Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Oct 3;221(10):e202106179.
doi: 10.1083/jcb.202106179. Epub 2022 Sep 14.

Engineered kinases as a tool for phosphorylation of selected targets in vivo

Katarzyna Lepeta  1 , Chantal Roubinet  2 , Milena Bauer  1 , M Alessandra Vigano  1 , Gustavo Aguilar  1 , Oguz Kanca  3 , Amanda Ochoa-Espinosa  4 , Dimitri Bieli  5 , Clemens Cabernard  6 , Emmanuel Caussinus  7 , Markus Affolter  1
Affiliations

Engineered kinases as a tool for phosphorylation of selected targets in vivo

Katarzyna Lepeta et al. J Cell Biol. .

Abstract

Reversible protein phosphorylation by kinases controls a plethora of processes essential for the proper development and homeostasis of multicellular organisms. One main obstacle in studying the role of a defined kinase-substrate interaction is that kinases form complex signaling networks and most often phosphorylate multiple substrates involved in various cellular processes. In recent years, several new approaches have been developed to control the activity of a given kinase. However, most of them fail to regulate a single protein target, likely hiding the effect of a unique kinase-substrate interaction by pleiotropic effects. To overcome this limitation, we have created protein binder-based engineered kinases that permit a direct, robust, and tissue-specific phosphorylation of fluorescent fusion proteins in vivo. We show the detailed characterization of two engineered kinases based on Rho-associated protein kinase (ROCK) and Src. Expression of synthetic kinases in the developing fly embryo resulted in phosphorylation of their respective GFP-fusion targets, providing for the first time a means to direct the phosphorylation to a chosen and tagged target in vivo. We presume that after careful optimization, the novel approach we describe here can be adapted to other kinases and targets in various eukaryotic genetic systems to regulate specific downstream effectors.

