Binary Micellar Solutions of Poly(Ethylene Oxide)-Poly(Styrene Oxide) Copolymers with Pluronic^® P123: Drug Solubilisation and Cytotoxicity Studies

Abstract

The non-commercial copolymers E₄₅S₈, E₄₅S₁₇ and their mixtures with Pluronic^® P123 (E₂₁P₆₇E₂₁) were studied as carriers of the model drug griseofulvin. Critical micelle concentration (cmc) (dye solubilisation method), drug solubilisation capacity (S_cp and S_h) determined by ultraviolet-visible (UV-Vis) spectroscopy and ¹H nuclear magnetic resonance (¹H NMR) and cytotoxicity (LDH activity in human neutrophils) were studied. E₄₅S₁₇ 1.0 wt.% dispersions presented colloidal aggregates limiting its S_cp in comparison to E₄₅S₈, but in 0.1 wt.% solutions this phenomenon seemed to be absent and E₄₅S₁₇ presented a higher S_cp. The mixtures that showed the best S_cp results contained 50% of P123 and presented low cmc. An evaluation of literature data suggested a minimum E_m content of 62% in E_mS_n copolymers below which the increase of S_n length does not lead to an increase of S_h. The results suggested no toxicity of the copolymers on human neutrophils, supporting the use of P123 and poly(styrene oxide) containing copolymers as drug carriers.

Keywords:
solubilisation capacity; binary micelles; griseofulvin; citotoxicity

Introduction

The potential of micelles formed from block copolymers for encapsulation and delivery of highly hydrophobic drugs has been widely examined. A major challenge in achieving this objective is the attainment of high drug loading into non-toxic surfactant micelles which remain stable on dilution but dissociate at the target site.¹ Low aqueous solubility is a problem in the formulation of many drugs like griseofulvin (Figure 1), a widely used antifungal antibiotic which has been used for many years as a model drug in studies of solubilisation by polymeric micelles.^2-7

Figure 1
Chemical structure of griseofulvin. The letters assign the different proton peaks observed by ¹H nuclear magnetic resonance (¹H NMR).

As described in many papers,^8-14 aqueous micellar solutions of block copolymers based on poly(ethylene oxide) as the hydrophilic component combined with a wide range of hydrophobic blocks have been investigated as vehicles for drug solubilisation. Nanoparticles with poly(ethylene oxide) surface show the ability to evade scavenging by the mononuclear phagocyte system, so resulting in increased circulation times in the blood, and are among the drug delivery systems most studied for this purpose.¹⁵ Copolymers of E_nP_mE_n type, with hydrophilic poly(oxyethylene) and hydrophobic poly(oxypropylene) blocks have been commercially available for several decades and a particular advantage of this family of copolymers is the so-called 'stealth' property of the poly(ethylene oxide) corona of their micelles. Here E denotes oxyethylene (OCH₂CH₂), P denotes oxypropylene (OCH₂CH(CH₃)), and subscripts m and n indicate chain length in P or E units. Concentrated solutions of certain E_nP_mE_n copolymers display thermally reversible gelation in the temperature range required for in situ gelling in topical use,^16,17 subcutaneous injection^11,18 and other vias.¹⁹ Such copolymers attract studies as drug solubilising agents showing, for example, improvements on the bioavailability of the studied drugs.^20,21 However, they present a relatively low solubilisation capacity.⁴ We have investigated a number of micellar solutions in which the core-forming blocks have been designed to be more hydrophobic and more compatible with the drug to be solubilised.^4,22,23 Cambón et al.²⁴ have also reported the advantage of using copolymers containing poly(styrene oxide) as drug carriers. For example, copolymers E₃₃S₁₄E₃₃ and E₃₈S₁₀E₃₈ presented potential application as chemotherapeutic carriers, enhancing more than 60 times doxorubicin apparent solubility (loading capacity of 1.8%) and presenting interesting biological properties. In previous works, we have used mixtures of copolymer E₆₂P₃₉E₆₂ (commercial notation Pluronic^® F87) with copolymer E₁₃₇S₁₈E₁₃₇ (S denotes styrene oxide: OCH₂CH(C₆H₅) to obtain a system with useful gelation characteristics (provided by F87) combined with a satisfactory drug-loading capacity (provided by poly(styrene oxide) micelle core) for drugs like griseofulvin.^3,4 For a given value of S_h (solubilisation capacity in terms of grams of solubilised drug per gram of hydrophobic block), the solubilisation capacity measured as S_cp will be increased if w_h, the weight fraction of S in the copolymer is increased, e.g., by replacing the triblock copolymer by a comparable diblock copolymer, and may be greatly increased if the chosen diblock copolymer forms cylindrical micelles.^23,25

Thus, the purpose of this work is the determination of the solubilisation capacity of binary micellar solutions of two non-commercial diblock copolymers E_mS_n (E₄₅S₈ and E₄₅S₁₇) with Pluronic^® P123. No studies of E₄₅S₁₇ have been reported up to date. E₄₅S₈ has already been studied as drug solubiliser,²³ but to our knowledge, studies of mixed micellar systems containing this copolymer were not reported up to date. The extent of solubilisation was determined by the more usual technique of ultraviolet (UV) spectroscopy and, for comparative purposes, absolutely by ¹H nuclear magnetic resonance (¹H NMR). The dye solubilisation method measured by fluorescence was used to determine the critical micelle concentration (cmc). In addition, to our knowledge, this is the first time that citoxicity studies using human neutrophils were performed in copolymers containing poly(ethylene oxide) and poly(styrene oxide) blocks.

