Glucose-Responsive Polymer Bearing a Novel Phenylborate

Xudong Deng , Rana Attalla , Lukas P. Sadowski , Mengsu Chen , Michael J. ...... Zachary Sharrett , Soya Gamsey , Jonathan Fat , Daniel Cunningham-Bry...
0 downloads 0 Views 115KB Size
Biomacromolecules 2003, 4, 1410-1416

1410

Glucose-Responsive Polymer Bearing a Novel Phenylborate Derivative as a Glucose-Sensing Moiety Operating at Physiological pH Conditions Akira Matsumoto, Syuhei Ikeda, Atsushi Harada, and Kazunori Kataoka* Department of Materials Science and Engineering, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan Received May 8, 2003; Revised Manuscript Received July 4, 2003

This study is devoted to the development of novel glucose-responsive polymers that operate under physiological conditions (pH 7.4, 37 °C), aiming for future use in a self-regulated insulin delivery system to treat diabetes mellitus. The approach involves the use of a newly synthesized phenylborate derivative {4-(1,6-dioxo-2,5-diaza-7-oxamyl) phenylboronic acid, DDOPBA} possessing an appreciably low pKa (∼7.8) as a glucose-sensing moiety, as well as the adoption of poly(N-isopropylmethacrylamide), PNIPMAAm, as the main chain that exhibits critical solution behavior in the range close to physiological temperature. Glucoseand pH-dependent changes in the critical solution behavior of the resultant copolymers were investigated at varying temperatures, revealing definite glucose sensitivities near the physiological conditions. Furthermore, DDOPBA moieties in the copolymers maintained constant apparent pKa values even when the temperature approaches the critical solution points of the main chain, indicating that spacing of the phenylborate moiety from the polymer backbone is a feasible way to minimize the microenvironment effect caused by a temperature-induced change in the hydration state of the polymer strands. Introduction Over the past several decades, stimuli-responsive polymer gels have attracted a great deal of research interest. A series of stimuli that includes heat,1-4 pH,5,6 electric fields,7 and light8 have been demonstrated to induce abrupt changes in the physical properties of polymer gels. This knowledge consequently enabled us to develop various types of stimuliresponsive, hence “self-regulated”, systems that are often called “intelligent” or “smart” materials. Among an extensive number of fields in which these systems have been successfully applied, applications in the medical field such as “drugdelivery systems” have been a constant research topic, revealing their potential and growing utility. One example of such an urgently demanded system, and what would surely be an attractive breakthrough for the treatment of diabetes, is a self-regulated insulin delivery system.9 The treatment for insulin-dependent diabetes mellitus (IDDM) practically conducted today is limited to insulin injection that significantly burdens patients. One of the more significant problems in this treatment is the difficulty in controlling the insulin dosage. An overdose of insulin causes an excessively decreased blood sugar level that may lead to serious hypoglycemia. These circumstances have inspired us to develop a self-regulated (and thus safer) insulin delivery system based on stimuli-responsive polymer gels. As a sensor moiety for the blood sugar level, we have paid particular attention to the uniqueness of phenylboronic acid. Phenylboronic acid and its derivatives are known to * To whom correspondence should be addressed. Telephone: +81-35841-7138. Fax: +81-3-5841-7139. E-mail: [email protected].

Figure 1. Equilibria of (alkylamido) phenylboronic acid.

form reversible covalent complexes with diol units, such as glucose.10-12 On the basis of this property, they have been primarily utilized as ligand moieties for affinity chromatography of polyol compounds.13,14 More recently, such applications as glyco-responsive polymer complex systems15,16 and synthetic mitogens for lymphocytes17,18 have been reported. As shown in Figure 1, phenylboronic acid compounds exist in equilibrium between the undissociated (neutral trigonal) and dissociated (charged tetrahedral) forms in an aqueous milieu. An increase in the glucose concentration induces a shift in the equilibrium, increasing the fraction of total borate anions (2 and 3) and decreasing the fraction of the uncharged form (1). The shift toward the charged phenylborates (2 and 3) occurs because only charged borates can form a stable complex with glucose in an aqueous milieu. The direct complexation of the uncharged form (1) with glucose is

