Open Access
8 August 2012 Förster resonance energy transfer photoacoustic microscopy
Author Affiliations +
Abstract
Förster, or fluorescence, resonance energy transfer (FRET) provides fluorescence signals sensitive to intra- and inter-molecular distances in the 1 to 10 nm range. Widely applied in the fluorescence imaging environment, FRET enables visualization of physicochemical processes in molecular interactions and conformations. In this paper, we report photoacoustic imaging of FRET, based on nonradiative decay that produces heat and subsequent acoustic waves. Estimates of the energy transfer efficiency by photoacoustic microscopy were compared to those obtained by fluorescence confocal microscopy. The experimental results in tissue phantoms show that photoacoustic microscopy is capable of FRET imaging with an enhanced penetration depth. Through its ability to three-dimensionally image tissue with scalable resolution, photoacoustic microscopy could be a beneficial biomedical tool to broaden the in vivo application of FRET.

1.

Introduction

A fluorophore in an excited state usually undergoes transitions of spontaneous emission, i.e., fluorescence, or fast nonradiative decay to the ground state. Förster, or fluorescence, resonance energy transfer (FRET) provides another mechanism for excited state decay in which energy is transferred from a donor fluorophore to an acceptor chromophore through nonradiative dipole-dipole coupling.1 The transfer rate varies inversely with the 6th power of the donor-acceptor distance with a half-maximum distance in the 1 to 10 nm range. Consequently, the FRET response can be used as a molecular ruler for measuring distances between biomolecules labeled with an appropriate donor and acceptor pair. The fluorescence quantum yield is determined by the ratio of the rate of fluorescence emission to the sum of all decay rates of the excited state. Therefore, fluorescence microscopy is well suited for FRET imaging.2,3

Photoacoustic (PA) microscopy is an emerging biomedical imaging modality based on detecting nonradiative decay.4 In the absence of photochemical decay, nonradiative decay converts the excited molecular state energy into heat. With pulsed light excitation, the heat induces thermoelastic expansion, generating acoustic waves in the medium. An ultrasonic transducer is used to detect the acoustic energy, which enables PA imaging of nonradiative decay processes.

FRET manifests through the reduction of donor fluorescence emission as a result of the nonradiative energy transfer to the acceptor, which increase both the fluorescence and PA emissions from the acceptor. The conversion efficiency of the transferred energy to photoacoustic emission depends on the quantum yield of the acceptor. In the case of a nonfluorescent acceptor, an increase of the PA signal strength becomes the solitary yield of the donor fluorescence quenching.

Compared with fluorescence techniques,5,6 PA imaging of FRET has inherent merits. By utilizing low ultrasonic scattering, PA imaging enables high-resolution, deeply penetrating imaging in biological tissue.79 Moreover, PA imaging is scalable with optical illumination and ultrasonic detection parameters. It can be designed to provide either submicron resolution at a shallow depth or a centimeter penetration depth while a high depth to resolution ratio is maintained.410 Therefore the ultimate applications of FRET in live cells up to organs in vivo can potentially benefit from the multiscale PA imaging capability. In this work, we demonstrate the initial experiments with PA imaging of FRET in solutions using a dual-modality PA and fluorescence confocal microscope, diagrammed in [Fig. 1(a)]. Absolute FRET efficiencies of solutions were quantified with and without intervening skin tissue.

Fig. 1

(a) Schematic of the dual-modality fluorescence and photoacoustic microscope. (b) Absorption spectra of Rhodamine 6G (donor) and DQOCI (acceptor) measured by a spectrophotometer, and emission spectrum of Rhodamine 6G measured by a fluorometer.

JBO_17_8_086007_f001.png

2.

Materials and Methods

Details about the imaging system design and performance have been published previously.11,12 The system employs a dye laser (CBR-D, Sirah) with tunable wavelengths in the range of 560 to 590 nm pumped by a 523 nm Nd:YLF laser (INNOSLAB, Edgewave); a microscope objective with an numerical aperture of 0.1; and a 600 nm long-pass filter for fluorescence emission imaging. The excited PA and fluorescence signals were detected by a 50 MHz ultrasonic transducer and a photomultiplier tube module, respectively.