PubMed Disclaimer

Figures

None
Graphical abstract
Figure 1.
Figure 1.
Schematic illustration of engineered kinases. (a) Overview of engineered kinase work concept. A synthetic kinase uses a small fluorescent protein binder (here: vhhGFP4, a GFP nanobody) to bring a constitutively active kinase domain (kinase) in close proximity to a fluorescent fusion protein (target). The persistence of the kinase domain around the fluorescent fusion protein allows for efficient phosphorylation (P) of the target. (b) Schematic illustration of non-muscle myosin II activation and actomyosin contractility. Non-phosphorylated Sqh/Myosin-II regulatory light chain (MRLC) assembles into an inactive compact molecule through a head to tail interaction. Reversible phosphorylation of Sqh at Ser21 (Ser19 in mammalian MRLC) results in myosin II molecule unfolding, allowing association with other myosin II molecules in an anti-parallel fashion and binding to the actin filaments through the head domains. An ATP-dependent conformational change in myosin II drives the actin filament sliding in an anti-parallel manner and results in contraction. (c) The structure of mammalian and Drosophila ROCK proteins. The Drosophila Rok kinase region shares ∼65% identity with the corresponding isolated domain of mammalian ROCK1 (Verdier et al., 2006). (d–f) The N-terminal kinase region (N-Rok) of Drosophila Rok (amino acid 1–452) was used for synthetic kinase constructs shown in (d–f). (d) Linear representation of N-Rok::vhhGFP4, N-RokDead::vhhGFP4, N-Rok::vhhGFP4-HA, N-RokDead::vhhGFP4-HA, vhhGFP4, N-Rok-HA and N-RokDead-HA. (e) Linear representation of N-Rok::dGBP1, in which a destabilized GFP-binding nanobody (dGBP1; Tang et al., 2016) substitutes vhhGFP4 from d. (f) Linear representation of N-Rok::2m22 and N-RokDead::2m22 in which a DARPin recognizing mCherry substitutes vhhGFP4 from d. In some constructs, the human influenza hemagglutinin tag (HA; black squares) was added to the C-terminus of the synthetic kinases to allow detection by immunofluorescence. N-RokDead contains a K116G single amino acid substitution (G) that abolishes catalytic activity. The proteins are aligned vertically with the N-Rok domain. Numbers refer to amino acid positions from N-terminus (N) to C-terminus (C).
Figure 2.
Figure 2.
N-Rok::vhhGFP4 is a functional enzyme and efficiently phosphorylates Sqh::GFP in vivo in a tissue-specific manner. (a) Expression of N-Rok::vhhGFP4 efficiently recruits Sqh::GFP at the cell cortex in interphasic cells. Stable S2 cell line expressing Sqh::GFP, transfected with N-Rok::vhhGFP4-HA, N-RokDead::vhhGFP4-HA, N-Rok-HA, or N-RokDead-HA. Sqh::GFP is shown in green, anti-HA staining in red (transfected cells), DAPI in dark blue. Graphs on the right show the ratio of cortical Sqh::GFP/cytoplasmic Sqh::GFP of a representative cell shown for each genotype. Scale bars are 5 µm. (b) Quantification of the data shown in a. n = 218 cells, data from three independent experiments. Bars indicate mean ± SD. Asterisks denote statistical significance, derived from unpaired and two-sided Mann–Whitney tests since normality tests showed non-normal distributions: ∗∗∗∗, P ≤ 0.0001; ∗∗, P ≤ 0.01; and n.s., not significant. (c) Stable S2 cell line expressing Sqh::GFP transfected with N-Rok::vhhGFP4-HA and stained with anti-Sqh2P antibody. The right panel shows the magnification of the indicated cell region, and the white arrow points to the enrichment of phospho-Sqh signal at the cell cortex. Scale bars are 5 µm. (d) Panels show lateral views of fixed Drosophila embryos at stage 13–14 (dorsal closure) expressing Sqh::GFP and the synthetic kinase in the engrailed domain (visualized by co-expression of mCherry-nls). Embryos were stained with anti-phospho-Sqh/MRLC antibody. The right panel shows magnification of the respective Sqh::GFP and p-Sqh images for each of the embryo genotypes shown on the left. White arrows point to the actomyosin cable around the dorsal hole, yellow arrows point to Sqh::GFP and p-Sqh foci. Please note that the “control” image is the same as in Fig. 4 a to have a single reference image for all N-Rok variants used. For every expressed synthetic kinase, the number of considered embryos is indicated (n). For N-Rok::vhhGFP4 the n number is given as a proportion of embryos in which clear p-Sqh clumps corresponding to Sqh::GFP foci were clearly visible to the total number of embryos included in the analysis. Scale bar, 20 μm; in the zoomed panel, 10 μm.
