Fluorescent beads (nanoparticles, nanospheres) are commonly used in fluorescence spectroscopy and microscopy. Due to the random distribution of dye and high dye to nanoparticle ratio, the fluorescence polarization observed from the beads is low. Therefore beads are not used for polarization study. We demonstrate that photoselective bleaching creates beads with highly polarized fluorescence. First, the beads were immobilized in a PVA polymer. Second, the beads-doped PVA film was exposed to the illumination within the dye absorption band. A progressive decrease of absorption was observed. Next, photophysical properties of photobleached and not bleached films dissolved in water were compared.
©2010 Optical Society of America
For decades, emission methods proved its utility for observing the locations and movement of molecules. High resolution techniques, like X-ray crystallography and Nuclear Magnetic Resonance provide exact positions and/or structural details of molecules and larger assembles but they are “blind” to any dynamic changes occurring in time. In contrast, fluorescence can be conveniently used in a time-resolved mode reporting changes occurring in a nanosecond time scale. In addition, fluorescence measures signals above a low background. Another advantage is a non-invasive detection. A fluorophore lifetime carries information not only about the fluorescent species itself but also about surrounding environment conditions. Therefore fluorescent dyes are often used to label macromolecules, such as proteins or DNA to study their movements and conformational changes. At present days fast development in optical imaging and detection make it possible to use fluorescence microscopy on live biological subunits even at the single molecule level detection [1–7]. Number of techniques supports observation of dynamics of individual molecules. Emission polarization or anisotropy is one of them. The usefulness of this method relies on ratiometric intensities measurements of two orthogonally polarized fluorescence signals following polarized excitation of the sample [8–10]. In addition to the requirements of maintaining proper polarization conditions in a polarization microscope, there is a need for extended molecule observation time. A handful of researchers have already demonstrated that application of deoxygenating compounds can prolong the time of fluorophore observation [11–14]. Alternatively, more stable emitter systems like quantum dots or microsphers (beads) can be used. The polymer-core nanoparticles with immobilized dyes proved their utility in studding live biological systems with dynamic flow tracking because of their spectral properties with usually high quantum yield, extinction coefficient and photostability. Nanospheres are widely employed now in the tissue imaging , biotechnology , as temperature sensors and have been used in a variety of applications that include diagnostics and biological assays. However a high local density of fluorescent molecules enforces a depolarization of observed fluorescence. As a result, these fluorescent nanoparticles cannot be studied using polarization methods.
In this paper, we present a method for preparation of highly polarized fluorescence beads. The photoselective bleaching eliminates fluorophores with transition moments along the electric vector of impinging light. The molecules with transition moments oriented along the direction of propagation of the light (perpendicular to the electric vector) are not excited and not photobleached. In effect, after sufficient photobleaching the fluorescent beads are highly polarized. To accomplish this goal we bleached fluorescent beads-doped PVA film with not polarized intensive light and dissolve later from polymer matrix into solution. We employed the steady state anisotropy, anisotropy decay and polarized Fluorescence Cross Correlation Spectroscopy (FCCS), which are the methods sensitive to the orientation of the transition dipole moment of the molecule and allow determining its changes under molecule movements.
We believe that bright polarized fluorescent nanoparticles will find applications in polarization assays of large macromolecules.
2. Materials and methods
Dark red fluorescent (660/680 nm) carboxylate modified microspheres (nanoparticles, beads) 0.02μm in diameter (lot: 426844) were from Invitrogen (Eugene, Oregon, USA). Low molecular weight (Mw 9,000-10,000 g/mol) PVA (polyvinyl alcohol) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Triton X-100 reduced form applied in 1% concentration of total amount of solution was obtained from Fluka (93424). Water used in all the experiments was deionized made up using a Milli-Q Synthesis A10 system produced by Millipore.
2.2 Sample preparation for optical studies
Fluorescence microspheres were fixed in 30% PVA film. PVA aqueous solution was prepared using standard method  by dissolving the powder in water heated to about 100°C under stirring. The mixture of nanospheres with PVA were poured to Petrie dishes and left for drying. Then, we measured absorption spectra of the fluorescent nanospheres-doped PVA film. The PVA strip without nanoospheres was used as a reference. Absorption spectra were recorded for beads before and after progressive photobleaching.
Next, we dissolved the same amount of bleached and not bleached films in deionized water (0.025g/1ml) and after addition of 1% of Triton X and 10min. sonication (to avoid aggregation of the beads) used these two samples for spectroscopic measurements. FSC experiment was performed for the adequately diluted samples.
