Letter pubs.acs.org/ac
γ‑PARCEL: Control of Molecular Release Using γ‑Rays Shuhei Murayama,† Jun-ichiro Jo,† Kazutaka Arai,† Fumihiko Nishikido,† Rumiana Bakalova,† Taiga Yamaya,† Tsuneo Saga,† Masaru Kato,‡ and Ichio Aoki*,† †
Molecular Imaging Center, National Institute of Radiological Sciences, 4-9-1 Anagawa, Inage-ku, Chiba 263-8555, Japan Graduate School of Pharmaceutical Sciences and GPLLI Program, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
‡
S Supporting Information *
ABSTRACT: We previously have developed the photoresponsive tetra-gel and nanoparticles for controlling the function of the encapsulated substance by UV irradiation. However, the penetration ability of the UV is not high enough. Here, we developed a radiation-responsive tetra-gel and nanoparticle based on γ-ray-responsive X-shaped polyethylene glycol (PEG) linker with a disulfide bond. The nanoparticle could retain small molecules and biomacromolecules. γ-Rays were used as a trigger signal because of their higher penetrating ability. This allowed a spatiotemporal release and control of the encapsulated substances from the nanoparticle in the deeper region, which is impossible by using light exposure (ultraviolet, visible, and near-infrared).
L
analysis of a living system and application in the prevention and treatment of diseases. The characteristic of the PARCEL mesh structure was derived from the X-shaped polyethylene glycol (PEG) linker, which is consisted of three functional units: (1) a mesh unit, (2) a connecting unit, and (3) a cleavable unit. If the cleavable unit of the linker is changed, the external signal that induces the cleavage of the mesh also changes, which can lead to various applications of the PARCEL method. UV light, which was used as a controlling signal in the PARCEL method, had a serious limitation. The penetration ability of the light stimuli is not high enough through the skin or tissue. γ-Rays were more appropriate as an external signal because of their deep-tissue penetration, allowing transmission of the signal deep into the body. Disulfide (−S−S−) and diselenide (−Se−Se−) bonds are known as radiation-sensitive units.15,16 The disulfide bond in proteins is stable in vivo.17 Replacement of the photocleavable unit to a γ-ray-cleavable unit such as a disulfide bond will allow the development of γray-responsive nanoparticles with preservation of their other properties. Cleavage of the S−S bonds in the gel will cause a chain reaction that releases the encapsulated compound.
iving systems are controlled by an accurate regulation of the time and place of molecular activation (e.g., activation of enzymes, gene expression), which causes signal transduction and specific cell signaling. To analyze the living system, controlling the method of the molecular activation is required. This is a reason for developing many kinds of molecular carriers.1−8 We have developed a method for molecular (functional) activation using nanoparticles prepared from a photocleavable X-shaped molecule.9−11 The molecules in the nanoparticles are (functionally) inactive because the shell of the nanoparticle restricts their interaction with other molecules in the environment. However, when the encapsulated molecules are released using external light, their activity is restored. The method can spatiotemporally control the function of various compounds. We named it “protein activation and release from cage by external light” (PARCEL).9−11 The nanoparticles have many advantages for functional control of the molecule: (1) the encapsulated molecule is physically trapped in the tetra armed gel mesh network; this control technique is applicable to various molecules,11 (2) the encapsulated molecule is stable in the mesh because the mesh structure protects it from the external environment or neighboring encapsulated molecules, (3) because the mesh size can be tuned to the encapsulated molecule, there is no limitation to the size of the encapsulated molecule, (4) because the nanoparticles are rapidly excreted by the urine and no accumulation is observed in other organs (demonstrated in mice), they may be considered safe for the organism,12 and (5) because the size and surface properties of the nanoparticles were easily controlled using synthetic conditions, the nanoparticles having different properties were easily prepared.13,14 Therefore, the PARCEL method has considerable potential for © XXXX American Chemical Society
