Biomacromolecules 2008, 9, 2007–2013
2007
Two Independent Ways of Preparing Hypercharged Hydrolyzable Polyaminorotaxane Basile Pe´re`s, Nicolas Richardeau, Nathalie Jarroux, Philippe Gue´gan, and Loı¨c Auvray* Laboratoire Mate´riaux Polyme`res aux Interfaces, UMR CNRS 7182, University of Evry, bld. Mitterrand, 91025 Evry Cedex, France Received March 6, 2008; Revised Manuscript Received April 4, 2008
The aim of this work is to synthesize new PEO-based polyrotaxanes from modified cyclodextrins. Two strategies are discussed and compared. In the first, a pseudopolyrotaxane was formed between R,ω- PEO dimethacrylate and R-cyclodextrin. A coupling reaction between 1-pyrenebutyric acid N-hydroxysuccinimide ester was carried out to block the cyclic molecules onto the PEO. Cyclodextrins of the supramolecular assemblies were then oxidized using sodium periodate and reacted with spermine to form a potentially highly charged polyrotaxane. In the second strategy, cyclodextrins were first modified, and used to form the polyrotaxane through the pseudopolyrotaxane synthesis followed by the blocking reaction. Acidic titration allowed quantifying the number of amine functions borne by the supramolecular assemblies through two variables: the number of rings per polymer chain and the number of spermine groups per cyclic molecule. The supramolecules obtained by both strategies are discussed.
1. Introduction Synthetic polymers have attracted a large amount of attention for gene transfer applications.1 Their low immunogeneicity and easy processability make these materials ideal candidates for use in a large scale therapeutic approach.2 Among the most studied polymers are poly(ethylene imine)3 and polylysine.4 However, these gene carriers have known drawbacks, that is, low transfection efficiency and possible toxicity.5 Recent works proved that the hydrolyzable character of supramolecular assemblies can enhance the nontoxicity of the vector and increase the delivery of DNA.6–8 This is why poly(ethylene oxide)-based polyaminorotaxane and R-cyclodextrins (R-CDs) have been synthesized.9 In that work, the cationic segments are grafted onto the hydroxyl functions of the R-CDs borne by the initially synthesized polyrotaxane. This is a classical strategy used by us (silylation)10 and other groups (reaction with succinic anhydride).11,12 The cationic functions are provided by the statistical modification of the hydroxyl functions of the R-CDs into aminoethylcarbamoyl derivatives6 or by an oligopeptide derivative.13 More recently, Ooya elegantly introduced disulfide functions in the blockers to allow the degradation of the supramolecular structure during delivery.9 Here, we propose two original ways to synthesize hydrolyzable polyaminorotaxanes. The first way consists in the modification of the R-CD borne by a premade polyrotaxane to obtain a new polyrotaxane based on poly(ethylene oxide) and an R-CDderived crown. In this case, the packing of crown ether on the PEO chain is not statistical,14 but controlled by the threading of the cyclic compound.15 The second way consists of the threading of the ammoniummodified R-CD-derived crown onto the poly(ethylene oxide) chain. In this case, crown ethers are synthesized by R-CD modification before being used in the supramolecular structure building. Thus, the number of spermine moieties on the R-CD derivative is controlled, and the control of the respective molarity * To whom correspondence should be addressed. E-mail: nathalie.jarroux@ univ-evry.fr.
of cyclic molecules and ethylene oxide units will be studied. The behavior of these structures in solution is investigated by SANS.
