Properties of a non-bioactive fluorescent derivative of differentiation-inducing factor-3, an anti-tumor agent found in Dictyostelium discoideum

ABSTRACT Differentiation-inducing factor-3 (DIF-3), found in the cellular slime mold Dictyostelium discoideum, and its derivatives, such as butoxy-DIF-3 (Bu-DIF-3), are potent anti-tumor agents. To investigate the activity of DIF-like molecules in tumor cells, we recently synthesized a green fluorescent DIF-3 derivative, BODIPY-DIF-3G, and analyzed its bioactivity and cellular localization. In this study, we synthesized a red (orange) fluorescent DIF-3 derivative, BODIPY-DIF-3R, and compared the cellular localization and bioactivities of the two BODIPY-DIF-3s in HeLa human cervical cancer cells. Both fluorescent compounds penetrated the extracellular membrane within 0.5 h and localized mainly to the mitochondria. In formalin-fixed cells, the two BODIPY-DIF-3s also localized to the mitochondria, indicating that the BODIPY-DIF-3s were incorporated into mitochondria independently of the mitochondrial membrane potential. After treatment for 3 days, BODIPY-DIF-3G, but not BODIPY-DIF-3R, induced mitochondrial swelling and suppressed cell proliferation. Interestingly, the swollen mitochondria were stainable with BODIPY-DIF-3G but not with BODIPY-DIF-3R. When added to isolated mitochondria in vitro, BODIPY-DIF-3G increased dose-dependently the rate of O2 consumption, but BODIPY-DIF-3R did not. These results suggest that the bioactive BODIPY-DIF-3G suppresses cell proliferation, at least in part, by altering mitochondrial activity, whereas the non-bioactive BODIPY-DIF-3R localizes to the mitochondria but does not affect mitochondrial activity or cell proliferation.


INTRODUCTION
The cellular slime mold Dictyostelium discoideum (D. discoideum) is a soil microorganism that, at the end of its life cycle, transforms into a multicellular fruiting body consisting of a stalk and spores. Differentiation-inducing factor-1 (DIF-1) ( Fig. 1A) is a putative morphogen that regulates cell fate by inducing the differentiation of prestalk cells and suppressing the differentiation of prespore cells (Kay et al., 1989;Kay et al., 1999;Morris et al., 1987). DIF-1 has been shown to function also as a modulator of chemotaxis in D. discoideum (Kuwayama and Kubohara, 2009). Differentiation-inducing factor-3 (DIF-3) (Fig. 1A) is the first metabolite formed during DIF-1 degradation and it has virtually no activity in the induction of prestalk cells and the modulation of chemotaxis (Kay et al., 1989;Kay et al., 1999;Kuwayama and Kubohara, 2009;Morris et al., 1988).
To study the cellular localization, function, and target proteins of DIF-3-like molecules in mammalian cells, we recently synthesized a green fluorescent derivative of DIF-3, BODIPY-DIF-3 (designated BODIPY-DIF-3G in this study) (Fig. 1B), and showed that BODIPY-DIF-3G localizes mainly to the mitochondria in HeLa human cervical cancer cells . We also showed that BODIPY-DIF-3G has similar biological activity as DIF-3 and Bu-DIF-3, suppressing cell proliferation in part, by acting as a mitochondrial uncoupler to disrupt mitochondrial function .
In the present study, we synthesized a red (orange) fluorescent derivative of DIF-3, BODIPY-DIF-3R (Fig. 1B), and showed that BODIPY-DIF-3R also localized mainly to the mitochondria. However, unlike BODIPY-DIF-3G and DIF-3, BODIPY-DIF-3R did not suppress HeLa cell proliferation nor induce any change in mitochondrial morphology or function. These results show that bioactive DIF-like molecules suppress cell proliferation at least in part via disturbance of mitochondrial activity, whereas BODIPY-DIF-3R localizes to the mitochondria but is not bioactive. Our results also indicate that the swollen mitochondria are morphologically, biochemically, and thus functionally different, from normal mitochondria and can be distinguished by staining with BODIPY-DIF-3G and BODIPY-DIF-3R.

Synthesis of fluorescent derivatives of DIF-3 and their effects on HeLa cell proliferation
We previously synthesized a green fluorescent derivative of DIF-3, BODIPY-DIF-3G (Fig. 1B), and elucidated its cellular localization and function in HeLa cells . Here, we synthesized another reagent for analyzing DIF-like molecules, a red (orange) fluorescent derivative of DIF-3, BODIPY-DIF-3R (Fig. 1B).

