this report, a novel approach for the intracellular delivery of histidine-tagged proteins was introduced where a Ni(II) chelate of octaarginine peptide bearing nitrilotriacetic acid [R8-NTA-Ni(II)] was used as a membrane-permeable carrier molecule. Significant internalization of histidine-tagged enhanced green fluorescent protein (EGFP) into HeLa cells was observed by confocal microscopic observation in the presence of R8-NTA-Ni(II). Nuclear condensation characteristic in apoptotic cell death was also induced in the cells treated with a histidine-tagged apoptosis-inducing peptide [pro-apoptotic domain peptide (PAD)], indicating that the cargo molecules really went through the membrane to reach the cytosol. The apoptosis-inducing activity of the peptide thus delivered was compared with that of the PAD peptide covalently connected with the octaarginine peptide.
INTRODUCTION
Peptide-mediated protein delivery into living cells has become recognized as a powerful tool for controlling cellular function and as a novel methodology in pharmaceutical research (1-3). The cellular introduction of the cargo molecules is easily conducted by simply conjugating cargo proteins with a peptide having a carrier function and adding these chimera molecules to the culture media. The internalization can be accomplished in a few minutes to substantially all cells. Proteins of molecular weight as high as ∼120 kDa were introduced by this methodology (4). One of the typical peptides to have this function is that derived from HIV-1 Tat protein [HIV-1 Tat (48-60): GRKKRRQRRRPPQ] (5). The peptide is highly basic and rich in arginines and lysines. We and others have shown that arginine or its guanidino moiety play a crucial role in exerting the carrier function (6-10). Using this methodology, various cellular functions such as the cell cycle and signal transduction have been successfully regulated (2). This methodology is thus regarded as an effective tool for understanding and controlling cellular functions. On the other hand, the histidine-tag method is known to be one of the most efficacious means for the purification of in vivo produced proteins (11). The trick in this method is to prepare proteins of interest so as to have an extra tag peptide segment comprising a continuous array of histidines of about six residues. By elution of the cell extracts containing the histidine-tagged proteins of interest through a column with immobilized Ni(II), the proteins are retained in the columns by chelating with Ni(II), whereas other molecules that lack histidine tags are not. Washout of nonspecifically adsorbed materials * Corresponding author. Fax: +81-774-32-3038, phone: +81774-38-3211; e-mail:
[email protected]. † Kyoto University. ‡ PRESTO, Japan Science and Technology Agency.
from the column followed by the elution with an imidazole-containing buffer will yield the desired purified proteins. This method is especially useful for the purification of a small amount of products from other cellular components, and various plasmids and kits are now available for the efficient production and purification of these histidine-tagged proteins (12). This method would be applicable not only for the purification of proteins but also for the detachable cross-link formation between proteins. In this report, we have developed a novel Ni(II) chelate carrier peptide and exemplified that histidine-tagged proteins were successfully delivered into the cells together with the Ni(II) chelate of octaarginine bearing nitrilotriacetic acid (NTA). EXPERIMENTAL PROCEDURES
Preparation of Octaarginine-Bearing Nitrilotriacetic Acid (R8-NTA). Preparation of the above compound was conducted as shown in Figure 1. The peptide chain was constructed by Fmoc-solid-phase peptide synthesis (Fmoc ) 9-fluorenylmethyloxycarbonyl) (13) on a 2-chlorotrityl chloride resin (14) using the standard protocol of a Shimadzu PSSM-8 peptide synthesizer. A combination of benzotriazole-1-yloxytrispyrrolidinophosphonium hexafluorophosphate (PyBOP) (15), 1-hydroxybenzotriazole (HOBt), and 4-methylmorpholine (NMM) was employed for a coupling system. The N-terminal was acetylated using acetic anhydride in the presence of NMM. Treatment of the peptide resin with acetic acid (AcOH)-trifluoroethanol (TFE)-dichloromethane (DCM) (1:1:8) at 20 °C for 1 h gave a protected peptide segment Ac-[Arg(Pbf)]8-Gly-OH [Ac ) acetyl; Pbf ) NG-2,2,4,6,7pentamethyldihydrobenzofuran-5-sulfonyl (16)]. After Ac[Arg(Pbf)]8-Gly-OH was treated with diisopropylcarbodiimide (DICDI) and N-hydroxysuccinimide (HOSu) (2 equiv each) in DMF at 25 °C for 2 h to form the corresponding succinimidyl ester, 5 equiv of N-(5-amino-
10.1021/bc034181g CCC: $27.50 © 2004 American Chemical Society Published on Web 05/04/2004
476 Bioconjugate Chem., Vol. 15, No. 3, 2004
Futaki et al.
Figure 1. Design and synthesis of octaarginine bearing nitrilotriacetic acid (R8-NTA).
