Fig 1.
Ligand-induced [35S]GTPγS binding of different fusions of Gαi1/q at the C-terminus of rat NTR1 mutant, TM86V-L167R.
(a) Schematic diagram of the longest NTR-Gα fusion construct. Residues after helix 7 are shown in circles and helix 8 (from Ser373) is indicated as a schematic projection. Potential palmitoylation sites (Cys386 and Cys388) are shown in yellow circles. In the longest construct, the C-terminus of the receptor was fused to the N-terminus of the Gαi1/q chimera with a TEV cleavage site (grey circles) and a (N5)-(G3S)2 linker (black circles) in between. In all the other shorter constructs, the Gαi1/q was directly fused to the different C-terminal amino acid positions. The arrows (empty) indicate the location of truncation sites as evaluated in panel b and the number after “Δ” indicates the first missing residue. (b) NT8-13-induced [35S]GTPγS assay on E. coli membranes containing the MBP-TM86V-L167R-Gαi1/q constructs. A complete or partial deletion of helix 8 in the fusion construct abrogates the signaling competency. Data are given as a mean ± s.e.m. of 2–7 independent experiments performed in at least in duplicates. *p<0.05 compared to the basal level calculated by Student’s t-test. The response (percentage increase in the binding of [35S]GTPγS above basal level) at 200 μM NT are 46.6±15% for Δ390-Gαi1/q (n = 5); 39.8±9% for Δ394-Gαi1/q (n = 7); 48.3±13% for Δ398-Gαi1/q (n = 4); 36.1±10% for Δ421-TEV-(N5)-(G3S)2-Gαi1/q (n = 7)).
Fig 2.
Schematic representation of the assembly of rNTR1-Gαi1/q, Gβ1 and Gγ1 subunits on a single baculovirus.
(a) (1) The construct of a rat NTR1-Gαi1/q fusion was cloned into the acceptor vector pFL with the help of ligation-independent cloning (LIC). (2) N-terminally RGS(His)10-tagged Gβ1 and N-terminally hemagglutinin (HA)-tagged Gγ1 were each cloned into an independent donor vector pIDC via LIC. (3) The Gβ1 and Gγ1 were then assembled onto a single pIDC plasmid with the help of a multiplication module (see panel b for detail), resulting in plasmid “pIDC βγ” (4) Using Lox-P sites (crossed circle, orange), an in vitro Cre recombination was performed. This resulted in one transfer vector on which all the three genes were present in a stoichiometry of 1:1:1, and each gene was under the control of a separate polyhedrin promotor (grey arrow). (5) DH10 EMBacY cells were transformed with the resulting transfer vector. There, the portion containing the expression modules (sequence between the two inverted black triangles) was integrated into the baculovirus genome via Tn7 transposition. (6) The virus genome was isolated and used to transfect insect cells, resulting in the formation of a first generation of baculovirus, which was used for high-level heterologous protein production. (b) Details of the multiplication module used to insert two (or more) genes into the pIDC vector. In the above-mentioned cloning scheme, the pIDC vector containing the Gβ1 subunit (“pIDC β”) was linearized using the homing endonuclease PI-SceI. The expression cassette (consisting of the promoter, the gene of interest and the polyadenylation site (not shown for clarity)) from the second pIDC vector containing the Gγ1 subunit (“pIDC γ”) was digested by PI-SceI and BstXI (recognition sequences are shown at the bottom of the panel). As the overhangs generated by digestion with BstXI (general recognition sequence: 5’-CCANNNNNNTGG-3’) was designed to be compatible to PI-SceI overhangs, the expression cassette could be inserted into the linearized “pIDC β” resulting in “pIDC βγ”. After ligation, the original PI-SceI site of the recipient vector was eliminated, while a new PI-SceI site was generated downstream of the newly inserted expression module, which, in principle, could again be used to integrate a new expression cassette.
Fig 3.
Expression and purification of rNTR1*-Gαi1/qβ1γ1 from insect cells.