ÀÀ
Access to both biologically functional and nonfunctional fluorescent DIF-3 derivatives could be powerful tools for studying structure-effect relationships and imaging the cellular localization and function of DIF-like molecules.
We then took a more detailed look at the intracellular localization of BODIPY-DIF-3R and BODIPY-DIF-3G under the same conditions in live HeLa cells by using high-magnification fluorescence microscopy and two fluorescent probes for mitochondria, MitoTrackerG and MitoTrackerR. BODIPY-DIF-3R was confirmed to localize mainly in the mitochondria, as shown by its co-localization with MitoTrackerG ( Fig. 3A), with no remarkable changes in mitochondrial morphology or cell morphology throughout the 3-day incubation period (Fig. 3B).
When cells were incubated for 0.5 h with 20 mM each of BODIPY-DIF-3G and BODIPY-DIF-3R, the two compounds were confirmed to co-localize to the mitochondria (Fig. 4). Incubation for 3 days with 20 mM BODIPY-DIF-3G induced many mitochondria to swell, which were stained well with BODIPY-DIF-3G and MitoTrackerR ( Fig. 5A) but were not stained with BODIPY-DIF-3R (Fig. 5B). In contrast, the morphologically normal, but not swollen, mitochondria were stained with both BODIPY-DIF-3G and BODIPY-DIF-3R (Fig. 5B).   Bu-DIF-3 is one of the more potent anti-tumor agents among the DIF-like molecules (Gokan et al., 2005), and it has been shown to induce mitochondrial swelling . Therefore, we compared the cellular localization of BODIPY-DIF-3G and BODIPY-DIF-3R in cells treated with 5 mM Bu-DIF-3. After 3 days, mitochondria swelled greatly in all the cells that were stained with BODIPY-DIF-3G and MitoTrackerR (Fig. 6A), but again, BODIPY-DIF-3R stained only the normal, but not swollen mitochondria (Fig. 6B). CCCP is a mitochondrial uncoupler (proton-specific ionophore) that induces mitochondrial swelling in HeLa cells . We next compared the cellular localization of BODIPY-DIF-3G and BODIPY-DIF-3R in CCCP-treated cells. When cells were incubated for 3 days with 10 mM CCCP, mitochondria swelled greatly in all the cells that were stained with BODIPY-DIF-3G and MitoTrackerR (Fig. 6C), but again, the swollen mitochondria were not stained with BODIPY-DIF-3R (Fig. 6D).
These results show that BODIPY-DIF-3R can penetrate the cell membrane and localize to normal mitochondria, but not to swollen mitochondria induced with the bioactive DIF-3 derivatives BODIPY-DIF-3G or Bu-DIF-3. Functionally, BODIPY-DIF-3R does not induce mitochondrial swelling. In addition, it is likely that the swollen mitochondria are morphologically, biochemically, and thus functionally different, from normal mitochondria and can be distinguished by staining with BODIPY-DIF-3G (or MitoTracker dyes) and BODIPY-DIF-3R.

Cellular localization of BODIPY-DIF-3G and BODIPY-DIF-3R in formalin-fixed HeLa cells
We next investigated the cellular localization of BODIPY-DIF-3G and BODIPY-DIF-3R in formalin-fixed HeLa cells (Fig. 7). As described previously , BODIPY-DIF-3G localized to mitochondria that were pre-stained well with MitoTrackerDR, and BODIPY-DIF-3R co-localized to mitochondria in formalin-fixed control cells (Fig. 7A). These results indicate that both BODIPY-DIF-3G and BODIPY-DIF-3R localize to normal-shaped mitochondria independently of the mitochondrial membrane potential and therefore that there may be target molecules of the DIF derivatives within mitochondria. However, CCCP-treated cells (Fig. 7B), in which there appeared to be swollen mitochondria, were hardly stained with MitoTrackerDR. Since accumulation of the probe in active mitochondria is dependent on the mitochondrial membrane potential, the membrane potential must be low in these swollen mitochondria. According to data from the manufacturer, the accumulation of MitoTrackerR in active mitochondria is also dependent on the membrane potential, and since MitoTrackerR was able to stain CCCP-treated swollen mitochondria (Fig. 6C), MitoTrackerR may be a more potent probe than MitoTrackerDR under our experimental conditions. BODIPY-DIF-3G stained both normal-shaped and swollen mitochondria even after fixation, but the swollen mitochondria were only weakly stained with BODIPY-DIF-3R (Fig. 7B), suggesting again that the swollen mitochondria were functionally, biochemically, and/or biophysically different from normal mitochondria.

