Cell Reports

Resource

The Rac-FRET Mouse RevealsTight Spatiotemporal Control of Rac Activityin Primary Cells and TissuesAnna-Karin E. Johnsson,1,4 Yanfeng Dai,1,4,5 Max Nobis,2 Martin J. Baker,1 Ewan J. McGhee,2 Simon Walker,1

Juliane P. Schwarz,2 Shereen Kadir,2 Jennifer P. Morton,2 Kevin B. Myant,2 David J. Huels,2 Anne Segonds-Pichon,1

Owen J. Sansom,2 Kurt I. Anderson,2 Paul Timpson,2,3,* and Heidi C.E. Welch1,*1Signalling Programme, Babraham Institute, Babraham Research Campus, Cambridge CB22 3AT, UK2Beatson Institute for Cancer Research, Switchback Road, Bearsden, Glasgow G61 1BD, UK3Garvan Institute of Medical Research and Kinghorn Cancer Centre, Cancer Research Program, St. Vincent’s Clinical School,Faculty of Medicine, University of New South Wales, NSW, 2010 Sydney, Australia4These authors contributed equally to this work5Present address: Research Centre for Animal Genetic Resources of the Mongolia Plateau, Inner Mongolia University,235 West University Road, 010021 Hohhot, China

*Correspondence: p.timpson@garvan.org.au (P.T.), heidi.welch@babraham.ac.uk (H.C.E.W.)

http://dx.doi.org/10.1016/j.celrep.2014.02.024

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

SUMMARY

The small G protein family Rac has numerous regula-tors that integrate extracellular signals into tightspatiotemporal maps of its activity to promote spe-cific cell morphologies and responses. Here, wehave generated a mouse strain, Rac-FRET, whichubiquitously expresses the Raichu-Rac biosensor.It enables FRET imaging and quantification ofRac activity in live tissues and primary cells with-out affecting cell properties and responses. Weassessed Rac activity in chemotaxing Rac-FRETneutrophils and found enrichment in leading-edgeprotrusions and unexpected longitudinal shifts andoscillations during protruding and stalling phases ofmigration. We monitored Rac activity in normal ordisease states of intestinal, liver, mammary, pancre-atic, and skin tissue, in response to stimulation or in-hibition and upon genetic manipulation of upstreamregulators, revealing unexpected insights into Racsignaling during disease development. The Rac-FRET strain is a resource that promises to fundamen-tally advance our understanding of Rac-dependentresponses in primary cells and native environments.

INTRODUCTION

The small G protein family Rac is an essential controller of actin

cytoskeletal dynamics and hence cell shape, adhesion, motility,

regulated secretion, and phagocytosis, as well as of gene

expression and reactive oxygen species (ROS) formation (Heas-

man and Ridley, 2008; Wennerberg et al., 2005). Rac is active

(i.e., able to bind downstream effectors) when guanosine

triphosphate (GTP)-bound and inactive when guanosine diphos-

Ce

phate (GDP)-bound. Its activation is catalyzed by at least 20

different DBL- or DOCK-type guanine nucleotide exchange fac-

tors (GEFs) (Rossman et al., 2005) and its inhibition by an equally

large number of Rac-GTPase-activating proteins (GAPs). Rac

downstream signaling specificity and the ensuing Rac-depen-

dent cell responses are largely conferred through the types of

GEFs and GAPs that couple Rac to any given upstream signal

(Rossman et al., 2005).

Forster resonance energy transfer (FRET) technology is widely

used to monitor protein/protein interactions, coupling fluoro-

phore pairs such as cyan fluorescent protein (CFP) and yellow

fluorescent protein (YFP) to two proteins of interest. Inter- and in-

tramolecular FRET probes have been used for a decade to visu-

alize Rac activity (Aoki andMatsuda, 2009; Hodgson et al., 2010;

Itoh et al., 2002; Kraynov et al., 2000). Intermolecular Rac FRET

reporters measure the interaction between separate molecules

that must be expressed to comparable levels and subcellular

distributions (Kraynov et al., 2000), which can be technically diffi-

cult, and they are prone to interfere with endogenous GTPase

signaling (Aoki andMatsuda, 2009; Hodgson et al., 2010). The in-

tramolecular ‘‘Raichu’’ (Ras superfamily and interacting protein

chimeric unit) Rac-FRET probe contains RAC1 as the signal

sensor and Pak-CRIB as the effector, CRIB being the CDC42/

Rac interactive binding motif of Pak, a Rac target that binds to

GTP-bound, but not GDP-bound, Rac. In Raichu-Rac, RAC1-

GTP binding to Pak-CRIB causes FRET from CFP to YFP (Itoh

et al., 2002). The probe is anchored into the plasma membrane

via a KRAS CAAX motif and hence monitors the balance of

endogenous Rac-GEF and Rac-GAP activities at the physiolog-

ically relevant subcellular localization of active RAC1 (Itoh et al.,

2002).

Rac-FRET biosensors have largely been used in transfection-

based experiments in order to correlate the localization of

Rac activity with cellular function. Rac is required for cell motility,

and use of Rac-FRET probes showed that active Rac localizes to

extending cell protrusions during many fundamental processes,

ll Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors 1153

including the leading edge of migrating cells (Itoh et al., 2002;

Kraynov et al., 2000; Machacek et al., 2009; Ouyang et al.,

2008), forming phagosomal cups during phagocytosis of

apoptotic cells (Nakaya et al., 2008), distal poles of daughter

cells during cell division (Yoshizaki et al., 2003), or developing

neurites during neurogenesis (Aoki et al., 2004). Combining

Raichu-Rac expression with downregulation of Vav-family

Rac-GEFs showed that phosphatidylinositol 3-kinase-driven

GEFmembrane targeting localizes Rac activity during neurogen-

esis (Aoki et al., 2005). Expression of an intermolecular Rac-

FRET reporter combined with downregulation of the Rac-GEF

TIAM1 showed that TIAM1 association with distinct scaffolding

proteins directs localized Rac activity depending on extracellular

stimulus (Rajagopal et al., 2010). Similarly, overexpression of a

Raichu-Rac-like probe combined with membrane-targeting of

TIAM1 or the Rac-GAP chimaerin in Madin-Darby canine kidney

(MDCK) cell cysts showed mislocalization of Rac activity to suf-

fice for luminal invasion (Yagi et al., 2012a, 2012b). Finally, use of

Raichu-Rac demonstrated apicobasal Rac activity gradients

at epithelial cell junctions driven by differential regulation of

TIAM1 through b2-syntrophin and Par-3 (Mack et al., 2012).

Raichu-Rac-derived probes are also beginning to be used for

monitoring Rac activity in whole tissues. Reporter expression in

Xenopus and zebrafish embryos showed localized RAC1 activity

in migrating cells during organ development (Kardash et al.,

2010; Matthews et al., 2008; Xu et al., 2012). A limitation of these

studies was that biosensor expression was transient. The

first transgenic Rac-FRET biosensor organism was generated

recently, a fly that conditionally expresses modified Raichu-

Rac in border cells. This revealed Rac activity gradients not

only inside cells, but between cell clusters, being highest in cells

leading in the direction of migration (Wang et al., 2010). First use

of Raichu-Rac-like probes inmammals was recently achieved by

transplantation of biosensor-expressing glioblastoma cells into

rat brain, thus enabling correlation of Rac activity with the

mode of tumor cell migration during invasion (Hirata et al.,

2012). Whereas this study was limited by biosensor expression

in cultured rather than primary cells, it clearly demonstrated

that the mammalian tissue microenvironment controls Rac

activity (Hirata et al., 2012).

There is therefore a need for measuring Rac activity in primary

mammalian cells and tissues for assessing its regulation by

physiologically and functionally relevant organ- or disease-spe-

cific environmental cues. Here, we report the development of a

Rac-FRET mouse strain, which ubiquitously expresses the orig-

inal intramolecular Raichu-Rac reporter to allow spatiotemporal

quantification of Rac activity in living primary mammalian cells

and tissues.

RESULTS

Generation of the Rac-FRET MouseWe generated a Rac activity reporter mouse strain by intro-

ducing the extensively validated intramolecular Raichu-Rac

FRET biosensor (Itoh et al., 2002) into the ROSA26 locus to

confer ubiquitous expression from the endogenous promoter.

To this end, we first generated Rac-FRETfl/fl, a strain in which

Raichu-Rac expression was conditionally prevented by a tran-

1154 Cell Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors

scriptional stop and crossed this with ‘‘deleter,’’ a strain with

X-chromosomalCre recombinase, to excise the stop and enable

constitutive Raichu-Rac expression before breeding out Cre

again (Figure S1). Homozygous knockin mice of the resulting

Rac-FRET strain were born at expected Mendelian ratio,

appeared healthy and fertile, lived normal life-spans, and ex-

hibited no untoward behaviors.

Raichu-Rac Expression Levels in Rac-FRETMouse Cellsand TissuesRaichu-Rac expression was detected in all tissues tested, at

approximately 20–25 ng/mg of tissue, although expression in

muscle tissue (heart, stomach, and skeletal muscle) was some-

what lower, at 2–10 ng/mg (Figure 1A). In primary Rac-FRET neu-

trophils, which express equal amounts of ubiquitous RAC1 and

hematopoietic RAC2, expression of the biosensor was 0.6% of

endogenous RAC1 and RAC2 (Figure 1B), and in primary E13.5

Rac-FRET mouse embryonic fibroblasts (MEFs), it was 8% of

endogenous RAC1 (Figure 1C). Therefore, Raichu-Rac protein

levels are low compared to endogenous Rac, as we had aimed

for, in order to prevent possible dominant-negative effects. As

Raichu-Rac expression did not cause obvious defects and

Rac-FRET mice appeared healthy, we continued to characterize

properties and responses of primary Rac-FRET MEFs and neu-

trophils in detail to determine if it affected cell function.

Primary Rac-FRET MEFs Function Normally and ShowHighly Localized Rac Activity upon PDGF or InsulinStimulationWe investigated proliferation, lifespan, and morphology of pri-

mary Rac-FRET MEFs. They showed normal proliferation rates,

low cell death (2%–4%) in culture, and normal cell morphologies

upon serum starvation and platelet-derived growth factor

(PDGF) stimulation (Figures S2A–S2C). Pak-CRIB pull-down

assays showed that endogenous RAC1 activity was normal in

serum-starved and PDGF-stimulated Rac-FRET MEFs, as was

the activity of the Rac target p38MAPK (Figures S2D and S2E).

Rac-FRET MEFs were assessed by ratiometric FRET imaging,

where increases in YFP/CFP fluorescence ratio induced by FRET

reflect increased Rac activity. Treatment with 50 ng/ml PDGF

stimulated the formation of lamellipodial protrusions and periph-

eral membrane ruffles, which showed significantly higher Rac

activity (FRET ratio of 1.25) than cell edges without protrusions

(FRET ratio of 1.1), and Rac activity remained high throughout

lamellipodia formation and ruffling (Figures 2A and 2B; Movie

S1). Therefore, Rac activity correlated spatially and temporally

with lamellipodia and ruffles, as expected from studies in

fibroblast-like cell lines (Itoh et al., 2002). Stimulation with

100 mg/ml insulin, which caused larger lamellipodia but fewer

dorsal ruffles, gave similar results (Figure 2C; Movie S2).

Rac-FRETNeutrophils FunctionNormally andShowHighRac Activity during Spreading and PolarizationWe and others have previously shown that Rac and its upstream

regulators play crucial roles in ROS formation, adhesion,

spreading, and chemotaxis of neutrophils (Deng et al., 2011;

Gu et al., 2003; Lawson et al., 2011; Roberts et al., 1999; Welch

et al., 2002, 2005), but analysis of Rac activity in these cells has

Figure 1. Raichu-Rac Expression Level in

Tissues and Primary Cells of the Rac-FRET

Mouse Strain

(A) Whole-tissue lysates were prepared from adult

Rac-FRET mice and blotted with anti-GFP anti-

body alongside recombinant CFP standards to

reveal the CFP portion of Raichu-Rac. a.u., arbi-

trary units; WT, wild-type.

(B) Raichu-Rac expression levels in total lysates

of the indicated numbers of diisopropyl fluo-

rophosphate-treated Rac-FRET neutrophils were

determined by western blotting alongside recom-

binant RAC1 and CFP standards using anti-RAC1

AB (green) and anti-GFP AB (red) to compare

biosensor expression with endogenous RAC1.

(C) Raichu-Rac expression levels in total lysates of

the indicated numbers of Rac-FRET MEFs were

determined by western blotting alongside recom-

binant RAC1 and CFP standards with antibodies

as in (B).

been hampered by the facts that they are short-lived and cannot

easily be transfected. Here, we determined if Rac-dependent

neutrophil responses are affected by Raichu-Rac expression.

Rac-FRET neutrophils developed normally in the bone marrow

(not shown), mounted normal ROS responses to the G protein

coupled receptor ligand N-formyl-methionyl-leucyl-phenylala-

nine (fMLP) and to phorbol myristate acetate (PMA), adhered

normally to glass or the integrin ligand poly-Arg-Gly-Asp, spread

normally, and underwent normal basal migration and fMLP-

stimulated chemotaxis in transwell assays, whereas Rac2�/�

cells showed the expected defects (Figures S3A–S3D; Lawson

et al., 2011; Roberts et al., 1999). Furthermore, the basal and

fMLP-stimulated activities of endogenous RAC1 and RAC2 (by

Pak-CRIB pull-down assays) and of the Rac target p38MAPK

were normal (Figures S3E–S3G). Therefore, Rac-FRET neutro-

phil responses were normal by all measures tested.

Rac is required for neutrophil spreading and polarization (Gu

et al., 2003; Roberts et al., 1999). We assessed Rac activity

during these responses in Rac-FRET neutrophils in the presence

or absence of the Rac inhibitor NSC23766. Upon first contact

Cell Reports 6, 1153–1164

with the coverslip (time 0), Rac activity

(normalized FRET ratio) was lower in

NSC23766-treated than control cells,

but increased initially in both upon adhe-

sion, reaching a maximal value of 1.23

within 8 min in control cells compared to

1.04 in NSC23766-treated cells, and

remained high throughout in control

cells whereas decreasing to 0.87 with

NSC23766 (Figures 3A and 3B). Control

cells spread from 90 to 130 mm2 within

8 min and remained spread, whereas

spreading was significantly dampened

by NSC23766 (Figure 3C). Rac activity re-

mained high in control cells, correlating

with their ability to start polarizing twice

as fast as NSC23766-treated cells, to

fully polarize and to maintain polarity (Fig-

ure 3D). Hence, combined use of NSC23766 and ratiometric

FRET imaging showed that Rac activity is required for Rac-

FRET neutrophil spreading and polarization, that it increases

globally throughout the cell during spreading, and that it remains

elevated during polarization.