FRET involves the nonradiative transfer of excited state energy from a fluorophore, the donor, to another nearby absorbing but not necessarily fluorescent molecule, the acceptor. The energy transfer efficiency E, which directly measures the fraction of photon energy absorbed by the donor that is transferred to the nearby acceptor, can be expressed as follows:1315

Eq. (1)

E=ktkf+kt+knr=[1+(r/R0)6]1,
where kf, kt, and knr are the donor fluorescence emission rate, the FRET rate from the donor to the acceptor, and other nonradiative decay rate of the donor, respectively; r is the intra- or inter-molecular distance; and R0 is the critical transfer distance characteristic of a given donor-acceptor pair, in the range of 1 to 10 nm. The distance r is an important parameter for describing biomolecules engaged in complex formation and conformational transition.

In our experiment, a pair consisting of a donor fluorophore (Rhodamine 6G) and a nonfluorescent acceptor chromophore (DQOCI) was examined. As observed with a fluorometer and a spectrophotometer, the emission spectrum of the donor overlaps considerably with the absorption spectrum of the acceptor [Fig. 1(b)], which is a requirement for a FRET pair. The energy transferred from the donor to the nonfluorescent acceptor increases the PA signal.

Förster’s theory predicts that, due to decreasing r, the energy transfer efficiency E increases with increasing acceptor concentration C.2,16 For an ensemble of donor and acceptor molecules, the FRET efficiency E is summed over the energy transfer channels to all acceptor molecules around the excited donor molecule, and is given by16,17

Eq. (2)

E=πxexp(x2)[1erf(x)],
where erf(x) is the error function, and x=C/C0. The critical concentration C0 is defined by C0=3/4πNAR03, where NA is Avogadro’s number and C0 corresponds to an average of one acceptor molecule in a sphere of radius R0. In our experiment, the concentrations of the acceptor varied between 104 and 103 molar, while the concentration of the donor was held constant as typically practiced in FRET solution studies.1820

The fluorescence quantum yields of donor Rhodamine 6G in the absence and presence of acceptor DQOCI, QD and QDA, are determined by the ratio of the fluorescence emission rate to the total decay rate, as given in Eqs. (3) and (4).

Eq. (3)

QD=kfkf+knr,

Eq. (4)

QDA=kfkf+knr+kt.

The subscripts D and A denote the presence of the donor and the acceptor, respectively. Therefore, the energy transfer efficiency E can be measured as

Eq. (5)

E=1QDAQD.
The fluorescence intensities F from the donor in the absence and presence of acceptor are given by

Eq. (6)

FDμa,D(λex)QDλexλf
and

Eq. (7)

FDAμa,D(λex)QDAλexλf,
where μa,D is the absorption coefficient of the donor at the excitation wavelength. The ratio of the excitation wavelength λex to the fluorescence emission wavelength λf accounts for the Stokes shift of fluorescence emission. For donor Rhodamine 6G, λexλf=523nm552nm=0.95.21,22

The PA amplitudes P can be calculated as

Eq. (8)

PDμa,D(λex)(1QDλexλf)
and

Eq. (9)

PDAμa,D(λex)(1QDAλexλf).
Note that PDA is attributed to the decay pathways via both kt and donor’s knr.

In fluorescence microscopy, the energy transfer efficiency E can be computed by

Eq. (10)

E=1FDAFD.
Similarly, the ratio of the PA amplitudes yields

Eq. (11)

PDAPD=1QDAλexλf1QDλexλf.

From Eqs. (5) and (11), one can compute the transfer efficiency as follows if the characteristic parameter QDλexλf of the donor is known:

Eq. (12)

E=(PDAPD1)(λfλexQD1).

