2.1.

Chemicals

R123, dimethylaminostyryl-methylpyridiniumiodine (DASPMI), C ₆–NBD –ceramide and verapamil were purchased from Molecular Probes (Eugene, OR), doxorubicin from Pharmacia-Farmitalia, Nerviano, Italy. The stock solution concentrations of R123 and DASPMI were determined by spectrophotometric methods.

2.2.

Cell Culture

The human breast carcinoma MCF-7 cell line and its derived multidrug resistant subline MCF-7/DX were a kind gift of Professor M. Manfait (Reims University, Reims, France). To confirm the data obtained on these cell lines, we repeated the experiments on the sensitive and resistant MCF-7 cell lines given by Dr. G. Arancia (Istituto Superiore di Sanita`, Roma, Italy), who also kindly provided the human lung carcinoma A549, a cell line that is resistant to doxorubicin. The MCF-7/DX and A549 cell lines are known to overexpress, respectively, the Pgp-170 glycoprotein⁴⁰ and the LRP.³⁸

Other carcinoma cells were also examined: the human ovary adenocarcinoma A2780 cell line and its multidrug resistant sublines A2780/706, A2780/388, A2780/1039, the human colon adenocarcinoma LoVo cell line and its MDR subline LoVo/DX, all kindly given by Dr. Dario Ballinari (Pharmacia, Milan, Italy).

The MCF-7, A549, A2780 cells and their MDR sublines were grown as monolayers in RPMI 1640 (Hyclone, Cramlington, UK) supplemented with 10 FCS (Hyclone), 2 mM L-glutamine, and antibiotics at 37 °C, 5 CO ₂. LoVo and LoVo/DX cells were grown in Ham’s F12K medium with 10 FCS (Hyclone), 2 mM L-glutamine, and antibiotics at 37 °C, 5 CO ₂. The resistant MCF-7/DX cells were cultured in the presence of 1 μM doxorubicin, while the LoVo/DX cells were maintained in the presence of 170 nM doxorubicin. Before the use, resistant cells were grown for at least 2–4 weeks in drug-free medium.

2.3.

Sample Preparation

The cells were seeded in 35 mm Petri dishes and allowed to grow, at 37 °C and 5 CO ₂, up to 50–70 confluence. At this stage the medium was removed and the cells were washed twice with phosphate buffered saline (PBS) and incubated at 37 °C and 5 CO ₂ in the growth medium containing the appropriate dye or drug for each specific experiment. For R123 staining, cells were incubated for 10 min in 1 μM R123 solution. This staining procedure has been optimized in order to allow cell viability and a high R123 accumulation. This last requirement is essential to study mitochondria in the MDR sublines, as R123 has been shown to be a substrate for outward transport in MDR cells. In the case of DASPMI, a concentration of 10μM was used with an incubation time of 30 min. After incubation, cells were washed twice with PBS. Few microliters of PBS were added before the coverslip was placed on to prevent cell drying.

For the vital staining of the Golgi apparatus⁴¹ 50 nmol of C ₆–NBD –ceramide were dissolved in 10 mL of Hepes-buffered minimal essential medium (HMEM) containing 0.34 mg/mL of defatted BSA. The solution was dialyzed for 12 h at 4 °C against 500 mL of HMEM. A final solution of the complex C ₆–NBD –ceramide/BSA at a concentration of 5 μM was used for a 30 min incubation of MCF-7 cells. Cells were then washed twice before measurements.

For the experiment with FCCP, cells were incubated with 1 μM R123 for 10 min as described earlier. After washing, 1 mL of PBS was added to the Petri dish. A LSCFM image was collected by the Nikon 20× air objective [numerical aperature (NA)=0.4] . Then 50 μL of the FCCP stock solution was added, to reach the final concentration of 100 μM.⁴² Cells were again imaged 10 min after the addition of FCCP, in the same optical conditions. For the verapamil experiment on MCF-7/DX cells, cells were preincubated in culture medium with 100 μM verapamil for 10 min at 37 °C, 5 CO ₂. Then a small volume of a R123 solution was added to reach the final concentration of 1 μM. After 10 min of this double incubation in the presence of R123 and verapamil, cells were washed with PBS. A few microliters of a 100 μM verapamil solution were then added to the cells before the coverslip was placed on. The whole procedure was completed in about 30 min. The R123 fluorescence images were then collected immediately after.