10.1021/bm034139o CCC: $25.00 © 2003 American Chemical Society Published on Web 08/19/2003

Glucose-Responsive Polymer

unstable in water because of its high susceptibility to hydrolysis.19 This shift in the direction of the charged phenylborates results in an increased hydrophilicity. Thus the change in the glucose concentration should have a significant effect on the solubility of amphiphilic polymer strands having pendant phenylborate moieties.20 Furthermore, in a cross-linked polymer network (polymer gel) with phenylborate moieties, the glucose-induced increase in the fraction of charged borates should generate a corresponding increase in the osmotic pressure arising from mobile counterions, resulting in an abrupt volume change of the gel, that is, a volume phase transition.21 We have previously reported that a gel composed of N-isopropylacrylamide (NIPAAm) and 3-acrylamidophenylboronic acid (AAPBA), referred to as an NB gel, exhibits a reversible volume change synchronized with a change in the glucose concentration, through which the sufficiently controlled and pulse-shaped release of insulin was achievable at 28 °C in a pH 9.0 solution.21 We expect this totally synthetic system, which includes no natural components such as proteins22,23 or enzymes24,25 that could potentially be denatured, to be highly stable in the human body and to exhibit a reliable functionality. However, this system was not able to adequately function under physiological conditions (37 °C, pH 7.4) for some latent molecular reasons as are described in the latter part of this paper. Hence, our ultimate goal is to establish such a system that is functional under these conditions for use in the human body. Accordingly, in this article, we report for the first time a synthetic approach toward the achievement of such a system. The approach presented here includes the utilization of a newly synthesized phenylborate derivative as to possess a lower pKa, as well as the adoption of a different type of structure for the main chain component. The glucose-responsive behavior of the resultant copolymer systems was investigated through the glucose- and pHdependent change in their lower critical solution temperature (LCST). In addition, the potentiometric titration of the copolymers under varying glucose concentrations and temperature were conducted to elucidate the disassociative behavior of the phenylborate moieties. All of the data were successfully correlated with the intended molecular design, and a discussion was presented in terms of the effect of the modulated copolymer structure on the glucose-responsive behavior. Materials and Methods Materials. N-Isopropylacrylamide (NIPAAm) was purchased from Kojin (Japan) and recrystallized from n-hexane. N-Isopropylmethacrylamide (NIPMAAm) was prepared by the reaction of methacryloyl chloride (Wako, Japan) with isopropylamine (Wako, Japan) in the same manner as has been previously reported26-28 and then recrystallized from a 20/80 mixture of toluene and hexane. 2,2′-Azobisisobutyronitrile (AIBN) (Wako, Japan) was recrystallized from methanol. 3-Acrylamidephenylboronic acid (AAPBA) was supplied from Wako (Japan) and was used without further purification. 4-Carboxyphenylboronic acid (CPBA) (Lan-