Experimental

Materials

Griseofulvin (molecular mass: 352.8 g moL^-1) was obtained from Sigma-Aldrich and was used in the form of finely ground (1 mm² mesh) powder. Differential scanning calorimetry (DSC) indicated a crystalline form with a melting point of 220.4 ºC and an enthalpy of fusion of 115.6 J g^-1. There was no detectable transition consistent with a glassy component. Copolymer E₂₁P₆₇E₂₁ (commercial notation P123) was a gift from ICI Surfactants (now Uniqema), and copolymers E₄₅S₈ and E₄₅E₁₇ were prepared in the School of Chemistry, Manchester. The molecular characteristics of the copolymers are given in Table 1, i.e., the number-average molar mass (M_n) and weight fraction E (%E) from ¹³C NMR spectroscopy, and the hydrophilic-lipophilic balance (HLB). Details can be found elsewhere.²³

Table 1
Selected properties of the nonionic block copolymers explored in this study

Copolymer	Mn / (g mol-1)	E / %	HLBa	Reference
E45S8	2940	67	13	23
E45S17	4020	49	10	OMIC School of Chemistry (Manchester)
E21P67E21 (P123)	5734	32	6.4	28

• ^a
  Hydrophilic-lipophilic balance (HLB) = E/5.

Critical micelle concentration (cmc)

Stock solutions were prepared by dissolving the copolymers in Milli-Q water and allowing 24 h for complete dissolution before diluting further to concentrations within the range 0.01-60 mg dm^-3. Solubilisation of 1,6-diphenyl-1,3,5-hexatriene (DPH) was used to determine the onset of micellization, as described for triblock copolyethers,²⁶ and before that for ionic surfactants.²⁷ DPH was dissolved in methanol and added to the copolymer solution, so that the final copolymer solution contained 1% (v/v) in methanol and 0.004 mmol L^-1 DPH, a mixture shown to provide the same values of the cmc as those obtained by other methods for copolymers in water alone. A F-4500 Hitachi fluorescence spectrophotometer was used in the experiments, with solution temperatures maintained at 25 ºC and 37 ºC (± 0.2 ºC).

Micellar size

The hydrodynamic diameter (δ_H) of the pure diblock copolymers E₄₅S₁₇ and the mixtures P123/E₄₅S₁₇ 50/50 and P123/E₄₅S₈ 50/50 mixtures in aqueous solutions were determined at 25 ºC, using the equipment Malvern Zetasizer Nano ZS model ZEN 3500. The systems were submitted to 30 scans with an acquisition time of 30 s for each scan. All the measurements were made in triplicate.

Solubilisation of griseofulvin

UV spectroscopy is commonly used to determine the extent of drug solubilisation, but involves an initial calibration procedure. In this study we used the UV spectroscopy and compared the solubilisation capacity determined by this method with that from the absolute method based on ¹H NMR spectroscopy.

UV spectroscopy quantification

A UV-Vis spectrometer (Hitachi U-2000) was calibrated by recording the absorbance (wavelength range 200-350 nm) of methanol solutions of griseofulvin (2-20 mg dm^-1) against a solvent blank. The strong absorbance at 292 nm gave a satisfactory Beer's law plot. In a solubilisation experiment, a 10 cm³ portion of a stock 1 wt.% copolymer solution was added to finely-ground (1 mm² mesh) griseofulvin powder (0.02 g). The mixture was stirred at 25 ºC for 72 h before being filtered (0.45 µm from Millipore) to remove any unsolubilised drug. The sample was then diluted with methanol to enable analysis by UV spectroscopy. The water content after dilution was low enough to allow the calibration for methanol solutions to be used without correction. After dilution the samples were held at 37 ºC for 72 h before analysis. Measurements were carried out in triplicate and the results were averaged.

¹H NMR spectroscopy quantification

Solutions with solubilised drug were prepared and filtered as described for the UV method, but were then freeze dried (24 h, 10^-3 mm Hg) to remove water. The entire sample was dissolved in CDCl₃ and its ¹H NMR spectrum recorded at ambient temperature (22 ºC) using a Bruker DRX 500 11.7 T (499.80 MHz for ¹H) spectrometer in a 5 mm inverse detection z-gradient probe at 298 K. The spectra were obtained using 90º rf pulse (9.20 µs), a spectral width of 12019 Hz, 256 transients with 64000 data points, an acquisition time of 2.73 s, and a relaxation delay of 10 s before being converted to 32000 data points and the phase and baseline corrected manually using the Bruker software. The amount of griseofulvin solubilised per gram of polymer was determined using appropriate peak integrals. All measurements were carried out in triplicate and the results averaged.