■
EXPEIMENTAL SECTION Materials. Tetra-poly(ethyl glycol)-amine (SUNBRIGHT PTE-050PA; Mn, 5328 g/mol) was purchased from the NOF Corporation (Tokyo, Japan). Hydrogen peroxide, N,N,N′,N′tetramethylethylenediamine (TEMED), triethylamine (TEA), Received: August 7, 2015 Accepted: November 3, 2015
A
DOI: 10.1021/acs.analchem.5b03030 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
Absorbance Measurement of the Released Nile Blue Chloride. After γ-ray irradiation (total dose: 10 Gy, 8 Gy/ min), 1050 μL of water was added to the gel and shaken for 15 min. A supernatant (200 μL each) that contained the released Nile blue chloride was poured into a 96-well plate. Then, the absorbance signal (638 nm) was measured with a multiplate reader (Infinite M200PRO, TECAN, Männedorf, Switzerland). Encapsulation of Proteins in γ-Ray-Responsive Nanoparticles. Proteins were encapsulated in γ-ray-responsive nanoparticles by combining 16 mM PEG-SS-Ac (50 μL), PEG-Ac13 (150 μL), 14 mg/mL trypsin solution (50 μL), 0.1 M APS (50 μL), and 0.1 M TEMED (50 μL) in that order at room temperature and subsequently vortexing the mixture for 20 min to obtain a dispersion of nanoparticles with diameters of 180 nm. The nanoparticle solution was filtrated using Vivaspin 6-30 K. DLS Measurement of the Nanoparticle Size Distribution. After filtration, the nanoparticles were analyzed using a dynamic light scattering (DLS) machine (Zetasizer Nano ZS, Malvern, UK). Binding Ratio Calculated from the NMR Spectrum. The binding ratio can be calculated from the number of remaining double bonds, as previously described.18 The 1H NMR spectrum of the gel shows a prominent signal at 6.1−6.2 ppm, which was assigned to the 8 protons of the terminal methylene group of the acrylic group. The proton signal at 2.6−2.9 ppm was used as an internal standard because these 6 protons do not participate in the reaction, and the integrated value of this peak was not changed by the reaction (Figure S1). The binding ratio was calculated from the following equation: Binding ratio (%) = {1 − (integrated value of proton signal at 6.1−6.2 ppm/8)/(integrated value of the proton signal at 2.6− 2.9 ppm/6)} × 100. In Vivo Model Using a Swine Tissue. To evaluate the release efficiency of the nanoparticles by the penetrated γ-ray through the body, we made an ex vivo sheltered model using a swine tissue (edible pole lump). The polypropylene conical tube was covered using the swine tissue (2 cm thickness). Then, the centrifuge tube with the nanoparticle solution was inserted into the conical tube, and the γ-ray (10 Gy) was irradiated through the tissue layer. γ-Ray Attenuation Rate with Swine Tissue. A cesium (Cs) point-like source (3.7MBq) was seated above the high sensitivity radiation detector module (C12137, Hamamatsu photonics K. K.) to estimate the abruption rate in the swine tissue. The tissue was put between the Cs source and the γ-ray detector (Figure S2). The gamma-ray detector outputs the dose equivalent value which is proportional to the physical dose. The attenuation ratio of tissue was calculated from the following equation: Shielded ratio (%) = {1 − (average radiation intensity with tissue − background value)/(average radiation intensity without tissue − background value) × 100}. Trypsin Activity Measurements Using Fluorescence. The released trypsin activity was measured using BODIPY FL casein, which is a substrate for fluorescence assays. After the γray irradiation, 100 μL of gel solution and 100 μL of 10 μg/mL BODIPY FL casein were added to each well. Then, the plate was incubated at 37 °C for 1.5 h, and the fluorescence signal (Ex 485 nm, Em 535 nm) was measured with a multiplate reader.
