ANTICANCER RESEARCH 33: 857-864 (2013)

The Influence of Phenothiazine Derivatives on Intracellular Accumulation of Cationic Cyanine Dye DiOC6(3) in LoVo-DX Cells

ANDRZEJ POŁA1, DANIELA MOSIĄDZ1, JOLANTA SACZKO2, TERESA MODRZYCKA1 and KRYSTYNA MICHALAK1

Departments of 1Biophysics and 2Medical Biochemistry, Wrocław Medical University, Wroclaw, Poland

Abstract. Aim: This study aimed to evaluate the influence phenomenon of human cancer cells, which is manifested by of phenothiazine derivatives (PDs) on the intracellular transport of a broad array of chemical compounds out of accumulation of cyanine dye DiOC6(3) in doxorubicin- cells (6, 7). Among MDR transporters, the earliest discovered resistant LoVo-DX cell line, with overexpression of P- was P-gp (8), hence, the molecular mechanism of its action glycoprotein. Materials and Methods: In order to maintain is best-understood. Most researchers suggest that P-gp acts a high expression level of P-gp, the LoVo-DX cells were as a pump that removes cytotoxic drugs, as well as other grown in the presence of doxorubicin (100 ng/ml). The time- structurally- and functionally-unrelated compounds, trying to dependent fluorescence signal (T-DFS) of the intracellular penetrate the cancer cell interior (4, 9-16). It seems that the accumulation of DiOC6(3), in the presence of PDs, was then electrical charge of transported molecules plays an important recorded. The rate constants k1, k2, k3 and amplitudes of T- role in recognizing compounds as substrates for MDR DFS, describing the intracellular accumulation process, transporters. P-gp has very high affinity for organic cations, were determined based on the respective theoretical whereas ABCC1 substrates share the property of being equation. Results: The values of k1 and k2 were dependent amphiphilic organic anions and glutathione or glucuronate on the hydrophobicity (logP) of the PDs used as drug conjugates (17, 18). It was found that amphiphilic drugs (19) resistance modulators. A rise of k1 and k2 values was and anions (20, 21), as well as cations (22, 23), can be observed when the logP of PDs increased. Conclusion: We transported by ABCG2. suggest that the k1 and k2 rate constants could be regarded Higher intracellular accumulation of anticancer drugs was as useful parameters for assessment of PDs as well as of observed in cell lines with low P-gp expression, as compared other compounds of potential application as reversers of to their resistant counterparts with P-gp overexpression (11, multidrug resistance. 24). Similarly, a lower intracellular concentration of cationic cyanine dyes was observed in cancer cells with high Multidrug resistance (MDR) of cancer cells is a basic reason expression of this transporter (25-27). Spectral properties of for frequent failure of chemotherapy in cancer treatment. The cyanines are very useful to determine the kinetic parameters molecular mechanism of MDR is multifactorial (1-3) but the describing the process of intracellular dye accumulation. The classical MDR phenotype is associated with overexpression kinetic parameters may be estimated from the model proposed of multidrug transporters belonging to the superfamily of by Wadkins and Houghton (26), which was established on the ATP-binding cassette proteins (ABC proteins), in the plasma basis of analyses of time-dependent fluorescence signals (T- membrane of cancer cells (4, 5). From the superfamily of DFS) during intracellular accumulation of cyanine dye in ABC proteins, mainly P-glycoprotein (P-gp; ABCB1), cancer cells with different P-gp expression levels. ABCC1 and ABCG2 are responsible for the MDR Many different chemical compounds can reverse or modulate MDR in cancer cells (5, 28). A very important group of compounds which in in vitro experiments has been shown to be MDR-reversing agents, are phenothiazine Correspondence to: Andrzej Poła, Department of Biophysics, derivatives (PDs). Phenothiazines are used in the treatment Wrocław Medical University, ul. Chałubińskiego 9, 50-368 Wrocław, Poland. E-mail: [email protected] of psychiatric disorders, and PDs are well-known substrates of P-gp (29-32). Key Words: LoVo, LoVo/DX, cyanine dyes, intracellular In the present article, we examined the influence of PDs, accumulation, phenothiazines. belonging to the perazine and promazine group, on the