Biomacromolecules, Vol. 4, No. 5, 2003 1411

caster, U.K.) and acryloyl chloride (Wako, Japan) were used as received. Ethylenediamine was purchased from Wako (Japan) and purified by distillation before use. Synthesis of 4-(1,6-Dioxo-2,5-diaza-7-oxamyl) Phenylboronic Acid (DDOPBA). The synthesis of 4-(1,6-dioxo2,5-diaza-7-oxamyl) phenylboronic acid (DDOPBA) was accomplished by a two-step reaction as shown in Figure 2. 4-(Chloroformyl) phenylboronic acid was first reacted with an excess amount of ethylenediamine to obtain 4-[(2aminoethyl)carbamoyl] phenylboronic acid (AECPBA), followed by the reaction with acryloyl chloride to yield DDOPBA. The detailed procedure is as follows: 4-Carboxyphenylboronic acid, CPBA (10 g, 60.3 mmol), was initially dried in vacuo in a two-neck round-bottom flask. After the flask was charged with argon, thionyl chloride (150 mL, 2.1 mol) was added under an argon atmosphere. The suspension was stirred and refluxed at 88 °C for 24 h to give 4-(chloroformyl) phenylboronic acid. After the remaining thionyl chloride was evaporated, the flask was charged with argon again. The content of the flask [4-(chloroformyl) phenylboronic acid] was suspended in 60 mL of distilled THF. The cooled suspension in an ice bath was then slowly added in dropwise manner to the cooled, distilled ethylenediamine (200 mL, 3.0 mol) in the presence of distilled triethylamine (10 mL, 71.9 mmol) under an argon atmosphere. After the mixture was stirred for 20 h, the ethylenediamine was evaporated, and then the residue was dissolved in 100 mL of distilled water. Thereafter, the pH of the solution was adjusted to ca. 4 by the addition of 1 N HCl. A white precipitate, assigned from the 1H NMR to be mainly the 2:1 reaction product of 4-(chloroformyl) phenylboronic acid with ethylenediamine [4,4′-(ethylenedicarbamoyl) phenylboronic acid], was filtered off. The filtrate was concentrated and then stored at 4 °C overnight to produce a white crystalline product. The collected crystals of 4-[(2-aminoethyl)carbamoyl) phenylboronic acid (AECPBA) were dissolved in water and recrystallized twice. Yield: 4.94 g (49% based on CPBA). 1H NMR (400 MHz, D2O) δ: 3.3 [NH2-CH2CH2-, 2H], 3.7 [NH2-CH2-CH2-, 2H], 7.8 [-NH-COC6H4-B(OH)2, 4H]. The synthesized 4-[(2-aminoethyl)carbamoyl) phenylboronic acid (AECPBA, 4 g, 19.2 mmol) was dissolved in 150 mL of freshly prepared 1 N NaOH under an argon atmosphere and the solution was stirred and cooled in an ice bath. Acryloyl chloride (5.2 mL, 57.6 mmol) was slowly added in drops to the solution. The reaction mixture was stirred for 24 h and concentrated. Subsequently, the pH of the solution was adjusted to ca. 4 and then the solution was stored stored at 4 °C overnight to produce a white crystalline product. The collected crystals of DDOPBA were again recrystallized from distilled water. Yield: 2.87 g (72% based on AECPBA). 1H NMR (400 MHz, D2O) δ: 3.5 [-NHCH2-CH2-NH-, 4H], 5.8 [CH2dCH-CO-NH-, 1H], 6.1-6.3 [CH2dCH-CO-NH-, 2H], 7.6 [-NH-C6H4B(OH)2, 4H]. Polymer Synthesis. Each copolymer was prepared by radical polymerization in methanol in a sealed glass ampule in vacuo at 60 °C for 24 h using AIBN as an initiator. The

1412

Biomacromolecules, Vol. 4, No. 5, 2003

Matsumoto et al.

Figure 2. Synthetic route of 4-(1,6-dioxo-2,5-diaza-7-oxamyl) phenylboronic acid (DDOPBA). Table 1. Composition of Phenylborate-Containing Polymersa

DDOPBA

yield [wt %]

borate contents in copolymerb [mol %]

10 10 20

50.1 85.7 48.6 70.2

10.5 9.5 10.0 18.4

feed composition [molar ratio] abbreviation

NIPAAm

PNIP-A-10.5 PNIP-D-9.5 PNIPM-D-10.0 PNIPM-D-18.4

90 90

NIPMAAm

AAPBA 10

90 80

a Initiator (azobisisobutylnitrile) at 10 mmol; monomer concentration of 1 mol/L; reaction temperature of 60 °C; reaction time of 24 h. b Determined from H NMR spectrum in DMSO at 80 °C.

