Figure 3
3T3-L1 adipocytes stably expressing HA-GLUT4-GFP were electroporated with siRNA to knockdown EFR3, or a corresponding Scr-siRNA at day 6 post-differentiation as described. Cells were stimulated with 100 nM insulin for 20 min, or left untreated (Basal), before fixation and staining for surface HA as decribed in Materials and Methods. dSTORM images were acquired and reconstructions calculated using ThunderSTORM. (A) Hierarchical density-based spatial clustering of applications with noise analysis for representative control and EFR3 knockdown basal and insulin-stimulated 3T3-L1 adipocytes. GLUT4 molecule coordinates were processed using a python HDBSCAN script written in house. Grey dots indicate single molecule localizations for representative whole cells (left panels). Red boxes indicate region of interests shown in the middle and right panels. The middle panels show coloured molecule clusters identified by HDBSCAN within an ROI (‘clustered only’) and coloured clusters plus grey non-clustered molecules identified by HDSCAN within an ROI (‘clustered and non-clustered’, right panels) are shown. The representative 4 μm × 4 μm ROI is highlighted by the red square. Scale bars = 10 μm. (B) L function analysis of GLUT4 molecule clustering in EFR3 knockdown basal and insulin-stimulated 3T3-L1 adipocytes. GLUT4 molecule coordinates were obtained using ThunderSTORM and its spatial pattern was analysed. We analysed the correlation of points using the variance stabilised L function, which is the transformed version of Ripley’s K function. The L function describes how many points (given by L(r)) can be found within a distance r of any arbitrary point. Empirical estimates of the centred L(r) function are shown in panels labelled (i) control basal and insulin-stimulated cells, (ii) EFR3 knockdown basal and insulin-stimulated cells, (iii) all experimental groups combined. The experiment was carried out independently 4 times on n=10 cells per group. Empirical estimates of the L function from each were pooled together to provide a weighted mean and the 83% (or 1.37σ) confidence interval for each group – see text for details. Complete spatial randomness, modelled from a Poisson process, is indicated by the dashed line. (C) GLUT4 surface density upon EFR3 knockdown. HA (surface GLUT4) localization density was determined from all cells analysed in the presence of insulin. The GLUT4 localisation values do not differ between Scr- and EFR3-siRNA treated groups (n.s.).
Insulin-stimulated GLUT4 dispersal is impaired upon EFR3 knockdown

3T3-L1 adipocytes stably expressing HA-GLUT4-GFP were electroporated with siRNA to knockdown EFR3, or a corresponding Scr-siRNA at day 6 post-differentiation as described. Cells were stimulated with 100 nM insulin for 20 min, or left untreated (Basal), before fixation and staining for surface HA as decribed in Materials and Methods. dSTORM images were acquired and reconstructions calculated using ThunderSTORM. (A) Hierarchical density-based spatial clustering of applications with noise analysis for representative control and EFR3 knockdown basal and insulin-stimulated 3T3-L1 adipocytes. GLUT4 molecule coordinates were processed using a python HDBSCAN script written in house. Grey dots indicate single molecule localizations for representative whole cells (left panels). Red boxes indicate region of interests shown in the middle and right panels. The middle panels show coloured molecule clusters identified by HDBSCAN within an ROI (‘clustered only’) and coloured clusters plus grey non-clustered molecules identified by HDSCAN within an ROI (‘clustered and non-clustered’, right panels) are shown. The representative 4 μm × 4 μm ROI is highlighted by the red square. Scale bars = 10 μm. (B) L function analysis of GLUT4 molecule clustering in EFR3 knockdown basal and insulin-stimulated 3T3-L1 adipocytes. GLUT4 molecule coordinates were obtained using ThunderSTORM and its spatial pattern was analysed. We analysed the correlation of points using the variance stabilised L function, which is the transformed version of Ripley’s K function. The L function describes how many points (given by L(r)) can be found within a distance r of any arbitrary point. Empirical estimates of the centred L(r) function are shown in panels labelled (i) control basal and insulin-stimulated cells, (ii) EFR3 knockdown basal and insulin-stimulated cells, (iii) all experimental groups combined. The experiment was carried out independently 4 times on n=10 cells per group. Empirical estimates of the L function from each were pooled together to provide a weighted mean and the 83% (or 1.37σ) confidence interval for each group – see text for details. Complete spatial randomness, modelled from a Poisson process, is indicated by the dashed line. (C) GLUT4 surface density upon EFR3 knockdown. HA (surface GLUT4) localization density was determined from all cells analysed in the presence of insulin. The GLUT4 localisation values do not differ between Scr- and EFR3-siRNA treated groups (n.s.).

Close Modal

or Create an Account

Close Modal
Close Modal