Figure S1.
Figure S1.
Binding of vhhGFP4 to GFP results in an increase of the fluorescence signal. All panels show stills from live-imaging with dorsal views of either sqh_Sqh::GFP or sqh_Sqh::mCherry embryos at stage 13/14 (dorsal closure) expressing either N-RokDead variants or vhhGFP4 nanobody alone in the en domain (visualized by co-expression of mCherry-nls). Note the enhancement of fluorescent signal due to the binding of vhhGFP4 or 2m22 to GFP and to mCherry, respectively. The right panel shows magnification of the respective images for each of the embryo genotypes shown on the left. Please note that the “control” image is the same as on Figs. 3 a and 4 b to have a single reference image for all N-Rok variants used. For every expressed synthetic kinase, the number of considered embryos is indicated (n). Scale bars, 50 μm; in the zoomed panel, 10 μm.
Figure 3.
Figure 3.
Synthetic Rok kinase modulates mechanical properties of cells through the phosphorylation of Sqh::GFP and myosin II activity. (a) Schematic illustration of dorsal closure process in the developing fly embryo. Dorsal closure was used as a model to assess myosin II activation by means of Sqh::GFP phosphorylation with synthetic Rok. In sqh_Sqh::GFP embryos, expression of N-Rok::vhhGFP4 leads to abnormal dorsal closure. All panels show stills from live-imaging with dorsal views of the developing sqh_Sqh::GFP embryos at stages 13/14–16 (dorsal closure) expressing variants of the synthetic kinase in the engrailed domain (visualized by co-expression of mCherry-nls). Note the yellow arrows pointing to the Sqh::GFP foci and actomyosin cable invaginations in N-Rok::vhhGFP4 panel. (b) Stills from live imaging with dorsal views of the developing sqh_Sqh::GFP embryo expressing N-Rok::vhhGFP4 in the engrailed domain mounted with gluing to the coverslip technique to show a more severe dorsal open phenotype than embryos imaged on a glass-bottom dish shown in panel a. (c) Dorsal closure was used to compare N-Rok::vhhGFP4 with the previously published effectors that are known to activate myosin II. All panels show stills from live-imaging with dorsal views of the developing sqh_Sqh::GFP embryos at stages 13/14–16 (dorsal closure) expressing myosin II activating tool in the engrailed domain (visualized by co-expression of mCherry-nls). Only for Fig. 3, b and c and corresponding movies, the embryos were imaged with gluing to the coverslip technique; all the other presented embryos were imaged on a glass-bottom dish. Scale bars: 50 μm. Images are representative of n embryos indicated for every expressed synthetic kinase variant.
Figure S2.
Figure S2.
All panels show stills from live-imaging with dorsal views of the developing sqh_Sqh::GFP or sqh_Sqh::mCherry embryos at stages 13/14–16 (dorsal closure) expressing variants of the synthetic kinase or the vhhGFP4 binder alone in the en domain (visualized by co-expression of mCherry-nls). Bottom panel shows dorsal views of embryo expressing SqhE20E21 under the control of enGal4 driver. Yellow arrows point to the Sqh::mCherry foci and actomyosin cable invaginations in N-Rok::2m22 panel. Please note that the “control” image is the same as on Figs. 3 a, 4 b, and Fig. S1, to have a single reference image for all N-Rok variants used. For every genotype, the number of considered embryos is indicated (n). Scale bars, 50 μm.
Figure 4.
Figure 4.
N-Rok::vhhGFP4Vi and N-Rok::dGBP1 are optimized variants of synthetic N-Rok. (a) Panels show lateral views of fixed sqh_Sqh::GFP embryos at stage 13–14 (dorsal closure) expressing N-Rok::vhhGFP4Vi in the engrailed domain (visualized by co-expression of mCherry-nls). Embryos were stained with anti-phospho-Sqh/MRLC antibody. The panel on the right shows the magnification of the respective Sqh::GFP and p-Sqh images for each of the presented embryos. White arrows point to the actomyosin cable around the dorsal hole, yellow arrows point to Sqh::GFP and p-Sqh foci. (b) Dorsal closure was used to compare the functionality of N-Rok::vhhGFP4Vi to N-Rok::vhhGFP4ZH-86Fb (Fig. 3 