Investigation of the Degradation Mechanisms of Poly (malic acid

Maria Elisa Martinez Barbosa, Sandrine Cammas, Martine Appel, and Gilles ... Faculté de Pharmacie 5, rue J.B. Clément, 92290 Châtenay-Malabry, France...
1 downloads 0 Views 107KB Size
Biomacromolecules 2004, 5, 137-143

137

Investigation of the Degradation Mechanisms of Poly(malic acid) Esters in Vitro and Their Related Cytotoxicities on J774 Macrophages Maria Elisa Martinez Barbosa, Sandrine Cammas, Martine Appel, and Gilles Ponchel* UMR CNRS 8612, Faculte´ de Pharmacie 5, rue J.B. Cle´ ment, 92290 Chaˆ tenay-Malabry, France Received July 30, 2003; Revised Manuscript Received October 27, 2003

Poly(β-malic acid) hydrophobic derivatives are promising polymers for biomedical and pharmaceutical applications. The objectives of the present work were to study the in vitro degradation profile of three PMLA hydrophobic derivatives and to evaluate their cytotoxicity before and after degradation. For this purpose, nanoparticles from poly(benzyl-malate) (PMLABe), poly(hexyl-malate) (PMLAHe), and poly(malic acid-co-benzyl-malate) (PMLAH/He) were prepared for degradation studies on standardized materials. Size exclusion chromatography (SEC) and 1H NMR indicated that degradation occurred by random hydrolysis of the polymer main chain for all three polymer derivatives. The presence of carboxyl groups on the side chain and their esterification with different alcohols varying hydrophilicities could affect the degradation rate. It was postulated that the degradation depended on the rate of diffusion of water into the core of the particles. The cytotoxicity of the polymer nanospheres as well as their degradation products were evaluated in vitro with J774 A1 murine macrophage-like cell line. The cytotoxicity depended on the degradation rate of the polymers and the amount of degradation products of low molecular weight produced. 1. Introduction Developments of original systems in the biomedical and pharmaceutical domains are nowadays based on the use of biopolymers answering to very strict conditions of applications. In the pharmaceutical field, more and more complex requirements are simultaneously requested in a single delivery system, stimulating the modification of already existing synthetic or natural polymers or the development of new polymers.1 In this respect, poly(malic acid) and its hydrophobic derivatives can be considered as a family of promising candidates.2 Poly(malic acid) (PMLA) is a polyester synthesized for temporary therapeutic applications.3-8 This polymer (Figure 1) has shown biocompatibility4 and is degradable by simple hydrolysis of the ester bond of the main chain, leading to the production of malic acid which can be considered as a nontoxic molecule.9 Interestingly, as can be seen from Figure 1, the main chain bears carboxylic groups which can be substituted, giving the opportunity either to modulate the overall hydrophobicity of the polymer or to introduce various bioactive ligands which can be used for giving specific properties to the polymer. The aim of the present work was to investigate the degradability of a series of polymers belonging to the family of PMLA and, simultaneously, the cytotoxicity of these polymers and their degradation products. Three derivatives were selected, including poly(benzyl malate) (PMLABe), poly(hexyl malate) (PMLAHe), and poly(malic acid-cobenzyl malate) (PMLAH/He), which exhibited varying * To whom correspondence should be addressed. Phone: 33 1 46 83 59 19. Fax 33 1 46 61 93 34. E-mail: [email protected].

Figure 1. Structure of poly(malic acid) (PMLA) and derivates, poly(hexyl malate) (PMLAHe) and poly(benzyl malate) (PMLABe).

