The mouse melanoma (B16/F10), human colon carcinoma cells (HCT-116), human gastric adenocarcinoma cells (AGS) and human skin fibroblasts (Detroit 551) were from American Type Cell Culture (ATCC)/(Bethesda, MD, USA). Dulbecco’s Modified Eagle’s medium (DMEM) was provided by Gibco (Auckland, New Zealand), Eagle’s Minimum Essential Medium (EMEM), McCoy’s 5A medium and Ham’s F-12K medium were from ATCC. Oroxylin A was supplied by MedChem Express (Monmouth Junction, NJ, USA). The caspase-9 activity assay kit was from R&D Systems (Minneapolis, MN, USA). Poly (D,L-Lactide-co-glycolide) (lactide:glycolide) (1:1) ester terminated (MW 24,000 - 38,000) (PLGA) and Solutol^®/Kolliphor HS15 (Macrogol) (15)-hydroxystearate, and all other reagents were obtained from Sigma-Aldrich (St Louis, MO, USA). Information of the properties of (PLGA) may be found in Rezvantalab et al. [10]. Information of the properties of Solutol^® HS15 is described in a review by Ali et al. [11].

The Oroxylin A-PLGA at 4.26 µg of Oroxylin A/mL in 1% PLGA and the Oroxylin A-Solutol at 1.99 µg of Oroxylin A/mL in 0.1% Solutol were prepared by Dr. Sean Mackay, University of Otago, Dunedin, New Zealand. Also, the preparation of Oroxylin A in whey protein-derived nanofibrils was donated by Hi-Aspect Ltd., Wellington, New Zealand.

Mouse melanoma cells (B16/F10) were cultured in DMEM, human colon carcinoma cells (HCT-116) in McCoy’s 5A medium, human gastric adenocarcinoma cells (AGS) in Hams F-12K medium and human skin fibroblasts in EMEM. All cultures were supplemented with 10% foetal bovine serum and penicillin (100 U/mL)/streptomycin (100 µg/mL). The test samples were added to the cells (5 × 10⁴ cells/mL), which were incubated using 96-well flat-bottomed multi-well plates for 24 hours at 37˚C in a 5% CO₂/95% air atmosphere. Each sample was assayed at least in triplicate. The carrier used was 1.5% dimethyl sulphoxide in Hank’s Balanced Salt Solution. Cells were then lysed and the cell concentrations were determined at 570 nm following reaction with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) [12]. Caspase-9 activity was measured using an assay kit supplied by R&D Systems. The caspase activity per cell in each culture was also determined by dividing the caspase activity by the absorbance value from the MTT assay.

Oroxylin A-PLGA nanoparticles were prepared using a double emulsion method (Figure 1) similar to that described by Ramalho et al. [13]. Initially, as proof of concept for the double emulsion method, a hydrophobic organic dye, 1,1’-dioctadecyl-3,3,3’,3’-tetramethylindocarboryamine perchlorate, was used as a substitute to Oroxylin A for PLGA encapsulation. Using this model, it was established that PLGA nanoparticles could be repeatedly formed using the double emulsion method, whereby insoluble 1,1’-dioctadecyl-3,3,3’,3’-trimethylindocarboryamine perchlorate could be suspended in phosphate-buffered saline (pH 7.4) to produce a pink suspension.

Figure 1. Schematic illustration depicting the preparation of PLGA nanoparticles encapsulating Oroxylin A.

Following this successful demonstration, we undertook the encapsulation of the desired agent, Oroxylin A. PLGA (100 mg) and Oroxylin A (approximately 1 mg) were dissolved in acetone (10 mL), while a 1% aqueous solution of Pluronic F127^® (1%, 200 mg in 20 mL) was prepared. The organic phase was added dropwise to the aqueous phase with continuous stirring (~1000 rpm). The solution was continuously stirred for a further 6 hours under a light air flow to allow the acetone to evaporate. Control PLGA nanoparticles were prepared in an identical manner, but without the inclusion of Oroxylin A. The PLGA suspensions were subsequently purified by centrifugation at 2500 rpm for 5 minutes, followed by 3500 rpm for 5 minutes. The pelleted material (larger micron-sized PLGA/Oroxylin A) was discarded. The remaining suspension was then dialysed against phosphate-buffered saline for 24 hours to remove any excess Pluronic F127^®.