a). Panels show stills from live-imaging with dorsal views of the developing sqh_Sqh::GFP embryos at stages 13/14–16 (dorsal closure) expressing N-Rok::vhhGFP4Vi in the engrailed domain (visualized by co-expression of mCherry-nls). Please note that both in a and b the “control” images are the same as on Figs. 2 d and 3 a, respectively, to have a single reference image for all N-Rok variants used. (c) Schematic illustration of the synthetic kinase with destabilized GFP nanobody (dGBP1) work concept. In a similar way as in Fig. 1 a, a synthetic kinase uses GFP to bring a constitutively active kinase domain (Rok KD) in close proximity to a fluorescent fusion substrate protein (target). The persistence of the kinase domain around the fluorescent fusion protein achieves efficient phosphorylation (P) of the target. In the absence of the GFP-target, dGBP1 nanobody is destabilized, and the whole nanobody-kinase fusion protein is targeted for degradation. (d) Panels show lateral views of stage 13–14 (dorsal closure) fixed sqh_Sqh::GFP or control yw embryos expressing N-Rok::dGBP1ZH-86Fb in the engrailed domain (visualized by co-expression of mCherry-nls). Embryos were stained with an anti-phospho-Sqh/MRLC antibody. The panel on the right shows the magnification of the respective Sqh::GFP and p-Sqh images. Yellow arrows point to Sqh::GFP and p-Sqh foci. (e) Dorsal closure was used to compare the functionality of N-Rok::dGBP1ZH-86Fb to N-Rok::vhhGFP4ZH-86Fb (Fig. 3 a) and N-Rok::vhhGFP4Vi (b). Panels show stills from live-imaging with dorsal views of the developing sqh_Sqh::GFP or control yw embryos expressing N-Rok::dGBP1 in the engrailed domain (visualized by co-expression of mCherry-nls). For every embryo genotype, the number of analyzed embryos is indicated (n). For N-Rok::dGBP1ZH-86Fb the n number is given as a proportion of embryos in which clear p-Sqh clumps corresponding to Sqh::GFP foci were visible to the total number of embryos included in the analysis. Scale bar, 50 μm in b and e; 20 μm in a and d; in the zoomed panel, 10 μm.
Figure S3.
Figure S3.
N-Rok::2m22 is a functional enzyme and efficiently phosphorylates Sqh::mCherry in vivo in a tissue-specific manner. Panels show lateral views of fixed sqh_Sqh::GFP (top panel) or sqh_Sqh::mCherry (bottom panel) embryos at stage 13–14 (dorsal closure) expressing N-Rok::2m22 in the en domain (visualized by co-expression of mCherry-nls in the top panel). Embryos were stained with anti-phospho-Sqh/MRLC antibody. The panel on the right shows magnification of the respective Sqh and p-Sqh images for the presented embryos. Yellow arrows point to Sqh::mCherry and p-Sqh foci. The number of embryos analyzed is indicated (n). Scale bars, 20 μm; in the zoomed panel, 10 μm.
Figure S4.
Figure S4.
Expression of N-Rok::vhhGFP4Vi in Tkv::YFP embryos has no effect on pMad levels. (a) Panels show lateral views of fixed Tkv::YFP embryos at stage 13/14 (dorsal closure) expressing N-Rok::vhhGFP4Vi in the en domain (visualized by co-expression of mCherry-nls). Embryos were stained with anti-phospho-Mad antibody. (b) Panels show lateral views of fixed embryos at stage 13/14 (dorsal closure) expressing Tkv::GFP and N-Rok::vhhGFP4Vi in the en domain. Embryos were stained with anti-phospho-Sqh/MRLC antibody. (c) Panels show stills from live-imaging with dorsal views of developing DE-Cad::GFP embryos at stages 13/14–16 (dorsal closure) expressing N-Rok::vhhGFP4Vi in the en domain (visualized by co-expression of mCherry-nls). (d) Panels show lateral views of fixed DE-Cad::GFP embryo at stage 13/14. Embryos were stained with anti-phospho-Sqh/MRLC antibody. Note the enhancement of fluorescent signal due to the binding of vhhGFP4 to DE-Cad::GFP in the en domain. The number of considered embryos is indicated (n). Scale bars, 20 μm in a, b, and d; 50 μm in c.
Figure 5.
Figure 5.
