Synthesis of PEG macro-CTA and triblock copolymers
PEG macro-CTA was synthesized according to the reaction procedure described previously and characterized by spectroscopic measurements [33]. It showed an increase in the UV absorbance at 280 nm confirming the inclusion of the xanthate group on the PEG structure, and presented the typical FTIR and 1H-NMR signals described in the literature. PEG macro-CTA: Yield 38%. FTIR (ν, cm− 1) 2948, 2877 (C–H), 1734 (> C = O), 1114 (C–O), 1060 (> C = S), 843 (> C–S). 1H-NMR (CDCl3): (δ, ppm), 4.63 (CH3–CH2–OC(O)), 4.40 (> CH–CH3), 4.23 (–CH2–OC(S)), 3.60–3.85 (–CH2–CH2–O), 1.60 (> CH–CH3), 1.30 (CH3–CH2–OC(O)). UV(> C = S), CHCl3: (π→π*, λmax = 280 nm) and (n→π*, λmax = 357 nm).
Block copolymers were synthetized from PEG macro-CTA. Both samples were characterized by FTIR and 1H-NMR. In S1 polymer, typical signals of the PVBz and PEG blocks were observed in agreement with our results previously reported (data not shown) [33]. The new copolymer S2 showed the FTIR bands located at 3062 cm− 1 (νC-H aromatic), 2921 cm− 1 and 2849 cm− 1 (νC-H aliphatic), 1723 cm− 1 (νC = O), and 1271 and 1115 cm− 1 (νCO–OR). 1H-NMR spectrum of this sample is shown in Fig. 1, including the structure and the peaks assignation based on the corresponding homopolymers spectra, which confirms the structure of the obtained polymer.
The degree of polymerization (DP), composition and ratio of the hydrophilic to hydrophobic blocks, are critical in order to produce vesicles by self-assembly. It is important to define and control these parameters to deepen in the relationship polymer structure/vesicle morphology knowledge to the purpose to obtain the desired vesicle morphology [38].
Based on the 1H-RMN results, composition of copolymer was estimated from the integral ratio of the peak at 3.6 ppm corresponding to methylene hydrogen of ethylene glycol (EG) unit, the signal at 7.7 ppm corresponding to the hydrogen of the aromatic ring of VB unit, and the signal at 4.9 ppm corresponding to the methyne hydrogen of DIPF unit, using the ratio of Eq. (1) for the S2 copolymer:
$${F}_{PEG}=\frac{{I}_{1}/4}{{I}_{1}/4+ {I}_{2}/2+{I}_{3}/1}$$
1
where FPEG is the molar fraction of EG unit in the copolymer and I1, I2 and I3 represent the 1H-NMR resonance peak areas at 3.6 ppm, 7.7 ppm and 4.9 ppm, respectively. Similarly, Fs of the other monomers in the block copolymers were calculated. The relative composition of the EG:BV:DIPF was 0.36:0.28:0.36. Thus, this copolymer presented a hydrophobic block with a relative composition of 44:56 (percentage mole fraction) of BV: DIPF, respectively.
Figure 2 shows the SEC results for the PEG macro-CTA and the triblock copolymers synthesized in this work and Table 1 summarizes the percent of conversion (C), number average molecular weight (Mn), polydispersity index (PDI), and fraction of hydrophilic block of synthetized copolymers. Macro-CTA presented a number-average molecular weight (Mn) of 8806 g/mol and a polydispersity index (PDI = Mw/Mn) of 1.1. After growth of hydrophobics blocks, an increase of the molecular weight and PDI were observed. The copolymer PVBz-b-PEG-b-PVBz showed a Mn of 22.00 kg/mol, PDI of 3.40, and f = 38. These parameters for P(VBz-co-DIPF)-b-PEG-b-P(VBz-co-DIPF) were: Mn = 31.50 kg/mol, PDI = 1.95 and f = 30. These results evidence the increase in molecular weight as a consequence of the hydrophobic blocks growth. In addition, the chain length obtained produce an f adequate for the formation of vesicles by self-assembly.
Table 1
Characteristics parameters of the synthesized copolymers.