Cytotoxicity study: determination of lactate dehydrogenase (LDH) assay

Human neutrophils were isolated by Lucisano and Mantovani's method^29,30 with slight modifications. Cells pellets were suspended in Hank's balanced salt solution (HBSS) containing 80-90% neutrophils with viability of 90 ± 2.0% established by Trypan Blue exclusion test. Human neutrophils (2.5 × 106 cells mL^-1) were incubated (15 min at 37 ºC) with HBSS (non-treated cells), Dimethyl sulfoxide (DMSO) (0.4% (v/v), vehicle-control), Triton X-100 (0.2% (v/v), standard cytotoxic drug) and following test drugs (1, 10, 50 and 100 µg mL^-1): diblock copolymers E_mS_n (E₄₅S₈ and E₄₅S₁₇), Pluronic^® P123, the binary mixtures of copolymers (P123/E₄₅S₈ 50/50 and P123/E₄₅S₁₇ 50/50, named as P50/S8 and P50/S17, respectively) or their versions loaded with griseofulvin (P123/E₄₅S₈G and P123/E₄₅S₁₇G, named as P50/S8G and P50/S17G, respectively).

The LDH activity was determined by a commercially available method (LDH liquiform of Labtest Diagnosis).

Statistical analysis

The results are expressed as mean ± standard deviation (SD). The statistical significance of differences between groups was determined by one-way ANOVA, followed by Tukey for multiple comparisons as a post hoc test. The significance level was set at p < 0.05.

Results and Discussion

Regarding binary mixtures, in tests made in our laboratories, the mixtures of P123 with copolymer E₄₅S₁₇ showed stable thermoresponsive properties only when the ratio was 50/50 at concentration 29% (m/m), with a transition temperature fluid-gel-fluid in the range 18-47 ºC. In addition, the systems P123/E₄₅S₈ 50/50 and 70/30 presented stable thermoresponsive properties in the concentrations 25-29% (50/50) and 23-29% (70/30), showing the potential to use such mixtures in subcutaneous administration of drugs, since they may form gel at body temperature (unpublished results).

Critical micelle concentration (cmc)

The arrows in the plots of fluorescence intensity of DPH against copolymer concentration (log scale) of two copolymer systems (E₄₅S₁₇ and P123/E₄₅S₁₇) in Figure 2 indicates the points at which a sharp increase in fluorescent intensity takes place, estimated as the cmc, due to the transference of DPH from the aqueous polar medium to the hydrophobic environment of the cores of the micelles.²⁸

Figure 2
Fluorescence intensity (DPH) versus concentration (log scale) for aqueous solutions of (o) E₄₅S₁₇ and (■) P123/E₄₅S₁₇ 30/70 mixture at 37 ºC.

As seen in Table 2, the obtained values of cmc were in the range 1 × 10^-2-2 × 10^-4 wt.%. The much higher concentration of the systems used in the solubilisation experiments (1.0 wt.%) is consistent with effectively complete micellization.

There was no considerable variation of cmc of diblocks E₄₅S₈ and E₄₅S₁₇ and their mixtures as the temperature changed from 25 to 37 ºC (see Table 2 and Figure 3). Typically, the cmc at 37 ºC was lower than at 25 ºC, especially for P123 (Table 2). Previous studies show that E_nP_mE_n copolymers have a more endothermic micellization process (DHº_mic around 200 kJ moL^-1)²⁶ than E_nS_m, E_nS_mE_n copolymers (DHº_mic ranging from 4 to 40 kJ moL^-1).^31,32

Values of cmc for E₄₅S₈ and E₄₅S₁₇ expressed in mol dm^-3 are much lower than those of P123 (E₂₁P₆₇E₂₁). Diblock copolymers have cmc values lower than their correspondent triblock copolymers (with the same molar mass and block composition), and the more hydrophobic is the copolymer, the lower its cmc.³³ This explains why the cmc of P123 (E₂₁P₆₇E₂₁) is the highest, since P123 is the less hydrophobic copolymer. The relative hydrophobicity between P and S units in a diblock copolymer is 1:12.¹¹ From this concept, we find that the EP copolymers which are equivalent to E₄₅S₈ and E₄₅S₁₇ would be E₄₅P₉₆ and E₄₅P₂₀₄, respectively, and then we can compare the hydrophobicities of the three copolymers by comparing their HLB's: 6.4 for P123, 5.2 for E₄₅P₉₆ and 2.8 for E₄₅P₂₀₄. The lower the HLB, the higher is the hydrophobicity and the lower is the cmc. Therefore, it was expected that E₄₅S₁₇ would have the lowest cmc. Besides this, diblock copolymers have cmc values lower than their correspondent triblock copolymers (with the same molar mass and block composition),³³ contributing to the lower cmc value of E₄₅S₁₇ in comparison to the triblock P123.

The cmc's (mol dm^-3) of the mixtures are similar to those of the pure copolymers E₄₅S₈ and E₄₅S₁₇ with the same order of magnitude (see Table 2). At 37 ºC all systems, including P123, have similar cmc's. The low cmc values of these systems bring potential stability of their micelles after blood dilution, turning these mixtures promising drug carriers in pharmacological applications.

The change in standard Gibbs free energy of micellization per mole of copolymer unimer can be related to cmc as shown in the following equation:

It was observed that the values of δGº_mic at both temperatures for the mixtures P123/E₄₅S₈ and P123/E₄₅S₈ were lower than the values for P123 and E₄₅S₈ indicating that the micellization process of mixtures is more spontaneous than the micellization process of pure surfactants, demonstrating a synergistic effect (Table 2). It was observed that all the values of δGº_mic were lower at body temperature than at room temperature, indicating that these systems may be used to spontaneously solubilize drugs in the body.