hydrophobicities, depending on different residues on the lateral chains. Because they are all water-insoluble polymers, nanoparticles were prepared from these polymers in order to perform degradation and cytotoxicities studies on reproducible materials exhibiting a standardized interface with different media in which degradation may occur. Polymer degradation kinetics were determined by size exclusion chromatography (SEC) and 1H NMR in different media, and attempts were made to correlate cytotoxicities of the nanoparticles on the J774 macrophage cell line to the amount of degradation products produced during the degradation process. 2. Materials and Methods 2.1. Materials. Poly(benzyl malate) (PMLABe), Mw ) 47 000 g/mol, Ip ) 1.3; poly(hexyl malate) (PMLAHe), Mw ) 16 000 g/mol, Ip ) 1.1; and poly(malic acid-co-benzyl malate) (PMLAH/He), Mw ) 20 900 g/mol, Ip ) 1.1, 10% mol of malic acid units were synthesized as described elsewhere.8 Potassium dihydrogen phosphate (KH2PO4) (Rectapur Prolabo), sodium azide (NaN3) (Sigma), sodium chloride (NaCl) (Sigma), and sodium hydroxide (NaOH) (Fluka) were used as received. Water was purified by reverse osmosis (MilliQ, Millipore). The solvents used were analytical grade and all other chemicals were commercially available reagent grade.

10.1021/bm0300608 CCC: $27.50 © 2004 American Chemical Society Published on Web 12/19/2003

138

Biomacromolecules, Vol. 5, No. 1, 2004

2.2. Nanoparticles Preparation and Characterization. Nanoparticles from PMLABe, PMLAHe, and PMLAH/He were prepared by nanoprecipitation as described elsewhere.10 Briefly, the polymer (25 mg for PMLABe or PMLAHe and 100 mg for PMLAH/He) was dissolved in 5 mL of acetone. This solution was added by dripping into 10 mL of water under magnetic stirring. The mixture was left under magnetic stirring for approximately 15 min. Further, the solvent was evaporated under reduced pressure at 35 °C (Rotavapor Bu¨chi 461 water bath RE 111). The mean diameter of the nanoparticles was determined after suitable dilution of bulk suspensions in milliQ water using dynamic laser light scattering (Nanosizer Coulter N4 Plus). 2.3. Degradation Study. Degradation studies were performed in aqueous pH 1.2 and pH 7.5 buffer solutions, using freshly prepared nanoparticles in order to avoid any artifact due to the possible degradation of the polymer in the suspension medium. Nanoparticles as well as degradation buffer solutions were prepared with milliQ water containing 0.05% w/w of NaN3 to avoid bacterial growth. The buffer solutions (pH 1.2; pH 7.5) were prepared according to the U.S. Pharmacopeia,11 without enzymes. At the beginning of the experiment, a series of vials containing 3 mL of nanoparticle suspensions were added to 5 mL of degradation buffers and incubated at 37 °C. The mean diameters of the nanoparticles contained in the corresponding vials were measured (Nanosizer) as a function of time, and the suspensions were lyophilized (Christ Alpha 1-4). The freeze-dried products were kept at -6 °C until their analysis by size exclusion chromatography (SEC) or by 1H NMR. SEC was used for the monitoring of the evolution of the molecular weight of the polymer during the degradation. Freeze-dried samples were dissolved in THF and then filtrated on a 0.2 µm filter. 50 or 100 µL were injected in a SEC apparatus equipped of an automated injector (Injector Waters 712 Wisp), a Viscotek Pump VE1121, a ViscoGEL column type GMHHR-M Mixed bed (7.8 mm × 30 cm, Viscotek), a column guard type HHR-H (6 mm × 4 cm), a differential refractometer (Viscotek T60A Dual Detector). 1 H NMR were performed in order to follow the changes in polymer structure during the degradation process. Solutions were prepared by dissolving freeze-dried samples in D6 acetone and filtrated on 0.2 µm. NMR spectra were obtained from a NMR Brucker Avance 400 apparatus, operating at 400 MHz. 2.4. Cytotoxicity Studies on J774 A1 Murine Macrophages. The J774 A1 murine macrophage-like cell line (ECACC catalog number 91051511) was maintained as an adherent culture in humidified atmosphere (5% CO2: 95% air) at 37 °C in complete cell medium consisting of RPMI 1640 medium with Glutamax-I (Gibco) supplemented with 10% (v/v) heat-inactivated fœtal calf serum (FCS;Gibco) added with 100 IU/mL penicillin and 100 IU/mL streptomycin (Gibco). In the experiments, cells were mechanically detached and counted on a Neubauer counting plate. Desired amounts of cells were adjusted to the required concentration (1 × 105 cell/mL) with complete cell medium and placed in 96-well plates at 200 µL per well, corresponding to 2 × 104

