Background

Curcumin, chemically known as diferuloyl methane, is a hydrophobic polyphenol derivedfrom the rhizome of the plant Curcuma longa (turmeric) of the Zingiberaceaefamily. Curcumin is known to suppress multiple signaling pathways and inhibit cellproliferation, invasion, metastasis and angiogenesis [1]. Its wide medical use includes anti-septic, analgesic, anti-inflammatory,anti-oxidant, anti-malarial and wound-healing [2]. In recent years, a particular interest was shown on the anti-oxidativeand anti-inflammatory properties of curcumin which might provide a therapeuticwindow for cancer treatment [3].

Curcumin is a yellow-colored tautomeric compound that is quite soluble in organicsolvents such as dimethoxy sulfoxide (DMSO), ethanol, methanol, chloroform oracetone. Upon dissolution in an organic solvent, curcumin absorbs light in thevisible wavelength range [4]. Turmeric contains three major analogues: curcumin, demethoxycurcumin(DMC), and bisdemethoxycurcumin (BDMC) and recently identified cyclocurcumin in lesssignificant amounts [5]. Commercially available curcumin mixture contains approximately 77%curcumin, 17% DMC and 3% BDMC as major components [6]. Although all three are highly active, curcumin is more efficient thanDMC and BDMC on various cell models [6, 7]. Contrary to these findings, studies on preclinical models ofcarcinogenesis have demonstrated that commercial grade curcumin – turmeric asa mixture - has the same inhibitory effect as pure curcumin [8, 9].

Pharmacologically regarded as safe, curcumin is nontoxic, even at relatively highdoses [10] such as 8 g per day [11]. As demonstrated recently, tumor cells are more sensitive to thecytotoxic activity of curcumin than normal cells [12]. In line with another study, the cellular uptake of curcumin was found tobe significantly higher in tumor cells compared to normal cells, which wasattributed to the differentiated membrane structure, protein composition and biggersize [13]. The lower uptake rate may explain the low toxicity of curcumin forhealthy cells.

The wide spectrum of pharmacological properties of curcumin is attributed to itsnumerous effects on several targets including transcription factors, growthregulators, adhesion molecules, apoptotic genes, angiogenesis regulators, andcellular signaling molecules [14]. Curcumin exerts anti-cancer activity mainly through blocking cell cycleprogression and triggering tumor cell apoptosis [15]. All three stages of carcinogenesis including initiation, promotion andprogression are suppressed by curcumin [16]. This is probably due to inhibition of the nuclear factor κB, whichplays a central role in regulating the expression of various genes involved in cellsurvival, apoptosis, carcinogenesis and inflammation. This efficacy makes curcuminto a potential therapeutic target [17]. Furthermore, curcumin affects various cell cycle proteins andcheckpoints involving downregulation of some of the cyclins and cyclin-dependentkinases (cdk), upregulation of cdk inhibitors, and inhibition of DNA synthesis [18]. However, the physiological response triggered by curcumin depends on thecell type, the concentration of curcumin (IC50: 2-40 μg/ml) andthe time of treatment [19]. For instance, curcumin treatment was reported to arrest cell growth atG2/M phase and induce apoptosis in human hepatoma cell line HepG2 [20, 21], whereas G0/G1 as well as G1/S phase arrests were reported for variousother cell lines [18].

Clinical use of curcumin remains very limited due to its extremely poor watersolubility (≈11 ng/ml) [22], and low bioavailability following oral administration [23]. Even when 10-12 g/ml of curcumin was administered orally in humans,curcumin levels in serum remained approximately at 50 ng/ml [24]. Several studies demonstrated that 10-50 μM(3.7-18.4 μg/ml) curcumin induces cell death primarily through apoptosis [25, 26]. However, the important question to be addressed is how to bring curcuminat these micromolar concentrations to the site of tumors while curcumin possessessuch a low bioavailability. Addressing this problem, targeted and triggered drugdelivery systems accompanied by nanoparticle technology have emerged as prominentsolutions [23]. Likewise, this study introduces emulsomes as a promising nanocarriersystem suitable for the delivery of curcumin.

Emulsomes are biocompatible vesicular systems comprising of a solid fat coresurrounded by phospholipid multi-layers (Figure 1) [27]. Due to the solid core, emulsomes can entrap higher amounts of lipophilicdrug compounds with a prolonged release time compared to emulsion formulationspossessing a liquid core [27–29]. Composed of fat and lipids, emulsomes are biocompatible. Thesecharacteristic properties make emulsomes to promising candidates for poorlywater-soluble therapeutic agents such as curcumin.