Table 2
Values of cmc and standard Gibbs free energy of micellization for E₄₅S₈, E₄₅S₁₇ and their mixtures with P123 at 25 °C and 37 °C

System	cmc / (mmol dm-3)	ΔGºmic / (kJ mol-1)	cmc / (mmol dm-3)	ΔGºmic / (kJ mol-1)
	25 °C		37 °C
P123a	2.44 × 10-2	-26.3	7.00 × 10-4	-36.5
P123/E45S8 50/50	9.20 × 10-4	-34.4	4.80 × 10-4	-37.5
P123/E45S8 30/70	6.60 × 10-2	-24.8	4.20 × 10-4	-37.8
E45S8	1.02 × 10-3	-34.2	6.80 × 10-4	-36.6
P123/E45S17 50/50	3.50 × 10-4	-36.8	3.10 × 10-4	-38.6
P123/E45S17 30/70	3.50 × 10-4	-36.8	2.90 × 10-4	-38.8
E45S17	4.70 × 10-4	-36.1	4.50 × 10-4	-37.6

• ^a
  Literature values.²⁸

Table 3 shows the cmc values at 25 ºC for diblock copolymers with similar ethylene oxide block length and different styrene oxide (S_n) block lengths. The cmc of E₅₀S_3.5 is much higher than the expected. Mai et al.³⁴ explain that the relatively high values of cmc found for this copolymer are consistent with their wider distribution of apparent hydrodynamic radius.

Figure 3 shows the plot of cmc (mol dm^-3) versus hydrophobic block length of diblock copolymers with similar ethylene oxide (E_m) block length and different styrene oxide (S_n) block lengths. Data from copolymers studied in the literature and data from this work were used to plot the graph.^23,31,34 To plot the graph, the values of log₁₀ (cmc; mol dm^-3) of the copolymers were adjusted to a common E block length, since the number of E units affect the cmc within a series of copolymers. We used the equation:

Figure 3
Critical micelle concentration (cmc) values at 25 °C for diblock copolymers with similar ethylene oxide (E_m) block length and different styrene oxide (S_n) block lengths. Literature values (o)^23,26,34 and this work (•).

Table 3
Solubilisation capacities S_cp and S_h (mg g^-1) of griseofulvin in 1 wt.% solutions of P123, E₄₅S₁₇, E₄₅S₈ and their mixtures at 25 °C and 37 °C. S₀ = 1.4 e 1.9 mg dL^-1 at 25 °C and 37 °C, respectively. Data obtained by UV-Vis spectroscopy

System	Scp / (mg g-1)	Sh / (mg g-1)	S/S0	Scp / (mg g-1)	Sh / (mg g-1)	S/S0
	25 °C			37 °C
P123	3.30 ± 0.07	4.9	3.6	5.10 ± 0.21	7.5	3.7
P123/E45S8 50/50	4.00 ± 0.13	7.9	3.8	7.50 ± 0.16	14.9	4.9
P123/E45S8 30/70	3.80 ± 0.06	8.7	3.7	5.60 ± 0.23	12.9	3.9
E45S8	7.40 ± 0.20	22.4	6.3	8.70 ± 0.08	26.4	5.6
P123/E45S17 50/50	2.80 ± 0.15	4.7	3.0	8.10 ± 0.17	13.6	5.3
P123/E45S17 30/70	2.40 ± 0.12	4.3	2.7	2.80 ± 0.19	5.0	2.5
E45S17	5.20 ± 0.07	10.2	4.7	7.40 ± 0.24	14.5	8.6

This equation, proposed by Booth, Atwood and Price,³⁵ where ν is the number of E units in the copolymer, was used to correct each value for a change in E block length from 100 units. It can be noted that the increase of hydrophobic block length reduces the cmc of the copolymers. Drug solubilizing agents with low values of cmc are interesting since they present higher micellar stability after blood dilution.

Solubilisation of griseofulvin and micellar size

With the data of drug solubilities obtained by UV-Vis spectroscopy, two parameters were investigated for the copolymers and their mixtures: S_cp, the solubilisation capacity expressed in mg of drug per g of polymer, and S_h, the solubilisation capacity expressed in mg of drug per g of hydrophobic block, where S_cp = S - S₀/m_cop and S_h = S_cp/W_h (S: solubility of drug in the micellar solution; S₀: aqueous solubility of drug; W_h: mass fraction of hydrophobic block). These values are shown in Table 4.

Table 4
Results of solubilisation capacity (S_cp) of griseofulvin in the copolymer solutions of E₄₅S₁₇ in the concentrations 0.1, 0.5 and 1 wt.% at 25 °C and 37 °C measured by UV-Vis spectrometry. S₀ = 1.4 e 1.9 mg dL^-1 at 25 °C and 37°C, respectively

Concentration / wt.%	Scp / (mg g-1)	Sh / (mg g-1)	Scp / (mg g-1)	Sh / (mg g-1)
	25 °C		37 °C
0.1	14.20 ± 0.18	28	22.20 ± 0.15	44
0.5	7.60 ± 0.09	15	8.70 ± 0.17	17
1.0	5.20 ± 0.07	10	7.40 ± 0.05	15

E₄₅S₈ and E₄₅S₁₇ presented satisfactory solubilisation capacity values when compared to Pluronic^® P123 (E₂₁P₆₇E₂₁). This is due the higher hydrophobicity of their poly(styrene oxide) blocks in comparison with the poly(propylene oxide) block (P₆₇) of Pluronic^® P123 (see Table 6). The values found for the P123 and E₄₅S₈ are in agreement with the data found in the literature: S_cp = 3.0 mg g^-1 at 25 ºC and S_cp = 3.8 mg g^-1 at 37 ºC for P123;²² and S_cp = 7.5 mg g^-1 at 25 ºC for E₄₅S₈.