acryloyl chloride (AC), acrylamide (AAm) ammonium persulfate (APS), methanol, diethyl ether, and 4-(4,6dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride n-hydrate (DMT-MM) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Nile blue chloride and trypsin from porcine were purchased from Sigma-Aldrich (St. Louis, MO). BODIPY casein was contained as a substrate in EnzChek Protease Assay Kit green fluorescence, which was purchased from Invitrogen Corporation (Carlsbad, CA). Melcaptopropionic acid and 2-aminoethanethiol hydrochloride were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Water was purified with a Milli-Q apparatus (Millipore, Bedford, MA). Synthesis of the γ-Ray-Responsive Linker (Compound 1). Melcaptopropionic acid and 2-aminoethanethiol hydrochloride were dissolved in water and stirred in a lightproof vial; then, hydrogen peroxide was added. The reaction was stirred at room temperature to obtain a disulfide mixture (3-((2aminoethyl) disulfanyl) propanoic acid and diacid and diamine). The diacid was separated on the basis of the pH values. The disulfide mixture was acrylated using acrylyl chloride with excess triethylamine at 0 °C. Diacrylamine, which was made from diamine, was separated on the basis of the pH values. The γ-ray-responsive linker (compound 1) as a white crystal was purified by recrystallization of the product. The structure of compound 1 was confirmed using 1H NMR. 1 H NMR (CDCl3): δ = 6.28 (dd, C(O)CHCH2), δ = 6.10, 5.70 (d, d, C(O)CHCH 2 ), δ= 3.69 (t, NHCH2CH2SS), δ = 3.01 (t SSCH2CH2COOH), δ = 2.96 (t, NHCH2CH2SS), δ = 2.77 (t, SSCH2CH2COOH). Synthesis of the γ-Ray-Responsive PEG Linker (PEGSS-Ac). Compound 1 (563 μmol) was dissolved in methanol and stirred. Then, tetra poly(ethylene glycol)-amine (tetraPEG-amine; 94 μmol) was added and stirred until all reactants were dissolved. Afterward, DMT-MM was added to start the synthesis without stirring in a lightproof vial. The reaction was conducted overnight at room temperature. The product was precipitated in diethyl ether on ice and filtered. The collected substance was washed with diethyl ether and dissolved in water. The aqueous solution was dialyzed (SpectraPor6, CO 1000 g mol−1) and freeze-dried to yield the tetra-acrylated PEG, which was referred to as PEG-SS-Ac. The structure of the linker was confirmed using 1H NMR. 1 H NMR (CD3OD): δ = 6.25 (dd, C(O)CHCH2), δ = 6.20, 5.62 (d, d, C(O)CHCH 2 ), δ = 3.5(br, NHCH2CH2SS), δ = 3.5 (br, NHCH2CH2CH2O), δ = 3.5 (br, [CH2CH2O]n, n ≈ 28), δ = 3.3 (s, OCH2C), δ = 3.2 (m, NHCH2CH2CH2O), δ = 2.96 (t, SSCH2CH2COO), δ = 2.85 (m, NHCH2CH2SS), δ = 2.60 (t, SSCH2CH2COO), δ = 1.7 (m, NHCH2CH2CH2O). Encapsulation of Nile Blue in the γ-Ray-Responsive Gel. Centimeter-sized hydrogel-encapsulated small molecules were prepared in a lightproof microtube. First, 200 μL of 3 M AAm, 50 μL of 16 mM PEG-SS-Ac, and 50 μL of 2 mg/mL Nile blue chloride solution were dispensed into the tube and shaken for 5 s. Then, 25 μL of 0.1 M APS and 25 μL of 0.1 M TEMED were added to the mixture to induce gelation at room temperature without stirring. After gelation, the gel was washed three times with water. γ-Ray Stimulation to Release the Encapsulated Compounds. The Model Gammator M (Radiation Machinery Corporation, New Jersey, USA) was used as a γ-ray irradiation source (Cs). Dose rate of γ-ray was 8 Gy/min. B
