Fluorescence, Phosphorescence, And Chemiluminescence

Analytical ChemistryReviewtypical workflows and precautions for quantitative analysis ofsingle-molecule superresolution images.17 Their guidelinesinclude potential pitfalls and essential control experiments.Sage et al. report the quantitative evaluation of softwarepackages for single-molecule localization microscopy (SMLM),noting that the quality of super-resolution images obtainedby SMLM depends largely on the software used to detectand accurately localize point sources.18 They focus on thecomputational aspects of super-resolution microscopy in orderto devleop metrics that reflect various trade-offs of SMLMsoftware packages in an effort to help users choose the bestsoftware for their needs.Recent advances in single-molecule switching nanoscopyhave greatly accelerated data acquisition speed and haveimproved temporal resolution of super-resolution imaging.However, Lin et al. noted that this technique had not beenquantified as to whether an increase in speed comes withcompromised image quality.19 They provide guidelines for optimizing the factors that affect spatial and temporal resolution sothat single-molecule switching nanoscopy at high speeds canachieve the same image quality as imaging at conventionalspeeds. In other work, Moerner and co-workers report that therotational mobility of single molecules affects localizationaccuracy in super-resolution fluorescence microscopy.20Long et al. have investigated the effects of fixed pattern noise(FPN) in semiconductor complementary metal oxide(sCMOS) cameras that had previously obstructed theirwidespread use in single molecule localization microscopy.21Surprisingly, they found that FPN leads to almost no effect onlocalization precision and that localization bias is usually 2 nmand thus can be neglected for most localization microscopyexperiments.Bleed-through or misidentification of probe species inmulticolor localization microscopy is known to create falsecolocalization and to artificially increase certain types ofcorrelation between two imaged species, affecting the reliabilityof information provided by colocalization and quantifiedcorrelation. Yet, surprisingly, the effect of bleed-through oncorrelation and methods for its correction had not previouslybeen systematically studied at typical rates of bleed-through inmulticolor imaging. Kim et al. recently presented a method ofbleed-through correction that can be applied to all types oflocalization microscopy (PALM, STORM, dSTORM, GSDIM,etc.), provided the rate of bleed-through can be reliablydetermined.22 In another work, McGorty et al. have describedthe correction of depth-dependent aberrations in 3D singlemolecule localization and super-resolution microscopy.23 Theydemonstrate that their method can maintain z localizationaccuracy over a large range of imaging depths between 0 and2.5 μm past the coverslip.Allen et al. have noted and addressed the importance ofproper sample preparation in single molecule localization-basedoptical nanoscopy.24 They have presented in-depth analyses ofall aspects of sample preparation for single molecule superresolution, including both live and fixed cell preparation, choiceof fluorophore, fixation and staining techniques, and imagingbuffer considerations. In a realated study, Whelan et al. havediscussed potentially overlooked preparative artifacts in singlemolecule localization microscopy (SMLM).25 In this study, theypresented three well-optimized fixation protocols for stainingmicrotubules, mitochondria, and actin with a discussion onvarious artifacts related to images obtained from samplesprepared using these protocols.same techniques are becoming quite common tools for researchers of widely varying interests and expertise. Single molecule fluorescence, fluorescence correlation spectroscopy (FCS),and Forster resonance energy transfer (FRET) are threeexamples of such techniques. An in depth review of each ofthese techniques is not possible in this relatively short section.However, some relevant reviews and advanced examples ofdiscoveries using these tools can be found in other sections ofthis Review and in the literature at large. Rather than focus onvarious applications or specific results, this section will brieflyhighlight a small subset of articles and reviews that acknowledgesome technical challenges in using these tools and/or offersome guidance on how results gained using these techniquescan be further improved. Although these techniques are relatedand occasionally overlap, the discussion below is divided intothree subsections: (1) single molecule fluorescence, (2) fluorescence correlation spectroscopy (FCS), and (3) Forster resonance energy transfer (FRET).Single Molecule Fluorescence. Recent advances incommercial imaging systems have allowed the use of singlemolecule fluorescence as a useful tool for many researchersof varying backgrounds. One such example involves the usesingle molecule and single-particle fluorescence microscopy bytraditionally trained synthetic chemists to investigate chemicalsystems by exploring the mechanisms of organic reactions,spatial distribution of chemical reactivity on surfaces, andthe phase of active catalysts. In a recent Perspectives article,Cordes et al. discuss the requisite photophysical and chemicalproperties of fluorescent reporters and highlight the primarychallenges in applying single-molecule techniques to chemicalquestions, with a goal of encouraging its broader use to observechemical