For the experiment on MCF-7/DX cells after a short verapamil treatment, cells were incubated in a 100 μM verapamil solution for 10 min as before. Then a R123 solution was added to the cells at a final concentration of 1 μM, removing immediately the medium in excess without any washing, in order to inspect immediately the cells. The coverslip was placed on the cells and the measurement was started about 12 min after the addition of verapamil.

2.4.

Confocal Fluorescence Microscopy

The intracellular distribution of R123, DASPMI, and C ₆–NBD –ceramide were studied by laser scanning confocal microscopy using the MRC-600 microscope (Bio-Rad, Hemel Hempstead, UK). The scanning head of the MRC-600 has been coupled to an upright epifluorescence microscope Nikon Optiphot-2 (Nikon, Tokyo, Japan) with the oil immersion objective Nikon Planapochromat 60× (NA 1.4). A 20× objective was instead employed when a larger field of view was required. A 25 mW argon laser was used for fluorescence excitation at 488 nm. The fluorescence emission was detected, through a long pass filter above 515 nm, by the fast photon counting mode of the MRC-600 microscope. The high sensitivity of the photon counting detection has allowed us to employ a low laser power excitation of 0.1 mW at the entry of the optical head. This is a crucial condition for the study of active mitochondria in living cells, since mitochondria are known to be damaged by light.^{39 43} In particular, we have observed that peripheral mitochondria can be seriously damaged by continuous irradiation. For their detection it has been therefore essential to employ a scanning excitation at low power.

The R123 distribution in the cell lines MCF-7, MCF-7/DX, and A549 was found to be fully reproducible under condition of cell viability (over 500 cells examined in 10 independent experiments for the MCF-7 cell lines).

2.5.

Image Analysis

In order to see the morphology of the two different pools of mitochondria, image processing has been performed by using the LaserPix program from Bio-Rad (Hemel Hempstead, UK). Two morphological filters have been used: the white top hat filter and the black top hat filter. The white top hat filter emphasizes the brighter structures and, by enhancing the contrast it makes possible to see the organelle shape. The black top hat displays only the structures contour, setting to black the fluorescent core of the structure itself. The effect of this latter filter can be well appreciated when a gray scale inversion is applied. In this work we have used the following filters setting: Kernel size=7×7; passes=1; strength=10.

3.1.

Different Mitochondria in Sensitive and Multidrug Resistant Cells

The intracellular distribution of rhodamine 123 has been studied by LSCFM in sensitive and multidrug resistant MCF-7 carcinoma cells. In sensitive cells [Fig. 1(a)], highly fluorescent mitochondria are observed in the cytoplasm with the exclusion of the Golgi region as described in the literature.^{33 34 43} To better visualize this localization we have studied in MCF-7 cells the distribution of C ₆–NBD –ceramide [see Fig. 1(b)], a vital stain of the Golgi system. By comparing Figs. 1(a) and 1(b), it is possible to see that mitochondria are crowded around the nucleus and spread toward the cell periphery. They display polymorphic structures, with spherical, tubular, and bean-like shapes.^{35 44 45} This morphology will be better appreciated after the image analysis reported in the next paragraph.

Figure 1

Fluorescence images of sensitive human breast carcinoma MCF-7 cells: (a) active mitochondria stained by rhodamine 123 and (b) the Golgi apparatus stained by the vital fluorescent probe C ₆–NBD -ceramide.

In multidrug resistant MCF-7/DX cells, instead, an unexpected result is found (Fig. 2). In these cells no active mitochondria are observed in the central cytoplasm, but only in the subplasmalemmal region, where they constantly display a “punctate” pattern. As will be shown by the image analysis in the next paragraph, these peripheral mitochondria exhibit a peculiar morphology with a star-lobed organization. The overall R123 fluorescence intensity of these mitochondria in MCF-7/DX cells is reduced by a factor of 2 compared with that of mitochondria in sensitive cells. This is an expected result since R123 has been shown to be a substrate for the enhanced MDR efflux mechanism.^{16 18} We should also note that MCF-7/DX cells have a different morphology compared with sensitive cells, forming looser colonies and displaying several filopodia. In the next section we will discuss the possible meaning of these morphological characteristics.

