Biosensors and Bioelectronics 46 (2013) 97–101

Contents lists available at SciVerse ScienceDirect

Biosensors and Bioelectronics

0956-56

http://d

n Corr

Wuhan

and Tec

fax: þ8

E-m

journal homepage: www.elsevier.com/locate/bios

Short communication

Monitoring of dual bio-molecular events using FRET biosensors basedon mTagBFP/sfGFP and mVenus/mKOk fluorescent protein pairs

Ting Su a,b, Shaotao Pan a,b, Qingming Luo a,b, Zhihong Zhang a,b,n

a Britton Chance Center for Biomedical Photonics, Wuhan National Laboratory for Optoelectronics–Huazhong University of Science and Technology, Wuhan

430074, Chinab MoE Key Laboratory for Biomedical Photonics, Department of Biomedical Engineering, Huazhong University of Science and Technology, Wuhan 430074,

China

a r t i c l e i n f o

Article history:

Received 13 December 2012

Received in revised form

6 February 2013

Accepted 11 February 2013Available online 27 February 2013

Keywords:

Fluorescent protein

FRET biosensor

Multi-parameter imaging

mTagBFP

sfGFP

mVenus

mKOk

63/$ - see front matter & 2013 Elsevier B.V. A

x.doi.org/10.1016/j.bios.2013.02.024

esponding author at: Britton Chance Cente

National Laboratory for Optoelectronics, Hua

hnology, Wuhan 430074, China. Tel.: þ86 27

6 27 87792034.

ail address: czyzzh@mail.hust.edu.cn (Z. Zhan

a b s t r a c t

Fluorescent protein (FP)-based Forster resonance energy transfer (FRET) biosensors are powerful tools

for dynamically measuring cellular molecular events because they offer high spatial and temporal

resolution in living cells. Despite the broad use of FP-based FRET biosensors in cell biology, imaging of

multiple molecular events (multi-parameter molecular imaging) in single cells using current FRET pairs

remains difficult because it usually requires a control group for additional data calibration. Hence,

spectrally compatible FRET pairs that do not require complex image calibration are the key to

widespread applications of FP-based FRET biosensors in multi-parameter molecular imaging. Here,

we report a new combination of spectrally distinguishable FRET pairs for dual-parameter molecular

imaging: mTagBFP/sfGFP (blue and green FP, B/G) and mVenus/mKOk (yellow and orange FP, Y/O). We

demonstrate that additional image correction is not necessary for these dual FRET pairs. Using these

dual FRET pairs, we achieve simultaneous imaging of Src and Ca2þ signaling in single living cells

stimulated with epithelial growth factor (EGF). By converting traditional FRET biosensors into B/G and

Y/O-based biosensors, additional applications are available to elucidate the dynamic relationships of

multiple molecular events within a single living cell.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Forster resonance energy transfer (FRET) is a physical pheno-menon in which an excited chromophore (donor) non-radiativelytransfers its energy to another chromophore (acceptor) when thedistance between donor and acceptor is between 1 and 10 nm.FRET, when applied to optical microscopy, enables researchersto visualize molecular interactions or conformational changesin single living cells, bypassing the intrinsic diffraction limit ofoptical microscopy. The advent of green fluorescent protein (GFP)and its combination with FRET greatly enhance the application ofthis technology within the field of cell biology (DiPilato andZhang, 2010; Yu and Xiao, 2012; Zhang et al., 2002, 2008). Theuse of fluorescent protein (FP)-based FRET has been particularlyadvantageous in the field of biosensors (Campbell, 2009). BecauseFRET technology is inherently quantitative, FRET-based biosensormeasurements are easily read in the ratiometric manner, greatly

ll rights reserved.

r for Biomedical Photonics,

zhong University of Science

87792033;

g).

simplifying experimental procedures. Amongst FP pairs, the CFP/YFP pair has been shown to be superior in providing sensitiveFRET signals, making it the preferred candidate for construction ofFP-based FRET biosensors (Carlson and Campbell, 2009). Althoughit is possible to monitor signals in spatially segregated subcellularlocations by targeting CFP/YFP-based biosensors to these loca-tions, spectral overlap between biosensors hinders the imaging ofmultiple signals within the same intracellular location. Moleculesin living cells form complex, interconnected networks, andthere is a strong demand from cell biologists to synchronouslystudy interacting molecules in single cells (Welch et al., 2011).The discovery and optimization of red-shifted FPs provide numer-ous candidate FRET pairs and make imaging of multiple pairs ofFP biosensors (multi-parameter imaging) spectrally possible(Chudakov et al., 2010).

