Supplementary Materials For: One-step split GFP staining for sensitive protein detection and localization in mammalian cells Lara Kaddoum 1,3, Eddy Magdeleine 1,3, Geoffrey S. Waldo 4, Etienne Joly 1,3, and Stéphanie Cabantous 2,3 1 CNRS, Institute of Pharmacology and Structural Biology (IPBS), Toulouse, France, 2 INSERM U563, Département Innovation Thérapeutique et Oncologie Moléculaire, Institut Claudius Régaud, Université Paul Sabatier, Toulouse, France, 3 Université Paul Sabatier, IPBS, Toulouse, France, and 4 Bioscience Division, MS-M888, Los Alamos National Laboratory, Los Alamos, NM, USA BioTechniques 49:727-736 (October 2010) doi 10.2144/000113512 Keywords: epitope tagging; protein localization; split GFP; green fluorescent protein; intracellular staining; protein fragment complementation Preparation of GFP 1-10 for in vitro assays Bacterial cell culture A solution of purified GFP 1-10 needs to be prepared to perform in vitro detection of fusion proteins. The engineered GFP 1-10 is ~50% soluble, and the inclusion body fraction is processed to take advantage of the enrichment and partial purification afforded by using inclusion bodies. A 3-mL overnight culture is used to inoculate 500 ml Luria-Bertani broth containing 35 μg/ml kanamycin. GFP 1-10 expression is induced in the exponential phase (2 h after inoculation) with 1 mm IPTG for 5 h at 37 C (this forces the GFP 1-10 to aggregate into inclusion bodies). After centrifugation, cell culture pellets are resuspended in 15 ml 100 mm Tris-HCl, ph 7.4, 0.1 M NaCl, 10% glycerol (v/v, TNG buffer), sonicated, and centrifuged again to recover pellet cell debris containing crude inclusion bodies (discard supernatant). Inclusion bodies purification Ten milliliters protein extraction reagent (e.g., Bugbuster, Novagen, EMD Chemicals, Gibbstown, NJ, USA) are added to the pellet and sonicated to resuspend the crude inclusion bodies. After centrifugation at 8,000 g for 20 min, the supernatant is removed. Three of these Bugbuster washes are performed, followed by two washes with 10 ml TNG buffer (this step enables residual detergent from the pellet mass). The pellet is weighed and resuspended in appropriate TNG volume to obtain a concentration of ~37.5 mg/ml washed pellet. Inclusion bodies are resuspended by sonication and dispensed into 1-mL aliquots in 1.7-mL microcentrifuge tubes. After centrifugation, the supernatant is removed. The pellet can be stored at -80 C for several months. Preparation of GFP 1-10 reagent for assay GFP 1-10 inclusion body pellet is dissolved by adding 1 ml 9 M urea containing 5 mm DTT and by incubating the tube in a 37 C water bath to help dissolution of the inclusion bodies. After complete dissolution, the microcentrifuge tube is centrifuged for 1 min at 16,000 g to remove any aggregated material, and the supernatant is transferred into a 50-mL Falcon tube containing 25 ml TNG buffer. The solution is mixed gently by inversion and filtered through a 0.2-μm syringe filter. This solution is ready to use for in vitro protein quantification assays. The remaining solution may be stored up to 2 weeks at -20 C. Supplementary Figure S1. FACS detection of MeCP2 in HEK cells. Top panels represent staining of nontransfected HEK cells (U), and bottom panels represent staining of G418-resistant HEK cells after transfection with MeCP2 (T), of which only around 15% are expressing detectable levels of the protein. (Left panels) Anti-MeCP2 immunostaining. Gray line in the upper left panel represents staining with secondary antibody only. (Right panels) Staining with GFP 1-10. The number adjacent to each peak corresponds to the mean fluorescence intensity of these peaks. 1
Supplementary Table S1. FACS fluorescence values of cells after GFP 1-10 staining. Time (h) 4 C Room temperature 37 C Mean fluorescence intensity of GFP Mean fluorescence intensity of GFP Mean fluorescence intensity of GFP MeCP2 MeCP2 N 2 A M e C P 2 0 2.21 6.10 2.76 2.21 8.51 3.85 2.17 11.55 5.32 0.5 2.15 54.25 25.23 1.95 71.05 36.44 1.68 79.15 47.11 1 2.11 90.58 42.93 1.86 137 73.66 1.64 132.16 80.59 1.5 2.15 127.49 59.30 2.02 207.21 102.58 1.72 170.01 98.84 2 2.17 152.61 70.33 2.07 248.05 119.83 1.79 186.01 103.92 3 2.23 181.06 81.19 2.29 313.4 136.86 1.98 210.97 106.55 4 2.35 214.80 91.40 2.74 381.97 139.41 2.29 241.44 105.43 6.5 2.81 278.81 99.22 3.75 445.08 118.69 2.64 276.32 104.67 9.5 3.31 313.40 94.68 4.61 461.38 100.08 3.16 305.05 96.53 11.5 3.59 355.45 99.01 5.23 465.55 89.02 3.43 316.23 92.20 24 6.04 453.16 75.03 8.58 445.08 51.87 4.96 299.61 60.41 Fixed and permeabilized cells were incubated with GFP 1-10 at three temperatures: 4ºC, RT, and 37ºC. Fluorescence was measured by FACS at different time points (left column). Signal/noise ratio is indicated for each data set in a third column by dividing mean fluorescence intensities of transfected cells (-MeCP2-GFP11) and control cells (). 2
Supplementary Figure S2: FACS detection of -H-Ras in HEK cells. Top panels represent staining of nontransfected HEK cells (U), and bottom panels represent staining of HEK cells after transient transfection with -H-Ras (T). (Left panels) Anti H-Ras immunostaining with anti H-Ras MAb (Pierce, Thermo Scientific). Gray line in the upper left panel represents staining with secondary antibody only. (Right panels) Staining with GFP 1-10. The number adjacent to each peak corresponds to the mean fluorescence intensity of these peaks. In transfected cells, these fluorescence intensities are indicated above for each cell population, for those expressing basal levels of H-Ras and transfected cells. The table summarizes these values. GFP 1-10 staining of transiently transfected HEK cells still yielded better signal-to-noise ratios than antibody staining (50 versus 13). 3
(a) (b) (c) Supplementary Figure S3. Absence of green and red channel interference for single staining of MeCP2 cells with anti-mecp2 polyclonal antibody or GFP 1-10 reagent. No green residual fluorescence is observed in the FITC channel when anti-mecp2 rhodamine immunostaining is performed (A). Inversely, GFP 1-10 staining reveals only fluorescence in the FITC channel with no overlap in the red channel (B). Unstained -MeCP2-GFP fusions display very slight red fluorescence, which is not correlated with MeCP2 localization but rather to background cell fluorescence (C). 4
(a) (b) (c) Supplementary Figure S4. Absence of green and red channel interference for single staining of HEK cells expressing H-Ras using anti H-Ras MAb or GFP 1-10 reagent. Anti H-Ras rhodamine immunostaining reveals membrane localization in the red channel for H-Ras in HEK cells, while no green residual fluorescence is observed (A). With GFP 1-10 staining, clear membrane localization is revealed in the FITC channel with some background cell fluorescence in the red channel, which is not related to H-Ras localization (B). Similar effect is observed in HEK cells as in cells for the GFP H-Ras fusion protein, with some staining of intracellular membrane compartments not seen with the tagged form (C). 5