1-carboxypentyl)iminodiacetic acid (Dojin, Japan) in H2O-DMF (1:3) was added to the mixture and allowed to react overnight. After evaporation of the solvent, the mixture was directly treated with trifluoroacetic acid (TFA)-ethanedithiol (EDT) (95:5) at 25 °C for 2 h. HPLC purification of the sample gave the desired compound. Yield from the starting resin, 31%. MALDI-TOFMS: 1611.7 [calcd for (M + H)+: 1611.8]. Retention time in HPLC, 16.2 min [Column, Cosmosil 5C18-AR-II (10 × 250 mm); eluate, A ) H2O containing 0.1% CF3COOH, B ) CH3CN containing 0.1% CF3COOH; gradient, 5-65% B in A over 40 min; flow, 2 mL/min; detection, 220 nm]. Rhodamine-Labeled Octaarginine Bearing Nitrilotriacetic Acid (Rho-R8-NTA). The title compound was similarly prepared as described above, except that the N-terminus of the protected octaarginine was labeled with tetramethylrhodamine using 5(6)-carboxytetramethylrhodamine N-hydroxysuccinimide ester (Sigma). The peptide corresponding to the main peak on HPLC was fractionated. Yield from the starting resin, 31%. MALDITOFMS: 2040.8 [calcd for (M + H)+: 2040.3]. Retention time in HPLC, 17.5 min [column, Symmetry 300 C18 (4.6 × 150 mm); eluate, A ) H2O containing 0.1% CF3COOH, B ) CH3CN containing 0.1% CF3COOH; gradient, 10-50% B in A over 40 min; flow, 1 mL/min; detection, 215 nm]. Preparation of Histidine-Tagged Enhanced Green Fluorescent Protein [H6-EGFP] and R8-Bearing EGFP (R8-EGFP). The EGFP protein bearing a (His)6 segment and a thrombin cleaving site on its N-terminus was prepared as follows. The gene of the EGFP derivative was created by PCR. The 720 bp fragment (679-1398) of pEGFP-N1 (Clontech) containing the EGFP domain was amplified by primers possessing the sequences for BamH I followed by (His)6 and thrombin sites and for the EcoR I site. The amplified fragment was cleaved by BamH I and EcoR I and cloned into pEV29b, which is a derivative of pET29b (Novagen) and has cloning sites identical to that of pEV3b (17). The sequence of the plasmid for H6-EGFP thus obtained was confirmed by a GeneRapid DNA sequencer (Amersham Biosciences). The H6-EGFP protein was overexpressed in the E. coli strain BL21(DE3)pLysS in soluble form and purified by a His‚ Bind Quick Column (Novagen) to give a single band in SDS-PAGE. R8-EGFP was similarly prepared using a PCR primer where the synthetic (Arg)8 gene was also
included between the thrombin site and the EGFP domain in the PCR primers. H-(His)6-(Gly)2-D(KLAKLAK)2-NH2 [histidine-tagged pro-apoptotic domain peptide, H6-PAD]. The peptide chain was constructed by Fmoc-solid-phase peptide synthesis on a Rink amide resin (18) by a Shimadzu PSSM-8 synthesizer using a PyBOP/HOBt/NMM coupling system. Fmoc-His(Trt), Fmoc-Gly, Fmoc-D-Lys(Boc), Fmoc-D-Leu, and Fmoc-D-Ala were employed as amino acid derivatives (Trt ) trityl, Boc ) tert-butyloxycarbonyl). Treatment of the peptide resin with TFA-EDT (95:5) and HPLC purification of the deprotected sample gave a pure peptide. Yield from the starting resin, 73%. MALDITOFMS: 2460.3 [calcd for (M + H)+: 2461.0]. Retention time in HPLC, 20.9 min [column, Cosmosil 5C18-AR-II (10 × 250 mm); eluate, A ) H2O containing 0.1% CF3COOH, B ) CH3CN containing 0.1% CF3COOH; gradient, 5-95% B in A over 45 min; flow, 2 mL/min; detection, 215 nm]. H-(Arg)8-(Gly)2-D(KLAKLAK)2-NH2 [R8-PAD]. The title peptide was prepared similarly to H6-PAD. Yield from the starting resin, 35%. MALDI-TOFMS: 2886.5 [calcd for (M + H)+: 2886.6]. Retention time in HPLC, 20.8 min [column, Cosmosil 5C18-AR-II (10 × 250 mm); eluate, A ) H2O containing 0.1% CF3COOH, B ) CH3CN containing 0.1% CF3COOH; gradient, 5-95% B in A over 45 min; flow, 2 mL/min; detection, 215 nm]. H-D(KLAKLAK)2-NH2 (PAD). The title peptide was prepared similarly to H6-PAD. Yield from the starting resin, 39%. MALDI-TOFMS: 1524.0 [calcd for (M + H)+: 1524.0]. Retention time in HPLC, 16.9 min [column, Cosmosil 5C4-AR-300 (4.6 × 150 mm); eluate, A ) H2O containing 0.1% CF3COOH, B ) CH3CN containing 0.1% CF3COOH; gradient, 5-95% B in A over 40 min; flow, 1 mL/min; detection, 215 nm]. Cell Culture. Human cervical cancer-derived HeLa cells (obtained from Riken BRC Cell Bank) were maintained in alpha-minimum essential medium (R-MEM) (Invitrogen) with 10% heat-inactivated calf serum (Invitrogen). Cells were grown on 60-mm dishes and incubated at 37 °C under 5% CO2 to approximately 70% confluence. A subculture was performed every 3-4 days. Protein Internalization and Microscopic Observation. For each assay, 5 × 104/mL cells were plated on a 35-mm glass-bottom dish (Iwaki) (2 mL/dish) and cultured for 48 h to attain complete adhesion. A typical