Because most of the current FP-based FRET biosensors areengineered on the basis of CFP and YFP pair, the simplest FPcombination is to adopt red-shift OFP/RFP (O/R) FRET pair suchas mOrange/mCherry (Ouyang et al., 2010) or mKOk/mLumin(Su et al., 2012). Unfortunately, the relatively low quantum yieldof RFP prevents O/R pairs from providing satisfactory sensitivityin FRET biosensors (Carlson and Campbell, 2009). Alternativesolutions involve the use of fluorescent proteins with unique

T. Su et al. / Biosensors and Bioelectronics 46 (2013) 97–10198

spectral properties (Ai et al., 2008; Nagai et al., 2009). Forexample, Wataru et al. introduced the ultramarine FP Sirius andmade a new combination of dual FRET pairs, including Sirius-mseCFP and Sapphire-DsRed (Nagai et al., 2009). Ai et al. (2008)generated the large stoke shift violet-excitable yellow-fluorescingFP mAmetrine and developed mAmetrine-tdTomato andmCitrine-mTFP1 dual FRET pairs. Although both dual FRET pairscan visualize dual-molecular events in single living cells, extracorrection, involving either linear un-mixing (Nagai et al., 2009)or calibration coefficients (Ai et al., 2008), is needed to acquirenon-interfered ratiometric images.

Recently, we reported a mVenus/mKOk (Y/O) pair that performsbetter than the mKOk/mLumin (O/R) pair in FRET biosensors(Su et al., 2012). In the present study, we introduce a new FRETpair that is composed of the blue fluorescent protein mTagBFP(Subach et al., 2008) and the green fluorescent protein sfGFP(Pedelacq et al., 2006). We show that the mTagBFP/sfGFP (B/G)-based FRET biosensor can be simultaneously imaged in the presenceof Y/O FRET-based biosensor without the need for further correction.Monitoring of two signals demonstrates the power of this combina-tion in acquiring kinetic information that is unattainable previously.

2. Materials and methods

2.1. Plasmid construction

Plasmids encoding each biosensor were genetically engineered.For details, please see the methods section of the Supplemental data.

2.2. Cell culture and transfection

HeLa cells were cultured at 37 1C in 5% CO2 in Dulbecco’smodified Eagle’s medium (DMEM) supplemented with 10% FBS,

Fig. 1. Characterization of the BG-Src biosensor in HeLa cells. (A). Pseudo-color BFP/

Graphical representation of (A) demonstrating the BFP/GFP ratio time course of the BG

upon sequential addition of EGF and the Src inhibitor PP1. (D). Photo-bleaching expe

illumination with a Violet blue 405 nm laser. Imaging parameters are identical with thos

scanned. Scale bar: 10 mm.

100 U/ml penicillin and 100 mg/ml streptomycin. The day beforetransfection, the cells were plated on 35-mm-diameter coverglass-bottom dishes (MatTek Corporation, Ashland, MA) at approxi-mately 50% confluence. Plasmids encoding the fluorescentbiosensors were transfected into the cells using the Lipofectamine2000 reagent according to the manufacturer’s instructions. Fortransfection of a single biosensor, 0.2 mg of DNA was used. Fordual transfections, equal amounts of DNA were used for eachsensor (0.1 mg each). For the EGF stimulation experiments, cellswere allowed to rest for 36 h after transfection and then serum-starved for 6–12 h in DMEM before EGF stimulation.

2.3. Ratiometric imaging

Ratiometric fluorescent imaging was performed with a con-focal laser-scanning microscope FV1000 (Olympus, Japan) andFluoView software (Version 1.5) (Olympus, Japan). Immediatelybefore imaging cells, DMEM was replaced with CO2-independentmedia to maintain cells at physiological pH (7.3). For the BG-Srcbiosensors, mTagBFP was excited at 405 nm, and mTagBFP andsfGFP emissions were collected at 440–480 nm and 500–560 nm,respectively. For the YO-Src and YO-TnC biosensors, mVenus wasexcited at 514 nm and mVenus and mKOk emissions werecollected at 524–540 nm and 560–620 nm, respectively. In thedual ratiometric imaging experiment with the BG-Src andY/O-based biosensors, the ‘‘time-control’’ function in the Fluo-View software was used to quickly switch between imaging ofeach biosensor. Under our imaging conditions, there was anapproximate 3-s delay between imaging of the two biosensors.