A spectrophotometric assay for the sensitive detection and quantification of Oroxylin A was developed based on the strong absorbance of Oroxylin A observed at 374 nm under alkaline conditions, which is due to the deprotonation of Oroxylin A to form the Oroxylin A anion (Figure 2).

Figure 2. Reaction scheme for the deprotonation of Oroxylin A under alkaline conditions.

To determine the concentration of Oroxylin A in the PLGA nanoparticle suspension, a standard curve of Oroxylin A was prepared by dissolving Oroxylin A in a mixture of acetone and pH 10 sodium borate buffer (1:1 acetone: buffer; 100 mM sodium borate adjusted with NaOH; 1 mM Oroxylin A). The stock solution was serially diluted to produce a concentration range between 0 and 0.33 mM and the absorbance was measured at 374 nm (Table 1, Figure 3).

Characterisation of the PLGA nanoparticle suspensions was performed primarily by Dynamic Light Scattering (DLS) (Figures 4-7). DLS was performed on all suspensions. Control PLGA nanoparticles displayed a hydrodynamic diameter of approximately 150 nm with excellent polydispersity (Figure 4). PLGA nanoparticles encapsulating Oroxylin A tended to have a slightly greater hydrodynamic diameter of approximately 160 nm with good polydispersity (Figure 5). The potential

Table 1. Absorbance of Oroxylin A at 374 nm under alkaline conditions (~pH 9) with increasing Oroxylin A concentration.

Figure 3. Standard curve of Oroxylin A alkaline conditions (pH ~9) with absorbance measured at 374 nm.

for the PLGA/Oroxylin A nanoparticle suspensions to be lyophilised and stored as a dry powder for re-suspension when required was also investigated. After lyophilisation, PLGA nanoparticles containing Oroxylin A could be easily re-suspended in deionized water or phosphate-buffered saline (pH 7.4) with only a small degree of aggregation observed by DLS (Figure 6). It was noted however, that re-suspending the PLGA nanoparticles containing Oroxylin A by ultrasonication (bath ultrasonicator, 40 kHz, 100 W, 60 seconds) caused significant aggregation by DLS (Figure 7). Therefore, PLGA nanoparticles containing Oroxylin A can be lyophilised and stored as a dry powder but are not recommended for re-suspension to be facilitated by ultrasonication.

The absorbance of control and Oroxylin A PLGA nanoparticles was measured under identical conditions by diluting (1:10) an aliquot of nanoparticle suspension with acetone:borate buffer (pH 10; 1:1 acetone:buffer; 100 mM sodium borate) and measuring absorbance. Under these conditions, it was observed that the cloudy appearance of the nanoparticle suspension clarified, suggesting either that PLGA is more soluble under these conditions, or that the PLGA is hydrolysed into lactic acid and glycolic acid. Absorbance of control PLGA nanoparticles (no Oroxylin A) had little absorbance at 374 nm (0.07, within error of baseline), while the absorbance of the Oroxylin A PLGA nanoparticle suspension corresponded to an Oroxylin Aconcentration of approximately 60 μM (A = 0.124; c = ~0.063 mM; batch variation ~0.057 - ~0.063 mM). Increasing the concentration of Oroxylin A used during the manufacturing process produced little benefit to the final Oroxylin A concentration in suspension using this method. To maximise the concentration of Oroxylin A, the PLGA/Oroxylin A nanoparticle suspension was concentrated by lyophilisation resulting in an increased Oroxylin A concentration of approximately 150 μM.

Attempts to solubilise Oroxylin A as a microemulsion were made using Kolliphor HS15 (Solutol). A solid quantity of Oroxylin A was dispersed in 1% Solutol via ultrasonication (60 minutes) and stirred continuously for 15 hours at 40˚C. The maximum concentration of Oroxylin A achieved using a 1% Solutol emulsion was approximately70 μM.