The factor QDλexλf can be photoacoustically measured by a comparative method22 using a nonfluorescent dye with a known absorption coefficient as a standard sample. The PA amplitude Ps from a nonfluorescent dye, Direct Red 81, at λex can be calculated as

Eq. (13)

PS=Kμa,S(λex)I,
where K is a proportionality coefficient related to ultrasonic detection, μa,S(λex) is the optical absorption coefficient of the standard sample at λex, and I is the incident light intensity.

The PA amplitude PD at λex from the donor Rhodamine 6G is given by

Eq. (14)

PD=Kμa,D(λex)(1QDλexλf)I,
where μa,D(λex) is the optical absorption coefficient of the donor at λex.

Taking the ratio of Eqs. (14) to (13) yields the factor QDλexλf:

Eq. (15)

QDλexλf=1PDμa,S(λex)PSμa,D(λex).
By using the PA microscope and a spectrophotometer, QDλexλf is measured to be 0.83±0.03 for Rhodamine 6G at an excitation wavelength of 523 nm.

The strong spectral overlap of the donor and acceptor required for FRET imaging leads to the acceptor bleed-through (ABT) contamination, i.e., the direct excitation of acceptor at the donor’s excitation wavelength.23 For quantitative treatment of FRET, we use a dual-wavelength method to identify and remove the ABT background. Because the absorption spectrum usually falls steeply on its long wavelength side, an excitation wavelength can be selected for the acceptor such that the donor has negligible absorption. In our experiment, the PA images were acquired at two wavelengths of 523 and 580 nm. The 580 nm light selectively excites only the acceptor DQOCI and produces almost no PA signal from the donor Rhodamine 6G, as indicated by the spectra in Fig. 1(b). Aided by the absorption spectrum, one can calculate the ABT PA signal PABT at 523 nm from DQOCI as follows:

Eq. (16)

PABT(523nm)=P(580nm)μa,A(523nm)μa,A(580nm),
where P(580 nm) is the PA amplitude measurements in the FRET dye system at 580 nm; μa,A(523nm) and μa,A(580nm) are the absorption coefficients of acceptor DQOCI at 523 nm and 580 nm, respectively.

The corrected PA signal PDA at λex is obtained as follows:24

Eq. (17)

PDA(λex)=PDAraw(λex)PABT(λex),
where PDAraw(λex) is the raw PA amplitude measured from the FRET dye system at λex.

Based on solutions containing a controlled amount of donor Rhodamine 6G and acceptor DQOCI, the PA and fluorescence imaging of FRET efficiency was analyzed. Seven stock ethanol solutions were prepared with concentrations of donor Rhodamine 6G and acceptor DQOCI as tabulated in Fig. 2(a). The sample solutions were injected into 7 glass tubes (VWR; inner diameter: 0.56 mm; outer diameter: 0.8 mm) and sealed using epoxy to avoid evaporation of the ethanol. PA and fluorescence images were acquired using the integrated dual-modality PA and fluorescence confocal microscope.

Fig. 2

(a) Photograph of the tube phantom and tabulation of the donor (Rhodamine 6G or R6G) and acceptor (DQOCI) concentrations. (b) Fluorescence microscopic image acquired at 523 nm. (c) B-scan cross-sectional photoacoustic microscopic image of the tubes acquired at 523 nm. (d) En face photoacoustic microscopic image acquired at 523 nm. (e) B-scan cross-sectional photoacoustic microscopic image of the tubes acquired at 580 nm. (f) En face photoacoustic microscopic image acquired at 580 nm.

JBO_17_8_086007_f002.png

3.

Results

Figure 2(b) shows the fluorescence images acquired at 523 nm. While the donor Rhodamine 6G is excited, fluorescence emission is detected in the presence of different concentrations of acceptor DQOCI. Compared with a solution containing only donor Rhodamine 6G (tube No. 1), the mixtures containing both donor Rhodamine 6G and acceptor DQOCI have diminished fluorescence signals (tubes No. 3, 5 and 7). More acceptor DQOCI made fluorescence quenching more effective. Figure 2(c) and 2(d) shows a cross-sectional B-scan PA image and an en face maximum-amplitude-projection (MAP) PA image of the tubes, respectively, both of which were acquired at 523 nm. It can be seen that the quenching of donor Rhodamine 6G fluorescence leads to a higher PA signal (tubes No. 3, 5 and 7). In comparison, the ABT backgrounds are relatively weak (tubes No. 2, 4, and 6). As expected, the PA images acquired at 580 nm, pictured in Fig. 2(e) and 2(f), show that the solution containing only donor Rhodamine 6G has negligible absorption (tube No. 1), and the solutions with both Rhodamine 6G and DQOCI yield PA signals of the same levels as the solutions with an equal concentration of acceptor DQOCI only (tubes No. 2-7).