2.4. Image processing

‘‘Image J’’ software (http://rsbweb.nih.gov/ij/) was used toprocess confocal images and to generate emission ratio images

GFP ratio images of BG-Src upon EGF (100 ng/ml) stimulation of HeLa cells. (B).

-Src biosensor. (C). BFP/GFP ratio time course of the BG-Src biosensor in HeLa cells

riment demonstrating stable intensities of BFP, GFP and the BFP/GFP ratio upon

e used in the subsequent dual-FRET experiments. A total of 120 image frames were

T. Su et al. / Biosensors and Bioelectronics 46 (2013) 97–101 99

for the BG-Src (or YO-Src and YO-TnC) biosensor. For the BG-Srcbiosensor, the original mTagBFP and sfGFP images were firstbackground subtracted at each time point. The updated imageswere then subjected to Gaussian blur (sigma value 1) for imagesmoothing. Because the sfGFP images had the largest signal-to-noise ratio, the sfGFP image at each time point was used togenerate a binary mask image with pixel values equal to 1 or 0.The mTagBFP and sfGFP images were then multiplied by thismask image. Finally, the updated mTagBFP and sfGFP imageswere used to generate the mTagBFP/sfGFP ratio images. For abetter presentation of ratio images, the brightness/contrast wasadjusted and a linear pseudo-color lookup table was applied.For quantification, the ratio values were normalized to theaverage value of the ratio before EGF stimulation. The aforemen-tioned procedures were also used to generate the mVenus/mKOk(or mKOk/mVenus) ratio images for the YO-Src (or YO-TnC)biosensor. Kinetic curves for the biosensors were plotted usingOriginPro 8 software.

3. Results and discussions

3.1. Characterization of the sfGFP/mTagBFP-based Src biosensor

(BG-Src) in living cells

The key to implementing dual-parameter imaging experiments isthat the two biosensors must be spectrally compatible. When usingdual-FRET biosensors, one FRET biosensor must be ratiometricallyimaged without interference from the other. By analyzing thespectra of current FPs, we determined that the B/G FRET pair mightbe combined with the Y/O FRET pair with minimal interference indual-FRET imaging. From all blue FPs and green FPs, mTagBFP(Subach et al., 2008) and sfGFP (Pedelacq et al., 2006) were chosenbased on their high brightness, fast maturation and high photo-stability (Crivat and Taraska, 2012; Day and Davidson, 2009).

Fig. 2. mTagBFP/sfGFP (B/G) and mVenus/mKOk (Y/O) FRET pairs are spectrally compat

excite the B/G-based FRET biosensor and a 514 nm laser was used to excite the Y/

transfection into HeLa cells. (C, D) Control experiments confirm that the ratio change o

and the ratio change of the Y/O-based calcium (YO-TnC) biosensor does not affect th

emission ratio of BG-Src without affecting the emission ratio of YO-C3 (C), and stimulat

the emission ratio of BG-Src (D). BG-Src is targeted to cytosol by adding nuclear expor

The substantial overlap between the emission spectra of mTagBFPand the absorbance spectra of sfGFP, together with their desiredoptical properties (quantum yield of mTagBFP and sfGFP is 0.63 and0.65, respectively; extinction coefficients of mTagBFP and sfGFPare 52,000 and 83,000 M–1 cm–1, respectively) result in highlysensitized emissions (Forster radius: 5.5 nm), and facilitate quan-titative measurements via direct ratio imaging.

To demonstrate the usefulness of the mTagBFP/sfGFP pair, weconstructed a FRET-based biosensor for Src kinase (denoted asBG-Src). The specific response element of BG-Src is a Src substratepeptide from p130cas and a SH2 domain from Src kinase and isthe same as the YC-based Src biosensor previously reported(Wang et al., 2005). During ratiometric imaging, the ratio changewas used to characterize the response of the BG-Src biosensor toSrc kinase. When expressed in HeLa cells, the BG-Src biosensorexhibited marked FRET response upon EGF stimulation, with a65% ratio change (Fig. 1A and B). The ratio change of the BG-Srcbiosensor was substantially larger than that of the initial Srcbiosensor based on ECFP/mCitrine (43%) (Wang et al., 2005) andwas comparable to the updated Src biosensor, which was basedon ECFP and circular permutated mVenus (77%) (Ouyang et al.,2008). Treatment with the Src-specific inhibitor PP1 readilyreversed the FRET response, indicating that BG-Src is specific toSrc kinase (Fig. 1C). The high photo-stability of mTagBFP andsfGFP ensures that the ratio of BG-Src biosensor is not distortedby UV illumination, as represented by the 405 nm laser in ourconfocal system (Fig. 1D).