(a) Purification scheme for the purification of rNTR1*-Gαi1/qβ1γ1. The rNTR1-Gαi1/q fusion and β1γ1 are co-expressed in insect cells using the MultiBac system, their co-expression leads to in vivo complex formation. After isolating the membrane, solubilization was carried out in a detergent of choice, (see text for detergents used). A ligand-affinity column was used to isolate the functional complex. The protein was then eluted from the NT ligand-affinity column using 3C protease (homemade, containing His6 tag) which cleaves the intact receptor-ligand complex from the column support; the 3C protease was then removed by a reverse Ni2+-NTA affinity column. The isolated complex was further polished using size exclusion chromatography (SEC) (b) Representative elution profile of purified fusion-protein complex on a Superdex 200 10/300 GL column (GE Healthcare). (c) SDS-PAGE Coomassie-stained gel of the pooled fractions (dashed lines in b) from the gel filtration column. (d) The stability of the purified complex was monitored by analytical SEC on the same column after incubating the protein for 6 days at 4°C. The SEC profile and SDS-PAGE image (b and c, respectively) are representative of more than fifteen independent purifications. d represents protein purified using an MNG:CHS detergent mixture. The asterisk (*) denotes that a mutant of rNTR1 was used: HTGH4-ΔICL3(B) (see text).
Fig 4.
Effect of ionic strength, pH, nucleotide analogs, DMSO concentration, and TCEP on the stability of the rNTR1*-Gα1i/qβ1γ1 complex.
(a) Analytical gel filtration showing the NaCl tolerance of the complex. The protein mixture was incubated for three days at 4°C. The complex was stable at 100 mM and eluted as a prominent monodisperse peak, but at 500 mM and 1 M NaCl, an aggregation peak was seen. Lower salt concentration (20 mM NaCl) affected the monodispersity of the complex. (b) The effect of various pH values was probed using analytical gel filtration, after incubating the protein for three days at 4°C in a buffer adjusted to the given pH value. Acidic pH at 5.5 or below was deleterious to the stability of the complex. (c) GDP and GTPγS analogs (100 μM, final concentration) were mixed with the purified complex and the mixture was incubated overnight at 4°C; as it can be seen, the nucleotide analogs caused a partial dissociation of the complex. The excess nucleotides in the buffer migrate at about 2.7 mL on the gel filtration column (not shown). (d) Purified protein was incubated without or with 100 μM TCEP for up to seven days at 4°C, however, there was no change of the monodisperse peak under either condition. (e) The effect of increasing concentrations of DMSO (2% and 4% (v/v), final concentration) was also tested; the mixture of DMSO and purified protein was incubated for a period of three days at 4°C. The stability of the complex was not affected by up to 4% (v/v) DMSO. All the analytical gel filtrations were performed on a Superdex 200 Increase 5/150 GL column (GE healthcare). The SEC profiles are representatives of experiment(s) performed once (a, b) or twice (c, d and e). * rNTR1 mutant used: HTGH4-ΔICL3(B).
Fig 5.
Comparison of GFP (or variant) affinity and ligand affinity purification.
(a) Schematic diagram of the construct used for comparing the GFP trap and the ligand pull-down methods. The NTSR1-EL mutant with deleted ICL3 (named EL _ΔICL3(B)) was fused to Gαi1/q15, a 3C-cleavable RGS-decahistidine tag was present at the N-terminus of Gβ1, while a deca-histidine tag followed by an enhanced YFP (eYFP) followed by a 3C protease cleavage site was present on the N-terminus of Gγ1. Prepared membranes were separated into two fractions. Numbers denote two schemes of purification. 1 –GFP-affinity scheme: membranes were solubilized in the presence of NT1 (a variant of NT8-13, see Section 4.9); batch binding with GFP-affinity beads; after washing elution was done using cleavage by 3C protease, the eYFP moiety was retained on the column while a mixture of protein and 3C protease was eluted; reverse Ni2+-NTA removed the co-eluted 3C protease; the flow-through was used for analytical SEC (aSEC). 2 –ligand-affinity scheme: membranes were solubilized in the absence of NT1; batch binding with NT-ligand affinity beads, after washing elution was done using cleavage by 3C protease; the mixture of protein, eYFP and 3C protease was eluted; reverse Ni2+-NTA removed the eYFP and the 3C protease; the flow-through was used for aSEC. (b) Comparison of aSEC runs of protein from the two columns. (c) The highest protein-containing fraction from aSEC was analyzed using SDS-PAGE followed by silver staining. Lanes: M- molecular weight marker, YFP- protein fraction for the GFP column eluate. NT lig.—Protein from the ligand-affinity column. *rNTR1 mutant used NTSR1-EL-ΔICL3(B).