AWD 131-138 br Fig A FT IR spectra focused on amide I
Fig. 2. (A) FT-IR spectra, focused on amide I, II, V, VI peaks (blue rectangles); (B) XRD diﬀractograms; and (C) DSC curves, focused on the glass transition temperature (Tg) and specific heat capacity ( Cp) of three α-mangostin loaded FNPs (EDClow-FNP, EDChigh-FNP, PEI-FNP) and the blank counterparts. Comparison between the free α-mangostin and physical mixture of the blank FNPs and free drug at the same amount as drug loading values using (D) DSC and (E) XRD are also demonstrated (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).
Fig. 3. (A) Aqueous solubility of α-mangostin loaded FNPs and the free drug; and dissolution profiles of α-mangostin loaded FNPs in (B) HEPES buﬀer and (C) HEPES + 0.5% Tween 80 (n = 3). Significant diﬀerences are noted between the free drug and FNP formulas (*, p < 0.01); between EDClow-FNP and other FNP formulas (**, p < 0.01) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).
3.5. Alpha mangostin dissolution test
In drug delivery research, the ability to favorably control the drug release rate is crucial, especially in the application of cancer che-motherapy. The dissolution profiles of α-mangostin loaded FNPs in non-sink condition HEPES and sink condition HEPES + 0.5% Tween 80 buﬀers are shown in Fig. 3B and C, respectively. Obviously, with Tween 80, all formulas showed rapidly α-mangostin released within 30 min in this sink condition, with no significant diﬀerence among them. On the other hand, without Tween 80, which simulates the real non-sink bio-logical condition, all formulas showed a zero-order sustained release of more than 3 days, with R2 > 0.9000. The drug release rate followed EDClow-FNP > PEI-FNP > EDChigh-FNP, which were well correlated with the solubility study. Based on the Noyes-Whitney’s equation, an increase in the diﬀusion layer thickness hinders the dissolution rate. In PEI-FNP particles, the coated PEI AWD 131-138 thickened this layer, con-sequently reduced the α-mangostin release rate. In addition, the dis-solution test conditions such as sink or non-sink is crucial for the par-ticle release pattern. In the sink condition, the particle dissolution is rapid, thus covering the diﬀerences between samples and prevents the interpretation of dissolution profiles . On the other hand, in the non-sink condition, the dissolution rate is slower due to the decrease in drug solubility, consequently allowing the clear discrimination between formulations .
3.6. Impact of intravenous diluent on the FNP properties
The final products of our study are lyophilized powders of α-man-gostin loaded FNPs for intravenous injection propose. Before injection, the powders need to be reconstituted in intravenous diluent such as dextrose 5% and NSS. Therefore, the stability of α-mangostin loaded FNPs in these two diluents was studied. After dispersion in NSS, all formulas showed particle aggregation with the surface charge of 0 mV in 2 h. However, in dextrose 5%, the particles were stable up to 24 h (Fig. 4A). This phenomenon could be explained by the diﬀerence in ionic strength between the two media. The NSS has an ionic strength of 0.15 M, whereas dextrose 5% is non-ionization nature. Based on the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, there are two main forces that control the interaction energy between two colloidal particles in a solution, including the attractive van der Waals force and the repulsive surface electrical force . Mathematically, this re-pulsive force originated from the electrical double layer interaction energy between two particles, which is dependent on the electrolyte concentration through the Debye constant κ (note that κ−1 is called the Debye length or the double layer thickness). Accordingly, an increase in solution ionic strength leads to an increase in the Debye constant, which in turn exponentially reduces the double layer energy, ultimately decreases the repulsive force . As a consequence, the FNPs get aggregate due to the dominance of the attractive van der Waals force. Therefore, dextrose 5% is considered a suitable infusion medium for re-
In addition, the α-mangostin loaded FNPs dissolution profiles in dextrose 5% are also demonstrated in Fig. 4B. The drug release pattern was consistent between formulations, which followed the order of EDClow-FNP > PEI-FNP > EDChigh-FNP. These results were in agree-ment to the dissolution profiles in HEPES buﬀer. Within 24 h, 15–30% α-mangostin was released into the diluents, depending on the for-mulation.