0250-7005/2013 $2.00+.40 857 ANTICANCER RESEARCH 33: 857-864 (2013)

intracellular accumulation of cationic carbocyanine dye Table I. Chemical structures of cyanine dye DiOC6(3) and phenothiazine derivatives from the promazine and perazine groups, and [DiOC6(3)] in colonic cancer cells with P-gp overexpression. Since both DiOC (3) and PDs are P-gp substrates, we assumed calculated values of logP for these compounds. *Calculated using the 6 software: SPARTAN ’08. that the presence of PDs in the cell suspension might affect the process of cyanine dye intracellular accumulation and, in consequence, changes in kinetic parameters describing this process should be detected. In the interpretation the results of our experiments a potential influence of ABCC1 and ABCG2 on intracellular accumulation of DiOC6(3) was excluded. The cationic dye used in our study was not a substrate in the case of ABCC1 protein. On the other hand, the expression ABCG2 protein was at the same level for both the LoVo and the LoVo-DX cell lines, what was confirmed in our laboratory (33).

Materials and Methods

Reagents. Promazine, chlorpromazine, trifluopromazine, prochlorpromazine, dimethyl sulfoxide (DMSO), DiOC6(3), dioxane, chloroform, butanol, acetone, ethanol, methanol, Dulbecco’s modified Eagle’s medium F12 Ham, cell dissociation solution non-enzymatic, doxorubicin hydrochloride, antibiotic antimycotic solution and fluorescein 5(6)-isothiocyanate (FITC) were purchased from Sigma, Poznań, Poland. Fetal bovine serum (FBS) was purchased from Gibco, Warsaw, Poland, L(+)-glutamine from BDH Prolabo, Wrocław, Poland and trifluoperazine from ICN Biomedicals Inc., Gdańsk, Poland. Perazine was supplied by FSP Galena, Wrocław, Poland. The antibody for P-gp was purchased from Santa Crus Biotechnology, Dallas, Texas, USA. Stock solutions of PDs (5 mM) and DiOC6(3) (400 μM) were prepared in DMSO. The chemical structure, kind of substituent at position 2 of the phenothiazine ring and calculated values of partition co-efficient (logP) for PDs used in these studies, are presented in Table I. Intracellular dye accumulation. The emission intensity of T-DFS Cell preparation. Doxorubicin-sensitive LoVo and doxorubicin- during intracellular accumulation of the dye was monitored at resistant LoVo-DX cells were obtained from Institute of λem=509 nm with excitation wavelength λex=482 nm. Fluorescence Immunology and Experimental Therapy (Polish Academy of measurements were carried-out with a Perkin-Elmer LS-50B Sciences, Wroclaw, Poland). Both cell lines were grown in spectrofluorimeter. A quartz cuvette, containing DiOC6(3) solution Dulbecco’s medium, supplemented with 10% FBS, L-glutamine and in cell medium (without FBS, glutamine, antibiotics) was placed in antibiotic solution. In order to maintain a high level expression of P- the spectrofluorimeter and then an appropriate volume of cell gp, the LoVo-DX cells were grown in the presence of doxorubicin suspension was added (sample was continuously stirred). The final (100 ng/ml). Cells were cultured at 37˚C in a humidified atmosphere concentration of DiOC6(3) was 0.4 μM and the cell density was 5 3 of 5% CO2. In log-phase growth, they were removed from the flask 2.5×10 cells/cm . To measure the kinetics of dye accumulation in surface with non-enzymatic cell dissociation solution. The density the presence of PD, an appropriate amount of PD in solution was of cells in suspension was determined using a Bürker mixed with the dye solution and then the cell suspension was added. haemocytometer and was usually about 106/cm3. The final concentration of PDs was 15 μM.