Figure 1
figure 1

Schematic drawing of CurcuEmulsome. CurcuEmulsome is composed of asolid tripalmitin core surrounded by phospholipid multi-layers. Thelipophilic load, i.e. curcumin, can locate itself in the inner core, as wellas inside the phospholipid layers of the nanocarrier.

Full size image

As recently demonstrated, the assembly of phospholipids and triglycerides to form astable dispersed emulsomes can be prepared by a dehydration-rehydration processfollowed by temperature-controlled extrusion [30]. In the present study, curcumin-emulsome nanoformulations, or so-calledCurcuEmulsomes, were formulated using the same methodology, and characterized withrespect to their structural and biophysical properties. HepG2 cell line was used asthe in vitro cellular model to study cellular uptake of CurcuEmulsomes andto evaluate the biological effects of the incorporated curcumin on cellularmorphology, as well as viability, compared to its free form. Cell cycle studies wereperformed to study CurcuEmulsome's effect on cell proliferation and implicitly toverify the incident of prolonged release of curcumin in biological level.

Results

CurcuEmulsome nanoformulation

The structural design of CurcuEmulsomes enables curcumin to be localized in theinner solid tripalmitin core, as well as inside the phospholipid layerssurrounding and stabilizing the nanocarrier (Figure 1). In contrast to free curcumin, poorly soluble in water(Figure 2A), curcumin incorporated intoCurcuEmulsomes is a colloidal solution (Figure 2B).Forming an intensive turbid suspension, CurcuEmulsomes achieved curcuminconcentrations up to 0.11 mg/ml (in range of 0.07-0.11 mg/ml) asestimated by absorbance measurements, where the latter value corresponds to a10,000-fold increase in solubility of curcumin (i.e. ≈11 ng/ml [22]).

Figure 2
figure 2

Enhanced solubility. (A) Free curcumin is poorly soluble in waterand macroscopic flakes are visible especially on the top and the bottomof the bottle. (B) Curcumin incorporated into CurcuEmulsomes, incontrast, is homogeneously dispersed in water.

Full size image

The aforementioned values correspond to the concentrations of curcuminincorporated into CurcuEmulsomes, as unincorporated curcumin in the solution wasalready spin down after a centrifugation process. Spin-down resulted in recoveryof incorporated curcumin – corresponding to 90% of total - within thesupernatant containing CurcuEmulsomes, indicating that a stable incorporationwas achieved.

Particular characterization of CurcuEmulsomes

TEM micrographs verified that CurcuEmulsomes are spherical in shape, and hence,similar in size and morphology to empty emulsomes (Figure 3A). DLS analysis with Zetasizer Nano ZS (Malvern Instruments Ltd,UK) determined the average diameter of ten distinct CurcuEmulsome formulationsas 286 ± 27 nm (polydispersity index of 0.34; conductivityof 0.17 ± 0.01 mS/cm) - where the plus-minus sign indicates themargin of average size of numerous CurcuEmulsome formulations made of the samecomposition. Consistent with this value, the mean diameter of empty emulsomeswas previously reported to be 297 ± 28 nm [30]. In addition, zeta potential of CurcuEmulsomes(37 ± 8 mV) is comparable to that of empty emulsomes(37 ± 7 mV) [30]. With the aid of the auto-fluorescence properties of curcumin, it waspossible to evidence the incorporation of curcumin into emulsomes and that theprepared nanocarrier system is a stable dispersed formulation in water(Figure 3B). Consequently, the presented datadeclines any significant influence of incorporated drug neither on the size noron the surface potential of the nanocarrier, which could further affect theparticular dispersity of the nanocarrier in water.

Figure 3
figure 3

Particular analysis of CurcuEmulsomes. A) Transmission electronmicrograph demonstrates CurcuEmulsomes are spherical in shape and have adiameter in range of 50–350 nm. B) Fluorescencemicroscopy image does not only confirm the incorporation of curcumininto CurcuEmulsomes, but also verifies the high-level dispersity ofCurcuEmulsomes in water. Bars correspond to 500 nm and2 μm, respectively.

Full size image

Tautomeric curcumin incorporated into CurcuEmulsomes in its enol form

Curcumin is a yellow-colored tautomeric compound that, upon dissolution in anorganic solvent, absorbs light in the visible wavelength range [4]. In nonpolar, i.e. aprotic solvents such as chloroform, the spectrumdisplays vibronic structure with λmax near 420 nm. Thisfeature corresponds to the fully conjugated form of the protonated enol [31]. In polar protic solvents such as DMSO, the vibronic features are nolonger resolved, and hence, the molar absorptivity decreases as solvent polarityincreases resulting in λmax shifts to nearly 430 nm [31]. In agreement with this [4, 32, 33], the UV–vis spectrum of CurcuEmulsomes displayed the sameλmax as curcumin in chloroform (420 nm), and differedfrom λmax of curcumin dissolved in DMSO (Figure 4A). Hence, curcumin incorporated in CurcuEmulsomes isevidently in its fully conjugated protonated enol form.