2.3 Steady-state measurements
The absorption spectra of red fluorescent beads were measured using a single beam Cary 50 Bio spectrometer (Varian Inc., Australia). The emission spectra of beads in the water were measured using Varian Cary Eclipse spectrofluorometer with excitation of 635nm. The spectrofluorometer was equipped with polarizer; (Manual Polarizer Accessory, Varian Inc., Australia) adjusted to proper positions for excitation and emission paths. We observed a remarkably high stability of dissolved photobleached nanospheres stored in the refrigerator at 4°C. One month after the preparation they still showed the same intensity and anisotropy.
2.4 Time-resolved measurements
Fluorescence decays and anisotropy study were performed using time-domain method implemented in FluoTime200 (PicoQuant GmbH, Berlin, Germany) equipped with a Hamamatsu microchannel plate detector (MCP). The emission monochromator was at 685nm and the Glan-Tylor polarizer was set to the proper conditions under measurements. For all the experiments we were using long pass wavelength glass filter 665nm. The excitation source was a 635nm pulsed laser diode driven by a PDL800 driver (PicoQuant GmbH, Berlin, Germany). Time-resolved fluorescence data were analyzed using the FluoFit software package (version 4.2.1, PicoQuant GmbH).
Fluorescence intensity decay data were fitted by the iterative convolution to the sum of exponents:
For anisotropy study, two fluorescence intensity decays were measured: III(t) and I┴?(t) for each sample solutions with either oriented or isotropic dyes in the beads. Relative to the vertical laser light polarization analyzing polarizer was set to vertical and horizontal orientation, respectively. From these decays, the anisotropy as a function of time was determined:18] using Cresyl Violet in methanol as a reference with near zero anisotropy. The anisotropy decay data were fited to the multiexponential function:
2.5 FCS (microscopy) measurements
The FCS (Fluorescence Correlation Spectroscopy) measurements were performed with MicroTime 200 (Picoquant GmbH, Berlin Germany), a time-resolved fluorescence microscope. As a source of light we used 635 pulsed diode laser (LDH-P-C-635B) driven by PDL 828 (Picoquant GmbH, Berlin, Germany) driver and operated at 20MHz repetition rate. Excitation was pre-cleaned (Chroma Technology bandpass filter z636/10x) and via single mode fiber optics coupled to the optical module. The excitation was naturally, linearly polarized from laser but additional horizontal polarizer was applied. After that, light was sent to the side port of Olympus IX71 inverted microscope and focused 10µm above the coverslip (Menzel-Glaser #1) inside a sample drop volume. We used Olympus UPlanSApo 60x magnification water objective, NA=1.2 to focus the excitation light and gather fluorescence. For the emission detection, a 647 razor long wavelength pass filter (Semrock LP02-647RS) was applied blocking the excitation light. Confocal type of measurements was achieved with 30µm pinhole. For the purpose of suppression afterpulsing influence effects typically observed with single photon avalanche photodiodes, the Fluorescence Lifetime Correlation Spectroscopy [19, 20] was adopted to analyze the autocorrelation of the time trace of fluorescence. The fluorescence light was split with polarizing beam splitter that splits light into orthogonal components detected by two identical Micro Photon Devices (PDM 1CTC) Avalanche Photo Diodes. Photon stream was collected by PicoHarp300 time-correlated single-photon counting module and data analysis (correlation) was performed with SymPhoTime (v. 220.127.116.11) software (Picoquant GmbH, Berlin, Germany). The experiments were typically performed with a laser power of 2µW in a focus.
Self similarity of the fluorescence fluctuation signal received form nanoparticle freely diffusing in the confocal volume can be extracted using autocorrelation analysis. The normalized fluorescence correlation function is defined by the following mathematical function [21–23]:
Auto- and cross-correlation curves due to translational diffusion through 3-dimentional Gaussian shaped volume were fitted with following expression:
3. Results and discussion
3.1 Absorption and fluorescence of nanospheres-doped PVA films
The aqueous PVA solution containing red nanospheres was dried in Petrie dish at room temperature for four days. For spectral measurements a strip 15mmX30mm, about half mm thick, was cut from the center of the film. Figure 1 presents the absorption spectra of the 24nm fluorescent beads immobilized in PVA film and progressively exposed to the unpolarized light from fiber optic illuminator (Oriel, model 77501).