Arginine Carrier Peptide Bearing Ni(II) Chelator
experiment employing 50 µM of R8-NTA-Ni(II) and 10 µM of H6-EGFP was conducted as follows. Prior to incubating with cells, R8-NTA in phosphate-buffered saline (PBS) (500 µM, 20 µL), NiCl2 in H2O (500 µM, 20 µL), and R-MEM containing 10% heat-inactivated calf serum (140 µL) were mixed together and allowed to stand for 15 min. H6-EGFP in PBS (100 µM, 20 µL) was then added to the above mixture and allowed to stand for another 1 h. The cells were incubated with the above mixture (total volume, 200 µL) at 37 °C for 3 h. The cells were washed three times with PBS. Distribution of the EGFP was analyzed using a confocal scanning laser microscope LSM 510 (Zeiss) equipped with an ×40 lens without fixing the cells. When the cells were treated with Rho-R8-NTA-Ni(II)/H6-EGFP complex, the complex was similarly prepared as stated above and applied to a Microcon YM-10 centrifugal filter (normal molecular weight limit 10 000) (Millipore) to remove excess RhoR8-NTA-Ni(II) before addition to the cells. When the cells were treated with a complex of R8-NTA-Ni(II) with H6EGFP at 4 °C, the cells and the medium containing R8NTA-Ni(II)/H6-EGFP were stored at 4 °C for 1 h prior to mixing with each other. After treatment for 1 h, the cells were washed with ice-cold PBS and subjected to the microscopic observation. In the case of observing nuclear condensation, the cells were incubated with the PAD peptides in R-MEM with 10% heat-inactivated calf serum at 37 °C for 24 h under 5% CO2, and the medium was changed to a fresh one. Hoechst 33258 was then added to the medium (final concentration, 50 µM), and the cells were incubated for another 30 min at 37 °C under 5% CO2. Cells were washed twice with PBS, soaked in a fresh R-MEM, and analyzed on an Olympus IX-70 fluorescence microscope using an × 20 lens. Flow Cytometry. To analyze the internalization of EGFP proteins by FACS, 1.5 × 105 HeLa cells in fresh culture medium (1.5 mL) were plated on a 12-well microplate (Iwaki) and cultured for 48 h in R-MEM containing 10% heat-inactivated calf serum. After complete adhesion the cells were incubated at 37 °C for 1 h with fresh medium (400 µL) containing R8-NTA-Ni(II)/ H6-EGFP prior to washing with PBS. The cells were then treated with 0.01% trypsin (Invitrogen) (400 µL) at 37 °C for 10 min prior to adding 600 µL of PBS. The cells were centrifuged at 2000 rpm for 5 min, and after removing the supernatant, they were washed with 1 mL of PBS and centrifuged at 2000 rpm for 5 min. After this washing cycle was repeated, the cells were suspended in PBS (1 mL) and subjected to fluorescence analysis on a FACScalibur (BD Biosciences) flow cytometer using 488nm laser excitation and an 515-545 nm emission filter. DNA Fragmentation Assay. HeLa cells were treated with a mixture of R8-NTA-Ni(II) (50 µM) and H6-PAD (10 µM) as described above for 24 h. DNA ladder formation was then examined by isolating oligonucleosomal DNA from cellar extracts and analyzing DNA by ethidium bromide staining after electrophoresis through 2% agarose gels, using Quick Apoptotic DNA Ladder Detection Kit (BioVision). Evaluation of Apoptotic Cells by FITC-AnnexinV. Cells were stained with FITC-Annexin-V according to the manufacturer’s specification using Vybrant Apoptosis Assay kit #3 (Molecular Probes). For analysis, 1.5 × 105 cells were plated on a 12-well plate as described above and incubated for 48 h. Cells were then incubated with fresh medium containing R8-NTA-Ni(II)/H6-PAD complex for 2 h, harvested by trypsinization and then washed with PBS. Cells were suspended in Annexin-V binding
Bioconjugate Chem., Vol. 15, No. 3, 2004 477
buffer (10 mM HEPES, 140 mM NaCl, and 2.5 mM CaCl2, pH 7.4) (200 µL), and then incubated with FITCAnnexin-V (FITC ) fluorescein isothiocyanate) (10 µL) and the propidium iodide (PI) working solution (100 µg/ mL, 2 µL) at room temperature for 15 min. After the incubation period, the cells were diluted with Annexin-V binding buffer (800 µL). Apoptotic cells, which were positive for FITC-Annexin-V and negative for PI staining, were analyzed by FACS. MTT Assay. The MTT [) 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl-2H-tetrazolium bromide] assay was conducted basically in the same manner as already reported (6). Cells (1.5 × 103/well) were cultured on 96-microtiter plates in R-MEM with 10% heat-inactivated calf serum in the presence of peptides. Prior to incubation with the cells, R8-NTA in R-MEM (500 µM, 5 µL), NiCl2 in