Rac Activity in Chemotaxing Rac-FRET NeutrophilsChemotaxing primary murine neutrophils are fast-migrating cells

(up to 20 mm/min) that extend transient probing lamellipodial pro-

trusions at their front and sides that typically last for 1 s and

longer-lived protrusions (several seconds), mostly at the leading

edge, that result in translocation. Here, we assayed Rac activity

during chemotaxis of Rac-FRET neutrophils in fMLP gradients

(Figure 4A; Movie S3). Ratiometric FRET imaging at 1 s frame

intervals showed that high Rac activity correlated spatially with

lamellipodial protrusions and low Rac activity with membrane re-

tractions. Temporally, Rac activity was as transient as the protru-

sions (Figures 4A and 4B). We used polar plots to visualize the

localization of Rac activity over time. Comparison with rubber

band plots (which depict cell outlines over time) and cell path

, March 27, 2014 ª2014 The Authors 1155

Figure 2. High Rac Activity in Lamellipodial Protrusions and Membrane Ruffles of PDGF-Stimulated Primary Rac-FRET MEFs

(A) Rac activity in PDGF-stimulated (50 ng/ml) Rac-FRET MEFs determined by ratiometric FRET live imaging. Frames were taken every 10 s (left: generation of

ratiometric image) and analyzed at protruding (top) and nonprotruding sections (bottom) along the cell edge (magnifications of boxes in the FRET image on the

left). Note that, for ratiometric FRET measurements, increase in FRET ratio equals increase in Rac activity.

(B) Quantification of mean Rac activity (FRET ratio ± SEM) in 88 basal and protruding sections along the cell edge of PDGF-stimulated Rac-FRET MEFs as in (A)

from 20 cells and three independent experiments. **p < 0.01 by paired Student’s t test.

(C) Rac activity in insulin-stimulated (100 mg/ml) Rac-FRETMEFs determined by ratiometric FRET imaging. Consecutive frames taken every 10 s, shown from 80 s

after addition of insulin (left: YFP/FRET and CFP images used to generate first ratiometric image). Arrow shows Rac activity at membrane ruffles.

plots showed that Rac activity is highest at the leading edge and

that bursts of high Rac activity typically last for 3–5 s (Figures 4C,

S4A, and S4B). In addition to the leading edge, average polar

plots showed bursts of Rac activity at the periphery and trailing

edge (Figures S4C and S4D). Detailed analysis of the time

dependence of Rac activity and the speed of movement of the

cell edge by Pearson’s correlation showed a direct correlation

(height of peak) between the extent of Rac activity (FRET ratio)

and speed of protrusion formation, without any time lag in the

order of seconds (alignment of peak with lag = 0 s; Figures 4D

and S4E). Therefore, high Rac activity is spatially, temporally,

and inmagnitude tightly correlated with lamellipodial protrusions

at the leading edge of chemotaxing neutrophils.

Oscillations of Rac Activity between the Front and Backof Chemotaxing Rac-FRET NeutrophilsTo investigate the bursts of Rac activity at the front versus back

of chemotaxing neutrophils further, we line-scanned Rac activ-

ity along the central longitudinal axis at 1 s intervals and dis-

played it as a function of time in kymographs (Figure 4E). This

revealed that Rac activity (normalized FRET ratio) is higher at

the front of chemotaxing cells than at the back, as expected,

but furthermore that the bulk of Rac activity shifts by 1.6 mm

toward the posterior of the cell during stalling phases (Figures

4F and S4F). Plotting the location of peak Rac activity on the

longitudinal axis against time revealed unexpected oscillations

1156 Cell Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors

of Rac activity between the front and back of chemotaxing neu-

trophils, and albeit these oscillations being noisy, they tended

toward periodicity. Oscillations of peak Rac activity occurred

in all cells analyzed and ranged from 6 to 12 s (mean ± SEM =

8.7 ± 0.4 s), with R2 values (fit of oscillations to periodicity)

ranging from 0.4 to 0.9 (mean = 0.7; Figure 4G). The oscillating

peak Rac activity was significantly greater than mean whether

localized at the front or the back (Figure S4G). To assess

whether the oscillations of Rac activity are characteristic of

chemotaxis, we compared control and NSC23766-treated

neutrophils. As expected, NSC23766 inhibited chemotaxis (Fig-

ure S4Hi), and both mean and peak Rac activity along the cen-

tral longitudinal axis were lower in NSC23766-treated cells than

in control cells (Figure S4Hii). We categorized NSC23766-

treated cells that retained some form of movement into those

migrating directionally (toward the chemoattractant) and those

which had lost directionality (moving nondirectionally, antidirec-

tionally, or without translocation). Oscillations of Rac activity

were significantly perturbed in both categories compared to

control cells but more so in cells that had lost directionality,

as seen by the significant decreases in R2 value from 0.71 to

0.51 and 0.36, respectively (Figure S4Hiii). These results

confirmed that oscillations of Rac activity along the central lon-

gitudinal axis of the cell are a hallmark of directional migration in

primary mouse neutrophils. Therefore, the use of ratiometric

FRET imaging in primary Rac-FRET neutrophils has allowed

A B

C D

Figure 3. High Rac Activity during Spreading and Polarization of Rac-FRET Neutrophils

(A) Spatiotemporal distribution of Rac activity in Rac-FRET neutrophils during spreading and polarization on glass coverslips was determined by ratiometric FRET

live imaging in the presence or absence of 75 mM NSC23766 (without preincubation), starting from the first point of contact of the cells with the coverslip.

Representative FRET ratio images are shown.

(B) Quantification of Rac activity (average cellular FRET ratio, normalized to control cells) in 69 control and 59 NSC23766-treated cells as in (A) from six inde-

pendent experiments ± SEM p < 0.0001 by ANOVA.

(C) Surface area of the cells in (B).

(D) Polarization of the cells in (B) was analyzed as: time to start polarizing (i), % of fully polarized cells (as defined by their ability to locomote) (ii), and % of cells

reverting to nonpolarized morphology over the 30 min of imaging (iii). *p < 0.05 by unpaired Student’s t test.

us to reveal unexpected insights into the complexities of Rac

signaling during chemotaxis.

Spatiotemporal Regulation of Rac Activity in MouseIntestinal Crypt CulturesTo examine the spatiotemporal control of Rac activity in a multi-

cellular environment, we generated primary three-dimensional

(3D)-intestinal crypt cultures (Myant et al., 2013; Sato et al.,

2009) from the Rac-FRET mouse and analyzed them by fluores-

cence-lifetime imaging microscopy (FLIM)-FRET imaging, where

a decrease in CFP fluorescence lifetime due to FRET reflects

increased Rac activity. We could readily detect Raichu-Rac

expression and image Rac activity in these cultures, which re-

sponded to stimulation with 200 nM PMAwith a time-dependent

increase that peaked after 30min and subsided after 90min (Fig-

ures 5A and 5B). Rac activity was also spatially regulated. Basal

Rac activity was higher in cells at the base of the crypts than in

distal cells, and cells at the base of the crypts also responded

more strongly to PMA treatment than distal cells (Figures 5A

and 5B). This demonstrates reversible and spatial regulation of

Rac activity within multicellular mammalian environments, as

previously observed in Drosophila (Wang et al., 2010).

Ce

Effects of Drug Treatment, Genetic Manipulation, andDisease Development on Rac Activity In VivoWe recently demonstrated that multiphoton microscopy can be

used for ex vivo and in vivo FLIM-FRET imaging of murine tumors

formed by implanted Raichu-RhoA-expressing pancreatic carci-

noma cells (Timpson et al., 2011) and that RHOA activity at the tip

of invading tumors correlates with invasion efficiency (McGhee

et al., 2011; Timpson et al., 2011). Here, we used similar imaging

techniques with the Rac-FRET mouse to examine Rac activity in

native host tissue.

Our recent work suggested that increased Rac activity

following loss of the tumor suppressor adenomatous polyposis

coli (APC) facilitates stem cell hyperproliferation at the base of in-

testinal crypts and colorectal cancer initiation via enhanced ROS

and nuclear factor (NF)-kB production (Myant et al., 2013). To

investigate this here, we first evaluated ex vivo multiphoton

FLIM-FRET imaging of Rac activity in normal intestinal tissue of

the Rac-FRET mouse. Rac activity could be measured within a

range of 0–150 mm from the base of crypts toward the villi and

was stimulated at the crypt base upon treatment with 200 nM

PMA (Figures 5C and 5D; Movie S4). We crossed the Rac-

FRET mouse to Vil-Cre-ERT2 APCfl/fl mice and induced intestinal

ll Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors 1157

A

B C

D E

F G

Figure 4. High Rac Activity Correlates with Protrusion Formation at the Leading Edge and Oscillates between the Front and Back of

Chemotaxing Rac-FRET Neutrophils(A) Ratiometric FRET live imaging of a representative Rac-FRET neutrophil chemotaxing toward 3 mM fMLP in an Ibidi chamber; images taken at 1 s intervals over

20 s (chemoattractant source is due south). For comparison of Rac activity with cell protrusions and retractions, cell perimeters at consecutive time points T (blue)

and T+1 s (red) were plotted using ImageJ plugin QuimP.

(legend continued on next page)

1158 Cell Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors

APC loss by tamoxifen treatment (Myant et al., 2013) to assess

the APC dependence of Rac activity in crypts by in vivo FLIM-

FRET imaging. APC loss led to increased Rac activity at the

base of crypts in Vil-Cre-ERT2 APCfl/fl Rac-FRET mice (Figures

5E and 5F), thus demonstrating the utility of the Rac-FRET

mouse as a tool for examining the regulation of Rac during dis-

ease initiation.

Next, we examined Rac activity in the pancreas in the context

of the stromal extracellular matrix (ECM) at depth. Upon stimula-

tion of Rac-FRET mouse pancreas with 200 nM PMA ex vivo,

increased Rac activity was observed (Figures 6A and 6B; Movie

S5). We crossed the Rac-FRET mouse to the Kras+/G12D

Trp53+/R172HPdx1-Cre (KPC) model of pancreatic ductal adeno-

carcinoma, in which gain-of-function p53R172H drives metastasis

on a KRASG12D background (Morton et al., 2010; Muller et al.,

2009). In vivo FLIM-FRET imaging of pancreatic tumors in KPC

Rac-FRET mice compared to normal pancreas in Rac-FRET

mice revealed that Rac activity was significantly upregulated in

tumors (Figures 6C and 6D). Such deregulated Rac activity

may partially explain the disruption of cell-cell and cell-matrix

adhesion and tumor dissociation, which characterize this inva-

sive and highly metastatic model (Morton et al., 2010; Muller

et al., 2009, 2013).

We also crossed the Rac-FRET mouse to the locally invasive

Polyoma-middle T (PyMT) breast cancer model. Rac activity

could readily be detected in isolated primary mammary tumors

of PyMT Rac-FRET mice and inhibited by NSC23766 treatment

(Figures 7A and 7B; Movie S6). In vivo FLIM-FRET imaging of

exposed tumor tissue of PyMT Rac-FRET mice injected with

NSC23766 showed that Rac was inactivated within 60 min of

NSC23766 administration and began to revert to control level

after 90 min (Figures 7C and 7D). This highlights the utility of

the Rac-FRET mouse for monitoring drug target activity and

clearance rates, which could be applied to guide the scheduling

and dosing of therapeutic intervention.

Finally, we have recently shown that the Rac-GEF PREX1 is

deregulated in melanoma and drives Rac-dependent invasion

(Li et al., 2011; Lindsay et al., 2011). To assess whether the

Rac-FRET mouse is useful for examining Rac activity in the

skin, while at the same time trialling tissue-specific expression,

(B) Enlargement of boxed sections shown in (A). The green-framed box shows

trailing edge.

(C) Polar plot of Rac activity (FRET ratio) around the perimeter of the cell shown i

starting in the center (1 s) and growing eccentrically outward over time.

(D) Pearson’s correlation between Rac activity (FRET ratio) at nodes around the ce

and cell edge velocity. FRET intensities were evaluated against edge speeds be

membrane protrusion/retraction speed. Red line shows average correlation from

3 mM fMLP over 100–120 s with images acquired every 1 s; gray traces show m

(E) Rac-FRET neutrophils chemotaxing toward 3 mM fMLP in an Ibidi chamber w

gitudinal axis, and Rac activity (FRET ratio) over time plotted as kymographs. Pro

achieved by averaging line scans for each segment (as detailed in Supplemental E

4.2 mm/min during stalling phases of migration.

(F) Average Rac activity (FRET ratio) in central longitudinal line scans of 25 cells fro

of the peak Rac activity from the front edge during protruding and stalling phases o

(mean of 25 cells; five experiments; paired t test p = 0.02).

(G) Rac activity oscillates between the front and the back of chemotaxing neu

representative chemotaxing neutrophil was plotted for each 1 s time frame in ord

activity over time, and best fit periodicity curves (purple) were applied to evaluate t

green, that of the back in red. Data shown are from one cell representative of 19

Ce

we crossed Rac-FRETfl/fl to K14-Cre mice to induce selective

Raichu-Rac expression in keratinocytes. Rac activity could

readily be detected by FLIM-FRET imaging of E15.5 K14-Cre

Rac-FRETfl/fl embryonic skin explants and stimulated by PMA

treatment (Figure S5; Movie S7). Combined with PREX1 defi-

ciency, this may be useful for examining PREX1 in melanoma

metastasis. Collectively, the assessment of Rac activity within

these and other organs, such as the liver (Figure S6; Movie

S8), emphasizes the general utility and scope of the Rac-FRET

mouse as a tool for monitoring the intricate mechanisms of regu-

lation of Rac signaling in amyriad of physiological processes and

disease states.

DISCUSSION

Here, we report the development of a Rac-FRET mouse that

ubiquitously expresses Raichu-Rac. We examined Rac activity

in primary MEFs, neutrophils, and intestinal crypts, as well as in

intact tissues (intestine, liver, mammary, pancreas, and skin), in

response to genetic or drug intervention. This revealed unex-

pected insights into Rac signaling during neutrophil chemotaxis,

tissue homeostasis, disease initiation, and disease progression.

We chose the Raichu-Rac reporter over other Rac-FRET bio-

sensors as the probe most likely to provide best signal-to-noise

ratio without affecting cell survival. A similar Raichu-Rac probe

with a RAC1 instead of a KRAS membrane-targeting cassette

gives higher background (Yoshizaki et al., 2003); another probe

that uses CRIB between CFP and YFP, measuring FRET inhibi-

tion upon binding of endogenous Rac (Graham et al., 2001), is

neither as sensitive as Raichu-Rac nor specific for Rac, as it

also binds CDC42-GTP; and finally a Raichu-like probe in which

YFP is replaced with more sensitive YPet (Ouyang et al., 2008)

suggested nonnegligible cytotoxicity in our hands. Recently,

progress in the development of Raichu sensors involving longer

linkers in the reporter molecule has been made, reducing basal

FRET further and thus increasing signal-to-noise (Komatsu

et al., 2011). A next step in reporter mouse development could

be the use of such an improved construct.

