Photosensitizer accumulation in spontaneous

0 downloads 0 Views 314KB Size Report
E-mail: [email protected]; Fax: 39 382 506-430; Tel: 39 382 506-412 b Istituto Nazionale ... in PDT8—with MDR-1 spontaneously resistant cells has been studied. RBAc is a .... color video monitor (Trinitron, Sony Co., Tokyo, Japan).
Photosensitizer accumulation in spontaneous multidrug resistant cells: a comparative study with Rhodamine 123, Rose Bengal acetate and Photofrin® Anna C. Croce,a Rosanna Supino,b Karen S. Lanza,a Donata Locatelli,a Piero Baglioni c and Giovanni Bottiroli *a a

Centro Studio Istochimica, CNR and Dip. Biologia Animale, Università, Pavia, Italy. E-mail: [email protected]; Fax: !39 382 506-430; Tel: !39 382 506-412 b Istituto Nazionale Tumori, Milano, Italy c Dipartimento di Chimica, Università, Firenze, Italy Received 14th September 2001, Accepted 15th October 2001 First published as an Advance Article on the web

The influence both of overexpression of multidrug transporter proteins and of phenotype changes occurring in cells developing spontaneous resistance on the accumulation of photosensitizer molecules was studied on two tumor-derived cell lines (B16, A2780) expressing the MDR-1 phenotype. Rhodamine 123, Rose Bengal acetate (a fluorogenic substrate that is restored to the native active molecule by specific enzyme activity inside cells) and Photofrin® were considered. The two resistant variants accumulate Rhodamine 123 to a lesser extent than the respective wild types. Treatment with verapamil markedly enhances Rhodamine 123 accumulation in resistant cells, blocking the drug’s extrusion. The amount of Rose Bengal is larger in resistant cells than in wild type cells. Verapamil does not affect drug accumulation, although it significantly impairs the efflux process. The results are explained by the enhancement of both membrane traffic and esterase activity resulting in intracellular Rose Bengal production that counterbalances the increased ability in the outward transport of resistant cells. Photofrin® is accumulated to a lower degree in resistant than in wild type cells. Verapamil does not alter the drug accumulation, although the release process is somewhat affected. Different intracellular turnovers of Photofrin® take place in the cell variants, and the release of the monomeric fluorescent fractions is greater in resistant than in wild type cells.

Introduction Tumor cells subjected to a long exposure to chemotherapeutic agents can develop a resistance to unrelated cytotoxic compounds, which is one of the most important causes of failure in cancer chemotherapy. The overexpression of multidrug transporter proteins is a frequent mechanism exploited by the resistant cells to overcome the effect of therapeutic treatments. The multidrug transporter proteins perform an ATP-dependent extrusion of cytotoxic agents from the cytoplasm and membrane loci, thus maintaining their cellular level at a subtoxic concentration. Two phosphoglycoproteins of 170 and 190 kDa (MDR-1, P170 and MRP, P190, respectively), members of the ATP-binding cassette transporter protein family, are involved in the cytotoxic drug extrusion.1,2 The MDR-1 outward transport system exhibits affinity towards a great number of compounds, consisting of relatively large, either uncharged or weakly basic molecules. In principle, MDR-1 related resistance could affect the intracellular accumulation of photosensitizers, thus reducing the efficacy of photodynamic therapy (PDT). Possible effects of MDR-1 related resistance on the cells’ interaction with photosensitizer molecules have been considered in the literature, but have led to conflicting results.3–7 In this work the interaction of Rose Bengal acetate (RBAc)—a compound previously proposed for a new strategy in PDT 8—with MDR-1 spontaneously resistant cells has been studied. RBAc is a hydrophobic molecule obtained by combination of RB with an acetate group, which acts as a quencher and affects the photophysical and photochemical properties of RB making it a fluorogenic substrate. The intracellular accumulation of active RB molecules depends on the balance of three processes: substrate influx from culture medium to cells, cell enzyme hydrolysis, which specifically splits the quencher DOI: 10.1039/b108346e

group restoring the original properties of the photosensitizer, and the product efflux from cells to culture medium.8 Two cell lines, B16, derived from a murine melanoma, and A2780, from a human ovarian carcinoma, and their respective doxorubicin (Dx) resistant variants 9,10 were used as models. Resistant cell lines were obtained by exposure to a low dose of Dx for a long time. With this procedure the induction of resistance is supposedly similar to what occurs in tumors in vivo under chemotherapeutic treatment, and the spontaneous selection of MDR-1 cells can be accompanied by different phenotypic and functional changes. The study has been carried out by comparing the response of RBAc to the potentially unfavorable conditions of resistant cells with that of Rhodamine 123 (Rh 123), a typical substrate for MDR-1 mediated outward transport,11 and of Photofrin®, a conventional photosensitizer, which at present is the most widely used in clinical practice.12,13

Materials and methods Photosensitizers RBAc was obtained as a colorless powder by acylation of RB (Fluka Chemie, Buchs, Switzerland), according to a procedure already described,8 and stored as a stock solution (1 × 10"2 M) in dimethyl sulfoxide. Photofrin® (stock solution 2.5 mg ml"1 in water) was kindly provided by QuadraLogic Technologies, Vancouver, BC, Canada. Rh 123 (Sigma Chem. Co., St. Louis, MO) was obtained as a powder and stored as a stock solution (0.5 mg ml"1 in water). The chemicals were diluted directly in the culture medium at the time of cell incubation. Photochem. Photobiol. Sci., 2002, 1, 71–78