Figure 4
figure 4

Spectral analysis of CurcuEmulsomes. (A) UV–vis absorbancespectra (excitation wavelength: 420 nm) and (B) emissionspectra of curcumin in various forms: (i) curcumin in DMSO (red), (ii)curcumin in chloroform (green), (iii) curcuemulsomes in water (1:100diluted) (yellow), and (iv) curcumin in water (blue).

Full size image

Like the absorbance spectrum, the emission spectrum of CurcuEmulsomes pursuedthat of curcumin in chloroform and showed a λmax at 500 nm(Figure 4B). Excitated at 420 nm, freecurcumin in DMSO showed an emission peak centered at 520 nm and curcumin inwater did not fluoresce.

Curcumin composition inside CurcuEmulsomes

Since turmeric as a mixture was demonstrated to have the same inhibitory effectas pure curcumin [8, 9], curcumin was used as purchased without any further purification.Therefore, the turmeric fed to the system contained all three analogues, i.e.curcumin, DMC and BDMC. HPLC analysis showed that the turmeric extract consistedof 78.1% curcumin, 17.7% DMC and 4.1% BDMC (Figure 5A), whereas CurcuEmulsomes comprised of 40.8% curcumin, 40.3% DMC and16.8% BDMC (Figure 5B). As curcumin analogues werethe only substances in CurcuEmulsomes raising a peak at 420 nm, emptyemulsomes did not show any peak in HPLC analysis (Figure 5C).

Figure 5
figure 5

Compositional analysis of CurcuEmulsomes. HPLC chromatograms showthe compositional distribution of curcumin and its analogues in(A) turmeric extract, (B) CurcuEmulsomes and(C) empty emulsomes. The HPLC data discloses that thecomposition of turmeric inside the nanocarrier is different than the onefed to the system. The plane curve of empty emulsomes declinesoccurrence of any interference of lipid composition of CurcuEmulsomes tothe analysis.

Full size image

Effect of CurcuEmulsomes on HepG2 cell viability

Previous studies demonstrated that 10–50 μM curcumin induces celldeath primarily through apoptosis [25]. Within this range, HepG2 cells were treated with CurcuEmulsomes andfree curcumin (in DMSO) of the same concentrations, respectively. Aftertreatment for 6, 24 and 48 hours, the cell viability was determined withCellTiter-Blue assay. As shown in Figure 6,CurcuEmulsomes showed no significant cytotoxicity until 24 hours, incontrast to free curcumin which demonstrated significant toxicity especially inthe early stage, i.e. after 6 hours. Nonetheless, on the long terms,incorporated curcumin preserved its biological activity, and thus, acted asefficient as free curcumin. Accordingly, after 48 hours 30 μMCurcuEmulsome lowered the viability of HepG2 to approximately 70%,40 μM CurcuEmulsome to approximately 50%, same percentages as observedwith free curcumin (Figure 6). In contrary, emptyemulsomes showed no significant effect on HepG2 cell viability.

Figure 6
figure 6

Cell viability profile of HepG2 cells treated with curcumin andCurcuEmulsomes. Cytotoxicity of CurcuEmulsomes, as well as freecurcumin (in DMSO) and empty emulsomes, to HepG2 cells were investigatedat various concentrations for 6, 24 and 48 hours compared tountreated cells. Cell viabilities are given in percentages relative tountreated cells. n.s., not significant; **comparable after48 hours.

Full size image

It is also important to mention that the viabilities recorded over 100%(Figure 6) might be due to the physicalinterference of the CurcuEmulsomes (not supported by our data since also thecells treated with curcumin at low doses showed the same response), as well asdue to the changes in cellular activities involved in redox reactions inresponse to curcumin and CurcuEmulsomes, as CellTiter-Blue is a fluorescentassay used to measure cell viability via non-specific redox enzyme activity [34]. Therefore, although the latter hypothesis is likely to be the case,the complete clarification merits further study.

Considering interference with cellular adhesion, curcumin and CurcuEmulsomescaused also morphological changes in HepG2 cells. Cells treated with freecurcumin and CurcuEmulsomes showed a round shape whereas untreated cellspreserved their flattened morphology (Figure 7).

Figure 7
figure 7

Effect of CurcuEmulsome on cell morphology. (A) Untreated HepG2cells preserved their flattened morphology throughout the study, whereascells treated with (B) curcumin or (C) CurcuEmulsomes(both 30 μM) showed a round shape after 24 hours ofexposure. Bars correspond to 50 μm.