detailed reaction conditions are summarized in the footnote of Table 1. After the polymerization, the reacted mixture was poured into a large excess of diethyl ether. The precipitated polymer was filtered off and dried in vacuo. The copolymer compositions were determined by 1H NMR (EX400, JEOL, Tokyo, Japan, DMSO-d6) from the peak intensity ratios between the methine proton on the isopropyl group either of NIPAAm or NIPMAAm (-CH(CH3)2, δ ) 3.8 ppm) and the aromatic protons either of AAPBA or DDOPBA (COC6H4, δ ) 7.8-7.9 ppm). Determination of Lower Critical Solution Temperature (LCST). The LCSTs of the 0.1 wt % polymer solutions were determined by observing the change in the turbidity at 500 nm using a Jasco model V-5500 spectrophotometer. Each sample was set in a thermostatic cell holder coupled with a ETC-505T controller. The heating rate was 0.2 °C/min. The LCST was defined as the temperature at the inflection point in the transmittance versus temperature curve. The measurements were conducted for a series of pH-conditioned buffer solutions (I ) 0.15) in the presence (5 g/L) and absence of glucose.

Potentiometric Titration. A potentiometric titration was performed at varied glucose concentrations and temperature conditions under an argon atmosphere using an automatic titrator system (DL25, Mettler Toledo, Germany) equipped with a pH electrode (glass electrode DG111-SC, Mettler Toledo, Germany). The calibration of the electrode was carried out using carbonate and phthalate buffers (Beckman, U.S.A.) at each experimental temperature. The measurement error was less than 0.01 pH units. Samples were prepared by dissolving each monomer (30 mg) or copolymer (100 mg) in 40 mL of freshly prepared 0.01 N NaOH solution (150 mM NaCl) adjusted to a series of glucose concentrations. The water was purified by a Milli-Q instrument (Millipore Corporation, U.S.A.). The titrant (0.01 N HCl, 150 mM NaCl) was added in 0.05 mL portions at 12-120 s intervals when the drift equilibrium reached the rate of 1 mV/s. The transmittance change at 600 nm was also monitored through a photoelectric probe (DP600, Mettler Toledo, Germany) during titration. Consequently, only the region of the curves where no precipitation occurred (no transmittance change was detectable) was used for the evaluation.

Glucose-Responsive Polymer

Biomacromolecules, Vol. 4, No. 5, 2003 1413

Results and Discussion The limitation of the pH conditions under which the NB system21 is operational is derived from the low acidity of the utilized phenylborate derivative (AAPBA, pKa ) 8.2). Assessed from the titration curves of the linear copolymer of NIPAAm with AAPBA (9:1 in molar ratio, pKa ) 8.9 at 15 °C) in the milieu at pH 7.4, the fraction of borate anions increases from 0.18 to 0.31 (the corresponding change in the LCST is from 13 to 15 °C) upon the addition of 5 g/L of glucose, whereas in the milieu at pH 9.0, the fraction increases more appreciably from 0.56 to 0.93 (the corresponding change in the LCST is from 22 to 36 °C) with the addition of the same amount of glucose, resulting in a sufficient glucose sensitivity by the system. Considering that the “glucose sensitivity” of the system is based on the glucose-induced fraction change between the uncharged (1, Figure 1) and the charged forms (2 and 3) of the phenylborates and that the shift in the equilibrium between 1 and 2 and 3 is caused by complexation between 2 and glucose, the use of a novel phenylborate derivative with a lower pKa {increased fraction of 2} was thought to be crucial to achieve a sufficient sensitivity at a lower (closer to physiological) pH condition. The chemical-structural design, which would decrease the pKa value of phenylborate, was attempted by introducing a stronger electron-withdrawing group into the phenyl ring. The electron-withdrawing effect should cause a decreased electron density on the boron atom, making the borate more acidic (decreasing pKa).29 Based on this strategy, a novel phenylborate-containing monomer, 4-(1,6-dioxo-2,5-diaza7-oxamyl) phenylboronic acid (DDOPBA) was synthesized (Figure 2). One may notice that DDOPBA is designed to possess the para-positioned carbamoyl group as its substituent and that the substituent constant (σ) given for the group is larger (0.62)30 than the value given for the meta-positioned amide group (0.21)30 as in the case of AAPBA (Figure 3), indicating the relatively strong electron-withdrawing effect of the para-positioned carbamoyl group. The linear relation between the substituent constants (σ) and the pKa values (Figure 3)31,32 predicts the reduced pKa value of DDOPBA as low as 7.5 due to the modified substituent effect, whereas the pKa value of AAPBA was formerly determined to be 8.2. To quantitatively evaluate such modified-structure effects on the glucose-sensitive behavior, the apparent pKa (pKa,app) values with varied glucose concentrations were examined via potentiometric titration for both the newly designed (DDOPBA) and the conventional phenylborate-containing monomer (AAPBA). The relations between the pKa,app values assessed from each titration curve and the glucose concentration at 25 °C are shown in Figure 4. Both monomers display significant decreases in their pKa,app with the increased glucose concentration due to a stabilization of the anionic borates through complexation with glucose. What should be noticed is that a comparison of these two kinds of monomers reveals the pKa,app values of DDOPBA to be approximately 0.4 lower than those of AAPBA over the entire range of analyzed glucose concentrations, indicating a more stable form of charged borates in DDOPBA as a result of the modified-structure effects stated above. The pKa value of

