Article pubs.acs.org/Langmuir
First Evidence of the Liposome-Mediated Deintercalation of Anticancer Drug Doxorubicin from the Drug−DNA Complex: A Spectroscopic Approach Anupam Das,† Chandan Adhikari,† Debasis Nayak,‡ and Anjan Chakraborty*,† †
Discipline of Chemistry, Indian Institute of Technology Indore, Indore, Madhya Pradesh, India Bioseciences and Biomedical Engineering, Indian Institute of Technology Indore, Indore, Madhya Pradesh, India
‡
S Supporting Information *
ABSTRACT: Biocompatible liposomes were used for the first time to study the deintercalation process of a prominent anticancer drug, doxorubicin (DOX), from doxorubicin-intercalated DNA (DOX− DNA complex) under controlled experimental conditions. The study revealed that anionic liposomes (DMPG liposomes) appeared to be the most effective to bring in the highest percentage of drug release while cationic liposomes (DOTAP liposomes) scored the lowest percentage of release. The drug release was primarily attributed to the electrostatic interaction between liposomes and drug molecules. Apart from this interaction, changes in the hydrophobicity of the medium upon addition of liposomes to the DNA−drug solution accompanied by lipoplex formation between DNA and liposomes were also attributed to the observed deintercalation. The CD and the timeresolved rotational relaxation studies confirmed that lipoplex formation took place between liposomes and DNA owing to electrostatic interaction. The confocal study revealed that in the postrelease period, DOX binds with liposomes. The reason behind the binding is electrostatic interaction as well as the unique bilayer structure of liposomes which helps it to act as a “hydrophobic sink” for DOX. The study overall highlighted a novel strategy for deintercalation of drug using biocompatible liposomes, as the release of the drug can be controlled over a period of time by varying the concentration and composition of the liposomes.
■
INTRODUCTION Doxorubicin (DOX), one of the most important family members of the anthracycline antibiotics, has been widely used as a potent antineoplastic agent for decades.1 The wellaccepted concept regarding the mode of action of DOX is that the drug inhibits (both by catalytic activity and poisoning) type IIA human topoisomerase,2 which is an enzyme that changes DNA topology by breaking and rejoining double-stranded DNA to step up the replication process. Despite intense and enormous efforts by different scientists for years, the exact mechanism of binding of DOX with DNA is not clear until now.3 Recently, Garcia et al. reported formation of two different types of DOX−DNA complexes (PD1 and PD2) using isothermal calorimetry.4 These complexes are nonfluorescent in nature and formed upon intercalation of DOX into DNA. The PD1 complex is formed at lower drug concentration ([DOX]/ [DNA] < 0.3) while PD 2 is formed at higher drug concentration.4 It has been reported that the origin of the biological activities of DOX depends on the features of the binding of DOX with DNA.5 The slow rate of dissociation from DNA is considered to be one of the most significant characteristics for a drug to be efficient as a cancer therapeutic.6 Considering the above two typical characteristics of DOX, one may say that kinetics of © XXXX American Chemical Society
association and dissociation of intercalators such as DOX are of great diagnostic significance. Previously, a number of techniques were employed to study the dissociation rate of the intercalator from DNA. These include T-jump relaxation for the measurement of fast kinetics7 and a modified footprinting technique to study the dissociation of drugs from specific binding sites.8 The detergent-sequestration technique using micelles (viz. sodium dodecyl sulfate) was used for deintercalation of hydrophobic and cationic drugs. Micelles acted as a hydrophobic sink for the dissociation of drug molecules.9−12 A few groups reported deintercalation by caffeine also.13,14 In the past decade, Westerlund and co-workers15 showed that the presence of surfactant molecules in monomeric as well as in micellar forms increases the rate of dissociation of cationic DNA-binding ligands both below and above the critical micelle concentration (CMC) of the respective surfactants. Recently, Mitra and co-workers reported a detailed thermodynamic study regarding the deintercalation mechanism of cationic dyes from DNA.16 Although there are a few reports of deintercalation by Received: April 29, 2015 Revised: November 23, 2015
A
DOI: 10.1021/acs.langmuir.5b03702 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
second section deals with the interactions of DOX with three different liposomes. The third section involves the study of the deintercalation of DOX from the DOX−DNA complex upon interaction with different liposomes. Here we successfully show that the deintercalation process of the drug molecules can be controlled over a time period, varying the composition and concentration of liposomes. The different interaction and deintercalation processes were established by applying steadystate and time-resolved fluorescence spectroscopy, circular dichroism, and confocal imaging.