Full size image

Uptake of CurcuEmulsomes by HepG2 cells

The uptake of CurcuEmulsomes in HepG2 cells could be evaluated by fluorescencemicroscopy analysis by the auto-fluorescence of curcumin (Figure 8). As previously reported [19], the cellular uptake was observed to be concentration-dependent aseach increase in concentration from 10 μM to 50 μM resultedin an increase in fluorescence intensity inside the cell (data not shown). Alongthe time of treatment, fluorescence microscopy analyses were performedsequentially after 6, 24 and 48 hours and information was collectedregarding the stepwise uptake mechanism and localization of curcumin andCurcuEmulsomes in HepG2. Accordingly, the fluorescence signal was limited to thecellular membrane for the first 6 hours, and widen to the innercompartments of the cells after 24 hours (Figure 8A). In agreement with Kunwar et al. (2008) [13], curcumin primarily localized in the cell membrane and subsequentlyaround the nucleus, most likely due to their compartmental lipophilicproperties. Moreover, in agreement with Mohanty et al. (2010) [26], cells treated with free curcumin showed the maximal fluorescenceintensity at 24 hours, which faded down significantly with time(Figure 8A). On the contrary, cells treated withCurcuEmulsomes did not exhibit any deterioration in the level of fluorescenceintensity neither after 24 nor 48 hours. This was attributed to theenhanced stability as well as to the gradual release of curcumin incorporatedinto the solid tripalmitin core of the nanocarrier. Hence, encapsulated curcuminremained protected from hydrolysis, and upon release, its biological activitypersisted alongside its fluorescence intensity for a longer period of time thanfree curcumin.

Figure 8
figure 8

Cellular uptake of CurcuEmulsomes by HepG2 cells. (A) Cells weretreated with 30 μM curcumin dissolved in DMSO, or incorporatedinto CurcuEmulsomes in 10% FCS MEM medium for 6, 24 and 48 hours.(Bars correspond to 50 μm). (B) Inset image of cellstreated with CurcuEmulsomes (30 μM) for 48 h. Arrowsindicate CurcuEmulsome accumulations upon cellular uptake.“N” indicates the cell nucleus. (Bar corresponds to10 μm).

Full size image

Previous thin-sectioning analysis of HepG2 cells treated with empty emulsomesdemonstrated that emulsomes are internalized in the cell within endosomes [30], resulting in an accumulation of the nanocarrier inside the cellbefore any sufficient release of the load could occur. Confirming this, thepresent data verified accumulation of CurcuEmulsomes inside the cytoplasm.Highly fluorescent spherical regions were discovered inside the cells treatedwith CurcuEmulsomes, which are ascribed to endosomes internalizing thenanocarriers. As indicated by arrows (Figure 8B),these regions were only detected for the cells exposed to CurcuEmulsomes for 24and 48 hours. This finding may explain why CurcuEmulsome causedcytotoxicity first after 24 hours.

Effect of CurcuEmulsomes on cell cycle

To explore the physiological effect of CurcuEmulsomes on cell proliferation, cellcycle analyses were performed on stable HepG2 cells with and without freecurcumin or CurcuEmulsomes. Flow cytometry analysis demonstrated that HepG2cells exposed to free curcumin (40 μM) for 24 hours weredifferentiated from untreated ones with a higher populations in the G2/M phase(35% instead of 18%) and with fewer fractions in the G0/G1 phase (55% instead of71%; Figure 9). Compared to the control, this resultsuggested that curcumin inhibited the growth of HepG2 by causing cell-cyclearrest in the G2/M phase. Remarkably, G2/M phase arrest declined after reachinga peak at 24 hours indicating that thereafter free curcumin lost itsactivity and cells started recovery. On the contrary, CurcuEmulsome treatment at40 μM resulted in a steady increase of cell population in G2/M phasefrom 19% to 22% and then to 26%, as population in G0/G1 phase decreases from 69%to 66% and then to 64%, from 6 to 24 hours and subsequently to48 hours, respectively. At 48 hours, the cell cycle profiles of cellstreated with curcumin and CurcuEmulsomes became comparable: around 26% of thecells in G2/M and 65% in G0/G1 phase (Figure 9). Cellcycle profiles of untreated cells remained unaltered throughout the experiment.Concisely, like free curcumin, CurcuEmulsome induced G2/M cell cycle arrest onHepG2 cells, but this was prolonged probably since curcumin was released insidethe cell gradually over time.

