Fluorescence of Supermolecules, Polymers, and Nanosystems

The history of molecular fluorescence is closely associated with the emission from plant extracts.N. Monardes, in his Historia Medicinal (Seville, 1565), was the firstto describe the blue opalescence of the water infusion of the wood of a Mexican tree used to treatkidney ailments. The strange optical properties of the wood, known as Lignum nephriticum(kidney wood), were later investigated by Kircher, Grimaldi, Boyle, Newton and many other scientists andnaturalists in the ensuing centuries. However, when G.G. Stokes published in 1852 the first correct relationshipbetween light absorption and fluorescence, his observations were based on the emission of quinine sulphatesolution, because in Europe the wood of Lignum nephriticum was no longeravailable and its botanic origin was unknown. An inspection of the works of sixteenth century Spanish missionariesand scholars who compiled information on the Aztec culture, such as Fr. Bernardino de Sahagun and FranciscoHernandez, indicates that pre-Hispanic Indian doctors had already noticed the blue color (fluorescence)of the infusion of coatli, a wood used to treat urinary diseases.Coatli wood was obtained from Eyserhardtia, a tree of the familyof Leguminosae, and is the most likely source of the exotic Lignum nephriticum. The wood of Eysenhardtia polystachyacontains large quantities of Coatline B, a rare C-glucosyl-α-hydroxydihydrochalcone. This compound gives rise to a fluorescentreaction product, in slightly alkaline water at room temperature, which is responsible for the blue emissionof Lignum nephriticum infusion.

The present review does not intend to be exhaustive: some applications of fluorescence relevant tofundamental and applied research will be illustrated with pertinent examples. In addition, technologicalor industrial applications will be exemplified.

On the optical irradiation of a complex absorptive medium, an ultrafast transfer of electronicenergy between chromophores or sites generally takes place, preceding any ensuing fluorescence. Such transferprocesses represent a redistribution of optically acquired energy and commonly lead to the indirectexcitation of other chromophore units. Consequently, electronic energy transfer is a major determinantof detail in the fluorescence spectrum of the system. Where a sequence of energy transfer steps canbe identified, the overall directionality of each sequence is important in determining any localizationof the electronic energy. It is therefore highly significant that it is possible to exercise directionalcontrol over energy transfer, not only through the operation of an intrinsic spectroscopic gradient, butalso by a variety of less well-known methods involving applied electrical or optical fields. Exploitationof the latter methods holds promise for advances in a wide range of technologies including opticalswitching and the production of energy harvesting materials.

The mathematical properties of the general luminescence decay law are described. Special attentionis paid to cases represented by continuous distributions of decay rate constants. Six important decay functionsare described in detail: stretched exponential (Kohlrausch), compressed hyperbola (Becquerel), Mittag–Leffler,Heaviside, Weibull, and truncated Gaussian.

We discuss the specific features of fluorescence that make it the method of choice to optically detectsingle molecules and other single luminescent nano-objects, such as color centers or semiconductor nanocrystals,in microscopy experiments. The main origin of the high signal-to-background ratio of fluorescence experimentsis the resonance enhancement of absorption by the fluorophore with respect to non-resonant Raman scatteringof the matrix. As a consequence, other methods with lower differences in resonance factors betweenprobe and matrix will fail to reach single-molecule sensitivity in far-field, even with diffraction-limitedexcitation volumes. We also briefly discuss signal-to-noise ratios in several new optical methods whichdo not rely on fluorescent probes.

Development of fluorescent chemosensors able to bind and signal either cations or anions in aqueous media is reviewed. Abundant examples in the literature support the versatility of polyamine-based chemosensors towards applications in water. The chemosensors described have a common structure consisting of a binding polyamine motif linked to a fluorophoric unit. Different sensing mechanisms are explored: photoinduced electron transfer from the amine lone pairs to the fluorophoric unit(s), photoinduced electron transfer between fluorophoric units, energy transfer, excimer and exciplex formation. The protonation equilibria of the polyamine chains render great versatility to the chemosensors since they are at the basis of the different sensing mechanisms and also allows the tuning between cation and anion binding regime.

The photophysical properties of fullerenes are selectively reviewed, with an emphasis on fluorescence.It is shown that fullerenes display several unusual photophysical properties, such as their fluorescenceanisotropy, and also that they can be used for a more complete understanding of general photophysicalprocesses like charge-transfer complexation and the heavy-atom effect. In spite of the work already carriedout, knowledge of the photophysics of fullerenes and derivatives is still incomplete, and much remains tobe done in this area.