Attempts to produce a liposome suspension of Oroxylin A were also made. Oroxylin A and soy phosphatidylcholine were dissolved in chloroform. The chloroform was removed under vacuum and the residual phospholipid/Oroxylin A film was rehydrated with pH 7.4 PBS to produce a final concentration of 1 mM Oroxylin and 10 mM soy phosphatidylcholine in suspension. The lipid suspension was subsequently extruded 15 times through 1000 nm followed by 100 nm polycarbonate membranes at 60˚C. The liposome suspension was then purified via dialysis against pH 7.4 PBS for 12 hours, during which the liposome suspension aggregated from solution. The process was repeated unsuccessfully using 0.5 mM Oroxylin A to 10 mM soy phosphatidylcholine.

Cell viability was measured by the MTT assay. The MTT assay is a colorimetric assay that detects the conversion of the yellow tetrazolium salt, MTT, to a purple formazan product produced by active mitochondria in living cells.

Poly (lactic-co-glycolic acid) (PLGA) was used as the platform delivery system for the encapsulation of Oroxylin A due to its widespread use in pharmaceutical science for the formulation of poorly water-soluble drugs [15][16]. The metabolic products of PLGA produce lactic acid and glycolic acid, which are subsequently eliminated as carbon dioxide and water via the Krebs cycle. Therefore, PLGA nanoparticles are biodegradable, biocompatible, and approved by the FDA as a pharmaceutical excipient.

Poly (lactic-co-glycolic acid) (PLGA) is a copolymer of lactic acid and glycolic acid [10]. Solutol HS 15 (Solutol) is a non-ionic surfactant and is composed of polyglycol mono- and di-esters of 12-hydroxystearic acid and free polyethylene glycol [11]. There is little evidence that PLGA or Solutol nanoparticles have intrinsic anti-cancer effect.

PLGA decomposes into H₂O and CO₂ that are eliminated in the body. Its polymeric nanoparticles degrade in vivo through hydrolysis of the ester bonds to its monomeric anions (lactate and glycolate). While D-Lactate is not further metabolised before excretion, L-Lactate is converted into CO₂, which is excreted through the lungs and is converted to pyruvate, which then enters the Krebs cycle. Glycolate is either directly excreted through the renal system or it can be oxidised to glyoxylate, which is subsequently converted into glycine, serine and pyruvate. The latter can again enter the Krebs cycle and be metabolised into CO₂ and H₂O [10].

PLGA-based nanoparticles encapsulating anti-cancer drugs have been widely used [10]. Notwithstanding the above, Solutol is known to reverse multi-drug resistance of cancer cells [17]. Also, Solutol-based nanoparticles encapsulating anti-cancer drugs have been used in several studies [18]-[22]. In these studies, Solutol is mixed with other polymers to form mixed micelles. Despite the limited evidence of Solutol alone affecting cancer cells, we found in this study that Solutol alone does decrease the viability of cancer cells (Table 5).

Figure 8. Some metabolic changes due to overexpression of UCP₂ in cancer cells. These changes include decrease in ROS, increase in COX2 activity and increase in PGE₂ levels.

Table 6.Oroxylin A reverses these changes. Hence, Oroxylin A inhibits UCP₂ expression, stimulates ROS, inhibits COX2 expression and lowers PGE₂ levels in cancer cells.

In this study, we observed that melanoma cell viability is not affected following administration of Oroxylin A to cancer cells. However, Zhu et al. [23] observed decreased cell viability of squamous cell carcinoma cells following Oroxylin A treatment after a lengthy incubation of 48 hours. Also, Zhao et al. [24] and Ha et al. [25] demonstrated decreased cell viability of human hepatocellular carcinoma cells and human colon cancer cells respectively following administration of Oroxylin A. However, our studies showed decreased cell viability following administration of Oroxylin A encapsulated nanoparticles to melanoma cells. We did not observe decreased cell viability following administration of non-encapsulated Oroxylin A (Table 2). The difference of our findings for Oroxylin A treatment of cancer calls and those of Zhu et al. [23], Zhao et al. [24], Ha et al. [25] may be due to the fact that different cancer cells have been used in each of these studies. We showed increased apoptosis following treatment of cancer cells with Oroxylin A encapsulated with Solutol but not with PLGA. Interestingly, Zhu et al. [23] showed Oroxylin A-loaded PDots caused increased apoptosis in squamous cells carcinoma.