This journal is © The Royal Society of Chemistry and Owner Societies 2002

71

Cell culture and treatments B16 and A2780 are stabilized cell lines derived from murine melanoma and human ovarian carcinoma, respectively. The resistant cell lines (B16/Dx and A2780/Dx) were obtained from the corresponding parental lines by sequential culturing in increasing Dx concentrations. Resistance indexes toward Dx of 200 and 19 were found for B16/Dx and A2780/ Dx, respectively. MDR-1 expression was checked in fresh cells by cytofluorometric determination of the Fluorescein– isothiocyanate-conjugated antibody specific for P-170 (IgG2a k, Clone C1, YLEM, Rome, Italy). The cell cultures were maintained in an RPMI 1640 medium, supplemented with 10% fetal calf serum, 1% antibiotics, 1% glutamine (GIBCO-BRL, Gaithersburg, MD, USA), and were routinely checked for the presence of mycoplasma with a Mycoplasma Elisa detection kit (Boheringer, Mannheim, Germany). For the experiments, cells were seeded in 6-well plates containing glass slides. After 48 hours the cells were incubated with Rh 123 (1.3 × 10"6 M), RBAc (5 × 10"6 M), or Photofrin® (5 µg ml"1), diluted to the final concentration from the stock solutions directly in the incubation medium, at 37 #C, in the dark. At least three sets of separate experiments were performed for each kind of determination. At pre-set times the coverslips were removed from the incubation medium, washed with phosphate buffered saline (PBS), air-dried and submitted to microspectrofluorometric measurements. For uptake studies, 2-deoxy--glucose (2-d-Dg), rotenone, and cytochalasin B (Sigma Chem. Co.) were added to cell cultures (final concentrations = 1 × 10"2, 1 × 10"5, and 5 × 10"6 M, respectively) 15 min before the photosensitizer; verapamil (Vp, 1 × 10"5 M; Knoll Pharmaceutics, Milan, Italy) was added just before the photosensitizer. For efflux studies, 2-d-Dg was administered to the cells 15 min before, and Vp just before the end of uptake time. The drugs were present throughout the release process in the drug-free medium. Enzyme assays Esterase activity on B16 and A2780 cell lines was evaluated on single cells, fixed with acetone (60%, 0 #C, 1 min), by means of a carboxylic esterase specific fluorogenic substrate, Fluorescein diacetate, under saturation conditions, according to the endpoint cytochemical procedure.14 The apparent Km was evaluated on cell homogenates obtained from the cells grown in flasks, detached with a cell scraper, suspended in PBS, counted and centrifuged. The pellet was suspended in distilled water (6.5 × 106 cells ml"1), without any chemical adjuvant for solubilisation, to avoid possible interference with the hydrolytic activity, and submitted three times to freezing and thawing. Km values were calculated from the kinetics of the appearance of Fluorescein emission at 530 ± 10 nm, under excitation at 488 ± 10 nm (substrate concentration range: 0.5–5.0 × 10"5 M). Cell organelle staining Organelle specific stainings were performed on living cells according to the following procedures. Mitochondria: Rh 123,15 2 µg ml"1, 5 min incubation. Endosomal compartment: Lucifer Yellow,16 10 µg ml"1, 15 min incubation. Endoplasmic reticulum: 3,3$-dihexyloxacarbocyanine iodide (DiOC6),17 0.5 µg ml"1, 10 min. All the incubations were performed at 37 #C and followed by three rinsings in PBS. Chemicals were purchased from Sigma Chemical Co. Photosensitizer evaluation Photosensitizer cell accumulation was evaluated both microscopically (fluorescence intensity measured on a fixed diaphragmed area of single cells, ∅ 4 µm) and on cell homogenates (ng drug per µg protein). 72

Photochem. Photobiol. Sci., 2002, 1, 71–78

According to the excitation/emission properties of the photosensitizers, the conditions for single cell measurements were as follows. Rh 123: excitation 436 nm (436 nm interference filter: T = 35%, HBW (half-width of the band) = ± 5 nm; TK 490 dichroic mirror), fluorescence intensity evaluated as the integrated area of the spectra in the 500–580 nm region; RB: excitation 500 nm (500 nm interference filter: T = 35%, HBW = ± 5 nm; TK 540 dichroic mirror), fluorescence intensity evaluated as integrated area in the 575 ± 10 nm range; Photofrin®: excitation at 405 nm (405 nm interference filter: T = 40%, HBW = ± 5 nm; TK 450 dichroic mirror), fluorescence intensity evaluated in the 635 ± 10 nm and 665 ± 10 nm ranges. Drug extraction was performed on cells detached from the coverslips by a cell scraper, and collected with 0.4 mL of NaOH (0.05 M). Aliquots (0.05 ml) of the solutions obtained were used to determine the protein content.18 From the remaining solutions, 0.3 ml were made up to 1 ml by the addition of PBS to the extracts of cells incubated with RBAc or Rh 123. For cells treated with Photofrin®, extracts were added to the cationic detergent cetyltrimethylammonium bromide (CTAB), at a final concentration above the critical micelle concentration (1.7 × 10"3 M), to unfold the porphyrin aggregated fractions.19–21 The photosensitizer concentration was determined fluorometrically by referring to calibration curves. Spectrofluorometric measurement conditions were as follows. Rh 123: excitation (exc.) 436 nm, emission (em.) 535 ± 10 nm; RB: exc. 500 nm, em. 575 ± 10 nm; Photofrin®: exc 405 nm, em. 635 ± 20 nm. Instrumentation Microspectrofluorometric analysis was performed under epiillumination by means of a microspectrograph (Leitz, Wetzlar, Germany), equipped with an optical multichannel analyzer using a 512 element intensified linear diode array detector (OMA III, EG&G-PRA, Princeton, NJ, USA). The excitation light was provided by a 75 W xenon lamp and by a 100 W mercury lamp, both with KG1 and BG38 antithermal filters, for RB or Rh 123 and Photofrin®, respectively. All measurements were performed with a 40× objective (Leitz, N.A. 0.6). Cell fluorescence images were recorded by means of a digital system Argus 100 VIM photon counting processor (Hamamatsu Photonics Deutschland GmbH, Herrsching am Ammersee, Germany) using a Hamamatsu C2400–09 ISIT camera mounted on a Leitz Orthoplan fluorescence microscope. Excitation and emission conditions are reported in the figure legends. All images were acquired with a 100× oil immersion iris objective (Leitz Neofluar, N.A. 1.32–0.60). To minimize photobleaching, the cells were centered under low-intensity transmitted light and the measuring time did not exceed 1 s. Photographs were obtained by means of a freeze frame-video (Polaroid, Cambridge, MA, USA) connected to a color video monitor (Trinitron, Sony Co., Tokyo, Japan). Cell homogenate measurements were performed by means of a spectrofluorometer (model SP-2; Applied Photophysics, London, UK) equipped with a single photon counting system (EG&G-Ortec, Oak Ridge, TN, USA). Statistical analysis The single cell data are presented as mean values ± SE (standard error). At least 50 cells were measured for each point of three repeated experimental sets. Significance levels are evaluated by means of the Mann–Whitney U test (normal distribution, Z statistic). For the cell extracts the mean values and the ranges of three repeated experiments are reported.