After subtraction of the ABT PA background using Eq. (17), the fluorescence and PA signals from the 0.5 mM pure donor Rhodamine 6G and the mixture solutions with different acceptor concentrations of 0.125, 0.25, and 0.5 mM acquired at 523 nm are plotted in Fig. 3(a) and 3(b). Based on the measurements of the donor Rhodamine 6G fluorescence emission in the presence and absence of the acceptor DQOCI, the FRET efficiencies calculated using Eq. (10) are shown in Fig. 3(c). The PA measurements of the FRET efficiencies based on Eq. (12) are shown in Fig. 3(d). The color-coded fluorescence and PA images shown in Fig. 3(c) and 3(d) clearly display the increase of FRET efficiency due to the increase of the acceptor concentration. The fluorescence and PA images of FRET efficiencies agree to each other with a correlation coefficient of 0.91. The systematic difference between the fluorescence and PA measurements possibly came from nonlinear effects, which were not modeled in Eqs. (6) and (8), but could be compensated for by calibrating the imaging system.

Fig. 3

(a) Fluorescence and (b) photoacoustic signals versus acceptor concentration showing the FRET effect. FRET efficiency map acquired using (c) the fluorescence microscope, and (d) the photoacoustic microscope.

JBO_17_8_086007_f003.png

To test the feasibility of imaging in biological tissue, freshly excised rat skin tissue was overlaid on the tubes, as shown in Fig. 4(a), and the same imaging procedure was repeated. The detected tubes in the fluorescence confocal image, pictured in Fig. 4(b), were severely blurred due to light scattering in the tissue. The PA cross sectional B-scan image of the tubes in Fig. 4(c) shows that the overlaid mouse skin had a thickness of 1mm. Again, the dual-wavelength PA images in Fig. 4(c) through 4(f) clearly show FRET quenching of the donor Rhodamine 6G fluorescence by acceptor DQOCI. Figure 4(g) shows the FRET efficiency map of the tissue phantom calculated from the PA image, whereas a similar map cannot be generated from Fig. 4(b) due to its poor signal-to-noise ratio. It has a correlation coefficient of 0.90 with the fluorescence measurements as shown in Fig. 3(c). The results demonstrate the potential for in vivo FRET imaging using the PA method.

Fig. 4

(a) Photograph of the tube phantom with overlaid mouse skin tissue in the same region as shown in Fig. 2(a). The tubes below are invisible due to tissue scattering. (b) Fluorescence microscopic image acquired at 523 nm. (c) B-scan cross-sectional photoacoustic microscopic image of the tubes acquired at 523 nm. (d) En face photoacoustic microscopic image acquired at 523 nm. (e) B-scan cross-sectional photoacoustic microscopic image of the tubes acquired at 580 nm. (f) En face photoacoustic microscopic image acquired at 580 nm. (g) FRET efficiency map acquired using the photoacoustic microscope.

JBO_17_8_086007_f004.png

4.

Conclusions

In conclusion, PA microscopy has been used to image FRET efficiencies through a 1-mm-thick skin tissue. Based on the relative increase of PA signals, absolute FRET efficiency can be quantified. Compared with confocal microscopy, PA microscopy offers better penetration into scattering biological tissue. Analogous to fluorescence imaging of FRET, PA imaging has to consider potential sources of artifacts, such as the bleed-through PA signals and excitation light fluence attenuation in tissue.25 Future work will extend PA FRET imaging to in vivo animal studies.

