Figure 1.
Visualisation of the STIM1-STIM1 interaction using a Venus-based BiFC system in resting mammalian cells.
Part A shows the principles of the Venus BiFC system. The two non-fluorescent fragments of Venus, VN (N terminus of Venus) and VC (C terminus of Venus), are each fused to one of a pair of interacting (test) proteins, A and B. The fusion proteins, VN-A and VC-B, do not fluoresce when expressed separately. If proteins A and B interact or associate, the two fluorescent fragments are brought together, and this facilitates reconstruction of the fluorescent protein. Part B is a schematic view of the fusion protein constructs used in this study. WT-STIM1 and STIM1 mutants were fused to the N- and C-terminal fragments of Venus. The functional domains of STIM1 include an EF-hand, a SAM domain, a transmembrane domain, a coiled-coil (CC) region, an ERM domain, a Ser/Pro-rich domain (SP), and a polylysine residue region (K). Part C shows images of HEK293T cells 24 h after transfection with plasmids encoding VN173-ST1, VC155-ST1 or VN173-ST1/VC155-ST1 (full length STIM1). VN173-ST1 or VC155-ST1 expressed alone did not emit fluorescence under excitation, but coexpression of VN173-ST1 and VC155-ST1 produced strong yellow fluorescence emission. Part D shows HEK293T cells coexpressing VN173-ST1 and VC155-ST1 under resting and TG-stimulated conditions. Coexpression of VN173-ST1 and VC155-ST1 produced fluorescence that displayed a fibrillar distribution in resting cells and that became punctate and localised to the ER-PM junctions following stimulation of cells with 2 uM TG. Scale bars, 20 µm.
Figure 2.
A C-terminal region of STIM1 (233–474) is critical for the oligomerisation of STIM1 in resting cells.
Part A shows images of HEK293T cells cotransfected with the VN173-ST1 and VC155-ST1, VN173-ΔSAM and VC155-ΔSAM, or VN173-ΔC and VC155-ΔC constructs. Images were acquired 24 h after transfection. Part B shows quantitative analysis of Venus-based BiFC efficiency measured from experiments such as those shown in Part A. Part C shows that VN173-ST1 and VC155-ST1, VN173-ΔSAM and VC155-ΔSAM, or VN173-ΔC and VC155-ΔC constructs were cotransfected into HEK293T cells, respectively, and these proteins were detected by immunoblotting with anti-STIM1 Ab(Upper). Tubulin protein was detected as a loading control (Lower). All data are given as mean±S.D. (n>50). The statistical significance was evaluated using a two-tailed Student’s t-test when compared with the combination of VN173-ST1 and VC155-ST1. ST1, ΔC and ΔSAM represent full length STIM1, truncated STIM1-ΔC (Δ233–474) and STIM1-ΔSAM (Δ132–200), respectively. The significance level indicated is as follows: **P<0.001. Scale bar, 20 µm.
Figure 3.
FRET measured by donor dequenching after acceptor photobleaching in HEK293T cells cotransfected with eCFP-STIM1 and eYFP-STIM1.
Part A: CFP-STIM1 images before and after photobleaching of the acceptor within the indicated region (Left column); acceptor YFP-STIM1 intensities before and after photobleaching in the indicated region (Right column). Part B: images acquired near the cell adhesion surface after stimulation of cells with 2 µM TG. Part C: the bar graphs representing FRET efficiency (E) are from 20 independent experiments such as those in part A. The efficiency was determined by the acceptor photobleaching method and was measured only in the (acceptor) bleached area. Cells outside the bleached region were used as controls. Part D: the bar graphs representing FRET efficiency (E) are from the 20 independent experiments in part B. All data are represented as mean±S.D. The significance levels indicated are as follows: **P<0.001. Scale bars, 20 µm.
Figure 4.
Ca2+ store depletion causes relocalisation of coexpressed BiFC-STIM1 and mCherry-STIM1, or coexpressed BiFC-STIM1 and Orai1-eCFP to ER-PM junctions to form colocalised puncta.
Part A: HEK293T cells were cotransfected with VN173-ST1, VC155-ST1 and mCh-STIM1 constructs. Live cells were examined under a confocal microscope at the cell footprint. Confocal images of the same cells were taken before (Rest) and 3 min after addition of TG (TG). In resting cells, mCh-STIM1 and BiFC-STIM1 exhibited a colocalised fibrillar distribution (Rest). Ca2+ store depletion caused BiFC-STIM1 and mCh-STIM1 to form colocalised puncta. Part B: HEK293T cells were cotransfected with VN173-ST1, VC155-ST1 and Orai1-eCFP constructs. After Ca2+ store depletion, Orai1-eYFP accumulated with BiFC-STIM1 as puncta localised near the plasma membrane. Scale bars, 20 µm.
Figure 5.
2-APB reverses or prevents BiFC-STIM1 relocalisation to puncta near the plasma membrane.
HEK293T cells were cotransfected with VN173-ST1 and VC155-ST1 constructs. After 24 h, live cells were examined under a confocal microscope at the cell footprint. Part A: VN173-ST1 and VC155-ST1 coexpressed in HEK293T cells redistributed from a fibrillar appearance (Rest) to puncta at the cell periphery after Ca2+ store depletion with TG. Puncta were visible at the cell footprint. The addition of 50 µM 2-APB rapidly reversed the relocalisation of BiFC-STIM1 to plasma membrane puncta (2-APB). Part B: 2-APB prevented BiFC-STIM1 relocalisation to puncta near the plasma membrane. Confocal imaging was carried out in experiments in which cells expressing BiFC-STIM1 were pretreated with 50 µM 2-APB for 5 min prior to store depletion with TG. Scale bars, 20 µm.
Figure 6.
Functional validation of the fusion protein of BiFC-STIM1 in HEK293T cells.
SOCE was assayed using Ca2+ microfluorimetry in HEK293T cells transfected with eYFP-STIM1 alone or with both VN173-ST1 and VC155-ST1 plasmids. The cells were first treated with TG (2 µM) in a Ca2+-free solution to empty the Ca2+ stores and subsequently switched to normal extracellular solution containing 2 mM Ca2+; this induced a transient increase in the cytosolic Ca2+ concentration. Part A: representative trace of Ca2+ dynamics in control HEK293T cells overexpressing pH-STIM1. Part B: representative trace of Ca2+ dynamics in cells cotransfected with both VN173-ST1 and VC155-ST1. Part C: mean data showing the effect of 50 µM 2-APB on SOCE from experiments such as those shown in A (n = 18 cells; **p<0.001). Part D: mean data showing the effect of 50 µM 2-APB on SOCE in experiments in part B (n = 18 cells; **p<0.001).