R-MEM (500 µM, 5 µL), and fresh R-MEM containing 10% heatinactivated calf serum (35 µL) were mixed together and allowed to stand for 15 min. H6-PAD in R-MEM (100 µM, 5 µL) was then added to the above mixture, and allowed to stand for another 1 h. The cells were incubated with the above mixture (total volume, 50 µL) at 37 °C under 5% CO2 for 24 h. MTT (Sigma) in PBS (0.05 mg/10 µL) was added to the above medium, and the cells were further incubated for 4 h. The precipitated MTT formazan was dissolved overnight in 0.04 N HCl in 2-propanol (100 µL). The absorbance at 570 nm was then measured. Cell viability was expressed as the ratio of the A570 of cells treated with peptide to the control samples. RESULTS
Design and Synthesis of Octaarginine Bearing Nitrilotriacetic Acid (NTA). Intracellular protein delivery using the Tat and other arginine-rich membranepermeable peptides has become recognized as a novel and promising technology for the modulation of cell functions (1-3). However, it has been regarded that covalent crosslinking between the cargo proteins and the carrier peptides is necessary for the cellular introduction. Because various histidine-tagged proteins have been prepared up to now, it would be beneficial if we could develop a means of bringing these proteins into cells for the study and modulation of cellular functions without reconstruction of plasmids or systems for expression and purification. Protein purification using the histidine-tags employs immobilized Ni(II) on a chelating adsorbent bearing a nitrilotriacetic acid (NTA) derivative, which occupies four positions in the metal coordination sphere of Ni(II). The remaining two ligand positions in the octahedral coordination sphere are available for selective protein interactions. We developed a novel carrier peptide bearing NTA (Figure 1). As a membrane-permeable carrier peptide, octaarginine was selected as a representative, which was connected to the N-(5-amino-1-carboxypentyl)iminodiacetic acid via a glycine residue as a spacer. In the presence of Ni(II), the carrier peptide R8-NTA, and the histidine-tagged protein should form a noncovalent complex which is expected to penetrate into cells. The synthetic scheme of the carrier peptide is shown in Figure 1, which is based on the condensation of the carrier peptide segment Ac-[Arg(Pbf)]8-Gly-OH with N-(5amino-1-carboxypentyl)iminodiacetic acid. Ac-[Arg(Pbf)]8Gly-OH was prepared by Fmoc-solid-phase peptide synthesis on a 2-chlorotrityl chloride resin followed by N-terminal acetylation using acetic anhydride and detachment from the resin by a treatment with AcOH-TFEDCM (1:1:8). After the C-terminus of Ac-[Arg(Pbf)]8-GlyOH was activated as a succinimidyl ester, excess N-(5-
478 Bioconjugate Chem., Vol. 15, No. 3, 2004
amino-1-carboxypentyl)iminodiacetic acid was added. Treatment of the sample with TFA-EDT (95:5) followed by HPLC purification yielded R8-NTA. The structure of the compound was ascertained by MALDI-TOFMS. Intracellular Delivery of Histidine-Tagged EGFP (H6-EGFP) by R8-NTA-Ni(II). To assess the ability of the Ni(II) chelate of R8-NTA [R8-NTA-Ni(II)] for intracellular delivery of histidine-tagged proteins, we prepared an N-terminus histidine-tagged enhanced green fluorescent protein (H6-EGFP), which was expressed in E. coli and purified using a Ni(II) chelate affinity column to be sufficiently pure. The apparent dissociation constant (KD) between NTA and H6-EGFP was reported to be 4.2 × 10-7 M (19). Internalization of the histidine-tagged EGFP was monitored by laser confocal microscopic observation after a 3-h incubation of the protein with HeLa cells at 37 °C. Prior to incubation with the cells, R8-NTA was mixed with NiCl2 and allowed to stand for 15 min, and H6EGFP was added to the above solution and allowed to stand for another 1 h (Figure 2A). The mixture was then added to the cells and incubated for 3 h. Significant internalization of H6-EGFP was observed when the cells were treated with H6-EGFP (10 µM) in the presence of R8-NTA-Ni(II) (50 µM), showing a very similar manner of cellular localization with that of R8-EGFP (a fusion protein of an octaarginine segment and EGFP) (Figure 2B, (i) and (ii)) Particulate signals from EGFP were observed which accumulated around the perinuclear area, suggesting that significant parts of these proteins were trapped in the cellular vesicles or formed aggregates in the cell. When the cells were treated with R8-NTANi(II)/H6-EGFP at 4 °C, no significant internalization of H6-EGFP was observed (data not shown). This fact suggested that an