Dose-response studies indicate that concentrations of 2.5 μM and higher sanguinarine suppress the growth of MCF-7 cells (Fig. ). There appears to be a relatively narrow window of growth-suppressive activity, with concentrations less than 2.5 μM exhibiting little effect, and 10 μM or higher resulting in a significant level of cell death in this cell line (data not shown). In addition, the effectiveness of sanguinarine is strongly influenced by cell density, with higher density cultures exhibiting a noticeable resistance to the same concentrations of sanguinarine effective in lower density cultures. Under the conditions used for this study, a single dose of 5 μM sanguinarine essentially completely suppressed MCF-7 proliferation for at least six days (Fig. ), while inducing only modest levels of cell death.
Figure 1 Dose-response characteristics of MCF-7 cells treated with sanguinarine. Responses of cells to a single dose of sanguinarine, showing effective suppression of cell growth in cultures receiving 5 or 7.5 μM sanguinarine. Data points are expressed (more ...)
To test for effects on certain cell cycle regulatory molecules, cultures were treated with 5 μM sanguinarine, fixed, and immunolabeled with various antibodies. Interestingly, changes in the localization patterns of cyclin D1 and topoisomerase II were noted. In control cells, cyclin D1 immunoreactivity is largely restricted to nuclei, which exhibit a range of labeling intensities (Fig. ). In cultures treated with 5 μM sanguinarine for 24 hours, however, punctate cytoplasmic deposits of immunoreactive material are evident, along with a general decrease in nuclear labeling (Fig. ). To test how persistent this cytoplasmic relocalization of cyclin D1 is, cells were treated with sanguinarine for 24 hours, followed by drug removal and recovery for 48 hours in drug-free medium. These experiments demonstrate that cytoplasmic foci of cyclin D1 are still present in many cells after two days of drug recovery (Fig. ).
Figure 2 Sanguinarine induces long-term disruptions in cyclin D1 trafficking. Cultures were fixed and immunolabeled with an antibody to cyclin D1 (green) and counterstained with Hoechst to visualize DNA (blue). DIC, differential interference microscopy. (a-c) (more ...)
The distribution of topoisomerase II is also altered following sanguinarine treatment (Fig. ). In control cells, topoisomerase II immunoreactivity is weakly distributed throughout the cytoplasm, and is also present in the nucleus of many cells; some cells display very strong nuclear labeling (Fig. ). Similar to the results for cyclin D1, cytoplasmic deposits of topoisomerase II become apparent after 24 hours of treatment with 5 μM sanguinarine (Fig. ). In many cultures, this redistribution of topoisomerase II is even more pronounced than that of cyclin D1. Careful examination of double-labeled cells indicates that most, but not all, of the cyclin D1 and topoisomerase cytoplasmic immunoreactive foci appear to be co-localized.
Figure 3 Topoisomerase II, but not NuMA, is also aberrantly localized following sanguinarine treatment. Cultures were fixed and double-labeled with antibodies to topoisomerase II (green), NuMA (red), and counterstained with Hoechst to visualize DNA (blue). DIC, (more ...)
To examine whether sanguinarine causes a non-specific redistribution of nuclear proteins in general, cultures were also labeled with antibodies to the nuclear matrix proteins lamin B1 and NuMA. Changes in the location or overall labeling pattern were not evident for either NuMA (Fig. ) or lamin B1 (not shown) in sanguinarine-treated cultures.
We then asked whether the relocalization of cyclin D1 and topoisomerase II in sanguinarine-treated cultures is accompanied by changes in the ability of these cells to carry out DNA synthesis. Following a 30 minute pulse of bromodeoxyuridine (BrdU), control cultures exhibit many strongly labeled cells, which display characteristic patterns of S-phase BrdU immunoreactivity (Fig. ). However, 2 hours of treatment with 5 μM sanguinarine markedly suppresses the ability of MCF-7 cells to incorporate BrdU into their DNA. Only a few weakly staining BrdU-positive foci remain in these cultures (Fig. ). Cultures pulsed with BrdU after 24 hours of sanguinarine treatment once again display many strongly labeled nuclei (Fig. ).
