Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: Methods Mol Biol. 2018;1821:87–106. doi: 10.1007/978-1-4939-8612-5_7

Characterization of Genetically Encoded FRET Biosensors for Rho-Family GTPases

SK Donnelly 1,2,#, V Miskolci 3,#, AM Garrastegui 1, D Cox 4,5, L Hodgson 6,7
PMCID: PMC6104821  NIHMSID: NIHMS984139  PMID: 30062407

Abstract

Genetically encoded FRET-based biosensors are increasingly popular and useful tools for examining signaling pathways with high spatial and temporal resolution in living cells. Here, we show basic techniques used to characterize and to validate single-chain, genetically encoded Förster resonance energy transfer (FRET) biosensors of the Rho GTPase-family proteins. Methods described here are generally applicable to other genetically encoded FRET-based biosensors by modifying the tested conditions to include additional/different regulators and inhibitors, as appropriate for the specific protein of interest.

Keywords: Rho GTPases, FRET, Biosensors, Fluorometry, Validation, Ratiometric analysis

1. Introduction

Fluorescent biosensors for Rho-family GTPases [1–19] have revealed new aspects of biology that had been inaccessible in the past, because of their superior ability to report dynamic changes in living cells and within subcellular compartments that render traditional approaches suboptimal. Biosensor designs have also evolved to optimize for better response, better stability, and improved optophysics. As end users, it is often easy and tempting to treat these biosensors as a monolith, without questioning the specific considerations that went into the design, validation, and characterization of these molecular constructs. As biosensor makers, we often fall into a situation whereby we pursue increased dynamic range of FRET response under very specific conditions without considering other minute molecular nuances, which may be even more important in understanding the biology of our target molecule. In order to reconcile these issues, we describe here basic guidelines for characterizing and validating these biosensor constructs. These methods are useful to not only biosensor designers but also the users of these tools. After obtaining a new biosensor designed by someone else, we often subject the sensor to these types of assays to validate in our own hands prior to performing biological experiments. This ensures a clearer understanding of how and what the sensor actually senses, and provides increased reliability for the data gathered using such tools. Biosensors are often engineered to achieve specific goals that may not be compatible with what is required in each individual circumstance, unless the limitations and features of specific biosensor are clearly understood.

In this chapter we describe these validation guidelines using Rac/Cdc42 family of GTPase biosensors as models [3, 4, 20]. These are monomeric, single-chain biosensors that contain FRET donor monomeric Cerulean1 [21], a tandem binding domain from PAK1 (PBD) that allows autoinhibitory control, circularly permuted monomeric Venus [22,23], and a C-terminally attached GTPase (see Fig. 1) [3,20]. These biosensors have been designed with good sensitivity, selectivity, ability to interact with appropriate upstream regulators, and ability to insert into correct membrane domains upon activation due to their intact C-terminal lipid modification motif. The validation and characterization approaches presented here will allow an appreciation of what exactly these sensor constructs are responding to in cells and their utilization will enable appropriate biological interpretations to be made based on their fluorescence readouts.

Fig. 1.

Fig. 1

Open in a new tab

Cartoon of the Rac2 biosensor design, based on the Rac1 biosensor [3], where it contains monomeric fluorescent proteins mCerulean1 and circularly permutated monomeric Venus as the FRET pair (a). The detection module PBD incorporates autoinhibitory mechanism to prevent spurious high FRET and to improve reversibility. Two red circles indicate H83/86D mutations [42] in the autoinhibitory PBD domain which prevents binding to other Rac GTPases when the autoinhibition is relieved through GTPase activation. (b) The biosensor depicted in (a) shown as a cDNA domain diagram

2. Materials

2.1. Cell Culture and Transfection

  1. LinXE cell line, a derivative of HEK293T [24].

  2. RAW/LR5, a monocyte/macrophage cell line [25].

  3. DMEM (4.5 g/mL glucose with 110 mg/L sodium pyruvate) supplemented with 10% fetal calf serum, penicillin (100 IU)/streptomycin (100 μg/mL), and 2 mM Glutamax: Used to cultivate LinXE cells using standard cell culture procedures.

  4. RPMI (with 300 mg/L l-glutamine) supplemented with 10% newborn calf serum, penicillin (100 IU)/streptomycin (100 μg/mL): Used to cultivate RAW/LR5 cells using standard cell culture procedures.

  5. Opti-MEM (Invitrogen).

  6. Plasmids: The biosensor constructs in pTriEX backbone or in other mammalian expression plasmid systems. Rho GTPase biosensors and their mutants are available from the author’s laboratory upon inquiry. Expression constructs for various Rho GTPase upstream regulators including GEF, GAP, and GDI can be obtained from sources such as Addgene.

  7. Lipofectamine 2000 reagent (Invitrogen).

  8. Fugene HD reagent (Roche).

  9. Dulbecco’s phosphate-buffered saline (DPBS), calcium and magnesium free: 0.2 g/L KCl, 0.2 g/L KH2PO4, 8 g/L NaCl, 1.15 g/L Na2HPO4 (anhydrous).

  10. 0.01% (w/v) Poly-l-lysine stock solution: Dilute 1:10 in DPBS for use.

  11. 6-Well tissue culture plates.

2.2. Fluorometry Assay

  1. 6-Well tissue culture plates.

  2. 3.7% Formaldehyde solution: Prepare from 37% formaldehyde stock solution in DPBS.

  3. 0.25% Trypsin.

  4. Spectrofluorometer with quartz cuvette and plate reader capability.

2.3. GST-PBD Pull-Down Assay

  1. PAK1-PBD-agarose slurry (Cytoskeleton, Inc.).

  2. Plasmids: pTriEX-EGFP-Rac1 (Q61L and T17N mutants; can be labeled with any fluorescent protein, i.e., EGFP, mCerulean, mVenus, mCherry).