Supramolecular radiative decay engineering allows systematic variations of the radiative decayrates, and therefore changes in the fluorescence lifetimes and intensities. Depending on whetherfluorescent dyes are immersed in macrocyclic host molecules with low or high polarizability, reducedor enhanced fluorescence lifetimes may result. Solvatochromic probes to “measure” thepolarizability inside such molecular container compounds are now at hand. Cucurbiturils, for example,are water-soluble host molecules, which possess a cavity with an exceptionally low polarizability,close to the gas phase. Placing fluorescent dyes inside cucurbiturils allows one to create in aqueoussolution a unique microenvironment, which approaches that of the gas phase and leads to unprecedentedphotophysical behavior. Accordingly, complexation by cucurbituril leads to prolonged fluorescencelifetimes, for 2,3-diazabicyclo[2.2.2]oct-2-ene (DBO) up to 1 μs, the same as that foundin the gas phase.

The properties of polymer materials are often determined by their interfaces. Polymer interfacesare usually much broader than inorganic interfaces, with values from 1 to 10 nm. This range matches thetypical length scale of Förster non-radiative resonance energy transfer (FRET). While the use of FRETin polymers was pioneered by Morawetz in the 1980s, the technique has only recently been extended to obtainquantitative detailed information on polymer interfaces and other nanostructured materials. A numberof systems with nanodomains and heterogeneous dye concentration profiles, ranging from block copolymer filmsand micelles, to polymer nanoparticles, latex film formation and polymer blends, have been successfullycharacterized by FRET.

Defocused imaging in wide-field fluorescence microscopy provides information on the 3D orientationof the transition dipole moment of single molecules with nanometer spatial resolution. In this chapter,the theoretical background of defocused imaging will be presented, followed by two experimental applications.As a first example, defocused imaging was used along with fluorescence lifetime measurements to provethe fact that the molecular orientation of dye molecules in a thin film has a significant affecton their fluorescence lifetime. This is attributed to the electromagnetic boundary condition effect. Asa second example, the power of the technique for following 3D molecular rotational reorientation (molecularrotational diffusion) in thin polymer films is demonstrated. Since many molecules can be monitored in parallel,both the temporal and the spatial heterogeneity of polymer dynamics can be addressed.

Organic semiconductor materials are attracting increasing interest for application as active materialsin large-area displays, light-emitting diodes, transistors, photodetectors, solar cells, etc. Elementaryprocesses like excitation energy transfer, charge photogeneration, charge transport, and photoluminescencequenching are at the basis of the function of the materials. In order to design new materials and optimizetheir function, a detailed knowledge of the various light-induced processes is required. In this contributionwe discuss how ultrafast spectroscopy can be used to study excited-state relaxation and quenching, energytransfer, and charge generation in neat conjugated polymers as well as in polymer–C60 blends.

Useful models for resonance energy transfer (RET) are reviewed for several geometries relevant tomembranes (planar, bilayer, multilayer) and uniform donor and acceptor probe distribution. Extensions fornon-uniform distribution of fluorophores are presented and discussed. Selected examples of quantitativeapplications of RET to these systems are described. It is illustrated how information about lipid phaseseparation (phase composition, domain size, partition coefficients, kinetics of lipid demixing) lipid–peptide(domain induction/membrane aggregation), lipid–protein (lipid selectivity in the annular region) andlipid–DNA (lipoplex structural characterization) interactions can be recovered from time-resolvedRET data.

Fluorescence correlation spectroscopy is a powerful technique for observing the diffusion offluorescent molecules. In buffer solution the autocorrelation analysis of the intensity fluctuations allowsthe calculation of the diffusion coefficient, and therefore the size of the molecule, as well as its interactions.In living cells, however, the situation is more complicated because the viscosity of the medium varies a lotfrom spot to spot. Moreover, the analysis of the autocorrelation curve is not always straightforward andthe chosen analysis method can sometimes be questioned. Autocorrelation curves obtained from living cellscan be fitted equally well with a multicomponent model or with an anomalous diffusion model (Wachsmuthet al., J Mol Biol, 298:677, 2000). The latter model introduces a global anomaly parameterthat describes the heterogeneity of the hindrance of the medium to the diffusing molecules in the cell.By applying the two models on a system of interacting proteins in living cells, we could compare theadvantages and disadvantages of both models. This article will give an introduction to fluorescence correlationspectroscopy as well as HIV-1 integrase, whose protein fragments have been used as a model in thisstudy. HIV-1 integrase has been shown to interact with a cellular protein, called lens epithelium-derivedgrowth factor or LEDGF/p75. This interaction will be used as a starting point for the comparison ofthe two models.