Figure 9
figure 9

Flow cytometric DNA histograms of HepG2 cells. Effect of freecurcumin and CurcuEmulsome on cell population of HepG2 cells wasinvestigated for 6, 24 and 48 hours after exposure. Like freecurcumin, CurcuEmulsomes induce G2/M cell cycle arrest. However, theexpansion in population of cell at G2/M phase occurs after48 hours, rather than 24 hours as for free curcumin. Thisdelay is attributed to the gradual release of curcumin from the solidtripalmitin core of the nanocarrier inside the cell. Cells are labeledwith DAPI for detection with Pacific Blue.

Full size image

Effect of CurcuEmulsomes on apoptosis

The apoptosis response of HepG2 to CurcuEmulsomes and free curcumin was analyzedby a Caspase 3/7 activity assay in which higher fluorescence intensitiescorrespond to higher level of apoptosis. Like free curcumin, CurcuEmulsomescaused a concentration-dependent increase in apoptosis with comparable apoptoticactivities at 24 and 48 hours (Figure 10).These results strongly suggested that the cytotoxicity of CurcuEmulsomes can beattributed to the induction of apoptosis and G2/M phase cell cycle arrest.

Figure 10
figure 10

Apoptotic activities of HepG2 cells treated with CurcuEmulsomes.Apoptosis responses of HepG2 cells to free curcumin and CurcuEmulsomeswere analyzed by a Caspase 3/7 activity assay. The level of fluorescenceintensity corresponds to the level of apoptosis.

Full size image

Discussion

The results of this study indicate that CurcuEmulsomes can successfully entrapcurcumin inside the inner solid matrix composed of tripalmitin surrounded byphospholipids. The stable formulations are spherical in shape and preserve thesurface characteristics of the nanocarrier. Most important, the solubility ofcurcumin is increased up to 0.11 mg/ml by means of CurcuEmulsomes,corresponding to an improvement in solubility by 10,000 times. Thus CurcuEmulsomescan achieve the effective concentrations of curcumin (i.e. 10–50 μM) [25, 26], and facilitate the delivery of bioactive molecules into the cell invitro.

In the literature, various encapsulation approaches like diblock copolymers [35, 36], hydrophobically modified starch [32], beta-casein micelles [37], lipid nanoemulsions [38], curcumin-rubusoside complexes [39], cyclodextrin assemblies [40, 41], liposomes [42], curcumin-nanodisk [4] and polymeric NanoCurc™ formulations [43] have been successfully applied to increase the solubility and thereby thedelivery of curcumin. Encapsulation of curcumin in a pluronic block copolymer showednot only anti-cancer activity comparable with free curcumin, but also demonstrated aslow and sustained release of curcumin [36]. Therefore, the aforementioned approaches, as well as CurcuEmulsomes,look promising to enable the effective use of curcumin in medical applications.

However, having partially the characteristics of both liposomes and emulsions,CurcuEmulsome approach possesses certain advantages over its alternatives. Likeliposomes, emulsomes are stabilized by phospholipid (multi-)layers as outermoststructure, and thus, there is no need for surfactants stabilizing thenanoformulation. This endows emulsomes high degree of biocompatibility attherapeutic applications. More detailed, in the absence of any synthetic surfactantssuch as poloxamers, polysorbates or doxycholate, the use of emulsomes as a drugdelivery system has demonstrable advantages, particularly for parenteraladministration of poorly water-soluble lipophilic drugs [27], such as curcumin. Alternatively, due to their colloidal nature,emulsomes can be passively taken up from the blood stream by macrophages of theliver and spleen after intravenous or intracardiac administration as demonstrated inearly in vivo studies [44, 45].

On the other hand, unlike lipid emulsions having a fluid core, emulsomes with a solidfat core can prolong the release of incorporated drugs - a property similar topolymeric nanoparticles [28, 46]. As previously demonstrated, zidovudine-emulsome formulations displayed aslow drug release profile in vivo (12–15% after 24 h) andprolonged the action at comparatively low drug doses [29]. Therefore, the developed CurcuEmulsomes would be expected not only tocircumvent the problems of low solubility and rapid elimination, but also to modifythe drug release profile thereafter, due to the presence of curcumin in the internalsolid lipid core.

Finally, having an analogous surface as liposomes [47], CurcuEmulsomes can further be tailored to fulfill specific requirementssuch as longer blood circulation or to enable cell targeting and active drugdelivery. For instance, Gill et al. (2011) coated emulsomes withO-palmitoyl amylopectin [48], whereas Pal et al. (2012) coated them with O-palmitoyl mannanboth with the aim of developing macrophage targeted systems [45]. In a recent study, we showed that crystalline bacterial cell surfacelayer (S-layer) proteins are capable to coat emulsomes and modify their entiresurface characteristics [30], e.g. by altering zeta-potential.

