Facile Synthesis of Acid-Labile Polymers with ... - ACS Publications

Article pubs.acs.org/Biomac

Facile Synthesis of Acid-Labile Polymers with Pendent Ortho Esters Jing Cheng,† Ran Ji,† Shi-Juan Gao,‡ Fu-Sheng Du,*,† and Zi-Chen Li*,† †

Beijing National Laboratory for Molecular Sciences, Key Laboratory of Polymer Chemistry and Physics of Ministry of Education, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, People's Republic of China ‡ CAS Key Laboratory of Pathogenic Microbiology and Immunology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, People's Republic of China S Supporting Information *

ABSTRACT: This work presents a facile approach for preparation of acid-labile and biocompatible polymers with pendent cyclic ortho esters, which is based on the efficient and mild reactions between cyclic ketene acetal (CKA) and hydroxyl groups. Three CKAs, 2-ethylidene1,3-dioxane (EDO), 2-ethylidene-1,3-dioxolane (EDL), and 2-ethylidene-4- methyl-1,3-dioxolane (EMD) were prepared from the corresponding cyclic vinyl acetals by catalytic isomerization of the double bond. The reaction of CKAs with different alcohols and diols was examined using trace of p-toluenesulfonic acid as a catalyst. For the monohydroxyl alcohols, cyclic ortho esters were formed by simple addition of the hydroxyl group toward CKAs with ethanol showing a much greater reactivity than iso-propanol. When 1,2- or 1,3-diols were used to react with the CKAs, we observed the isomerized cyclic ortho esters besides the simple addition products. Biocompatible polyols, that is, poly(2-hydroxyethyl acrylate) (PHEA) and poly(vinyl alcohol) (PVA) were then modified with CKAs, and the degree of substitution of the pendent ortho esters can be easily tuned by changing feed ratio. Both the small molecule ortho esters and the CKA-modified polymers demonstrate the pH-dependent hydrolysis profiles, which depend also on the chemical structure of the ortho esters as well as the polymer hydrophobicity.

■

INTRODUCTION pH-Sensitive polymers and the relevant materials have been extensively studied regarding their applications in the fields of smart drug/gene delivery and diagnosis because of the numerous pH gradients in human body.1−3 In general, pHsensitive polymers can be categorized into two groups: those with ionizable weak acid or base and those with acid-cleavable linkage.4 The polymers of the later category are usually neutral and may have better biocompatibility. Various acid-labile linkers, such as hydrazone, imine, cis-aconitic amide, acetal/ketal, ortho ester, and so on, have been used to build up the versatile pHsensitive polymers, polymer−drug conjugates, hydrogels, and micro/nanoparticles.5−9 Of these, the acid-degradable polymers or nanoparticles containing ketal/acetal linkages in the backbone or as the pendent groups have attracted great interest partly due to their tunable hydrolysis profile and the neutral degradation products, which is very important for curing the acute inflammatory diseases.10−23 Recently, acetalderivatized dextran or cyclodextrin were synthesized by the reaction of dextran or cyclodextrin with 2-methoxypropene, providing a facile approach to prepare acid-labile polymers with pendent ketal/acetals.24,25 Ortho ester is one of the most acid-labile functionalities available. Generally speaking, ortho ester is more susceptible to a weakly acidic environment than ketal or acetal with a similar structure.26 Poly(ortho ester)s, an important family of the aciddegradable polymeric biomaterials, have been extensively investigated over the past 40 years because of their unique © 2011 American Chemical Society

characteristics including the good biocompatibility, pHdependent surface degradation, and reproducible production.27,28 Ortho esters have also been applied to prepare pH-sensitive lipids or acid-degradable polymers that showed potential for drug/gene delivery.29−34 While most of the aforementioned poly(ortho ester)s have ortho ester units in the backbone, in recent years, we and others reported a family of acid-labile polymers with pendent cyclic ortho esters.35−39 Although these polymers exhibit promising properties as potential drug carriers, they have to be prepared by (co)polymerization of the individual monomers which are synthesized through multistep reactions with relatively low yields. Ketene acetals (KAs) represent a well-known family of compounds which have been used as reagents for organic synthesis or as monomers for polymerization.40−42 KAs have extraordinary reactivity with compounds containing active hydrogen including alcohols, however, the structural effect of the alcohols on their reactivity toward KAs has not been studied in detail.43 On this basis, we report a facile method to synthesize acid-labile polymers with pendent cyclic ortho esters by modifying multihydroxyl polymers (polyols) with cyclic ketene acetals (CKAs) (Scheme 1). Various synthetic or natural biocompatible polyols with primary or secondary hydroxyl group, such poly(vinyl alcohol), poly(2-hydroxyethyl acrylate), dextran, and Received: October 10, 2011 Revised: December 11, 2011 Published: December 16, 2011 173