Martinez Barbosa et al.

cell per well. After 24 h, most of the cells were adherent, and the nonadherent cells were removed before adding the nanoparticles suspensions. Cytotoxicity studies were performed on freshly prepared nanoparticles suspensions and on nanoparticles suspensions incubated for 11, 21, and 31 days at pH 7.5. In this latter case, 30 mL of PMLABe, PMLAHe, and PMLAH/He nanoparticles suspensions were prepared as indicated in above. Then, 6 mL of each suspension were placed in 10 mL of pH 7.5 buffer and kept at 37 °C during 11, 21, and 31 days. Whatever the origin of the samples, the water contained in the nanoparticles suspensions was exchanged by complete cell medium before putting the suspensions in contact with the cells as follows. 10 mL of complete cell medium was added into 10 mL of the nanoparticle suspension, and water was evaporated under reduced pressure at 35 °C to obtain a final volume of 10 mL. In a first experiment, 2 × 104 cells per well were incubated for 1, 4, and 24 h with freshly prepared nanoparticles suspensions, previously filtered on a 0.45 µm sterile filter, at increasing concentrations ranging from 1 to 2500 µg of polymer/mL for the PMLABe and the PMLAHe, 1-10 000 µg of polymer/mL for the PMLAH/He, respectively. After the corresponding incubation time, the medium was removed and the cellular viability was estimated using the assay based on the tetrazolium salt MTT (3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide).12 During incubation, MTT is absorbed in the cells (dead and live) and the tetrazolium ring is cleaved by an enzymatic reaction, which is catalyzed by the dehydrogenase enzyme produced in active mitochondria. This reaction leads to a degradation product. The amount of this degradation product, which can be detected and quantified by spectrophotometry, is proportional to the number of living cells. Results were expressed as the concentration of the studied substance which is able to kill 50% of initial living cells (concentration necessary for Inhibiting 50% of living cells, IC50%). Experiments were performed in triplicate. Additionally, blank samples were obtained by following the nanoparticles preparation protocol without using polymer. These samples were treated and tested as described above in order to assess any possible effect of acetone traces on the cellular viability. No cytotoxic effects could be evidenced for all blank samples under examination. In a second experiment, the cytotoxicity of nanoparticle suspensions incubated in the pH 7.5 medium were determined. 2 × 104 cells per well were incubated for 1, 4, and 24 h, with filtered (0.45 µm sterile filter) degraded nanoparticles suspensions at increasing concentrations ranging from 1 to 940 µg of polymer/mL for the PMLABe and the PMLAHe, 1-2500 µg of polymer/mL for the PMLAH/He, respectively. After the preset incubation time, the medium was removed and the cellular viability was estimated using the MTT conversion test as described above. Experiments were performed in triplicate. Blank samples prepared as described above were similarly incubated in the degradation buffer. No cytotoxic effects could be seen on these blank samples.

Biomacromolecules, Vol. 5, No. 1, 2004 139

Degradation Mechanisms of PMLA Esters Table 1. Physicochemical Characteristics of PMLABe, PMLAHe, and PMLAH/He Polymers and the Corresponding Nanoparticles Prepared by Nanoprecipitation (n ) 3) polymer

Mwa PMLABe 47 000 PMLAHe 16 000 PMLA H/He 20 900

nanoparticles

polymean size polymolecularitya Tgb ( S.D.c dispersityc index (°C) (nm) index 1.3 1.1 1.05

37 -10 -11

96 ( 24 102 ( 32 163 ( 51

0.087 0.162 0.169

a Determined by SEC in THF; Polystyrene standard. b Measured by DSC. c Measured using dynamic laser light scattering (Nanosizer Coulter N4 plus).

Figure 2. Typical chromatograms obtained by size exclusion chromatography (SEC). Case of PMLAH/He at initial time (a) and after 3 weeks of incubation at pH 7.5 and 37 °C (b).