  3. Cell lysis buffer for pulldown: 50 mM Tris–HCl pH 7.4, 500 mM NaCl, 50 mM MgCl2, 1% (w/v) Triton X-100, protease inhibitor cocktail (Sigma), 1 mM phenylmethanesulfonyl fluoride (PMSF) (add at the very last minute).

  4. 5× Gel loading buffer with dithiothreitol (DTT): 0.25 M Tris–Cl pH 6.8, 10% (w/v) sodium dodecylsulfate (SDS), 0.25% (w/v) bromophenol blue, 0.5 M DTT, 50% glycerol.

  5. 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting supplies/apparatus.

  6. Anti-GFP antibody (Roche, clones 7.1 and 13.1).

  7. Anti-β-actin antibody (Santa Cruz Biotechnology, Clone AC-15).

  8. Ponceau S solution.

2.4. Imaging Experiments for In-Cell Validation

  1. 25 mm Round coverslips #1.5 thickness (Warner Instruments).

  2. 3.7% Formaldehyde solution made from 37% stock formaldehyde solution and DPBS.

  3. BWD buffer for live-cell experiments [20, 25]: 20 mM HEPES, pH 7.4, 125 mM NaCl, 5 mM KCl, 5 mM glucose, 10 mM NaHCO3, 1 mM KH2PO4, 1 mM CaCl2, 1 mM MgCl2, pH 7.4.

  4. Ham’s F-12K medium (phenol red free) (Crystalgen).

  5. FluoroBrite DMEM without phenol red (Invitrogen).

  6. Attofluor chamber system or other compatible live-cell chamber system that can accommodate 25 mm coverslips, or Cell-View glass-bottom culture dishes (Greiner Bio-One).

  7. Fluorescence inverted microscope for imaging FRET in living cells [26].

3. Methods

3.1. Transfection of Biosensor and Regulators

For validation and characterization, it is critical to achieve the best possible transfection efficiency and cell health.

  1. Treat 6-well plates with the diluted poly-l-lysine solution for 5 min at 24 °C and aspirate prior to plating cells (no need to rinse).

  2. Plate LinXE cells on poly-l-lysine-pretreated 6-well dishes at 1.25 × 106 cells per well in normal culture medium, 1 day before transfection (see Note 1). Plate as many wells as needed for different conditions plus one additional well for the negative control. Wells are typically set up in triplicates.

  3. For transfection, follow the manufacturer’s protocols for Lipofectamine 2000 reagent. We use 4 μg total DNA per well together with 10 μL per well of Lipofectamine 2000 reagent, in 100 μL Opti-MEM solutions each for dispersing the DNA and dispersing the Lipofectamine 2000 reagent. The amount of biosensor DNA as well as regulator DNA must be optimized based on titration, ensuring that the total amount of DNA is kept the same per well (using empty pcDNA3.1 vector to make up the deficit as required) (see Fig. 2) (see Note 2). The total volume of medium plus the transfection mix in wells should be 2 mL. We do not change the medium following transfection.

  4. 24 h following transfection cells are ready for pull-down experiments (see Subheading 3.3). 48 h following transfection cells are ready for fluorometric assays (see Subheading 3.2) (see Note 3).

Fig. 2.

Fig. 2

Open in a new tab

Representative, normalized spectra of Cdc42-GDI-binding biosensor [1] showing titration of biosensor DNA during transfection. This biosensor detects binding of GDI to Cdc42. As such, titrating the biosensor in transfection reveals a saturation point against endogenous GDI in LinXE cells (800 ng of biosensor DNA in total 4 μg of DNA transfected per a well of a 6-well plate). Reproduced from Hodgson et al. 2016 Nature Chemical Biology [1]

3.2. Fluorometric Analysis of Rho GTPase Biosensors

Testing biosensor mutants and responses to upstream regulators in vitro in fluorometry assays is absolutely critical to characterize how the sensor responds (see Fig. 3) (see Note 4). If biosensor responses are not carefully and exhaustively tested in this manner, the biosensor cannot be trusted to provide biologically meaningful readouts. We suggest that the following versions of the biosensors should be tested (Rac GTPase mutations are quoted as an example (see Fig. 3), analogous mutations should be used for other family members of GTPases): (1) wild-type version; (2) Q61L and G12V constitutively activated mutants (F28L can also be a good choice as a fast-cycling activated mutant); (3) T17N dominant negative mutant; (4) effector-binding mutants T35S and/or Y40C; and (5) if interested in showing the effect of GEF interaction G15A and/or D118A mutants in addition to the T17N mutant.

Fig. 3.

Fig. 3

Open in a new tab

Typical validation and characterization fluorometry for a GTPase biosensor. In this example, a biosensor for Rac2 GTPase is shown [20]. (a) Representative, normalized spectra of Rac2 biosensor showing emission scans between 450 and 600 nm upon excitation at 433 nm. Active = G12V mutant; active + GDI = G12V mutant plus 2× excess GDI; inactive = T17N mutant. (b) Fluorometric validations of Rac2 biosensor, showing various mutants with or without 2× excess GDI co-expression. Mutants of Rac2 that are able to bind GDI (WT, G12V, T40C) show low FRET whereas the Q61L mutant which does not bind GDI shows high FRET. (c) The wild type and mutants of Rac2 biosensor with or without H83/86D mutations (GTPase-binding deficient) [42] in PBD1 (2× PBD H/D = both PBD1 and PBD2 are GTPase-binding deficient), with or without 2× excess GDI. (d) The wild-type version of Rac2 biosensor co-expressed with Rac2-targeting (in green) or non-Rac2-targeting GEFs, with or (e) without 2× excess GDI. (f) The wild-type Rac2 biosensor co-expressed with 4× excess targeting (in red) or non-targeting GAPs. Fluorometry data: Average ±SEM of 3 independent experiments performed in triplicates. *p < 0.00001 versus WT alone; #p < 0.00001 versus G12V alone. Originally published in The Journal of Immunology. Veronika Miskolci, Bin Wu, Yasmin Moshfegh, Dianne Cox, and Louis Hodgson. 2016. Optical tools to study the isoform-specific roles of small GTPases in immune cells. J. Immunol. Vol: 196(8): 3479–93. Copyright © 2016 The American Association of Immunologists, Inc

