Letters

Modulating Fluorescence Resonance Energy Transfer in ConjugatedLiposomes

Xuelian Li, Matthew McCarroll, and Punit Kohli*

Department of Chemistry and Biochemistry, Southern Illinois UniVersity, Carbondale, Illinois 62901

ReceiVed May 11, 2006. In Final Form: August 17, 2006

We report here a novel system where the rate of energy transfer is based on changes in the spectral overlap betweenthe emission of the donor and the absorption of the acceptor (J) as well as changes in the quantum yield of the acceptor.We use the fluorophore dansyl as the donor and polydiacetylene (PDA) as the acceptor to demonstrate the modulationof FRET through conformationally induced changes in the PDA absorption spectrum following thermal treatment thatconverts the PDA backbone of the liposome from the blue form to the red form. Energy transfer was found to besignificantly more efficient from dansyl to the red-form PDA. These findings support the basis of a new sensingplatform that utilizesJ-modulated FRET as an actuating mechanism.

Fluorescence resonance energy transfer (FRET) is a processwhereby energy is nonradiatively transferred from a donormolecule to an acceptor molecule through long-range dipoleinteractions. Recently, FRET has been exploited for manyapplications including DNA sensing,1a-c proteomics,1d andprobing various processes in living cells.1e,f These applicationsall rely on the remarkable dependence of the FRET efficiencyon the donor-acceptor separation distance. Most FRET-basedsensing schemes are based on the modulation of the separationdistance to trigger a sensing response. The use of the other factorsin the Forster equation, such as spectral overlap and donor-acceptor dipole orientation, have been underutilized in thedevelopment of sensing systems.2 The goal of the work reportedhere is to investigate FRET (the feasibility of building sensingsystems) based primarily on the modulation of the spectral overlapand acceptor quantum yield.

The rate of energy transfer,kET, between a donor and anacceptor molecule is given by

whereK is a constant,k2 is an orientation factor between donorand acceptor molecules,Q is the quantum yield of the donor,Jis the overlap between the donor emission spectrum and absorptionacceptor spectrum, andr is the donor-acceptor distance.3 Thewide success of FRET-based sensing systems is based on ther6

dependence ofkET that results in extremely large differencesbetween the “on” and “off” states of the sensor. Although thedistance dependence is the dominant term in eq 1,kET is alsolinearly dependent onJ, Q, andk2. We report here a novel systemthat utilizes changes inJ and changes in the quantum yield ofthe acceptors for fluorescence amplification for the rate of energytransfer between donor and acceptor. We use the fluorophoredansyl as the donor and polydiacetylene (PDA) as the acceptorto demonstrate the modulation of FRET through conformationallyinduced changes in the PDA absorption spectrum.

Polydiacetylenes are intriguing conjugated polymers becausetheir electronic absorption spectrum changes dramatically in

* Corresponding author. E-mail: pkohli@chem.siu.edu.(1) (a) Tyagi, S.; Kramer, F. R.Nat. Biotechnol.1996, 14, 303-308. (b)

Zhang, P.; Beck, T.; Tan, W.Angew. Chem., Int. Ed.2001, 40, 402-405. (c) Liu,B.; Bazan, G. C.Proc. Natl. Acad. Sci. U.S.A.2005, 102, 589-593. (d)Kumaraswamy, S.; Bergstedt, T.; Shi, X.; Rininsland, F.; Kushon, S.; Xia, W.;Ley, K.; Achyuthan, K.; McBranch, D.; Whitten, D.Proc. Natl. Acad. Sci. U.S.A.2004, 101, 7511-7515. (e) Zal, T.; Gascoigne, N. R. J.Biophys. J.2004, 86,3923-39. (f) Verveer, P. J.; Wouters, F. S.; Reynolds, A. R.; Bastiaens, P. I. H.Science2000, 290, 1567-1570.

(2) Clapp, A. R.; Medintz, I. L.; Mauro, J. M.; Fisher, B. R.; Bawendi, M. G.;Mattoussi, H.J. Am. Chem. Soc.2004, 126, 301-310.