Determination of P-gp expression by immunofluorescence. Cells in Mathematical analysis. The mathematical analysis of the intracellular growth medium were plated on cover glasses and were grown for accumulation of carbocyanine dyes was given by Wadkins and one or two days. They were fixed in 4% formalin buffer and washed Houghton (26). In our study, an equation (Eqtn 1) was applied to with PBS. The intracellular localization of P-gp was determined with estimate parameters characterizing dye accumulation in cells. The antibody detector labeled with FITC. Samples were attached to observed fluorescence intensity F(t) (T-DFS) for cyanine dye in the microscope slides using fluorescence mounting medium. form of monomers during its accumulation in cancer cells is given by: Fluorescence was monitored using a confocal scanning laser microscope (Carl Zeiss GmbH, Jena, Germany) with excitation wavelength λex of 470±20 nm and fluorescence emission recorded at λem=525±20 nm. (Eqtn 1)

858 Poła et al: Effect of PDs on DiOC6(3) Uptake in LoVo-DX Cells

Results

Confocal microscopy studies. Using confocal microscopy, we determined the expression of transporting membrane protein, P-gp, in both cell lines. We observed a significantly higher expression of P-gp in plasma and mitochondrial membranes of LoVo-DX cells, as compared to LoVo cells (Figure 1).

Intracellular accumulation of DiOC6(3) in LoVo-DX and LoVo cells determined by fluorescence spectroscopy. The changes of T-DFS during intracellular accumulation of 5 3 DiOC6(3) for both cell lines (2.5×10 /cm ) are shown in Figure 2. The estimated value of rate constants k1 and k2 were higher for LoVo cells than for LoVo-DX cells. Like the rate constants the amplitudes A1 and A2 were also higher for LoVo cells (Table II). The solid lines in Figure 2 represent the theoretically calculated curves obtained by fitting of experimental data to Equation 1.

Accumulation of DiOC6(3) in LoVo-DX cells in the presence of PDs determined by fluorescence spectroscopy. The addition of a given PD to the DiOC6(3) buffer solution (without the cells) caused a slight (less than 1%) decrease in the amplitude of T-DFS for the dye monomer (data not shown). This decrease of amplitude did not influence the shape of the T-DFS curve measured in the presence of LoVo- DX cells, with the given PD in the sample. The T-DFS registered during intracellular accumulation of DiOC6(3) in LoVo-DX cells, in the presence of compounds from both groups of PDs, are shown in Figures 3 and 4. The T-DFS was registered after the addition of LoVo-DX cell suspension (final density of the cells was 2.5×105 cells/cm3) to buffer a solution containing both the dye and PD. The k1 and k2 rate constants were dependent on the partition co-efficient (logP) of PDs. With an increase of logP, an increase of rate constants was observed (Figure 5). The solid lines in Figures 3 and 4 represent the theoretically calculated curves obtained by fitting of experimental data to Equation 1. The average values of parameters obtained in at least three independent experiments are listed in Table II.

Figure 1. Expression of P-gp in LoVo cells (A) and in LoVo/Dx cells (B). Discussion

Currently, optical dyes including cyanines are often used in where k1 represents an effective rate constant of transition of the investigations of living cells (27, 34, 35). In aqueous dye from buffer solution to plasma and mitochondrial membranes, solutions, some cyanine dyes usually exist in different forms, k2 and k3 are rate constants for dye aggregation in mitochondria and such as fluorescent monomers and J-aggregates, non- its intracellular traffic, respectively. A1 is the amplitude of processes fluorescent H-aggregates/dimers, or various mixed described by rate constants k1 and k2. The parameters A2 and F0 aggregates (36-41). In the lipid membrane (plasma represent the amplitude of the process described by rate constant k 3 membrane), the efficacy of aggregation was higher than and the fluorescence intensity of the background, respectively. The kinetic parameters were estimated by fitting the experimental data observed in pure water (42-44). We assumed that the (fluorescence signal F) to the equation (1), using the Sigma Plot 8.0 hydrophobic interaction of cationic dye DiOC6(3) with Software (Systat Software Inc., London, UK). plasma and mitochondrial membranes and the negative