These versions of the biosensor should be expressed together with exogenous, excess GDI (2–4-fold excess by DNA quantity) to determine if GDI binds and reduces FRET as expected in some conditions (i.e., WT, G12V, effector-binding mutants) but not in conditions that are known not to bind GDI (i.e., Q61L, T17N for Rac1/2, R66E). The co-expression of GEF-DHPH domains (active, catalytic fragments) that are either targeting or non-targeting should also be shown. This experiment should be done with or without exogenous GDI to show (1) rescue above and beyond GDI-mediated repression of activity, and (2) in the absence of excess GDI the recovery of FRET up to similar levels as the constitutively activated mutant versions of the biosensor. Similarly, targeting and non-targeting GAPs should be co-expressed to show the attenuation of FRET in response to GAPs. See Note 4 for further details. In these experiments include a negative control transfected with empty pcDNA3.1 vector or equivalent for background autofluorescence correction.

3.2.1. Adherent Cell Analysis

  1. Directly fix cells in 6-well dishes using 3.7% formaldehyde in DPBS, 2 mL per well. Leave at room temperature in a dark location for 20 min (see Note 5).

  2. Gently rinse cells three times with DPBS, add 1 mL of DPBS, and immediately scan the plate in a fluorometer (see Note 6).

  3. The fluorometry conditions are set up as follows: excitation wavelength = 433 nm; emission wavelengths scanned between 450 and 600 nm at 3 nm increments; and 1-s integration at the photomultiplier tube (PMT) per increment, with appropriate correction factors for the signal and the reference channels that are provided by the manufacturer. For adhered cell scanning using plate reader, the entry, intermediate, and exit slits are all set to 5 nm (see Note 7).

3.2.2. Detached Cell Analysis

  1. Rinse cells once gently with DPBS, suction out, and add 1 mL of 0.25% trypsin per well. Ensure that the trypsin wets the entire surface of the well, then suction out excess trypsin, and incubate at room temperature for 1–2 min.

  2. Using 500 μL ofice-cold DPBS, detach and resuspend cells by using a vigorous pipetting action, collect into 1.5 mL microcentrifuge tubes, and place on ice. Transfer the detached cell suspension to a 500 μL quartz fluorescence cuvette and immediately perform an emission scan (see Note 8).

  3. The basic setting for the detached measurement conditions is identical to adherent cell condition (see Subheading 3.2.1), the only difference being that slit widths of 1–2 nm are used instead of 5 nm to cut down on the incident light intensity and attenuate brightness (see Note 9).

3.2.3. Analysis of the Results

  1. Export the data in such a way to obtain the signal/reference output (S/R). A typical reference is the photodiode output but in older fluorometer models signal from a concentrated rhodamine solution may be used as well.

  2. Subtract the background autofluorescence S/R obtained from the “mock”-transfected negative control (see Note 10) at each emission wavelength from the raw data.

  3. Normalize the emission spectrum against the intensity at the peak of FRET donor (CFP emission at 470 nm) fluorescence emission. The FRET peak is at 525 nm and an example scan is shown in Fig. 3a (see Note 11).

  4. Compare the resulting FRET/donor peaks at 525 nm between various conditions to see if the responses are as expected (see Fig. 3). Conditions where biosensor is active or being activated by upstream regulators should result in high FRET. Dominant negative mutant should yield low FRET (see Note 12), as should effector and GTPase-binding mutations or GAP/GDI co-expressions. Co-expression of activating GEF fragment in the absence of exogenous GDI should result in similar activation levels as the constitutively activated mutant biosensor expression (see Note 13).

  5. Perform Western blot of the cell lysates from the conditions used in fluorometric analysis to confirm expression of full-length biosensor mutants. Confirm the expression of tagged GEF and GAP domains used in the fluorometry analysis using Western blot, additionally probing for β-actin as a loading control.

3.3. Biosensor Effector Competition

This analysis characterizes the extent of biosensor binding to endogenous cellular effector targets, which may result in overexpression/signaling artifacts (see Note 14). The proximity of the components within single-chain biosensors results in an apparent high local concentration of each domain of the biosensor; thus the chance of the biosensor interacting with endogenous cellular binding targets, as opposed to the built-in affinity domain, is low. However it is important to verify that this is indeed the case for each new sensor. Here we assess the binding characteristics of the biosensor by providing exogenous binding domain in excess in a competitive binding assay.

  1. Transfect the following conditions into LinXE cells as in Subheading 3.1: (a) untransfected; (b) fluorescent protein (FP)-labeled constitutively active Rac GTPase; (c) FP-labeled dominant negative Rac GTPase; (d) constitutively active version of the Rac GTPase biosensor; and (e) constitutively active version of the Rac GTPase biosensor that contains the GTPase-binding deficient (H83D-H86D) mutations in the built-in affinity domains (in this case the PAK-binding domain—PBD).

  2. 24 h following transfection, collect cells by brief trypsinization, resuspend in cold DPBS, and then pellet by centrifuging at 300 × g for 3 min. Keep on ice at all times.

  3. Lyse the cells by adding 500 μL of cold cell lysis buffer for pulldown (protease inhibitor cocktail and PMSF added just prior to this step), and pipette vigorously to disperse all cells. Let the cell lysis proceed on ice for 30 min.

  4. Centrifuge at 20,000 × g at 4 °C for 15 min to clear the lysate.

  5. Transfer the supernatant to clean 1.5 mL microcentrifuge tubes. Remove and reserve 50 μL each of the cell lysates as “10% input” fractions. Mix the 10% input fractions with 5× loading buffer with DTT, boil at 95 °C for 5 min, and set aside for Western blotting.

  6. Add 10–15 μL equivalent of PAK-PBD agarose bead slurry to each sample tube and incubate at 4 °C with rotation for 1–2 h (see Note 15).