Together, our results demonstrate that the mTagBFP/sfGFP paircan be used as a FRET biosensor.

3.2. mTagBFP/sfGFP and mVenus/mKOk FRET pairs are spectrally

compatible for dual-FRET imaging in living cells

To image two pairs of FRET biosensors in the same cellularlocation, it is imperative that imaging of one biosensor does not

ible in vivo. (A). Spectra of the B/G and Y/O FRET pairs. A 405 nm laser was used to

O-based biosensor. (B). Spectral bleed-though of the two FRET biosensors after

f the BG-Src biosensor does not affect the Y/O-based caspase-3 (YO-C3) biosensor,

e BG-Src biosensor. Stimulation of HeLa cells with EGF (100 ng/ml) changed the

ion with histamine (100 mM) changed the emission ratio YO-TnC without affecting

t sequence (NES) to the 30-end of the biosensor.

Fig. 3. Simultaneous monitoring of Src and Ca2þ signal in single living HeLa cells using dual biosensors. (A). Pseudo-color ratio images of HeLa cells expressing dual

biosensors (BG-Src and YO-TnC biosensors). EGF (100 ng/ml) was added at time 0. (B) Emission ratio (black curve) time course of the BG-Src biosensor and emission ratio

(red curve) time course of YO-TnC in (A). Scale bar: 10 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of

this article.)

T. Su et al. / Biosensors and Bioelectronics 46 (2013) 97–101100

affect imaging of the other. In the case of mTagBFP/sfGFP- andmVenus/mKOk-based FRET biosensors, spectral contamination isattributed to crosstalk between sfGFP and mVenus (Fig. 2A). Togain quantitative insights, we transfected one biosensor in HeLacells and measured bleed-through emissions into channels usedby the other biosensor. In the YO-Src biosensor, only 1% ofmVenus fluorescence bled into the sfGFP channel. However, inBG-Src biosensor, 7% of sfGFP fluorescence bled into the mVenuschannel (Fig. 2B). To determine the effect of this contamination onthe ratio of each biosensor, both B/G- and Y/O-based biosensorswere co-transfected into HeLa cells. Addition of EGF stimulated adramatic response from the BG-Src biosensor but the Y/O-basedcaspase-3 biosensor (YO-C3) remained stable (Fig. 2C), indicatingthat fluorescence contamination from BG-Src does not distort theratio of YO-C3. Similarly, a ratio change in the Y/O-based calciumbiosensor (YO-TnC) did not influence the ratio of the BG-Srcbiosensor (Fig. 2D). In summary, our results demonstrate that theB/G- and Y/O-based biosensors can work together to monitor dualmolecular events in the same cellular location.

3.3. Simultaneous imaging of Src and calcium signal in single living

cells using B/G-Src and Y/O-TnC biosensor, respectively

Upon binding to exogenous ligands, many plasma membranereceptors initiate multiple signaling pathways. In many cases,a complicated signaling network is activated, causing diverseeffects ranging from cytoskeletal changes to gene expressionalterations (Schlessinger, 2000). One example is EGF-inducedsignal transduction; it is well known that EGF can trigger Srcsignaling in cytosol and Ca2þ release from the ER into the cytosol(Oda et al., 2005). To gain insight into the dynamics of these twoprocesses, BG-Src and YO-TnC were co-transfected into HeLa cells(Supplemental Fig. S1). Upon EGF administration, cytosolic levelsof Ca2þ were robustly increased, as indicated by YO-TnC, peakingat approximately 2 min and then decreasing back to baselinelevels (Fig. 3). In contrast, Src signaling, as indicated by BG-Src,was sustained throughout the imaging period, consistent with itsrole in regulating long-term events such as STAT-mediated geneexpression and changes in cell morphology (Yeatman, 2004).In the other example, simultaneous imaging of cytoplasmic andplasma membrane Src signaling was achieved in single living cellsusing BG-Src and YO-Src biosensors, respectively (SupplementalFig. S2).