Table 1 Intracellular contents of Rh 123, RBAc, and Photofrin® in B16 and A2780 wild types and resistant variants at 120 min incubation time. Fluorescence intensity (FI, mean values ± SE) of resistant cells are reported as percentages of wild type cells. Cell extract values are reported as the overall amount per protein weight and as percentages of resistant vs. wild type cells a

Cell lines

Single cell FI 540 ± 40nm

B16 100 ± 5.3 B16/Dx 39 ± 3.1 c A2780 100 ± 4.1 A2780/Dx 42 ± 3.8 c

Photofrin®

Rose Bengal b

Rh 123 (Cell extract drug/ng)/ (protein/µg)

Single cell FI 575 ± 10nm

(Cell extract drug/ng)/ (protein/µg)

Single cell FI 635 ± 10nm

(Cell extract drug/ng)/ (protein/µg)

10.29 (9.50–10.80) 5.60 (4.80–5.90) (54%) 15.30 (14.09–16.00) 8.56 (7.92–8.98) (56%)

100 ± 3.6 130 ± 3.4 c 100 ± 4.2 119 ± 4.7 c

5.40 (4.80–10.00) 10.40 (9.80–11.00) (192%) 3.80 (3.30–4.20) 5.89 (5.30–6.20) (155%)

100 ± 6.8 70 ± 4.2 c 100 ± 5.7 82 ± 3.1 c

2.35 (2.20–2.43) 2.11 (17.80–24.50) (90%) 1.78 (1.68–1.88) 1.68 (1.59–1.77) (94%)

Ratio of cells protein content (µg 10"6 cells) vs. cell volume: B16 = 0.81, B16/Dx = 0.56, A2780 = 0.72, A2780/Dx = 0.56. b Under the conditions here used, the RB fluorescence intensity values measured on the cells can be directly related to the photosensitizer content, due to the hydrophobic effect exerted by proteins on RB aggregation.8 Statistics for FI values, Mann–Whitney U test, Z statistic: resistant vs. respective wild type. c p ≤ 0.01.

a

Results Photosensitizer accumulation in B16 and A2780 cells Rhodamine 123. A quantitative analysis of the intracellular amount of Rh 123 in B16 and A2780 cell lines was performed after 120 min of incubation (Table 1). The capability of photosensitizer accumulation, evaluated both as drug amount per protein content of cell extracts and as fluorescence intensity per fixed area of single cells, is higher in wild type than in resistant cells, for both cell lines. The cell fluorescence intensity increases with the photosensitizer incubation time faster in wild type than in resistant cells for both lines (Fig. 1A and B). When the cells are trans-

Fig. 2 Time course of the persistence of Rh 123 fluorescence in cells loaded for 120 min, and transferred in a drug-free medium in B16 (!), B16/Dx (∆) cells (A), and in A2780 (!), A2780/Dx (∆) cells (B); untreated (——) and Vp-treated (- - -). Each point of the curves is the mean of three independent experiments. SE ≤ 5%.

retention ability similar to or greater than that of the respective wild type cells. In wild type cells Vp does not induce significant changes in the Rh 123 fluorescence values, apart from a slight reduction during the uptake.

Fig. 1 Time course of the appearance of Rh 123 intracellular fluorescence in B16 (!), B16/Dx (∆) cells (A), and in A2780 (!), A2780/Dx (∆) cells (B); untreated (——) and Vp-treated (- - -). Each point of the curves is the mean of three independent experiments. SE ≤ 5%.

ferred in a drug-free medium, the fluorescence intensity decreases more rapidly in resistant cell variants than in the corresponding wild type lines (Fig. 2A and B). Administration of Vp, a specific MDR-1 reversing agent,22 results in an enhancement of the Rh 123 accumulation process in both B16/Dx and A2780/Dx cells, over all the incubation time. As to the release process, the effect of Vp is more evident in the initial stages considered, at which resistant cells exhibit a