  7. Wash three times with 1 mL of lysis buffer; the first wash should be done with the lysis buffer containing protease inhibitors (see Note 16).

  8. After the final wash, leave approximately 80 μL of the solution at the bottom of the tube with the slurry, add appropriate amount of the 5× loading buffer with DTT, boil at 95 °C for 5 min, and set aside for Western blotting.

  9. Set up SDS-PAGE gel, and proceed with Western blotting. Example blots are shown in Fig. 4 (see Notes 17 and 18).

Fig. 4.

Fig. 4

Open in a new tab

Western blot of PAKI-PBD-agarose pulldown of fluorescent protein (FP)-tagged, constitutively active and dominant negative GTPase for assay control, and the constitutively active version of a GTPase biosensor with functional or GTPase-binding-deficient mutations in the built-in affinity-binding domain, overexpressed in LinXE cells. Total lysates and β-actin were used as loading controls. Ponceau S staining is also shown to indicate the presence of GST-PBD in the pull-down fractions. Lane 1: untransfected control. Lane2: constitutively active mutant version of the biosensor with GTPase-binding-deficient mutations in the built-in affinity-binding domain. Lane 3: constitutively active mutant version of the biosensor with functional built-in affinity domain. Lane 4: dominant negative mutant version of the GTPase tagged with an FP. Lane 5: constitutively active mutant version of the GTPase tagged with an FP. Pulldown and 10% lysate input: detected using anti-GFP antibody

3.4. Biosensor Characterization in Cells Using Microscopy Imaging

Once the biosensor is characterized and validated in vitro, we normally further characterize it in target cells (fibroblasts, tumor cells, etc.). This involves expression of mutant versions of the sensor and measurement of the ratio of FRET/donor in cells using a microscope, along with live-cell exogenous stimulation experiments to ensure that the sensor responds to physiologically relevant stimuli.

  1. Transfect constitutively active and dominant negative versions of the biosensor into target cells of interest.

  2. 24 h after transfection detach and replate cells onto appropriately prepared 25 mm coverslips (see Note 19).

  3. Time before fixation following replating depends on the cell type and how long they may take to attach. For fixation, treat cells on coverslips with 3.7% formaldehyde in DPBS at room temperature for 20 min, shielded from light.

  4. Rinse three times with DPBS, mount the coverslips, and image them. Use ratiometric analysis [27] to process the data. Example results are shown in Fig. 5.

  5. For live-cell characterizations, mount transfected cells with the wild-type version of the biosensor on coverslips onto live-cell imaging chamber such as the Attoflour chamber (Molecular Probes), or such as that shown previously [26], and stimulate using known stimuli (see Note 20). General considerations of live-cell ratiometric imaging technique are beyond the scope of this chapter, but more details can be found elsewhere [27]. An example of such an exogenous stimulation experiment is shown in Fig. 6 (see Note 21).

Fig. 5.

Fig. 5

Open in a new tab

Ratiometric images of transiently overexpressed constitutively active (G12V) or dominant negative (T17N) Rac2 biosensor in RAW/LR5 cells [20]. Scale bar = 10 μm. Quantification of whole-cell Rac2 activities are also shown, n = at least 15 cells/condition, mean ± SEM, p < 0.00001. Originally published in The Journal of Immunology. Veronika Miskolci, Bin Wu, Yasmin Moshfegh, Dianne Cox, and Louis Hodgson. 2016. Optical tools to study the isoform-specific roles of small GTPases in immune cells. J. Immunol. Vol: 196 (8): 3479–93. Copyright © 2016 The American Association of Immunologists, Inc

Fig. 6.

Fig. 6

Open in a new tab

Live-cell imaging of Rac1 and Rac2 activation dynamics during fMLP response by macrophages [20], shown as a proof-of-principle validation of exogenous stimulation. Time-lapse biosensor activity image (top panel) and DIC (bottom panel) of RAW/LR5 cells with (a) Rac1 and (b) Rac2 biosensor and fMLP-receptor stably expressed, stimulated with 100 nM fMLP; scale bar = 5 μm. (c) Quantification of whole-cell average Rac1 and Rac2 activities, normalized to average whole-cell activity prior to stimulation (baseline levels during the first 4 min before stimulation). Data are the Ave. ±SEM of 11 cells for Rac1 and 10 cells for Rac2; black *p < 0.05, 10–270 s versus 0 s (Rac1); red *p < 0.05, 10–480 s versus 0 s (Rac2), one-tailed, paired Student’s t-test was used. Originally published in The Journal of Immunology. Veronika Miskolci, Bin Wu, Yasmin Moshfegh, Dianne Cox, and Louis Hodgson. 2016. Optical tools to study the isoform-specific roles of small GTPases in immune cells. J. Immunol. Vol: 196(8): 3479–93. Copyright © 2016 The American Association of Immunologists, Inc

Acknowledgments

This work was supported by American Cancer Society Lee National Denim Day Postdoctoral Fellowship PF-15-135-01-CSM (S.-D.), Irma T. Hirschl Career Scientist Award (L.H.), and NIH grants: T32GM007491 (V.M.); R01 GM071828 and P01 CA100324 (D.C.); and CA205262 (L.H.). A.M.G. was supported by the Summer Undergraduate Research Program (SURP) of the Albert Einstein College of Medicine, Graduate Division of Biomedical Sciences. Sara K. Donnelly and Veronika Miskolci contributed equally to this work.

4 Notes

1.

Cells should be monodispersed for best transfection efficiency. LinXE or any other HEK293T derivative detaches easily even if no trypsin is used. However, without trypsin, the cells tend to come off in clumps, which significantly reduces transfection efficiency following replating. Thus, always trypsinize them carefully and sufficiently so that cells are detached and monodispersed.

2.

