Tripartite split-GFP assay to identify selective intracellular nanobody that suppresses GTPase RHOA subfamily downstream signaling

Abstract

Strategies based on intracellular expression of artificial binding domains present several advantages over manipulating nucleic acid expression or the use of small molecule inhibitors. Intracellularly-functional nanobodies can be considered as promising macrodrugs to study key signaling pathways by interfering with protein-protein interactions. With the aim of studying the RAS-related small GTPase RHOA family, we previously isolated, from a synthetic phage display library, nanobodies selective towards the GTP-bound conformation of RHOA subfamily proteins that lack selectivity between the highly conserved RHOA-like and RAC subfamilies of GTPases. To identify RHOA/ROCK pathway inhibitory intracellular nanobodies, we implemented a stringent, subtractive phage display selection towards RHOA-GTP followed by a phenotypic screen based on F-actin fiber loss. Intracellular interaction and intracellular selectivity between RHOA and RAC1 proteins was demonstrated by adapting the sensitive intracellular protein-protein interaction reporter based on the tripartite split-GFP method. This strategy led us to identify a functional intracellular nanobody, hereafter named RH28, that does not cross-react with the close RAC subfamily and blocks/disrupts the RHOA/ROCK signaling pathway in several cell lines without further engineering or functionalization. We confirmed these results by showing, using SPR assays, the high specificity of the RH28 nanobody towards the GTP-bound conformation of RHOA subfamily GTPases. In the metastatic melanoma cell line WM266-4, RH28 expression triggered an elongated cellular phenotype associated with a loss of cellular contraction properties, demonstrating the efficient intracellular blocking of RHOA/B/C proteins downstream interactions without the need of manipulating endogenous gene expression. This work paves the way for future therapeutic strategies based on protein-protein interaction disruption with intracellular antibodies.

Keywords: RHO-ROCK signaling; RHOA GTPase; nanobodies; single domain antibody (sdAb); tripartite split-GFP.

Figure 1

The tripartite split GFP assay demonstrates the intracellular interaction and the selective recognition of the active RHO conformation by the selected intrabodies. (A) Principle of the tripartite split-GFP complementation assay adapted to assess hs2dAb/RHO interaction. β-strand 10 (GFP10) and β-strand 11 (GFP11) are fused to RHOA mutants (either DN RHOA N19 or CA RHOA L63) and hs2dAb, respectively. These constructions are transfected in a MRC5 cell line that constitutively expresses the detector fragment GFP1–9 (β-strands 1–9). When protein interaction occurs, GFP10 and GFP11 strands are tethered and then spontaneously associate with GFP1–9 fragment to form a full-length GFP. If the two proteins do not interact, GFP10 and GFP11 are not tethered and entropy is too high to allow complementation with GFP1–9. (B) Percentage of reconstituted GFP (rGFP) fluorescent cells analyzed by flow cytometry for the indicated transfection conditions. P-values were calculated using a Student’s t test. **, p<0.01.

Figure 2

The RH28 is selective for active RHOA conformation in living cells. Percentage of reconstituted GFP (rGFP) fluorescent cells analyzed by flow cytometry for the indicated transfection conditions. (A) constitutively active mutants (B) wild-type proteins. P-values were calculated using a Student’s t test.*p<0.05; **p<0.01.

Figure 3

RH28 competes with RBD for active RHOA recognition. (A) Principle of the tripartite split-GFP complementation assay adapted to assess hs2dAb/RBD competition. Hs2dAb were transfected in the triSFP RHOA cell line, 24 hours before doxycycline induction. Among cells that express hs2dAb, a competition with RBD would lead to a decrease in the intensity of the rGFP fluorescence in mCherry positive cells. (B) For each hs2dAb, rGFP fluorescence intensity was quantified among the 4 different populations of mCherry positive cells (i.e., among the 4 cell populations ranked according to increasing levels of hs2dAb (or RBD) expression). (C) RHOA G-LISA competition assay with 10-fold dilutions of hs2dAb. Results were analyzed with two-way ANOVA model. Absorbance at 485 nm reflects RHOA-GTP captured by the coated RBD. P-values were calculated using a Student’s t test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 4

RH28 intrabody efficiently inhibits actomyosin contractility and blocks RHOA/ROCK pathway in melanoma cancer cells. (A) RH28 selectivity validation in WM266.4 lentiviral cell line. Expression of RH28 and NR-hs2dAb was induced or not with doxycycline at 1µg/ml. After 20 hours of induction, cells were harvested and cleared cell lysates were incubated with Ni-NTA beads for 45 min. Endogenous RHOA proteins was revealed with corresponding antibodies and hs2dAb were revealed with myc-tag antibody. (B) Representative images of cell line morphology in 3D collagen drops after 6 hours of hs2dAb expression. WM266.4 cell lines were seeded in collagen and phenotypes were analyzed 24 hours post doxycycline induction. At 40X magnification, elongated or rounded cell shape could be observed. Scale bar: 50 µm. (C) Representative images and quantification of gel contraction after 72 hours of treatment by doxycycline. (D) Representative immunoblot and quantification analysis of myosin light chain phosphorylation (pMLC2) status in the 3 different cell lines seeded in collagen drops. P-values were calculated using a Student’s t test.*p<0.05; **p<0.01.