Sample name | Composition | time [h] | C [%] | Mn [Kg/mol] | PDI | Fa [%] |
Macro-CTA | PEG-xanthtate | - | 38 | 8.8 | 1.10 | - |
S1 | PVBz-b-PEG-b-PVBz | 16 | 41 | 22 | 3.4 | 38 |
S2 | P(VBz-co-DIPF)-b-PEG-b-P(VBz-co-DIPF) | 17 | 39 | 31.5 | 1.95 | 30 |
a Based on SEC results Percent of conversion (C), number average molecular weights (Mn), and polydispersity index (PDI). |
Self-assembly of the triblock copolymers
It has long been recognized that polymersome or polymeric vesicles made of self-assembled amphiphilic block copolymers are very promising nanostructures [39]. They appear to be highly stable, and present the advantage to encapsulate hydrophilic drugs in the core and hydrophobic drugs in the shell. Thus, they can act as biocompatible devices with wide application in healthcare. In this sense, polymersomes development is a field of great interest and intense research [40]. Our group is devoted to the development of delivery systems for bone pathologies and, in view of these applications the amphiphilic triblock copolymers previously described were self-assembled to generate polymersomes. We were also interested in the effect of the nature of the hydrophobic block on the nanoparticle characteristics. Figure 3 shows TEM images and the size distributions of polymersomes prepared from samples S1 and S2. We found that amphiphilic triblock copolymers formed spherical particles (Fig. 3a and b), but particles obtained from S2 (statistical fumaric copolymer as hydrophobic block) presented smaller average diameter than those obtained from S1 (PVB as hydrophobic block) (Fig. 3b and d). Furthermore, in S2 sample a broad size distribution was observed, although the content of large particles is very small.
Polymersomes obtained from S1 and S2 samples were also analyzed by dynamic light scattering (DLS) and the intensity distribution curves are presented in Fig. 4. It was observed that sample S1 shows a monomodal distribution with a hydrodynamic Z-average diameter (ZD) of 163 ± 64 nm and polydispersity index (PDI) of 0.25 ± 0.03 which indicated a narrow size range, considering that the shoulder near 450 nm is of much lower intensity. In comparison, sample 2 exhibited a monomodal distribution with a hydrodynamic Z-average diameter (ZD) of 119 ± 17 nm and polydispersity index (PDI) of 0.18 ± 0.01 indicating a narrower size range than sample S1.
For both samples, the size differences measured by the TEM and DLS techniques, are in agreement with those previously reported by us and other researchers [33, 34, 41, 42]. These results are consistent with previous studies indicating the size of the polymersomes is influenced by not only the molecular weight of the hydrophobic block, but its chemical nature, composition and rigidity [6]. In our work we found that particles prepared from random copolymer as hydrophobic block have smaller size than those prepared from polyvinylbenzoate as hydrophobic block, emphasizing the influence of the hydrophobic block composition in particle formation.
Characterization and entrapment efficiency of risedronate
The morphology of the vesicles from samples S1 and S2 incorporating risedronate were characterized by TEM and DLS. In Fig. 5 shows the TEM images of both samples together with the corresponding size distribution. It can be observed that sample S1 exhibit a greater average diameter than those of sample S2. Besides, both size distributions are narrow with a very small contribution from large particles. These results are similar to those observed in empty vesicles (Fig. 3).
The curves intensity vs size distribution obtained for dynamic light scattering (DLS) of both samples incorporating Risedronate (R) are presented in Fig. 6 (S1 + R and S2 + R). Both samples presented a bimodal distribution with a main peak (P1) corresponding to the one with the highest integral intensity and a second peak (P2) corresponding to a second smaller size population. The maximum value of distribution in intensity of each peak along with the PDI values are presented in Table 2. We also included the values for the samples without drug (S1 and S2) for comparison purposes.
Table 2
Maximum of the intensity distribution of each particle´s population and polydispersity index.