Tracheal system of Drosophila embryo develops abnormally due to the excessive myosin II phosphorylation by engineered N-Rok kinase. (a) All panels show lateral views of fixed sqh_Sqh::GFP embryos at stages 14–16 expressing the indicated variant of synthetic Rok kinase in the tracheal system (btl expression domain visualized by co-expression of mCherry-nls). Embryos were stained with an anti-phospho-Sqh/MRLC antibody. For every expressed synthetic kinase, the number of embryos analyzed is indicated (n). For N-Rok::dGBP1, the n number is given as a proportion of embryos in which clear p-Sqh clumps corresponding to Sqh::GFP foci were visible, to the total number of embryos included in the analysis. Scale bar, 20 μm. (b) Aberrant tracheal development in sqh_Sqh::GFP embryos expressing the indicated variant of synthetic Rok kinase in the tracheal system. All panels show stills from live imaging with lateral views of sqh_Sqh::GFP embryos at stages 11/12, 14, and 17, corresponding to the stages depicted on the schematic drawing above. The panel on the right shows magnification of the indicated region from the last stage. Yellow arrows point to Sqh::GFP foci visible already at stage 11/12; asterisks indicate clusters of tracheal cells around the Sqh::GFP foci. Scale bars, 50 μm; in the zoomed panel, 20 μm.
Figure 6.
Figure 6.
Effect of constitutive activation of actomyosin with N-Rok::dGBP1 in dorsal branches on filopodia formation and cell migration. (a) Schematic illustration of dorsal branch formation, showing cell junction rearrangements during cell intercalation. Cell intercalation is brought about by the migration of tip cells towards the fibroblast growth factor (FGF) source. (b) All panels show stills from live imaging with dorsolateral views of dorsal branches of sqh_Sqh::GFP kniGal4 control embryos and sqh_Sqh::GFP kniGal4 Rok::dGBP1ZH-86Fb embryos; LifeAct-mRuby (magenta) was used to visualize pools of F-actin. Arrows indicate the same dorsal branch for all time points; arrowheads point to the tip cell detached from the stalk cells. Scale bars, 20 or 5 μm on the magnification panel.
Figure 7.
Figure 7.
Effect of constitutive activation of actomyosin with N-Rok::dGBP1 in the kni domain on cell intercalation. (a) Confocal projections showing lateral views of fixed sqh_Sqh::GFP kniGal4 control and sqh_Sqh::GFP kniGal4 Rok::dGBP1 embryos stained for DE-Cad (red) and vermiform (blue). Note the presence of ectopic Sqh::GFP foci in the embryos expressing the engineered kinase (arrowheads). The bottom panel shows the magnification of a single branch from the respective image of stage 16 embryos to show the characteristic pattern of lines and small rings of AJs, corresponding to auto- and intercellular AJs, respectively. Yellow arrows point to the rings corresponding to intercellular AJs. (b) Stills from time-lapse movies showing projections of dorsal branches of sqh_Sqh::GFP kniGal4 control embryos and sqh_Sqh::GFP kniGal4 Rok::dGBP1 embryos; α-Cat::mCherry (magenta) marks cell junctions. In the course of the experiments with UAS α-Cat::mCherry stock, we noticed background expression of α-Cat::mCherry in the epidermis, regardless of the driver used, and even in the absence of a driver. However, we verified a robust signal increase with the use of enGal4 driver. Therefore, the observed background of α-Cat::mCherry expression, although limiting a clearer view of the imaged embryos (especially at the beginning of branch emerging), did not prevent the use of this line for studying intercalation in dorsal branches via live imaging. Scale bars, 10 or 5 μm on the magnification panel.
Figure 8.
Figure 8.
Schematic illustration of engineered synthetic Src kinases. (a) Schematic illustration of the synthetic Src kinase work concept. In a similar way as in Fig. 1 a and Fig. 4 c, synthetic kinase uses GFP to bring a constitutively active kinase domain (Src KD) in close proximity to a fluorescent fusion substrate protein (target). The persistence of the kinase domain around the fluorescent fusion protein allows for efficient phosphorylation (P) of the target. In the absence of the GFP-target, dGBP1 nanobody is destabilized, and the whole nanobody-kinase fusion protein is targeted for degradation. (b) The structure of Drosophila Src42A protein. The C-terminal part of Drosophila Src42A (amino acid 214–517) spanning the ATP-binding site, Src catalytic domain, and the crucial regulatory tyrosine residue Y511 was used for synthetic kinase constructs shown in (c–e). (c) Linear representation of dGBP1-HA::Src Y400E, dGBP1-HA::Src Y400D and dGBP1-HA::SrcDead in which N-terminal dGBP1 with HA tag (black squares) was fused with C-terminal part of Src