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179

Biomacromolecules

Article

For the reaction of CKAs with 1,2- or 1,3-diols, the diol (1.05 mmol) and TsOH were dissolved in THF first, followed by the addition of CKA (1.0 mmol) under magnetic stirring. The reaction mixture was stirred at 30 °C for 15 min. After removal of the evaporated organic solvents at reduced pressure, ∼10 μL of the mixed ortho esters was dissolved in 0.5 mL of CDCl3 and analyzed by 1H NMR spectroscopy. Then, 50 μL of deuterated hydrogen chloride (DCl) in D2O (20%) was charged into the NMR tube, which was shaken for 30 s and incubated for 5 min at ambient temperature to completely hydrolyze the ortho esters. The tube was directly used for 1H NMR analysis. Our preliminary experiments showed that more than 95% of the diol monoesters remained in the CDCl3 phase while most part of the diols were extracted into the D2O phase. Acid-Triggered Hydrolysis of the Ortho Esters. To characterize the structure of the hydrolysis products, ∼10 μL of cyclic ortho ester was dissolved in 0.5 mL of D2O in a NMR tube, to which 20 μL of DCl in D2O (20%) was added. After being shaken for 30 s and incubated at ambient temperature for an additional 5 min, the tube was analyzed by 1H NMR spectroscopy. In the case of hydrolysis kinetics measurement, 10 mg of the ortho ester was dissolved in 0.5 mL of deuterated phosphate buffer (PB, 1.0 mM, pD 8.4), and the measured 1H NMR spectrum was used as that of the 0 time point. After adding 10 μL of 0.5 M deuterated buffer (pD: 5.4, 6.5 and 7.5), the solution was quickly mixed and the 1H NMR spectra were recorded at specific time points. For pD 8.4, the hydrolysis experiment was carried out in 1.0 mM deuterated phosphate buffer. Modification of Polyols with CKAs. Two polyols, poly(2hydroxyethyl acrylate) (PHEA) and poly(vinyl alcohol) (PVA), were modified with CKAs, and the reaction was carried out in anhydrous DMSO at 30 °C for 1 h. Take the reaction of PHEA with 2-ethylidene1,3- dioxolane (EDL) as an example. PHEA (1.0 g) was azeotropically dried with toluene prior to use and thoroughly dissolved in 10 mL of anhydrous DMSO in a flame-dried flask. After sequential addition of TsOH (1.0 mg in 0.1 mL THF) and 0.52 g EDL (5.2 mmol in 1.5 mL THF), the mixture was stirred at 30 °C for ∼30 min followed by adding one drop of TEA to quench the reaction. Then, the reaction mixture was dialyzed against water containing 0.1% TEA for 5 h at ambient temperature (MWCO: 3500) and lyophilized. The crude viscous product was dissolved in 10 mL of CH2Cl2 and precipitated from a mixed solvent of diethyl ether and hexane containing 1% TEA. The precipitate was dried in vacuum, yielding 1.0 g of the purified polymer (EDL-PHEA-1). Other CKA-modified PHEAs or PVAs were prepared by the similar procedure with yields of 60−80% and characterized by 1H NMR spectroscopy. pH-Dependent Hydrolysis of the CKA-Modified PHEAs. Hydrolysis of the CKA- modified PHEAs was monitored by 1H NMR spectroscopy in a mixed solvent of deuterated acetone and D2O (1:4, v/v) at 26 °C. Briefly, the polymer (80 mg) dissolved in deuterated acetone (0.5 mL) was added dropwise into 2.0 mL of deuterated PB (1.0 mM, pH 8.5) under stirring. Then, 0.6 mL of the formed emulsion was charged into a NMR tube, and the 1H NMR spectrum was measured, which was used as that at the 0 time point. After addition of 10 μL of 0.5 M deuterated buffer (pD: 5.4 and 7.5), the solution was quickly mixed and the 1H NMR spectra were recorded at specific time points. For the hydrolysis at pD 8.4, the experiment was carried out in 1.0 mM deuterated PB.

Scheme 1. Modification of Polyols by CKA

so on, can be used as the backbones. Because CKA reacts quickly with the hydroxyl group under gentle conditions, a large number of ortho ester based bioerodable polymers with a wide range of properties can thus be easily prepared in a highthroughput way. In addition, the hydrolysis products of the acid-labile polymers will be nontoxic and biometabolized if appropriate CKAs are used for the modification.