Table 5
Solubilisation capacity of griseofulvin (S_cp and S_h) at 25 ºC, micellar association number (N_w) and ethylene oxide weight percentage (%E) of diblock copolymers with similar ethylene oxide (E_n) block lengths and different styrene oxide (S_n) block lengths

Copolymer	Scp / (mg g-1)	Sh / (mg g-1)	Nw	%E	HLB	Reference
E50S3.5	2.8	16	20	84	16.8	34
E50S5.1	4.0	17	35	78	15.6	34
E51S6.5	4.9	18	50	74	14.8	34
E45S8	7.5	22	80	67	13.4	23
E45S10	11.2	29	103	62	12.4	23
S13E60	14.1	37	100	63	12.6	34
S15E63	11.2	28	140	61	12.2	23
E45S17	5.2	10	155	49	9.8	present work
S17E65	11.7	27	150	58	11.6	23
S20E67	12.4	27	189	55	11.0	23

Figure 4
¹H NMR spectrum in CDCl₃ of griseofulvin in (a), griseofulvin and E₄₅S₈ recovered from solubilisation at 37 °C in (b) and griseofulvin and E₄₅S₁₇ recovered from solubilisation at 37 °C in (c). Expansion of the drug peaks was made in the region 5.3-6.3 ppm.

It is interesting to evaluate the dependence of the drug solubilisation capacity (S_cp) with temperature and cmc. Low values of cmc mean that the copolymers are more hydrophobic, what confers to them higher solubilisation capacities. Values of cmc of E₄₅S₈ and E₄₅S₁₇ are similar to each other and lower than that of P123 (see Table 3); this explains the higher solubilisation capacities of E₄₅S₈ and E₄₅S₁₇ in comparison to P123.

Observing the chemical structure of E₄₅S₁₇, it was expected that its solubilization capacity would be higher than that of E₄₅S₈, due its higher hydrophobicity (it contains 9 units of S more than E₄₅S₈). However this was not observed. The reason may be the limited solubility of this polymer, since it aggregates itself forming a turbid system at 1.0 wt.%, what affects the solubilisation of griseofulvin in the system. This observation suggests that a %E of 49% (the %E of E₄₅S₁₇) is too low to turn a copolymer of E_mS_n type soluble in water.

Quantification experiments using ¹H NMR were performed to confirm the results obtained at 37 ºC, and the values of S_cp found using this technique were 8.7 mg g^-1 for E₄₅S₈ and 6.8 mg g^-1 for E₄₅S₁₇, in agreement with the UV-Vis quantification. Figure 4 shows the spectra of griseofulvin and the copolymers encapsulated with griseofulvin. Attributions were based on the work of Rekatas et al.³⁶ and Ribeiro et al.⁷ See Figure 1 for griseofulvin proton attributions.

Silva et al.³⁷ observed that, in aqueous solutions, P123 (E₂₁P₆₇E₂₁) and F127 (E₉₈P₆₇E₉₈) form micelles, while L121 (E₅P₆₈E₅) form aggregates that eventually separate from the solution forming a crystalline liquid phase.³⁷ This is due to the very low HLB of L121 (2.0) in comparison to that of F127 (HLB = 14) and that of P123 (HLB = 6.4).

As already mentioned, E₄₅S₁₇ at 1 wt.% solutions shows turbidity, a behavior which seems similar to that reported for L121. This behavior is probably due the high hydrophobicity of the copolymers. Since the relative hydrophobicity between P and S unit in a diblock copolymer is 1:12, we found the E_mP_n copolymer equivalent to E₄₅S₁₇ and calculated the HLB of this copolymer in order to compare its hydrophobicity with that of L121 (E₅P₆₈E₅). We found that E₄₅P₂₀₄ (12 × 17 units of S = 204) is the E_mP_n with a hydrophobicity equivalent to E₄₅S₁₇. The HLB of E₄₅P₂₀₄ is 2.8, only a little higher than that of L121.

The graphs of scattering intensity versus hydrodynamic diameter for the copolymer E₄₅S₁₇ are in Figure 5. The wide unimodal peak found for E₄₅S₁₇ at 1 wt.% confirms the presence of agglomerates. The average diameter at 25 ºC for the copolymer E₄₅S₁₇ was 29 ± 3 nm at 0.1 wt.% with a polydispersity of 0.255 and 66 ± 2 nm at 1.0 wt.% with a polydispersity of 0.230, respectively. The average diameter at 25 ºC for the copolymer E₄₅S₈ was 19 ± 2 nm at 0.1 wt.% with a polydispersity of 0.200 and 17 ± 2 nm with a polydispersity at 1.0 wt.% of 0.091, respectively.