synthetic surfactants, currently there are no reports of biocompatible entities having a cytotoxicity level lower than that of other synthetic surfactants. In the present effort, for the first time, we propose a novel approach for deintercalation of an anticancer drug doxorubicin hydrochloride (DOX) (Scheme 1) with the help of biocompatible small unilammelar vesicles (SUV). Scheme 1
■
EXPERIMENTAL SECTION
Materials and Solution Preparation. Doxorubicin hydrochloride (DOX), calf thymus DNA (ctDNA), and different phospholipids were purchased from Sigma-Aldrich and used without further purification. The structure of DOX and the lipids are shown in Scheme 1. Phosphate buffer salts were purchased from Merck. In all cases, we used Milli-Q water to prepare the solutions. A stock solution of DOX was prepared by dissolving a weighed amount in 0.01 M phosphate buffer (Na2HPO4 and NaH2PO4) of pH ∼ 7.4. The working solutions of different concentrations of DOX were prepared by diluting the stock solution with an exact amount of buffer. ctDNA was dissolved in phosphate buffer and stirred overnight to solubilize the DNA fibers. The exact concentration of DNA solution was determined by measuring the absorbance at 260 nm (O.D. at λ260). Both the DOX and the DNA solutions were kept in the dark at 4 °C. The liposome preparation was conducted following the rapid ethanol injection method as reported previously.19,20 In brief, an aqueous solution of DOX (2 μM) was placed in a round-bottom flask, and the temperature was kept above the phase transition temperature of the respective lipid. The phase transition temperatures (Tm) of DMPC, DMPG, and DOTAP are 24 °C, 23 °C, and 0 °C, respectively. To each roundbottom flask containing aqueous DOX (2 μM) solution, the desired amount of ethanolic lipid solution was added by rapid injection (above their phase transition temperature), and the mixture was equilibrated for more than 1 h. Lipid concentration of the final liposomal solution was 1 mM, and the injected ethanol was less than 1% (v/v) of the solution. Mixed liposomes were prepared by mixing the two lipids according to the weight ratio (e.g., DMPC:DMPG (9:1)). SDS solution was prepared by dissolving an exact amount of SDS in phosphate buffer. Spectroscopic Techniques. Steady-state absorption spectra were recorded with a Varian UV−vis spectrophotometer (model: Cary 100). Fluorescence titrations were performed using a Fluoromax-4p spectrofluorometer from Horiba JobinYvon (model: FM-100). For DOX−DNA titration, we prepared a set of solutions in different volumetric flasks which contained 2 μM DOX and different concentrations of DNA (from 0 to 1.23 mM). In similar fashion, the liposomes solutions were prepared, keeping the DOX concentration constant (2 μM). In brief, we prepared a set of solutions of liposomes where the DOX concentration was constant (2 μM) and the liposome concentration was varied from 0 to 1 mM. For the deintercalation process, we varied the concentration of liposomes while the concentrations of DNA and DOX were kept fixed. The samples were excited at 490 nm (λex = 490 nm) and the emission peak appeared at ∼595 nm (λem = 595 nm). The fluorescence spectra were corrected for the spectral sensitivity of the instrument. The excitation and emission slits were 2/2 nm for almost all the emission measurements. Throughout all the titration experiments we maintained pH = 7.4, ionic strength (I) = 0.01 M, and temperature (T) = 25 °C. For time-correlated single-photon counting (TCSPC), we used a picosecond TCSPC machine from Horiba (Fluorocube-01-NL). The experimental setup for TCSPC has been described previously.19,20 The samples were excited at 482 nm using a picosecond diode laser (model: Pico Brite-482L), and the decays were collected at 595 nm. The signals were collected at magic angle (54.75°) polarization using a photomultiplier tube (TBX-07C) as detector which has a dark count
It has been found that liposomes are the most suitable surfactant entities to entrap both the hydrophobic and the hydrophilic drug molecules due to their unique physicochemical properties.17 The surface charge of the liposomes can easily be modulated using different colipids. The advantages of using liposomes over micelles, reverse micelles, and other artificial vesicles are that they are biocompatible and hence nontoxic.18 In the present work, we used three different types of liposomes, namely DMPC liposomes, DMPC:DMPG (9:1, described as DMPG) liposomes, and DMPC:DOTAP (9:1, described as DOTAP) liposomes. Among the above-mentioned liposomes, DMPG, DMPC, and DOTAP are anionic, neutral, and cationic in charge, respectively. Thus, by varying the composition of phospholipids, we can easily tune the overall charge of the mixed liposomes. The three above-mentioned phospholipids were selected in preference to other surfactants, keeping in mind their low cytotoxicity level compared with other synthetic surfactants. We employed steady-state and time-resolved spectroscopy to unravel the deintercalation process. Circular dichroism and time-resolved anisotropy experiments were conducted to study the liposome−DNA interaction. The postrelease fate of DOX was revealed by confocal study. We divided the manuscript into three sections. In the first section, we illustrate the interaction of DOX with ctDNA. The B
DOI: 10.1021/acs.langmuir.5b03702 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 1. (a) Fluorescence quenching of DOX (2 μM) with increasing concentration of ctDNA. Fluorescence intensity of DOX vs DNA concentration plot is inset. (b) Time-resolved decays of DOX at different concentrations of DNA (c) Time-resolved anisotropy decays of DOX at different concentrations of DNA. The experiments were carried out at 25 °C, pH 7.4, and 0.01 M ionic strength, and 2 μM DOX concentration. less than 20 Hz. The full width half-maximum (fwhm) of instrument response function of our setup is ∼140 ps. The data analysis was performed using IBH DAS Version 6 decay analysis software. Throughout all the titration experiments we maintained pH = 7.4, ionic strength (I) = 0.01 M, and temperature (T) = 25 °C. The decays were fitted with a multiexponential function. n
D(t ) =
⎛ −t ⎞ ⎟ ⎝ τi ⎠
∑ ai exp⎜ i=1
For the confocal imaging of samples, we used a confocal microscope from Olympus, model no. IX-83. A multiline Ar laser (gas laser) with an excitation of 488 nm was used. The observation mode was LSM, scan mode was XY, and scan direction was one way. The liquid samples were dropped on glass slides and fixed with the coverslips before imaging.