Within the 4 μg DNA per well specified, one must account for different amounts and mixtures of DNA components to achieve the desired combinations of biosensors and regulators. Typically, a titration experiment of the biosensor DNA concentration should be performed first to determine the minimal concentration of biosensor DNA needed to overcome endogenous cellular GTPase regulators, especially the GDI. In the example given in Fig. 2, a Cdc42 biosensor-GDI interaction [1] is used to illustrate the saturation point of endogenous GDI binding as a function of the biosensor DNA concentration being transfected. This is always an important consideration if the biosensor is not a strong mutant version (i.e., constitutive active or dominant negative) but rather wild-type or other weaker mutants such as effector-binding mutations.

3.

Prior to pull-down experiments or fluorometric analysis, it is good practice to image the biosensors in fluorescence under 20–40 × magnification to observe the expression patterns. This should be used to confirm if the patterns of localizations are appropriate to the particular mutant versions (and/or with upstream regulators) being expressed. The observations are best performed using the yellow fluorescent protein (YFP) excitation/emission settings, which directly excites the FRET acceptor. Direct excitation/emission of FRET acceptor is not affected by presence or absence of FRET; thus, this can be used to track the localization of biosensors in cells.

4.

Additional GTPase mutations that may be useful can be found in Takai et al. (1994) [28]. In addition, versions of the biosensor in which a constitutively activated mutation (e.g., Q61L) is present but with H83/86D (GTPase-binding deficient mutations) in the PAK1 PBD domain [29] should be used to show the binding specificity. The H83/86D mutation within the PBD domain together with an additional Q61L constitutively activating mutation in the GTPase is also important in the pull-down experiment to determine effector competition. For the upstream regulators, exogenous GDI at 2–4-fold excess should be tested. In the regulator co-expression experiments, each GEF/GAP should be titrated to achieve the best response while determining the toxicity limits (GEF/GAP overexpressions are highly toxic). Typical range for GEF/GAP co-expression is 1:0.5–10, biosensor DNA:regulator DNA [20]. In all cases, total DNA quantity should be 4 μg per a 6-well plate, if using the Lipofectamine 2000 protocol. In the case of a deficit, the balance is achieved by using an empty, pcDNA3.1 vector.

5.

For the Rac biosensors, cellular signaling pathways are such that Rac activity is dependent on cell adhesion [30, 31]. This means that by trypsinizing and detaching cells, the signaling pathways are altered and the resulting measurements of biosensor may not be reliable. We have found this to be the case for the wild-type and other weaker mutant versions of the biosensor and when expressing the wild-type sensor together with various upstream mutants. This effect is not as obvious when dealing with the constitutively active and/or dominant negative mutants as their signaling capacity is strong enough that they appear to overwhelm this effect. We have found this not to be a problem with our Rho biosensors [7, 10]. Therefore, depending on what proteins are being tested and what their native signaling pathways require, one must decide on how to measure the biosensor readout (detached or attached). The contra argument for the adhered cell measurement is a possible reduction in the signal-to-noise ratio of measurements in some cases, as compared to direct measurement of concentrated cell suspensions in quartz cuvettes, which produce the maximum signal-to-noise ratio.

6.

Most plate readers are designed to accommodate 96 wells or greater number of wells per plate. When scanning 6-well plates, we normally set up multiple sampling positions (typically 7–8) within a well of a 6-well plate and take an average of these measurements.

7.

Our fluorometer is configured with a set of double-grating-based monochromators for the excitation and emission paths; thus, an intermediate slit is needed. If a single grating system is used, entry and exit slits should be set at 5 nm.

8.

The detached cell suspensions, as a function of time on ice, will begin to aggregate and form stringy cell clusters. This will differentially affect light scattering during fluorescence emission scans. It is therefore important to pipette the samples vigorously to resuspend and break up any clusters and aggregates quickly just prior to placing the sample into the cuvette for measurement. Thus, processing one 6-well plate at a time is highly recommended.

9.

For detached cell measurements, an appropriate quartz cuvette should be used (cuvette should be able to hold 500 μL). The slit widths of the fluorometer should be adjusted so that one does not saturate the PMT and stays within the linear range of signal measurement. Consult the manufacturer’s instructions for the specific fluorometer to determine the linear limit of detection of the particular PMT.

10.

In both adherent and detached cell conditions, measure one well with “mock”-transfected condition, containing no fluorescence but otherwise transfected with empty pcDNA3.1 vector or equivalent. This will produce the emission spectra of cellular autofluorescence which will be subtracted from the raw data.

11.

Optimizations: Good transfection efficiency is critical. 24 h post-transfection, we find that individual cellular expression levels of fluorescence may be brighter but not as many cells are expressing fluorescence compared to 48 h post-transfection. Ultimately in our hands, 48 h post-transfection produces the best result. Different GEFs and GAPs could show differential strength of effects as well. Titration experiments should always be performed to find the best condition to achieve the maximum extent of response. To test biosensors other than Rho GTPases, this technique described here can be readily extended, by co-expressing various upstream activators and inhibitors and then measuring the appropriate FRET response. There has been a debate as to whether or not to normalize the fluorescence emission spectra to the donor peak emission. We do this because it makes the comparison of changes in FRET very intuitive. In addition, ratiometric FRET imaging in cells is such that FRET emission channel is normalized against the donor emission channel to calculate ratio, which is analogous to how we present this data in fluorometry. However if one wishes to see how the donor peak drops and the FRET peak rises and vice versa (as per the definition of a FRET response), then one must calibrate against the total expression levels of the biosensor by measuring the acceptor direct excitation/emission, and scale the donor excitation/emission scans according to it in order to overlay the concentration-corrected spectra.

12.

