Fig. 3.

DPP9-l localizes to the nucleus and is active there. a HeLa cells transfected with HA-tagged DPP9-l or DPP9-S were subjected to indirect immunofluorescence using confocal microscopy for detection of the HA tag. Simultaneously cells were also analyzed for endogenous RanBP2 as a marker protein for the nuclear rim. Hoechst staining was applied to visualize the nucleus. Assays were performed at least three independent times. Shown is a representative image. For control, cells were stained only with secondary antibody (Supplementary Fig. S2A). b HA-tagged DPP9-S and FLAG-tagged DPP9-l were simultaneously overexpressed in HeLa cells and subjected to indirect immunofluorescence, detecting the HA tag and the FLAG tag as described in a. c Chart showing the subcellular distribution of DPP9-l and DPP9-S. For quantification of subcellular localization, transfected cells were grouped into the following three categories: N>C (stronger signals in the nucleus), N=C (equal signal intensities in the nucleus and in the cytoplasm), and C>N (stronger signals in the cytosol). Quantification was performed from at least two independent experiments, from at least four slides with more than 400 cells expressing the respective DPP9 protein being counted using the HA antibody for detection. d DPP9-l or DPP9-S-transfected HeLa cells were subjected to subcellular fractionation (T totals, C cytosolic fraction, M membranous fraction, and N nuclear fraction) followed by Western-blot analysis of 10 μg of each fraction using the commercial Abcam DPP9 (total) antibody. Lamin A/C was used as a marker for the nuclear fraction, tubulin as a marker for the cytosolic fraction. Note that no protease inhibitors were used during the fractionation procedure to enable subsequent DPP9 activity measurements in the various fractions (Supplementary Fig. S2B). The faster migrating band of Lamin A/C represents a degradation product of this protein due to the lack of protease inhibitors. Shown is one representative result of at least three independent experiments. e 3D stack of a nucleus of a HeLa cell overexpressing the long form of DPP9. Serial confocal images (every 0.5 μm) of a Z-stack with merge signals are shown. RanBP2 is used as a marker for the nuclear rim showing that DPP9-l is localized within the nucleus. f DPP9-l is localized within the nucleus and not at the outer nuclear rim. DPP9-l-overexpressing HeLa cells were fixed and permeabilized by 0.01 % Digitonin so that the nuclear membrane stayed intact. Cells were subjected to indirect immunofluorescence, detecting the HA tag. Protein localizations were analyzed as described in a. g HeLa cells transfected with HA-tagged DPP9-l were analyzed by indirect immunofluorescence using the commercial antibody (DPP9 total), which detects both DPP9 forms. Simultaneously, cells were also analyzed for endogenous RanBP2 as a marker protein for the nuclear rim. Hoechst staining was applied to visualize the nucleus. Assays were performed at least three independent times. Shown is a representative image. h HA-tagged DPP9-l was mutated in the start methionine of DPP9-S and replaced by an Alanine (DPP9-l M30A). The localization of this mutant in HeLa-overexpressing cells was analyzed as in g