Figure 5
Scattering intensity versus hydrodynamic diameter at 25 °C for the copolymer E₄₅S₁₇ at 0.1 wt.% in (a) and 1.0 wt.% in (b).

With the aim to evaluate the influence of E₄₅S₁₇ agglomerates on the solubilisation capacity, griseofulvin solubilisation experiments were also performed in 0.1 wt.% and 0.5 wt.% solutions of this copolymer. The results are shown in Table 4.

The aggregation presented by diblock E₄₅S₁₇ adversely affects the solubility of griseofulvin. When the concentration of E₄₅S₁₇ is 0.1 wt.%, the aggregation phenomenon seems to be absent and relatively high solubilisation capacity is reached. The low cmc values of E₄₅S₁₇ (Table 2) ensures a virtually complete micellization at 0.1 wt.%. Observing the substantial increase of S_cp in this concentration, we may infer that as the polymer concentration increases, the percentage of polymer molecules in aggregate form increases and the percentage of molecules in micellar form, available to solubilise the drug, decreases, causing a relatively low S_cp for E₄₅S₁₇ at 1.0 wt.%.

Micellar size measurements by light scattering were performed to evaluate if a co-micellization process took place in the binary mixtures of E₄₅S₁₇ and E₄₅S₈ with P123 at 1 wt.%. As shown in Figure 6, the appearance of a single peak for both mixtures instead of two peaks, confirm the formation of binary micelles in these systems.³⁸ The values of hydrodynamic diameter for P123/E₄₅S₈ 50/50 and P123/E₄₅S₁₇ 50/50 at 1 wt.% were 17 ± 1 nm with a polydispersity of 0.106 and 30 ± 3 nm with a polydispersity of 0.248, respectively.

Figura 6
Scattering intensity versus hydrodynamic diameter at 25 °C for the mixtures P123/E₄₅S₈ 50/50 at 1 wt.% in (a) and P123/E₄₅S₁₇ 50/50 at 1 wt.% in (b).

Data of griseofulvin solubilisation obtained at 25 ºC using UV-Vis for diblock copolymers with similar ethylene oxide (E_m) block length and different styrene oxide (S_n) block lengths are shown in Table 6. These data were used to plot a graph (see Figure 6) that shows the variation of griseofulvin solubilisation capacity (S_h) in relation to the number of S units (hydrophobic block length, n).

The increase in hydrophobic block length (n) in the range 3.5-13 increased the drug solubility and the association micellar number (N_w). However, in the range 13-20, the increase in n leads to a decrease in solubilisation capacity S_h, with a more pronunciated effect in E₄₅S₁₇. Lee et al.³⁹ reported that the formation of molecular aggregates observed for L121 copolymer promotes inefficiency of drug solubilisation. The smooth decrease observed for the copolymers S₁₅E₆₃, S₁₇E₆₅, S₂₀E₆₇ may be associated with a minimum %E, below which the increase of hydrophobic block length does not lead to an increase of solubilisation capacity S_h. For this copolymer series, this %E would be 62% (Figure 7).

Figure 7
Solubilisation capacity (S_h) of diblock copolymers with similar ethylene oxide (E_m) block lengths and different styrene oxide (S_n) block lengths. 25°C: (o) literature dates^23,34 and (■) present work.

In a previous study of a copolymer series of E_nP_mE_n type, an increase on partition coefficient was observed as the temperature and the hydrophobic block (P_m) length increased.⁴⁰

The mixtures that showed the best S_cp results were P123/E₄₅S₈ 50/50 and P123/E₄₅S₁₇ 50/50, with values equal to 7.5 and 8.1 mg g^-1, respectively, at 37 ºC.

Pierri and Avgoustakis⁴¹ studied the encapsulation and release of griseofulvin from micelles of polylactide-polyethyleneglycol copolymers named as PLA(X)-PEG(Y), where X and Y are the molecular masses of the blocks in KDa. They found that PLA(4)-PEG(5) encapsulated only 6.5 mg of drug per g of copolymer, an amount lower than the S_cp values at 37 ºC for E₄₅S₈, E₄₅S₁₇ and the mixtures P123/E₄₅S₈ 50/50 e P123/E₄₅S₁₇ 50/50.

Cytotoxicity of polymers in human neutrophils

The possible toxic effects of the polymers in human neutrophils were investigated by measurements of LDH activity in cell suspension. We found that diblock copolymers E_mS_n (E₄₅S₈ and E₄₅S₁₇) and Pluronic^® P123 did not induce a significant increase in the LDH activity release by human neutrophils when compared with control group (DMSO 0.4%) (Figure 8). In addition, neither the binary mixtures of copolymers (P123/E₄₅S₈ 50/50 and P123/E₄₅S₁₇ 50/50, named in Figure 8 as P50/S8 and P50/S17, respectively) or their versions loaded with griseofulvin (P123/E₄₅S₈G and P123/E₄₅S₁₇G, named in Figure 9 as P50/S8G and P50/S17G, respectively) promoted the increase in the LDH activity (Figure 9). The measurement of LDH activity, enzyme present in the cell cytoplasm, is a marker of intact membrane with considerable sensitivity. In this study the polymers alone or associated with griseofulvin did not affect the cell viability assessed by LDH activity, suggesting the absence of toxicity on the cell membrane human neutrophils.