As an interface between different biological compartments, membranes guarantee an efficient exchangeof matter, energy and/or signals. For this purpose, such an interface has to be designed as a verydynamic system, yet with a non-random distribution of its components, lipids and proteins. A delicatebalance of lipid and protein interactions is the basis of tightly regulated mechanisms to concentrate moleculesat the site of interest at a specific time and, thereby, exclude unwanted components. In order toelucidate this highly intricate architecture, the top-down approach—by looking at the intact cell—isbest complemented by a bottom-up strategy, by building the whole complexity starting from a minimalnumber of components. Within this framework, model membranes are key systems to isolate the biologicalmachinery and identify its function. In this paper, we review research on biomimetic membranes for opticalmicrospectroscopy. In particular, we focus on giant unilamellar vesicles (GUVs), and their applicationto studies on domain assembly and on membrane curvature and deformations. In order to build complexity,efforts must be made towards mimicking cellular compositions, by using GUVs with native lipid compositions,to reconstitute (complexes of) membrane proteins and to include components of an artificial cytoskeletonunderneath the bilayer. Novel exciting avenues lie ahead in the arena of membrane biophysics, many of whichare strongly coupled to the promising developments of optical technologies.

Carbon nanotubes, upon surface modification through defects-targeted functionalization, exhibit strongphotoluminescence in the visible and into the near-IR region. In this chapter, the general features andpolarization characteristics of the photoluminescence are highlighted, mechanistic issues addressed, andpotential applications explored. The similarities and differences between the defect-derived emission andthe bandgap fluorescence (emission from individualized single-walled carbon nanotubes) are also discussed.

We have studied lipopolyamine–DNA complex formation by fluorescence correlation spectroscopy(FCS). Two lipopolyamines, N ⁴,N ⁹-dioleoylspermineand N ¹-cholesteryl spermine carbamate, wereused to condense linear calf thymus DNA and two plasmid DNAs: pGL3 (5.3 kilobase pairs) and pEGFP(4.7 kilobase pairs). PicoGreen

^® (PG), a high-affinity DNA intercalating agent that only fluoresceswhen intercalated, was used in our FCS study. In this study, the ConfoCor I set-up upgraded with TimeHarp 200was used. FCS directly visualizes the condensation process by tracking changes in diffusion coefficientsand particle numbers. We were able to define the fluorescent signalling behaviour of PG through the processfrom dye binding to dye release and then dye quenching. Dye release was suggested as the indicator forDNA conformational change, but not for nanoparticle formation. Dye quenching, through the observation oflifetime change, is a more important event accurately and sensitively reporting that a singlenanoparticle exists.

We report on the observation and application of morphology-dependent resonances (MDR) (or whisperinggallery mode) emission from individual micron-sized particles. MDR emission has been observed through confocaland two-photon fluorescence microscopes and when the particles are trapped in a quadrupole ion trap.The emission has been collected as a series of optical section slices through the micron particles,and the emission resolved temporally using two-photon excitation.

We report on the progress that has been made in the area of luminescence sensing and encoding by makinguse of microparticles and nanoparticles prepared from plastic materials. These are quite different fromparticles built up from metal sulfides (such as the so-called quantum dots, “Q-dots”; see Michaletet al., Science 307:538, 2005), other semiconductor materials, metalnanoparticles (mainly gold) (see DL Feldheim, CA Foss (eds) Metal Nanoparticles:Synthesis, Characterization, and Applications, p 338, Marcel Dekker, 2002), or glass andits modifications including certain sol–gels. Plastic nanoparticles may contain magnetic beads inorder to facilitate separation from the sample solution. All the particles described here are doped withfluorescent dyes, which is in contrast to particles where the material itself displays intrinsic luminescence.Unlike the case of Q-dots, the color of plastic beads can be varied to a wide extent irrespective oftheir size, as can be the decay times and even anisotropy. This, in fact, is a most attractive featureof such beads and makes them superior in many cases despite the undisputed utility of other types of particlesin certain fields.

The area of beads was almost exclusively occupied until 10 years ago by polystyrene beads (alsoreferred to as latex beads) 0.1–5 μm in diameter. They are widely used in bioassays andflow cytometry because they can be manufactured with good reproducibility (usually by emulsion polymerization)and because they are rather inert. Other applications include agglutination tests, particle capture ELISAs(e.g., Abbott's IMx and AxSym), solid-phase assays (often used for pregnancy testing), scintillation proximityassays, luminescent oxygen channeling immunoassay (LOCI), and bead–FRET assays. Nanoparticles havebeen used for labeling purposes, particularly in the context of protein arrays and DNA arrays. In recentyears, beads have been fluorescently dyed for purposes of encoding, for example in combination with opticalfiber arrays and in microwells, and in methods for homogeneous multiplexed high-throughput screening. Theanalytical information may be the color of the fluorescence, its intensity (or—even better—theratio of two intensities), decay time, anisotropy, or combinations thereof.

In the first section we will describe dyes for doping plastic particles. The second section willreport on chemical sensing with addressable micro- and nanospheres, and the third on the use of dyed microparticlesin sensing pH values. We will also report on luminescence lifetime encoded microbeads as carriers for multiplexedbioassays (Sect. 4), the use of dyed polymer microparticles in simultaneous sensing of oxygen and temperature(Sect. 5), and on nanobead labels for homogeneous protein assays and protein arrays (Sect. 6).