Figure 3. Relationship between pKa values of substituted phenylboronic acid and substituent constants. All of the substituent constants were from ref 29. As the substituent constant for the para-positioned carbamoyl, σ+ was adopted, which is the value corrected to fit on the meta-substitution-based line. The pKa value for the amide at the metaposition (AAPBA) was determined by the author. The value for metasubstituted NO2 was from ref 28. All of the rest were from ref 31.

Figure 4. Glucose-dependent changes in the apparent pKa values of phenylborate-containing monomers at 25 °C, as determined from potentiometric titrations: (a) AAPBA (O); (b) DDOPBA (b).

DDOPBA at 25 °C (in the absence of glucose) was determined to be 7.79, which has a good correlation with the calculated values from the σ vs pKa line shown in Figure 3. Thus, the new molecular design has achieved a phenylborate-containing monomer endowed with a lowered pKa (DDOPBA), which should exhibit a sufficient glucose sensitivity at the lower pH closer to physiological conditions. Then, PNIPAAm-based copolymers bearing approximately 10 mol % of phenylborate with different pKa’s, namely, AAPBA (pKa ) 8.2) and DDOPBA (pKa ) 7.8), were prepared. The obtained copolymers, poly(NIPAAm-coAAPBA)[89.6/10.5] and poly(NIPAAm-co-DDOPBA)[90.5/ 9.5], were referred to as PNIP-A-10.5 and PNIP-D-9.5, respectively. Figures 5 and 6 show the pKa,app values of the

1414

Biomacromolecules, Vol. 4, No. 5, 2003

Figure 5. Temperature dependence of the apparent pKa values determined for PNIP-A-10.5 (b) and AAPBA (O).

Figure 6. Temperature dependence of the apparent pKa values determined for PNIP-D-9.5 (b) and DDOPBA (O).

respective copolymers in comparison with each corresponding monomer determined at various temperatures. Both copolymers have somewhat higher pKa,app values than those of the corresponding monomers, indicating a weakened acidity of the phenylborate moieties in these copolymers. This may be attributed to the nature of the polyelectrolytes where the phenylborate moieties are spatially trapped; that is, as the dissociation proceeds, the concentration of negative charges in the immediate vicinity of the phenylborates is increased and thus further dissociation of the phenylborates is impeded.33 Nevertheless, as expected from the pKa values of the monomers, PNIP-D-9.5 showed a lower pKa,app than PNIP-A-10.5 over the entire range of evaluated temperatures, reflecting the more-stabilized state of the charged borates in DDOPBA. Interestingly enough, PNIP-A-10.5 displays a tendency of increasing pKa,app as the temperature increases closer to the precipitation point (LCST) as seen in Figure 5, whereas PNIP-D-9.5 (Figure 6) and both monomers show almost constant values over the evaluated temperature range. Although further investigations are needed to give a clear interpretation of the phenomenon, one possibility is to assume

Matsumoto et al.