■

EXPERIMENTAL SECTION

Materials. p-Toluenesulfonic acid (TsOH), D2O, DCl, D3PO4 (85 wt % in D2O), NaOD (40 wt.% in D2O), DMSO-d6, acetone-d6 (J&K Chemical Ltd.), 2,2-dimethyl-1,3- propanediol, 1,3-propanediol, 2,4pentanediol, and tris(triphenylphosphine) ruthenium(II) dichloride ((Ph3P)3RuCl2, Alfa Aesar) were used as received. CDCl3 (J&K Chemical Ltd.) was treated with anhydrous Na2CO3 prior to use. Acrolein, ethylene glycol, 1,2-propanediol, ethanol, iso-propanol, triethyl amine (TEA), and other solvents were purchased from Beijing Chemical Reagent Company. Regarding the synthesis and characterization of CKAs and ortho esters, all the solvents were distilled under sodium or calcium hydroxide prior to use. Stock solution of TsOH in anhydrous THF (10 mg/mL) was prepared for convenient use. The deuterated buffers (pD 5.4, 6.5, 7.5, and 8.5) were prepared using D2O, D3PO4, and NaOD and calibrated by pH meter. Synthesis of Cyclic Ketene Acetals. Acrolein (0.5 mol), diol (0.5 mol), toluene (250 mL), and TsOH (40 mg) were placed in a 500 mL flask equipped with a magnetic stirrer, Dean−Stark trap, and condenser. A total of 10 mL of saturated NaCl solution with NaCl crystal (∼2 g) was placed in the trap to avoid the acrolein dissolved in the aqueous phase. The mixture was refluxed in an oil bath (120 °C) for 6 h until about 9 mL of water was collected in the trap. After cooling, the solution was removed by rotary evaporation and 200 mL of CH2Cl2 was added. The solution was washed with 30 mL of 1% Na2CO3 thrice and with 30 mL of saturated NaCl solution, respectively. Then, the separated organic phase was dried over anhydrous K2CO3, and distilled at reduced pressure to afford vinyl acetal. The yields of various vinyl acetals were in the range of 40−65%. For the catalytic isomerization, vinyl acetal (0.2 mol) and (Ph3P)3RuCl2 (50 mg) were charged into a 50 mL flask fitted with a magnetic stirrer, argon inlet, and refluxing condenser. The reaction mixture was bubbled with argon gas for 10 min and heated to and kept at 125 °C in an oil bath for 5 h until the isomerization finished, as monitored by 1H NMR spectroscopy. Then, the mixture was distilled to give the ketene acetal under reduced pressure (∼10 mmHg). The isomerization yield of various CKAs were in the range of 70−90%. All the CKAs and their precursors (vinyl acetals) were characterized by 1 H NMR spectroscopy (Figure S1). Reaction of CKAs with Alcohols or Diols. CKA (1.0 mmol) and quantitative alcohol were thoroughly dissolved in 1.0 mL of anhydrous THF (in a 5 mL flask) under magnetically stirring. After adding 10 μL of p-toluenesulfonic acid (TsOH) in THF (10 mg/mL), the solution was stirred in an oil bath (30 °C) for the designed time, and the reaction was quenched by adding one drop of triethylamine (TEA). After removal of the organic solvent at reduced pressure by rotary evaporation, ∼10 μL of the residue was taken for 1H NMR analysis (in 0.5 mL of CDCl3).

■

RESULTS AND DISCUSSION Although different methods have been used to prepare ketene acetals, the isomerization of the double bond of the α,β-unsaturated aldehyde-derived acetals in the presence of (Ph3P)3RuCl2 developed by Crivello is one of the simple approaches.44 We first synthesized three cyclic ketene acetals (CKAs) following Crivello’s method. They are different in the number of cyclic atoms (EDO vs EDL) or the substituted groups (EDL vs EMD) (Scheme 2). The structures of these CKAs were confirmed by the 1 H NMR spectra (Figure S1). 174

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179

Biomacromolecules

Article

isomerized ortho ester A1 (Scheme 4). The 1H NMR spectrum revealed that both A1 and A′1 were formed, however, it was

Scheme 2. Synthetic Route of 2-Ethylidene-1,3-dioxane (EDO), 2-Ethylidene-1,3-dioxolane (EDL), and 2Ethylidene-4-methyl-1,3-dioxolane (EMD)a

Scheme 4a

a Reagents and conditions: (i) acrolein, TsOH, toluene, 120 °C, 5 h; (ii) (Ph3P)3RuCl2, 125 °C, 4 h.

Addition Reaction of CKAs with Different Monoalcohols. Ketene acetals (KA) have extraordinary reactivity with compounds containing active hydrogen, such as H2O and alcohols, due to the presence of the strong nucleophilic center in the KA molecule. KAs react readily with primary or secondary alcohols to form ortho esters, while they are inert to the tertiary alcohols.40,43 However, the effect of the structure of the alcohols on their reactivity toward KAs has not been studied in detail. Here, we investigated the reactions between EDO and two alcohols (ethanol and iso-propanol), and the results are summarized in Scheme 3. EDO reacted rapidly with ethanol

a

Reaction conditions: (i) TsOH, THF, r.t., 30 min; molar ratio of CKA to the diol: 1.0/1.05. (ii) CDCl3/20% DCl in D2O (10:1, v/v), r.t., 5 min.

difficult to directly determine their ratio because of the poorly resolved spectrum (Figure 1A). According to literature and our

Scheme 3. Schematic Illustration of the Reactions between CKA or Cyclic Ortho Ester with Monohydroxyl Alcohola

a

Concentration of TsOH: 0.1 mg/mL; solvent: THF.