Acknowledgments

This work was sponsored in part by National Institutes of Health grants R01 EB000712, R01 EB008085, R01 CA134539, U54 CA136398, R01 CA157277, and R01 CA159959. L.W. has a financial interest in Microphotoacoustics, Inc. and Endra, Inc., which, however, did not support this work.

References

1. 

X. F. WangB. Herman, Fluorescence Imaging Spectroscopy and Microscopy, John Wiley & Sons Inc., New York (1996). Google Scholar

2. 

T. Forster, “Energy migration and fluorescence,” J. Biomed. Opt., 17 011002 (2012). http://dx.doi.org/10.1117/1.JBO.17.1.011002 JBOPFO 1083-3668 Google Scholar

3. 

A. PeriasamyS. S. VogelR. M. Clegg, “FRET 65: a celebration of Forster,” J. Biomed. Opt., 17 011001 (2012). http://dx.doi.org/10.1117/1.JBO.17.1.011001 JBOPFO 1083-3668 Google Scholar

4. 

L. V. WangS. Hu, “Photoacoustic tomography: in vivo imaging from organelles to organs,” Science, 335 (6075), 1458 –1462 (2012). http://dx.doi.org/10.1126/science.1216210 SCIEAS 0036-8075 Google Scholar

5. 

V. Gaindet al., “Deep-tissue imaging of intramolecular fluorescence resonance energy-transfer parameters,” Opt. Lett., 35 (9), 1314 –1316 (2010). http://dx.doi.org/10.1364/OL.35.001314 OPLEDP 0146-9592 Google Scholar

6. 

V. Gaindet al., “Towards in vivo imaging of intramolecular fluorescence resonance energy transfer parameters,” J. Opt. Soc. Amer. A, 26 (8), 1805 –1813 (2009). http://dx.doi.org/10.1364/JOSAA.26.001805 JOAOD6 1084-7529 Google Scholar

7. 

S. Mallidiet al., “Multiwavelength photoacoustic imaging and plasmon resonance coupling of gold nanoparticles for selective detection of cancer,” Nano Lett., 9 (8), 2825 –2831 (2009). http://dx.doi.org/10.1021/nl802929u NALEFD 1530-6984 Google Scholar

8. 

Y. Wanget al., “In vivo three-dimensional photoacoustic imaging based on a clinical matrix array ultrasound probe,” J. Biomed. Opt., 17 061208 (2012). http://dx.doi.org/10.1117/1.JBO.17.6.061208 JBOPFO 1083-3668 Google Scholar

9. 

D. RazanskyC. VinegoniV. Ntziachristos, “Multispectral photoacoustic imaging of fluorochromes in small animals,” Opt. Lett., 32 (19), 2891 –2893 (2007). http://dx.doi.org/10.1364/OL.32.002891 OPLEDP 0146-9592 Google Scholar

10. 

L. V. Wang, “Multiscale photoacoustic microscopy and computed tomography,” Nat. Photonics, 3 (9), 503 –509 (2009). http://dx.doi.org/10.1038/nphoton.2009.157 1749-4885 Google Scholar

11. 

Y. Wanget al., “In vivo integrated photoacoustic and confocal microscopy of hemoglobin oxygen saturation and oxygen partial pressure,” Opt. Lett., 36 (7), 1029 –1031 (2011). http://dx.doi.org/10.1364/OL.36.001029 OPLEDP 0146-9592 Google Scholar

12. 

Y. Wanget al., “Integrated photoacoustic and fluorescence confocal microscopy,” IEEE Trans. Biomed. Eng., 57 (10), 2576 –2578 (2010). http://dx.doi.org/10.1109/TBME.2010.2059026 IEBEAX 0018-9294 Google Scholar

13. 

C. BerneyG. Danuser, “FRET or no FRET: a quantitative comparison,” Biophys. J., 84 (6), 3992 –4010 (2003). http://dx.doi.org/10.1016/S0006-3495(03)75126-1 BIOJAU 0006-3495 Google Scholar

14. 