Figure 7. Glucose- and pH-dependent changes in the LCST of PNIPA-10.5 (O, no glucose; b, glucose concentration of 5 g/L) and PNIPD-9.5 (4, no glucose; 2, glucose concentration of 5 g/L).

that polymer coils exist in a less-expanded state due to the relatively reduced degree of hydration as the temperature increases closer to the precipitation point (LCST).25 This may cause a decreased local permittivity around the phenylborate moieties or an even more emphasized polyelectrolyte effect than is described above, resulting in an increased pKa,app value with a temperature increase for PNIP-A-10.5. The negligible temperature-induced increase in the pKa,app of PNIP-D-9.5 may then be due to the increased distance from the main chain and, therefore, the greater mobility of the phenylborate moieties, reducing a microenvironment effect induced by the polymer strands. In a good agreement with the above discussion is the pHdependent change in the LCST of these copolymers. As seen in Figure 7, the LCST of both PNIP-A-10.5 and PNIP-D9.5 increases at the higher pH and in the presence of glucose where the number of charged borates increases. Worth noticing is that the pH region at which the LCST of PNIPD-9.5 shows a considerable change has been lowered as compared with the case of PNIP-A-10.5, being consistent with their pKa,app values. Additionally, careful observation of Figure 7 reveals that the change in LCST with pH of PNIP-D-9.5 is steeper than that of PNIP-A-10.5 and becomes more remarkable in the higher temperature range. This is also consistent with the different temperature dependencies of their pKa,app values observed in Figures 5 and 6. Namely, the dissociation of phenylborates in PNIP-A-10.5 is inhibited with increased temperature due to the increasing pKa,app (Figure 5), and thus a rise in the LCST with a pH increase is correspondingly suppressed. On the other hand, in PNIPD-9.5, the dissociation of phenylborates is not affected by increasing the temperature, as can be seen in Figure 6, thus leading to a constant increase in the LCST with pH. As the next stage of the molecular design, for the purpose of improving the temperature conditions under which our system can operate, we changed the backbone structure of the copolymer from poly(N-isopropylacrylamide) (PNIPAAm) to poly(N-isopropylmethacrylamide) (PNIPMAAm). It is generally accepted that the LCST of amphiphilic polymers

Glucose-Responsive Polymer

Figure 8. Glucose- and pH-dependent changes in the LCST of PNIPD-9.5 (4, no glucose; 2, glucose concentration of 5 g/L) and PNIPMD-10.0 (0, no glucose; 9, glucose concentration of 5 g/L).

decreases with increased hydrophobicity of their structure. Nevertheless, PNIPMAAm, containing the R-methyl group and thus being more hydrophobic than PNIPAAm, exhibits higher LCST than PNIPAAm.26-28 The well-accepted explanation for this contradictory behavior of the higher LCST of PNIPMAAm than of PNIPAAm is the restricted free rotation of the main chains of PNIPMAAm induced by the R-methyl group, preventing the hydrophobic groups from aggregating in the most favorable manner.27,28 The glucoseand pH-dependent changes in the LCST of poly(NIPMAAmco-DDOPBA) [90.0/10.0], referred to as PNIPM-D-10.0, are shown in Figure 8 with those of PNIP-D-9.5 as the control. While PNIPM-D-10.0 maintains a similar pH dependency to that of the control, the temperature range where the curves of PNIPM-D-10.0 appear has drastically shifted to the higher temperature range, resulting in a conspicuous glucose sensitivity (a difference between LCSTs in the presence and the absence of glucose) at around physiological temperature, at least in a slightly alkaline condition (∼pH 8), due to the use of PNIPMAAm, which produces a higher LCST than that of PNIPAAm. It is interesting to note that, in the case of PNIPM-D-18.4 (Figure 9) with the higher borate content than PNIPM-D10.0, the LCST of the polymer significantly decreases because of an increase in the fraction of hydrophobic (uncharged) DDOPBA at a lower range of pH. However, as the pH increases, a more drastic change in LCST and thus a larger sensitivity is accomplished because of a more appreciable change in the amount of the anionic charge. Moreover, around the physiological pH of 7.4, a large-enough glucose-induced LCST change can be seen, indicating a sensitivity equivalent to that of the poly(NIPAAm-coAAPBA) (NB) system, which was achieved at pH 9.21 Conclusions In an attempt to establish a synthetic route to adjust the phenylborate-based glucose-responsive system (NB system) to be operational under physiological conditions, modifica-