The ROSA26 locus was chosen for its ubiquitous expression,

and the endogenous ROSA26 promoter, rather than stronger

a protrusion at the leading edge of the cell, the red box a retraction at the

n (A) over time; perimeters depicted as perfect circles, 1 circle/1 s time frame,

ll perimeter and 0.4 mm from the cell edge (schematic example shown in insert)

tween T�5 s and T+5 s to test for time dependence between Rac activity and

seven independent experiments with a total of 133 cells chemotaxing toward

eans of individual experiments.

ere live imaged at 1 s intervals, line scans performed through the central lon-

filing of steep (protruding) versus flat (stalling) sections of the kymograph was

xperimental Procedures). Average speedwas 15 mm/min during protruding and

m five independent experiments. Gray and black dotted lines show the distance

f migration. The extent of the retrograde shift was 1.6 mmduring stalling phases

trophils. Maximum Rac activity (blue) along the central longitudinal axis of a

er to allow an assessment of the spatial localization of the point of highest Rac

he oscillations. The position of the front of the cell at each time point is traced in

cells analyzed.

ll Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors 1159

Figure 5. Spatiotemporal Distribution of Rac Activity in Intestinal Tissue

(A) PMA stimulation of Rac activity at the base is stronger than in distal cells of Rac-FRET duodenal crypt cultures. Representative fluorescence image of an

intestinal crypt culture (left) with Raichu-Rac (blue) and corresponding FLIM-FRET fluorescence lifetimemaps of Rac activity before (middle) or after (right) 200 nM

PMA treatment. In the FLIM-FRET images, arrows highlight the crypt base and insets show enlarged boxed sections. Note that, for FLIM-FRET measurements,

decrease in fluorescence lifetime equals increase in Rac activity.

(B) Quantification of Rac activity in intestinal crypt cultures as in (A) upon stimulation with 200 nM PMA for 0–90 min. Mean fluorescence lifetime ± SEM of 21–31

cells at varying positions in the base (black) or the distal section of the crypt (gray), as indicated by the schematics, for each time point and location. *p < 0.05,

***p < 0.001, and ****p < 0.0001 by unpaired Student’s t test between indicated time and 0-time control.

(C) Rac activity at the base of Rac-FRET intestinal crypts is stimulated by PMA ex vivo. Representative FLIM-FRET images before (left) and after (right) stimulation

of freshly isolated intestinal crypt tissue with 200 nM PMA for 15 min. For each pair of images, the left-hand panel shows a representative fluorescence image of

intestinal crypts expressing Raichu-Rac (blue), the right-hand one a corresponding fluorescence lifetime map.

(D) Quantification of Rac activity (fluorescence lifetime; mean ± SEM) at the base of 20 intestinal Rac-FRET mouse intestinal crypts before and after PMA

stimulation ex vivo as in (C).

(E) Rac activity at the base of intestinal crypts is increased upon tissue-specific APC loss in vivo. Representative FLIM-FRET images at the base of intestinal crypts

of live Vil-Cre-ERT2 Rac-FRET control mice (left) and Vil-Cre-ERT2 APCfl/fl Rac-FRET mice with intestinal tissue-specific APC deletion (right). For each pair of

images, the left-hand panel shows a representative fluorescence image of intestinal crypts expressing Raichu-Rac (blue) and the second harmonic generation

(SHG) signal from host ECM components (white), the right-hand one a corresponding fluorescence lifetime map.

(F) Quantification of Rac activity (fluorescence lifetime; mean ± SEM) from 206 cells at the base of intestinal crypts in control mice and 197 cells in APC-deleted

mice as in (E); three independent regions each. **p < 0.05 by unpaired Student’s t test.

exogenous promoters, to deliberately aim for low expression

levels to prevent conceivable dominant-negative effects. We

introduced Raichu-Rac with a floxed stop cassette for either

ubiquitous or conditional expression, depending on the choice

of Cre-recombinase strain. The Rac-FRET mouse expresses

Raichu-Rac constitutively and ubiquitously. Our assessment

of MEFs and neutrophils showed that we gauged expression

level well, high enough for detection, but not affecting cell

1160 Cell Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors

properties or responses. However, particularly in neutrophils,

signal strength is limiting, so development of a strain with

stronger inducible expression may be useful. Crossing of Rac-

FRETfl/fl to K14-Cre mice demonstrated that the conditional

Rac-FRETfl/fl strain is useful for tissue-specific expression of

Raichu-Rac. Such targeted expression of the reporter will allow

the assessment of Rac activity in specific cell types within het-

erogeneous tissues. For example, Lgr5-Cre mice (Barker et al.,

A

C

B

D

Figure 6. High Rac Activity in Pancreatic Tumors

(A) Rac activity is stimulated by PMA in Rac-FRETmouse pancreas ex vivo. From the left, panels show a fluorescence image of freshly isolated pancreatic tissue,

an enlargement of the region analyzed, with Raichu-Rac expression in blue and SHG signal from host ECM in white, a corresponding Rac activity (fluorescence

lifetime) map, and images of Rac activity before (top) and after (bottom) treatment with 200 nM PMA for 15 min.

(B) Quantification of pancreatic Rac activity (fluorescence lifetime;mean ±SE) as in (A) from 30 cells in three regions before and after stimulation with 200 nMPMA

for 15 min. *p < 0.05 by unpaired Student’s t test. ns, not significant.

(C) Increased Rac activity in pancreatic tumors in vivo. Representative FLIM-FRET images comparing normal pancreas in live Rac-FRET mice and pancreatic

tumors in live KPC Rac-FRETmice. For each pair of images, the left-hand panel shows a representative fluorescence image of Raichu-Rac (blue) and SHG signal

from host ECM (white), the right-hand one a corresponding FLIM-FRET fluorescence lifetime map.

(D) Quantification of Rac activity in Rac-FRET and KPC Rac-FRET pancreas in vivo as in (C). Mean fluorescence lifetime ± SEM of 222 Rac-FRET and 461 KPC

Rac-FRET cells from three to four regions/mouse, three mice/genotype. ***p < 0.001 by unpaired Student’s t test.

2007) could be used in the future to express Raichu-Rac specif-

ically in stem cells of the small intestine to investigate further the

increased Rac activity we observed upon deletion of the tumor

suppressor APC. Alternatively, the Rac-FRET mouse could be

crossed to a strain expressing a red fluorescent protein specif-

ically in a cell type of interest, for in vivo colabeling and

comparing Rac activity between specific cell types within com-

plex tissues. We also crossed Rac-FRETfl/fl with Cre-ERT2 mice

for drug-inducible expression (Feil et al., 1996). Cre-ER+/T2

Rac-FRET+/fl neutrophils expressed similar levels of Raichu-

Rac upon tamoxifen treatment than Rac-FRET neutrophils and

showed normal cell responses. Thus, inducible Raichu-Rac

expression is another useful feature of the Rac-FRETfl/fl strain.

Crossing the Rac-FRET mouse to various models of disease

has demonstrated its utility for assessing the effects ofmutations

that recapitulate human disease etiology on Rac activity in a

time- and tissue-specific manner, even in organs and tissues

that contain complex mixtures of cell types, without affecting tis-

sue homeostasis. For example, in colorectal cancer, APC loss

often leads to hyperproliferation of intestinal stem cells, and

RAC1 is a critical mediator of this process through its roles in

ROS production and NF-kB signaling (Myant et al., 2013). There-

fore, it is unsurprising that we found the Rac pathway to be active

upon APC loss. We suspect that, in normal intestinal tissue, the

basal and PMA-stimulated Rac activities were also higher in cells

at the base of crypts than in distal cells because the highly pro-

liferative intestinal stem cells are located there. As another

Ce

example for possible future applications, crossing the Rac-

FRET mouse with Tyr-Cre strain for melanoblast-specific

expression (Delmas et al., 2003), with Prex1�/� mice or with mu-

rine models of melanoma, could provide insights into the cell-

type- and stage-specific roles of Rac deregulation during mela-

noma progression (Li et al., 2011; Lindsay et al., 2011).

Ratiometric FRET microscopy was useful for measurement of

Rac activity in isolated primary Rac-FRET neutrophils andMEFs.

Similarly, ratiometric FRET was chosen in Rac-FRET reporter-

expressing zebrafish embryos, because of their optical transpar-

ency (Kardash et al., 2010; Xu et al., 2012). In thick tissue or

organ sections, multiphoton FLIM-FRET imaging allowed us to

monitor tissues at depth and with high resolution, both ex vivo

and in vivo, in the context of the host tissue and environmental

cues. Considering recent advances in imaging and image

analysis, which allow millisecond resolution of spatiotemporal

dynamics (Hinde et al., 2013), it should be possible to adapt

imaging systems to analyze Rac activity alongside Rac-depen-

dent cell responses in any type of tissue.

Use of the Rac-FRET mouse allowed us to observe the exqui-

site spatiotemporal regulation of Rac activity in primary neutro-

phils, which are very different from even their closest model

cell lines, e.g., in their formation of more-transient and probing

lamellipodial protrusions during chemotaxis, which we show

here to be accompanied by equally transient and locally

restricted bursts of Rac activity. During the initial phase of

neutrophil spreading, we observed some increase in Rac

ll Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors 1161

Figure 7. Rac Activity in Mammary Tumors Is Inhibited by NSC23766

(A) Rac activity in PyMT Rac-FRET mouse mammary tumors is inhibited by NSC23766 ex vivo. Representative FLIM-FRET images of freshly isolated mammary

tumors of PyMTRac-FRETmice without (left) or with (right) treatment with 50 mMNSC23766 for 60min ex vivo. For each pair of images, the left-hand panel shows

a representative fluorescence image of tissue expressing Raichu-Rac (blue) and SHG signal from host ECM (white), the right-hand one a corresponding Rac

activity (fluorescence lifetime) map.

(B) Quantification of Rac activity in PyMT mammary tumors as in (A). Mean fluorescence lifetime ± SEM of 144 control cells and 74 NSC23766-treated cells from

two to three different regions/group. ***p < 0.001 by unpaired Student’s t test.

(C) Rac activity inmammary tumors is inhibited by treatment with NSC23766 in vivo. Representative FLIM-FRET images of Rac activity in mammary tumors of live

PyMT Rac-FRET mice before (left) and 60 min after (right) intraperitoneal (i.p.) injection of NSC23766 (4 mg/kg). Order of images as in (A).

(D) Quantification of Rac activity (fluorescence lifetime; mean ± SEM) in 210 cells within mammary tumors of PyMT Rac-FRET mice in vivo 0–90 min after i.p.

injection of NSC23766. **p < 0.01 and ***p < 0.001 by unpaired Student’s t test.

activity, even in NSC23766-treated cells (though lower than in

controls). This could simply be due to incomplete inhibition, but

it is also possible that NSC23766-insensitive Rac-GEFs might

contribute to this phase. Rac-GEFs from the Vav family, for

example, are nonresponsive to NSC23766 (Gao et al., 2004)

and known to be required for neutrophil spreading (Lawson

et al., 2011). This possibility could be investigated in the future

by analysis of Rac-FRET neutrophils with added Rac-GEF

deficiencies. During neutrophil chemotaxis, we observed unex-

pectedbehaviors ofRacactivity in addition to theexpected accu-

mulation at the leading edge, including a retrograde shift during

stalling phases and a yo-yoing between the front and back of

the cell. It is unsurprising that these phenomena have not been

observed previously, as only a combination of FRET technology

and the use of primary cells affords sufficient spatiotemporal res-

olution. Wider and slower waves of Rac activity had previously

been observed in HL60 cells (Weiner et al., 2007), and it seems

possible that the oscillations in primary cells are mechanistically

related. It will be interesting to determine if specific upstream reg-

ulators mediate these oscillations and if these Rac-GEFs are

required for the fast migration mode of neutrophils.

Different types of small G proteins of the Rho family are

activated dependently of each other. Combined use of Rac,

CDC42, or RHOA activity probes and high-resolution imaging

of single-membrane protrusions showed that RhoA is activated

at the tip of forming cell protrusions whereas Rac and CDC42

1162 Cell Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors

activity patterns are wider (Machacek et al., 2009). Furthermore,

transplantation of glioblastoma cells expressing Rac, CDC42, or

RHOA reporters into rat brain suggested that the balance of Rac,

CDC42, and RHOA activities dictates modes of cancer cell inva-

sion (Hirata et al., 2012). Transgenic biosensormice for a number

of different types of small G proteins are currently being gener-

ated (Goto et al., 2013), although not as conditional alleles and

at much higher expression levels. Such GTPase reporter mice

will facilitate future comparisons between different small G pro-

teins tremendously, and different strains will doubtless prove

appropriate for different applications.

In conclusion, the Rac-FRET mouse strain has enabled us to

monitor the intricate and dynamic regulation of the small G pro-

tein Rac, an essential controller of distinct biological responses

depending on timing, location, and signaling context. Future

use of the Rac-FRET mouse as a tool, alone or in combination

with deficiencies in Rac-GEFs, Rac-GAPs, or other upstream

regulators, should fundamentally advance our insight into the

signaling networks that drive Rac-dependent cell responses

and enable us to expand our knowledge of Rac signaling in pri-

mary cells and complex multicellular physiological and disease

states.

EXPERIMENTAL PROCEDURES

Detailed protocols can be found in Supplemental Experimental Procedures.

Generation of the Rac-FRET Reporter Mouse Strain

Modified Raichu-1011X plasmid (Itoh et al., 2002) was introduced into stop-

eGFP-ROSA26TV (Addgene 11739) to generate the Rac-FRET targeting

vector. Upon germline transmission, Rac-FRETfl/fl strain was crossed with

deleter to induce ubiquitous expression of Raichu-Rac. The resulting homozy-

gous Rac-FRETKi/Ki knockin mouse strain was called Rac-FRET for brevity.

Raichu-Rac expression in cells and tissues of the Rac-FRETmousewas deter-

mined by western blotting.

Rac-FRET MEF and Neutrophil Isolation and Responses

Primary E13.5 MEFs were isolated upon timed mating of Rac-FRETKi/fl mice

and cultured for up to 15 days. Proliferation assays were done by cell counting,

cell cycle assays by propidium iodide staining, and morphology assays by

image analysis of tetramethylrhodamine isothiocyanate-phalloidin-stained

cells. Primary neutrophils were freshly isolated from bone marrow for each

experiment. Adhesion and spreading were assayed on glass or integrin-ligand

surfaces, ROS formation by luminol assay, and chemotaxis by transwell

assays using 3 mm pore filters (Lawson et al., 2011; Welch et al., 2005) or Ibidi

chamber assays using Ibidi sticky slide IV0.4. Endogenous Rac activity was

determined in Rac-FRET MEF and neutrophil lysates by Pak-CRIB pull-

down and p38MAPK phosphorylation by western blotting.

Tissues and Mouse Strains for Analysis of Rac Activity in Live

Tissues and Organs

Intestinal crypt cultures (Sato et al., 2009) and skin explants (Mort et al., 2010)

were prepared as described. Vil-Cre-ERT2 APCfl/fl Rac-FRET mice for imaging

of Rac activity in intestinal crypts following APC loss, KPC Rac-FRET mice for

imaging of Rac activity in pancreatic tumors, PyMT Rac-FRET mice for

imaging of Rac activity in mammary tumors, and K14-Cre Rac-FRETfl/fl mice

for tissue-specific expression of Raichu-Rac and imaging of Rac activity in

embryonic skin were generated as detailed in Supplemental Experimental

Procedures.

Imaging

Rac activity was assessed in Rac-FRET MEFs and neutrophils by ratiometric

FRET imaging using an Olympus Cell R imaging system. Pairs of images

were acquired sequentially every 15 s for spreading neutrophils and every

10 s for MEFs. For chemotaxing neutrophils, both channels were acquired

simultaneously at 1 frame/s and cells tracked using ImageJ plugin QuimP11.