(3) Lakowicz, J. R.Principles of Fluorescence Spectroscopy; Kluwer Academic/Plenum: New York, 1999; Chapter 13.

kET ) Kk2QJ

r6(1)

© Copyright 2006American Chemical Society

OCTOBER 10, 2006VOLUME 22, NUMBER 21

10.1021/la061340m CCC: $33.50 © 2006 American Chemical SocietyPublished on Web 09/13/2006

response to stress applied to the conjugated ene-yne backboneand can exist in either a blue or red form. These polymeric systemshave been used as liposome-based colorimetric sensors for thedetection of influenza viruses,4a cholera toxins,4b antimicrobialpeptides,4c and antibody-antigen interactions.4d In this work,we have chosen to use PDA as the FRET acceptor because thedonor-acceptor overlap integral will be modulated in responseto changes in the absorption spectrum. As a proof of concept,the conversion from the blue form to the red form can beaccomplished by heating.5 We have chosen dansyl and PDA asthe donor-acceptor pair for the following reasons: (1) Thespectral overlap between dansyl emission and PDA absorptionin its blue form is relatively small and increases significantly inthe red form when PDA is transformed into its red form (Figure1B). (2) Dansyl emission is sensitive to the local environment.3

This solvent-sensitive emission property of dansyl fluorophorescan be utilized to probe the local environment of bilayer PDAliposomes during heating of the PDA liposomes. (3) Thebackground emission from the blue form of PDA is very lowbecause of an extremely small quantum yield (∼10-4). Impor-tantly, in the red form, the quantum yield of PDA increasesnearly 2 orders of magnitude to a value of 0.02.6 These changesin the quantum yield of PDA increase the value of this emissionshift in sensing applications.

We have produced a FRET platform with a relatively fixeddonor-acceptor distance by synthesizing bilayered liposomesfollowing the self-assembly of a mixture of dansyl-tagged 10,12-pentacosdiyoic acid (1) and 10,12-pentacosdiyoic acid (2)monomers (Figure 1A).8 Dialysis results indicate that>95% ofall dansyl-tagged fluorophores were covalently bound in theliposomes.8

Figure 2A shows the absorption spectrum of PDA at varioustemperatures. The dominant peak centered at 630 nm resultsfrom the blue-form PDA at 298 K.4,5 As the temperature of thePDA solution is increased, the intensity of the peak at 630 nmdecreased while two distinct peaks (centered at 495 and 540 nm)

whose intensities grow with increasing temperature of the solutionwere observed. In fact, we observe an isosbestic point at awavelength of 555 nm (Figure 2A) that strongly suggests thatthere are two species and that thermal-induced stress leads to theconversion of the PDA blue form to the PDA red form. Theseobservations are consistent with previous reports.5 The overlapbetween the absorption spectrum of PDA and the emissionspectrum of dansyl (J) increases with thermal treatment, ultimatelyleading to an increase in the FRET efficiency (kET) from dansylto PDA (eq 1). For example, we have experimentally determinedthat the overlap intergral (J) increases from a value of 1.1× 1010

M-1 cm-1 nm4 in the blue form to 2.52× 1010 M-1 cm-1 nm4

in the red form, following thermal treatment. This represents an∼130% increase in theJ value.7

Dansyl is a solvent-polarity-sensitive probe whose emissionwavelength can provide useful information regarding its localenvironment in the liposome bilayers.3 The emission spectrumfor the dansyl-containing liposomes is an anomalously broadpeak centered at 484 nm, indicating that the dansyl fluorophoresare located in the liposome solution with environments rangingfrom nonpolar to those of moderate polarities (Figure 2B).Followingconversion to the red form, thebroademissionspectrumbecomes more well defined with peaks appearing at 465, 519,and 560 nm (Figure 2B). The peaks at 465 and 519 nm can beattributed to the dansyl moiety existing in one of two well-defined environments of differing polarities, corresponding tothose of hexane and ethanol, respectively. The peak at 560 nmcorresponds to the emission of the red-form PDA and can betentatively attributed to the FRET process. The magnitude of theincreased intensity of the peak at 560 nm is dependent on thetemperature of the thermal treatment. In fact, there was a modest