The 10mm light guide from the illuminator was attached to the center of the strip. The bleached area was used for absorption and fluorescence measurements. The absorption intensity decreased with the time substantially. Fluorescence (not shown) decreased proportionally to the absorption. After 8 hours exposure to the light both, absorption and fluorescence signals decreases about 20 fold. We noticed that absorption increased when the strip was tilted. Molecules with transition moments in the plane of the strip were substantially eliminated and molecules with perpendicular transition moments still remain in the sample.
3.2 Steady-state fluorescence of dissolved nanospheres
Next, we dissolved in 1ml of water 25mg of both, bleached and unbleached portions of the PVA strip. We observed fluorescence from the sample of bleached area, which had no emission in the film.
An oriented bead system is a suitable model to carry out studies on the fluorescence anisotropy. For this purpose, beads embedded in the PVA films were dissolved in water. Figure 2 presents emission spectra of the unbleached (a) and photobleached (b) beads recorded with the magic angle direction to the polarization of the electric vector of exciting light. The observed spectra are similar for both, bleached and unbleached samples. We note that the intensity of the sample containing bleached nanospheres was only 10 fold smaller than for unbleached sample (compare top and bottom).
Figure 2 shows also steady-state anisotropies for both samples after dissolving in water. Very low anisotropy was observed for unbleached sample (top) and very high, close to the limiting value of 0.4, for bleached (bottom). Before the bleaching, transition dipole moments are symmetrically distributed around the axis of light propagation. The formation of anisotropic distribution of transition moments in the bleached nanospheres is presented in Fig. 3 .
Non-polarized light interacts with fluorophore-labeled nanospheres immobilized in the polymer. This interaction results in photodestruction of the dye molecules with dipole moments aligned in the plane of the polymer. Transition dipole moments (dyes) which left after photobleaching lie along the axis of the light propagation. These molecules were not excited in the process of photobleaching.
3.3 Time-resolved measurements of dissolved nanospheres
The first step in quantitative measurements of orientation is to determine whether the probe posses distinct polarization so that the transition dipole moment of the fluorophore reflects the orientation of the bead. Figure 4 shows the anisotropy decays of unbleached and bleached nanospheres.
In the case of unbleached sample, two correlation times are recovered, 4.46ns and 0.35ns with associated anisotropies 0.05 and 0.15, respectively (Table 1 ). The same figure (Fig. 4 light line) shows the decay of anisotropy of bleached nanosperes.
The anisotropy decay can be fitted with two exponents with correlation times of 140.1, and 0.3ns, and associated anisotropies of 0.31 and 0.06, respectively. The anisotropy associated with the longer correlation times increased significantly upon photobleaching. We conclude that polarized nanospheres are good for monitoring the anisotropy.
Next, we measured lifetimes of unbleached and bleached nanospheres. The intensity decays are presented in Fig. 5 and decay parameters are summarized in Table 2 . Analysis of the fluorescence decays for untreated beads resulted in the fluorescence lifetimes of 4.83ns (α1=47%) and 2.74ns (α2=53%) yielding an (amplitude-weighted) average fluorescence lifetime <τ>=3.73ns. The light treatment (photobleaching) yielded fluorescence lifetime components 5.03ns (α1=80%) and 1.1ns (α2=20%) resulting in increase of the average lifetime to <τ>=4.26ns.
We realized that in the photobleaching process many dye molecules were eliminated as shown in Fig. 6 . This reduces an excitation energy migration (homo transfer ) between dyes and possible quenching. In unbleached nanospheres the fluorescence of the red dye is partially self-quenched. This finding is important because longer lifetime is preferred in anisotropy measurements.
3.4 Fluorescence correlation spectroscopy measurements of dissolved nanospheres
We examined mobility of the fluorescent nanospheres by means of Fluorescence Lifetime Correlation Spectroscopy (FLCS), which measures intensity fluctuations of the fluorophores diffusing through a small confocal volume and analyzing them simultaneously with lifetime of the samples. In FLCS data acquisition, the excitation is pulsed and two independent timings are performed for every detected photon. We used FLCS method because typical autocorrelation function recorded with one APD detector contains afterpulsing time constant of the order of a few hundred nanoseconds which make it difficult to recognize residue from rotational correlation effects. In this setup, the fluorescence light was split into two parts by polarizer cube and directed to two detectors in order to extract translational diffusion and to measure part of rotational correlation curve from those data as well. Figure 7(a) shows the measured auto- and cross-correlation functions for unbleached nanospheres. Autocorrelation functions for detectors that recorded polarized emission perpendicularly and in parallel to the excitation light are presented with green and blue lines, respectively. A cross-correlation of the experimental signals from two polarizations is presented with black circles and a red line (fit). We recovered comparable values of diffusion coefficients D from auto- and cross correlation in one component fits with D=15µm2/s. The diffusion time was τD=2.6ms. The translational diffusion coefficient obtained experimentally was very well related to the Stokes-Einstein relationship D=kT/6πηr, where k is the Boltzmann constant, T the absolute room temperature, η the solution (water) viscosity, and r the hydrodynamic radius of the micrspheres. From calculations, we expected D=18.1µm2/s. We did not found equal amplitudes of autocorrelation functions (which are reversely related to the concentration of the sample) because of photoselection effect obtained using polarized excitation light. The larger amplitude of autocorrelation function corresponds to the detector that recorded polarized emission perpendicularly (green line) to the excitation and the lower amplitude value to the parallel polarization (blue line).