endocytic pathway predominantly contributed to the cellular uptake of H6-EGFP, as reported for the HIV-1 Tat peptide (20). Equimolar amounts of H6-EGFP and R8-NTA-Ni(II) (10 µM each) gave somewhat less efficient internalization of the protein (data not shown). On the other hand, little internalization of H6-EGFP was observed in the absence of R8-NTA-Ni(II) (Figure 2B, (iii)). No significant improvement in the cellular uptake of H6-EGFP was observed with the addition of R8 peptide [(Arg)8-amide] (50 µM) that lacked the NTA moiety (data not shown). FACS analysis of the cells treated with H6-EGFP (10 µM) in the presence of R8-NTA-Ni(II) (50 µM) for 3 h showed that the amount of the internalized protein was almost half of that of the cells treated with R8-EGFP (10 µM) (Figure 2C). When the cells were treated with the double amount of H6EGFP (20 µM) in the presence of R8-NTA-Ni(II) (100 µM), the amount of the internalized protein became almost comparable to that for R8-EGFP (10 µM). The above studies indicated that histidine-tagged proteins were successfully brought into the cells with the aid of R8NTA-Ni(II), although the efficiency in the delivery of cargo proteins was not comparable to that using a covalently attached arginine peptide segment. Interestingly, R8-NTA retained carrier ability even when R8-NTA was not treated with NiCl2 (Figure 2B, (iv)); though considerably less compared with the case when R8-NTA was complexed with NiCl2, some amount of H6-EGFP was still brought into the cells. Presumably, some divalent cations present in the cells or culture medium formed a complex with R8-NTA to help the internalization of H6-EGFP. Less efficiency in the delivery using the R8-NTA-Ni(II)/H6-EGFP system in comparison with that using R8EGFP suggested that H6-EGFP partially dissociated
Futaki et al.
from R8-NTA-Ni(II) in the cells, presumably due to the presence of other cations and nucleophiles in the cells and the culture medium to compete for the coordination site of Ni(II). Acidification of the endosomic particles may also reduce the binding of H6-EGFP to R8-NTA-Ni(II). To obtain information on this matter, we prepared rhodamine-labeled R8-NTA (Rho-R8-NTA) and examined the cellular localization in comparison with that of H6EGFP. Figure 2D shows a picture of the cells when treated with a mixture of Rho-R8-NTA-Ni(II) and H6EGFP. The fluorescent signals from these molecules showed particulate distribution in the cells. Superimposition of these pictures yielded yellow signals. This fact suggested that a considerable part of H6-EGFP retained the complex structure with Rho-R8-NTA-Ni(II). On the other hand, signals of the respective colors were simultaneously observed, suggesting that some part of these components dissociated from each other and formed vesicles or aggregates abundant with one of these components. To confirm that proteins delivered into the cells can actually modulate biological functions and to estimate the efficiency of the delivery system of histidinetagged proteins using R8-NTA-Ni(II), we conducted other experiments to introduce an apoptosis-inducing peptide into the cells. Delivery of Apoptosis-Inducing Peptide. Ellerby et al. reported that an amphiphilic basic antibacterial peptide showed apoptosis-inducing activity when delivered in mammalian cells [pro-apoptotic domain peptide (PAD): D(KLAKLAK)2-amide] (21). The mechanism of apoptosis caused by this peptide is attributed to interaction with the mitochondrial membranes to disturb the mitochondrial structure. They selectively targeted this peptide to cancer cells by attaching a cyclic RGD segment to it. We have prepared a PAD peptide bearing a H6 segment (H6-PAD, Figure 3A) and assessed the internalization efficiency through its apoptosis-inducing activity in comparison with PAD covalently linked with membrane-permeable octaarginine (R8-PAD). Condensation of the nucleus is one of the typical phenomena of apoptotic cells. We first confirmed that the intracellular delivery of the PAD peptide caused apoptosis. When the cells were treated with R8-NTA-Ni(II)/ H6-PAD for 24 h, significant condensation of the nucleus was observed (Figure 3B). Similarly, condensation of the nucleus was induced by R8-PAD. On the other hand, little change was observed in the size of the nucleus after treatment of the cells with the PAD peptide (10 µM) (data not shown). Even when the peptide concentration was raised to 50 µM, only a slight extent of apoptosis was observed. The above result indicated that intracellular delivery of the PAD peptide using arginine