Figure 4 Sanguinarine transiently inhibits DNA synthesis. Control and drug-treated cells were pulsed with 10 μM BrdU for 30 minutes prior to fixation, and subsequently labeled with an antibody to BrdU (green) and counterstained with Hoechst (blue). (a (more ...)
Because sanguinarine is itself fluorescent, its uptake and distribution in living MCF-7 cells were studied with vital epifluorescence imaging methods. For these experiments, both Hoechst 33342 and sanguinarine were added to cultures to compare their uptake and distribution patterns. Both compounds could be visualized and differentiated using different lasers and filter sets provided with a Nikon C1 confocal microscope. Time-lapse studies show that cells take up Hoechst over 2 hours, resulting in a progressive increase in fluorescence that is restricted to nuclei (Fig. ). Sanguinarine also strongly labels nuclei, but displays a number of significant differences from Hoechst. It is more rapidly taken up by cells, first becoming visible in small cytoplasmic aggregates a few minutes after drug addition (not shown). These aggregates decrease in intensity at about the same time nuclear fluorescence increases, and eventually only nuclear fluorescence is evident. Sanguinarine nuclear fluorescence is shorter-lived than Hoechst, and begins to decrease in intensity after a few hours (Fig. ). Sanguinarine nuclear fluorescence is noticeably diminished six hours after drug addition, and cells are essentially non-fluorescent after 24 hours of sanguinarine treatment (data not shown).
Figure 5 Vital imaging of sanguinarine uptake and distribution in MCF-7 cells by laser-scanning confocal microscopy. 5 μg/ml Hoechst and 10 μM sanguinarine were simultaneously added to living cultures and images taken with both blue (Hoechst) and (more ...)
To verify that most of the cells treated with 5 μM sanguinarine in the immunofluorescent studies were in fact still viable, a number of different fluorescent probes routinely used to assess the apoptotic status of cells were used, including Hoechst, tetramethylrhodamine methyl ester (TMRM), calcein AM, and ethidium homodimer. Although TMRM and ethidium homodimer both fluoresce red, they label different structures in living versus dead cells: cells with intact plasma membranes exclude ethidium homodimer and are unlabeled, whereas apoptotic cells with compromised plasma membranes exhibit red nuclei after incubation with this DNA-binding dye. In contrast, mitochondria with polarized inner membranes are strongly labeled by TMRM in living cells, whereas dying or dead cells with depolarized mitochondria are unlabeled by this probe. Calcein AM is hydrolyzed by esterases and retained in living cells with intact plasma membranes, imparting a green fluorescence, while Hoechst is a cell permeant blue fluorescent DNA-binding dye that reveals chromatin patterns in both living and dead or dying cells. Thus, living, viable cells exhibit an overall green cytoplasmic fluorescence, red mitochondia, and blue nuclei with normal patterns of euchromatin and heterochromatin, while apoptotic cells that are well along the cell death pathway lack both the green cytoplasmic calcein and the red mitochondrial TMRM signals, and contain dual-labeled red and blue nuclei that display condensed chromatin patterns. By these criteria, most MCF-7 cells treated with 5 μM sanguinarine for 24 hours (which are morphologically identical to the cells displaying altered cyclin D1 and topoisomerase II trafficking in the immunolabeling experiments), are not apoptotic (Fig. ). Overall, a slight variation exists between groups of cells in both control and sanguinarine-treated cells with respect to both the intensity and textural labeling of heterochromatin by Hoechst. In the fields shown in Fig. and , the sanginarine-treated group of cells appears to be somewhat more heterochromatic and brightly stained than the controls, but the converse situation often exists, depending on the field of view. These patterns may reflect modest differences in the ability of Hoechst to penetrate individual MCF-7 cells or clones of cells at this concentration and timepoint. These subtle differences are clearly distinguishable from the extensive nuclear alterations characteristic of apoptosis.
Figure 6 Most cells treated with 5 μM sanguinarine for 24 hours are not apoptotic. Multiparameter imaging was conducted with control (a-d) and sanguinarine-treated (e-h) cultures, using differential interference (DIC) microscopy (a and e), a fluorescein (more ...)