859 ANTICANCER RESEARCH 33: 857-864 (2013)

Figure 2. Time-dependent fluorescence signal (T-DFS) recorded during Figure 3. The time-dependent fluorescence signal (T-DFS) recorded intracellular accumulation of [DiOC6(3)] in LoVo-DX and LoVo cells. during intracellular accumulation of DiOC6(3) in LoVo-DX cells without The concentration of DiOC6(3) added was 0.4 μM, the cell density was modulator (1) and in the presence of promazine (2), chlorpromazine (3) 5 3 2.50×10 cells/cm . The solid lines are theoretical curves obtained by and triflupromazine (4). The points represent experimental data and the fitting of experimental data to Equation 1 (see Material and Methods). solid lines are theoretical curves obtained by fitting of the experimental data to Equation 1. The dye concentration in the sample was 0.4 μM, 5 3 the cell density was 2.5×10 cells/cm , λex=482 nm, λem=509 nm. potential of these membranes were the most important factors affecting the dye molecules to move from the aqueous phase to the hydrophobic cell compartments. On the other hand, P-gp overexpressed in the plasma and mitochondrial membrane of cancer cells hinders free displacement of dye molecules from aqueous solution to hydrophobic compartments of the cells. The fast increase of T-DFS (Figure 2), after addition of LoVo-DX or LoVo cells to the buffer solution of the dye, was caused by transition of dye monomers to plasma and mitochondrial membranes. A higher final value (at 250 s) of T-DFS in the case of LoVo cells indicated a greater intracellular amount of dye monomers in these cells than in LoVo-DX cells (Figure 2). This was caused by intensive removal of the dye molecules out of LoVo-DX cells by P-gp, acting as an active transporter. On the other hand, a higher value of k1 in LoVo Figure 4. The time-dependent fluorescence signal (T-DFS) recorded cells than in LoVo-DX cells indicated a faster passage of the during intracellular accumulation of DiOC6(3) in LoVo-DX cells without dye molecules to hydrophobic cell compartments in this cell modulator (1) and in the presence of trifluoperazine (2), perazine (3) and line. This observation confirms that P-gp overexpression in prochlorperazine (4). The points represent experimental data and the LoVo-DX cells hinders free passage of the dye from aqueous solid lines are theoretical curves obtained by fitting of the experimental data to Equation 1. The dye concentration in the sample was 0.4 μM, solution to plasma and mitochondrial membranes. However, 5 3 the cell density was 2.5×10 cells/cm , λex=482 nm, λem=509 nm. higher values of rate constant k2 for LoVo cells point to faster intracellular aggregation of the dye in these cells as compared to LoVo-DX cells. This was caused by the action of P-gp, which resulted in a high intracellular dye gradually when a hydrogen atom at this position is substituted concentration in LoVo cells, which in turn, favored by –Cl and by –CF3. A similar relationship between logP and aggregation. the type of substituent in position 2 of the phenothiazine ring (- Hydrophobicity (logP) of PDs (promazine and perazine H<-Cl<-CF3) was also observed experimentally for the derivatives) used in our studies depends on the nature of promazine group by Kitamura et al. (45, 46) and for new- substituent in position 2 of the phenothiazine ring and increases synthesized PDs in our earlier research (47).

860 Poła et al: Effect of PDs on DiOC6(3) Uptake in LoVo-DX Cells

Table II. The kinetics parameters (rate constants – k1, k2, k3, and amplitudes of fluorescence signal – A1 and A2, see Materials and Methods) of DiOC6(3) accumulation in LoVo and LoVo-DX cell lines, without and in the presence of phenothiazine derivatives.