Pearson’s correlation was used to analyze the time dependence between

Rac activity and speed of cell edge movement and polar plots (Ferguson

et al., 2007) for the spatiotemporal representation of Rac activity at the cell

perimeter. In addition, line scans of Rac activity were performed along the

central longitudinal axis.

Imaging of Rac activity in various intact tissues and organs was done using

multiphoton FLIM-FRET imaging. Crypt cultures were imaged at depths of

0–100 mm and freshly isolated ex vivo crypts (Myant et al., 2013) at

0–150 mm. For in vivo imaging of intestinal, pancreas, or mammary tissue,

mice were terminally anesthetized. Multiphoton FLIM-FRET imaging was car-

ried out on a Nikon Eclipse TE2000-U inverted microscope with an Olympus

long working distance 20 3 0.95 numerical aperture water immersion lens

using a scan head specifically designed for multiphoton excitation. Data

were analyzed using the time-correlated single-photon counting fluorescence

lifetime analysis functionality of ImSpectorPro (LaVison Biotec).

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures,

six figures, and eight movies and can be found with this article online at

http://dx.doi.org/10.1016/j.celrep.2014.02.024.

ACKNOWLEDGMENTS

We are extremely grateful to Miki Matsuda for allowing us to generate the Rac-

FRET mouse with the Raichu-Rac construct and for constructively discussing

the choice of construct with us at the beginning of the project. We acknowl-

Ce

edge the great skill of the people from the Babraham and Beatson animal

facilities and the passionate care they take in looking after our mice. We also

thank James Conway for critical reading of the manuscript. The project was

funded by BBSRC core funding, BBSRC grant BB/I02154X/1, CRUK core

funding, and NHMRC and ARC funding. Work in the O.J.S. lab was funded

through European Union PRIMES project grant FP7-HEALTH-2011-278568.

Received: November 20, 2013

Revised: February 5, 2014

Accepted: February 15, 2014

Published: March 13, 2014

REFERENCES

Aoki, K., and Matsuda, M. (2009). Visualization of small GTPase activity with

fluorescence resonance energy transfer-based biosensors. Nat. Protoc. 4,

1623–1631.

Aoki, K., Nakamura, T., and Matsuda, M. (2004). Spatio-temporal regulation of

Rac1 and Cdc42 activity during nerve growth factor-induced neurite

outgrowth in PC12 cells. J. Biol. Chem. 279, 713–719.

Aoki, K., Nakamura, T., Fujikawa, K., and Matsuda, M. (2005). Local phospha-

tidylinositol 3,4,5-trisphosphate accumulation recruits Vav2 and Vav3 to

activate Rac1/Cdc42 and initiate neurite outgrowth in nerve growth factor-

stimulated PC12 cells. Mol. Biol. Cell 16, 2207–2217.

Barker, N., van Es, J.H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen,

M., Haegebarth, A., Korving, J., Begthel, H., Peters, P.J., and Clevers, H.

(2007). Identification of stem cells in small intestine and colon by marker

gene Lgr5. Nature 449, 1003–1007.

Delmas, V., Martinozzi, S., Bourgeois, Y., Holzenberger, M., and Larue, L.

(2003). Cre-mediated recombination in the skin melanocyte lineage. Genesis

36, 73–80.

Deng, Q., Yoo, S.K., Cavnar, P.J., Green, J.M., and Huttenlocher, A. (2011).

Dual roles for Rac2 in neutrophil motility and active retention in zebrafish he-

matopoietic tissue. Dev. Cell 21, 735–745.

Feil, R., Brocard, J., Mascrez, B., LeMeur, M., Metzger, D., and Chambon, P.

(1996). Ligand-activated site-specific recombination in mice. Proc. Natl. Acad.

Sci. USA 93, 10887–10890.

Ferguson, G.J., Milne, L., Kulkarni, S., Sasaki, T., Walker, S., Andrews, S.,

Crabbe, T., Finan, P., Jones, G., Jackson, S., et al. (2007). PI(3)Kgamma has

an important context-dependent role in neutrophil chemokinesis. Nat. Cell

Biol. 9, 86–91.

Gao, Y., Dickerson, J.B., Guo, F., Zheng, J., and Zheng, Y. (2004). Rational

design and characterization of a RacGTPase-specific small molecule inhibitor.

Proc. Natl. Acad. Sci. USA 101, 7618–7623.

Goto, A., Sumiyama, K., Kamioka, Y., Nakasyo, E., Ito, K., Iwasaki, M., Eno-

moto, H., and Matsuda, M. (2013). GDNF and endothelin 3 regulate migration

of enteric neural crest-derived cells via protein kinase A and Rac1. J. Neurosci.

33, 4901–4912.

Graham, D.L., Lowe, P.N., and Chalk, P.A. (2001). A method to measure the

interaction of Rac/Cdc42 with their binding partners using fluorescence reso-

nance energy transfer between mutants of green fluorescent protein. Anal.

Biochem. 296, 208–217.

Gu, Y., Filippi, M.D., Cancelas, J.A., Siefring, J.E., Williams, E.P., Jasti, A.C.,

Harris, C.E., Lee, A.W., Prabhakar, R., Atkinson, S.J., et al. (2003). Hematopoi-

etic cell regulation by Rac1 and Rac2 guanosine triphosphatases. Science

302, 445–449.

Heasman, S.J., and Ridley, A.J. (2008). Mammalian Rho GTPases: new in-

sights into their functions from in vivo studies. Nat. Rev. Mol. Cell Biol. 9,

690–701.

Hinde, E., Digman, M.A., Hahn, K.M., and Gratton, E. (2013). Millisecond

spatiotemporal dynamics of FRET biosensors by the pair correlation function

and the phasor approach to FLIM. Proc. Natl. Acad. Sci. USA 110, 135–140.

Hirata, E., Yukinaga, H., Kamioka, Y., Arakawa, Y., Miyamoto, S., Okada, T.,

Sahai, E., and Matsuda, M. (2012). In vivo fluorescence resonance energy

ll Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors 1163

transfer imaging reveals differential activation of Rho-family GTPases in glio-

blastoma cell invasion. J. Cell Sci. 125, 858–868.

Hodgson, L., Shen, F., and Hahn, K. (2010). Biosensors for characterizing the

dynamics of rho family GTPases in living cells. In Current Protocols in Cell

Biology, Chapter 14, J.S. Bonifacino, M. Dasso, J.B. Harford, J. Lippincott-

Schwartz, and K.M. Yamada, eds. (New York: John Wiley & Sons), pp. 11–26.

Itoh, R.E., Kurokawa, K., Ohba, Y., Yoshizaki, H., Mochizuki, N., and Matsuda,

M. (2002). Activation of rac and cdc42 video imaged by fluorescent resonance

energy transfer-based single-molecule probes in the membrane of living cells.

Mol. Cell. Biol. 22, 6582–6591.

Kardash, E., Reichman-Fried, M., Maıtre, J.L., Boldajipour, B., Papusheva, E.,

Messerschmidt, E.M., Heisenberg, C.P., and Raz, E. (2010). A role for Rho

GTPases and cell-cell adhesion in single-cell motility in vivo. Nat. Cell Biol.

12, 47–53, 1–11.

Komatsu, N., Aoki, K., Yamada, M., Yukinaga, H., Fujita, Y., Kamioka, Y., and

Matsuda, M. (2011). Development of an optimized backbone of FRET biosen-

sors for kinases and GTPases. Mol. Biol. Cell 22, 4647–4656.

Kraynov, V.S., Chamberlain, C., Bokoch, G.M., Schwartz, M.A., Slabaugh, S.,

and Hahn, K.M. (2000). Localized Rac activation dynamics visualized in living

cells. Science 290, 333–337.

Lawson, C.D., Donald, S., Anderson, K.E., Patton, D.T., and Welch, H.C.

(2011). P-Rex1 and Vav1 cooperate in the regulation of formyl-methionyl-

leucyl-phenylalanine-dependent neutrophil responses. J. Immunol. 186,

1467–1476.

Li, A., Ma, Y., Yu, X., Mort, R.L., Lindsay, C.R., Stevenson, D., Strathdee, D.,

Insall, R.H., Chernoff, J., Snapper, S.B., et al. (2011). Rac1 drives melanoblast

organization during mouse development by orchestrating pseudopod- driven

motility and cell-cycle progression. Dev. Cell 21, 722–734.

Lindsay, C.R., Lawn, S., Campbell, A.D., Faller, W.J., Rambow, F., Mort, R.L.,

Timpson, P., Li, A., Cammareri, P., Ridgway, R.A., et al. (2011). P-Rex1 is

required for efficient melanoblast migration and melanoma metastasis. Nat.

Commun. 2, 555.

Machacek, M., Hodgson, L., Welch, C., Elliott, H., Pertz, O., Nalbant, P., Abell,

A., Johnson, G.L., Hahn, K.M., and Danuser, G. (2009). Coordination of Rho

GTPase activities during cell protrusion. Nature 461, 99–103.

Mack, N.A., Porter, A.P., Whalley, H.J., Schwarz, J.P., Jones, R.C., Khaja,

A.S., Bjartell, A., Anderson, K.I., and Malliri, A. (2012). b2-syntrophin and

Par-3 promote an apicobasal Rac activity gradient at cell-cell junctions by

differentially regulating Tiam1 activity. Nat. Cell Biol. 14, 1169–1180.

Matthews, H.K., Marchant, L., Carmona-Fontaine, C., Kuriyama, S., Larraın, J.,

Holt, M.R., Parsons, M., and Mayor, R. (2008). Directional migration of neural

crest cells in vivo is regulated by Syndecan-4/Rac1 and non-canonical Wnt

signaling/RhoA. Development 135, 1771–1780.

McGhee, E.J., Morton, J.P., Von Kriegsheim, A., Schwarz, J.P., Karim, S.A.,

Carragher, N.O., Sansom, O.J., Anderson, K.I., and Timpson, P. (2011).

FLIM-FRET imaging in vivo reveals 3D-environment spatially regulates

RhoGTPase activity during cancer cell invasion. Small GTPases 2, 239–244.

Mort, R.L., Hay, L., and Jackson, I.J. (2010). Ex vivo live imaging of melano-

blast migration in embryonic mouse skin. Pigment Cell Melanoma Res. 23,

299–301.

Morton, J.P., Timpson, P., Karim, S.A., Ridgway, R.A., Athineos, D., Doyle, B.,

Jamieson, N.B., Oien, K.A., Lowy, A.M., Brunton, V.G., et al. (2010). Mutant

p53 drivesmetastasis and overcomes growth arrest/senescence in pancreatic

cancer. Proc. Natl. Acad. Sci. USA 107, 246–251.

Muller, P.A., Caswell, P.T., Doyle, B., Iwanicki, M.P., Tan, E.H., Karim, S., Lu-

kashchuk, N., Gillespie, D.A., Ludwig, R.L., Gosselin, P., et al. (2009). Mutant

p53 drives invasion by promoting integrin recycling. Cell 139, 1327–1341.

Muller, P.A., Trinidad, A.G., Timpson, P., Morton, J.P., Zanivan, S., van den

Berghe, P.V., Nixon, C., Karim, S.A., Caswell, P.T., Noll, J.E., et al. (2013).

1164 Cell Reports 6, 1153–1164, March 27, 2014 ª2014 The Authors

Mutant p53 enhances MET trafficking and signalling to drive cell scattering

and invasion. Oncogene 32, 1252–1265.

Myant, K.B., Cammareri, P., McGhee, E.J., Ridgway, R.A., Huels, D.J., Cor-

dero, J.B., Schwitalla, S., Kalna, G., Ogg, E.L., Athineos, D., et al. (2013).

ROS production and NF-kB activation triggered by RAC1 facilitate WNT-

driven intestinal stem cell proliferation and colorectal cancer initiation. Cell

Stem Cell 12, 761–773.

Nakaya, M., Kitano, M., Matsuda, M., and Nagata, S. (2008). Spatiotemporal

activation of Rac1 for engulfment of apoptotic cells. Proc. Natl. Acad. Sci.

USA 105, 9198–9203.

Ouyang, M., Sun, J., Chien, S., and Wang, Y. (2008). Determination of hierar-

chical relationship of Src and Rac at subcellular locations with FRET biosen-

sors. Proc. Natl. Acad. Sci. USA 105, 14353–14358.

Rajagopal, S., Ji, Y., Xu, K., Li, Y., Wicks, K., Liu, J., Wong, K.W., Herman, I.M.,

Isberg, R.R., and Buchsbaum, R.J. (2010). Scaffold proteins IRSp53 and

spinophilin regulate localized Rac activation by T-lymphocyte invasion and

metastasis protein 1 (TIAM1). J. Biol. Chem. 285, 18060–18071.

Roberts, A.W., Kim, C., Zhen, L., Lowe, J.B., Kapur, R., Petryniak, B., Spaetti,

A., Pollock, J.D., Borneo, J.B., Bradford, G.B., et al. (1999). Deficiency of the

hematopoietic cell-specific Rho family GTPase Rac2 is characterized by ab-

normalities in neutrophil function and host defense. Immunity 10, 183–196.

Rossman, K.L., Der, C.J., and Sondek, J. (2005). GEF means go: turning on

RHO GTPases with guanine nucleotide-exchange factors. Nat. Rev. Mol.

Cell Biol. 6, 167–180.

Sato, T., Vries, R.G., Snippert, H.J., van de Wetering, M., Barker, N., Stange,

D.E., van Es, J.H., Abo, A., Kujala, P., Peters, P.J., and Clevers, H. (2009).

Single Lgr5 stem cells build crypt-villus structures in vitro without a mesen-

chymal niche. Nature 459, 262–265.

Timpson, P., McGhee, E.J., Morton, J.P., von Kriegsheim, A., Schwarz, J.P.,

Karim, S.A., Doyle, B., Quinn, J.A., Carragher, N.O., Edward, M., et al.

(2011). Spatial regulation of RhoA activity during pancreatic cancer cell inva-

sion driven by mutant p53. Cancer Res. 71, 747–757.

Wang, X., He, L., Wu, Y.I., Hahn, K.M., and Montell, D.J. (2010). Light-medi-

ated activation reveals a key role for Rac in collective guidance of cell move-

ment in vivo. Nat. Cell Biol. 12, 591–597.

Weiner, O.D., Marganski, W.A., Wu, L.F., Altschuler, S.J., and Kirschner, M.W.

(2007). An actin-based wave generator organizes cell motility. PLoS Biol. 5,

e221.

Welch, H.C., Coadwell, W.J., Ellson, C.D., Ferguson, G.J., Andrews, S.R.,

Erdjument-Bromage, H., Tempst, P., Hawkins, P.T., and Stephens, L.R.

(2002). P-Rex1, a PtdIns(3,4,5)P3- and Gbetagamma-regulated guanine-

nucleotide exchange factor for Rac. Cell 108, 809–821.

Welch, H.C., Condliffe, A.M., Milne, L.J., Ferguson, G.J., Hill, K., Webb, L.M.,

Okkenhaug, K., Coadwell, W.J., Andrews, S.R., Thelen, M., et al. (2005).

P-Rex1 regulates neutrophil function. Curr. Biol. 15, 1867–1873.