(4) (a) Charych, D. H.; Nagy, J. O.; Spevak, W.; Bednarski, M. D.Science1993, 261, 585-588. (b) Pan, J.; Charych, D.Langmuir1997, 13, 1365-1367.(c) Kolusheva, S.; Boyer, L.; Jelinek, R.Nat. Biotechnol.2000, 18, 225-227.(d) Kolusheva, S.; Kafri, R.; Marina, K.; Jelinek, R.J. Am. Chem. Soc.2001, 123,417-422.

(5) (a) Okada, S.; Peng, S.; Spevak, W.; Charych, D.Acc. Chem. Res.1998,31, 229-239. (b) Hofmann, U. G.; Peltonen, J.Langmuir2001, 17, 1518-1524.(c) Carpick, R. W.; Mayer, T. M.; Sasaki, D. Y.; Burns, A. R.Langmuir2000,16, 1270-1278.

(6) Olmsted, J., III.; Strand, M.J. Phys. Chem.1983, 87, 4790-4792.

(7) The overlap function,J(λ), between dansyl fluorophore emission and PDAabsorption (Figure 1) is calculated using the following equation (see chapter 13of ref 3 for more details):J(λ) ) ∫ F(λ) ε(λ)λ4δλ/∫ F(λ)δλ, whereF(λ) is thefluorescence intensity of the donor (dimensionless),ε(λ) is the extinction coefficient(M-1 cm-1), andλ is the wavelength (nm).J(λ) is in M-1 cm-1 nm.4

(8) Supporting information accompanying this communication.

Figure 1. (A) Structures of two monomers used in this study. (B)Corrected normalized absorption absorption spectra of blue- (bluecurve) and red-form (red curve) PDA liposome solutions (1 mM)at 298 K. The green curve is the emission spectrum of the PDAliposome solution composed of monomers1and2 (mol ratio [2]/[1]) 1000, [monomertotal] ) 1 mM).

Figure 2. (A) UV-vis and (B) fluorescence responses (λex ) 337nm) of a liposome solution composed of ([2]/[1] ) 1000, [mono-mertotal] ) 1 mM) for different thermal treatments. The insets in Aare photographs of vials containing blue- and red-form PDA. UV-vis and fluorescence spectra were taken at 298 K by first heatingthe liposome solutions to a given temperature for 10 min followedby equilibrating the solution to 298 K for 10 min. The dotted curvein B is the emission spectrum of the same solution but withoutdansyl fluorophores.

8616 Langmuir, Vol. 22, No. 21, 2006 Letters

increase (∼2×) when the solution was treated at 328 K, whereasthe intensity increased 10-fold after treating the liposome solutionat 343 K (Figure 2B).

The peak centered at∼560 nm is attributed to the PDAemission5b,c,9 following the FRET from dansyl donors to PDAacceptors (vide infra). To support this conclusion, we performedtwo control experiments. First, the direct excitation at 470 nmof liposomes shows an emission peak centered at∼560 nm thatincreases in intensity following incremental thermal treatmentof the solution (Figure 1S, Supporting Information). Theliposomes were derived from monomers of [1] and [2] (mol ratio[2]/[1] ) 1000, [monomertotal] ) 1 mM). There is reasonableassurance that dansyl does not contribute to the observedfluorescence because of its extremely low molar extinctioncoefficient at 470 nm10and the fact that the 560 nm peak (Figure2B) is indeed from FRET-derived PDA fluorescence. Second,when the same liposome solution without dansyl is excited ina manner similar to that used in the FRET experiment (λex) 337nm), the direct PDA emission was∼20 times smaller than thatof the similar solution containing both donor and acceptor (Figure2B). These controlled experiments strongly support the conclusionthat the emission peak at 560 nm is from PDA emission as aresult of the FRET process.