Effects of DIFs on mitochondrial O 2 consumption
We have previously shown that the bioactive compounds DIF-3, Bu-DIF-3, and BODIPY-DIF-3G act like mitochondrial uncouplers such as CCCP in that they also increase mitochondrial O 2 consumption . We examined the effects of BODIPY-DIF-3R on mitochondrial O 2 consumption in isolated mouse liver mitochondria in vitro by using a Clark-type oxygen electrode. As described previously , BODIPY-DIF-3G at 20-100 mM increased basal O 2 consumption in a dose-dependent manner (Fig. 8). In contrast, the non-bioactive compound, BODIPY-DIF-3R, had no significant effect up to the maximum 100 mM dose. These results suggest that the inability of BODIPY-DIF-3R to suppress cell proliferation is due to its inability to disturb mitochondrial O 2 consumption.

DISCUSSION
In addition to D. discoideum being studied as a model organism in cell and developmental biology, other cellular slime molds have recently been shown to produce many pharmacologically active in HeLa cells treated for 3 days with BODIPY-DIF-3G. Cells were incubated for 3 days with BODIPY-DIF-3G (20 mM) and then for 0.5 h with Hoechst (0.1 mg/ml) and MitoTrackerR (0.1 mM) (A) or BODIPY-DIF-3R (20 mM) (B). Cells were washed free of the additives and observed by using high-magnification fluorescence microscopy. Three-dimensional (3D) images were constructed from z-stacked two-dimensional (2D) images, and two representative 2D-projections of the 3D images are shown. BODIPY-DIF-3G and MitoTrackerR co-localized to mitochondria, and the swollen mitochondria were stained with BODIPY-DIF-3G and MitoTrackerR, but not BODIPY-DIF-3R. Scale bars: 20 mm.   compounds (Kikuchi et al., 2004;Kikuchi et al., 2005;Kikuchi et al., 2006;Kikuchi et al., 2010;Kikuchi et al., 2012;Kikuchi et al., 2013). This has led to the suggestion that cellular slime molds are ''uncultivated drug resources.'' Among the compounds reported so far, the polyketides DIF-1 and DIF-3 (Morris et al., 1987;Morris et al., 1988) are the most promising and wellstudied potential anti-tumor drugs (Asahi et al., 1995;Gokan et al., 2005;Kubohara et al., 1995;Kubohara, 1997;Kubohara, 1999). The mechanisms of action of DIF-1 and DIF-3 have been analyzed in several different types of tumor cells but remain to be elucidated (Jingushi et al., 2012;Jingushi et al., 2013;Kanai et al., 2003;Kubohara et al., 1995;Kubohara, 1997;Kubohara, 1999;Kubohara and Hosaka, 1999;Shimizu et al., 2004;Takahashi-Yanaga et al., 2003;Takahashi-Yanaga et al., 2006). To obtain useful tools for the analyses of the cellular localization and function of DIF-like molecules, we previously synthesized the green fluorescent derivative BODIPY-DIF-3G . In the present study, we described the red (orange) fluorescent derivative BODIPY-DIF-3R. These two different fluorescent DIF-3 derivatives enabled us to simultaneously study their effects on cell proliferation, mitochondrial function, and morphology. Unexpectedly, unlike BODIPY-DIF-3G and DIF-3, BODIPY-DIF-3R did not significantly affect cell proliferation or mitochondrial morphology although it penetrated the cell membrane and localized to normal mitochondria (Fig. 2). In addition, BODIPY-DIF-3G, Bu-DIF-3, and CCCP induced mitochondrial swelling (Fig. 2D, Fig. 5, Fig. 6), whereas BODIPY-DIF-3R did not (Fig. 2D, Fig. 3). Finally, we showed that DIF-3, Bu-DIF-3, and BODIPY-DIF-3G promoted mitochondrial O 2 consumption in a dose-dependent manner , while BODIPY-DIF-3R did not (Fig. 8). These results suggest that the pharmacologically active DIF-like molecules suppress cell proliferation at least in part by increasing O 2 consumption rates similar to the action of mitochondrial uncouplers. One reason for the pharmacological inactivity of BODIPY-DIF-3R may be its relatively large molecular mass and/or structure (Fig. 1B). The molecular targets of these DIF-like molecules remain to be identified.
It is important to note that swollen mitochondria induced by BODIPY-DIF-3G, Bu-DIF-3, or CCCP could be stained with BODIPY-DIF-3G, but not with BODIPY-DIF-3R (Figs 5, 6), which was the case even in formalin-fixed cells (Fig. 7), suggesting that swollen mitochondria are functionally, biochemically, and/or biophysically different from normal mitochondria. Elucidation of these differences (e.g. by the use of the two BODIPY-DIF-3 derivatives described in this study) may reveal an unidentified target protein of functional DIF-like molecules in mitochondria.