peptides effectively promoted the induction of apoptosis for the target cells. Induction of apoptosis by the PAD peptide was further confirmed by DNA fragmentation assay (Figure 3C). Typical ladder formation of nuclear DNA isolated from the cells treated with H6-PAD (20 µM) in the presence of 5 equiv of R8-NTA-Ni(II) was observed in the gelelectrophoresis (Figure 3C, lane 4). Apoptosis inducing activity of the PAD peptides was assessed by a FACS analysis using FITC-conjugated Annexin-V (FITC-Annexin-V) (22). Annexin-V is a human anticoagulant that has high affinity for binding to phosphatidylserine (PS). In normal healthy cells, PS in the plasma membrane is oriented toward the cytoplasm. As an early event of apoptosis, the translocation of PS to the extracellular environment is observed, which can be evaluated with Annexin-V staining followed by FACS
Arginine Carrier Peptide Bearing Ni(II) Chelator
Bioconjugate Chem., Vol. 15, No. 3, 2004 479
Figure 2. (A) Complex formation of H6-EGFP with the Ni(II) chelate of R8-NTA [R8-NTA-Ni(II)] and (B) confocal microscopic observation of the HeLa cells treated with R8-NTA-Ni(II)/H6-EGFP for 3 h. The R8-NTA-Ni(II)/H6-EGFP complex was formed as described in the Experimental Section. (C) FACS analysis of the internalized EGFP proteins. HeLa cells were treated with H6EGFP (10 µM (blue) or 20 µM (red)) in the presence of 5 equiv of R8-NTA-Ni(II). Green line represents the cells treated with R8EGFP (10 µM) and black line control cells (untreated). (D) Confocal microscopic observation of the cells treated with rhodaminelabeled R8-NTA-Ni(II) [Rho-R8-NTA-Ni(II)] and H6-EGFP. Rho-R8-NTA-Ni(II) and H6-EGFP were mixed together at a molar ratio of 5:1 to facilitate the complex formation and then applied to a Microcon YM-10 centrifugal filter (normal molecular weight limit 10 000) (Millipore) to remove excess Rho-R8-NTA-Ni(II) before addition to the cells (final concentration of H6-EGFP, 10 µM).
analysis. In combination with the cell impermeable nucleic acid stain propidium iodide (PI), apoptotic cells (Annexin-V positive/PI negative) can be differentiated from necrotic (Annexin-V positive/PI positive) and normal (Annexin-V negative/PI negative) cells. HeLa cells were treated with R8-NTA-Ni(II)/H6-PAD or R8-PAD for 2 h. FACS analysis showed that the treatment with H6-PAD (10 µM) in the presence of 5 equiv of R8-NTA-Ni(II)
induced apoptosis in cells to the same extent to those treated with a typical apoptosis-inducing agent, staurosporine (400 nM). H6-PAD (20 µM) in the presence of R8NTA-Ni(II) (100 µM) gave comparable effect with R8-PAD (10 µM) to induce apoptosis in >40% of cells, four times higher than those treated with staurosporine (400 nM). The apoptosis-inducing activity of the peptides was also assessed in terms of the cell viability determined by the
480 Bioconjugate Chem., Vol. 15, No. 3, 2004
Futaki et al.
Figure 3. (A) Primary structure of histidine-tagged pro-apoptotic domain peptide (H6-PAD) and its related peptides. Amino acids are expressed by a one-letter abbreviations (H, histidine; G, glycine; R, arginine; K, lysine; L, leucine; A, alanine). (B) Nuclear condensation in apoptotic cells treated with PAD peptides for 24 h. (C) Fragmentation of the chromosomal DNA by the PAD peptides. Lane 1 contains a size marker, lane 2 contains DNA isolated from control cells (untreated), and lane 3-5 contain DNA from cells treated with PAD peptide (10 µM), H6-PAD (20 µM) in the presence of R8-NTA-Ni(II) (100 µM), and staurosporine (400 nM) for 24 h, respectively. (D) Apoptosis analysis by Annexin-V staining. Cells untreated and treated with H6-PAD (10 µM (1) and 20 µM (2)) in the presence of 5 equiv of R8-NTA-Ni(II), R8-PAD (10 µM) (3), and staurosporine (400 nM) (4) were stained with FITC-Annexin-V in the presence of PI and analyzed by flow cytometry. The percentages of cells positive for Annexin-V and negative for PI were shown. (E) Apoptotic cell death induced by the PAD peptides. The viability of the cells was determined by the MTT [3-(4,5dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] assay. HeLa cells were incubated in the medium containing a peptide at the indicated concentration for 24 h. Treatment of the cells with the PAD peptide at the concentration of 10 µM did not cause significant cytotoxicity to the cells. H6-PAD 10 µM (5) and H6-PAD 20 µM (5) represent the results of the cells treated with H6-PAD (10 and 20 µM) in the presence of 5 equiv of R8-NTA-Ni(II), respectively. Error bars represent the mean ( standard deviation of three to four samples.
