3.1 The fluorescent substrate is transported by DAT and VMAT2
The fluorescent reporter dye is marketed as a substrate of the plasma membrane monoamine transporters (the dopamine, norepinephrine and serotonin transporters; DAT, NET and SERT). Since this dye is an analog of MPP+, which is transported by DAT and VMAT2, we hypothesized that this dye is also a VMAT2 substrate. To test this, we generated HEK293 cell lines stably expressing human DAT, human VMAT2 or both (HEK-DAT, HEK-VMAT2, or HEK-DAT/VMAT2, respectively). These cells were incubated with the assay reagent (dye concentration, 4 μM) for 1 hour and images were collected by confocal microscopy (). Cells that do not express DAT (HEK cells and HEK-VMAT2 cells; ) do not take up the dye and show no fluorescence. However, cells that express DAT (HEK-DAT and HEK-DAT/VMAT2; ) are able to take up the compound and fluoresce. In addition, the presence of VMAT2 in the cells changes the localization of the dye. In HEK-DAT and HEK-DAT/VMAT2 cells, the dye localizes to the cell body in a similar pattern as is seen in HEK-DAT cells, but there is also additional punctate staining in extensions from the cell body. This punctual localization is seen only in the presence of active VMAT2 (, ). The extensions contain a concentration of mCherry-VMAT2-positive puncta. These puncta represent a vesicle-like compartment that contains functional VMAT2. Their localization to extensions from the cell body likely reflects the partial neuronal profile of HEK293 cells (Shaw et al.). Despite their renal origin, these cells express many genes characteristic of neuronal cells. In addition, during the creation of this stable cell line, we selected cells for further propagation that developed such extensions.
The fluorescent substrate is transported by DAT and VMAT2
Uptake of the fluorescent substrate in HEK-DAT/mCherry-VMAT2 cells
Additionally, since the dye is a substrate for DAT and VMAT2, we assessed whether the dye impairs dopamine uptake of dopamine by these transporters. In HEK-DAT/VMAT2 cells and in vesicles isolated from HEK-VMAT2 cells, preincubation with the dye (4 μM) did not impair uptake of 3H-DA (data not shown). Therefore, competition of the dye with dopamine is unlikely to be a concern at concentrations of dye used in the assay.
We also tested the toxicity of the reagent by WST-1 assay since the dye is an analog of MPP+,. There was no toxicity observed 24 hours after the cells were treated with the concentration of dye used in the assay (4 μM) according to the protocol (15 minute incubation with HBSS buffer, 2 hour incubation with assay reagent, wash and replace with media) (data not shown).
3.2 Characterization of HEK-DAT/mCherry-VMAT2 stable cell line
To create a method for measuring this VMAT2-dependent difference in localization, we created a stable cell line expressing hDAT and a fusion of mCherry and hVMAT2 (HEK-DAT/mCherry-VMAT2). We verified that these cells contained functional DAT and VMAT2 by radioactive uptake assays. These assays showed that inhibition of DAT by GBR12909 almost completely blocked uptake of 3H-DA into the cells, as expected (). They also demonstrate that inhibition of VMAT2 by tetrabenazine (TBZ) lowers the capacity of the cell to take up DA (). This is not due to an inhibition of DAT by TBZ, as TBZ does not inhibit uptake in HEK-DAT cells (). The radioactive uptake assays verified that the cells contain functional DAT and VMAT2 and that they respond to the appropriate inhibitors. Vesicles isolated from these cells have TBZ-dependent uptake (IC50 = 54 nM; 95% CI: 40 – 75 nM) (). In intact cells, inhibition of DAT by nomifensine blocks uptake in HEK-DAT (IC50 = 2.4 μM; 95% CI: 1.7 – 3.3 μM) and HEK-DAT/mCherry-VMAT2 cells (IC50 = 5.5 μM; 95% CI: 3.8 – 8 μM) (). TBZ inhibits uptake in HEK-DAT/mCherry-VMAT2 cells (IC50 = 219 nM; 95% CI: 144 – 331 nM) but not in HEK-DAT cells (). The IC50 for TBZ in whole cells is higher than the IC50 in purified vesicles (). This is likely due to low permeability of TBZ, such that the concentration inside the cell at the vesicle is much lower than the concentration of TBZ applied in the media. The concentration of TBZ inside the cell, at the vesicle, is most likely closer to the values calculated for purified vesicles. This difference between purified vesicle fractions and whole cells underscores the need for an assay to assess vesicular function in an intact cellular environment.
HEK-DAT/mCherry-VMAT2 cells contain functional DAT and VMAT2
3.3 Subcellular localization of dye
To characterize the localization of the dye, we initially examined uptake of the fluorescent substrate in HEK-DAT cells. In these cells, uptake of the dye is blocked by nomifensine, as expected (). In addition, the dye primarily localized to the mitochondria, shown by colocalization with MitoTracker (). Mitochondrial localization is expected based on the structural similarity of the dye to MPP+. In contrast, in HEK-DAT/mCherry-VMAT2 cells, the dye only partially localizes to the mitochondria as indicated by colocalization of the dye with MitoTracker (). In these cells, the dye also localizes to mCherry-positive puncta in extensions from the cell body as indicated by colocalization of the dye with mCherry (). This punctal localization is TBZ-dependent and can be visualized as a loss of colocalization between mCherry and the reporter dye (). When VMAT2 function is blocked by TBZ, the localization of the dye is primarily mitochondrial, similar to HEK-DAT cells. Uptake in these cells (HEK-DAT/mCherry-VMAT2) is also blocked by nomifensine (). Together, this demonstrates that the reporter dye is transported by VMAT2 into a VMAT2-containing compartment and that in the absence of VMAT2 activity, either in cells that do not express VMAT2 or by pharmacological inhibition with TBZ, localization to this compartment is lost. Similar mitochondrial localization is seen in HEK-DAT cells and TBZ-treated HEK-DAT/mCherry-VMAT2 cells (, ).
Uptake of the fluorescent substrate in HEK-DAT cells
3.4 Quantification of TBZ-dependent uptake
As mentioned above, the localization of the reporter dye to mCherry positive puncta is TBZ-dependent. Therefore, we carried out experiments to determine if this dependence could be quantified and used to generate dose-response curves and calculate IC50
values. Confocal images clearly show a loss of the fluorescence from the dye in mCherry-positive puncta with increasing concentrations of TBZ, indicated by a loss of green fluorescence in the mCherry-positive regions of the cells (). This change was quantified by thresholding the red channel, defining regions of interest based on that threshold, applying those regions to the green channel and measuring the standard deviation of pixel intensities in the green channel within those regions (). By defining a region of interest based on mCherry staining, the areas of mitochondrial staining are excluded from the analysis. Thus, this strong mitochondrial localization does not complicate the data analysis. Standard deviation of pixel intensities within these mCherry-defined regions of interest provides a measure of the punctate fluorescence. When VMAT2 activity is high, fluorescence within these regions is highly punctate and a high standard deviation of pixel values is calculated. As VMAT2 activity is inhibited by increasing concentrations of TBZ, fluorescence in these regions becomes less punctate and a low standard deviation of pixel values is calculated. From these values, a punctate-diffuse index was calculated by dividing the standard deviation in each image by the mean of the standard deviations in all control (vehicle treated) images (Goldstein et al., 2000
). The IC50
calculated by this method (IC50
= 142 nM; 95% CI: 33 – 608 nM) is similar to the IC50
calculated from the radioactive uptake assays (IC50
= 219 nM; 95% CI: 144 – 331 nM).
TBZ sensitive uptake in HEK-DAT/mCherry-VMAT2 cells