2. Experimental Section Materials. R-CD was kindly supplied by Wacker and dried prior to use under vacuum at 70 °C. R,ω-Dimethacrylate poly(ethylene glycol) j n ) 1100 g · mol-1) was purchased from Polysciences, Inc. The (M average molar mass of polymer samples was determined by size exclusion chromatography by the suppliers and the degree of functionalization jf ) 2 was verified by 1H NMR spectroscopy. Sodium periodate (NaIO4), ammonium persulfate, and spermine were purchased from Acros and used as received. Ultra pure water was prepared by passing through a Quantum Ultrapure Organex Cartridge (QTUM000EX, Millipore) and dimethylsulfoxide (DMSO) and dimethylformamide (DMF) were purchased from Merck and used as received. Measurements. Size Exclusion Chromatography. Size exclusion chromatography (S.E.C.) was performed in DMF with both UV (λ ) 345 nm) and refractometric detectors. The column was a HR 5 E Styragel column (Waters) and the flow rate of DMF, used as eluent, was 0.3 mL/min. Calibration was done with PEO standards. 1 H NMR. 1H NMR spectra were recorded on a Bruker at 300 MHz in dimethylsulfoxide (DMSO-d6) or D2O. Neutron Scattering Experiments. Small-angle neutron scattering (SANS) measurements were performed at the Laboratoire Le´on Brillouin (“ORPHEE” reactor, CEN Saclay) on the “PACE” spectrometer. The scattering vector length q ranges from 10-3 to 10-1 Å-1. Neutron beam wavelength of 6 Å was used with a sample to detector distance of 3.2 m. The samples (2% solution in water) were used without neutralization, meaning that the compound was not very charged, and loaded into Hellma quartz neutron cells with a 2 mm optical path length. The cells were placed in a sample-passing holder, and scattering for each sample was measured for about 4 h at 20 °C. The temperature was controlled to (1 °C. The data have been corrected by background subtraction and water normalization as usual. MALDI-TOF. MALDI-TOF characterizations of polymers by MALDI-TOF were performed using a Perseptive Biosystems Voyager-DE Pro STR time of flight mass spectrometer, equipped with a nitrogen laser (λ ) 337 nm). External calibration using PEO
10.1021/bm800247c CCC: $40.75 2008 American Chemical Society Published on Web 06/03/2008
2008
Biomacromolecules, Vol. 9, No. 7, 2008
Pérès et al.
Scheme 1. Schematic Representation of R-CD Oxidation Reaction
was performed with the same matrix as in the experiments. Data were collected with a laser power set just above the ionization threshold of the matrix to avoid fragmentation and to maximize the resolution. Typically, five acquisitions corresponding to 50 shots were realized for each sample to ensure representative mass spectra. All investigated matrices and analytes were dissolved in THF. The concentration of polymer solutions was 10-3 M (analyte concentration). Solutions of matrices had a concentration comprising between 5 × 10-3 M to 5 × 10-1 M, to take into account the matrix solubility. For all analyses, typically 20 µL of the analyte solution were added to a 20 µL solution of matrix and thoroughly mixed then centrifuged. About 0.5 µL of the resulting mixture was spotted onto the sample plate and allowed to air-dry at room temperature. This method of sample preparation is usually called “dried droplet” and is easier to process than the “sandwich” samples preparation method. Oxidation of r-CD. R-CD (1 g, 6 mmol of glucopyranose units) was dissolved in 20 mL of deionized water. To this solution was added sodium periodate (6.2 mmol) and the mixture was vigorously stirred in the dark at room temperature for 72 h. The product was directly purified by dialysis in cellulose ester tubing (500 Da cutoff) for 7 h, at 25 °C, to remove iodate and unreacted periodate. The product was isolated by lyophilisation. The yellow/white powder, depending on the drying time, was stored at 4 °C in the dark. The product was finally characterized by MALDI-TOF spectrometry and 1 H NMR spectroscopy. This compound is the 2, 5, 7, 10, 12, 15, 17, 20, 22, 25, 27, 30-dodecaformyl- 4, 9, 14, 19, 24, 29hexahydroxymethyl 30-crown-12. For the sake of clarity, this compound will be named