MTT assay (Figure 3E). The peptide-induced apoptotic cell death caused a decrease in the population of live cells among the total cells. The HeLa cells were treated with H6-PAD (10 and 20 µM) in the presence of 5 equiv of R8NTA-Ni(II) for 24 h. In such cases, viability of the cells ranged around 65% and 17%, respectively, compared with untreated cells. These values were comparable to those for R8-PAD (5 µM) (62%) and R8-PAD (10 µM) (15%). On the other hand, PAD (10 µM) showed no significant cytotoxicity under the given condition. Viability of the R8 peptide (octaarginine amide) (10 µM) was judged to be 88% (data not shown). Thus, although the internalization efficiency of the PAD peptide was a slightly less effective in the case of using the R8-NTA-Ni(II)/H6-PAD system as compared with R8-PAD, it was judged that the H6-PAD peptide was actually delivered into the cytosol to interact with the mitochondrial membranes. These facts were suggestive of the effectiveness of the R8-NTANi(II) system for the intracellular delivery of histidinetagged proteins. DISCUSSION
In this paper, we have shown that histidine-tagged peptides and proteins were efficiently brought into the
cells using a membrane-permeable octaarginine peptide bearing nitrilotriacetic acid (R8-NTA). The histidinetagged method is one of the most popular methods for the efficient purification of proteins, and numerous proteins bearing the tag have been prepared. Using this approach, these histidine-tagged proteins are expected to be delivered into cells without modifying the plasmids and expression systems. The results obtained here also raised the possibility of bringing other chelating molecules and noncovalently assembled supramolecular complexes into cells. Other tagged systems can be applicable to the arginine-peptide mediated protein delivery into cells. Although application of the Tat peptide chelated with 99mTc was reported for in vivo imaging and radiotherapy (23), this is, to the best of our knowledge, the first report on the delivery achieved while retaining a chelating complex structure as a ‘coupler’ between carrier peptides and cargo proteins. As shown in Figure 2D, a certain population of the carrier peptide (R8-NTA) and cargo H6-EGFP were observed without significant colocalization. This is presumably because of the nonspecific competition of intracellular nucleophilic compounds with the coordination center of Ni(II) to H6-EGFP. Protonation of histidine in
Arginine Carrier Peptide Bearing Ni(II) Chelator
H6-EGFP by acidification of endocytic particles may also explain this fact. Although the apparent internalization efficiency of the proteins is hampered by partial dissociation of the complexes, the cargo molecules can be unequivocally delivered into cytosol to control cellular function as shown in the delivery of the PAD peptide. This characteristic may be useful when the covalent attachment of carrier peptides is undesirable for the exertion of their function or cellular localization of the cargo proteins. Particulate cellular distribution of the cargo EGFP suggested the involvement of endocytosis in the internalization. It is also suggested that the cargo proteins diffuse into cytosol to give considerably less intense signals than when retained in endosomes. Thinking about this methodology having been successfully applied in modulating various cellular functions, the modulation may be achieved even when only a small portion of the cargo proteins trapped in endosome are released into cytosol. Thus, assessment of the internalization efficiency of proteins not only by microscopic observation or FACS analysis (corresponding to the total cellular uptake of proteins) but also by activities induced by the cargo proteins (reflecting the amount of the cytosol-released proteins) becomes crucial. It would be implausible that the chelating molecules with dissociation constants of micromolar range can internalize into cells bearing their complex structure by direct penetration through the membranes with an unfolding process. This fact will be a criterion for discussing the mechanism of membrane permeation of the arginine-rich carrier systems. ACKNOWLEDGMENT