(EtOH) or iso-propanol (i-PrOH) at 30 °C in THF, and the reactions were completed in 20 min. When a mixture of ethanol and iso-propanol reacted with EDO in stoichiometry (1:1:1 in molar ratio), the main product was 2-ethoxy-2-ethyl-1,3dioxane (Ia) formed from ethanol, with little 2-ethyl-2-isopropoxy-1,3- dioxane (Ib) detected (Figure S2). This indicates that the less sterically hindered primary alcohol is much more active than the secondary alcohol. Once the cyclic ortho ester (Ib) was formed, however, the iso-propoxy group was not easily replaced by an ethoxy group. We also examined the effects of temperature (15−40 °C in THF) and solvent (CHCl3 and DMSO at 30 °C), and the reaction between EDO and ethanol can be completed in 30 min. Similar results were obtained for the other two CKAs (EDL and EMD). The 1H NMR spectra of the obtained cyclic ortho esters are shown in Figures S3−S5. Addition Reaction of CKAs with 1,2- or 1,3-Diols. Some polyols, such as poly(vinyl alcohol) (PVA) and dextran, have 1,3- or 1,2-diol units. When CKAs are used to modify these polyols, the isomerized structures may be formed besides the simple addition product of hydroxyl group to CKA. To demonstrate this issue, EDO was first reacted with 2,4-pentanediol, a model compound of 1,3-diol. There are two possible cyclic ortho esters resulting from the reaction of EDO and 2,4-pentanediol; one is the simple addition product A′1, and the other is the

Figure 1. 1H NMR spectra of the cyclic ortho esters prepared by the reaction of EDO with 2,4-pentandiol (A), and the in situ acidhydrolyzed products (B) in CDCl3. Peak intensities of proton j′ (Ij′ ∼ 5.1 ppm) and proton b′ (Ib′ ∼ 2.3 ppm) are used to calculate the molar ratio of the two monoesters: A1% = (2Ij′/Ib) × 100% = 65%. *Denotes the proton signals coming from 1,3-propanediol monopropionate, the hydrolyzed product of EDO. EDO rapidly reacts with trace of water in the solvent.

results, when 2-alkyloxy-1,3-dioxalane or 2-alkyloxy-1,3-dioxane are hydrolyzed in a mildly acidic solution, cleavage of the exocyclic alkyloxy takes place preferentially, resulting in the formation of glycol monoester or 1,3-propanediol monoester, but with little formation of the corresponding diols (Figures S3−S5).27,37,45,46 Thus, we can reasonably assume that the hydrolysis of cyclic ortho ester A1 produces only 2,4-pentanediol monoester (H-A1), while 1,3-propanediol monoester (H-A′1) comes from the ortho ester A′1 (Scheme 4). In other words, the ratio of the two cyclic ortho esters (A1 and A′1) can be obtained by measuring the molar ratio of the two monoesters. 175

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179

Biomacromolecules

Article

Table 1. Dominating Cyclic Ortho Esters by Reaction of the CKAs with Different 1,3- or 1,2-Diolsa

a

The value represents the percent molar content of the dominating ortho ester.

Scheme 5. Reaction of the CKAs with Different 1,2- or 1,3-Diols

As shown in Figure 1B, after in situ acid-triggered hydrolysis of the cyclic ortho esters, the ratio of the two monoesters can be determined by comparing the intensities of peaks at ∼5.1 and ∼2.3 ppm; and accordingly, the molar ratio of A1 to A′1 was 0.65:0.35, in favor of the isomerized ortho ester A1. We further used a variety of 1,3- or 1,2-diols to react with the CKAs following a similar condition and analyzed the formed cyclic ortho esters by in situ hydrolysis method (Figures S6−S9). The preferred cyclic ortho ester and its percent molar content in the mixture of the two competing cyclic ortho esters for each of the reactions are summarized in Table 1. When EDO was used as the CKA, the isomerized cyclic ortho esters dominated, and the main products were the ortho esters with substituted 1,3dioxane or 1,3-dioxalane structures. By contrast, in the case of EMD, the predominant products were the cyclic ortho esters with 4-methyl-1,3-dioxalane ring, which were produced by simple addition of the diols to EMD. Interestingly, the dominating ortho esters by reactions of EDO with 1,2-propanediol and EMD with 1,3-propanediol were identical (A3 vs C2), having almost the same percentage. For the other two pairs of reactions, EDO with glycol versus EDL with 1,3-propanediol (A4 vs B2) or EDL with 1,2propanediol versus EMD with glycol (B3 vs C3), a similar trend was obtained. Furthermore, when the reaction was carried out in CDCl3 as in situ monitored by 1H NMR spectroscopy, the similar phenomenon was observed (data not shown). These results demonstrate that the proportion of the two competing cyclic ortho esters is thermodynamically controlled, mainly depending on their stability (Scheme 5). pH-Dependent Hydrolysis of Various Cyclic Ortho Esters. Three pairs of ortho esters, which are different in the number of the cyclic atoms or in the substituent on the rings, were studied to demonstrate their pH-dependent hydrolysis behaviors (Table 2). The experiments were carried out in the deuterated aqueous buffer, as monitored by 1H NMR