R-CD-derived crown(12) aldehyde. Yield 0.696 g. (0.7 mmol, 70%) of white solid. Spermine Grafting on r-CD-Derived Crown(12) Aldehyde. A solution of R-CD-derived crown(12) aldehyde (0.5 g, 3 mmol of initial glucopyranose groups) in 5 mL of deionized water was slowly dropped in a basic solution containing 1.25 equiv amount of spermine to glucopyranose units dissolved in 2.5 mL of deionized water at pH ) 11 (adjusted by NaOH addition). The mixture was gently stirred at room temperature for 24 h. The amine-based derivatives were obtained after reduction of the imine function with an excess of NaBH4 (4 equiv) and stirred for 72 h at room temperature. The product was freeze-dried and precipitated in THF to remove salts and then dried. The product was then freeze-dried from water. For the sake of clarity, this compound will be named R-CD-derived crown(n) spermine, with n being the number of spermine moieties per cyclic molecule. Yield 1.07 g (0.47 mmol, 94%) of white solid. 1H NMR: 1.2-1.4 (m, 8H,-CH2-CH2-CH2), 1.6-2.4 (m, 14H, -CH2-NH-), 2.6-4.2 (m, 6H, CH-O or CH2-O). 13 C NMR: 102.6 (6C, O-CH-O), 70.8 (12C, -CH-O), 58.2 (6C, -O-CH(-O-)-CH2-OH), 55.5 (6C, -O-CH(-CH-)-CH2-OH), 49.1
(18C, -CH-CH2-NH-), 46.8 (18C, CH2-CH2-NH-), 39.3 (6C, CH2-NH2), 36.0 (6C, -NH -CH2-CH2-CH2-NH-), 31.3 (6C, CH2CH2-NH-CH2-CH-O), 26.8 (12C, NH-CH2-CH2-CH2-CH2-NH-). Preparation of the Inclusion Compounds: Pseudopolyrotaxane POE/Modified r-CD. A total of 0.3209 g of R,ω-dimethacrylate j n ) 1100 g · mol-1) was added to a saturated poly(ethylene glycol) M aqueous solution of modified R-CD (2.9 g/20 mL). The mixture was stirred at room temperature for 2 h. According to Harada’s findings,16 the stoichiometries of ethylene oxide units and R-CD in the pseudopolyrotaxane were 2:1 (2nOE ) nR-CD). Compared to the POE-based pseudopolyrotaxane and native R-CD, the mixture did not become turbid, and the complex was collected by freeze-drying to give the pseudopolyrotaxane. At this step, no characterization was performed. The supramolecular assembly was analyzed after the blocking reaction that was undertaken as previously reported.20 Synthesis of the Polyrotaxane Based On Modified r-CD or Native r-CD. A mixture of 1 g of the pseudopolyrotaxane POE/modified R-CD, 340 mg of 1-pyrenebutyric acid N-hydroxysuccinimide ester (npy ) 5 ndim) and 700 mg of ammonium persulfate (the initiator) were introduced into a reaction flask. The reaction started at room temperature with the addition of 17.5 mL of dimethyl sulfoxide in water (33/66; v/v). The concentration of the system must not be lower than 0.1 g/1.75 mL. Roughly, 15 min after the addition of the solvent mixture, the reaction was quenched by dipping the reaction flask in liquid nitrogen, followed by freeze-drying. The resulting solid was dissolved in dimethylformamide or dimethylsulfoxide. It was then precipitated in diethyl ether, filtered, washed, and dried at 60 °C in a drying chamber. The conversion polyrotaxane was determined by SEC using a free R-CD quantification, as already discussed,10 and was at least 95%. Modification of the Polyrotaxane Based On Native r-CD. Polyrotaxane (100 mg, 0.009 mmol of R-CD) was dissolved in 500 L of a solvent blend of DMSO/H2O (30/70, v/v). To this solution was added sodium periodate (0.009 mmol), and the mixture was vigorously stirred in the dark at room temperature for 7 days. The reaction was quenched by dipping the reaction flask in liquid nitrogen, followed by freeze-drying. The resulting solid was dissolved in dimethylsulfoxide. The product was filtered to remove the salt and then evaporated. A solution of this oxidized polyrotaxane (0.1 g, 0.09 mmol of CD) in 1.2 mL of deionized water was slowly dropped in a basic solution containing 1.25 equiv amount of spermine to glucopyranose units dissolved in 0.5 mL of deionized water at pH ) 11 (adjusted by NaOH addition). The mixture was gently stirred at room temperature for 24 h. The amine-based derivatives were obtained after reduction of the imine function with an excess of NaBH4 (4 equiv) and stirred for 7 days at room temperature. The product was extracted in ethanol and then dried.