Rose Bengal acetate. At 120 min incubation, larger amounts of RB are retained in B16/Dx and A2780/Dx than in the respective wild type cells (Table 1). During RBAc incubation (Fig. 3A and B) the fluorescence intensity shows a rapid linear increase in the first 20 min, followed by a slower trend. A fluorescence intensity higher in resistant than in the wild type cells is found at each incubation time considered. The kinetics of RB efflux exhibit a biphasic trend in B16 cell lines, both wild type and resistant variant (Fig. 4A). During the first 10 minutes a fast drug release is observed, and the slope of the first part of the curve is slightly higher in B16/Dx than in the B16 cell line. The slower part of the process is accompanied by a partial recovery of the fluorescence values at 30 minutes, which is more evident in wild type than in resistant cells. In A2780/Dx cells the release process shows a decrease of the fluorescence intensity of about 30% at 15 min, while no significant reduction of the fluorescence intensity is observed in the respective wild type (Fig. 4B). Vp treatment does not affect the trend of the appearance of RB fluorescence in wild cells. In resistant cells, Vp treatment Photochem. Photobiol. Sci., 2002, 1, 71–78

73

results in a reduction at short incubation times, followed by a small increase in the appearance of RB fluorescence (Fig. 3A and B).

Fig. 3 Reaction progress curves for the increase of intracellular fluorescence intensity upon RBAc incubation of B16 (!), B16/Dx (∆) cells (A), and in A2780 (!), A2780/Dx (∆) cells (B); untreated (——) and Vp-treated (- - -). Each point of the curve is the mean of three independent experiments. SE ≤ 5%.

As to the release process, Vp treatment favors RB retention only in the resistant variants (Fig. 4A and B). For up to 30 min of release the fluorescence values in Vp-treated B16/Dx cells are about twice than those of untreated B16/Dx cells, and even greater than those of the wild type. In A2780/Dx Vp increases the fluorescence intensity of resistant cells to values similar to those of the wild type cells.

Fig. 4 Time course of the persistence of RB fluorescence in cells incubated with RBAc for 120 min, and transferred in a drug-free medium. B16 (!), B16/Dx (∆) cells (A), and A2780 (!), A2780/Dx (∆) cells (B); untreated (——) and Vp-treated (- - -). Each point of the curve is the mean of three independent experiments. SE ≤ 5%.

74

Photochem. Photobiol. Sci., 2002, 1, 71–78

Treatment with 2-d-Dg—a poison that reduces the availability of metabolic energy, thus affecting the membrane traffic and internalization processes of extracellular molecules— results in a reduction in RB intracellular accumulation, more evident in the B16/Dx than in the B16 cell line (Fig. 5A and B).

Fig. 5 Effect of metabolic poisons on the reaction progress curves, upon RBAc treatment, and on the persistence, upon transfer in an RBAc-free medium, of RB intracellular fluorescence in B16 cells (A) and B16/Dx (B) cells, (!, ∆) untreated, (- -"- -) 2-d-Dg and (- -#- -) cytochalasin B treated cells. Each point of the curve is the mean of three independent experiments. SE ≤ 5%.

Similar but lesser effects are induced by treatment with cytochalasin B, a poison that affects the actin-based micro-filament system involved in membrane traffic.16 On RB-loaded cells transferred in a RBAc-free medium, 2-d-Dg favors the photosensitizer retention. This is more evident in the B16/Dx than in the B16 cell line. Similar results were induced also by rotenone, an inhibitor of the respiratory chain (data not shown). Since, apart from the balance between the influx of the substrate and product efflux, the intracellular accumulation of RB depends on intracellular esterase activity, the hydrolytic properties of the cells were also considered. Esterase cytochemical assay shows that the enzyme activity is increased by about 50% more in B16/Dx cells than in B16, while the overall substrate– enzyme affinity properties (Km) are quite comparable. In A2780/ Dx cells, the esterase activity is about 20% higher than in the wild type, while the overall substrate–enzyme affinity is reduced (Table 2). Photofrin®. At 120 min incubation, the capability of photosensitizer accumulation, evaluated both as drug amount per protein content on cell extracts and as fluorescence intensity per fixed area on single cells, is higher in wild type than in resistant cells, for both B16 and A2780 cell lines. In A2780 cells the amount of photosensitizer at all the incubation times considered is larger in wild type than in resistant cells (Fig. 6A and C). The (665/635) nm ratio values, which give information on the balance between unfolded oligomers and monomers, are higher in resistant than in wild type cells at all the incubation times (Fig. 6B). In B16 cells the trend of the 635 nm cell fluorescence vs. incubation time is more complex, the signal amplitude at short incubation time being higher in resistant than in wild type cells (Fig. 7A and C). The (665/635)

Table 2 Esterase activity characterisation in B16 and A2780 wild types and resistant cell variants. Esterase activity (arbitrary unit (AU) mean value ± SE) from resistant variant vs. respective wild type cells are evaluated by means of a cytochemical assay. Km values (mean and range values) are evaluated from cell homogenates (details are reported in Materials and methods) a

a

Cell line

B16

B16/Dx

A2780

A2780/Dx

Esterase activity Km/M

100 ± 2.1 0.91 × 10"5 (0.98–0.84 × 10"5)

150 ± 3.2 b 1.2 × 10"5 (1–1.37 × 10"5)

100 ± 1.8 0.6 ± 0.05 × 10"5 (0.58–0.72 × 10"5)

122 ± 2.2 c 3.3 ± 0.01 × 10"5 (3.1–3.5 × 10"5)

Statistics: Mann–Whitney U test, Z statistic: resistant variant vs. respective wild type. b p ≤ 0.01. c p ≤ 0.05.