Sample | P1 [nm] | P2 [nm] | PDI |
S1 | 137 ± 4 | 468 ± 73 | 0.25 ± 0.01 |
S1 + R | 144 ± 2 | 519 ± 73 | 0.22 ± 0.01 |
S2 | 166 ± 11 | 63 ± 8 | 0.18 ± 0.01 |
S2 + R | 179 ± 4 | 67 ± 5 | 0.18 ± 0.01 |
In both cases (S1 and S2), data presented seems to show a small increase in the size of the vesicles by incorporating the hydrophilic drug. However, such size increase does not correspond to statistically significant differences. This result is concordant with hydrophilic properties of the encapsulated drug, that remains included in the internal aqueous portion of the vesicles, without affecting their dimensions.
To evaluate efficiency of encapsulation (amount of active principle encapsulated in the polymersomes) we carried out RP-HPLC assays. The analysis of RP-HPLC chromatograms revealed the presence of the characteristic peak of risedronate at tR = 1.824 ± 0.004 min. Additionally, chromatograms of polymersomes risedronate-loaded samples after extraction, presented a peak corresponding to the drug. The analysis of the area under the peak revealed that the content of risedronate in polymersomes was around 4 ± 2 and 12 ± 2 mg of risedronate per gram of polymersomes for samples S1 + R and S2 + R, respectively. In S1 + R sample, the bisphosphonate concentration for osteogenic in vitro effect was suboptimal, while in S2 + R risedronate concentration was in the optimal range to achieve the same osteogenic effect [43].
3.3 Biocompatibility assays
In living beings, macrophage cells from immune system act as important immune sentinels and are able to react against every particle sensed as foreign body [44]. To determine the in vitro biocompatibility of synthesized polymersomes, we selected an established cell line of murine macrophage/monocytes.
S2 derived–polymersomes allowed cell proliferation at same extent as control condition, without differences over time, as determined by crystal violet method (Fig. 7a). This effect of S2 polymersomes on RAW264.7 proliferation was independent of the tested concentration. In accordance, our previous cytotoxicity studies using RAW 264.7 macrophages incubated with S1 polymersomes demonstrated the absence of cytotoxicity during the same tested period of time [33].
Then, we evaluated mitochondrial activity through the MTT bioassay. This assay is based on the reduction of MTT by mitochondrial succinate dehydrogenase to form an insoluble dark blue formazan product [45]. Only viable cells with active mitochondria reduce significant amounts of MTT to formazan, indicating cell viability. We found that cell viability was preserved in all studied conditions over time (Fig. 7b).
Similar results were obtained with S2-risedronate loaded polymerosomes (S2 + R) they allowed cell proliferation at same extent as control condition, without differences over the culture time (Fig. 7c).
Next, we designed a series of experiments to evaluate a possible anti-inflammatory activity. RAW264.7 cells were incubated either in presence of polymersomes and /or bacterial lipopolysaccharide (LPS) for 24 h, and then we evaluated cell proliferation and nitric oxide production. We found that LPS treated cell proliferate less (80%, p<0.01) compared to basal condition (Fig. 8a), both in presence of 1% of FBS or in DMEM without FBS. While, nanoparticles reverted this toxic effect (nanoparticles alone had no effect on cell proliferation, Fig. 8a). In accordance, NO production was increased in the presence of LPS (42 folds vs basal, Fig. 8b, p<0.0001), while 10 µL of nanoparticles partially reverted this effect (35 folds vs basal, Fig. 8b, p<0.0001). On the other hand, nanoparticles induce a moderate increase on NO production (13 fold vs basal, Fig. 8b) without affecting cell viability or proliferation (Fig. 7). To determine if the effect on nitrite determination was an interference of polymersomes, the control nitrite curve was performed in the presence or absence of polymersomes. Figure 8c shows that there are no differences between points, suggesting that there is other mechanism than direct quenching of nitrite in culture media.
The observed absence of response using RAW264.7 macrophages was not totally surprising, since it has been previously described that those particles containing PEG residues in its structures has low protein affinity as well as induces no immune response [46]. Moreover, it has been demonstrated that liposomes obtained from PEG modified polymers are more resistant to protein interaction, plasma clearance and macrophage uptake [47]. Although additional studies are needed in order to evaluate the release profile of the drug and the effects that it can cause, our findings open a lot of opportunities in the design of therapies based on bisphosphonates-loaded polymersomes.