Wennerberg, K., Rossman, K.L., and Der, C.J. (2005). The Ras superfamily at a

glance. J. Cell Sci. 118, 843–846.

Xu, H., Kardash, E., Chen, S., Raz, E., and Lin, F. (2012). Gbg signaling controls

the polarization of zebrafish primordial germ cells by regulating Rac activity.

Development 139, 57–62.

Yagi, S., Matsuda, M., and Kiyokawa, E. (2012a). Chimaerin suppresses Rac1

activation at the apical membrane to maintain the cyst structure. PLoS ONE 7,

e52258.

Yagi, S., Matsuda, M., and Kiyokawa, E. (2012b). Suppression of Rac1 activity

at the apical membrane of MDCK cells is essential for cyst structure mainte-

nance. EMBO Rep. 13, 237–243.

Yoshizaki, H., Ohba, Y., Kurokawa, K., Itoh, R.E., Nakamura, T., Mochizuki, N.,

Nagashima, K., andMatsuda,M. (2003). Activity of Rho-family GTPases during

cell division as visualized with FRET-based probes. J. Cell Biol. 162, 223–232.

Cell Reports, Volume 6

Supplemental Information

The Rac-FRET Mouse Reveals

Tight Spatiotemporal Control of Rac Activity

in Primary Cells and Tissues

Anna-Karin E. Johnsson, Yanfeng Dai, Max Nobis, Martin J. Baker, Ewan J. McGhee,

Simon Walker, Juliane P. Schwarz, Shereen Kadir, Jennifer P. Morton, Kevin B. Myant,

David J. Huels, Anne Segonds-Pichon, Owen J. Sansom, Kurt I. Anderson, Paul

Timpson, and Heidi C.E. Welch

Figure S1, related to Figure 1 Johnsson et al. Generation of the Rac-FRET mouse strain

B

fl/+ +/+

1.5 2.0 2.5 3.0

0.6

0.8 1.0

0.4

floxed allele

C D

6.5

9.4

23

4.3

floxed allele

Wt allele

fl/+ +/+

9.4

23

6.5

floxed allele

Wt allele

fl/+ +/+

E floxed allele

Cre-deleted Rac-FRET Ki allele

0.6 0.8 1.0

0.4

0.6 0.8 1.0

0.4

fl/fl Ki/fl Ki/Ki

A

Supplemental p3

Legend for Figure S1, related to Figure 1. Generation of the Rac-FRET mouse strain.

(A) Schematic of Rac-FRET mouse strain generation. Raichu-Rac, consisting of YFP, the CRIB domain of

Pak, Rac1, CFP, and the K-Ras CAAX box, was modified to introduce a Kozak translation initiation

sequence at its 5’ end and cloned into the STOP-eGFP-ROSA26TV targeting vector. The vector backbone

was removed and the targeting insert consisting of the 5’ recombination arm, a splicing acceptor site

(SA, white diamond), the loxP-flanked (black triangles) Neomycin resistance (Neo, N)/STOP cassette

(STOP, S), Raichu-Rac, and the 3’ recombination arm was electroporated into ES cells. Neomycin-

resistant ES clones were submitted to a PCR screen (B) and positive clones subjected to Southern

blotting (C and D) to confirm the correct insertion into the ROSA26 genomic locus, using 5’ and 3’

Southern probes outside of the targeted region and making use of the fact the targeting introduced new

EcoR1 and EcoR5 sites. The position of the 5’ and 3’ probes (white boxes) and sizes of restriction

fragments are to scale. Once germline transmission had been achieved, the resulting Rac-FRETfl/fl strain

was crossed with ‘deleter’, a Cre recombinase expressing strain for ubiquitous deletion of loxP-flanked

sequences. After excision of the Neomycin resistance/STOP cassette, which was monitored by PCR (E),

the Cre recombinase allele was crossed out again through selective breeding, resulting in the Rac-Fret

mouse strain.

(B) PCR screen of ES cells. Neomycin-resistant ES clones were submitted to a PCR screen for the

presence of the floxed allele using a forward primer upstream of the 5’ recombination arm and a reverse

primer within the splice acceptor site. Clones that yielded the expected 1356 bp PCR product were

further subjected to Southern blotting.

(C) Southern blot of EcoR1-digested Wt and targeted ES cell genomic DNA with 5’ probe.

(D) Southern blot of EcoR5-digested Wt and targeted ES cell genomic DNA with 3’ probe.

Supplemental p4

(E) PCR screen of mouse biopsies for the presence of floxed and cre-deleted Rac-FRET Ki alleles. Primers

are listed in Supplemental Experimental Procedures.

Figure S2, related to Figure 2 Johnsson et al. Normal function and morphology of Rac-FRET MEFs

A B

C

D E

24 48 72 96 Hours in culture

0

5

10

15

20 M

EFs

/ ml (

x 10

5 )

ME

Fs (%

)

Rac-FRETfl/fl

Rac-FRET

Rac-FRETfl/fl Rac-FRET

- + - + PDGF 0

20

40

60

80

100

ME

Fs /

cate

gory

(%)

a

b

c d

GTP- Rac1

Rac1

Rac-FRETfl/fl Rac-FRET

- + - + PDGF

Rac

-GTP

(% o

f tot

al)

0

0.2

0.4

0.6 p=0.06 p=0.04

phospho-p38MAPK

p38MAPK

Rac-FRETfl/fl Rac-FRET

- + - + PDGF

Pho

spho

-p38

MA

PK

(AU

)

0

20

40

60

80

100 p=0.04 p=0.08

i ii b a

10 µm

10 µm

10 µm

10 µm

d c

Supplemental p6

Legend for Figure S2, related to Figure 2. Normal function and morphology of Rac-FRET MEFs.

(A) Rac-FRET MEF numbers increase during culture at the same rate as those from Rac-FRETKi/fl and Rac-

FRETfl/fl littermates, and cell death is low throughout. Numbers of living (filled symbols) and dead MEFs

(open symbols) from Rac-FRET (circle), Rac-FRETfl/fl (triangle) and Rac-FRETKi/fl (crosses) littermate E13.5

embryos after culture for the indicated periods of time. Data are mean ± SEM of n>6 embryos/genotype

from 3 timed matings. Statistics are 2-way Anova.

(B) Proliferation rates and proportions of apoptotic cells in exponentially growing cultures of Rac-FRET

and Rac-FRETfl/fl littermate MEFs are comparable. Rac-FRET (gray) or Rac-FRETfl/fl (black) MEFs from

littermate E13.5 embryos were propidium-idodine stained and their stages in cell cycle analysed by flow

cytometry. n≥5 embryos/genotype from 5 matings. Paired t-test showed non-significance.

(C) Serum-starved Rac-FRET MEFs display normal morphologies (lamellipodia, peripheral or dorsal

ruffles and spreading) and normal PDGF-dependent stimulation of dorsal ruffling. Serum-starved Rac-

FRET or Rac-FRETfl/fl MEFs were stimulated with or without 50 ng/ml PDGF for 5 min, fixed, TRITC-

phalloidin stained, and their basal or ‘active Rac’ morphologies assessed by fluorescence microscopy. (i)

Photos show representative cell morphologies (a) basal, (b) lamellipodia and edge ruffles, (c) dorsal

ruffles and (d) spreading. (ii) Scoring of >1000 cells per condition for morphologies as in (i).

(D) Raichu-Rac expression does not affect endogenous Rac1 activity in Pak-CRIB pull down assays with

Rac-FRET MEFs. GTP-loading of endogenous Rac1 is normal in Rac-FRET MEFs. Pak-CRIB pull down assays

in total lysates of Rac-FRET (Rac-FRETKi/Ki) and Rac-FRETfl/fl MEFs from littermate E13.5 embryos grown on

glass slides, serum-starved and stimulated with 50 ng/ml PDGF for 5 min. 20 % of the pull down and 2%

of the total lysate samples loaded per lane. Data are mean Rac activity ± SEM of 4 independent

experiments. Statistics are paired two-tailed t-test.

Supplemental p7

(E) Basal and PDGF-stimulated p38MAPK phosphorylation is normal in Rac-FRET MEFs. Total lysates from

Rac-FRET (Rac-FRETKi/Ki) or Rac-FRETfl/fl MEFs prepared as in (D) were western blotted for phospho- and

total p38MAPK. Data are mean p38MAPK phosphorylation ± SEM relative to unstimulated Rac-FRETfl/fl from 3

independent experiments. Statistics are paired two-tailed t-test.

Figure S3, related to Figures 3 and 4 Johnsson et al. Normal Rac-FRET neutrophil responses

B D C

E F G

A i ii iii iv

RO

S pr

oduc

tion

(AU)

Time (min)

Unprimed TNFα-primed

**

***

No. o

f cel

ls p

er fi

eld

of v

iew

Supplemental p9

Figure S3, related to Figures 3 and 4. Normal Rac-FRET neutrophil responses.

(A) Isolated Rac-FRET neutrophils mount a normal ROS response to the GPCR ligand fMLP, both with or

without prior TNFα priming, and to PMA, whereas Rac2-/- neutrophils, used as controls, showed the

expected defects. (i), Representative kinetic traces of ROS production by fMLP-stimulated WT and Rac-

FRET neutrophils. (ii-iv), Total ROS production, determined as area under the curve, upon stimulation

with 10 µM fMLP (ii), priming with 500U/ml TNFα for 45 min prior to fMLP stimulation (iii), or

stimulation with 500 nM PMA (iv). Data are mean % of the stimulated WT response. For WT and Rac-

FRET, n = mean ± SEM of 4 individual experiments in (ii) and (iii) and 3 in (iv). For Rac2-/-, n = mean ± SEM

of 3 experiments in (ii) and (iii) and 2 (± range) in (iv). Difference from WT was calculated by one-sample

t-test and considered significant at *p<0.05.

(B) Rac-FRET neutrophils adhere normally to glass or the integrin ligand pRGD, whereas Rac2-/-

neutrophils show the expected defects. Mean number of WT or Rac-FRET cells per field of view that

adhered to uncoated or pRGD-coated glass coverslips within 15 min, determined by image analysis as

detailed in Supplemental Experimental Procedures, from n = 3 (± SEM) individual experiments for WT

and Rac-FRET, or 2 (± range) for Rac2-/-.

(C) Rac-FRET neutrophils spread normally under basal and fMLP-stimulated conditions, whereas Rac2-/-

neutrophils show the expected defects. Rac-FRET neutrophils spread normally on pRGD coated

coverslips. Mean cell area of neutrophils that had been spreading on pRGD-coated coverslips for 15 min,

determined by image analysis as detailed in Supplemental Experimental Procedures, from n = 3 (± SEM)

individual experiments for WT and Rac-FRET, or 2 (± range) for Rac2-/-.

(D) Rac-FRET neutrophils undergo normal basal migration and fMLP-stimulated chemotaxis in transwell

assays, whereas Rac2-/- cells show the expected lack of motility. Basal and 3µM fMLP stimulated

migration of WT, Rac-FRET or Rac2-/- bone marrow leukocytes through 3 µm transwell filters within 40

Supplemental p10

min; number of migrated neutrophils determined post-assay by flow cytometry and expressed as % of

total neutrophils in n=2 experiments ± range. Statistics are one-way ANOVA with Dunnett’s post hoc

test; difference form Wt considered signify cant at p<0.05.

(E) Endogenous Rac1 activity (GTP-loading) is normal in fMLP-stimulated Rac-FRET neutrophils. WT and

Rac-FRET neutrophils were stimulated for 10s with 10 µM fMLP followed by pull down of active Rac

using Pak-CRIB sepharose and Rac1 western blotting. Top: GTP-loading of Rac1 was determined by

densitometric analysis of western blots and expressed as % of total Rac1; n = mean (± SEM) of 4

experiments. Statistics are Student’s t-test; difference from WT not significant. Bottom: representative

Rac1 western blots of pull down samples (GTP) and 2% of the total lysate controls (Total). GTPγS was

used as a positive control for GTP-loading.

(F) Endogenous Rac2 activity (GTP-loading) is normal in fMLP-stimulated Rac-FRET neutrophils.

Experiments as in (E), but with Rac2 western blotting.

(G) The Rac target p38MAPK shows normal levels of phosphorylation in fMLP-stimulated Rac-FRET

neutrophils. WT or Rac-FRET neutrophils were stimulated with 0.3 or 1 µM fMLP and total cell lysates

subjected to phospho-p38MAPK and total p38 western blots. Levels of phospho-p38 were determined by

densitometry and compared to total p38 for loading. Top: Phospho-p38MAPK as % of unstimulated WT.

N=mean (± SEM) of 3 experiments. Statistics are Student’s t-test; difference from WT not significant.

Figure S4, related to Figure 4 Johnsson et al. High Rac activity correlates with protrusion formation at the leading edge and oscillates between the front and back of chemotaxing Rac-FRET neutrophils.

A B D

Inac

tive

Act

ive

C

G

Time (s)

Rac

act

ivity

(bac

kgro

und-

ad

just

ed F

RE

T ra

tio)

Max intensity Mean intensity

Rac

act

ivity

(%

diff

eren

ce fr

ont/b

ack)

F E 1.9

1.8

1.7

1.6

1.5

1.4

1.3

1.2

15

10

5

0 100 80 60 40 20 0 120

Control NSC23766

Mig

ratio

n m

ode

(%)

H 100

80

60

40

20

0

Non- directional

non-trans- locating

anti- directional

directional

directional

*

****

2.0

1.5

1.0

0

Rac

act

ivity

(FR

ET ra

tio)

Control NSC23766 Max Mean Max Mean

0.8

0.6

0.4

0.2

0 Perio

dici

ty o

f Rac

act

ivity

os

cilla

tions

(R2 )

Control NSC23766 Directed Other Directed Other

Not applic.

*

***

i ii iii

Supplemental p12

Figure S4, related to Figure 4. High Rac activity correlates with protrusion formation at the leading

edge and oscillates between the front and back of chemotaxing Rac-FRET neutrophils.

(A) Rubber band plot of the cell outline (dark green = start; bright yellow = end) of the 20 time frames of

neutrophil chemotaxis shown in Figure 4A.

(B) Migration path (centre = start) of the 20 time frames shown in Figure 4A.

(C) Polar plot of the average Rac activity (FRET ratio; red = high, blue = low) around the cell perimeter of

41 cells from 4 independent chemotaxis experiments over 120s. For ratiometric FRET measurements:

increase in FRET ratio = increase in Rac activity.

(D) Average migration path of 41 cells as in (C) from 4 independent experiments over 120s.

(E) Schematic of QuimP nodes around the cell perimeter and 0.4 µm from the cell edge used to calculate

Pearson correlation between Rac activity and cell edge velocity of cells analysed in Figure 4D.

(F) Difference in Rac activity (FRET ratio) between the front and back of chemotaxing neutrophils as in

Figure 4F during stalling and protruding migration (paired T-test Rac activity at the front versus the back,

**p<0.005). No significant change in front/back ratio of Rac activity between protruding and stalling

phases of migration, despite the retrograde shift of peak Rac activity during stalling phases (Figure 4F).