In Fig. 7(b), we report data obtained by auto- and crosscorrelation functions using horizontally polarized light of excitation applied for bleached beads. The bleach bead preparation studied was identically to the one described in the above experiment. Under those conditions, single channel analysis received for both the detectors that recorded either polarized emission perpendicularly (green) or in parallel (blue) to the excitation shows the same gain of the autocorrelation function amplitude in the range of delay times of about 10−4 – 10−3ms.
Such a time scale is characteristic for the fluctuations due to rotation of the fluorescence bleached beads with its absorption transition dipole moment relative to the linear polarized excitation light (for comparison see decay anisotropy data Fig. 4). Because of rather large noise of the autocorrelation curve analysis in region of rotational correlation time, we also performed crosscorrelation analysis. Resulting function shows different behavior with drop of the amplitude for short correlation times. This is an anti-correlation and it is usually recorded for highly bright molecules . Drop of crosscorrelation function in region of fast lag times results from the fact that molecules after excitation emit the fluorescence signal to the detector that recorded polarized emission in one direction, whereas the other detector does not record any signal at the same time. Such a behavior was not observed for non-bleached nanospheres because of Förster Resonance Energy Transfer (FRET) when the molecules after energy transfer lose the preferential direction of excitation. For calculating the translational diffusion coefficient, we ignored the fast part of the correlation function, which is characteristic for rotations. We obtained D=15µm2/s. The result also shows that the majority of beads enter the confocal volume as single particles. Only a very small part of spikes in the time trace originate from aggregates, which can be observed for longer lag times (in range of 101 to 102ms) but they did not influence the calculated value of the translational diffusion coefficient.
In this study, a simple and rapid method for preparation of polarized fluorescent nanospheres is described. In order to achieve this goal the fluorescent nanospheres were immobilized in PVA films, bleached of non polarized light propagation and rinsed to the aqueous solution. Such prepared nanoparticles have a high anisotropy value and a reasonable brightness. We attempted to estimate the reduction in observed fluorescence. First, we compared absorptions of dissolved bleached and unbleached nanospheres and found that optical density decreased about 14 fold upon bleaching. This tells that in a bleached sample remain about 7 percent of dye molecules. However, fluorescence signal decreased only about 10 fold (see Fig. 2). Discrepancy between absorption and fluorescence data can be explained if one takes into account process of self quenching which is stronger for unbleached beads. In fact, the lifetime of bleached nanospheres increases significantly (see Fig. 5 and Table 2). Using oriented fluorescent nanoospheres as a macromolecule label may provide a high initial anisotropy and a good brightness. Thus, polarized nanospheres with a proper functionalization or surface modification will find applications in biophysics and biology, e.g., in polarization assays. FCS is the method of choice to determine local concentrations and molecular mobility parameters. We demonstrated that this method can be used to study both, single beads and/or aggregated cluster. FLCS data also resulted in rotational correlation.
This work was supported by Texas ETF (CCFT, Z.G.), Austrian FWF grant P20454-N13 (ZFP) and by ARPATP Project 000130-0042-2007 (IG). RL dedicates this paper his mentor professor Stanislaw Krawczyk.
References and links
1. P. V. Jena, P. S. Shirude, B. Okumus, K. Laxmi-Reddy, F. Godde, I. Huc, S. Balasubramanian, and T. Ha, “G-Quadruplex DNA Bound by a Synthetic Ligand is Highly Dynamic,” J. Am. Chem. Soc. 131(35), 12522–12523 (2009). [CrossRef] [PubMed]
4. D. Evanko, “Nature Milestones in Light Microscopy. Milestone 17. Single molecules in the dark,” Nature (October): (2009), doi:.