The dominant negative mutant (T17N mutant version in the case of Rac/Cdc42 GTPases) is a very important condition to test in order to show that the biosensor is appropriately designed so that the binding domain does not spuriously bind to an otherwise inactive GTPase and produce high FRET. During design and validation of biosensors, the dominant negative condition could produce high FRET for the following reasons: (1) the difference in native affinity of the binding domain for the on versus off states of the GTPase is not high enough (not selective enough) and thus the proximity of the domains within a single-chain construct results in inappropriate interactions due to apparent high local concentration, yielding high FRET, or (2) cellular GEFs bind to the dominant negative mutant [32–34] and sterically impinge on the FRET pair of fluorescent proteins resulting in high FRET. To address the first issue, we have successfully incorporated an autoinhibitory regulation into the binding domain that reduces the binding affinity when GTPase is in the off state, but still allows efficient binding when the GTPase is activated [3]. This approach substantially reduced FRET in the dominant negative mutant, ensuring correct interaction of the domain to the active but not inactive GTPase. Additional approaches could include the use of point mutations to slightly reduce the binding affinity of the domain for the GTPase [4]. To determine if GEF binding is the cause of increased FRET in the T17N dominant negative mutant, other GTPase mutants that are known to cause strong binding to GEFs [35, 36], e.g., G15A and D118A, can be assessed for their impact on FRET. If they result in similarly high FRET levels as the T17N mutant, this issue can be typically corrected by modifying the linker structures within the biosensor (details of the structure design are beyond the scope of this discussion). In some GTPase sensor designs, only GTPase-binding defective mutants and not dominant negative mutants were used for the validation of the biosensor. This is inadequate and could produce biosensors with moderate-to-low off rates, which would result in a bias toward activation and high FRET when used in cells. Thus, a proper functional validation of biosensor must include the dominant negative mutant version.

13.

Biosensors containing non-monomeric versions of fluorescent proteins can result in spuriously high FRET upon overexpression of the constitutively active version of the sensor. This is due to the tendency of these fluorescent proteins to dimerize and/or oligomerize coupled with the confinement of the active versions of the sensor to the plasma membrane, which reduces the degrees of freedom of its components compared to overexpression in cytoplasmic/non-membrane-targeted locations. This situation results in an aggregation-driven FRET, both inter- and intramolecularly. The extent of this effect can be determined by performing a careful titration of expression levels and measuring the FRET/donor response as a function of the concentration of the expressed biosensor, similar to Fig. 2, to best characterize the concentration cutoff to eliminate intramolecular FRET effects due to aggregation. Importantly, the monomeric versions of the biosensor should be used to sidestep this problem in order to obtain the most reliable readouts in live-cell imaging experiments where the membrane loading-associated aggregation could skew the data toward high activation.

14.

If the GTPase built into the biosensor binds to endogenous effector targets and can signal downstream when activated (as opposed to only binding to the built-in binding domain within the biosensor), this would result in overexpression artifacts. This is suboptimal because it would directly amplify the signaling pathway of interest, thus making the biological interpretation of results difficult. Our biosensors do not bind to exogenously supplied excess binding domain when activated; thus, the overexpression artifacts are minimized. However, this does not mitigate the dominant negative effect stemming from overexpression of these biosensors. This is due to the fact that our sensors respond to all upstream regulators with minimal signaling to downstream targets so they could diminish the upstream signaling fidelity by acting as “signaling sinks” when overexpressed. Thus, we try to minimize this issue by achieving low expression levels of the biosensor in target cells that is just above the signal-to-noise ratio sufficient for data acquisition. This typically translates to a signal-to-noise ratio of 2–3:1 comparing the dimmest part of the cell in the foreground to the background fluorescence levels on the microscope [37, 38]. In addition, control analysis should be performed to determine if expression of biosensors in cells of interest impacts the major phenotype that one wishes to study. We perform this type of analysis routinely by quantifying the rate of protrusion and other cell motility parameters. There are various methods previously described to achieve these imaging conditions and expression levels of the biosensor, including the stable expression of biosensors using viral transduction and inducible expression strategies [27, 37, 39, 40].

15.

Increasing the amount of PAK1-PBD-agarose slurry added could push toward more competitive binding. This parameter should be optimized based on how the control lanes in the resulting blot look. We normally use 10–20 μL slurry volume per tube under these conditions.

16.

For centrifugation of beads, we use 30 s at 500 × g at room temperature in a bench top centrifuge. When removing the unbound lysate and/or the lysis buffer during wash steps, do not vacuum suction or otherwise try to remove every last bit of the solution. Best approach is to use a P1000 pipet and manually remove the solution so that approximately 80–100 μL remains at the bottom of the microcentrifuge tube every time. This will ensure that the slurry is not disturbed.

17.

The Roche anti-GFP antibody will detect all GFP variants including mCerulean and mVenus; thus we normally use this antibody to detect for both the controls and the biosensor bands. In addition, GTPase-specific antibody can also be used but these could sometimes give background in the 10% lysate input blot because of nonspecific interactions that are common in direct GTPase detection. Ponceau S solution can be used to detect the PAK1-PBD in the bound fractions in order to control for equal loading. Incubate the freshly transferred membrane with Ponceau S for 5–10 min, and then wash with distilled water. Alternatively, an anti-GST antibody could be used.

18.

Expected results from the pull-down experiment: The pull-down experiment is designed to show that other cellular GTPase effectors would not compete against binding to activated GTPase within the biosensor. As such, a mutant version of the biosensor containing a constitutively activated GTPase and a functional binding domain should not be pulled down by supplying an excess, exogenous binding domain (in the case of Rac subfamily GTPases, PAK1-PBD is used) [3, 20]. In the case of Rho-subfamily GTPases, Rhotekin-RBD is used [7, 10]). This is because the active GTPase mutant within the biosensor would preferentially bind the integral binding domain due to the high local proximity and availability. When appropriate mutations are introduced into the integral binding domain that precludes it from binding to the constitutively activated GTPase within the biosensor, exogenously supplied excess binding domain should now be able to compete for the binding to the constitutively activated GTPase within the biosensor, resulting in this biosensor mutant being pulled down. The fluorescently tagged constitutively active or the dominant negative mutants of GTPase are used as controls for the pulldown by exogenous binding domain: the constitutively active GTPase should be pulled down, but the dominant negative GTPase should not be pulled down. The 10% input lysate blot is used to show the expression of mutants of these biosensors and the control GTPases in cells.