4 4 4 Cell line Phenothiazine derivative k1 × 10 k2 × 10 k3 × 10 A1 A2

LoVo - 2186±132 168±6 325±17 47.5±1.9 76.0±0.9 LoVo-DX - 1161±151 150±5 139±2 27.6±8.8 54.2±0.3

LoVo-DX Promazine 1950±166 393±43 179±7 13.9±0.3 33.0±0.1 LoVo-DX Chlorpromazine 2550±316 418±57 168±7 13.4±0.2 32.6±0.2 LoVo-DX Triflupromazine 3954±378 535±50 202±5 12.8±0.2 32.1±0.2

LoVo-DX Perazine 2267±238 394±43 191±7 11.1±0.2 27.1±0.2 LoVo-DX Prochlorperazine 2688±254 415±45 195±8 11.1±0.2 24.3±0.1 LoVo-DX Trifluoperazine 2989±339 435±76 158±7 12.2±0.2 32.7±0.2

For all investigated PDs, the shape of the T-DFS curve was similar (Figures 3 and 4). The final value (after 250 s) of T- DFS was lower in the presence of PD compared to T-DFS of the dye solution without cells. The values of both k1 and k2 were dependent on the partition coefficient (logP) of PDs. With an increase of logP, an increase of rate constant was observed (Figure 5), and in our opinion, for such kind of relationship, P-gp activity is responsible. Under the conditions of our experiments, cyanine dye DiOC6(3) and phenothiazine were simultaneously present in plasma, mitochondrial and most likely also in endoplasmic reticulum membranes. P-gp is able to transport both DiOC6(3) and PDs across cell membranes since these compounds are well-known substrates of this drug transporter. On the basis of our experiments and according to results of earlier investigations on the intracellular accumulation of other dyes in the presence of different modulators of P-gp (12, 25, 48, 49), we suggest that phenothiazines are removed from membranes with higher efficacy than cyanine dye DiOC6(3). While PD is present in a membrane, P-gp pumps-out this compound across the membrane with high efficacy; DiOC6(3) dye escapes this pumping-out and can be more effectively incorporated from aqueous phase into plasma membrane of LoVo-DX cells. Such a mechanism is most likely responsible for the observed rise of k1 value and the increase of intracellular concentration of dye in line with an increase of partition coefficient values of the PDs employed. In this way, aggregation was promoted, which caused a shift of equilibrium between monomer and H-aggregate/dimer towards non-fluorescent H- aggregate/dimer, consequently, a lowering intensity of T-DFS was observed. Aggregation of DiOC6(3) in mitochondria could be described by k2 rate constant. The value of k2 also Figure 5. The plots of k (A) and k (B) rate constants versus logP of increased with logP of PDs. In contrast, k3 was not dependent 1 2 phenothiazine derivatives: promazine group (filled symbols) and on the hydrophobicity of PDs. In all our experiments, A 1 perazine group (opened symbols). Pearson correlation co-efficient and value was approximately half the A value. This difference 2 2 2 p value for promazine group: r =0.880, p=0,0090 for k1 and r =0.784, 2 could be a result of an overlapping of two processes – fast p=0.0005 for k2; for perazine group: r =0.999, p=0.0002 for k1 and 2 accumulation of dye monomers in cell membranes, and r =0.989, p=0.000004 for k2.