The colloidal characteristics of the emulsome evidence its robust character andindicate its potential in versatile use for lipophilic therapeutic agents other thancurcumin. As previously reported [28–30], the size of emulsomes (a mean diameter of 286 nm, Figure 3) is predominantly determined by the phospholipid totripalmitin ratio, and evidently, incorporation of curcumin did not influenceneither particle size nor zeta potential characteristics. Moreover, the particlesizes can be tuned by altering the phospholipid to solid lipid ratio [29].

Although curcumin, DMC and BDMC show only very small chemical modifications withrespect to their number of methoxy groups, a decrease in hydrophobicity in the orderof curcumin > DMC > BDMC is known [49]. Therefore, a shift in the ratio of the analogues inside the lipophilicfat core should be expected, but not in terms of a relative decrease of curcumincompared to DMC and BDMC (Figure 5). Hence, this resultcontradicts with the relative hydrophobicity of the analogues, as well as thefindings of Rungphanichkul et al. (2011), where encapsulation ofcurcuminoids in non-ionic surfactant based liposomes, so-called niosomes, favoredthe incorporation of curcumin rather than its analogues [50]. Although some thermodynamic parameters such as the polarity, as well asthe molecular electrostatic interactions of curcuminoids with charged groups oflipid compounds, such as hexadecylamine, are thought to play a role in thisselective incorporation process, the complete clarification of this finding meritsfurther study.

Biological efficacy of CurcuEmulsomes was studied in vitro on HepG2 cellline model. In line with earlier studies on emulsomes [29], the delay in cytotoxicity is attributed to the slow release of curcuminentrapped inside the solid core of emulsomes. Hence, on the short terms thecytotoxic effect of CurcuEmulsomes remains limited. Nevertheless, CurcuEmulsomesdisplayed prolonged biological activity and acted as efficiently as free curcumin onlong terms (Figure 6).

Like free curcumin, CurcuEmulsomes caused morphological changes in HepG2 cells wheretreated cells distinguished from untreated ones by their round shape. Based on AFMstudies, Jiang et al. (2012) demonstrated the effect of curcumin oncytoskeletal arrangement of HepG2 cells and, combined with flow cytometric analysis,correlated this morphological effect with the upregulated expression of tubulin [21]. The latter caused disorganization of the well-organized, filamentousnetwork of healthy cells as deduced from the adopted round shape. Therefore,delivering curcumin into the cell, CurcuEmulsomes must be initiating the same effect(Figure 7).

Indicating for an enhanced stability, fluorescence images demonstrated thatincorporated curcumin preserve its fluorescence intensity for longer times comparedto free curcumin (Figure 8A). Parallel to our previouscross-sectional analysis of cells treated with empty emulsomes [30], the fluorescence microscopic data verified the accumulation ofCurcuEmulsomes inside the cytoplasm upon their uptake by the cell (Figure 8B). Accordingly, CurcuEmulsomes accumulate inside the cellbefore any sufficient release of the load could occur. This finding may explain whyCurcuEmulsomes caused cytotoxicity only after 24 hours (Figure 6).

Cell cycle analysis demonstrated that CurcuEmulsomes cause a prolonged induction ofG2/M cell cycle arrest where the peak of G2/M phase rose steadily from 6 to48 hours (Figure 9). In the contrary, free curcuminresults in a sharp increase after 24 hours which declined after 48 hours.These findings, in line with cytotoxicity data, corroborate the slow and sustainedrelease of curcumin from CurcuEmulsomes into the cells. Cell cycle analyses wereonly performed for 48 hours because the low viability profiles of treated HepG2cells (Figure 6) did not allow longer investigations.However, speculatively, a further increase in G2/M phase arrest might be predicteddue to the slow release profile of emulsomes.

Conclusions

Introducing a new nanocarrier system for curcumin, the present study illustrates theparticular characteristics of CurcuEmulsomes and investigates the delivery ofcurcumin into the cell in vitro, where HepG2 cell line is used as a model.In summary it may be concluded that i) curcumin can be incorporated into theemulsomes, ii) the incorporation enhances the poor water solubility of thisbioactive polyphenol, iii) upon incorporation, biological activity as well asfluorescence integrity of curcumin is preserved, iv) delivered within a solid lipidcore, curcumin is gradually released into the cell, thereby resulting in prolongedcytotoxicity and cell cycle arrest on HepG2, v) due to its prolonged activity, theincorporated curcumin acts, on long terms, as efficient as free curcumin dissolvedin organic solvent. Consequently, enabling curcumin to reach its effectiveconcentrations inside the cell, the presented approach may allow therapeuticapplications of curcumin, and with future perspectives, provide an alternativeplatform for the delivery of hydrophobic bioactive agents whose medical use isotherwise limited.