Hypercharged Polyaminorotaxane
Biomacromolecules, Vol. 9, No. 7, 2008
2009
Figure 1. MALDI-TOF spectrum of R-CD-derived crown(12) aldehyde obtained with 2-nitrophenyl octyl ether: NPOE as matrix (10-1M in THF) of the products. Table 1. Effect of Initial Spermine Concentration on the Modification of the R-CD-Derived Crown(12) Aldehydea run
nCD (10-4 mol)
nspermine (10-4 mol)
[concentration of reactive compounds] in mol · L-1
yield %
spermine number per CD (RMN)
spermine number per CD (acidic titration)
1 2 3
2.08 2.08 2.08
4.54 8.65 13.1
1.56 1.70 1.82
93 92 94
2 3 6
2 3 6
a
In water at room temperature over a period of 96 h.
Scheme 2. Grafting of Spermine Moities on Oxidized R-CD
3. Results and Discussion We first described the modification of R-CDs that is used to control the charge density along the polyrotaxane chain. Then, the polyrotaxane synthesis is investigated by two different strategies. 3.1. r-CD Modification. The R-CD modification was carried out in a two-step procedure adapted from Azzam et al.17–19 The secondary hydroxyl groups were first oxidized. Then, the spermine was grafted by reductive amination. a. Oxidation of R-CD. The R-CD was oxidized using sodium periodate, as described by Immel et al. (Scheme 1).20 As reported, this reaction is specific to vicinal hydroxyl groups. The oxidation reaction was performed until completion (all the secondary hydroxyl groups were transformed into aldehyde groups18). MALDI-TOF spectrometry was conducted using 2-nitrophenyl octyl ether: NPOE as a matrix in THF was used to prove the reaction (Figure 1). Analysis of the MALDI TOF spectra showed that the peaks corresponding to native R-CDs have totally disappeared and new signals at lower atomic mass units (uma) were detected (Figure 1). The ions at m/z 983.7378 and 999.0219 uma correspond to (2+Na)+ and (2+K)+. Due to the analyte composition, the major peak corresponds to the R-CD-derived crown(12) aldehyde, where the cationisation with Na+ is higher than the intensity of the K+ adducts and is identified with an accuracy greater than 0.1 amu. Three other peaks at 1001.0307, 1019.0346, and 1037.0457 uma were detected and deserve some comments. The difference between these peaks is 18 amu and demonstrates the ability of the R-CD-derived crown(12) aldehyde to complex water or to be reversibly bound to the aldehyde functions. No data allows the discrimination between these two interpretations, but the interesting feature is that these structures are stable under the MALDI TOF analysis conditions, up to three complexed/ bound H2O molecules. The origin of the water in our experiments may be attributed to the lyophization step. To confirm the water attribution, the product was dissolved in benzene and freeze-dried before mass spectrometry characterization. Signals attributed to associated water were then turned down.
The NMR characterization carried out in DMSO-d6 provided evidence on the disappearance of the secondary hydroxyl groups. Nevertheless, as already reported,18 quantification of the aldehyde functions by 1H NMR is very difficult. b. Grafting of Spermine. Grafting of spermine onto R-CDderived crown(12) aldehyde is presented in Scheme 2. Table 1 presents the synthesis conditions that were used for this reaction, carried out at room temperature in water. The molar ratio spermine/CD was increased from 2.18 to 6.29. As expected, at constant reaction time, the number of grafted spermine increases with the increasing ratio. Figure 2 shows the NMR spectra of run 3. News peaks attributed to the presence of spermine are clearly evidenced at 1.2, 1.35, and 2.4 ppm. Peaks assignment shows that the characteristic peaks of the R-CD, such as the anomeric hydrogen peak, are shifted upfield and under the peak of solvent. Oxidation reaction leads to a new molecule that has NMR chemical shift largely different from native CD, as observed for other modifications.21 To confirm the NMR quantification, titration was also carried out, and both results were consistent as shown in Table 1. The grafting of spermine on R-CD-derived crown(12) aldehyde seems to be limited to one spermine per glucopyranose unit (run 3), as previously discussed.18 To confirm that assertion, MALDI TOF analysis was also carried out on run 1, Table 1, and the spectrum is presented in
2010
Biomacromolecules, Vol. 9, No. 7, 2008
Pérès et al.