Table 2. Half-Life Time (t1/2, min) of the pH-Dependent Hydrolysis of Various Cyclic Ortho Estersa pH ortho ester

temperature (°C)

5.4

6.5

7.5

8.5

Ia IIa IIIa IIIa Ib IIb IIIb

15 15 15 23 23 23 23

∼1 ∼2

∼2 ∼3

14 20 36 15 ∼8 12 17

50 120 ∼1900b ∼720b 28 80 ∼560b

a

Nomenclature: Ia, 2-ethoxy-2-ethyl-1,3-dioxane; IIa, 2-ethoxy-2ethyl-1,3-dioxalane; IIIa, 2-ethoxy- 2-ethyl-4-methyl-1,3-dioxalane; Ib, 2-ethyl-2-iso-propoxy-1,3-dioxane; IIb, 2-ethyl-2-iso-propoxy- 1,3dioxalane; IIIb, 2-ethyl-2-iso-propoxy-4-methyl-1,3-dioxalane. bEstimated by extrapolating the kinetic curves.

spectroscopy. Figure 2 shows the representative time-dependent 1 H NMR spectra of ortho ester Ia hydrolyzed at pH 7.5. The hydrolysis degree was estimated by monitoring the change in integration intensity of peaks at ∼2.3 ppm (proton b′) using the peak at ∼4.8 ppm (HDO) as a standard. Compound Ia is very sensitive to the mildly acidic environment, showing a half-life time of 14 min even though in the neutral medium. By the same method, hydrolysis kinetics of the other ortho esters was also evaluated (Figures S10−S13), and the half-life times at different pHs are summarized in Table 2. As expected, all of these ortho esters were hydrolyzed faster with the decrease of pH. When the exocyclic alkyloxy group was identical, at the same pH and temperature, hydrolysis rate decreased in the order of Ia > IIa > IIIa, and Ib > IIb > IIIb. For acid-triggered hydrolysis of ortho esters, it is generally accepted that there are three reaction stages.46 The first stage, that is, generation of the dialkoxycarbocation intermediate, which consists of the probable concerted proton transfer and 176

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179

Biomacromolecules

Article

Figure 2. 1H NMR (300 MHz) spectra of ortho ester Ia (20 mg/mL) at different hydrolysis times at 15 °C in deuterated PB (10 mM, pH 7.5). The relative amount of the remaining ortho ester (Ia) is calculated by the formula: 1 − (It − I0)/(I∞ − I0), where I0, It, and I∞, denote the peak intensity of proton b′ (∼2.4 ppm) at 0 min (zero), t min, and of the completely hydrolyzed sample, respectively. The spectrum at 0 min is measured in the buffer of pH 8.5, and # denotes the spectrum of the completely hydrolyzed sample. The proton peak of HDO (∼4.8 ppm) is used as the internal standard, and * denotes the proton signals of impurities.

C−O bond cleavage,47,48 is the rate-determining step in the mildly acidic or neutral media.49 The faster hydrolysis rate of the six-membered cyclic ortho ester (Ia or Ib) as compared to the five-membered counterpart (IIa or IIb) can be explained by the well-known anomeric (i.e., stereoelectronic) effect. For Ia or Ib, the molecule adopts an almost ideal chair conformation, which makes the exocyclic oxygen highly basic and is helpful for C−O breaking and leaving of the protonated exocyclic alkoxy group.26,50 Regarding the hydrolysis of ortho esters with fiveand six-membered rings, Jones et al. reported an opposite trend: the five-membered ortho ester hydrolyzed faster than the six-membered one.51 This discrepancy might be due to the difference in substituents on the rings. In Jones’ molecules, the amide group might affect conformation of the ortho esters via intramolecular hydrogen bonding, which would influence the hydrolysis. Modification of PHEA and PVA by CKAs. PHEA and PVA were used as the polyols not only for their good biocompatibility, but also for their different hydroxyl groups. PHEA has the primary hydroxyl groups while PVA possesses the secondary hydroxyl groups in the form of the 1,3-diol structure. PHEA was first modified in anhydrous DMSO by EDL with various feed ratios. As shown in Figure 3, the degree of substitution (DS) of the ortho ester can be simply tuned by changing the feed ratio of EDL to hydroxyl group. However, DS was generally lower than the corresponding feed ratio, which is mainly attributed to the partial hydrolysis of the ortho ester during the postreaction purification procedure. By increasing the feed ratio of EDL to hydroxyl group and adopting a modified purification protocol, DS can be increased up to 95% (Figure S16). Following a similar procedure, PHEA was also modified by EDO and EMD (Figure S17). To further clarify the polymer structure, EDO-modified PHEA was completely hydrolyzed in acidic D2O. Only the proton signals