861 ANTICANCER RESEARCH 33: 857-864 (2013) formation of permanent or temporal non-fluorescent H- References aggregate/dimer there. Displacement of dye molecules from aqueous phase to hydrophobic cell compartments should 1 Gottesman MM and Pastan I: Biochemistry of multidrug generate a strong increase of fluorescence intensity. However, resistance mediated by the multidrug transporter. Annu Rev inside membranes aggregation occurred, non-fluorescent H- Biochem 62: 385-427, 1993. 2 Simon SM and Schindler M: Cell biological mechanisms of aggregates/dimers appeared very quickly, and a decrease of multidrug resistance in tumors. Proc Natl Acad Sci USA 91: A1 amplitude was observed. When higher values of A2 3497-504, 1994. amplitude, as compared to A1 are observed, it could be 3 Larsen AK and Składanowski A: Cellular resistance to suggested that dye monomers are formed again as a result of topoisomerase-targeted drugs: from drug uptake to cell death. H-aggregate/dimer dissociation. This process is described by Biochim Biophys Acta 1400(1-3): 257-274, 1998. 4 Fardel O, Lecureur V and Guillouzo A: The P-glycoprotein rate constant k3 and this rate constant is not dependent on the logP of PDs. multidrug transporter. Gen Pharmac 27(8): 1283-1291, 1996. 5 Krishna R and Mayer LD: Multidrug resistance (MDR) in By analyzing the data obtained, we conclude that cancer: Mechanisms, reversal using modulators of MDR and the intracellular accumulation of DiOC6(3) apparently proceeds role of MDR modulators in influencing the pharmacokinetics of in a similar way, as proposed by Cabrini and Verkman (50) anticancer drugs. Eur J Pharm Sci 11: 265-283, 2000. for cyanine dye DiSC3(5). The authors suggested that 6 Klein I, Sarkadi B and Varadi A: An inventory of the human DiSC3(5) entered the outer leaflet of the lipid bilayer in the ABC proteins. Biochim Biophys Acta 1461: 237-262, 1999. form of a monomer and then dimerization of the dye 7 Sharom FJ: ABC multidrug transporters: Structure, function and occurred. The created dimers moved to the inner leaflet of the role in chemoresistance. Pharmacogenomics 9: 105-127, 2008. 8 Juliano RL and Ling V: A surface glycoprotein modulating drug lipid bilayer, where they underwent fast dissociation. We permeability in Chinese hamster ovary cell mutants. Biochim suggest that the process of DiOC6(3) accumulation in LoVo- Biophys Acta 455: 152-162, 1976. DX doxorubicin-resistant cancer cells proceeds as follows: 9 Ford JM and Hait WN: Pharmacology of drugs that alter dye monomers from the aqueous solution enter plasma multidrug resistance in cancer. Pharmacol Rev 42: 155-199, 1990. membrane and this process is characterized by rate constant 10 Higgins CF and Gottesman MM: Is the multidrug transporter a flippase? Trends Biochem Sci 17(1): 18-21, 1992. k1, then cyanine is accumulated in mitochondria where aggregation/dimerization occurrs, characterized by rate 11 Roepe PD: The role of the MDR protein in altered drug translocation across tumor cell membranes. Biochim Biophys constant k . Afterwards, as a result of dissociation of H- 2 Acta 1241(3): 385-405, 1995. aggregates/dimers in mitochondrial membrane, dye 12 Ford JM: Experimental reversal of P-glycoprotein-mediated monomers appeared in the cytoplasm. multidrug resistance by pharmacological chemosensitisers. Eur J Cancer 32A: 991-1001, 1996. Conclusion 13 Sharom FJ: The P-glycoprotein efflux pump: How does it transport drugs? J Membrane Biol 160: 161-175, 1997. 14 Loo TW and Clarke DM: Determining the structure and In the presence of PDs, kinetic parameters characterizing the mechanism of the human multidrug resistance P-glycoprotein process of dye accumulation, especially rate constants k1 using cysteine-scanning mutagenesis and thiol-modification and k2, were changed. Influence of PDs on rate constants techniques. Biochim Biophys Acta 1461: 315-325, 1999. was dependent on the partition co-efficients of the 15 Loo TW and Clarke DM: Do drug substrates enter the common compounds. Taking into account that DiOC6(3) drug-binding pocket of P-glycoprotein through gates? Biochem. accumulation was enhanced in the presence of PDs and Biophys. Res Commun 329: 419-422, 2005. kinetic parameters characterizing dye accumulation differed 16 Wiese M and Pajeva IK: Structure−activity relationships of multidrug resistance reversers. Curr Med Chem 8(6): 685-713, 2001. from those determined without PDs, the kinetic rate 17 Litman T, Druley TE, Stein WD and Bates SE: From MDR to constants might serve as useful indicators of drug MXR: New understanding of multidrug resistance systems, their resistance-reversing activity of MDR modulators. An properties and clinical significance. Cell Mol Life Sci 58: 931- important factor for the ability of the studied PDs to inhibit 959, 2001. P-gp transport activity is their high partition co-efficient, 18 Wang RB, Kuo CL, Lien LL and Lien EJ: Structure–activity which facilitates the interaction of these compounds with P- relationship: Analyses of P-glycoprotein substrates and gp inside the membrane. inhibitors. J Clin Pharm Ther 28: 203-228, 2003. 19 Borst P and Elferink RO: Mammalian ABC transporters in health and disease. Annu Rev Biochem 71: 537-592, 2002. Acknowledgements 20 Mao Q and Unadkat JD: Role of the breast cancer resistance protein (ABCG2) in drug transport. The AAPS J 7: 118-133, 2005. This work was supported by a Wroclaw Medical University 21 Leslie EM, Deeley RG, Susan PC and Cole SPC: Research Grant (No. 1410). We thank Dr Michał Majkowski for Multidrug−resistance proteins: Role of P-glycoprotein, MRP1, photography an Dr Kamila Środa-Pomianek for reviewing the MRP2, and BCRP (ABCG2) in tissue defense. Toxicol Appl manuscript. Pharmacol 204: 216-237, 2005.