19.

For mouse embryonic fibroblasts (MEF/3T3), we routinely use fibronectin-coated coverslips. These are produced by treating the coverslips in 10 μg/mL fibronectin in DPBS at room temperature for 1 h [10]. For breast tumor cells, we use gelatin-matrix-coated coverslips which are prepared as described previously or fibronectin coating depending on the assay [41]. For macrophage cell line RAW/LR5, we directly plate them onto cleaned coverslips. Transfection procedure for RAW/LR5 cells using the Fugene HD reagent was previously described in detail [39].

20.

For imaging, we routinely use BWD buffer supplemented with appropriate amount of serum or bovine serum albumin. Ham’s F-12K medium without phenol red or FluoroBrite DMEM without phenol red is also an excellent choice for live-cell imaging.

21.

Cells can be starved following established protocols for the specific cell line of interest and stimulated using any number of methods, including serum stimulation, use of chemokine/cytokine, and growth factors such as PDGF and EGF. This approach is important to show that the biosensor properly responds to cellular signaling pathways initiated by well-known ligands under established protocols. It is ideal if biochemical pull-down results are available that match the observed kinetic of stimulated activation of the GTPase in question.

References

  • 1.Hodgson L, Spiering D, Sabouri-Ghomi M, Dagliyan O, DerMardirossian C, Danuser G, Hahn KM (2016) FRET binding antenna reports spatiotemporal dynamics of GDI-Cdc42 GTPase interactions. Nat Chem Biol 12:802–809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ioannou MS, Bell ES, Girard M, Chaineau M, Hamlin JN, Daubaras M, Monast A, Park M, Hodgson L, McPherson PS (2015) DENND2B activates Rab13 at the leading edge of migrating cells and promotes meta-static behavior. J Cell Biol 208:629–648 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Moshfegh Y, Bravo-Cordero JJ, Miskolci V, Condeelis J, Hodgson L (2014) A Trio-Rac1-Pak1 signalling axis drives invadopodia disassembly. Nat Cell Biol 16:574–586 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Hanna S, Miskolci V, Cox D, Hodgson L (2014) A new genetically encoded single-chain biosensor for Cdc42 based on FRET, useful for live-cell imaging. PLoS One 9: e96469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Donnelly SK, Bravo-Cordero JJ, Hodgson L (2014) Rho GTPase isoforms in cell motility: don’t FRET, we have FRET. Cell Adhes Migr 8:526–534 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bravo-Cordero JJ, Hodgson L, Condeelis JS (2014) Spatial regulation of tumor cell protrusions by RhoC. Cell Adhes Migr 8:263–267 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Zawistowski J, Sabouri-Ghomi M, Danuser G, Hahn K, Hodgson L (2013) A RhoC biosensor reveals differences in the activation kinetics of RhoA and RhoC in migrating cells. PLoS One 8:e79877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Bravo-Cordero JJ, Oser M, Chen X, Eddy R, Hodgson L, Condeelis J (2011) A novel spatiotemporal RhoC activation pathway locally regulates cofilin activity at invadopodia. Curr Biol 21:635–644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Machacek M, Hodgson L, Welch C, Elliott H, Pertz O, Nalbant P, Abell A, Johnson GL, Hahn KM, Danuser G (2009) Coordination of Rho GTPase activities during cell protrusion. Nature 461:99–103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Pertz O, Hodgson L, Klemke RL, Hahn KM (2006) Spatiotemporal dynamics of RhoA activity in migrating cells. Nature 440:1069–1072 [DOI] [PubMed] [Google Scholar]
  • 11.Rosenberg BJ, Gil-Henn H, Mader CC, Halo T, Yin T, Condeelis J, Machida K, Wu YI, Koleske AJ (2017) Phosphorylated cortactin recruits Vav2 guanine nucleotide exchange factor to activate Rac3 and promote invadopodial function in invasive breast cancer cells. Mol Biol Cell 28:1347–1360 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Reinhard NR, van Helden SF, Anthony EC, Yin T, Wu YI, Goedhart J, Gadella TW, Hordijk PL (2016) Spatiotemporal analysis of RhoA/B/C activation in primary human endothelial cells. Sci Rep 6:25502. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kedziora KM, Leyton-Puig D, Argenzio E, Boumeester AJ, van Butselaar B, Yin T, Wu YI, van Leeuwen FN, Innocenti M, Jalink K, Moolenaar WH (2016) Rapid remodeling of invadosomes by Gi-coupled receptors: dissecting the role of rho GTPases. J Biol Chem 291:4323–4333 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.van Unen J, Reinhard NR, Yin T, Wu YI, Postma M, Gadella TW, Goedhart J (2015) Plasma membrane restricted RhoGEF activity is sufficient for RhoA-mediated actin polymerization. Sci Rep 5:14693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Timmerman I, Heemskerk N, Kroon J, Schaefer A, van Rijssel J, Hoogenboezem M, van Unen J, Goedhart J, Gadella TW Jr, Yin T, Wu Y, Huveneers S, van Buul JD (2015) A local VE-cadherin and Trio-based signaling complex stabilizes endothelial junctions through Rac1. J Cell Sci 128:3514. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Martin K, Reimann A, Fritz RD, Ryu H, Jeon NL, Pertz O (2016) Spatio-temporal co-ordination of RhoA, Rac1 and Cdc42 activation during prototypical edge protrusion and retraction dynamics. Sci Rep 6:21901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Fritz RD, Menshykau D, Martin K, Reimann A, Pontelli V, Pertz O (2015) SrGAP2-dependent integration of membrane geometry and slit-Robo-repulsive cues regulates fibroblast contact inhibition of locomotion. Dev Cell 35:78–92 [DOI] [PubMed] [Google Scholar]