Figure 8
Evaluation of (a) Pluronic^® P123, (b) E₄₅S₈ and (c) and E45S17 toxicity measured by lactate dehydrogenase (LDH) activity in human neutrophils. Data from two to eight samples. *Versus control group (DMSO 0.4%, v/v). Results represent means ± the standard error of the mean (SEM) (p < 0.05; ANOVA and Tukey as the post hoc test). For more details see item cytotoxicity study of Experimental section.

Figure 9
Evaluation of P50/S8 in (a), P50/S17 in (b), P50/S8G in (c) and P50/S17G in (d) toxicity measured by lactate dehydrogenase (LDH) activity in human neutrophils. Data from two to eight samples. *Versus control group (DMSO 0.4%, v/v). Results represent means ± the standard error of the mean (SEM). (p < 0.05; ANOVA and Tukey as the post hoc test).

Conclusions

The E_mS_n type copolymers studied, E₄₅S₈ and E₄₅S₁₇, and their mixtures with P123 showed low cmc values, an advantage to apply them as drug solubilizing agents for pharmacological applications.

E₄₅S₁₇, with a weight percentage of poly(ethylene oxide) (%E) of 49%, formed insoluble/colloidal aggregates that limited the drug solubilisation capacity in comparison to E₄₅S₈ at 1.0 wt.%. In E₄₅S₁₇ 0.1 wt.% solutions, this phenomenon seemed to be absent and a better solubilisation capacity was observed. An evaluation of data of solubilisation capacity S_h (weight of drug/weight of hydrophobic block) collected from the literature indicated that there is a minimum %E of 62%, below which the increase of hydrophobic block length does not lead to an increase of in S_h in E_mS_n type copolymers.

Preliminary studies showed that the systems P123/E₄₅S₈ 50/50, P123/E₄₅S₈ 30/70 and P123/E₄₅S₈ 50/50 presented thermoresponsive gelling properties provided by Pluronic^® P123. Among the mixtures, the systems P123/E₄₅S₈ 50/50 and P123/E₄₅S₁₇ 50/50 showed the best solubilisation results. Considering these properties, these binary mixtures have the potential to be used in subcutaneous administration of drugs.

Besides this, E₄₅S₈, E₄₅S₁₇, Pluronic^® P123 and their mixtures did not affect the cell viability assessed by LDH activity, suggesting the absence of toxicity on the cell membrane human neutrophils, which supports the use of Pluronic^® P123 and copolymers containing poly(styrene oxide) as drug solubilising agents in pharmaceutical formulations.

Acknowledgments

This work was supported by the CNPq (SAO and NMPSR, project number 550857/2010-9), CAPES (CLM), PRONEX, CNPq/FUNCAP and CENAUREMN.