Figure 2

Fluorescence images of rhodamine 123 in multidrug resistant MCF-7/DX cells.

The differences in the mitochondria of MCF-7 and MCF-7/DX cells have been confirmed by a different potentiometric probe, DASPMI. Again active mitochondria are observed through the whole cytoplasm in sensitive cells [Fig. 3(a)], whereas in resistant cells they are confined in the cell periphery [Fig. 3(b)].

Figure 3

Fluorescence images of DASPMI, a potentiometric probe of mitochondria, (a) in human breast carcinoma sensitive MCF-7 cells, and (b) in multidrug resistant MCF-7/DX cells.

We have verified the effect of a proton ionophore on R123 fluorescence in MCF-7 and MCF-7/DX cells. Since potentiometric probes are accumulated and retained by the mitochondria membrane potential, no mitochondria fluorescence is expected when the potential is dissipated. Indeed, no R123 fluorescence was detected in both cell lines after the treatment with the uncoupler FCCP (data not shown). This experiment confirms that the two potentiometric probes stain active mitochondria in sensitive and MDR cells. We have also studied the intracellular fluorescence of R123 in human colon adenocarcinoma LoVo cells and in human ovarian carcinoma A2780 cells, as well as in their MDR sublines. In these two sensitive cell lines, active mitochondria were found in the central region of the cytoplasm, displaying the morphology seen in sensitive MCF-7 cells (data not shown). None of these cells showed punctate mitochondria in the subplasmalemmal region. In the LoVo and A2780 MDR sublines, no R123 intracellular fluorescence was instead observed, a result that might be due to a very efficient R123 efflux in these MDR carcinoma cells. Apparently, MCF-7/DX cells have, also in presence of MDR, a high R123 retention that could explain the staining of peripheral mitochondria, not observed under the same conditions in the LoVo and A2780 MDR cell lines. In order to overcome this difficulty we have examined resistant cells not expressing Pgp-170.

We have, therefore, studied the R123 fluorescence in human lung carcinoma A549 cells. These cells, which display a low responsivity to doxorubicin,³⁸ are characterized by the overexpression of the LRP. As can be seen in Fig. 4, they display highly fluorescent mitochondria in the perinuclear region of the cytoplasm. These mitochondria are similar for localization and morphology to those observed in sensitive MCF-7 cells [Fig. 1(a)]. In addition, a system of mitochondria can be seen in the cell periphery (Fig. 4), with an average fluorescence intensity 4–5 times weaker than that of central mitochondria. Also in this case, to visualize the morphology of these peripheral mitochondria, we have performed the image analysis that is reported in the next paragraph.

Figure 4

(a) Fluorescence image of rhodamine 123 in human lung carcinoma A549 cells. In addition to highly fluorescent central mitochondria, mitochondria with a lower fluorescence are also present in the periphery of the cell. (b) Enlarged image of peripheral mitochondria. (c) The fluorescence image b has been processed by the black top hat filter (LaserPix software, Bio-Rad). (d) The processed image c after gray scale inversion.

3.2.

Morphology of Central and Peripheral Mitochondria

In order to compare the morphology of the central and peripheral mitochondria, image analysis procedures have been applied to the confocal R123 fluorescent images of MCF-7 and A549 cells (Figs. 1, 2, and 4). The morphological filters white top hat and black top hat (LaserPix, Bio-Rad) have been used for this study. A high magnification image of a sensitive MCF-7 cell and its image analysis are shown in Fig. 5. The well-known filamentous, bean-shaped and polymorphic morphology of mitochondria^{33 34 45} can be easily identified by the direct inspection of the fluorescent data [Fig. 5(a)], but it is better appreciated after filter treatments [Figs. 5(c)–5(e)].

Figure 5

(a) Fluorescence image of rhodamine 123 in a single sensitive MCF-7 cell. (b) Enlarged view of central mitochondria. Image analysis of b is performed: in (c) by the white top hat filter and in (d) by the black top hat filter (LaserPix software, Bio-Rad); (e) same image as in d after gray scale inversion.