Figure 2. 1H NMR 300 MHz spectrum of R-CD-derived crown(6) spermine (4, run 3) in D2O at room temperature.
Figure 3. MALDI TOF spectrum of run 1, R-CD-derived crown(2) spermine (4, run 1) at 10-3 M in ammonium acetate 10-3 M DHB 10-1 M in THF. Scheme 3. Polyaminorotaxanes Synthesis from Polyrotaxane
Figure 3. As evidenced in the spectra, modification of R-CDderived crown by spermine is a statistical reaction, and monomodified R-CD-derived crown coexists with R-CD-derived crown, having two spermine functions. One has to keep in mind that low molecular weight compounds have a higher desorption efficiency and their magnitude on spectra cannot be taken for granted. R-CD-derived crown having three spermine functions are not evidenced by MS characterization. Titration measurements are in good agreement with the MALDI-TOF measurement that is a mean value of two spermine functions per cyclic compounds for this reaction. 3.2. Polyrotaxane Synthesis. Synthesis of R-CD-based polyrotaxane has been extensively studied.22 The synthesis of polyaminorotaxanes was first carried out by the spermination of premade supramolecular assemblies, then the synthesis of polyaminorotaxane was performed from the R-CD-derived crown(n) spermine. 3.2.1. Polyaminorotaxanes Synthesis by Modification of Premade Polyrotaxane. The synthesis of PEO-based polyrotaxane with R-CD rings was carried out using a radical blocking reaction.23 Conversions of polyrotaxane formation are high, thanks to the heterogeneous reaction medium.10 As previously
reported, the DMSO/H2O (v/v) ratio is a straightforward way to control the amount of CD threaded onto the PEO chain. Indeed, an increase of the DMSO ratio in the solvent mixture leads to polyrotaxane with a low amount of threaded CD.10 Scheme 3 recalls the synthetic pathway for the polyaminorotaxanes synthesis. Such synthesis procedure leads to polyrotaxane having one R-CD ring per two ethylene oxide units.10 Then the vicinal hydroxyl functions of the R-CDs of the polyrotaxane are oxidized, leading to aldehyde functions. Spermine is allowed to react with the R-CD-derived crown
Hypercharged Polyaminorotaxane
Biomacromolecules, Vol. 9, No. 7, 2008
2011
Figure 4. Chromatogram of the different polyrotaxanes in DMF obtained during each step of modification. Scheme 4. Polyaminorotaxanes Synthesis from R-CD-Derived Crown(2) Spermine
adehyde-based polyrotaxane through a standard reductive amination. The supramolecular structure is preserved during all the synthetic steps, as witnessed by the SEC chromatograms (Figure 4). The reaction time was adjusted to allow the complete modification of R-CD-derived crown adehyde on the polyrotaxane; the much longer reaction time (one week) than for free CD can be attributed to hydrogen bonding between R-CDs and to the steric hindrance of the hydroxyl functions in the polyrotaxane. Full modification was witnessed by 1H NMR characterization, that is, the disappearance of the hydroxyl functions signal accompanied by the appearance of a new signal corresponding to the spermine groups. The number of spermine groups borne by the R-CDs in the supramolecular assembly was evaluated by acidic titration. The result of acidic titration indicates 279+ charges per polyrotaxane. Taking into account that each polyrotaxane has an average of 11 R-CDs (SEC analysis) and that each spermine group allows the formation of four charges, one can calculate that one CD is modified by six spermine groups on average. 3.2.2. Polyaminorotaxanes Synthesis from Modified R-CD. R-CD-derived crown(2) spermine is selected as the cyclic compound because it allows the formation of supramolecular assembly with a controlled low amount of spermine per ring, namely, two spermine grafts per R-CD-derived crown. The reaction strategy is reported in Scheme 4. R-CD-derived crown(2) spermine is threaded with PEO to form the pseudopolyrotaxane. In this case, the supramolecular assembly cannot be isolated because it did not give rise to a precipitate, as observed for native R-CD. The viscous medium was freeze-dried and led to a paste that was further used for polyrotaxane synthesis.