Methods

Materials

Curcumin, glyceryl tripalmitate (tripalmitin, purity ≥99%),1,2-dipalmitoyl-rac-glycero-3-phosphatidylcholine (DPPC, 99%), glutaraldehydesolution (50%), glycerol (99%) and Dulbecco’s Phosphate Buffered Saline(PBS, 10X) were purchased from Sigma-Aldrich GmbH, Germany. Hexadecylamine (HDA,≥99%), uranyl acetate dehydrate (≥98%), methanol (99.5%) andchloroform (≥99.8%) were obtained from Fluka Chemika, Germany. Cholesterol(>98%) was purchased from Avanti Polar-Lipids, US. Dimethyl sulfoxide (DMSO)was purchased from Riedel-de Haën (Sigma Aldrich, Germany).4′,6-Diamidino-2-phenylindole dihydrochloride (DAPI) was purchased fromAppliChem GmbH, Germany. All chemicals were used as purchased without anyfurther purification.

Cell line

HepG2 (human liver hepatocellular carcinoma cell line) was obtained from AmericanType Culture Collection (Rockville, MD; ATCC HB-8065). HepG2 cells were culturedin Minimal Essential Medium (MEM) with Earlea’s Salts medium (PAA,Pasching, Austria), supplemented with 10% fetal calf serum (FCS) albumin and 1%antibiotic/antimycotic (both PAA) at 37°C in a humidified atmosphere of 5%CO2 and 95% air.

CurcuEmulsome preparation

Curcumin and tripalmitin with a weight ratio of 2:5 were dissolved in chloroform.DPPC, cholesterol and HDA with a molar ratio of 10:5:4 were dissolved separatelyin chloroform. Both lipid solutions were mixed and the organic solvent wascompletely removed using a rotary evaporator (Rotavapor R-215, Büchi,Switzerland) under reduced pressure at 474 mbar and 60°C. The formeddry film was hydrated with MilliQ water, the temperature was set to 80°Cand the solution was rotated until the lipid film was resuspended. The obtainedproduct was homogenized by high pressure extrusion system with heating control(Avestin Liposo Fast LF-50, Ottawa, Canada). At a temperature of 66°C andunder an overpressure above 10 bars, the solution was passed multiple timesthrough 800-nm and 400-nm polycarbonate filters (Nucleopore Track-Etch Membrane,Whatman, UK). Immediately after extrusion, the obtained emulsome suspension wasplaced on ice for 10 min. CurcuEmulsome preparations were centrifuged at13,200 rpm (16,100 g) for 10 minutes to spin down unincorporatedcurcumin. The CurcuEmulsome suspension, i.e. the supernatant, was stored at4°C until further characterization and cell culture studies. Emptyemulsomes were prepared as described above but without curcumin [30].

Quantification of curcumin by absorbance measurements

A 1 mg/ml stock solution of curcumin was prepared in DMSO. A standard curve,generated by successive dilution of the stock solution (5, 10, 20, 50,100 μg/ml) in a 96-well microplate (Cellstar, Greiner Bio-One GmbH,Frickenhausen, Germany), was used to determine curcumin concentrations insamples prepared by dilution of CurcuEmulsome suspension 1:10 in DMSO. Sampleabsorbance was measured at 430 nm on Infinite F200 plate reader (TECAN,Austria).

Compositional analysis of CurcuEmulsomes

The composition of CurcuEmulsomes was determined by HPLC. CurcuEmulsomeformulation was dissolved in methanol to disrupt its structure. The sample wassubjected to sonication for 3 min at 170 W (Transsonic T 460, Elma,Germany) followed by centrifugation at 14,680 rpm (20,238 g) for10 min at 25°C (Centrifuge 5424, Eppendorf, Germany). The clearsupernatant was analyzed using reverse phase isocratic mode (RP-HPLC) on SummitHPLC systems (Dixon/ThermoFisher Scientific, Germany). In brief, 10 μlof the sample was injected automatically in the injection port and analyzed onC18 column (Nucleosil 120-3C18, 150×4 mm, Macherey-Nagel, Germany) with themobile phase consisting of acetonitrile and 2% acetic acid (40:60, v/v) at33°C [51]. The amount of curcumin was quantified by UV detection at 420 nmwith UV/VIS-Detector UVD 170U/340U (Dionex, Germany). The compositionaldistribution of curcumin in the sample was determined from the peak areacorrelated with the standard curve. The total HPLC analysis time was 20 minper sample, with curcumin, DMC and BDMC eluting at retention times of 17.3, 15.4and 13.7 min, respectively.