Biomacromolecules, Vol. 4, No. 5, 2003 1415

Figure 9. Glucose- and pH-dependent changes in the LCST of PNIPM-D-10.0 (0, no glucose; 9, glucose concentration of 5 g/L) and PNIPM-D-18.4 (], no glucose; [, glucose concentration of 5 g/L).

tions were made on the molecular structure of the polymer system. First, intending to optimize the operational pH condition, a novel phenylboronic acid-containing monomer, 4-(1,6-dioxo-2,5-diaza-7-oxamyl) phenylboronic acid (DDOPBA), which was designed to possess a lower pKa, was prepared. It was demonstrated that the system is operational at lower (closer to physiological) pH conditions with the use of DDOPBA because of the lowered pKa effect, as well as its excellent temperature dependency. Second, to improve the operational temperature conditions, the backbone structure of the system was changed from PNIPAAm to PNIPMAAm. As a result, the temperature region in which the system is functional was drastically increased, leading to a conspicuous glucose sensitivity of the system around the physiological temperature. Furthermore, it was demonstrated that a poly(NIPMAAm-co-DDOPBA) system with an increased borate content becomes operational under physiological pH conditions. All of these accomplished improvements in the operational conditions of the system and the confirmed consistency with the modulated chemical structures suggest that further improvement of the system via synthetic methodology can be readily achievable. We expect this totally synthetic system, which is composed of no natural components such as proteins22,23 or enzymes24,25 that could potentially be denatured, to be highly stable in the human body with a reliable functionality. Hence, we may conclude that, upon further improvement, this system may serve as a landmark selfregulated glucose-sensitive device that may overcome many of the difficulties in the treatment of diabetes carried out today. Acknowledgment. The authors thank Professor T. Aoyagi, Department of Nanostructured and Advanced Materials, Graduate School of Science and Engineering, Kagoshima University, for his technical advice on the synthesis of DDOPBA. The present work is supported in part by a Grant-in-Aid for Scientific Research on Priority