reactions molecule by molecule.12Single molecule fluorescence microscopy has a wide range ofuses, one of which involves biological investigations insideliving cells with millisecond- and nanometer-scale resolution.The power of single-molecule-based methods and its increasingaccessibility have led to many exciting results. However, optimizing new single-molecule experiments can be challenging, inparticular when super-resolution imaging and tracking areapplied to living cells. In a recent review, Haas et al. summarizecommon obstacles to live-cell single-molecule microscopy andalso describe methods developed and applied to overcomingthese challenges.13Shivanandan et al. have reviewed the challenges in quantitative single molecule localization microscopy (SMLM), ahighly useful tool for quantitative biological experiments ranging from molecular biology to neuroscience.14 This reviewincludes a discussion of applications of SMLM in quantitativebiology, as well as some of the challenges involved and somesolutions that have been proposed. In another work, Endesfelderet al. discuss some key principles of single-molecule superresolution techniques; pointing out pitfalls, highlighting recentdevelopments, and identifying opportunities for the future.15Sauer and co-workers suggest that achieving super resolution is now public domain as a result of the availability ofcommercial instruments and open-source reconstructionsoftware.16 They also note that localization microscopy remainsprone to user errors. For example, high emitter densitieswith inappropriate photoswitching rates can give rise to theappearance of artificial membrane clusters. Thus, singlemolecule movies recorded to reconstruct these images mustbe carefully investigated when investigating membrane organization and cluster analysis. In other work, Coltharp et al. describe172DOI: 10.1021/acs.analchem.5b04109Anal. Chem. 2016, 88, 170 202

Analytical ChemistryReviewForster Resonance Energy Transfer. The physicalprocess of Förster resonance energy transfer (FRET) waselucidated more than 6 decades ago and has since become apowerful tool for biomedical research. A wide range of FRETapproaches have been described with each having corresponding advantages and disadvantages. Ma et al. have recentlyreviewed FRET applications in protein studies.32 They summarize the basic components of FRET techniques, establishedquantification methods, as well as potential pitfalls andillustrated all of these by example applications.Conjugation between streptavidin (SA) and biotin has beenwidely used to link donors and acceptors for investigatingdistance-dependent FRET, but a contradictory finding hasrecently been reported that FRET of a common system of(QD-SA)-(biotin-DNA-dye) (donor, quantum dot (QD);acceptor, small organic fluorescent dye; and linker, deoxyribosenucleic acid (DNA) molecule via SA-biotin conjugation) lost itsdependence on the number of DNA base pairs when using aphosphate-buffered saline (PBS) solution. Thus, Saremi et al.have reported a re-evaluation of biotin streptavidin conjugationin FRET applications in an effort to resolve these conflictingresults.33 They have found that the conflict was caused by theionic strength of the adopted buffer solutions. FRET was foundto lose the DNA length dependence at relatively high ionicstrengths.Warner and co-workers have recently explored the FRETphenomena in organic nanomaterials for potential applicationsin sensor and optoelectronic devices.34 In this regard, specialtypes of organic salts with relatively low melting points, giventhe acronym Group of Uniform Material Based of Organic Salts(GUMBOS), are employed. Two cyanine-based GUMBOS ofvariable methane chain length were used to synthesize binarynanomaterials (nanoGUMBOS). FRET between binarycyanine nanoGUMBOS of altered methane chain exhibitedemission in the visible to near-infrared region of the electromagnetic spectrum, demonstrating potential applications inoptoelectronics. These binary organic nanomaterials alsoexhibited high thermal- and photostability. Moreover, thisapproach demonstrated that tuning emission spectra producedby use of FRET can be easily achieved by changing the moleratio of donor and acceptor during nanomaterials formation.In another study from this group, carbazole based GUMBOSwere synthesized that displayed intramolecular FRET.35 Thesecarbazole-based GUMBOS exhibited multiple emissions fromthe second excited singlet state (S2), first excited singlet state(S1), as well as from an intramolecular charge transfer state.This study suggests a significant overlap between the S2emission and S1 absorption such that intramolecular FRET isobserved.Ensemble FRET results can be analyzed in a variety of ways,and due to experimental artifacts, the results obtained fromdifferent analyses are not always consistent. In an effort todetermine optimal analysis for use in nanodrop fluorometry,Kelliher, et al. performed both ensemble and single-moleculestudies of FRET on oligomers of double-stranded DNA andcompared the single-molecule results to those obtained usingvarious ensemble FRET analyses.36 It was found that analyzingthe increase of acceptor fluorescence is less likely to introduceerrors, as compared with analyzing the fluorescence intensityof the donor in the absence and presence of the acceptor.In another study, Kruger et al. describe single molecule FRETdata analysis procedures for FRET efficence determination.37They outline the parameters needed for FRET efficiencyFluorescence