4. Discussion

Here, we demonstrate that a combination of FRET pairs, whichis based on mTagBFP/sfGFP and mVenus/mKOk FPs, allows for

simultaneous kinetic measurements of two molecular eventswithin the same subcellular location without the need for furtherimage correction. The traditional YC-based biosensors canbe easily converted to mTagBFP/sfGFP or mVenus/mKOk-basedbiosensors. We thus expect that more dual signaling events willbe simultaneously studied, providing previously unattainabledynamic molecular information.

Acknowledgments

We thank Dr. Atsushi Miyawaki for providing mVenus andmKO. The authors also thank the Analytical and Testing Center(Huazhong University of Science and Technology) for spectralmeasurements. This work was supported by the National BasicResearch Program of China (Grant no. 2011CB910401), ScienceFund for Creative Research Group of China (Grant no. 61121004),National Natural Science Foundation of China (Grant no. 81172153),and National Science and Technology Support Program of China(Grant no. 2012BAI23B02).

Appendix A. Supporting information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.bios.2013.02.024.

References

Ai, H.W., Hazelwood, K.L., Davidson, M.W., Campbell, R.E., 2008. Nature Methods 5(5), 401–403.

Campbell, R.E., 2009. Analytical Chemistry 81 (15), 5972–5979.Carlson, H.J., Campbell, R.E., 2009. Current Opinion in Biotechnology 20 (1), 19–27.Chudakov, D.M., Matz, M.V., Lukyanov, S., Lukyanov, K.A., 2010. Physiological

Reviews 90 (3), 1103–1163.Crivat, G., Taraska, J.W., 2012. Trends in Biotechnology 30 (1), 8–16.Day, R.N., Davidson, M.W., 2009. Chemical Society Reviews 38 (10), 2887–2921.DiPilato, L.M., Zhang, J., 2010. Current Opinion in Chemical Biology 14 (1), 37–42.Nagai, T., Tomosugi, W., Matsuda, T., Tani, T., Nemoto, T., Kotera, I., Saito, K.,

Horikawa, K., 2009. Nature Methods 6 (5), 351–353.Oda, K., Matsuoka, Y., Funahashi, A., Kitano, H., 2005. Molecular Systems Biology 1,

0010 2005.Ouyang, M., Huang, H., Shaner, N.C., Remacle, A.G., Shiryaev, S.A., Strongin, A.Y.,

Tsien, R.Y., Wang, Y., 2010. Cancer Research 70 (6), 2204–2212.Ouyang, M., Sun, J., Chien, S., Wang, Y., 2008. Proceedings of the National Academy

of United States of America 105 (38), 14353–14358.Pedelacq, J.D., Cabantous, S., Tran, T., Terwilliger, T.C., Waldo, G.S., 2006. Nature

Biotechnology 24 (1), 79–88.Schlessinger, J., 2000. Cell 103 (2), 211–225.Su, T., Zhang, Z., Luo, Q., 2012. Biosensors and Bioelectronics 31 (1), 292–298.Subach, O.M., Gundorov, I.S., Yoshimura, M., Subach, F.V., Zhang, J.H., Gruenwald,

D., Souslova, E.A., Chudakov, D.M., Verkhusha, V.V., 2008. Chemistry & Biology15 (10), 1116–1124.

T. Su et al. / Biosensors and Bioelectronics 46 (2013) 97–101 101

Wang, Y., Botvinick, E.L., Zhao, Y., Berns, M.W., Usami, S., Tsien, R.Y., Chien, S.,2005. Nature 434 (7036), 1040–1045.

Welch, C.M., Elliott, H., Danuser, G., Hahn, K.M., 2011. Nature reviews Molecularcell biology 12 (11), 749–756.

Yeatman, T.J., 2004. Nature Reviews Cancer 4 (6), 470–480.

Yu, H.B., Xiao, Y., 2012. Journal of Innovative Optical Health Sciences 5 (3), 145–149.Zhang, J., Campbell, R.E., Ting, A.Y., Tsien, R.Y., 2002. Nature Reviews Molecular

Cell Biology 3 (12), 906–918.Zhang, Z., Yang, J., Lu, J., Lin, J., Zeng, S., Luo, Q., 2008. Journal of Biomedical Optics

13 (1), 011006.