1416

Biomacromolecules, Vol. 4, No. 5, 2003

Area (A) “Molecular Synchronization for Design of New Materials System” from the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT) and also by a Grant for 21st Century COE Program “Human-Friendly Materials based on Chemistry” from MEXT. References and Notes (1) Hirokawa, Y.; Tanaka, T. J. Chem. Phys. 1984, 81, 6379-6380. (2) Hoffman, A. S.; Afrassiabi, A.; Dong, L. C. J. Controlled Release 1986, 4, 213-222. (3) Okano, T.; Bae, Y. H.; Jacobs, H.; Kim, S. W. J. Controlled Release 1990, 11, 255-265. (4) Schild, H. G. Prog. Polym. Sci. 1992, 17, 163-249. (5) Tanaka, T.; Fillmore, D.; Sun, S. T.; Nishio, I.; Swislow, G.; Shah, A. Phys. ReV. Lett. 1980, 45, 1636-1639. (6) Siegel, R. A.; Firestone, B. A. Macromolecules 1988, 21, 32543259. (7) Tanaka, T.; Nishio, I.; Sun, S. T.; Ueno-Nishio, S. Science 1971, 218, 467-469. (8) Suzuki, A.; Tanaka. T. Nature 1990, 346, 345-237. (9) For some examples of related approaches, see: Qiu, Y.; Park, K. AdV. Drug DeliVery ReV. 2001, 53, 321-339. (10) Boeseken, J. AdV. Carbohydr. Chem. 1949, 47, 189-210. (11) Aronoff, S.; Chen, T.; Cheveldayoff, M. Carbohydr. Res. 1975, 40, 299-309. (12) Foster, A. B. AdV. Carbohydr. Chem. 1957, 12, 81-115. (13) Weith, H.; Wiebers, J.; Gilham, P. Biochemistry 1970, 9, 43964401. (14) Gelijkens, C.; Deleenheer, A. J. Chromatogr. 1980, 183, 78-82. (15) Kitano, S.; Kataoka, K.; Koyama, K.; Okano, T.; Sakurai, Y. Makromol. Chem. Rapid Commun. 1991, 12, 227-233. (16) Kitano, S.; Koyama, K.; Kataoka, K.; Okano, T.; Sakurai, Y. J. Controlled Release 1992, 19, 161-170.

Matsumoto et al. (17) Miyazaki, H.; Kikuchi, A.; Koyama, Y.; Okano, T.; Sakurai, Y.; Kataoka, K. Biochem. Biophys. Res. Commun. 1993, 195, 829-836. (18) Uchimura, E.; Otsuka, H.; Okano, T.; Sakurai, Y.; Kataoka, K. Biotechnol. Bioeng. 2000, 72, 307-314. (19) Lorand, J. P.; Edwards, J. D. J. Org. Chem. 1959, 24, 769-774. (20) Kataoka, K.; Miyazaki, H.; Okano, T.; Sakurai, Y. Macromolecules 1994, 27, 1061-1062. (21) Kataoka, K.; Miyazki, H.; Bunya, M.; Okano, T.; Sakurai, Y. J. Am. Chem. Soc. 1998, 120, 12694-12695. (22) Kim, S. W.; Pai, C. M.; Makino, K.; Seminoff, L. A.; Holmberg, D. L.; Gleeson, J. M.; Wilson, D. E.; Mack, E. J. J. Controlled Release 1990, 11, 193-201. (23) Sato, S.; Jeong, Y.; Macrea, J. C.; Kim, S. W. J. Controlled Release 1984, 1, 67-77. (24) Ishihara, K.; Muramoto, N.; Shinohara, I. J. Appl. Polym. Sci. 1984, 29, 211-217. (25) Kost, J.; Horbett, T. A.; Ratner, B. D.; Singh, M. J. Biomed. Mater. Res. 1985, 19, 1117-1133. (26) Fujishige, S.; Kubota, K.; Ando, I. J. Phys. Chem. 1989, 93, 33113313. (27) Kubota, K.; Hamano, K.; Kuwahara, N.; Fujishige, S.; Ando, I. Polym. J. 1990, 22, 1051-1057. (28) Kubota, K.; Fujishige, S.; Ando, I. Polym. J. 1990, 2, 15-20. (29) Shinghal, R. P.; Ramamurthy, B.; Govindraj, N.; Sarwar, Y. J. Chromatogr. 1991, 543, 17-38. (30) Swan, C. G.; Lupton, E. C., Jr. J. Am. Chem. Soc. 1968, 90, 43284337. (31) Washburm, R. M.; Billig, F.; Bloom, M.; Albrught, C. F.; Levens, E. AdV. Chem. Ser. 1961, 32, 218-220. (32) Ingold, C. K. Structure and Mechanism in Organic Chemistry; Cornell University Press: New York, 1953; Chapter XIII, pp 738, 741, 750. (33) Elias, H. G. Macromolecules Synthesis and Materials; Plenum Press: New York and London, 1997; Vol. 2, pp 811-813.

BM034139O