This work was partly supported in part by Grants-inAid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. LITERATURE CITED (1) Futaki, S., Goto, S., Suzuki, T., Nakase, I., and Sugiura, Y. (2003) Structural variety of membrane permeable peptides. Curr. Protein Pept. Sci. 4, 87-96. (2) Wadia, J. S., and Dowdy S. F. (2003) Modulation of cellular function by TAT mediated transduction of full length proteins. Curr. Protein Pept. Sci. 4, 97-104. (3) Wright, L. R., Rothbard, J. B., and Wender, P. A. (2003) Guanidinium rich peptide transporters and drug delivery. Curr. Protein Pept. Sci. 4, 105-124. (4) Schwarze, S. R., Ho, A., Vocero-Akbani, A., and Dowdy, S. F. (1999) In vivo protein transduction: delivery of a biologically active protein into the mouse. Science 285, 1569-1572. (5) Vive`s, E., Brodin, P., and Lebleu, B. (1997) A truncated HIV-1 Tat protein basic domain rapidly translocates through the plasma membrane and accumulates in the cell nucleus. J. Biol. Chem. 272, 16010-16017. (6) Futaki, S., Suzuki, T., Ohashi, W., Yagami, T., Tanaka, S., Ueda, K., and Sugiura, Y. (2001) Arginine-rich peptides: An abundant source of membrane-permeable peptides having potential as carriers for intracellular protein delivery. J. Biol. Chem. 276, 5836-5840. (7) Suzuki, T., Futaki, S., Niwa, M., Tanaka, S., Ueda, K., and Sugiura, Y. (2002) Possible existence of common internaliza-
Bioconjugate Chem., Vol. 15, No. 3, 2004 481 tion mechanisms among arginine-rich peptides. J. Biol. Chem. 277, 2437-2443. (8) Futaki, S., Nakase, I., Suzuki, T., Zhang, Y., and Sugiura, Y. (2002) Translocation of branched-chain arginine peptides through cell membranes: flexibility in the spatial disposition of positive charges in membrane-permeable peptides. Biochemistry 41, 7925-7930. (9) Wender, P. A., Mitchell, D. J., Pattabiraman, K., Pelkey, E. T., Steinman, L., and Rothbard, J. B. (2000) The design, synthesis, and evaluation of molecules that enable or enhance cellular uptake: peptoid molecular transporters. Proc. Natl. Acad. Sci. U.S.A. 97, 13003-13008. (10) Umezawa, N., Gelman, M. A., Haigis, M. C., Raines, R. T., and Gellman, S. H. (2002) Translocation of a β-peptide across cell membranes. J. Am. Chem. Soc. 124, 368-369. (11) Hochuli, E., Dobeli, H., and Schacher, A. (1987) New metal chelate adsorbent selective for proteins and peptides containing neighbouring histidine residues. J. Chromatogr. 411, 177-184. (12) Terpe, K. (2003) Overview of tag protein fusions: from molecular and biochemical fundamentals to commercial systems. Appl. Microbiol. Biotechnol. 60, 523-533. (13) Atherton, E., and Sheppard, R. C. (1989) Solid-Phase Peptide Synthesis, A Practical Approach, IRL Press, Oxford. (14) Barlos, K., Chatzi, O., Gatos, D., and Stavropoulos, G. (1991) 2-Chlorotrityl chloride resin. Studies on anchoring of Fmoc-amino acids and peptide cleavage. Int. J. Pept. Protein Res. 37, 513-520. (15) Coste, J., Le-Nguyen, D., and Castro, B. (1990) PyBOP: A new peptide coupling reagent devoid of toxic byproduct. Tetrahedron Lett. 31, 205-208. (16) Carpino, L. A., Shroff, H., Triolo, S. A., Mansour, E. M. E., Wenschuh H., and Albericio F. (1993) The 2,2,4,6,7pentamethyldihydrobenzofuran-5-sulfonyl group (Pbf) as arginine side chain protectant. Tetrahedron Lett. 34, 7829-7832. (17) Nagaoka, M., and Sugiura, Y. (1996) Distinct phosphate backbone contacts revealed by some mutant peptides of zinc finger protein Spl: effect of protein-induced bending on DNA recognition. Biochemistry 35, 8761-8768. (18) Rink, H. (1987) Solid-phase synthesis of protected peptide fragments using a trialkoxy-diphenyl-methylester resin. Tetrahedron Lett. 28, 3787-3790. (19) Schmid, E. L., Keller, T. A., Dienes, Z., and Vogel, H. (1997) Reversible oriented surface immobilization of functional proteins on oxide surfaces. Anal. Chem. 69, 1979-1985. (20) Richard, J. P., Melikov, K., Vive`s, E., Ramos, C., Verbeure, B, Gait, M. J., Chernomordik, L. V., and Lebleu, B. (2003) Cell-penetrating peptides. A reevaluation of the mechanism of cellular uptake. J. Biol. Chem. 278, 585-590. (21) Ellerby, H. M., Arap, W., Ellerby, L. M., Kain, R., Andrusiak, R., Rio, G. D., Krajewski, S., Lombardo, C. R., Rao, R., Ruoslahti, E., Bredesen, D. E., and Pasqualini R. (1999) Anticancer activity of targeted pro-apoptotic peptides. Nat. Med. 5, 1032-1038. (22) Danielson, S. R., Wong, A., Carelli, V., Martinuzzi, A., Schapira, A. H., and Cortopassi, G. A. (2002) Cells bearing mutations causing Leber’s hereditary optic neuropathy are sensitized to Fas-induced apoptosis. J. Biol. Chem. 277, 5810-5815. (23) Polyakov, V., Sharma, V., Dahlheimer, J. L., Pica, C. M., Luker, G. D., and Piwnica-Worms, D. (2000) Novel Tatpeptide chelates for direct transduction of technetium-99m and rhenium into human cells for imaging and radiotherapy. Bioconjugate Chem. 11, 762-771.
BC034181G