862 Poła et al: Effect of PDs on DiOC6(3) Uptake in LoVo-DX Cells

22 Schinkel AH and Jonker JW: Mammalian drug efflux 38 Zakharova GV and Chibisov AK: Effect of poly(methacrylic transporters of the ATP−binding cassette (ABC) family: An acid) on photoinduced processes with cyanine dyes in aqueous overview. Adv Drug Deliv Rev 55(1): 3-29, 2003. solutions. High Energy Chem 35(6): 417-422, 2001. 23 Young AM, Allen CE and Audus KL: Efflux transporters of the 39 Rao TVS, Huff JB and Bieniarz Ch: Supramolecular control of human placenta. Adv Drug Deliv Rev 55(1): 125-132, 2003. photophysical properties of cyanine dyes. Tetrahedron 54: 24 Toffoli G, Simone F, Gigante M and Boiocchi M: Comparison 10627-10634, 1998. of mechanisms responsible for resistance to idarubicin and 40 Tatikolov AS and Costa SMB: Effects of normal and reverse daunorubicin in multidrug−resistant LoVo cell lines. Biochem micellar environment on the spectral properties, isomerization Pharmacol 48(10): 1871-81, 1994. and aggregation of a hydrophilic cyanine dye. Chem Phys Lett 25 Neyfakh AA: Use of fluorescent dyes as molecular probes for 346: 233-240, 2001. the study of multidrug resistance. Exp Cell Res 174: 168-176, 41 Zhang Y, Junfeng X, Yalin T, Guangzhi X and Wenpeng Y: 1998. Aggregation behaviour of two thiacarbocyanine dyes in aqueous 26 Wadkins RM and Houghton PJ: Kinetics of transport of solution. Dyes and Pigments 76: 88-93, 2008. dialkyloxacarbocyanines in multidrug-resistant cell lines 42 Sidorowicz A, Poła A and Dobryszycki P: Spectral properties of overexpressing P-glycoprotein: Interrelationship of dye alkyl 3,3’-diethyloxadicarbocyanine included in phospholipid chain length, cellular flux and drug resistance. Biochemistry 34: liposomes. J Photochem Photobiol B 38: 94-97, 1996. 3858-3872, 1995 43 Sidorowicz A, Mora C, Jabłonka S, Poła A, Modrzycka T, 27 Kanofsky JR and Sima PD: Preferential cytotoxicity for Mosiądz D and Michalak K: Spectral properties of two betaine- multidrug-resistant K562 cells using the combination of a type cyanine dyes in surfactant micelles and in the presence of photosensitizer and a cyanine dye. J Photochem Photobiol B 54: phospholipids. J Mol Struct 744: 711-716, 2005. 136-144, 2000. 44 Maruszewski K, Sidorowicz A, Poła A and Stręk W: Effect of 28 Teodori E, Dei S, Scapecchi S and Gualtieri F: The medicinal phosphatidylcholine on spectral properties and photostability of chemistry of multidrug resistance (MDR)−reversing drugs. Il 3,3’-diethyloxadicarbocyanine iodide (DODCI) – a potent Farmaco 57: 385-415, 2002. antitumor photosensitizer. J Mol Struct 444: 147-153, 1998. 