The blocking reaction of pseudopolyrotaxane by radical coupling10 optimized in our laboratory permits to isolate the corresponding polyrotaxane with a high conversion. The characterization of the different polyrotaxane was achieved by SEC in DMF where the peak characteristic of free R-CD-derived crown(6) spermine (i.e., not borne by the PEO chain) can be quantified. The chromatogram (Figure 5) demonstrates the presence of objects of high molecular weight and the disappearance of free R-CD-derived crown(2) spermine. The starting reaction conditions are two ethylene oxide units for one R-CD-derived crown(2) spermine. No extractions are carried out and, as witnessed by Figure 5, only a negligible amount of low molecular weight compound is detected, providing an argument for a polyrotaxane structure of 11 R-CD-derived crown(2) spermine units threaded by one PEO chain. To confirm this analysis, an acidic titration of amine functions was used and we concluded there were 90 charges per polyrotaxane. Taking into account the formation of four charges per spermine group, one concludes there were almost 11 R-CD-derived crown(2) spermine units per PEO chain, as determined by the SEC analysis. 3.2.3. SANS Characterization of Polyaminorotaxane. In our synthetic procedures, the more densely charged polyrotaxane is obtained through the modification of premade polyrotaxane by spermine grafts. The internal structure of these supramolecular assemblies was then investigated by SANS. These first experiments were conducted in aqueous solution, as reported for gene vector synthesis24 and we have studied the behavior of the scattered intensity. The examined samples have a welldefined asymptotic behavior exhibited, Figure 6, where we show
2012
Biomacromolecules, Vol. 9, No. 7, 2008
Pérès et al.
Figure 5. Chromatogram of polyrotaxane in DMF obtained from R-CD-derived crown(2) spermine.
Figure 6. q · I(q)/c vs q for polyrotaxane exhibiting a rod-like asymptotic behavior: polyrotaxane issue of polyaminorotaxanes synthesis from polyrotaxane 1% in D2O.
that the normalized quantity q · I(q)/c (c being the concentration) reaches a well-defined limit at large scattering vector. This is typical of a rod-like local behavior, and the persistence length is close to the full extended length of the PEO: 70 Å. This observation was already reported for densely packed CD onto a PEO chain.10,25 The ring opening of the glucopyranose units does not change the internal structure of our R-CD-derived crown-based polyrotaxane. It is demonstrated that the cytoplasm has a pH close to 6.5. We thus added DCl to our samples to drop the pH from 10 to 6.5. In such conditions, a very rapid degradation of our supramolecular structure was observed, as demonstrated by the low diffusion intensity whatever the vector of diffusion. This result proves the high degradability of our supramolecular assembly in physiological conditions.
4. Conclusion New hydrolyzable polyaminorotaxanes were designed through two pathways. In a first strategy, a premade R-CDbased polyrotaxane was grafted by spermine moieties. The only way to control the grafting density is to fully modify each R-CDs by six spermine moieties. The new R-CD-derived crown(6) spermine-based polyrotaxanes were characterized by NMR and SEC to demonstrate that the CD are completely modified and that the supramolecular structure is retained.
These conclusions are supported by the titration results. Reduction of the number of spermine moieties per R-CD may also be possible through this strategy, but a lack of control of the distribution of these spermine groups along the polyrotaxane structure is evident. To circumvent this drawback, we designed another strategy that is the modification of R-CD followed by the polyrotaxane synthesis. We were able to achieve a fair control of the modification of the CD, and we selected the R-CD-derived crown(2) spermine to construct a polyrotaxane. The supramolecular structure is demonstrated by SEC and NMR, and a high conversion is achieved. These new supramolecular structures open new routes for gene transfer applications, and this very versatile polyaminorotaxane construction will be an easy way to control the amount of spermine moieties per cyclic molecules, as well as the number of rings per macromolecule as already demonstrated.10 These methods could be used to adjust the structure of the polyrotaxane chain and the charge density along the polyrotaxane chain.