Fig. 6 Time course of (A) the appearance of Photofrin® intracellular 635 nm fluorescence in A2780 (!) and A2780/Dx (∆) cells, and of (B) (665/635) nm ratio values; (——) untreated and Vp-treated (- - -) cells. For each experimental set the ratio value is obtained as the mean of the ratio values calculated from single spectra at each experimental time. Each point of the curves is the mean of three independent experiments. SE ≤ 5% for fluorescence intensity measurements; SE ≤ 3% for ratio values. (C) Time course of the intracellular overall amount of Photofrin® per protein content in A2780 (!) and A2780/Dx (∆) cells. Error bars show the range of the measured values.

Fig. 7 Time course of the appearance of (A) Photofrin® intracellular 635 nm fluorescence in B16 (!) and B16/Dx (∆) cells, and of (B) 665/ 635) nm ratio values; (——) untreated and Vp-treated (- - -) cells. For each experimental set the ratio value is obtained as the mean of the ratio values calculated from single spectra at each experimental time. Each point of the curves is the mean of three independent experiments. SE ≤ 5% for fluorescence intensity measurements; SE ≤ 3% for ratio values. (C) Time course of the intracellular overall amount of Photofrin® per protein content in B16 (!) and B16/Dx (∆) cells. Error bars show the range of the measured values.

nm ratio values always increase with incubation time in resistant cells, while in wild type cells an increase is observed only up to 30 min (Fig. 7B). Vp treatment during Photofrin® uptake does not affect significantly both the 635 nm cell fluorescence intensity and the trend of (665/635) nm ratio values for either wild type or resistant variants of the B16 and A2780 lines. Treatment with cytochalasin and 2-d-Dg slightly reduced the accumulation of Photofrin® in both the B16 and B16/Dx cells. The values of the (665/635) nm fluorescence intensity ratio did not exhibit appreciable changes (data not shown).

The photosensitizer intracellular retention was studied on cells loaded with Photofrin® for 60 and 120 min, transferred in a drug-free medium. In B16, the 60 min loaded cells show a reduction of the 635 nm fluorescence intensity that is more pronounced in the resistant than in the wild type cell line; the (665/635) nm ratio values are comparable for both wild type and resistant cells (Fig. 8A and B). In the 120 min loaded cells the fluorescence intensity reduction is faster in the wild type than in the resistant cells; the ratio values exhibit appreciable changes during the efflux process only in resistant cells (Fig. 8C and D). Photochem. Photobiol. Sci., 2002, 1, 71–78

75

on the extracellular medium, confirm these findings (data not shown). Photosensitizer localization Treatment with RBAc results in an intracellular orange fluorescence localized mainly in the cytoplasm, with a perinuclear accumulation in both B16 and B16/Dx cells, these latter being characterized by a more pronounced polarized distribution. A diffusion of the fluorescence all over the cytoplasm is observed at longer incubation times, which is more evident in the case of the B16 cells (Fig. 10A–D).

Fig. 8 635 nm fluorescence persistence (A, C) and (665/635) nm ratio values (B, D) in 60 min (A, B) and 120 min (C, D) Photofrin® loaded B16 cells, transferred in a drug-free medium. (!) B16, (∆) B16/Dx cells; (——) untreated, (- - -) Vp-treated cells. Ratio values calculated as described in Fig. 7. Each point of the curve is the mean of three independent experiments. SE ≤ 5% for fluorescence intensity measurements; SE ≤ 3% for ratio values.

In A2780, a slower decrease of the 635 nm fluorescence values in the resistant variant than in the wild type is observed for cells loaded for both 60 and 120 min. The (665/635) nm ratio values are higher in resistant than in wild type cells, mainly in 120 min loaded cells (Fig. 9A–D).

Fig. 10 Typical fluorescence distribution patterns of RB (A–D) and porphyrin (E–H) in B16 (A, B, E, F) and B16/Dx cells (C, D, G, H) incubated with 5 × 10"6 M RBAc or 5 µg ml"1 Photofrin® for 60 min (A, C, E, G) and 120 min (B, D, F, H). Exc.: 500 nm interference filter, em. > 540 nm (RB); exc.: 405 nm interference filter, em. > 600 nm (Photofrin®). The sensitivity of the camera was adjusted according to the fluorescence signal. Scale bar represents 14 µm.

In the case of Photofrin® a marked redistribution of the fluorescence occurs during incubation. At 60 min incubation Photofrin® fluorescence is mainly localized on the plasma membrane in most cells, both wild type and resistant (Fig. 10E and G). At 120 min incubation, a diffuse cytoplasmic fluorescence, along with some brighter areas, is observed in both the cell lines (Fig. 10F–H). The typical fluorescence patterns of cell constituents obtained upon cytochemical specific staining are shown in Fig. 11A–F. Mitochondria seem less abundant in B16/Dx cells, where they are localized preferentially around the nuclei, whereas in B16 they are diffused in the cytoplasm. The endosomal compartment and endoplasmic reticulum show quite similar patterns. The former exhibits a cytoplasmic, polar localization around the nucleus in both B16/Dx and B16 cells. The latter exhibits a more defined and homogeneous localization around the nucleus, with a reticular pattern particularly evident in B16 resistant cells.

Discussion

Fig. 9 635 nm fluorescence persistence (A, C) and (665/635) nm ratio values (B, D) in 60 min (A, B) and 120 min (C, D) Photofrin® loaded A2780 cells, transferred in a drug-free medium. (!) A2780, (∆) A2780/Dx cells; (——) untreated, (- - -) Vp-treated cells. Ratio values calculated as described in Fig. 7. Each point of the curve is the mean of three independent experiments. SE ≤ 5% for fluorescence intensity measurements; SE ≤ 3% for ratio values.

