br the PDI values increased implying that
the PDI values increased, implying that the more complex (e.g. layered) structures appeared in the system .
Micellar structures formed by the AA-CUR bioconjugate were not mechanically strong and stiﬀ enough to allow their imaging by scan-ning Amyloid Beta-Peptide 1-40 microscopy. Thus we have reinforced them by the cross-linking process. Alginate can be physically cross-linked upon addition
of the calcium ions, due to the reorganization of the polymeric chains around the calcium ions and the formation of so-called egg-box struc-tures. A separate experiment was performed in order to adjust the op-timal calcium chloride concentration for the crosslinking process, using a series of solutions containing the same amount of the polymer and decreasing concentration of calcium chloride. Prior to the crosslinking
D. Lachowicz et al.
Changes in the average diameters and zeta potential values of the micellar solutions of AA-CUR for various calcium chloride/polymer ratios. The optimal calcium chloride/polymer ratio used in further experiments is shown in bold.
the concentration of the AA-CUR bioconjugate was adjusted to 0.6 mg/ ml, to assure that the micelles were indeed formed. The results are shown in Table 1.
The highest absolute value of the zeta potential was found for the calcium ion/bioconjugate ratio of 0.15. The higher zeta potential en-sures the higher colloidal stability of the aqueous micellar dispersion. That corresponds well with the lowest PDI value obtained under these experimental conditions. Therefore we have chosen this particular calcium ion/bioconjugate ratio to crosslink the AA-CUR micelles for further studies. The resulting, stiﬀer individuals were then character-ized by scanning electron microscopy. The SEM image have revealed the regular, spherical objects of the size within the range of 100–200 nm. The energy-dispersive X-ray (EDX) analysis allowed to detect, next to silicon substrate and sputtered gold, also carbon and oxygen from AA-CUR and relatively high amount of calcium, thus confirming the presence of calcium ions in the observed structures (see Fig. 4B).
This observation was confirmed by the AFM visualization. The re-sults of the atomic force microscopy measurements are presented in Fig. 5. The analysis of the AFM images allows to define the average size of the obtained micelles as being around 120–150 nm, which is in a good agreement with the SEM analysis of the crosslinked particles of AA-CUR bioconjugate .
3.3. Curcumin release studies
The normal level of calcium in the human blood is in the range of 0.09 mg/ml (9 mg/dl). This corresponds perfectly with the concentra-tion of calcium chloride we have selected previously for the micelles’ crosslinking process used to obtain SEM images. We have, therefore, decided to study the curcumin release process from the calcium cross-linked micelles, as we believe the micelles will be crosslinked by the calcium present in the blood when the release occurs. To assess the process of curcumin release from the obtained cross-linked micelles the simple model of the lipid membrane environment was applied. Curcumin was released from the bioconjugate due to the hydrolysis of the ester bond between the alginate chain and curcumin moiety. Considering that curcumin is highly hydrophobic and unstable in aqueous media, it was solubilized in the oleic acid phase (mimicking the situation when bioconjugate interacts with the cell membrane of cancer cell) and its concentration in the oleic phase was evaluated spectrophotometrically. The aqueous micellar dispersion of the bio-conjugate was added to the oleic acid to form a suspension, as these phases do not mix. The suspension was shaken while incubated at 37 °C, and the concentration of curcumin in the oleic acid phase was measured after various times using UV–Vis spectroscopy. The release procedure has been described in detail in Supplementary Materials. The results of the release studies are presented in Fig. 6. Curcumin was released gradually without so-called “burst release” until the plateau was reached after about 5 h.
The observed process was relatively fast, taking into account that pH of bioconjugate solution was around 7. That can be explained considering that oleic acid increased the hydrolysis rate, as it provided local acidic environment once the bioconjugate reached the interface. European Polymer Journal 113 (2019) 208–219
In that regard the oleic acid was well chosen as a model phase, as it not only mimicked the lipid environment of the cell membrane, but it also provided the lower local pH typically observed in cancer tissue.
3.4. Evaluation of cytotoxicity of alginate-curcumin bioconjugate micelles against several mouse cancer cell lines
Many in vitro and in vivo studies demonstrated that curcumin ex-hibits remarkable antiproliferative activity towards various cancer cell lines . To verify biological activity of alginate-curcumin bioconju-gate micelles the experiments with several mouse cancer cell lines: mammary carcinoma 4T1, melanoma B16F10 and colon carcinoma cell line CT26-CEA and MC38-CEA were performed. The cells were exposed to diﬀerent doses of AA-CUR bioconjugate for 48 h and then viability of the cells was measured by MTT assay. As demonstrated in Fig. 7A, dose dependent sensitivity of all tested cancer cells to AA-CUR bioconjugate was observed. The strongest cytotoxicity of AA-CUR micelles was found for the concentration of 0.7 mg/ml. AA-CUR used in this concentration decreased viability of cancer cells by 80% or more. High cytotoxic ac-tivity of AA-CUR was also confirmed by microscopic analysis of cell morphology (Fig. 7B). First of all, the density of cancer cells treated with AA-CUR was significantly lower than that for control cells. In addition, the presence of characteristic large cells with altered mor-phology has been observed, which may serve as an evidence of the apoptosis induction by AA-CUR. At the same time free curcumin dis-solved in DMSO (5.95 µg/ml) had weak eﬀect on cell viability (decrease in viability to approximately 80%, data not shown). That discrepancy between the activity of AA-CUR and free curcumin is most likely due to the much lower bioavailability of free curcumin compared to curcumin in the conjugate.