29 Ford JM, Prozialeck WC and Hait WN: Structural features 45 Kitamura K, Imayoshi N, Goto T, Shiro H, Mano T and Nakai Y: determining activity of phenothiazines and related drugs for Second derivative spectrophotometric determination of partition inhibition of cell growth and reversal of multidrug resistance. coefficients of chlorpromazine and promazine between lecithin Mol Pharmacol 35(1): 105-115, 1989. bilayers vesicles and water. Anal Chim Acta 304: 101-106, 1995. 30 Bosch I and Croop J: P-glycoprotein multidrug resistance and 46 Kitamura K, Goto T and Kitade T: Second−derivative cancer. Biochim Biophys Acta 1288(2): 37-54, 1996. spectrophotometric determination of partition coefficients of 31 Wesołowska O, Molnar J, Motohashi N and Michalak K: phenothiazine derivatives between human erythrocyte ghost Inhibition of P-glycoprotein transport function by N- membranes and water. Talanta 46: 1433-1438, 1998. acylphenothiazines. Anticancer Res 22: 2863-2868, 2002. 47 Poła A, Michalak K, Burliga A, Motohashi N and Kawase M: 32 Varma MV, Ashokraj Y, Dey CS and Panchagnula R: P- Determination of lipid bilayer/water partition coefficient of new glycoprotein inhibitors and their screening: A perspective from phenothiazines using the second derivative of absorption spectra bioavailability enhancement. Pharmacol Res 48(4): 347-59, 2003. method. Eur J Pharm Sci 21: 421-427, 2004. 33 Wesołowska O, Wiśniewski J, Środa K, Krawczenko A, 48 Homolya L, Holló Z, Germann U, Pastan I, Gottesman MM and Bielawska-Pohl A, Paprocka M, Duś D and Michalak K: 8- Sarkadi B: Fluorescent cellular indicators are extruded by the Prenylnaringenin is an inhibitor of multidrug resistance- multidrug−resistance protein. J Biol Chem 268: 21493-21496, associated transporters, P-glycoprotein and MRP1. Eur J 1993. Pharmacol 644: 32-40, 2010. 49 Wang E, Casciano CN, Clement RP and Johnson WW: Active 34 Plasek J and Sigler K: Slow fluorescent indicators of membrane transport of fluorescent P-glycoprotein substrates: Evaluation as potential: a survey of different approaches to probe response markers and interaction with inhibitors. Biochem Biophys Res analysis. J Photochem Photobiol B 33: 101-124, 1996. Commun 289: 580-585, 2001. 35 Delaey D, van Laar F, De Vos D, Kamuhabwa A, Jacobs P and 50 Cabrini G and Verkman AS: Mechanism of interaction of the de Witte P: A comparative study of the photosensitizing cyanine dye DiSC3(5) with renal brush-border vesicles. J characteristics of some cyanine dyes. J Photochem Photobiol B Membrane Biol 90: 163-175, 1986. 55: 27-36, 2000. 36 Herz AH: Dye−dye interactions of cyanines in solution and at AgBr surfaces. Photogr Sci Eng 18: 232-334, 1974. 37 Avdeeva VI, Semenkina VG and Shapiro BI: A new type of Received December 21, 2012 aggregate of cyanine dyes. High Energy Chem 40(5): 348-350, Revised February 6, 2013 2006. Accepted February 6, 2013

863