Oroxylin A is an inhibitor of UCP₂ [7]. We consider that the effects outlined above of Oroxylin A encapsulated nanoparticles on cancer cells are due, at least in part, to the inhibition of UCP₂ by Oroxylin A. UCP₂ converts the carbohydrate metabolism of cancer cells from mitochondrial oxidative phosphorylation to aerobic glycolysis. It does this by uncoupling mitochondrial oxidative phosphorylation. Baffy et al. [8] have suggested that recoupling mitochondrial oxidative phosphorylation would be a novel way to treat cancer. As Oroxylin A is an inhibitor of UCP₂, this agent would be a suitable candidate to cause the recoupling of mitochondrial oxidative phosphorylation.

Recently, Zhu et al. [23] investigated the effect of Oroxylin A-Loaded PDots on squamous cell carcinoma cells. Both Oroxylin A alone and Oroxylin A-Loaded PDots inhibited cell growth, however Oroxylin A-Loaded PDots were more potent. Zhu et al. [23] also found that Oroxylin A-Loaded PDots increased apoptosis and that Oroxylin A and Oroxylin A-loaded PDots were effective in decreasing size and weight of implanted tumours in mice.

Zhao et al. [24] studied synergism of Oroxylin A and fluorouracil on liver cancer viability and observed that Oroxylin A alone inhibited cell growth, increased apoptosis and reduced growth of implanted tumours. Similarly, Ha et al. [25] carried out parallel studies on the synergism of Oroxylin A and 5-fluorouracil on colon cancer cells. They found Oroxylin A inhibited cell growth, enhanced apoptosis and reduced growth of implanted tumours. When combined with 5-fluorouracil, Oroxylin A was more effective than Oroxylin A alone in decreasing weight and size of implanted tumours [24][25]. Other studies have shown that other inhibitors of UCP₂, when combined with anti-cancer agents, are more effective in decreasing cell viability and increasing ROS production and apoptosis than the inhibitors alone [2] (reference therein).

In non-small cell lung cancer, Oroxylin A reverses hypoxia-induced cisplatin resistance [26]. The combination of Oroxylin A and gefitinib in A549 cells instigated more apoptotic cells than other studied combinations and gefitinib alone [27].

The effects of encapsulated Oroxylin A nanoparticles and Oroxylin A PDots on cancer cells have been demonstrated in cancer cells in culture. Maeda [9] has shown that certain high molecular weight substances are more effective in the treatment of whole tumours (Enhanced Permeability and Retention Effect—EPR Effect). Although the studies carried out by us have been on cultured cancer cells, Zhu et al. [23] have shown that Oroxylin A-Loaded PDots do shrink tumour size and weight. Despite our studies with cultured cancer cells, it is interesting that we only achieved a decrease in cell viability and an increase in apoptosis using Oroxylin A nanoparticles and not using Oroxylin A alone.

In this work, we treated cancer cells with PLGA or Solutol alone. We found that Solutol decreased cancer cell viability, but PLGA did not. It appears that the effect of Solutol alone may be due to its properties as a detergent.

Lawson et al. [28], Lawson et al. [29] and Stefanadis et al. [30][31] have shown that there is an increase in the temperature of cancers compared to normal tissue. This temperature increase has been attributed to increased heat production by these tumours [32]. UCP₂ has been shown to release heat [8] and so its overexpression in cancer is at least one source of the extra heat production in cancer [32]. Conversely, inhibition of UCP₂ by Oroxylin A will reduce the heat production in cancers. The heat production in cancer also results in increased entropy production in these tumours. Increased entropy production is a feature of a departure from the steady state, where entropy production is a minimum. Growing tumors are a departure from the steady state of normal tissue [32].

Tuli et al. [6] have reviewed the anti-cancer potential of Oroxylin A and noted that apoptosis is lacking in cancer cells, resulting in immortal malignant cells. They then proceed to cite numerous studies showing that Oroxylin A enhances apoptosis in cancer cells. Consistent with these findings, our studies have shown that Oroxylin A encapsulated in Solutol enhances apoptosis as shown by increased caspase-9 activity.