3. Results 3.1. Nanoparticles Preparation and Characterization. The nanoprecipitation method allowed for the easy and reproducable preparation of PMLABe, PMLAHe, and PMLAH/He nanoparticle suspensions in the absence of any surfactants. The formation of the particles results from the auto-aggregation of the preformed insoluble polymer chains in the water medium. As shown in Table 1, the nanoparticles obtained had a diameter typically smaller than 200 nm. These water suspensions were very stable when stocked at 4 °C, and their diameter was kept constant for 48 weeks (data not shown). However, the physicochemical stability of the suspension at 37 °C in the buffer solutions used for degradation studies depended on the pH of the suspension medium. A tendency to aggregation could be noticed in pH 1.2 buffer solution. The particles were more stable in pH 7.5 buffer solution. 3.2. Polymers Degradation Studies. 3.2.1. Polymer Degradation Studied by Size Exclusion Chromatography. Changes in the molecular weight of the polymers constituting the nanoparticles incubated in pH 1.2 and 7.5 buffer solutions were estimated by size exclusion chromatography (SEC) after increasing incubation times. As an example, Figure 2 shows the chromatograms obtained for PMLAH/He at initial time and after 3 weeks of incubation at pH 7.5. For all three types of polymers, the chromatograms showed an unimodal weight distribution at any degradation time. Figures 3 and 4 show the time course of the normalized molecular weight (ratio of the polymer molecular weight at time t to the initial polymer molecular weight at to) of the polymers when the nanoparticles were incubated in pH 1.2 and pH 7.5 media at 37 °C, respectively.

Figure 3. Normalized molecular weight (measured molecular weight/ initial molecular weight) of the polymers ([ PMLABe, 9 PMLAHe, and 2 PMLAH/He) recovered in the nanoparticles suspensions preserved at pH 1.2 and 37 °C as a function of time.

Figure 4. Normalized molecular weight (measured molecular weight/ initial molecular weight) of the polymers ([ PMLABe, 9 PMLAHe, and 2 PMLAH/He) recovered in the nanoparticles suspensions preserved at pH 7.5 and 37 °C as a function of time.

A regular decrease of the molecular weight was observed for the three polymers, at both pH values (Figures 3 and 4). The slowest decrease of the molecular weight was obtained for the PMLABe while the fastest was observed for PMLAH/ He. After 8 weeks of incubation, the Mw/Mw,initial ratio was still 0.25 for PMLABe, whereas it was almost zero for the two other polymers. Additionally, it can be seen that the profiles were very similar whatever the degradation medium, suggesting that pH was not an important factor in the degradation mechanism. 3.2.2. Polymer Degradation Studied by 1H NMR. Polymer degradation was also analyzed by 1H NMR to determine the molecular structure of the degradation products formed during the incubation of nanoparticles in pH 1.2 and 7.5 buffer solutions at 37 °C. Figure 5 shows the 1H NMR spectrum of PMLAHe as an example of the samples which were studied. For each polymer, spectra of initial polymers and those of degradation products obtained in both degradation media were very similar, suggesting either that no degradation occurred during time or that the molecular structure of the polymer was preserved during the degradation process and, thus, remained similar to that of the initial product. Indeed, no peaks corresponding to the production of hexanol or benzyl alcohol (which should have resulted from an hydrolysis of the lateral esters of the polymers, as shown in Figure 6) nor to the production of a large amount of the esters which should have resulted from successive hydrolysis of the terminal units, were observed. 3.2.3. Degradation of the Nanoparticles. Polymer degradation studies were undertaken using nanoparticles in order to standardize the surface of contact between the polymer and

140

Biomacromolecules, Vol. 5, No. 1, 2004

Martinez Barbosa et al.