(G) Rac activity oscillates between the front and the back of chemotaxing neutrophils. The maximum

and mean Rac activities (FRET ratios) along the central longitudinal axis of chemotaxing neutrophils as in

Figure 4G are plotted for each time-frame to allow a comparison of the spatial localization of the point

of highest Rac activity (Figure 4G) with the intensity of Rac activity for each 1s time frame.

(H) Chemotaxis, Rac activity and oscillations of Rac activity between the front and the back of migrating

neutrophils are inhibited by NSC23766 treatment. Rac-FRET neutrophils were subjected to chemotaxis

assays as in Figure 4 and above, except that they were pretreated for 5 min either in the presence or

absence of 75 µm NSC23766 and then assayed with or without the inhibitor. (i) NSC23766 inhibits

directional migration towards the chemoattractant. Migration of cells which showed any movement at

Supplemental p13

all was categorised as directional migration towards the chemoattractant, non-directional, anti-

directional or non-translocating migration. Data is from 29 control and 47 NSC23766-treated cells. (ii)

The maximum and mean Rac activities (FRET ratios) along the central longitudinal axis of chemotaxing

neutrophils was assessed as in Figure S4G and the average of all time-frames plotted for the same

control and NSC23766-treated cells as in (i). (iii) The oscillations of Rac activity (FRET ratio) along the

central longitudinal axis of chemotaxing neutrophils are perturbed by NSC23766 treatment. Cells were

categorised into directionally migrating cells and cells without directional migration towards the

chemoattractant (non-directional, anti-directional or non-translocating) and analysed as in Figure 4G for

oscillations of Rac activity along the central longitudinal axis. Plotted are mean R2 values ± SEM which

depict the periodicity of these oscillations from 13 control cells, 4 directionally migrating NSC23766-

treated cells and 8 non-directionally migrating NSC23766-treated cells. Statistics are unpaired t-test.

Figure S5, related to Figures 5-7 Johnsson et al. PMA stimulates Rac activity in skin explants of E15.5 K14-Cre Rac-FRETfl/fl mice

B

Fluo

resc

ence

life

time

(ns)

Time after PMA (min)

1.95

2.00

2.05

2.10

2.15

2.20

2.25

2.30

0 5 10 15 20 25 30

***

***

E15.5 K14-Cre Rac-FRETfl/fl skin ex vivo pre-PMA

E15.5 K14-Cre Rac-FRETfl/fl skin ex vivo post-PMA

Activ

e

In

activ

e 3.0

2.5

2.0

1.5

1.0

Activ

e

In

activ

e

3.0

2.5

2.0

1.5

1.0 50 μm

50 μm

A

Supplemental p15

Figure S5, related to Figures 5-7. PMA stimulates Rac activity in skin explants of E15.5 K14-Cre Rac-

FRETfl/fl mice.

(A) Representative FLIM-FRET images of Rac activity in skin explants of E15.5 embryos from K14-Cre Rac-

FRETfl/fl mice which express Raichu-Rac selectively in keratinocytes, shown before (top panels) and after

(bottom panels) stimulation with 400 nM PMA for 30 min. For each pair of images, the left-hand panel

shows a representative fluorescence image of the skin explants expressing Raichu-Rac (blue), and the

right-hand panel shows a corresponding Rac activity (fluorescence lifetime) map. Insets show

enlargements of the cells in the boxed sections.

(B) Quantification of Rac activity in keratinocytes as in (A) upon PMA stimulation for up to 30 min. Mean

fluorescence lifetime ± SEM for 150 cells before and 40-60 cells for each time-point of PMA stimulation

in 4 different regions. *** p<0.001 by unpaired Student t- test.

B

Fluo

resc

ence

life

time

(ns)

PMA

2.1

1.9

1.7

1.5

1.6

1.8

2.0

1.4 - +

*

Figure S6, related to Figures 5-7 Johnsson et al. Stimulation of Rac activity in in the liver by PMA treatment ex vivo.

Rac-FRET liver ex vivo post-PMA

Activ

e

In

activ

e

3.0

2.5

2.0

1.5

1.0

Rac-FRET liver ex vivo pre-PMA

Activ

e

In

activ

e

3.0

2.5

2.0

1.5

1.0 40 µm

40 µm

A

Supplemental p17

Figure S6, related to Figures 5-7. Stimulation of Rac activity in in the liver by PMA treatment ex vivo.

(A) Representative FLIM-FRET images of Rac activity in the liver of Rac-FRET mice before (top panels) or

after (bottom panels) stimulation with 200 nM PMA for 15 min ex vivo. For each pair of images, the left-

hand panel shows a representative fluorescence image of the Rac-FRET mouse liver expressing Raichu-

Rac (blue) and the right-hand panel shows a corresponding Rac activity (fluorescence lifetime) map.

(B) Quantification of Rac activity (fluorescence lifetime; mean of 302 cells ± SEM before or after

treatment of the liver with PMA ex vivo. * p < 0.05, by unpaired Student t- test.

Supplemental p18

Supplemental Movie Legends

Movie S1, related to Figure 2. Ratiometric FRET live-imaging of PDGF-stimulated primary Rac-FRET

MEFs. Ratiometric FRET live-imaging of Rac activity at lamellipodial protrusions and membrane ruffles of

PDGF stimulated (50 ng/ml) Rac-FRET MEFs; frames taken every 10s. The left panel shows the YFP-FRET

movie (CFP excitation, YFP emission filter), the middle panel the CFP movie, and the right panel the

resulting FRET ratio movie. The time of PDGF addition is denoted by an empty frame. The cell shown is

representative of 20 imaged under these conditions from 3 independent experiments. For ratiometric

FRET measurements: increase in FRET ratio = increase in Rac activity.

Movie S2, related to Figure 2. Ratiometric FRET live-imaging of Insulin-stimulated primary Rac-FRET

MEFs. Ratiometric FRET live-imaging of Rac activity at lamellipodial protrusions and membrane ruffles of

a section from a representative Insulin stimulated (100 µg/ml) Rac-FRET MEF; frames taken every 10s.

The left panel shows the YFP-FRET movie (CFP excitation, YFP emission filter), the middle panel the CFP

movie, and the right panel the resulting FRET ratio movie. The time of Insulin addition is denoted by an

empty frame. The cell shown is representative of 11 imaged under these conditions from 2 independent

experiments.

Movie S3, related to Figures 4 and S4. Ratiometric FRET live-imaging of primary Rac-FRET neutrophils

chemotaxing in an fMLP gradient. Ratiometric FRET live-imaging movie of a representative Rac-FRET

neutrophil chemotaxing towards 3 µM fMLP in an Ibidi chamber. Images were taken at 1s intervals over

101s. The chemoattractant was added before the start of imaging, and the chemoattractant source is

due-South of the movie.

Supplemental p19

Movie S4, related to Figure 5. Multiphoton imaging of Rac-FRET mouse intestine. Ex vivo multiphoton

z-stack imaging of Rac-FRET mouse intestine to visualize the 3D-space investigated from the base of the

crypts towards the villi (0-150 µm). Raichu-Rac expression within the crypt is shown in blue (signal in

intestine of non Rac-FRET mice is negligible) and the corresponding host ECM content in red.

Movie S5, related to Figure 6. Multiphoton imaging of Rac-FRET mouse pancreas. Ex vivo multiphoton

z-stack imaging of Rac-FRET mouse pancreas to show the 3D-space investigated. Raichu-Rac expression

is shown in blue (signal in pancreas of non Rac-FRET mice is negligible) and host ECM in white.

Movie S6, related to Figure 7. Multiphoton imaging of mammary tumors of PyMT Rac-FRET mice. Ex

vivo multiphoton z-stack imaging (maximum projection) of PyMT Rac-FRET mouse mammary tumors to

show the 3D-space investigated. Raichu-Rac expression in the tumor is shown in blue (signal in breast

tumours of non Rac-FRET mice is negligible) and host ECM in white.

Movie S7, related to Figures 5-7 and S5. Multiphoton imaging of embryonic skin from K14-Cre Rac-

FRETfl/fl mice. Ex vivo multiphoton z-stack imaging (maximum projection) of keratinocytes in embryonic

skin from K14-Cre Rac-FRETfl/fl mice to show the 3D-space investigated. Expression of Raichu-Rac is

shown in blue (signal in embryonic skin of non Rac-FRET mice is negligible) and host ECM in white.

Movie S8, related to Figures 5-7 and S6. Multiphoton imaging of Rac-FRET mouse liver. Ex vivo

multiphoton z-stack imaging of Rac-FRET mouse liver to show the 3D-space investigated. Raichu-Rac

expression in the liver is shown in blue (signal in liver of non Rac-FRET mice is negligible) and host ECM

in red.

Supplemental Experimental Procedures

Generation of the Rac-FRET reporter mouse strain:

Raichu-1011X plasmid (Itoh et al., 2002) encoding the original intra-molecular FRET reporter for Rac

activity (a gift from Michiyuki Matsuda, Kyoto, Japan) was modified by PCR to introduce an AscI

restriction site and Kozak translation initiation consensus sequence 5’ of the Raichu-Rac cassette, using

primers CACGGCGCGCCACCATGGTGAGCAAGGGCGAG and CTCTTTCTCGAGGGCGGCGGTC. The resulting

PCR fragment was digested with XhoI and ligated into EcoRI-digested Raichu-1011X that had been

treated with Klenow DNA polymerase to generate a blunt end and then digested further with XhoI. The

resulting modified Raichu-1011X plasmid was verified by sequencing.

To generate the Rac-FRET targeting vector, the AscI/Kozak-modified Raichu-1011X plasmid was

digested with SalI, treated with Klenow to generate a blunt end and then digested further with AscI. The

AscI/blunt restriction fragment was ligated into AscI/XmaI-digested STOP-eGFP-ROSA26TV vector

(Addgene 11739), a targeting vector originally generated by the Rajewsky lab (Sasaki et al., 2006) which

is widely used for ubiquitous gene expression from the endogenous ROSA26 promoter on mouse

chromosome 6 (Zambrowicz et al., 1997) and works on the principle that floxed genes become

expressed after Cre recombinase-mediated deletion of a neomycin resistance/STOP cassette that is

flanked by two loxP sites. The Rac-FRET targeting vector was digested with ApaLI to remove the vector

backbone, yielding a targeting fragment consisting of the 5’ recombination arm, a splice acceptor site,

the loxP-flanked neomycin resistance/STOP cassette, Raichu-Rac, and the 3’ recombination arm. The

purified targeting fragment was electroporated by the Babraham Gene Targeting Facility into C57Bl/6

Bruce4 ES cells. Neomycin-resistant ES clones were screened by PCR using primers upstream of the 5’

recombination arm (CCCACCGCCCCACACTTATTG) and within the splice acceptor site

(GACCGCGAAGAGTTTGTCCTCAAC). Genomic DNA from clones that yielded the expected 1356 bp PCR

Supplemental p20

product was digested with EcoRI or EcoRV and subjected to Southern blotting to confirm recombination,

using 1 kb 5’ and 3’ probes that hybridised outside of the targeted region, making use of the introduced

EcoRI and EcoRV restriction sites.

ES cells from one positive clone were injected into albino C57Bl/6 blastocysts and the resulting

chimeras crossed with C57Bl/6 mice. After germline transmission, the resulting strain, called Rac-

FRETfl/fl, was crossed with ‘deleter’ strain which expresses Cre recombinase from the X chromosome for

ubiquitous excision of loxP-flanked sequences (Schwenk et al., 1995), to induce the ubiquitous

expression of Raichu-Rac by removal of the floxed neomycin resistance/transcriptional STOP cassette.

Finally, the Cre recombinase allele was crossed out again by selective breeding. The resulting

homozygous Rac-FRETKi/Ki knock-in (Ki) mouse strain was called Rac-FRET for brevity. We also trialled

drug-inducible expression of Raichu-Rac by crossing Rac-FRETfl/fl with Cre-ERT2 strain (Feil et al., 1996),

which inducibly expresses Cre-recombinase from the ROSA26 locus upon tamoxifen treatment, for

analysis of F1 generation Cre-ERT2 Rac-FRETfl mice.

Genotyping:

Genotyping of Rac-FRETfl/fl was routinely done by PCR, with a forward primer targeted 5’ of the splice

acceptor (GTCGCTCTGAGTTGTTATCAGTAAGGGAGC) and reverse primers in the Neo cassette of the

mutated allele (TCGCCTTCTATCGCCTTCTTGA) or in the Wt allele (AACCCCAGATGACTACCTATCCTCC),

resulting in 453 bp and 393 bp fragments, respectively. Genotyping of Rac-FRETKi/Ki (after Cre-mediated

excision of the Neo/STOP cassette) was done using the Rac-FRETfl/fl forward primer and a reverse primer

within the Pak-CRIB domain (GAAGCACTGCAGGCCGTA), yielding a 585 bp band. Breeding-out of the Cre

recombinase allele after excision of the Neo/STOP cassette was monitored using primers

TACCTGGCCTGGTCTGGACACAGTG and ATGGCTAATCGCCATCTTCCAGCAG, testing for the absence of the

350 bp Cre band.

Supplemental p21

Mouse husbandry:

Rac-FRET, Rac-FRETfl/fl and Rac2-/- (Roberts et al., 1999) strains were bred and housed in specific

opportunistic pathogen free isolators. C57Bl/6 mice from our Biological Services Breeding Unit (used as

Wt controls), Cre-ErT2 (Feil et al., 1996), and Cre-ErT2 Rac-FRETfl mice were bred and housed in

individually vented cages in a shower-in barrier facility. Rac-FRET mice were bred either from

heterozygous pairs (Rac-FRETKi/fl x Rac-FRETKi/fl) for generating Rac-FRETKi/Ki (Rac-FRET), Rac-FRETKi/fl and

Rac-FRETfl/fl littermates, or from homozygous Rac-FRET pairs, and both bred equally well. Tamoxifen-

injected Cre-ErT2 Rac-FRETfl/fl animals were housed in individually vented cages in a shower-in barrier

experimental facility. Mice were used at between 8-14 weeks of age and were sex- and age-matched

where possible. For treatment of animals used in FLIM-FRET imaging experiments, see each relevant

section. All animal work was carried out under the control of the British Home Office and the local

Animal Welfare and Experimental Ethics Committee.

Quantification of Raichu-Rac expression in Rac-FRET cells and tissues:

The expression level of Raichu-Rac protein in cells and tissues of the Rac-FRET mouse was determined

using standard curves of recombinant CFP and/or Rac1, as available Rac1 antibodies did not recognize

the Rac1 protein within the biosensor, presumably due to the deletion of the 16 most C-terminal

residues of Rac1 in the biosensor molecule. For multiple-tissue comparison, 2% w/v whole-tissue lysates

from adult Rac-FRET mice were prepared by rinsing isolated tissues/organs in PBS, finely mincing with a

razor blade on ice, boiling in 1.3x Laemmli sample buffer and trituration through 19 and then 25-gauge

needles. For determination of Raichu-Rac expression in primary neutrophils, freshly prepared bone-

marrow derived mature neutrophils were incubated with 7 mM DFP for 10 min at RT to prevent

proteolysis, washed twice in DPBS and boiled in SDS-sample buffer. For measuring Raichu-Rac

Supplemental p22

expression in primary mouse embryonic fibroblasts (MEFs; see below), Rac-FRET MEFs were cultured for

12 days post-isolation, washed twice in PBS, lysed in ice-cold lysis buffer (50 mM Tris, pH 7.4, 100 mM

NaCl, 1.2% NP-40, 10% glycerol, 2 mM MgCl2, 1 mM DTT, 0.1 mM PMSF, 10 µg/ml antipain, 10 µg/ml

aprotinin and 10 µg/ml leupeptin) for 10 min on ice, and lysates boiled in SDS-sample buffer. Lysates

were subjected to SDS-PAGE alongside protein standards of recombinant purified EE-Rac1 (Welch et al.,

2002) and CFP (MBL International) and western blotted with Rac1 (Millipore, 053989) and GFP (Santa

Cruz, SC-8334) antibodies. To quantify Raichu-Rac, fluorescence detection with Licor Biosciences goat

anti-rabbit 680RD (926-68071) and goat anti-mouse 800RD (926-32210) or standard ECL coupled with

ImageJ were used and blots analysed on the basis that our GFP antibody recognized YFP and CFP to

equal extent and that Rac1 and Rac2 are expressed at equal levels in neutrophils (Li et al., 2002).