Synthesis of BODIPY-DIF-3G and BODIPY-DIF-3R
BODIPY-DIF-3G was synthesized as described previously , and BODIPY-DIF-3R was synthesized in six reaction steps as described below (Fig. 1B). The synthesized compounds were stored as 10 mM solutions in DMSO at 220˚C.

Cell proliferation assay
HeLa cells were cultured in 12-well plates for 3 days at 2.5-5610 3 cells/well in 1 ml of DMEM-FBS containing 0.2% DMSO or 20 mM of DIF-3, BODIPY-DIF-3G, or BODIPY-DIF-3R. Cells were treated with trypsin and the detached cells were used for direct cell number count using a hematometer, and relative cell number was calculated.
For high-magnification observation and multi-color imaging, cells were incubated for 0.5 h or 3 days with 2 ml DMEM-FBS containing additives in 35-mm tissue culture treated plastic 81156 m-dishes (ibidi, Martinsried, Germany). The cells were washed twice with 2 ml of PBS(2), submerged in 2 ml of HEPES buffer (pH 7.4) and observed by using a Keyence BZ-9000 fluorescence microscope (Osaka, Japan) equipped with an oil immersion 1006 lens (CFI Plan Apo VC100XH) and multi-filters that distinguish up to four fluorescent probes simultaneously. Digitized images of z-stack sections were taken at 0.4mm intervals, which were then treated (haze-reduced) with the Keyence BZ analyzer software (for deconvolution fluorescence imaging) and compiled into three-dimensional (3D) images; when 3D images were constructed, nonlinear adjustment was performed to obtain clear (high contrast) images without haze. All color images are presented in pseudo colors.

Observation of formalin-fixed cells
HeLa cells were incubated for 3 days with 2 ml of DMEM-FBS in 35mm tissue culture plastic 81156 m-dishes in the presence or absence of CCCP (10 mM), and then incubated for a further 0.5 h with Hoechst 33342 (0.1 mg/ml) and MitoTrackerDR (0.2 mM), which are dyes that are retained in nuclei and mitochondria, respectively, after fixation. Cells were washed twice with PBS(2) and fixed for 20 min at room temperature with 2 ml of 3.7% (v/v) formaldehyde in PBS(2). After being washed 3 times with PBS(2), the fixed cells were incubated for 0.5 h with 20 mM BODIPY-DIF-3G and BODIPY-DIF-3R in 2 ml of DMEM-FBS, washed 3 times with PBS(2), and observed and analyzed with a Keyence BZ-9000 fluorescence microscope and BZ analyzer software as described above.

Preparation of mitochondria-enriched cell fraction and measurement of O 2 consumption
Mitochondria were isolated from mouse liver (ICR, 7-10-week-old females) by differential centrifugation as described previously (Kabuyama et al., 2010;Kubohara et al., 2013). Mitochondrial O 2 consumption was determined by using a Clark-type oxygen electrode (Strathkelvin Instruments Ltd., North Lanarkshire, Scotland) as described (Gottlieb et al., 2002;Wegrzyn et al., 2009). The mitochondria-enriched fraction was incubated in O 2 measurement buffer (225 mM mannitol, 75 mM sucrose, 10 mM KCl, 0.1 mM EDTA, 3 mM phosphate, 5 mM succinate, 5 mM glutamate, 20 mM Tris-HCl; pH 7.4) in the presence of 1% DMSO, BODIPY-DIF-3G (20 or 100 mM), or BODIPY-DIF-3R (20 or 100 mM) at 30˚C. After recording the mitochondrial respiration State 4 reaction, an aliquot of ADP was added to a final concentration of 200 mM to induce State 3 respiration. However, since DIF-related molecules have been shown to affect State 4 respiration , the rate of O 2 consumption during State 4 was calculated and compared in this study.

Statistics
Significance was assessed by unpaired (two-tailed) Student's t-test. Values were considered significantly different when the P value was less than 0.05.