Interestingly, in our FRET experiments the changes in theemission intensity of the donor and acceptor were not proportional,as is the case in typical FRET. Following temperature treatmentsat increasing temperatures, the donor emission decreased to aconstant value at 328 K, whereas the acceptor emission continuedto increase up to 343 K (Figure 2B), indicating that an additionalprocess affected the acceptor emission. With all other parametersbeing constant, the increase inJ cannot alone account for the>10 times increase in the acceptor intensity. It was previouslyreported that the temperature treatment of a PDA solution atT> 320 K results in a considerable increase in the fraction ofred-form PDA at the expense of the blue form9 and that thequantum yield of PDA is more than 2 orders of magnitude greaterin the red form.6Our observations are consistent with the literature,and we believe the large increase in the quantum yield of the redPDA partially contributes to the dramatic increase in the 560 nmemission peak. It should be noted that we have not specificallyexamined the effect of the donor-acceptor distance,r, and thedonor quantum yield dependence on FRET in this system. Ourpreliminarily TEM data suggests that diacetylene liposomes aremechanical robust up to 343 K, hence we can expect the separationdistance to remain relatively constant in a given liposome system.To investigate this possible effect, we are now designing

fluorophore-tagged diacetylene monomers where we can sys-tematically change the donor-acceptor separation distance.

We now discuss an issue of FRET versus direct PDA excitationfor sensing applications. The emission of a fluorophore dependson the extinction coefficient, quantum yield of fluorophore, andinstrumental parameters such as the light source, detector, andoptics as well. Considering the extremely low quantum yield(∼0.02) and moderate extinction coefficient of the red-PDA form(∼4000 M-1 cm-1, depending upon polymerization conditionssuch as polymerization time, presence of oxygen in the solution,etc.), the emission intensity from PDA through direct PDAexcitation would be extremely low, especially for concentration<0.01 mM. The amplification of fluorescence can be achievedby the FRET mechanism where the donor can transfer energyto PDA, which will significantly improve the sensitivity of thesensor. There also exists opportunities for sensors based onenhanced fluorescence amplification where fluorescence am-plification is through both FRET (from donor to PDA) and directPDA excitation mechanisms. We will report on the sensors basedon these strategies in a future publication.

Even though successful colorimetric sensors have beendeveloped from PDA that are based on stress-induced spectralshifts,4 an opportunity exists to develop fluorescence-based PDAbased sensors of superior sensitivity that utilizeJ-modulatedFRET as an actuating mechanism. On the basis of these concepts,we are now developing FRET-based assays for highly specific(bio)chemical sensing applications in which fluorophore-taggedPDA liposomes will contain the appropriate receptors (such asantibodies, enzymes, etc.). Analyte-receptor binding will perturbthe FRET efficiency, leading to sensitive indirect detection.

In conclusion, the use ofJ to modulate FRET efficiency ina model liposome system has been investigated. We found thatan increase in FRET resulted in temperature-induced changes inJ, as was predicted. The observed increases in FRET could not,however, be fully explained by changes inJ and were partiallyattributed to the higher quantum yield of red-form PDA.

Acknowledgment. We thank Professor Lori Vermulen forthe use of her pulsed sonicator and Professors Dan Dyer andBoyd Goodson for critically reading the manuscript. We thankthe National Science Foundation (Grant 0609349), MTC, andORDA at SIUC for partial funding of this work and the IMAGEcenter (SIUC) for assisting with electron microscopy studies.NMR spectra were collected using an NMR spectrometerpurchased through an NSF grant (MRI-0421012).

Supporting Information Available: Details of the synthesisand characterization of monomers and liposomes (Figures 1S-3S andScheme 1S). This material is available free of charge via the Internetat http://pubs.acs.org.

LA061340M

(9) (a) Carpick, R. W.; Mayer, T. M.; Sasaki, D. Y.; Burns, A. R.Langmuir2000, 16, 4639-4647.

(10) The extinction coefficient of dansyl-amide in water at 470 nm is∼18 M-1

cm-1 (Chen, R. F.Anal. Biochem.1968, 25, 412-416).

Letters Langmuir, Vol. 22, No. 21, 20068617