Rac-FRET neutrophil isolation and responses:

Isolated bone-marrow derived mature neutrophils were used unless otherwise indicated. Neutrophils

were freshly isolated by Percoll Plus gradient for each experiment as described, either at RT (Lawson et

al., 2011; Welch et al., 2005) or 4°C (Ferguson et al., 2007) depending on the subsequent assay, typically

to a purity of 80- 90% as determined by May-Gruen-Giemsa staining of cytospins. Neutrophils adhesion

and spreading assays were performed on glass or integrin-ligand surfaces (Lawson et al., 2011), and ROS

formation determined by luminol assay (Lawson et al., 2011; Welch et al., 2005), as previously

described. Transwell chemotaxis assays were done with crude bone marrow cells and neutrophils

identified post-assay by flow cytometry, as described (Lawson et al., 2011; Welch et al., 2005). Ibidi

chamber chemotaxis assays were performed with Ibidi sticky slide IV0.4 coated with 10% mouse serum in

HBSS+ (HBSS with Ca2+ and Mg2+, 15 mM Hepes, pH7.4, all endotoxin free) for 30 min at RT and then

washed three times in HBSS+. Neutrophils were added to the channel at 2.5x107 cells/ml and, without

letting cells settle, 5 µl of 3µM fMLP containing 5x106/ml latex beads was added to one well of the slide

Supplemental p23

while the same volume was aspirated from the opposing well immediately thereafter, forming a steep

and short-lived gradient. The latex beads were allowed to settle and were used as an indicator for the

area of steepest gradient, according to the manufacturer’s instructions (Ibidi application note 01).

Rac-FRET MEF isolation, culture and responses:

Rac-FRETKi/fl mice were timed-mated and E13.5 MEFs isolated by incubating diced tissue with 0.05%

trypsin-EDTA at 37°C for 25 min, triturating the cell suspension and passing it through 40 μm cell

strainers. MEFs were cultured in complete medium (DMEM, high glucose, with glutamine; Invitrogen],

10% batch-adjusted heat-inactivated FBS [PAA], 0.1 mM β-mercaptoethanol, 1unit/ml penicillin and

1μg/ml streptomycin) for up to four passages or 15 days. For cell proliferation assays, Rac-FRETfl/fl, Rac-

FRETKi/fl and Rac-FRET MEFs from littermate embryos at 3-5 days post-isolation were plated at

0.8x105/ml in triplicate in 12-well plates. After 24, 48, 72 or 96 h, cells were trypsinised, stained with

0.04% trypan blue/PBS, and counted by hematocytometer. For cell cycle assays, Rac-FRETfl/fl and Rac-

FRET MEFs from littermate embryos were grown overnight in 10 cm dishes at 3x106 cells/dish,

trypsinised, washed in PBS, fixed in ice-cold 70% ethanol/PBS on ice for 30 min, washed, stained with 50

µg/ml propidium iodide in PBS containing 0.1 mg/ml RNAse for 10 min at RT, and analysed by flow

cytometry. For cell morphology assays, Rac-FRETfl/fl and Rac-FRET MEFs from littermate embryos at 3-5

days post-isolation were plated onto sterile glass coverslips at 2x105/well in complete medium in 6-well

plates. After 24 h, MEFs were washed twice with DMEM and serum-starved overnight in DMEM/0.1%

fatty-acid free BSA before stimulation with 50 ng/ml PDGF-BB (Source Bioscience) for 5 min. Cells were

fixed in 4% paraformaldehyde in 50 mM Pipes, pH 6.8, 1 mM EGTA, 1 mM MgCl2 for 15 min at RT,

washed twice in PBS, permeabilised in PBS/0.1% Triton X100 for 10 min, and washed again. F-actin was

stained using 2 μg/ml TRITC-phalloidin in PBS for 1 h in the dark. Slides were washed in H2O and

mounted using AquaPolymount (PolySciences, 18606-20). Samples were blinded and imaged using a

Supplemental p24

Zeiss AxioImager. 10-12 photos were taken across the width of each coverslip and all cells scored for

basal or ‘active Rac’ morphologies (spreading, lamellipodia, peripheral and dorsal membrane ruffles).

Biochemical assays:

For Pak-CRIB assays in neutrophils, 1-2x107 Rac-FRET or Wt bone-marrow derived neutrophils/ml were

pre-warmed at 37°C in DPBS (with Ca2+ and Mg2+) and stimulated for 10s with 10 µM fMLP. The reaction

was stopped by addition of 5 vol ice -cold lysis buffer (see above). Cells were lysed on ice for 3 min,

insoluble material removed by centrifugation at 13,400 xg for 3 min at 4°C, and GTP-bound Rac isolated

using immobilised GST-Pak-CRIB as described (Welch et al., 2005), with 2% of the lysates being set aside

for monitoring total Rac levels. For measuring maximal cellular Rac activity, some total lysate samples

were complemented with 63 µM GTPγS and 10 mM EDTA and incubated at 30oC for 30 min, before

stopping the reaction by the addition of 60 mM MgCl2. Samples were western blotted with monoclonal

Rac1 (05-389, Millipore) or polyclonal Rac2 antibody (07-604, Millipore) and analysed using ImageJ. For

MEFs, 1x106 Rac-FRET or Rac-FRETfl/fl cells at 3-5 days post-isolation from littermate embryos were

grown on 57x64 mm glass slides (Agar Scientific, Stansted, UK) in complete medium for 24 h, serum-

starved overnight in DMEM/0.1% fatty-acid free BSA and stimulated for 5 min with 50 ng/ml PDGF.

Medium was aspirated, dishes transferred onto iced metal trays, cells scraped into 1 ml ice-cold lysis

buffer, incubated for 3 min, and insoluble material removed by centrifugation at 13,400 xg for 3 min at

4°C. Pak-CRIB pull down was performed as above and samples western blotted with monoclonal Rac1

antibody (05-389) and analysed by ImageJ. To measure p38MAPK phosphorylation, Rac-FRET and Wt

neutrophils were prepared as for Pak-CRIB assays, stimulated with 0.3 or 1 µM fMLP for 45 s, the

reaction stopped with 10 vol DPBS (with Ca2+ and Mg2+) and cells spun at 13,400 xg for 10s at 4°C. Ice-

cold anti-phosphatase lysis buffer (20 mM Tris pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton

X100, 2.5 mM sodium pyrophosphate, 1mM β-glycerophosphate, 1 mM sodium orthovanandate, 0.1

Supplemental p25

mM PMSF, 10 µg/ml leupeptin and 10 µg/ml aprotinin) was added, samples incubated on ice for 3 min,

spun for 3 min as above, and the supernatant boiled in SDS-sample buffer. Western blots were

performed using phospho-p38MAPK and total p38MAPK antibodies (Cell Signalling 9211 and 9212). For

p38MAPK phosphorylation assays in Rac-FRET MEFs, samples were prepared as for Pak-CRIB assays and

analysed as for neutrophils.

Imaging and quantification of Rac activity in MEFs and neutrophils by ratiometric FRET analysis:

For Rac-FRET neutrophils and MEFs, all images were acquired using an Olympus Cell^R imaging system

comprising an Olympus IX81 microscope, Olympus 60×1.4 NA objective, Till Photonics Polychrome V

polychromator light source, Andor iXon EM CCD camera (Hamamatsu Orca ER used for MEF

experiments) and a Solent Scientific environment chamber to maintain 37°C. Wide-field epifluorescence

images were collected using 425 nm excitation with a 10 nm band width and two emission filters 483/32

for CFP and 542/27 for YFP with an exposure time of 500 ms. Sequential acquiring of images was used

for neutrophil spreading experiments and for MEFs, with pairs of images being acquired every 15s for

spreading neutrophils and every 10s for MEFs. For imaging of chemotaxing neutrophils, an OptoSplit II

Image splitter (Cairn Research Ltd, Faversham, UK) was used with an associated ImageJ plugin so that

both channels were acquired simultaneously at 1 frame/s. The CFP and YFP fluorescence intensity

images were background-subtracted and median-filtered (over a circle with a 2-pixel radius; ImageJ). A

pixel-by-pixel YFP/CFP ratio image was generated using Ratio-Plus plugin (created by Paulo Magalhães,

available from ImageJ), which allowed setting a cut-off for pixel intensity, thereby making it possible to

exclude weakly fluorescing pixels. The cut-off was chosen based on the histogram after background

subtraction, of an area devoid of cells.

Supplemental p26

For imaging Rac-FRET MEFs, cells 5-10 days post-isolation were grown on 22 mm glass coverslips

for 24 h in complete medium, washed in PBS and serum-starved overnight in DMEM with 0.1% fatty-acid

free BSA. Imaging was begun before addition of stimulus, and an empty frame was inserted into the

movies to mark the time of addition of either 50 ng/ml PDGF or 100 µg/ml Insulin. CFP and YFP-FRET

(CFP excitation, YFP emission filter) videos were mean-filtered (radius of 1 pixel; ImageJ function which

mean-corrects each pixel to all adjacent pixels in a 3x3 square) and bleach-corrected using

(www.embl.de/eamnet/html/bleach_correction.html), and mean background pixel intensity was

subtracted with a clipping value of max-mean intensity. Rac activity was presented in pseudo-colours

using the ImageJ 16-colour scheme.

For analysing Rac activity in spreading and polarising neutrophils, the cells were added to a dish

containing coverslips and either HBSS with Ca2+/Mg2+ and Rac inhibitor NSC27633 or HBSS with

Ca2+/Mg2+ alone. Imaging commenced from the first point of contact of the cells with the coverslip

(without any preincubation period apart from the time it took the cells to settle onto the coverslips,

usually around 2 mins), and cells were manually kept in focus as they spread. For analysis, the average

FRET ratio for each frame was normalized to the average FRET ratio of the first image from the same

cell, and FRET ratio in NSC23766-treated cells was normalised to that in control cells. The timing of the

start of polarisation, the % of fully polarised cells, and the % of polarised cells reverting to non-polar

morphology were assessed by manually counting these events for all cells that were included in the

FRET ratio analysis.

Neutrophils chemotaxing towards 3 µM fMLP in an Ibidi chamber were imaged at a frame

interval of 1s and tracked using ImageJ plugin QuimP11 (www.warwick.ac.uk) to determine the speed of

movement between frames for a number of nodes on the cell perimeter and the maximal 3x3 FRET ratio

pixel intensity within a chosen distance, here 0.4 µm from the outer node. Nodes were grouped into 5%

bins for each frame, and FRET ratio compared up to 5s forwards and backwards in time with the speed

Supplemental p27

of protrusion or retraction. Pearson correlation was used to analyse the time-dependence between Rac

activity and speed of cell edge movement. Schematic cell traces displaying cell outlines at times T and

T+1s were generated using QuimP11. Polar plots (Ferguson et al., 2007) were used for spatiotemporal

representation of Rac activity at the cell perimeter, with the circumference of the cell represented as a

circle and time represented as excentric circles from the centre outwards. In addition, line scans were

performed along the longitudinal axis of chemotaxing neutrophils each 1s and the pixel with the highest

Rac activity along this line determined for each frame. De-noising was applied to the raw image using

NDSAFIR software (Boulanger et al., 2010). From this, the periodicity of the Rac activity signal shifting

longitudinally throughout the cell over time was calculated using Fourier transformation.

3D intestinal crypt cultures and freshly isolated crypts for ex vivo FLIM-FRET imaging of Rac activity:

Intestinal crypts were cultured as described (Sato et al., 2009). In short, crypts from the Rac-FRET mouse

were purified by incubating the small intestine in 2 mM EDTA for 30 min in cold PBS. After several

washes, crypts were embedded in 20 µl/well growth-factor reduced Matrigel (BD Bioscience) and plated

in a 24-well plate. Each well contained 0.5 ml Advanced DMEM/F12 (Life Technologies) supplemented

with N2 and B27 (Life Technologies). In addition, Noggin (100 ng/ml, Peprotech), EGF (50 ng/ml,

Peprotech) and R-spondin1 (500 ng/ml, R&D Systems) were added and renewed every other day. 3D

crypts were passaged every 7 days and allowed to grow for 4 days after passaging prior to multiphoton

based FLIM-FRET imaging of Rac activity at depths of 0-100 µm in response to stimulation with 200 nM

phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich) for up to 90 min. Ex vivo FLIM-FRET imaging of

Rac activity (see below) in freshly isolated crypts (Myant et al., 2013) was performed from the base of

crypts towards the villi at depths of 0-150 µm.

Supplemental p28

Vil-Cre-ERT2 APCfl/fl Rac-FRET mice for in vivo FLIM-FRET imaging of Rac activity at the base of intestinal

crypts following APC loss:

Vil-Cre-ERT2 APCfl/fl mice (el Marjou et al., 2004; Myant et al., 2013; Sansom et al., 2007) were crossed to

the Rac-FRET mouse to generate Vil-Cre-ERT2 APCfl/fl Rac-FRET mice. Recombination of the APC allele was

induced with 1 intraperitoneal (ip) injection of 80 mg/kg tamoxifen per day for 4 days. Tamoxifen-

induced Vil-Cre-ERT2 Rac-FRET mice were used as controls. Analysis was performed on day 4 after

induction (Myant et al., 2013). To permit FLIM imaging, intestinal tissue was treated with 50 mg/ml (-)-

scopolamine-N-butylbromide for 10 min to minimise peristalsis. Tissue was maintained on a 37°C heated

stage.

KPC Rac-FRET mice for ex vivo and in vivo FLIM-FRET imaging of Rac activity in pancreatic tumours:

For imaging of Rac activity in pancreatic tumours, the Rac-FRET mouse was crossed to the Kras+/G12D

Trp53+/ R172H Pdx1-Cre (KPC) mouse model of pancreatic cancer in which Pdx1-Cre targets tissue specific

expression of active KrasG12D and mutant p53R172H to the pancreas (Hingorani et al., 2005), resulting in

invasive and metastatic pancreatic ductal adenocarcinoma which recapitulates the pathology of the

human disease (Heid et al., 2011; Hruban et al., 2006; Morton et al., 2010; Muller et al., 2009; Muller et

al., 2013; Olive and Tuveson, 2006; Tan et al., 2013). KPC Rac-FRET mice were used at 138, 145 and 156

days of age (average 146), control mice without induction of K-RasG12D and p53R172H expression at 113,

209 and 113 days (average 145); all were on mixed C57Bl6/FvB genetic background. For ex vivo imaging,

the pancreas was dissected out and FLIM-FRET imaged immediately (see below) within a 2-3h period.