Vp treatment during the release process favors the persistence of 635 nm fluorescence intensity in the resistant variants of the two cell lines, for both 60 and 120 min loaded cells. No appreciable effects are observed for the wild types. Fluorescence intensity measurements of the released porphyrin, performed 76

Photochem. Photobiol. Sci., 2002, 1, 71–78

The cell interaction of photosensitizer molecules characterized by different physico-chemical properties was investigated on two tumor-derived cell lines (B16, A2780), where spontaneous selection for Doxorubicin resistance resulted in MDR-1 overexpression as well as in other phenotype changes. Rh 123, the typical substrate for the outward transport system associated with MDR-1, is accumulated by resistant cells to a much lesser extent than the wild type ones, and Vp treatment induces an enhancement of Rh 123 accumulation in resistant cells to amounts quite comparable to those of wild type cells. These results are explained by the Vp impairment of the marked drug efflux process in resistant cells, and confirm the MDR-1 activity of our resistant cell models. RBAc treatment results in a significantly greater accumulation of RB in B16/Dx and A2780/Dx than in wild type cells.

Fig. 11 Typical localization patterns of the intracellular constituents in B16 (A, C, E) and B16/Dx cells (B, D, F) recorded for living cells. Mitochondria: Rh 123, 2 µg ml"1, 5 min incubation, exc. UG1 filter (380–420 nm), em. > 510 nm (A, B); Endoplasmic reticulum: DiOC6, 0.5 µg ml"1, 10 min incubation, exc. BG12 filter (400–480 nm), em. > 530 nm, (C, D); endosomal compartment: Lucifer Yellow, 10 µg ml"1, 15 min incubation, exc. UG1 filter (380–420 nm), em. > 510 nm, (E, F). The sensitivity of the camera was adjusted according to the fluorescence signal. Bar = 14 µm.

These data are in apparent contrast with the increased ability of resistant cells to release the product. The marked impairment of RB efflux induced by VP only in the resistant cells indicates the participation of MDR-1 in this process, despite the anionic nature of RB. A similar result has already been shown in the case of a fluorogenic substrate, the acetoxymethyl (AM) ester derivative of the Fluorescein analog calcein.23 In fact, RB can act as an amphipatic molecule and partition into lipids.24 On this basis the RB interaction with P-170 cannot be excluded, considering the possibility that P-170 can act as a flippase, driving the cell extrusion of uncharged molecules localized in membrane bilayers.1,25,26 The MDR-1 multidrug transporter was also reported to mediate the extrusion of the hydrophobic (AM) calcein esters before the intracellular hydrolysis of the fluorogenic substrates took place.27,28 A similar activity is confirmed by the finding that the extent of RB fluorescence recovery during the efflux process attributable to the hydrolysis of the fluorogenic substrate accumulated in the cells is less in the resistant than in the respective wild type cells. The greater intracellular accumulation of RB in resistant cells during uptake indicates that the improved extrusion processes are largely counterbalanced by other phenomena, such as substrate internalization and/or intracellular hydrolysis. The role of the substrate internalization process is supported by 2-dDg, which induces a more marked impairment of RB accumulation in B16/Dx than in the wild type. The result indicates that, apart from a diffusion process, a mediated transport, e.g. adsorptive endocytosis, is more efficient in resistant cells favoring the internalization of the hydrophobic substrate. This suggestion is in agreement with an enhancement in the mediated transport activity, related to increases in both membrane fluidity and plasma membrane traffic, already reported for pleiotropic anthracycline resistant cells.29,30 The involvement of the membrane traffic in RBAc internalization may also explain why the amount of RB accumulated in resistant cells is less than could be expected under Vp treatment. This result might be due to the ability of Vp to impair the plasma membrane traffic.29 Further support of an increase of mediated transport in resistant cells is provided by the RB intracellular localization: 31 the polar pattern ascribable to the endosomal compartment is more pronounced in B16/Dx than in B16 cells. Moreover, the mediated internalization pathway results in a delivery of the substrate to sites where hydrolytic activity is exerted. As a consequence, more pronounced RB production is expected in resistant than in wild-type cells, since esterase activity is higher