6. L. Edman, Z. Földes-Papp, S. Wennmalm, and R. Rigler, “The fluctuating enzyme: a single molecule approach,” Chem. Phys. 247(1), 11–22 (1999). [CrossRef]
8. J. N. Forkey, M. E. Quinlan, M. A. Shaw, J. E. T. Corrie, and Y. E. Goldman, “Three-dimensional structural dynamics of myosin V by single-molecule fluorescence polarization,” Nature 422(6930), 399–404 (2003). [CrossRef] [PubMed]
9. S. Syed, G. E. Snyder, C. Franzini-Armstrong, P. R. Selvin, and Y. E. Goldman, “Adaptability of myosin V studied by simultaneous detection of position and orientation,” EMBO J. 25(9), 1795–1803 (2006). [CrossRef] [PubMed]
10. A. P. Bartko and R. M. Dickson, “Imaging three-dimensional single molecule orientations,” J. Phys. Chem. B 103(51), 11237–11241 (1999). [CrossRef]
11. M. Tokunaga, K. Kitamura, K. Saito, A. H. Iwane, and T. Yanagida, “Single molecule imaging of fluorophores and enzymatic reactions achieved by objective-type total internal reflection fluorescence microscopy,” Biochem. Biophys. Res. Commun. 235(1), 47–53 (1997). [CrossRef] [PubMed]
12. A. Yildiz, J. N. Forkey, S. A. McKinney, T. Ha, Y. E. Goldman, and P. R. Selvin, “Myosin V walks hand-over-hand: single fluorophore imaging with 1.5-nm localization,” Science 300(5628), 2061–2065 (2003). [CrossRef] [PubMed]
13. S. A. Rosenberg, M. E. Quinlan, J. N. Forkey, and Y. E. Goldman, “Rotational motions of macro-molecules by single-molecule fluorescence microscopy,” Acc. Chem. Res. 38(7), 583–593 (2005). [CrossRef] [PubMed]
15. A. M. De Grand, S. J. Lomnes, D. S. Lee, M. Pietrzykowski, S. Ohnishi, T. G. Morgan, A. Gogbashian, R. G. Laurence, and J. V. Frangioni, “Tissue-like phantoms for near-infrared fluorescence imaging system assessment and the training of surgeons,” J. Biomed. Opt. 11(1), 10 (2006). [CrossRef]
16. E. Meiss, H. Konno, G. Groth, and T. Hisabori, “Molecular processes of inhibition and stimulation of ATP synthase caused by the phytotoxin tentoxin,” J. Biol. Chem. 283(36), 24594–24553 (2008). [CrossRef] [PubMed]
17. R. Luchowski, P. Sarkar, S. Bharill, G. Laczko, J. Borejdo, Z. Gryczynski, and I. Gryczynski, “Fluorescence polarization standard for near infrared spectroscopy and microscopy,” Appl. Opt. 47(33), 6257–6265 (2008). [CrossRef] [PubMed]
18. B. Valeur, “Molecular Fluorescence: Principles And Applications,” 125–198 (2006).
19. J. Enderlein and I. Gregor, “Using fluorescence lifetime for discriminating detector afterpulsing in fluorescence-correlation spectroscopy,” Rev. Sci. Instrum. 76(3), 5 (2005). [CrossRef]
21. E. L. Elson and D. Magde, “Fluorescence correlation spectroscopy. I. Conceptual basis and theory,” Biopolymers 13(1), 1–27 (1974). [CrossRef]
22. D. Magde, E. L. Elson, and W. W. Webb, “Thermodynamic fluctuations in reacting system: measurements by fluorescence correlation spectroscopy,” Phys. Rev. Lett. 29(11), 705–708 (1972). [CrossRef]
24. R. Luchowski, E. G. Matveeva, I. Gryczynski, E. A. Terpetschnig, L. Patsenker, G. Laczko, J. Borejdo, and Z. Gryczynski, “Single Molecule Studies of Multiple-Fluorophore Labeled Antibodies. Effect of Homo-FRET on the Number of Photons Available Before Photobleaching,” Curr. Pharm. Biotechnol. 9(5), 411–420 (2008). [CrossRef] [PubMed]
25. M. L. Barcellona, S. Gammon, T. Hazlett, M. A. Digman, and E. Gratton, “Polarized fluorescence Correlation spectroscopy of DNA-DAPI complexes,” Microsc. Res. Tech. 65(4-5), 205–217 (2004). [CrossRef]