Figure 3. 1H NMR (300 MHz) spectra of PHEA in D2O and the EDL-modified PHEAs in CDCl3. Molar feed ratios of EDL to hydroxyl group of PHEA are 0.6 (EDL-PHEA-1), 0.8 (EDL-PHEA-2), and 1.0 (EDL-PHEA-3), respectively. The degrees of substitution (DS) in the final polymers are estimated by comparing the intensity of peak d (Id) and peak f + f′ (If+f′), being 44, 58, and 74%, respectively (DS = Id/If+f′).

of PHEA and 1,3-propanediol monopropionate were detected, indicating that PHEA reacted with EDO through the simple addition of a hydroxyl group to the carbon−carbon double bond as ethanol did (Figure S18). 177

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179

Biomacromolecules

Article

Figure 4. Hydrolysis kinetics of the CKA-modified PHEAs.

neutral medium. By contrast, the other two polymers, EDOPHEA and EDL-PHEA, showed a significant hydrolysis, even at pH 8.5. This demonstrates that EMD may be a better candidate than the other two CKAs for the preparation of ortho estercontaining polymers that are stable enough in weakly basic media but sensitive enough to mildly acidic environments.

By contrast, when PVA was modified by EMD with a feed ratio of 1:1 in anhydrous DMSO, the isomerized structure of a six-membered cyclic ortho ester on the polymer backbone was observed, besides the pendent five-membered cyclic ortho ester, which was formed by the simple addition of a hydroxyl group to EMD (Figure S19). These results are very consistent with those for the model reactions of CKAs with the corresponding small monohydroxyl alcohol and 1,3-diols. Cytotoxicity of two EMD-modified polymers, EMD-PVA and EMD-PHEA, was evaluated by a MTT assay in Hela cells, using poly(ethylene glycol) methyl ether (mPEG 5k) and branched poly(ethylenimine) (PEI 25k) as negative and positive controls, respectively (Figure S20). Compared to PEI 25k, the two acid-labile polymers showed much lower cytotoxocity. However, the cell viabilities for both EMDPVA and EMD-PHEA were lower than that for mPEG 5k at high polymer concentration, which might be attributed to their hydrophobic character. Hydrolysis of CKA-Modified PHEAs. Three CKA-modified PHEAs with the similar degree of substitution (∼44%, Figure S17), that is, EDO-PHEA, EDL-PHEA, and EMD-PHEA, were studied on their pH-dependent hydrolysis in a mixed solvent of deuterated phosphate buffer and acetone (Figure 4). As expected, all the polymers showed significant pH-dependent hydrolysis behaviors: the lower the pH, the faster the hydrolysis rate. At pH 8.5, the hydrolysis rate followed the order of EDO-PHEA > EDLPHEA > EMD-PHEA, which is consistent with the results of their corresponding small model compounds (Table 2). However, at the same pH, the polymers hydrolyzed much slower than their small model compounds. This can be attributed to the hydrophobic aggregation of the polymers in aqueous buffers.16,52,53 Interestingly, EMD-PHEA was quite stable at pH 8.5 but showed a moderate hydrolysis rate in the

■

CONCLUSION

■

ASSOCIATED CONTENT

The reaction of CKAs with different alcohols and the hydrolysis of their ortho ester products were studied in detail. Less sterically hindered primary alcohol is much more active than the secondary one. When 1,2- or 1,3-diols react with CKAs, the isomerized cyclic ortho esters are formed; the ratio of the competing ortho esters is mainly determined by their stability. Hydrolysis of the cyclic ortho esters is pH-dependent and greatly affected by their structure. Facile synthesis of the acidlabile polymers with pendent cyclic ortho esters has been achieved by modifying the biocompatible polyols with CKAs. The chemical structure and hydrolysis kinetics of these acidlabile polymers can be easily manipulated by using different polyols and CKAs or by changing the feed ratio. The mild and efficient reaction condition, structural diversity of the polyols and CKAs, and ease of obtaining the starting materials make the present approach very potential for the preparation of biocompatible and acid-sensitive polymeric drug carriers.