DOI: 10.1021/acs.analchem.5b03030 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry Scheme 1. Synthesis of a γ-Ray-Responsive Linker (PEG-SS-Ac)
■
RESULTS AND DISCUSSION In this study, we designed a new PEG linker, which has an −S− S− group as a cleavable unit for a monomer of the tetra-gel. This linker was cleaved by γ-ray irradiation, and it was synthesized from γ-ray-responsive compound19 and amine derivatized X-shaped PEG (28 repeated PEG units in each hand) using esterification20 (Scheme 1). We expect that the new γ-ray-sensitive structure will allow a controlled release of molecules in the deeper region, which is impossible using a light-sensitive structure that is prepared using the conventional PARCEL method previously published in our paper.9,10 The first step of the study was to confirm that the γ-rayresponsive gel, which was made from γ-ray-responsive PEG-SSAc and AAm, could retain molecules in the mesh structure and release these molecules in a controlled manner using external radiation. γ-Ray-unresponsive gel that was made from PEG-Ac and AAm was prepared for comparison. Nile blue was chosen as an encapsulated molecule, and it was added to the preparation solution of each gel. The released amounts of Nile blue before and after γ-irradiation were measured on the basis of its absorbance at 638 nm and calculated using a calibration curve (Figures 1 and S3). There was a negligible leakage of Nile blue from the γ-ray-responsive gel in the absence of γ-irradiation and from the γ-ray-unresponsive gel after γ-irradiation. After the γ-irradiation, a large amount of Nile blue molecules was released from the γ-ray-responsive gel that was thought to derive from cleavage of disulfide bonds and subsequent mesh degradation. The gel mesh should be small enough to retain small molecules, such as Nile blue, and to control their release using external γ-irradiation on the principle of the PARCEL method. The second step of the study was to use the gel to control the molecular activity within the microspace. For this purpose, it was necessary to prepare γ-ray-responsive nanoparticles because the γ-ray-responsive gels are not sufficiently small to introduce into the microspace. The previously used PEG linkers and PEG-SS-Ac have similar structures, and we anticipated that the nanoparticle could be generated by the same preparation method.10 Using a controlled radical polymerization with PEG-SS-Ac and PEG-Ac (noncleavable linker) (mixture ratio of 1:3) by vortexing for 20 min, we obtained nanoparticles with an average diameter of ∼180 nm. The binding ratio of the acryloyl group of the linker in the nanoparticle solution was evaluated using the NMR spectrum, and the value was 56% (Figure S1) which was higher than that
Figure 1. Regulation of the release of small molecules (Nile blue) from γ-ray-responsive terta-gel using γ-irradiation (10 Gy). SS gel: γray-responsive; PEG gel: gel without γ-ray-responsive moieties (for comparison). The data are means ± SD of three independent experiments.