  • 18.Fritz RD, Letzelter M, Reimann A, Martin K, Fusco L, Ritsma L, Ponsioen B, Fluri E, Schulte-Merker S, van Rheenen J, Pertz O (2013) A versatile toolkit to produce sensitive FRET biosensors to visualize signaling in time and space. Sci Signal 6:rs12. [DOI] [PubMed] [Google Scholar]
  • 19.Pertz O (2010) Spatio-temporal Rho GTPase signaling—where are we now? J Cell Sci 123:1841–1850 [DOI] [PubMed] [Google Scholar]
  • 20.Miskolci V, Wu B, Moshfegh Y, Cox D, Hodgson L (2016) Optical tools to study the isoform-specific roles of small GTPases in immune cells. J Immunol 196:3479–3493 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Rizzo MA, Springer G, Segawa K, Zipfel WR, Piston DW (2006) Optimization of pairings and detection conditions for measurement of FRET between cyan and yellow fluorescent proteins. Microsc Microanal 12:238–254 [DOI] [PubMed] [Google Scholar]
  • 22.Nagai T, Ibata K, Park ES, Kubota M, Mikoshiba K, Miyawaki A (2002) A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat Biotechnol 20:87–90 [DOI] [PubMed] [Google Scholar]
  • 23.Nagai T, Yamada S, Tominaga T, Ichikawa M, Miyawaki A (2004) Expanded dynamic range of fluorescent indicators for Ca(2+) by circularly permuted yellow fluorescent proteins. Proc Natl Acad Sci U S A 101:10554–10559 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Brunet JP, Cotte-Laffitte J, Linxe C, Quero AM, Geniteau-Legendre M, Servin A (2000) Rotavirus infection induces an increase in intra-cellular calcium concentration in human intestinal epithelial cells: role in microvillar actin alteration. J Virol 74:2323–2332 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cox D, Chang P, Zhang Q, Reddy PG, Bokoch GM, Greenberg S (1997) Requirements for both Rac1 and Cdc42 in membrane ruffling and phagocytosis in leukocytes. J Exp Med 186:1487–1494 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Spiering D, Hodgson L (2012) Multiplex imaging of Rho family GTPase activities in living cells. Methods Mol Biol 827:215–234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Spiering D, Bravo-Cordero JJ, Moshfegh Y, Miskolci V, Hodgson L (2013) Quantitative ratiometric imaging of FRET-biosensors in living cells. Methods Cell Biol 114:593–609 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Takai Y, Kaibuchi K, Sasaki T, Tanaka K, Shirataki H, Nakanishi H (1994) Rho small G protein and cytoskeletal control. Princess Takamatsu Symp 24:338–350 [PubMed] [Google Scholar]
  • 29.Frost JA, Khokhlatchev A, Stippec S, White MA, Cobb MH (1998) Differential effects of PAK1-activating mutations reveal activity-dependent and -independent effects on cytoskeletal regulation. J Biol Chem 273:28191–28198 [DOI] [PubMed] [Google Scholar]
  • 30.Del Pozo MA, Kiosses WB, Alderson NB, Meller N, Hahn KM, Schwartz MA (2002) Integrins regulate GTP-Rac localized effector interactions through dissociation of Rho-GDI. Nat Cell Biol 4:232–239 [DOI] [PubMed] [Google Scholar]
  • 31.del Pozo MA, Price LS, Alderson NB, Ren XD, Schwartz MA (2000) Adhesion to the extracellular matrix regulates the coupling of the small GTPase Rac to its effector PAK. EMBO J 19:2008–2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Feig LA (1999) Tools of the trade: use of dominant-inhibitory mutants of Ras-family GTPases. Nat Cell Biol 1:E25–E27 [DOI] [PubMed] [Google Scholar]
  • 33.Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A (1992) The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell 70:401–410 [DOI] [PubMed] [Google Scholar]
  • 34.Hart MJ, Eva A, Zangrilli D, Aaronson SA, Evans T, Cerione RA, Zheng Y (1994) Cellular transformation and guanine nucleotide exchange activity are catalyzed by a common domain on the dbl oncogene product. J Biol Chem 269:62–65 [PubMed] [Google Scholar]
  • 35.Waheed F, Speight P, Dan Q, Garcia-Mata R, Szaszi K (2012) Affinity precipitation of active Rho-GEFs using a GST-tagged mutant Rho protein (GST-RhoA(G17A)) from epithelial cell lysates. J Vis Exp 61:pii 3932 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Wong KW, Mohammadi S, Isberg RR (2006) Disruption of RhoGDI and RhoA regulation by a Rac1 specificity switch mutant. J Biol Chem 281:40379–40388 [DOI] [PubMed] [Google Scholar]
  • 37.Hodgson L, Shen F, Hahn K (2010) Biosensors for characterizing the dynamics of rho family GTPases in living cells. Curr Protoc Cell Biol Chapter 14:Unit 14.111–26 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Hodgson L, Pertz O, Hahn KM (2008) Design and optimization of genetically encoded fluorescent biosensors: GTPase biosensors. Methods Cell Biol 85:63–81 [DOI] [PubMed] [Google Scholar]
  • 39.Miskolci V, Hodgson L, Cox D (2017) Using fluorescence resonance energy transfer-based biosensors to probe Rho GTPase activation during phagocytosis. Methods Mol Biol 1519:125–143 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Wu B, Miskolci V, Sato H, Tutucci E, Kenworthy CA, Donnelly SK, Yoon YJ, Cox D, Singer RH, Hodgson L (2015) Synonymous modification results in high-fidelity gene expression of repetitive protein and nucleotide sequences. Genes Dev 29:876–886 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Bravo-Cordero JJ, Moshfegh Y, Condeelis J, Hodgson L (2013) Live cell imaging of Rho GTPase biosensors in tumor cells. Methods Mol Biol 1046:359–370 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Frost JA, Xu S, Hutchison MR, Marcus S, Cobb MH (1996) Actions of Rho family small G proteins and p21-activated protein kinases on mitogen-activated protein kinase family members. Mol Cell Biol 16:3707–3713 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES