Calixarenes (1) are a family of readily synthesized and functionalized macrocycles which have found applications in a number of research areas. In particular the ability of derivatives to complex anions (2, 3) and cations (4, 5) has been exploited for the development of sensors. Recently, interest has focused on the use of calixarene derivatives in biological systems be it as drugs, drug delivery systems, or imaging agents. In the design of drug molecules, particular attention has been paid to calixarenes as antimicrobials (6–8), as vaccines (9) and as anticancer agents (10). In the field of drug delivery, agents for nucleic acids based on single calixarenes (11–13) have shown promise and we have recently shown that larger arrays of cationic calixarene, so-called multicalixarenes, are able to transfect DNA effectively (14). In addition a number of studies have combined the ability of calixarene derivatives to complex cations with their low toxicity in the development of noncovalently labeled protein MRI imaging agents (15, 16).
While the nontoxic (17) and nonimmunogenicity (18, 19) status of calixarenes has been established, less information is available on how calixarenes are processed within cells and their cellular fate, both of which are important features which must be characterized before such derivatives are taken further for clinical development. We recently demonstrated that fluorescently labeled calixarene have the potential to track progress of the macrocycle in cells, as an NBD labeled cationic derivative was readily and speedily taken up into cells and provided a very stable, nontoxic read-out system to investigate cellular localization (20). The absorbance and emission spectra for the NBDCalAm showed that it can easily be visualized using a GFP filter (20).
Thus, the preparation of fluorescent derivatives of calixarenes provides useful information for researchers developing calixarenes as drugs and as drug delivery systems on how these derivatives interact with the cell surface and on their internalization methods. In addition, these labeled molecules, themselves offer potential as stains of intracellular processes
We demonstrate here, using live-cell studies, that a cationic fluorescently labeled calixarene (NBDCalAm) acts as a lysosomotropic compound without any induction of cell death, unlike other previously described lysosomotropic compounds (21, 22). Thus this NBDCalAm provides us with a novel, remarkable easy to use, exciting tool to investigate lysosomal function and kinetics in living cells over longer periods of time.
Materials and Methods
Cells and Materials
CHO.CCR5 cells, HeLa RC-49 cells, and THP-1 cells have been described previously (20, 23, 24). CHO and HeLa cells were routinely grown in DMEM supplemented with 10% bovine calf serum, 100 units of penicillin/ml, 100 μg of streptomycin/ml, 2 mM L-glutamine at pH 7.4 and 37°C in a H2O-saturated, 5% CO2 atmosphere. THP-1 cells were grown in RPMI supplemented with 10% bovine calf serum, 100 units of penicillin/ml, 100 μg of streptomycin/ml, 2 mM L-glutamine at pH 7.4. Filipin, sucrose, 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI), sodium azide, brefeldin A and methyl-β-cyclodextrin (MCD) were purchased from Sigma (Poole, UK). Bafilomycin A1, nocodazole, and monensin were from Calbiochem (Darmstadt, Germany). BODIPY® TR ceramide, and LysoTracker® Red were from Molecular Probes (Eugene, OR). All other biochemicals and cell culture reagents were obtained from Fisher Scientific (Loughborough, UK). NBDCalAm was synthesized as previously described (20) and stock solutions were prepared at 500 mM in water. The final concentration of NBDCalAm in all experiments was 10 mM.
Cells were grown in small dishes overnight, washed with PBS or HBSS and then incubated with filipin (5 μg/ml), MCD (10 mM), nocodazole (1 μM), bafilomycin A1 (1 μM), sodium azide (10 mM) or left untreated as control cells. After 1 hour NBD-labeled calixarene (NBDCalAm) (10 mM) was added and cells were incubated for different time points. NBDCalAm that had not entered the cells was removed by washing with HBSS and cells were incubated with medium for different time periods. Single stained cells were taken to an inverted Leica DM IL HC and images were acquired. For visualization of the nucleus, cells were incubated with DAPI (1μg/ml) for 5 minutes and then analyzed.
Visualization of mitochondria with pdsred-mito
To visualize mitochondria, cells were grown in small dishes overnight and then transfected with pdsred-mito using Fugene® (Roche, Burgess Hill, UK) according to the manufacturer's instruction. 24 hours after transfection cells were loaded with NBDCalAm. After a further 24 hours cells were washed three times with HBSS and analyzed under the microscope.
Visualization of the Golgi complex with BODIPY-labeled sphingolipids
To stain the Golgi in intact cells, BODIPY TR ceramide was added as a complex with bovine serum albumin. Cells grown in small dishes were rinsed three times with PBS. The cells were then incubated for 30 minutes at 4°C with 5 μM of the BODIPY TR ceramide/bovine serum albumin solution. The samples were rinsed several times with ice-cold HBSS and incubated in fresh medium for a further 30-minute period at 37 °C, before being prepared for microscopy as above.
Colocalization of NBDCalAm with acidic vesicles
Cells were plated in small dishes overnight and were loaded with NBDCalAm 48 hours before the LysoTracker Red stain. Cells were then exposed for 15minutes at 37°C to LysoTracker Red (50 nM) These conditions are optimal for labeling and for minimizing interference with intravesicular pH (25). Cells were then washed three times with ice-cold PBS before they were prepared for microscopy.
Pulse Chase Experiments
Cells were loaded for 24 to 48 hours with NBDCalAm. Pictures were taken to verify correct loading after 24 and 48 hours. Sodium azide, monensin, nocodazole, brefeldin A, or bafilomycin A1 were added to cells and subsequent pictures were taken over different time periods.
For each assay, monolayer cells were harvested from the flask with PBS/EDTA and washed twice with PBS before incubation with 10 mM NBDCalAm for different timeperiods. Suspension cells were washed with PBS twice before addition of 10 mM NBDCalAm for different timeperiods. Incubation was stopped by washing cells once with PBS. The cells were then treated with trypsin for 10 minutes at 37°C and washed twice with PBS. Fluorescence analysis was performed with an Accuri® C6 Flow Cytometer. Unstained cells were used to set-up the machine. Ten thousand cells per sample were acquired.
Nonfixed cells were analyzed using a Leica DM IL HC microscope with eye piece 10× magnification and the objective 63× magnification. Imaging was done with a Leica DFC 420 camera with 5 Megapixel. The filter used for NBDCalAM was GFP with excitation at 470 and 40 bandwidth, emission at 525 and 50 bandwidth. For the commercial LysoTracker Red, BODIPY TR ceramide or pdsred-mito the filter N3 with excitation at 546 and 12 bandwidth, emission at 600 and 40 bandwidth was used and for DAPI, the filter A4 with excitation at 360 and 40 bandwidth, emission at 470 and 40 bandwidth was used. Cells were grown in small petri dishes and pictures were taken from living, nonfixed cells. Colocalization analysis was performed using the Leica LAS Image Overlay Module. The Pearson's correlation coefficient (Rr) was calculated using the Volocity 5.2 software.
A solution of NBDCalAm in distilled water with an absorbance κ ∼ 1.0 was exposed to just over 8 mW of light centered in a narrow band (FWHM 20 nm) around 452 nm. The light source was a high pressure xenon arc lamp (Müller Elektrotechnik LAX1450) which was filtered using a monochromator (PTI model 101 with 1,200 lines mm−1 grating) and measured downstream of the sample position with a thermopile sensor (Molectron PM3 and Coherent 3Sigma power meter). The absorption spectrum was monitored continuously at right angles to the irradiation using a tungsten lamp and an Ocean Optics HR2000 fiber optic spectrometer. The spectrum was recorded automatically at 20-minute intervals for several hours.
NBDCalAm is Rapidly Taken up into Cells and is Stable over Long Periods of Time
We recently showed that NBDCalAm is taken into the cells without involvement of cellular endocytosis mechanisms (20). These experiments were performed in paraformaldehyde fixed cells and not living cells. There is discussion as to whether fixing of cells actually changes the uptake of compounds into cells and can even mask the endocytotic uptake of substances into the cells (26). Therefore the uptake of NBDCalAm into living cells in real time was investigated. Endocytosis via clathrin-coated pits was inhibited using sucrose (20, 24) and caveolae related pathways were inhibited with filipin and MCD (20, 27). The observations in living cells were exactly as previously described for fixed cells. The NBDCalAm is taken into the cells rapidly despite the blocking of clathrin-coated pit and caveolae connected endocytosis mechanisms (data not shown). We therefore postulate that the NBDCalAm is not taken into the cells via active endocytosis using conventional pathways, but is more likely to cross to the inside directly via a plasma membrane carrier as has been suggested for polyamines (28)
Although those experiments clearly showed that the fixing of cells had no effect on NBDCalAm uptake into cells, all further experiments were performed on living, nonfixed cells, to prevent any alterations of the cellular machinery due to the preparation procedures for the cells.
Time-course analysis of the uptake shows that the NBDCalAm rapidly gains entry to the cells where it remains very stable over long periods of time. After 1 to 3 minutes incubation with the compound, the intracellular localization is already visible (data not shown). We quantified the uptake kinetics in three different cell lines (HeLa, CHO, THP-1) using flow cytometry, on trypsin-treated cells (26), and obtained similar results to the immunofluorescence staining (Figs. 1a and 1b), with an uptake constant of 1 to 3 minutes in the different cell lines (Fig. 1c). In addition, when cells are incubated for 2 minutes with NBDCalAm before washing off the excess of the compound and incubating the cells in dye-free medium, the signal remains strong over 72 hours. This demonstrates that NBDCalAm is taken up rapidly (inside the 2-minute window) and it also shows that the fluorescence is very stable and does not diminish once the cells are incubated in dye-free medium after staining (Fig. 2). At the concentration necessary for microscopy work the NBDCalAm has no cytotoxic effect on the cells. CHO cells grow in the presence of NBDCalAm at a similar rate to nontreated cells as previously shown (20). NBDCalAm is visible in nearly all daughter cells, which indicates that a very small concentration is necessary to achieve visibility inside the cell. The NBDCalAm has a prolonged photostability. After 5 hours of irradiation as described above, the sample absorbance had changed from 1.046 to 0.931, a decrease of only 10.8%. Over the whole irradiation period the sample absorbed a total energy of approximately 800 J and is therefore very suitable for fluorescence-based applications.
Intracellular Distribution of NBDCalAm Changes over Time
Interestingly, the time-course experiments clearly show that the intracellular distribution of the compound after 72 hours is very different from the initial observations (Fig. 2). Whereas up to 2 to 3 hours after introducing the compound into the cells, it is localized preferably in the cytoplasm around the nucleus, presumably in areas associated with the Golgi apparatus as can be seen in Figure 3. There is a change over time and after about 3 to 4 hours up to 72 hours the distribution of NBDCalAm becomes more punctuated in the cells, which can be indicative of a primary enrichment in the Golgi-apparatus and subsequent distribution to acidic vesicles.
NBDCalAm Colocalizes with Golgi-Apparatus and Acidic Vesicles
To analyze the exact localization of NBDCalAm inside the cells, various cellular localisation markers were used. The distribution of NBDCalAm after several hours is suggestive of either colocalization in acidic vesicles or alternatively in mitochondria. Therefore CHO cells were transfected with pdsred-mito, a plasmid which expresses a mitochondrial marker and after transfection cells were incubated with NBDCalAm. Forty-eight hours after transfection cells were analyzed and it became clear that the NBDCalAm does not colocalize with mitochondria, even if the staining pattern is very similar (Fig. 3). The Pearson's correlation coefficient (Rr) was calculated as 0.00600483 using the Volocity 5.2 software. This clearly shows that the NBDCalAm does not colocalise with the mitchondria.
Hence a double stain of CHO cells with Golgi-tracker and NBDCalAm was performed. Up to 3 to 4 hours after introducing the compound to cells, we observed colocalization between the compound and the Golgi apparatus, indicating that the NBDCalAm indeed accumulates in the Golgi apparatus after entering the cell (Fig. 4).
Incubation with NBDCalAm for longer time periods (4 to 72 hours) or 2 minutes incubation with NBDCalAm followed by a 4 to 72 hours waiting period, leads to a distinct pattern and no colocalization with the Golgi-tracker (Fig. 5). Once this pattern was detectable in cells using microscopy, then a double staining with LysoTracker Red was performed. In cells that were treated in this manner, a high proportion of overlay between NBDCalAm and LysoTracker Red was visible (Fig. 5), with a calculated Pearson's coefficient of 0.693257, indicating a good degree of colocalization between the two stains. For single red stains or green stain, no bleed through was detectable for the opposite filter (data not shown). These results are a good indication that the NBDCalAm is indeed localizing in acidic vesicles, especially lysosomes, after several hours in the cell.
NBDCalAm is Transported to Acidic Vesicles via Golgi Apparatus
After identifying the intracellular localization of NBDCalAm, the transport mechanisms were characterized in more detail. Incubation of CHO cells with a diverse set of inhibitors gave more information about the transport of the compound through the CHO cells. Preincubation of cells with brefeldin A, which inhibits transport from Golgi to the lysosomes (24), blocks nearly all lysosomal localization of the compound. Similarly monensin, which changes endosomal pH, bafilomycin A1, which blocks the activity of vacuolar ATPases and sodium azide, which depletes the intracellular ATP-pool, completely prevent any localization of the compound in the lysosomes (Fig. 6A). These results together show that NBDCalAm is transported from the Golgi apparatus to lysosomes and it demonstrates the importance of pH and the activity of vacuolar ATPases for the localization of NBDCalAm.