shown in b. (d) Linear representation of Src Y400E, Src Y400D, and SrcDead for which the exact same part of Src42A was used as for the constructs presented in c. These constructs lack the dGBP1 binder, and an in-frame START codon (ATG) was added N-terminally. (e) Linear representation of Src Y400E::dGBP1-HA, Src Y400D::dGBP1-HA, SrcDead::dGBP1-HA and Src::dGBP1-HA in which the order is changed: the dGBP-1-HA binder was fused C-terminally to the Src part and an in-frame START codon (ATG) was added N-terminally. In Src::dGBP-1-HA the regulatory tyrosine Y400 is left unmutated. SrcDead constructs contain a K276M single amino acid substitution (M) that abolishes catalytic activity. Numbers refer to amino acid positions from N-terminus (N) to C-terminus (C).
Figure 9.
Figure 9.
Engineered dGBP1-HA::Src is a functional enzyme and efficiently phosphorylates bsk::GFP in vivo. (a) Panels show lateral views of fixed Drosophila embryos at stages 13–14 (dorsal closure) expressing synthetic Src kinase variants in the engrailed domain in the absence of GFP-fusion substrate. Embryos were stained with anti-phospho-JNK antibody that recognizes Bsk, the Drosophila JNK ortholog, and an anti-phosphotyrosine antibody (pY), that detects tyrosine phosphorylated proteins in all species. (b) Panels show lateral views of fixed Drosophila embryos at stages 13–14 (dorsal closure) expressing bsk::GFP and synthetic Src kinase variants in the engrailed domain. Embryos were stained as in panel a. The right panel in both a and b shows the magnification of the respective pJNK and pY images for each of the embryo genotypes shown on the left. Arrows point to the areas of increased pJNK and pY signal in the engrailed domain where the active kinase is expressed. For every expressed synthetic kinase, the number of considered embryos is indicated (n). Scale bar, 25 μm; in the zoomed panel, 10 μm.
Figure S5.
Figure S5.
Engineered Src::dGBP1-HA is a functional enzyme and efficiently phosphorylates Bsk::GFP in vivo while being innocuous in the absence of the GFP target. (a) Panels show lateral views of fixed Drosophila embryos at stage 13–14 (dorsal closure) expressing synthetic Src kinase variants without the nanobody in the engrailed domain. Embryos were stained with anti-phospho-JNK antibody that recognizes Bsk, the Drosophila JNK ortholog and an anti-phosphotyrosine antibody (pY), that detects tyrosine phosphorylated proteins in all species. (b) Panels show lateral views of fixed Drosophila embryos at stage 13–14 (dorsal closure) expressing Bsk::GFP and synthetic Src kinase variants with C-terminal dGBP1 expressed in the engrailed domain. Embryos were stained as in panel a. The right panel in both a and b shows magnification of the respective pJNK and pY images for each of the embryo genotypes shown on the left. Arrows point to the areas of increased pJNK and pY signal in the engrailed domain where the active kinase is expressed. For every expressed synthetic kinase, the number of considered embryos is indicated (n). Scale bar, 25 μm; in the zoomed panel, 10 μm.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Affolter, M., and Caussinus E.. 2008. Tracheal branching morphogenesis in Drosophila: New insights into cell behaviour and organ architecture. Development. 135:2055–2064. 10.1242/dev.014498 - DOI - PubMed
    1. Aguilar, G., Matsuda S., Vigano M.A., and Affolter M.. 2019. Using nanobodies to study protein function in developing organisms. Antibodies. 8:16. 10.3390/antib8010016 - DOI - PMC - PubMed
    1. Allan, C., Burel J.-M., Moore J., Blackburn C., Linkert M., Loynton S., MacDonald D., Moore W.J., Neves C., Patterson A., et al. . 2012. OMERO: Flexible, model-driven data management for experimental biology. Nat. Methods. 9:245–253. 10.1038/nmeth.1896 - DOI - PMC - PubMed
    1. Amano, M., Chihara K., Nakamura N., Kaneko T., Matsuura Y., and Kaibuchi K.. 1999. The COOH terminus of Rho-kinase negatively regulates rho-kinase activity. J. Biol. Chem. 274:32418–32424. 10.1074/jbc.274.45.32418 - DOI - PubMed
    1. Amano, M., Nakayama M., and Kaibuchi K.. 2010. Rho-kinase/ROCK: A key regulator of the cytoskeleton and cell polarity. Cytoskeleton. 67:545–554. 10.1002/cm.20472 - DOI - PMC - PubMed

Publication types

LinkOut - more resources

-