S Supporting Information *

NMR spectra of the CKAs, ortho esters, polymers, and their hydrolysis kinetics; results of cell viability evaluation; and more experimental details. This material is available free of charge via the Internet at http://pubs.acs.org. 178

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179

Biomacromolecules

■

Article

(29) Masson, C.; Garinot, M.; Mignet, N.; Wetzer, B.; Mailhe, P.; Scherman, D.; Bessodes, M. J. Controlled Release 2004, 99, 423−434. (30) By, K.; Nantz, M. H. Angew. Chem., Int. Ed. 2004, 43, 1117− 1120. (31) Wang, C.; Ge, Q.; Ting, D.; Nguyen, D.; Shen, H. R.; Chen, J. Z.; Eisen, H. N.; Heller, J.; Langer, R.; Putnam, D. Nat. Mater. 2004, 3, 190−196. (32) Huang, Z. H.; Guo, X.; Li, W. J.; MacKay, J. A.; Szoka, F. C. J. Am. Chem. Soc. 2006, 128, 60−61. (33) Tang, R. P.; Palumbo, R. N.; Ji, W. H.; Wang, C. Biomacromolecules 2009, 10, 722−727. (34) Yuan, Z. F.; Huang, J. Y.; Liu, J.; Cheng, S. X.; Zhuo, R. X.; Li, F. Nanotechnology 2011, 22, 335601. (35) Huang, X. N.; Du, F. S.; Ju, R.; Li, Z. C. Macromol. Rapid Commun. 2007, 28, 597−603. (36) Huang, X. N.; Du, F. S.; Cheng, J.; Dong, Y. Q.; Liang, D. H.; Ji, S. P.; Lin, S. S.; Li, Z. C. Macromolecules 2009, 42, 783−790. (37) Qiao, Z. Y.; Du, F. S.; Zhang, R.; Liang, D. H.; Li, Z. C. Macromolecules 2010, 43, 6485−6494. (38) Tang, R. P.; Ji, W. H.; Wang, C. Macromol. Biosci. 2010, 10, 192−201. (39) Tang, R. P.; Ji, W. H.; Panus, D.; Palumbo, R. N.; Wang, C. J. Controlled Release 2011, 151, 18−27. (40) McElvain, S. M. Chem. Rev. 1949, 45, 453−492. (41) Cao, L. W.; Wu, Z. H.; Pittman, C. U. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 2841−2852. (42) Undin, J.; Plikk, P.; Finne-Wistrand, A.; Albertsson, A. C. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 4965−4973. (43) McElvain, S. M.; Curry, M. J. J. Am. Chem. Soc. 1948, 70, 3781− 3786. (44) Crivello, J. V.; Malik, R.; Lai, Y. L. J. Polym. Sci., Part A: Polym. Chem. 1996, 34, 3091−3102. (45) Chiang, Y.; Kresge, A. J.; Salomaa, P.; Young, C. I. J. Am. Chem. Soc. 1974, 96, 4494−4499. (46) Ahmad, M.; Bergstrom, R. G.; Cashen, M. J.; Chiang, Y.; Kresge, A. J.; McClelland, R. A.; Powell, M. F. J. Am. Chem. Soc. 1979, 101, 2669−2677. (47) Bunton, C. A.; De Wolfe, R. H. J. Org. Chem. 1965, 30, 1371− 1375. (48) Bergstrom, R. G.; Cashen, M. J.; Chiang, Y.; Kresge, A. J. J. Org. Chem. 1979, 44, 1639−1642. (49) Chiang, Y.; Kresge, A. J.; Lahti, M. O.; Weeks, D. P. J. Am. Chem. Soc. 1983, 105, 6852−6855. (50) Li, S. G.; Dory, Y. L.; Deslongchamps, P. Tetrahedron 1996, 52, 14841−14854. (51) Bruyere, H.; Westwell, A. D.; Jones, A. T. Bioorg. Med. Chem. Lett. 2010, 20, 2200−2203. (52) Pitt, C. G.; Shah, S. S. J. Controlled Release 1995, 33, 397−403. (53) Mahmoud, E. A.; Sankaranarayanan, J.; Morachis, J. M.; Kim, G.; Almutairi, A. Bioconjugate Chem. 2011, 22, 1416−1421.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (Z.L.); [email protected] (F.D.).

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (21174002, 50973002, and 21090351). We thank Prof. Wenlin Huang (Institute of Microbiology, CAS) for his kind help regarding cytotoxicity evaluation.