in the former than in the latter cells. In fact, a similarity in the extent of the increase in hydrolytic activity with that of the amount of RB was found in both B16 and A2780 resistant cells. It is worthy of note that changes in enzyme activities, for example, catechol oxidase,32 have already been found to accompany the induction of Dx resistance. Photofrin® is accumulated to a lower extent in resistant than in wild type cells. The differences are less pronounced when the 635 nm fluorescence intensity is considered with respect to the drug amount per protein content. This can be explained by the chemico-physical nature of Photofrin®,12,19–21 and indicates that relatively larger amounts of both partially unfolded 665 nm fluorescent and highly aggregated non-fluorescent porphyrins, with respect to the 635 nm fluorescent monomeric species, are present in wild type than in resistant cells. The relatively higher occurrence of monomers in resistant cells can be related to a more active unfolding process, as indicated by the increase in the (665/635) nm ratio values during the drug treatment. The influence of the disaggregation on the persistence of the 635 nm cell fluorescence was demonstrated by the efflux studies. In particular, in B16 cells after 60 min incubation—when the (665/635) nm ratio values are comparable for both cell lines— the monomer release is faster in resistant than in wild type cells. This fact induces a continuing relative enrichment of the aggregate fraction, resulting in an increase of the (665/635) nm ratio. In 120 min loaded cells—when the (665/635) nm ratio values are greater in resistant than in wild type cells—the monomer release in B16/Dx cells is relatively slow at short times, and increases at long times. This behavior indicates that the availability of the monomer for the efflux process depends on the unfolding process, and can be mediated by the intracellular localization of porphyrins. In fact, the cytoplasmic localization of porphyrins after 120 min of incubation supports the idea that the monomerization process is followed by diffusion inside the cell, before release. Different cell mechanisms may be involved in the monomer release process, such as extrusion related to the membrane recycle, passive diffusion 20,33 and P-170 activity. At least a partial involvement of the MDR-1 transporter is supported by Vp treatment, which improves the persistence of the 635 nm fluorescence only in the resistant cells. The association of porphyrin species with plasma membrane, and the interaction of Photofrin® with the glycoprotein responsible for MDR-1, have been already reported.34,35 In conclusion, the results obtained show that the retention of the photosensitizers is influenced to different degrees by both MDR-1 outward transport activity and phenotype changes accompanying the resistance induction. The accumulation of Rh 123, a weakly cationic molecule, is greatly influenced by the MDR-1 expression of resistant cells. In the case of Photofrin®, the lower photosensitizer accumulation in resistant than in wild type cells seems to be mainly related to the drug intracellular turnover, although a partial MDR-1 effect on monomer extrusion cannot be excluded. The increased accumulation of RB in resistant cells is the result of multiple processes, and can be mainly ascribed to the improvement in the membrane traffic and to the increase in enzyme activity, which counterbalance the effects on the extrusion process greatly related to MDR-1 resistance.

Acknowledgements This work has been supported by the Italian National Research Council, C.N.R. The authors are grateful to Mrs Paola Veneroni, Dipartimento di Biologia Animale, Università, Pavia, for cultivating cells, and to Dr Marilena Colantuono, Istituto Tumori Milano, Division of Experimental Oncology: B, for assistance in cell treatment. Photochem. Photobiol. Sci., 2002, 1, 71–78

77

References 1 J. A. Endicott and V. Ling, The biochemistry of P-glycoproteinmediated multidrug resistance, Annu. Rev. Biochem., 1989, 158, 137–171. 2 S. P. C. Cole, G. Bhardwaj, J. H. Gerlach, J. E. Mackie, C. E. Arant, K. C. Almquist, A. J. Stewart, E. U. Kurtz, A. M. V. Duncan and R. G. Deeley, Overexpression of a novel transporter gene in multidrug resistant human lung cancer cell line, Science, 1992, 58, 1650–1654. 3 D. Kessel and C. Erickson, Porphyrin photosensitization of multidrug resistant cell types, Photochem. Photobiol., 1992, 55, 397–399. 4 S. F. Purkiss, M. F. Grahn and S. Williams, Haematoporphyrin derivative-photodynamic therapy of colorectal carcinoma, sensitized using verapamil and adriamycin, Surg. Oncol., 1996, 5, 169–175. 5 J. B. Mitchell, E. Glastein, K. H. Cowan and A. Russo, Photodynamic therapy of multi-drug resistant cell lines, Proc. Am. Assoc. Cancer Res., 1988, 29, A315. 6 G. Singh, B. C. Wilson, S. M. Sharkey, G. P. Browman and P. Deschamps, Resistance to photodynamic therapy in radiation induced fibrosarcoma-1 and chinese hamster ovary-multi-drug resistant cells in vitro, Photochem. Photobiol., 1991, 54, 307– 312. 7 R. Marchesini, A. Gritti, A. Colombo, T. Dasdia and A. E. Sichirollo, Effectiveness of photodynamic therapy after Photofrin II sensitization in a multi-drug resistant human breast carcinoma cell line in vitro, 3rd Biennial Meeting of the International Photodynamic Association, Buffalo, New York, 1990, IV/1. 8 G. Bottiroli, A. C. Croce, P. Balzarini, D. Locatelli, P. Baglioni, P. Lo Nostro, M. Monici and R. Pratesi, Enzyme-assisted cell photosensitization: A proposal for an efficient approach to tumor therapy and diagnosis. The Rose Bengal fluorogenic substrate, Photochem. Photobiol., 1997, 66, 374–383. 9 R. Supino, E. Prosperi, F. Formelli, M. Mariani and G. Parmiani, Characterization of a doxorubicin-resistant murine melanoma line: studies on cross-resistance and its circumvention, Br. J. Cancer, 1986, 54, 33–42. 10 B. Stefanska, M. Dzieduszycka, M. M. Bontemps-Gracz, E. Borowski, S. Martelli, R. Supino, G. Pratesi, M. de Cesare, F. Zunino, H. Kusnierczyk and C. Redzikowski, 8,11-dihydroxy-6[(aminoalkyl)amino]-7H-benzo[e]perimidin-7-ones with activity in multidrug-resistant cell lines: synthesis and antitumor evaluation, J. Med. Chem., 1999, 42, 3494–3501. 11 H. Tapiero, J. N. Munck, A. Fourcade and T. J. Lampidis, Cross resistance to rhodamine 123 in adriamycin- and daunorubicinresistant Friend leukemia cell variants, Cancer Res., 1984, 44, 5544–5549. 12 J. Moan, S. Sandberg, T. Christensen and S. Elander, Hematoporphyrin derivative: chemical composition, photochemical and photosensitizing properties, in Porphyrin Photosensitization, ed. D. Kessel and T. J. Dougherty, Plenum Press, New York and London, 1983, pp. 165–180. 13 T. J. Dougherty, C. J. Gomer, B. W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan and Q. Peng, Photodynamic therapy, J. Natl. Cancer Inst., 1998, 90, 889–905. 14 Z. Lojda, R. Gossrau and T. H. Schiebler, in Enzyme Histochemistry: A Laboratory Manual, Springer Verlag, Berlin, Heidelberg, New York, 1979. 15 L. V. Johnson, M. L. Walsh and L. B. Chen, Localization of mitochondria in living cells with Rhodamine 123, Proc. Natl. Acad. Sci. U. S. A., 1980, 77, 990–994. 16 E. Holtzman, Cellular Organelles, in Lysosomes, ed. P. Siekevitz, Plenum Press, New York and London, 1989. 17 R. W. Sabnis, T. G. Deligeorgiev, M. N. Jachak and T. S. Dalvi, DiOC6 (3): a useful dye for staining the endoplasmic reticulum, Biotech. Histochem., 1997, 72, 253–258. 18 O. H. Lowry, B. J. Rosebrough, A. I. Farr and R. J. Randall, Protein