References and Notes (1) Seymour, L., Kataoka, K.; Kabanov, A. Self assembling complexes for genes deliVery; Kabanov, A., Felgner, P., Seymour, L., Willey, L., Eds.; 1998, 219. (2) Eliyahu, H.; Barenholz, Y.; Domb, A. J. Molecules 2005, 10, 34–64. (3) Boussif, O.; Lezouac´h, F.; Zanta, M. A.; Mergny, M. D.; Sherman, D.; Demeneix, B.; Berh, J. P. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 7297–7301. (4) Midoux, P.; Mendes, C.; Legrand, A.; Raimond, J.; Mayer, R.; Monsigny, M.; Roche, A. C. Nucleic Acid Res. 1993, 21, 871–878. (5) Verbaan, F. J.; Oussoren, C.; Dam, I. M.; Takakura, Y.; Hashida, M.; Crommelin, D. J.; Hennink, W. E.; Storm, G. Int. J. Pharm. 2001, 214, 99–101. (6) Ooya, T.; Yamashita, A.; Kurisawa, M.; Sugaya, Y.; Maruyama, A.; Yui, N. Sci. Technol. AdV. Mater. 2004, 5, 363–369. (7) Yamashita, A.; Yui, N.; Ooya, T.; Kano, A.; Maruyama, A.; Akita, H.; Kogure, K.; Harashima, H. Nature Protoc. 2006, 1, 2861–2869. (8) Young, Y. K.; Sengoku, Y.; Ooya, T.; Park, K. D.; Yui, N. Sci. Technol. AdV. Mater. 2005, 6, 484–490. (9) Ooya, T.; Choi, H. S.; Yamashita, A.; Yui, N.; Sugaya, Y.; Kano, A.; Maruyama, A.; Akita, H.; Kogure, K.; Harashima, H. J. Am. Chem. Soc. 2006, 128, 3852–3853. (10) Jarroux, N.; Guegan, P.; Cheradame, H.; Auvray, L. J. Phys. Chem. B. 2005, 109, 23816–23822. (11) Eguchi, M.; Ooya, T.; Yui, N. J. Controlled Release 2004, 96, 301–307. (12) Ooya, T.; Eguchi, M.; Ozaki, A.; Yui, N. Intern. J. Pharm. 2002, 242, 47–54. (13) Yui, N.; Ooya, T.; Kawashima, T.; Saito, Y.; Tamai, I.; Sai, Y.; Tsuji, A. Bioconjugate Chem. 2002, 13, 582–587. (14) Huang, F.; Gibson, H. W. Prog. Polym. Sci. 2005, 30, 982–1018. (15) Harada, A.; Kamach, M. Macromolecules 1990, 23, 2821–2823.
Hypercharged Polyaminorotaxane (16) Harada, A.; Li, J.; Kamachi, M. Macromolecules 1994, 27, 4538–4543. (17) Azzam, T.; Raskin, A.; Makovitzki, A.; Brem, H.; Vierling, P.; Lineal, M.; Domb, A. J. Macromolecules 2006, 35, 3852–3853. (18) Kandra, L.; Liptak, A.; Jodal, I.; Nanasi, P.; Szejtli, J. J. Inclusion Phenom. 1984, 2, 869–875. (19) Szejtli, J.; Kandra, L. J. Inclusion Phenom. 1987, 5, 639–643. (20) Immel, S.; Lichtenthler, F. W.; Linder, H. J.; Nakagawa, T. Tetrahedron: Asymmetry 2001, 12, 2767–2774. (21) Harabagiu, V.; Simionescu, B. C.; Pinteala, M.; Merrienne, C.; Mahuteau, J.; Gue´gan, P.; Cheradame, H. Carbohydr. Polym. 2004, 56, 301–311.
Biomacromolecules, Vol. 9, No. 7, 2008
2013
(22) Wenz, G.; Han, B. H.; Muller, A. Chem. ReV. 2006, 106 (3), 782– 817. (23) Jarroux, N.; Guegan, P.; Buchmann, W.; Tortajada, J.; Cheradame, H. Macromol. Chem. Phys. 2004, 205, 1206–1217. (24) Mockey, M.; Bourseau, E.; Chandrashekhar, V.; Chaudhuri, A.; Lafosse, S.; Le Cam, E.; Quesniaux, V. F. J.; Ryffel, B.; Pichon, C.; Midoux, P. Cancer Gene Ther. 2007, 802–814. (25) Karino, T.; Okumura, Y.; Ito, K.; Shibayama, M. Macromolecules 2004, 6177–6182.
BM800247C