of photocleavable linker (PEG-photo-Ac) (48%).19 That indicates the nanoparticles made from PEG-SS-Ac are more complex than the nanoparticles made from PEG-photo-Ac. This high-density mesh structure is suitable for retention of smaller compounds than the previous particle.10 The diameter of the nanoparticle decreased after γ-irradiation, and their average size became ∼120 nm (Figure 2). The diameter was larger than that of the photoresponsive nanoparticles after irradiation (∼60 nm). It is thought that the large nanoparticles remain after irradiation because the recombination reaction of the disulfide bonds of the PEG-SS-Ac occurred within the nanoparticle. The recombination does not occur in photoresponsive units. After the cleavage of disulfide bonds using γirradiation, the free thiyl radicals that are generated can attack other bonds and make new bonds. The recombination makes new, larger meshes that provide new routes for release of the encapsulated compound. To evaluate the release of the encapsulated compounds, which was triggered by γ-irradiation, we prepared two types of nanoparticles from γ-ray-cleavable linker and γ-ray-noncleavable C
DOI: 10.1021/acs.analchem.5b03030 Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Letter
■
SUMMARY AND CONCLUSIONS We designed and synthesized the γ-ray-responsive PEG linker with a disulfide bond. Polymerization of the linker with AAm allows development of a γ-ray-responsive tetra-gel that can encapsulate small molecules and release them using γirradiation. Without AAm, polymerization of the disulfide linker allows development of γ-ray-responsive nanoparticles that can spatiotemporally control protein activity with γ-ray radiation as a trigger stimulus. Because of the penetrability of γrays, the presence of swine tissue does not affect the trigger signal to the nanoparticles. The γ-ray-responsive nanoparticles are appropriate for intravenous application and could make the response target be located in deep tissue, which is impossible to achieve with the conventional PARCEL method with light. In conclusion, a variety of molecules (from small to large molecules) were effectively released and activated within the light shielding area upon irradiation of γ-ray-responsive tetra-gel or nanoparticles with γ-ray. Because the encapsulated molecules were physically trapped by the mesh structure, γ-PARCEL should be amenable to a variety of proteins. In the future, by improving the stability of the gel, this kind of method should permit the spatiotemporal control of molecular activity in the deep area. It will be a powerful tool for analysis and control of living systems.
Figure 2. DLS measurements of the size of the nanoparticles before (blue) and after (red) γ-irradiation.
linker, respectively. Trypsin was encapsulated within both nanoparticles as a model protein. The trypsin-loaded nanoparticles were subjected to 10 Gy γ-irradiation, which is a smaller irradiation dose for γ-ray responsive carriers than previous reports.16 To demonstrate the controlled release in the deep region, triggered by γ-rays, a muscle tissue from swine (20 mm thick) was placed between the radiation source and the sample. The released trypsin was quantitatively measured via degradation of its substrate BODIPY-casein (Figure 3). In the
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b03030. 1 H NMR spectrum; attenuation rate of γ-ray muscle tissue from swine; calibration curve of the absorbance of Nile blue (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by JSPS KAKENHI Grant Number 26860032, Core-to-Core Program, A. Advanced Research Networks, an internal grant of NIRS, and the Center of Innovation Program (COI stream from JST).
■
Figure 3. Regulation of the enzymatic activity of the γ-ray-responsive nanoparticle using γ-irradiation. The data are means ± SD of three independent experiments.
REFERENCES
(1) Thomas, K. R.; Folger, K. R.; Capecchi, M. R. Cell (Cambridge, MA, U. S.) 1986, 44, 419−428. (2) Smithies, O.; Gregg, R. G.; Boggs, S. S.; Koralewski, M. A.; Kucherlapati, R. S. Nature 1985, 317, 230−234. (3) Evans, M. J.; Kaufman, M. H. Nature 1981, 292, 154−156. (4) Fire, A.; Xu, S.; Montgomery, M. K.; Kostas, S. A.; Driver, S. E.; Mello, C. C. Nature 1998, 391, 806−811. (5) Chang, C.; Fernandez, T.; Panchal, R.; Bayley, H. J. Am. Chem. Soc. 1998, 120, 7661−7662. (6) Hahn, E. M.; Muir, W. T. Angew. Chem., Int. Ed. 2004, 43, 5800− 5803. (7) Endo, M.; Nakayama, K.; Kaida, Y.; Majima, T. Angew. Chem. 2004, 116, 5761−5763. (8) Yazawa, M.; Sadaghiani, M. A.; Hsueh, B.; Dolmetsch, E. R. Nat. Biotechnol. 2009, 27, 941−947. (9) Murayama, S.; Kato, M. Anal. Chem. 2010, 82, 2186−2191.
nanoparticles with γ-ray-cleavable linker (PEG-SS-Ac), a large amount of trypsin was released after 10 Gy γ-ray irradiation. In contrast, the enzymatic activity was not detected in control nanoparticles without PEG-SS-Ac. Trypsin leakage from the nanoparticles before the irradiation was negligible. Although the γ-rays attenuated to