■
(1)
Here D(t) denotes normalized fluorescence decay and ai is the normalized amplitude of decay components τi, respectively. The average lifetime was obtained from the equation n
⟨τ ⟩ =
∑ aiτi i=1
(2)
The quality of the fit was judged by reduced chi square (χ2) values and corresponding residual distribution. The acceptable fit has a χ2 near unity. For the anisotropy decays, we used a motorized polarizer in the emission side. The emission intensities at parallel and perpendicular polarizations were collected alternatively until a certain peak difference between parallel (I∥) and perpendicular (I⊥) decay was achieved. The same software was also used to analyze the anisotropy data. The timeresolved anisotropy was described with the following equation:
r(t ) =
I (t ) − GI⊥(t ) I (t ) + 2GI⊥(t )
RESULTS AND DISCUSSION
DOX is known to undergo dimerization under physiological conditions at higher concentration.21 The process of dimerization and aggregation depends upon the ionic strength of the buffer used for the solution preparation. To overcome the dimerization problem, we used a very low concentration of DOX (2 μM) and at the same time the buffer strength (0.01 M) was also kept very low. Recent studies from Blanchard and co-workers showed that DOX is susceptible to chemical changes upon exposure to radiation.22 They reported that DOX undergoes several reactions, including a photoreduction to form a dihydroquinone in aqueous solution upon exposure to light.22 Besides, the complex formation of DOX is highly dependent on the functionality of the hydroxymethylketone and methoxy substituents of DOX.23,24 In our experiments, we maintained all the experimental conditions properly so that photochemical changes of DOX did not take place. The titration of DOX versus ctDNA (Figure 1(a)) reveals that the fluorescence intensity of aqueous DOX gradually decreases with increasing concentration of ctDNA. More than 95% quenching takes place at the maximum concentration of ctDNA (1.23 mM). The extreme amount of quenching (∼95%) led us to assume that almost all the DOX binds to DNA and a negligible amount of DOX remains unbound in the aqueous solution. The Stern−Volmer quenching constant obtained from the fluorescence measurements is ∼3 × 104 M−1 (±5%) (Supporting Information (SI) Figure 1). It is revealed that the drop in fluorescence intensity is rapid in the initial stage upon addition of DNA. However, at a higher concentration of DNA the drop in fluorescence intensity takes place gradually. We
(3)
where r(t) is the rotational relaxation correlation function, I∥(t) and I⊥(t) are the parallel and perpendicular components of the fluorescence, and G is the correction factor. Circular dichroism (CD) spectra were recorded by using a Jasco J815 spectrometer (Jasco, Tokyo, Japan). Far-ultraviolet (UV) (200− 310 nm) spectra were recorded in 0.1 cm path length cell (Hellma, Muellheim/Baden, Germany) using a step size of 0.5 nm, bandwidth of 1 nm, and scan rate of 20 nm/min. Each sample spectrum was obtained by averaging five scans and corrected with the solvent spectrum. For all the CD experiments, we maintained pH = 7.4, ionic strength (I) = 0.01 M, and temperature (T) = 25 °C. C
DOI: 10.1021/acs.langmuir.5b03702 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 2. (a, b) Emission spectra of DOX in the presence of DMPG and DOTAP liposomes, respectively. (c) Plot of fluorescence intensity (λex = 490 nm) of DOX as a function of total concentration of liposome. The curve represents the nonlinear regression fit to the experimental data using eq 6. (d) The time-resolved decay curves of DOX upon addition of the liposomes to an aqueous buffer solution of DOX. The experiments were carried out at 25 °C, pH 7.4, 0.01 M ionic strength, and 2 μM DOX concentration.
plotted fluorescence intensity of DOX against the concentration of DNA (Figure 1a, inset). The feature is quite similar to the other cationic dye as reported by Mitra et al. in the case of intercalation of other cationic dyes to DNA.25,26 Recently, quenching of DOX in the presence of ctDNA has been extensively described by Garcia and co-workers.4 In the presence of DNA, two nonfluorescent complexes (PD1 and PD2) of DOX are formed depending on the drug to DNA molar ratio (CD/CP).4 At a lower CD/CP value (