In vitro cytotoxicity assay

Cytotoxicity of CurcuEmulsomes was examined by CellTiter-Blue Cell ViabilityAssay (Promega, Germany) as described previously by Ucisik et al.(2013) [30]. Briefly, HepG2 cells were seeded in 96-well microtiter plates at adensity of 10,000 cells per well in a final volume of 300-μL culturemedium. After 24 h, the cell culture media were aspirated and the cellswere treated with 100-μl culture medium containing free curcumin (in DMSO)or CurcuEmulsomes at various concentrations. Other cells were left untreated asnegative control. DMSO content in total cell medium was kept below 0.15% toavoid any influence of DMSO to HepG2. Fluorescence intensity of cells wasrecorded using Infinite F200 plate reader (TECAN, Austria) with a560(20)Ex/595(35)Em fluorescence intensity filter(TECAN, Austria).

Cell cycle analysis

HepG2 cells were seeded in cell culture flasks at a density of 500,000 cells per25 cm2. After two days of incubation cell medium was changedwith 5-ml culture medium containing free curcumin (40 μM) orCurcuEmulsome (40 μM). Other cells were left untreated as negativecontrol. Cells were harvested, washed three times with PBS, then counted andresuspended in PBS at concentration of 1×106 cells/ml. Of eachsample 3×105 cells (300 μl) were stained with2 μg/ml DAPI in methanol for 15 min at room temperature in thedark. Subsequently, cells were centrifuged at 1250 rpm for 5 min,resuspended in ice-cold FACS buffer (Becton Dickinson (BD), Austria) andimmediately analyzed via a FACSCanto II Flow Cytometer equipped with a BDFACSDiva acquisition and analysis program (BD, Austria). Samples, stained withDAPI, were excited with a 405-nm blue laser and the emitted light in the regionof 450(50) nm (Pacific Blue) was recorded. Data from at least 10,000 cell countswere collected for each data file. Gating was set properly to exclude celldebris, cell doublets, and cell clumps.

Apoptosis test

The apoptosis response of HepG2 cells to CurcuEmulsomes and free curcumin in DMSOwere analyzed by Cell Meter Caspase 3/7 Activity Apoptosis Red FluorescenceAssay Kit (AAT Bioquest, Biomol, Germany). Briefly, HepG2 cells were seeded in96-well microtiter plates at a density of 10,000 cells per well in a finalvolume of 100-μL culture medium. After 24 h, the cell culture mediawere aspirated and the cells were treated with a medium containing free curcumin(in DMSO) or CurcuEmulsomes at various concentrations for 6, 24 and 48 h.Other cells were left untreated as negative control. At the time of analysis,the medium was replaced and equal volume of Z-DEVD-ProRed™ Reagent Assaywas added to each well. Following incubation of cells at room temperature for70 min in the dark, the fluorescence intensity atEx/Em = 535(25)/635(35) nm was monitored by Infinite F200 platereader.

Ultraviolet–visible (UV–vis) absorbance spectroscopy

UV–vis absorbance spectroscopy was performed on a U-2900 UV/Visspectrophotometer (Hitachi, Japan). Samples were scanned from 300 to700 nm.

Fluorescence spectroscopy

Fluorescence spectra were obtained on a Perkin Elmer LS 55 luminescencespectrometer (Perkin Elmer, UK). Curcumin samples were excited at 420 nm,and emission was monitored from 430 to 600 nm (2.5-nm slit width).

Dynamic and phase analysis light scattering

Diluted in 1 mM KCl solution (pH 6.3) CurcuEmulsomes with a final DPPCconcentration of 4 μg/ml were analyzed by the Zetasizer Nano ZS(Malvern Instruments Ltd, UK) for their particle size distribution (DynamicLight Scattering; DLS) and zeta potential characteristics (Phase Analysis LightScattering; M3-PALS) as previously described [30].

Electron microscopy

The shape and the integrity of CurcuEmulsomes were analyzed by a FEI TecnaiG2 20 Transmission Electron Microscope (TEM) at 120 kVequipped with FEI Eagle 4 k camera (FEI Europe, The Netherlands) afternegative staining with uranyl acetate (1% in MilliQ) as described by Ucisiket al. (2013) [30].

Fluorescence microscopy

Nikon Eclipse TE2000-S fluorescence microscope (Nikon, Melville, NY) was used tovisualize the samples. The images of cell cultures were taken with 20× and40× objectives, and those of CurcuEmulsomes preparation with 100x oilimmersion objective. Curcumin incorporated in emulsomes was detected using aCyan Fluorescent Protein (CFP)-Fluorescence filter (Excitation at436/20 nm; Emission at 480/40 nm).

Authors’ information

This study embodies a part of MHU’s PhD study.