PyMT Rac-FRET mice for ex vivo and in vivo FLIM-FRET imaging of Rac activity in mammary tumours:

The Rac-FRET mouse was crossed with the locally invasive Polyoma-middle T (PyMT) breast cancer

model (Guy et al., 1992) to monitor whether NSC23766-dependent inhibition of Rac activity could be

Supplemental p29

observed in the tumors. PyMT Rac-FRET mice on mixed C57Bl6/FvB genetic background were aged

between 80-120 days, with tumours forming after ~40-50 days. Primary tumours were FLIM-FRET

imaged (see below) ex vivo 14-21 days after formation within 2-3h after being dissected out. In vivo

imaging was done as described below.

K14-Cre Rac-FRET mice for tissue-specific expression of Raichu-Rac and ex vivo FLIM-FRET imaging of Rac

activity in embryonic skin:

Rac-FRETfl/fl mice were crossed with K14-Cre strain (Dassule et al., 2000) to induce tissue-specific

expression of Raichu-Rac in the skin. The experimental set up for ex vivo imaging of Rac activity in the

skin of embryos homozygous for Rac-FRET was as described (Mort et al., 2010). Briefly, a section of skin

from the back of E15.5 K14-Cre Rac-FRET embryos was removed, placed epidermis side-down onto a

glass-bottomed dish and covered with a nuclepore membrane (Whatman). Growth Factor Reduced

Matrigel (BD bioscience) was applied to the top to immobilize the sample and incubated for 10 min at

37°C, followed by the addition of culture medium (phenol red-free DMEM with HEPES (GIBCO), 10% FBS,

penicillin, streptomycin). For PMA treatment, 400 nM was added directly to the culture medium for 0-30

min prior to FLIM-FRET imaging.

Multiphoton FLIM-FRET imaging and analysis of Rac activity as fluorescence lifetime:

To permit in vivo imaging, mice were terminally anesthetized using 1:1 hypnorm-H2O + 1:1 hypnovel-

H2O as described (Abdel-Latif et al., 2005; McGhee et al., 2011; Timpson et al., 2011), before being

restrained on a 37°C heated stage and the respective tissue (pancreas, breast, liver or intestine) being

surgically exposed. Where indicated, NSC23766 (Tocris Bioscience) was administered by ip injection

(4mg/kg) (Myant et al., 2013) and Rac activity assessed over 90 min.

Multiphoton-excited FLIM-FRET measurements were conducted using typical scan parameters

Supplemental p30

of 150×150 μm field of view over 512×512 pixels scanned at 400 Hz with the acquisition of 100 frames.

Pixel dwell time was 5 µs with a total acquisition time of approximately 120s. Typical laser power as

measured at the sample plane was ~30 mW. Control and tumour tissues expressing Raichu-Rac could be

imaged at depths of up to 150 μm. Imaging was carried out on a Nikon Eclipse TE2000-U inverted

microscope with an Olympus long working distance 20×0.95 NA water immersion lens. The excitation

source used was a Ti:Sapphire (Cameleon) femtosecond pulsed laser, operating at 80 MHz and tuned to

a wavelength of 830 nm. A scan head specifically designed for multiphoton excitation was used (Trim-

scope, LaVision Biotec, Germany) to control beam scanning and data acquisition. Non-descanned

detectors (NDD) were used (Hamamatsu H6780-01-LV 1M for <500 nm detection and H6780-20-LV 1M

for >500 nm detection), located at the back focal plane of the objective. A dichroic filter (Chroma 455

nm DCXR) was used to spectrally separate the second harmonic generated signal, when present, from

the CFP emission of the Rac-FRET mouse. Band pass filters (Semrock 470/40 and Chroma 500 nm LP)

were used to further filter the emission for the CFP channel. For the measurement of fluorescence

excited state lifetime, a 16-anode PMT (FLIMx16, LaVision Biotec, Germany) was used in a time-

correlated single photon counting (TCSPC) scheme. Detector array sped-up data acquisition by enabling

detection of multiple photons per excitation event. The increased photon count rate enabled accurate

lifetime to be measured in situations not previously accessible in vivo. The TCSPC detector was used

with a gain setting of 250 (0-255 full range) and configured with 75 time bins, each of which were 0.04

ns wide, based on the 80 MHz repetition rate of the laser as described (McGhee et al., 2011; Nobis et al.,

2013; Timpson et al., 2011).

Raw data was analyzed using the built-in TCSPC fluorescence lifetime analysis functionality of

ImSpectorPro (LaVison Biotec, Germany). Single exponential fit with offset was performed on

fluorescence decay data from specific regions and lifetime fit recorded. Statistically significant

differences were assessed by unpaired Student's t-test on multiple data sets. Lifetime maps were also

Supplemental p31

produced using ImspectorPro, where single exponential fit was performed for each pixel across the time

bins spanning the peak to the end of decay. An intensity threshold was applied with the value set to the

average background pixel value for the summed data set. A 3x3 smoothing operation was performed on

the raw data while calculating the FLIM map. A standard rainbow colour look up table with the limits

between 0 and 3.0 ns, 1 and 3.0 ns or 0 and 4 ns, as stated, was used to display all FLIM maps for ease of

comparison between individual tissue types and activation states. In the lifetime colour maps, low basal

Rac activity is represented as warm yellow/red colours, while high Rac activity is represented as blue,

and areas of low signal to noise ratio in which an accurate lifetime measurement could not be achieved

are black. A 5x5 median filter was applied to Tau maps.

Supplemental p32

Supplemental References

Abdel-Latif, D., Steward, M., and Lacy, P. (2005). Neutrophil primary granule release and maximal superoxide generation depend on Rac2 in a common signalling pathway. Can. J. Physiol. Pharmacol. 83, 69-75.

Boulanger, J., Kervrann, C., Bouthemy, P., Elbau, P., Sibarita, J.B., and Salamero, J. (2010). Patch-based nonlocal functional for denoising fluorescence microscopy image sequences. IEEE Trans. Med. Imaging 29, 442-454.

Dassule, H.R., Lewis, P., Bei, M., Maas, R., and McMahon, A.P. (2000). Sonic hedgehog regulates growth and morphogenesis of the tooth. Development 127, 4775-4785.

el Marjou, F., Janssen, K.P., Chang, B.H., Li, M., Hindie, V., Chan, L., Louvard, D., Chambon, P., Metzger, D., and Robine, S. (2004). Tissue-specific and inducible Cre-mediated recombination in the gut epithelium. Genesis 39, 186-193.

Feil, R., Brocard, J., Mascrez, B., LeMeur, M., Metzger, D., and Chambon, P. (1996). Ligand-activated site-specific recombination in mice. Proc. Nat. Acad. Sci. U. S. A. 93, 10887-10890.

Ferguson, G.J., Milne, L., Kulkarni, S., Sasaki, T., Walker, S., Andrews, S., Crabbe, T., Finan, P., Jones, G., Jackson, S., et al. (2007). PI(3)Kγ has an important context-dependent role in neutrophil chemokinesis. Nat. Cell Biol. 9, 86-91.

Guy, C.T., Cardiff, R.D., and Muller, W.J. (1992). Induction of mammary tumors by expression of polyomavirus middle T oncogene: a transgenic mouse model for metastatic disease. Mol. Cell. Biol. 12, 954-961.

Heid, I., Lubeseder-Martellato, C., Sipos, B., Mazur, P.K., Lesina, M., Schmid, R.M., and Siveke, J.T. (2011). Early requirement of Rac1 in a mouse model of pancreatic cancer. Gastroenterology 141, 719-730, 730 e711-717.

Hingorani, S.R., Wang, L., Multani, A.S., Combs, C., Deramaudt, T.B., Hruban, R.H., Rustgi, A.K., Chang, S., and Tuveson, D.A. (2005). Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell 7, 469-483.

Hruban, R.H., Adsay, N.V., Albores-Saavedra, J., Anver, M.R., Biankin, A.V., Boivin, G.P., Furth, E.E., Furukawa, T., Klein, A., Klimstra, D.S., et al. (2006). Pathology of genetically engineered mouse models of pancreatic exocrine cancer: consensus report and recommendations. Cancer Res. 66, 95-106.

Supplemental p33

Itoh, R.E., Kurokawa, K., Ohba, Y., Yoshizaki, H., Mochizuki, N., and Matsuda, M. (2002). Activation of rac and cdc42 video imaged by fluorescent resonance energy transfer-based single-molecule probes in the membrane of living cells. Mol. Cell. Biol. 22, 6582-6591.

Lawson, C., Donald, S., Anderson, K., Patton, D., and Welch, H. (2011). P-Rex1 and Vav1 cooperate in the regulation of fMLF-dependent neutrophil responses. J. Immunol. 186, 1467-1476.

Li, S., Yamauchi, A., Marchal, C.C., Molitoris, J.K., Quilliam, L.A., and Dinauer, M.C. (2002). Chemoattractant-stimulated Rac activation in wild-type and Rac2-deficient murine neutrophils: preferential activation of Rac2 and Rac2 gene dosage effect on neutrophil functions. J. Immunol. 169, 5043-5051.

McGhee, E.J., Morton, J.P., Von Kriegsheim, A., Schwarz, J.P., Karim, S.A., Carragher, N.O., Sansom, O.J., Anderson, K.I., and Timpson, P. (2011). FLIM-FRET imaging in vivo reveals 3D-environment spatially regulates RhoGTPase activity during cancer cell invasion. Small GTPases 2, 239-244.

Mort, R.L., Hay, L., and Jackson, I.J. (2010). Ex vivo live imaging of melanoblast migration in embryonic mouse skin. Pigment Cell Melanoma Res. 23, 299-301.

Morton, J.P., Timpson, P., Karim, S.A., Ridgway, R.A., Athineos, D., Doyle, B., Jamieson, N.B., Oien, K.A., Lowy, A.M., Brunton, V.G., et al. (2010). Mutant p53 drives metastasis and overcomes growth arrest/senescence in pancreatic cancer. Proc. Nat. Acad. Sci. U. S. A. 107, 246-251.

Muller, P.A., Caswell, P.T., Doyle, B., Iwanicki, M.P., Tan, E.H., Karim, S., Lukashchuk, N., Gillespie, D.A., Ludwig, R.L., Gosselin, P., et al. (2009). Mutant p53 drives invasion by promoting integrin recycling. Cell 139, 1327-1341.

Muller, P.A., Trinidad, A.G., Timpson, P., Morton, J.P., Zanivan, S., van den Berghe, P.V., Nixon, C., Karim, S.A., Caswell, P.T., Noll, J.E., et al. (2013). Mutant p53 enhances MET trafficking and signalling to drive cell scattering and invasion. Oncogene 32, 1252-1265.

Myant, K.B., Cammareri, P., McGhee, E.J., Ridgway, R.A., Huels, D.J., Cordero, J.B., Schwitalla, S., Kalna, G., Ogg, E.L., Athineos, D., et al. (2013). ROS production and NF-kappaB activation triggered by RAC1 facilitate WNT-driven intestinal stem cell proliferation and colorectal cancer initiation. Cell Stem Cell 12, 761-773.

Nobis, M., McGhee, E.J., Morton, J.P., Schwarz, J.P., Karim, S.A., Quinn, J., Edward, M., Campbell, A.D., McGarry, L.C., Evans, T.R., et al. (2013). Intravital FLIM-FRET imaging reveals dasatinib-induced spatial control of src in pancreatic cancer. Cancer Res. 73, 4674-4686.

Supplemental p34

Olive, K.P., and Tuveson, D.A. (2006). The use of targeted mouse models for preclinical testing of novel cancer therapeutics. Clinical cancer research : an official journal of the American Association for Cancer Res. 12, 5277-5287.

Roberts, A.W., Kim, C., Zhen, L., Lowe, J.B., Kapur, R., Petryniak, B., Spaetti, A., Pollock, J.D., Borneo, J.B., Bradford, G.B., et al. (1999). Deficiency of the Hematopoietic Cell-Specific Rho Family GTPase Rac2 Is Characterized by Abnormalities in Neutrophil Function and Host Defense. Immunity 10, 183-196.

Sansom, O.J., Meniel, V.S., Muncan, V., Phesse, T.J., Wilkins, J.A., Reed, K.R., Vass, J.K., Athineos, D., Clevers, H., and Clarke, A.R. (2007). Myc deletion rescues Apc deficiency in the small intestine. Nature 446, 676-679.

Sasaki, Y., Derudder, E., Hobeika, E., Pelanda, R., Reth, M., Rajewsky, K., and Schmidt-Supprian, M. (2006). Canonical NF-kappaB activity, dispensable for B cell development, replaces BAFF-receptor signals and promotes B cell proliferation upon activation. Immunity 24, 729-739.

Sato, T., Vries, R.G., Snippert, H.J., van de Wetering, M., Barker, N., Stange, D.E., van Es, J.H., Abo, A., Kujala, P., Peters, P.J., et al. (2009). Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459, 262-265.

Schwenk, F., Baron, U., and Rajewsky, K. (1995). A cre-transgenic mouse strain for the ubiquitous deletion of loxP-flanked gene segments including deletion in germ cells. Nucl. Acids Res. 23, 5080-5081.

Tan, E.H., Morton, J.P., Timpson, P., Tucci, P., Melino, G., Flores, E.R., Sansom, O.J., Vousden, K.H., and Muller, P.A. (2013). Functions of TAp63 and p53 in restraining the development of metastatic cancer. Oncogene. Published online. doi 10.1038/onc.2013.287.

Timpson, P., McGhee, E.J., Morton, J.P., von Kriegsheim, A., Schwarz, J.P., Karim, S.A., Doyle, B., Quinn, J.A., Carragher, N.O., Edward, M., et al. (2011). Spatial regulation of RhoA activity during pancreatic cancer cell invasion driven by mutant p53. Cancer Res. 71, 747-757.

Welch, H.C., Coadwell, W.J., Ellson, C.D., Ferguson, G.J., Andrews, S.R., Erdjument-Bromage, H., Tempst, P., Hawkins, P.T., and Stephens, L.R. (2002). P-Rex1, a PtdIns(3,4,5)P3- and Gβγ-regulated guanine-nucleotide exchange factor for Rac. Cell 108, 809-821.

Welch, H.C., Condliffe, A.M., Milne, L.J., Ferguson, G.J., Hill, K., Webb, L.M., Okkenhaug, K., Coadwell, W.J., Andrews, S.R., Thelen, M., et al. (2005). P-Rex1 regulates neutrophil function. Curr. Biol. 15, 1867-1873.

Zambrowicz, B.P., Imamoto, A., Fiering, S., Herzenberg, L.A., Kerr, W.G., and Soriano, P. (1997). Disruption of overlapping transcripts in the ROSA beta geo 26 gene trap strain leads to widespread expression of beta-galactosidase in mouse embryos and hematopoietic cells. Proc. Nat. Acad. Sci. U. S. A. 94, 3789-3794.

Supplemental p35