In Figs. 6(a) and 6(b) the apical region of MCF-7/DX cells is reported. Here the morphology of the less fluorescent peripheral mitochondria is significantly improved by image filtering [Figs. 6(c)–6(e)]. This image analysis shows a star-lobed organization not yet reported for mitochondria in living cells.

Figure 6

(a) Confocal fluorescence image of rhodamine 123 in multidrug resistant MCF-7/DX cells. (b) An enlarged view of peripheral mitochondria. Image analysis of B is performed: in (c) by the white top hat filter (LaserPix software, Bio-Rad) and in (d) by the black top hat filter; (e) same image as in d after gray scale inversion.

The same analysis has been performed on the R123 fluorescence image of an A549 cell reported in Fig. 4. Here the morphology of peripheral mitochondria is hardly appreciated by a visual inspection of the data in Fig. 4, due to their low fluorescence intensity. The application of the black top hat filter allows to recognize a star-lobed organization also in this case. Surprisingly, it is identical to that found for peripheral mitochondria in resistant MCF-7/DX cells.

3.3.

Two Populations of Mitochondria Coexist in MCF-7/DX Cells

To explore what happens to the peripheral mitochondria when an MDR revertant is supplied to MCF-7/DX cells, we have studied the effect of verapamil, a well-known revertant of multidrug resistance. As can be seen from Fig. 7, when resistant MCF-7/DX cells are incubated with R123 in the presence of verapamil, a drastic change in the fluorescence distribution is observed: peripheral mitochondria disappear and a pool of mitochondria similar to that of sensitive cells becomes fluorescent, with an intensity at the level of sensitive cells.

Figure 7

Rhodamine 123 fluorescence in multidrug resistant MCF-7/DX cells after verapamil incubation. This treatment induces a drastic change in the system of active mitochondria: (a) in the adhesion plane peripheral mitochondria are absent; only central mitochondria are seen through the cytoplasm, as in sensitive MCF-7 cells; (b) peripheral subplasmalemmal mitochondria are absent also in the apical plane.

This result could be explained in two alternative ways. One possibility is that verapamil might induce a dislocation of peripheral mitochondria toward the center of the cell and a change in their morphology. Alternatively, two different populations of mitochondria could coexist in the cell. If this is the case, central mitochondria might be present in resistant cells, although not active.

To discriminate between these two cases, R123 fluorescence distribution has been studied in resistant cells after a short treatment with verapamil, as explained in Sec. 2. A series of z-section images has been collected in order to explore the peripheral and central regions of the cytoplasm, where the two types of mitochondria are expected to be found. As can be seen in Fig. 8, at the initial stage of the experiment, after 12 min of verapamil incubation, both peripheral and central mitochondria are simultaneously fluorescent. Peripheral mitochondria are localized in the apical region [Fig. 8(a)] while central mitochondria are evident in the adhesion plane 4 μm below [Fig. 8(b)]. At longer times (about 45 min after the addition of verapamil and 30 min after the addition of R123) the fluorescence of peripheral mitochondria disappears and only central mitochondria remain fluorescent as already observed in the experiment of Fig. 7. In a control experiment performed in absence of verapamil but under the same R123 incubation conditions (30 min), only peripheral mitochondria were observed. This excludes that the staining of central mitochondria could be due to a longer R123 incubation time.

Figure 8

Rhodamine 123 fluorescence in multidrug resistant MCF-7/DX cells after a short treatment with verapamil (10 min). Central and peripheral mitochondria are simultaneously active: (a) peripheral mitochondria are observed in the apical plane; (b) both central and peripheral mitochondria can be seen in the adhesion plane.

This experiment, therefore, seems to indicate that two populations of mitochondria are simultaneously present in multidrug resistant MCF-7/DX cells: a peripheral population of active mitochondria and a central population of mitochondria with no membrane potential, not detectable by potentiometric probes. In these cells verapamil seems to induce a dissipation of membrane potential in peripheral mitochondria, and to reset the potential of central mitochondria to the level of sensitive cells.