Figure 5. 1H NMR spectra in D6 acetone of poly(hexyl malate) at initial time (a) and after incubation for 4 weeks in phosphate buffer pH 7.5 (b).

the degradation media. It was expected that the degradation of the polymers could lead progressively to the degradation of the nanoparticles. To assess the physical degradation of the particles, the size measurements were undertaken during degradation studies, but nanoparticles had a tendency to agglomerate, especially at pH 1.2, making impossible their mean diameter measurement. 3.3. Cytotoxicity of Nanoparticles and Degradation Products. An in vitro study of the cytotoxicity of the PMLABe, PMLAHe, and PMLAH/He nanoparticles and their degradation products (nanoparticles incubated in the pH ) 7.5 degradation buffer and kept at 37 °C during 11, 21, and 31 days) on J774 A1 murine macrophage-like cells were determined after 1, 4, and 24 h of contact with the cells. The cellular viability was estimated using the MTT test.12 Results were expressed as the concentration of the

studied substance which is able to kill 50% of initial living cells (Concentration necessary for Inhibiting 50% of living cells, IC50%). The values of IC50% are summarized in Table 2. For the nanoparticles of each polymer, the values of the IC50% decreased when the degradation time increased, and for the same degradation time, the values of IC50% varied according to the nature of the polymer. PMLAH/He nanoparticles had IC50% values lower than those of PMLABe. Finally, the values of IC50% decreased when the cellular incubation time increased. Table 2 shows that nanoparticles of PMLA derivatives present a low toxicity and that PMLABe nanoparticles are less toxic than those of PMLAH/ He. The amounts of oligomers or low molecular weight products (2500 >2500 >2500 >940 >940 400 >940 >940 400 >940 >940 450

>2500 2000 300 400 300 250 400 300 300 400 200 100

1600 100 60 1000 100 50 1000 300 40 900 100 40

0

11

21

31

a Estimated IC50% (µg/mL) (concentration corresponding to the death of 50% of the cells), for PMLABe, PMLAHe, and PMLAH/He nanoparticles incubated with J774 A1 murine macrophages for 1, 4, and 24 h, respectively. T0 corresponds to freshly prepared (nondegraded) nanoparticles suspensions. T11, T21, and T31 correspond to nanoparticles suspensions progressively degraded by incubation in a pH 7.5 buffer (37 °C) during 11, 21, and 31 days respectively (n ) 3).

Table 3. Oligomers Formed during Degradationa oligomers (%)

degradation time (week)

PMLABe

PMLAHe

PMLAH/He

0 1 2 3 4

0 0 0 nd 1.7

0 0.5 0.8 nd 9.2

0.3 1.7 3.6 23.4 39.7

a Values were determined from chromatograms obtained by SEC during degradation studies by comparing the partial area under the curve from 0 to 1500 g/mol to the total area of these chromatograms.

are tabulated in Table 3. As could be expected, the amount of degradation products increased as a function of the degradation time. The rate of production of these oligomers increased within the series PMLABe, PMLAHe and PMLAH/He. 4. Discussion 4.1. Degradation of PMLA Derivatives. The degradation of three PMLA derivatives in the physical form of nanoparticles has been investigated. It would have been of interest to compare with unmodified PMLA. However, due to its

high water solubility, it was impossible to prepare nanoparticles, not feasible making to include PMLA in the present study. As shown in Figure 6, three possible mechanisms could be proposed for the degradation of PMLA derivatives. As a first possibility, the degradation could take place by hydrolysis of the lateral chain, leading to the formation of the corresponding alcohol. As a second possibility, the degradation could occur at the end of the main polymer chain by a hydrolysis of the terminal ester bonds, thus leading to the formation of the corresponding monoester. As a third possibility, the degradation could take place on an ester bond randomly located in the main polymer chain, thus leading to the production of reduced molecular weight polymer chains. Although a combination of these different mechanisms cannot be excluded, different results strongly suggested that a random degradation of the main polymer chain was the more likely degradation process. SEC analysis of the polymers during degradation showed unimodal chromatograms for all three types of polymers under study and for all degradation time (Figure 2), suggesting that, whatever the nature of the PMLA derivative, the degradation occurred by statistical hydrolysis of the main chain in polymers of lower molecular weight. Hydrolysis of the short lateral chain or hydrolysis at the end of the main chain were not likely because these mechanisms would have led to the production of very low molecular weight products (as alcohols or monoesters), which should have resulted in bimodal SEC chromatograms. NMR data confirmed that no alcohols or very low molecular weight esters were formed during the degradation process because the spectra remained very similar when time elapsed, indicating that degradation occurred by random hydrolysis of the main polymer chain. This result was in good agreement with various proposals in the literature. Indeed, this mechanism has already been described in the case of PMLA.5 Moreover, Cammas et al.13 have shown previously in the case of PMLABe that the production of malic acid, which would have been produced by hydrolysis of the lateral chain, was very low during the conservation of nanoparticles suspensions in water. The rates of degradation of the polymers were considerably influenced by the fact that the synthesized derivatives were water insoluble polymers. For this reason, the polymers were used in the form of nanospheres for standardizing the contact of the polymers with the degradation medium during degradation studies. The structural properties of these nanospheres influenced the rate of degradation of the polymer. As can be seen in Figures 3 and 4, the pH had no effect on the hydrolysis whatever the polymer. This result was rather surprising because it would be expected that an acidic or basic pH would favor hydrolysis of esters bonds, compared to neutrality. In the conditions prevailing in the core of the nanospheres, it can be imagined that hydrolysis could be controlled by the rate of penetration and diffusion of water into the nanosphere, allowing the hydrolysis of the ester functions and thus leading to the formation of an increasing number of carboxylic functions within the nanoparticles. In turn, such a phenomenon would result in an acidification