References

• ¹
  Tyrrell, Z. L.; Shen, Y.; Radosz, M.; Prog. Polym. Sci.2010, 35, 1128.
• ²
  Elworthy, P. H.; Patel, M. S.; J. Pharm. Pharmacol.1982, 34, 543.
• ³
  Pinho, M. E. N.; Costa, F. M. L. L.; Filho, F. B. S.; Ricardo, N. M. P. S.; Yeates, S. G.; Attwood, D.; Booth, C.; Int. J. Pharm.2007, 328, 95.
• ⁴
  Ribeiro, M. E.; Vieira, I. G.; Cavalcante, I. M.; Ricardo, N. M.; Attwood, D.; Yeates, S. G.; Booth, C.; Int. J. Pharm.2009, 378, 211.
• ⁵
  Oliveira, C. P.; Ribeiro, M. E. N. P.; Ricardo, N. M. P. S.; Souza, T. V. P.; Moura, C. L.; Chaibundit, C.; Yeates, S. G.; Nixon, K.; Attwood, D.; Int. J. Pharm.2011, 421, 252.
• ⁶
  Oliveira, C. P.; Vasconcellos, L. C. G.; Ribeiro, M. E. N. P.; Ricardo, N. M. P. S.; Souza, T. V. P.; Costa, F. M. L. L.; Chaibundit, C.; Yeates, S. G.; Attwood, D.; Int. J. Pharm.2011, 409, 206.
• ⁷
  Ribeiro, M. E.; de Moura, C. L; Vieira, M. G.; Gramosa, N. V.; Chaibundit, C.; de Mattos, M. C.; Attwood, D.; Yeates, S. G.; Nixon, S. K.; Ricardo N. M.; Int. J. Pharm.2012, 436, 631.
• ⁸
  Adams, M. L.; Lavasanifar, A.; Kwon, G. S.; J. Pharm. Sci.2003, 92, 1343.
• ⁹
  Gaucher, G.; Dufresne, M. H.; Sant, V. P.; Kang, N.; Maysinger, D.; Leroux, J. C.; J. Controlled Release2005, 109, 169.
• ¹⁰
  Savic, R.; Eisenberg, A.; Maysinger, D.; J. Drug Targeting2006, 14, 343.
• ¹¹
  Attwood, D.; Zhou, Z.; Booth, C.; Expert Opin. Drug Delivery2007, 4, 533.
• ¹²
  Chiappetta, D. A.; Sosnik, A.; Eur. J. Pharm. Biopharm.2007, 66, 303.
• ¹³
  Alvarez-Lorenzo, C.; Sosnik, A.; Concheiro, A.; Curr. Drug Targets2011, 12, 1112.
• ¹⁴
  Khoee, S.; Kavand, A.; Eur. J. Med. Chem.2014, 73, 18.
• ¹⁵
  Rabanel, J. M.; Hildgen, P.; Banquy, X.; J. Controlled Release2014, 185, 71.
• ¹⁶
  Schmolka, I. R.; J. Biomed. Mater. Res.1972, 6, 571.
• ¹⁷
  Rodeheaver, G. T.; Kurtz, L.; Kircher, B. J.; Edlich, R. F.; Ann. Emerg. Med.1980, 9, 572.
• ¹⁸
  Gong, C.; Qi, T.; Wei, X.; Qu, Y.; Wu, Q.; Luo, F.; Qian, Z.; Curr. Med. Chem.2013, 20, 79.
• ¹⁹
  Ozbılgın, N. D.; Saka, O. M.; Bozkır, A.; Drug Delivery2014, 21, 140.
• ²⁰
  Abdelbary, G.; Makhlouf, A.; Pharm. Dev. Technol.2014, 19, 717.
• ²¹
  Moretton, M. A.; Cohen, L.; Lepera, L.; Bernabeu, E.; Taira, C.; Hocht, C.; Chiappetta, D. A.; Colloids Surf., B2014, 122, 56.
• ²²
  Attwood, D.; Booth, C. In Colloid Stability and Application in Pharmacy, 3^rd ed.; Tadros, T. F., ed.; Wiley-VCH: Weinheim, 2007, ch. 2.
• ²³
  Crothers, M.; Zhou, Z.; Ricardo, N. M. P. S.; Yang, Z.; Taboada, P.; Chaibundit, C.; Attwood, D.; Booth, C.; Int. J. Pharm.2005, 293, 91.
• ²⁴
  Cambón, A.; Rey-Rico, A.; Barbosa, S.; Soltero, J. F. A.; Yeates, S. G.; Brea, J.; Loza, M. I.; Alvarez-Lorenzo, C.; Concheiro, A.; Taboada, P.; J. Controlled Release2013, 167, 68.
• ²⁵
  Zhou, Z.; Chaibundit, C.; D'Emanuele, A.; Lennon, K.; Attwood, D.; Booth, C.; Int. J. Pharm.2008, 354, 82.
• ²⁶
  Alexandridis, P.; Holzwarth, J. F.; Hatton, T. A.; Macromolecules1994, 27, 2414.
• ²⁷
  Chattopadhyay, A.; London, E.; Anal. Biochem.1984, 139, 408.
• ²⁸
  Alimi-Guez, D.; Leborgne, C.; Pembouong, G.; van Wittenberghe, L.; Mignet, N.; Scherman, D.; Kichler, A.; J. Gene Med.2009, 11, 1114.
• ²⁹
  Lucisano, Y. M.; Mantovani, B.; J. Immunol.1984, 132, 2015.
• ³⁰
  Kabeya, L. M.; Kanashiro, A.; Azzolini, A. E.; Soriani, F. M.; Lopes, J. L.; Lucisano-Valim, Y. M.; Res. Commun. Mol. Pathol. Pharmacol.2002, 111, 103.
• ³¹
  Crothers, M.; Attwood, D.; Collett, J. H.; Yang, Z.; Booth, C.; Taboada, P.; Mosquera, V.; Ricardo, N. M. P. S.; Martini, L. G. A.; Langmuir2002, 18, 8685.
• ³²
  Yang, Z.; Crothers, M.; Ricardo, N. M. P. S.; Chaibundit, C.; Taboada, P.; Mosquera, V.; Kelarakis, A.; Havredaki, V.; Martini, L.; Valder, C.; Langmuir2003, 19, 943.
• ³³
  Booth, C.; Attwood, D.; Macromol. Rapid Commun.2000, 21, 501.
• ³⁴
  Mai, S.; Booth, C.; Kelarakis, A.; Havredaki, V.; Ryan, A. J.; Langmuir2000, 16, 1681.
• ³⁵
  Booth, C.; Attwood, D.; Price, C.; Phys. Chem. Chem. Phys.2006, 8, 3612.
• ³⁶
  Rekatas, C. J.; Mai, S. M.; Crothers, M.; Quinn, M.; Collett, J. H.; Attwood, D.; Heatley, F.; Martini, L.; Booth, C.; Phys. Chem. Chem. Phys.2001, 3, 4769.
• ³⁷
  Silva, R. C.; Olofsson, G.; Schillén, K.; Loh, W.; J. Phys. Chem. B2002, 106, 1239.
• ³⁸
  Jindal, N.; Metha, S. K.; Colloids Surf., B2015, 129, 100.
• ³⁹
  Lee, E. S.; Oh, Y. T.; Youn, Y. S.; Nam, M.; Park, B.; Yun, J.; Kim, J. H.; Song, H. T.; Oh, K. T.; Colloids Surf., B2011, 82, 190.
• ⁴⁰
  Kadam, Y.; Yerramilli, U.; Bahadur, A.; Colloids Surf., B2009, 72, 141.
• ⁴¹
  Pierri, E.; Avgoustakis, K.; J. Biomed. Mater. Res., Part A2005, 75, 639.

Publication Dates

• Publication in this collection
  Nov 2015

History

• Received
  26 June 2015
• Accepted
  18 Aug 2015