■

REFERENCES

(1) Lee, E. S.; Gao, Z. G.; Bae, Y. H. J. Controlled Release 2008, 132, 164−170. (2) Ganta, S.; Devalapally, H.; Shahiwala, A.; Amiji, M. J. Controlled Release 2008, 126, 187−204. (3) Park, I. K.; Singha, K.; Arote, R. B.; Choi, Y. J.; Kim, W. J.; Cho, C. S. Macromol. Rapid Commun. 2010, 31, 1122−1133. (4) Gillies, E. R.; Frechet, J. M. J. Pure Appl. Chem. 2004, 76, 1295− 1307. (5) Oishi, M.; Sasaki, S.; Nagasaki, Y.; Kataoka, K. Biomacromolecules 2003, 4, 1426−1432. (6) Gu, J. X.; Cheng, W. P.; Liu, J. G.; Lo, S. Y.; Smith, D.; Qu, X. Z.; Yang, Z. Z. Biomacromolecules 2008, 9, 255−262. (7) Gao, W. W.; Chan, J. M.; Farokhzad, O. C. Mol. Pharm. 2010, 7, 1913−1920. (8) Zhang, Y. L.; Tao, L.; Li, S. X.; Wei, Y. Biomacromolecules 2011, 12, 2894−2901. (9) Jin, Y.; Song, L.; Su, Y.; Zhu, L. J.; Pang, Y.; Qiu, F.; Tong, G. S.; Yan, D. Y.; Zhu, B. S.; Zhu, X. Y. Biomacromolecules 2011, 12, 3460− 3468. (10) Basko, M.; Kubisa, P. Macromolecules 2002, 35, 8948−8953. (11) Tomlinson, R.; Heller, J.; Brocchini, S.; Duncan, R. Bioconjugate Chem. 2003, 14, 1096−1106. (12) Wu, W. T.; Pang, W. M.; Wang, Y. S.; Zhu, Q. R.; Lu, F.; Xu, G. Y. Polymer 2005, 46, 3132−3140. (13) Bulmus, V.; Chan, Y.; Nguyen, Q.; Tran, H. L. Macromol. Biosci. 2007, 7, 446−455. (14) Chan, Y.; Wong, T.; Byrne, F.; Kavallaris, M.; Bulmus, V. Biomacromolecules 2008, 9, 1826−1836. (15) Paramonov, S. E.; Bachelder, E. M.; Beaudette, T. T.; Standley, S. M.; Lee, C. C.; Dashe, J.; Frechet, J. M. J. Bioconjugate Chem. 2008, 19, 911−919. (16) Yang, S. C.; Bhide, M.; Crispe, I. N.; Pierce, R. H.; Murthy, N. Bioconjugate Chem. 2008, 19, 1164−1169. (17) Zhang, L.; Bernard, J.; Davis, T. P.; Barner-Kowollik, C.; Stenzel, M. H. Macromol. Rapid Commun. 2008, 29, 123−129. (18) Tang, X. Z.; Pan, C. Y. J. Biomed. Mater. Res. 2008, 86A, 428− 438. (19) Falco, E. E.; Patel, M.; Fisher, J. P. Pharm. Res. 2008, 25, 2348− 2356. (20) Chen, W.; Meng, F. H.; Li, F.; Ji, S. J.; Zhong, Z. Y. Biomacromolecules 2009, 10, 1727−1735. (21) Chen, W.; Meng, F. H.; Cheng, R.; Zhong, Z. Y. J. Controlled Release 2010, 142, 40−46. (22) Fiore, V. F.; Lofton, M. C.; Roser-Page, S.; Yang, S. C.; Roman, J.; Murthy, N.; Barker, T. H. Biomaterials 2010, 31, 810−817. (23) Chen, D. Y.; Li, N. J.; Gu, H. W.; Xia, X. W.; Xu, Q. F.; Ge, J. F.; Lu, J. M.; Li, Y. G. Chem. Commun. 2010, 46, 6708−6710. (24) Bachelder, E. M.; Beaudette, T. T.; Broaders, K. E.; Dashe, J.; Frechet, J. M. J. J. Am. Chem. Soc. 2008, 130, 10494−10495. (25) Zhang, J. X.; Jia, Y.; Li, X. D.; Hu, Y. Q.; Li, X. H. Adv. Mater. 2011, 23, 3035−3040. (26) Deslongchamps, P.; Dory, Y. L.; Li, S. G. Tetrahedron 2000, 56, 3533−3537. (27) Heller, J.; Barr, J.; Ng, S. Y.; Abdellauoi, K. S.; Gurny, R. Adv. Drug Delivery Rev. 2002, 54, 1015−1039. (28) Heller, J.; Barr, J. Biomacromolecules 2004, 5, 1625−1632. 179

dx.doi.org/10.1021/bm201410c | Biomacromolecules 2012, 13, 173−179