R. M. Clegg, “Fluorescence resonance energy transfer,” Curr. Opin. Biotechnol., 6 103 –110 (1995). http://dx.doi.org/10.1016/0958-1669(95)80016-6 CUOBE3 0958-1669 Google Scholar

15. 

E. A. Jares-ErijmanT. M. Jovin, “FRET imaging,” Nat. Biotechnol., 21 (11), 1387 –1395 (2003). http://dx.doi.org/10.1038/nbt896 NABIF9 1087-0156 Google Scholar

16. 

T. Forster, “Transfer mechanisms of electronic excitation energy,” Radiation Res. Supp., 2 326 –339 (1960). http://dx.doi.org/10.2307/3583604 RARSAM 0485-8611 Google Scholar

17. 

B. MeerG. CokerS.-Y. Chen, Resonance energy transfer : theory and data, VCH publishers, Inc., New York (1994). Google Scholar

18. 

M. C. Adamset al., “Real-time picosecond measurements of electronic-energy transfer from DODCI to Malachite green and DQOCI,” Chem. Phys. Lett., 66 (3), 428 –434 (1979). http://dx.doi.org/10.1016/0009-2614(79)80310-3 CHPLBC 0009-2614 Google Scholar

19. 

J. B. BirksK. N. Kuchela, “Energy transfer in organic systems II: solute-solute transfer in liquid solutions,” Proc. Phys. Soc., 77 (5), 1083 (1961). 0370-1328 Google Scholar

20. 

A. KawskiE. KutenJ. Kaminski, “Fluorescence quenching and nonradiative energy-transfer in solutions,” J. Phys. B-Atomic Molec. Opt. Phys., 6 (9), 1907 –1916 (1973). http://dx.doi.org/10.1088/0022-3700/6/9/034 0953-4075 Google Scholar

21. 

J. M. DixonM. TaniguchiJ. S. Lindsey, “PhotochemCAD 2: a refined program with accompanying spectral databases for photochemical calculations,” Photochem. Photobiol., 81 (1), 212 –213 (2005). http://dx.doi.org/10.1562/2004-11-06-TSN-361.1 PHCBAP 0031-8655 Google Scholar

22. 

J. N. Miller, “Relative fluorescence quantum yields using a computer-controlled luminescence spectrometer,” Analyst, 108 (1290), 1067 –1071 (1983). http://dx.doi.org/10.1039/an9830801067 ANLYAG 0365-4885 Google Scholar

23. 

A. Periasamy, “Fluorescence resonance energy transfer microscopy: a mini review,” J. Biomed. Opt., 6 (3), 287 –291 (2001). http://dx.doi.org/10.1117/1.1383063 JBOPFO 1083-3668 Google Scholar

24. 

Y. Chenet al., “Characterization of spectral FRET imaging microscopy for monitoring nuclear protein interactions,” J. Microsc., 228 (Pt. 2), 139 –152 (2007). http://dx.doi.org/10.1111/jmi.2007.228.issue-2 JMICAR 0022-2720 Google Scholar

25. 

B. T. Coxet al., “Two-dimensional quantitative photoacoustic image reconstruction of absorption distributions in scattering media by use of a simple iterative method,” Appl. Opt., 45 (8), 1866 –1875 (2006). http://dx.doi.org/10.1364/AO.45.001866 APOPAI 0003-6935 Google Scholar
© 2012 Society of Photo-Optical Instrumentation Engineers (SPIE) 0091-3286/2012/$25.00 © 2012 SPIE
Yu Wang and Lihong V. Wang "Förster resonance energy transfer photoacoustic microscopy," Journal of Biomedical Optics 17(8), 086007 (8 August 2012). https://doi.org/10.1117/1.JBO.17.8.086007
Published: 8 August 2012
Lens.org Logo
CITATIONS
Cited by 40 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Luminescence

Fluorescence resonance energy transfer

Rhodamine

Photoacoustic microscopy

Photoacoustic spectroscopy

Absorption

Resonance energy transfer

Back to Top