|G-Protein Signaling Through Tubby Proteins|
Sandro Santagata, Titus J. Boggon, Cheryl L. Baird, Carlos Gomez, Jin Zhao, Wei Song Shan, David G. Myszka, Lawrence Shapiro
Supplemental Figure 2. ClustalW sequence alignment of the C-terminal domain region from tubby homologues. Residues involved in binding PtdIns(4,5)P2 in the mouse tubby co-crystal structure are boxed in red. Residues that are identical in a majority of the aligned sequences are highlighted in yellow. The aligned sequences correspond to the following GenBank accession codes: Tubby (Mus musculus), 1717822; TULP1 (Homo sapiens), 6715610; TULP2 (H. sapiens), 4507737; TULP3 (H. sapiens), 4507739; rat (Rattus norvegicus), 6981686; chicken (Gallus gallus), 12719458; drosophila (Drosophila melanogaster), 7291244; elegans (Caenorhabditis elegans), 3875712; arabidopsis (Arabidopsis thaliana), 2829918.
Supplemental Figure 3.A. Substitutions of the palmitoylation sites of Gαq* alter its capacity to induce transit of tubby to the nucleus. Cysteines 9 and 10 at the N-terminus of Gαq*(Q209L) which anchors Gαq proteins to the cell membrane when palmitoylated, were mutated to serines. When co-transfected into Neuro-2A cells along with pGFPtub-fl, Gαq*(Q209L) resulted in predominant nuclear accumulation of tubby while the cysteine mutant Gαq*(C9S,C10S, Q209L) was significantly hindered in its capacity to mediate nuclear translocation of tubby from the cell membrane to nuclear compartment. Images were acquired by CLSM from coverslips generated 24 hours after transfection. B. G proteins β2 and γ2 along with phospholipase C β1 do not alone mediate shuttling of tubby into the nucleus. While Neuro-2A cells co-transfected with expression constructs for Gαq* and GFP-tubby demonstrate a substantial nuclear distribution of tubby, co-transfection with G proteins β2 and γ2 and PLC-β1 did not induce translocation of GFP-tubby into the nuclear compartment. Images were acquired by CLSM from coverslips prepared 24 hours post-transfection.
Supplemental Figure 4. GST TULP3 C-terminal domain binds to a similar pattern of phosphorylated inositide lipid head groups as the C-terminal domain of tubby as determined through analysis using PIP strips (18).
Supplemental Figure 5. Amino acid substitutions in the 5HT2c receptor that decrease PI(4,5)P2 hydrolysis significantly reduce GFP tubby translocation to the nucleus. Neuro-2A cells plated on coverslips in six-well plates were transfected (Qiagen effectene reagent) with 100 ng of pGFPtub-fl and 10 ng of wild type or double mutant (33) pCDNA5HT2c receptor. Coverslips were processed 24 hours post transfection, and images acquired by confocal laser scanning microscopy (CLSM). As the amino acid substitutions are known to significantly reduce but not completely abolish the levels of PI(4,5)P2 hydrolysis, it was critical to perform the experiments in the appropriate dose response range.
Supplemental Figure 6. Receptor mediated activation of phospholipase Cγ does not induce tubby translocation from the plasma membrane to the nuclear compartment. 293T cells were transfected using a calcium phosphate transfection protocol with pGFPfl-tub and with either cDNA expression constructs for the 5HT2c receptor, epidermal growth factor receptor (EGFR), a chimeric molecule encoding the extracellular domain of EGFR and the intracellular portion of fibroblast growth factor receptor (FGFR) or a chimeric molecule of the extracelular region of EGFR and the intracellular region of trkB, a receptor for nerve growth factor (trkB). Images were acquired from living cells 24 hours post-transfection using Nikon Eclipse TE200 inverted microscope and a DKC 5000 Sony Digital Photocamera. Only the 5HT2c receptor, which signals through Gαq and PLC-β, induced nuclear localization of tubby. This provides further evidence of the specificity of tubby activation; hydrolysis of PI(4,5)P2 is insufficient in the absence of Gαq activation.
Supplemental Figure 7.A. Inhibition of PKC does not block GFP-tubby translocation to the nucleus. Neuro-2A cells were transfected with GFP-tubby and AchR-M1. Twenty-four hours post-transfection, the cells were preincubated for 10 minutes at 37�C in media with either DMSO (no inhibitor) or 10μM of the PKC inhibitor Bisindolylmaleimide I (BIM-I) dissolved in DMSO. BIM-I acts as a competitive inhibitor for the ATP binding sites of PKC α,ΒI,ΒII,γ,δ?and ε. After 10 minutes, 100μM acetylcholine was added to the media. As demonstrated by the image BIM-I did not effect the efficiency of GFP tubby translocation in response to stimulation of the AchR-M1. The reddish-orange hue in the cytoplasm of the BIM-I stained cells represents auto-fluorescence of the inhibitor. B. Ca2+ influx using the ionophore ionomycin did not trigger dissociation of GFP-tubby from the cell membrane. Neuro-2A cells (six-well plates) were transfected with 400 ng pGFPtub-fl and were incubated with either DMSO alone (no inhibitor) or 10μM ionomycin (dissolved in DMSO). Cells were incubated for 48 hours and were regularly monitored for localization of GFP tubby from the initiation of the experiment. At no point in this period did the localization of tubby differ between the treated and untreated cells. Images were acquired at 6 hours post ionomycin addition using a Nikon Eclipse TE200 inverted microscope and a DKC 5000 Sony Digital Photocamera.