78

Photochem. Photobiol. Sci., 2002, 1, 71–78

19

20 21

22

23

24

25 26 27

28

29

30

31 32

33 34

35

measurements with the Folin phenol reagent, J. Biol. Chem., 1951, 193, 265–275. R. W. Redmond, E. J. Land and T. G. Truscott, Aggregation effects on the photophysical properties of porphyrins in relation to mechanisms involved in photodynamic therapy, in Methods in Porphyrin Photosensitization, ed. D. Kessel, Plenum Press, New York and London, 1984, pp. 293–302. G. Bottiroli, R. Ramponi and A. C. Croce, Quantitative analysis of intracellular behaviour of porphyrins, Photochem. Photobiol., 1987, 46, 663–667. R. Cubeddu, R. Ramponi and G. Bottiroli, Disaggregation effects on hematoporphyrin derivative in the presence of surfactants at different concentrations: temperature dependence, Proc. SPIE-Int. Soc. Opt. Eng., 1987, 701, 316–319. T. Tsuruo, H. Iida, S. Tsukagoshi and Y. Sakurai, Overcoming of vincristine resistance in P388 leukemia in vivo and in vitro through enhanced cytotoxicity of vincristine and vinblastine by verapamil, Cancer Res., 1981, 41, 1967–1972. G. Liminga, P. Nygren, S. Dhar, K. Nilsson and R. Larsson, Cytotoxic effect of calcein acetoxymethyl ester on human tumor cell lines: drug delivery by intracellular trapping, Anti-Cancer Drugs, 1995, 6, 578–585. T. A. Dahl, O. Valdes-Aguilera, W. R. Midden and D. C. Neckers, Partition of Rose bengal anion from aqueous medium into a lipophilic environment in the cell envelope of Salmonella typhimurium: implications for cell-type targeting in photodynamic therapy, J. Photochem. Photobiol., B, 1989, 4, 171–184. M. M. Gottesman and I. Pastan, Biochemistry of multidrug resistance mediated by the multidrug transporter, Annu. Rev. Biochem., 1993, 62, 385–427. F. Higgins and M. M. Gottesman, Is the multidrug transporter a flippase?, Trends Biochem. Sci., 1992, 17, 18–21. L. Homoloya, Z. Hollo, U. Germann, I. Pastan, M. M. Gottesman and B. Sarkadi, Fluorescent cellular indicators are extruded by the multidrug resistance protein, J. Biol. Chem., 1993, 268, 21493– 21496. C. Marbeuf-Gueye, M. Salerno, P. Quidu and A. Garnier-Suillerot, Inhibition of the P-glycoprotein- and multidrug resistance proteinmediated efflux of anthracyclines and calceinacetomethyl ester by PAK–104P, Eur. J. Pharmacol., 2000, 391, 207–216. M. Sehested, T. Skovsgaard, B. van Deurs and H. Winther-Nielsen, Increased plasma membrane traffic in daunorubicin resistant P388 leukaemic cells. Effect of daunorubicin and verapamil, Br. J. Cancer, 1987, 56, 747–751. M. Sehested, T. Skovsgaard, B. van Deurs and H. Winther-Nielsen, Increase in nonspecific adsorptive endocytosis in Anthracycline and vinca alkaloid-resistant, Ehrlich ascite tumor cell lines, J. Natl. Cancer Inst., 1987, 78, 171–179. G. Bottiroli, A. C. Croce, M. Biggiogera, K. S. Lanza, S. Fiorani and D. Locatelli, Photosensitizer damage targets in Rose Bengal acetate treated cells, Lasers Surg. Med., 2001, Suppl. 13, 41. R. Supino, M. Mariani, A. Colombo, E. Prosperi, A. C. Croce and G. Bottiroli, Comparative studies on the effects of doxorubicin and differentiation inducing agents on B16 melanoma cells, Eur. J. Cancer, 1992, 28A, 778–783. A. C. Croce, E. Wyroba and G. Bottiroli, Uptake and distribution of haematoporphyrin derivative in the unicellular eukaryote Paramecium, J. Photochem. Photobiol., B, 1990, 6, 405–417. W. S. L. Strauss, R. Sailer, M. H. Gschwend, H. Emmert, R. Steiner and H. Schneckenburger, Selective examination of plasma membrane-associated photosensitizers using total internal reflection fluorescence spectroscopy: correlation between photobleaching and photodynamic efficacy of protoporphyrin IX, Photochem. Photobiol., 1998, 67, 363–369. S. L. Gibson, M. K. Al-Shavi, A. E. Senior and R. Hilf, Inhibition of the ATPase activity of P-glycoprotein photosensitization of multidrug-resistant cells in vitro, Photochem. Photobiol., 1995, 61, 390–396.