The microtubule network is responsible for correct transport of vesicles throughout the cells. Therefore the use of nocodazole, a microtubule inhibitor is an additional way to analyze vesicle transport in cells. Interestingly, nocodazole does not change the localization of the compound, however, it changes the dimensions of the lysosomes. They appear larger and less spread out around the cells, which is indicative of a lack of transportation along the microtubule network. Even after nocodazole treatment, there is still a complete overlay between NBDCalAm and the LysoTracker Red (Figs. 6A and 6B).
A functional vacuolar ATPase is essential to keep NBDCalAm in acidic vesicles. The experiments described above showed that the NBDCalAm was locating in acidic vesicles. To investigate whether vacuolar ATPases have any role in keeping the NBDCalAm in the acidic vesicles, pulse chase experiments were performed. Cells were loaded with NBDCalAm for 48 hours and excess compound was washed off. CHO cells were then incubated with inhibitors for up to 12 hours and analyzed using a microscope. When NBDCalAm is already localized in acidic vesicles, then treatment of cells with brefeldin A has no effect on localization of the compound, since brefeldin A inhibits transport out of the Golgi apparatus. However monensin, bafilomycin A1, and sodium azide all change the localization of NBDCalAm in the cells. Treatment with these inhibitors leads to a diffuse staining of the cell, without any specific localization of the fluorescent compound in acidic vesicles visible (Fig. 7). These results indicate that pH, a functional vacuolar ATPase and the presence of ATP are all important for holding NBDCalAm in the acidic vesicles.
Recently it has become apparent that understanding the function of lysosomes is important not only for gaining an insight into the biology of healthy cells, but also in a variety of different disease settings.
Vacuolar-type H+-ATPases-driven proton pumping and organellar acidification are essential for vesicular-trafficking along both the exocytotic and endocytotic pathways of eukaryotic cells (29). Deficient function of vacuolar ATPases and defects in vesicular acidification have been recognized as important mechanisms in a variety of human diseases and are emerging as potential therapeutic targets (30, 31). Acidity in cells has been shown to have a role in resistance to chemotherapy (32), proliferation (33), and metastatic behaviour (34). Inhibition of the vacuolar ATPases by small interfering RNA as well as pharmacological inhibition through proton pump inhibitors leads to tumor cytotoxicity and marked inhibition of human tumor growth in xenograft models (35–37). In light of the recently reported use of antibodies to trigger lysosomal release, the development of novel, nontoxic markers to monitor this process is of great relevance (38).
Our report provides clear evidence that a fluorescent labeled calixarene localizes in acidic vesicles and can therefore be a useful tool for cell biology studies. The NBDCalAm is taken up into the cells rapidly and localizes over time into the lysosomes. This is consistent with other studies (39) which indicate that compounds featuring weakly basic amines, such as the anilines in NBDCalAM, accumulate in cellular compartments with low internal pH. There is no apparent localization in other cellular organelles such as mitochondria. Only very little background staining can be seen inside the cells with the main fluorescence in acidic vesicles. We therefore estimate that the NBDCalAm does not bind nonspecifically to cellular organelles and components. The fluorescent compound is stable over long periods of time without showing any cytotoxic effects for the cells, and is even distributed into daughter cells after cell division. Cells can be assayed numerouse times after staining with the NBDCalAm, whithout experiencing any toxicity. The NBDCalAm shows a very good photostability and is very easy to use. The stain can be performed in complete cell medium or in PBS without any concernable differences to the outcome. A 5-minute incubation period with the NBDCalAm will lead to a localization in acidic vesicels after several hours with no apparent loss of fluorescent signal or cellular blebbing if the cells are incubated for long stretches of time in stain free medium, unlike the commercially available LysoTracker® Red. This allows for a short incubation time of the cells with the staining solution and makes monitoring and assaying over long periods of time without a negative effect on either the stain or the cells possible. Pulse-chase experiments confirmed that the compound is actively kept in the lysosomes by vacuolar ATPases and provides therefore an ideal marker to analyze the correct functioning of ATPases in healthy or diseased cells.
Taken together our data show that the NBD labeled calixarene provides a very stable and sensitive marker for lysosomes, and provides several novel and exciting advantages over some commercially available lysosomal markers. Unlike some of the commercial lysosomal markers, we do not observe a decrease in the fluorescent signal once the cells are incubated in dye-free medium after staining. On the contrary, the NBDCalAM is very stable even in dye-free medium over longperiods of time and therefore offers novel experimental opportunities for live-imaging over hours in dye-free medium without suffering the loss of fluorescent signal.
The authors thank Dr. Stephen Ashworth for his help in analyzing the physical properties of the compound and Dr. Paul Thomas for his help in analyzing the fluorescence data.