142

Biomacromolecules, Vol. 5, No. 1, 2004

within the core of the nanosphere which is favorable to further hydrolysis of the polymer. However, the pH of these acidic microdomains was apparently not modified by the outer buffer solutions, explaining the lack of the effect of the pH on the degradation rate. Further experimental evidence can be given that the rate of degradation of the polymers was controlled by water penetration in the nanosphere. As can be seen from Figures 3 and 4, the normalized molecular weight decreased more rapidly for the copolymer containing 10 mole percent of malic acid, PMLAH/He, which gave a more pronounced hydrophilic character to this polymer2 compared to the others (PMLAHe and PMLABe), suggesting that the degradation rate was inversely related to the hydrophobic character of the polymers. PMLABe was the most hydrophobic polymer because of the introduction of very hydrophobic benzyl groups in the polymer. It is known that the initial molecular weight of the polymers can influence their degradation rate. The influence of this parameter cannot be excluded for these derivatives. Finally, it has been suggested that the chain flexibility and mobility of the polymer, as represented by the value of the glass transition temperature, could influence the degradation rate14 and could also contribute to the ability of water to penetrate in the nanospheres. A similar mechanism as been described in the case of poly(D,L-lactic acid).15 In this respect, the observed difference in Tg could result in different mobilities of the polymer chains at 37 °C (experimental temperature). It can be suggested that the PMLABe chains mobility (Tg ) 37 °C) was lower than the one of PMLAHe and PMLAH/He (Tg ) -10 and -11 °C, respectively). Therefore, it could explain that the rate of degradation was higher for PMLAHe and PMLAH/He than for PMLABe. Finally, considering the structure of the nanoparticles during the degradation process, no direct relationship could be established between the degradation process and the morphology of the particles. However, aggregation phenomena suggested that the surface and possibly the core of the particles was quite rapidly affected by the degradation process. The knowledge of the degradation mechanism of this family of polymer has interesting practical consequences. Indeed, it should be possible to promote the degradation of these polymers by forming statistical copolymers containing a varying amount of hydrophilic units of acid malic, giving a method for modulating the degradation rate of the polymer. However, when solid structures such as nanoparticles are required, the hydrophilic/lipophilic balance of the copolymer should be correctly adjusted in order to adjust simultaneously the degradation rate and to maintain the water insolubility of the polymer, which is necessary for forming the particles in aqueous media. 4.2 Cytotoxicity of the PMLA Polymers and of the Corresponding Nanospheres. To perform cytotoxicity studies, it is necessary to choose a cell line which is representative of the organ which is likely to be mostly exposed to the polymer. As shown in this study, PMLA derivatives can be used to prepare nanoparticles which can be of interest for pharmaceutical applications. Drug targeting is one of these applications, which is a drug delivery strategy consisting of

Martinez Barbosa et al.

Figure 7. Relationship between cytotoxicity, as indicated by the IC50%, and the amount of low molecular weight polymers (under 1500 g/mole) present during the toxicity studies after 1 h of incubation in the presence of J774 A1 murine macrophage-like cells. Amount of oligomers (