Correlation Spectroscopy. Fluorescencecorrelation spectroscopy (FCS) is a powerful tool for accuratedetermination of translational diffusion coefficients. In anIUPAC Technical Report, Enderlein considered several of themost common sources of optical aberrations and their impacton the outcome of conventional FCS measurements.26 A newvariant of FCS, dual-focus FCS, which is robust against most ofthe considered aberrations, was also described in this report.Sanguigno et al. noted that FCS experiments performed onflat elements, such as membranes, show unusually high relativeerrors as compared to experiments in aqueous solution.27 FCSmeasurements on flat surfaces have generally been interpretedwith certain hypotheses; the membrane is assumed to beperfectly flat, motionless, and aligned with the optical axes. Theauthors investigated the robustness of these hypotheses, in anattempt to understand how misalignments and thermalfluctuations affect temporal correlation of intensity fluctuationcollected during measurements on membranes.Imaging FCS using array detectors has been used to quantifythe number, mobility, and organization of biomolecules in cellsand organisms. However, Wohland and co-workers have notedthat there have not been any systematic studies on the errors inthese estimates that are introduced due to instrumental andexperimental factors.28 They further investigate the limitationsthat current state-of-the-art detectors place on time resolution,signal-to-noise ratio, and total measurement time. This wasachieved by using a combination of simulations and experiments on lipid bilayers to provide characteristic performanceparameters and guidelines that govern accuracy and precision ofdiffusion coefficient and concentration measurements incamera-based FCS. Guidelines are provided for an efficientexperimental design for camera-based FCS to extract information on mobility, concentration, and heterogeneity.Although FCS is a powerful tool for investigation ofmolecular dynamics using fluorescent proteins as molecularlyspecific labels, Wohland and co-workers note that FCS dataanalyses and interpretation using fluorescent proteins remain achallenge due to typically low signal-to-noise ratio of FCS dataand correlated noise in autocorrelated data sets.29 Fittingprocedures that ignore these two important issues can providesimilarly good fits for multiple competing models. Bayesianmodel selection accounts for the highly correlated noise that ispresent in FCS data sets and additionally penalizes modelcomplexity to prevent over interpretation of FCS data. Thus,the authors applied Bayesian model selection to evaluate FCSdata from fluorescent proteins assayed in vitro and in vivo.They found that Bayesian model selection was a robustprocedure for determining appropriate transport and photophysical models for fluorescent proteins when suitable modelsare provided. In other studies, Schwille and co-workers presenta graphical user interface (PyCorrFit) for fitting theoreticalmodel functions to experimental FCS data.30 The programfeatures a set of tools specialized in FCS data evaluation.FCS is typically used at nanomolar concentrations and thislimitation is generally thought to be fundamentally related tothe technique itself. However, Laurence et al. report that thelimitation to nanomolar concentrations is not fundamental butinstead due to detector limits as well as laser fluctuations.31 Useof a high count rate detector system and application of laserfluctuation corrections allowed FCS measurements up to 38 μMwith the same signal-to-noise as at lower concentrations withoutthe need for nanoconfinement approaches previously used toincrease the concentration range of FCS.173DOI: 10.1021/acs.analchem.5b04109Anal. Chem. 2016, 88, 170 202

Analytical ChemistryReviewS/N and photostability as well as better tumor-to-backgrounduptake.45McCarley and co-workers have recently synthesized a novelprobe based on trimethyl-locked quinone propionic acid(Q3PA) attached via linker to a naphthalimide for cellularimaging of a cancer enzyme.46 An overexpressed cancer enzymepresent in the cytosol of human tumor cell, NAD(P)H/quinone oxidoreductase-1, selectively reduces Q3PA andproduced bright emission of naphthalimide (reporter) whichis used to image cancer cells. The quantum yield of thereporter, formed after reduction by the cancer enzyme, is 95times more fluorescent than the probe. In another study, theturn on fluorescent probe based on a naphthalimide derivativewas synthesized for rapid and selective imaging of biologicalthiols composed of glutathione, cysteine, and homocysteine.47Tumor pH-sensitive magnetic nanogrenades (termedPMNs) have been developed and used as cancer theranosticagents. PMNs are comprised of self-assembled iron oxidenanoparticles and pH-responsive ligands. They target tumorsvia surface-charge switching triggered by the acidic tumormicroenvironment. In acidic subcellular compartments, they arein a highly active state that turns on MR contrast, fluorescence,and photodynamic therapeutic activity. Therapeutic efficacywas achieved in heterogeneous drug-resistant tumors.48 Inanother example of probes possessing both MR and opticalimaging modalities, the Meade group synthesized multimericcontrast agents possessing three Gd(III) chelates and an IR-783dye. Biodistribution of a PEGylated derivative resulted inobservable fluorescence in xenograft MCF7 tumors and renalclearance via MR imaging.49A new theranostic prodrug that undergoes hydrogenperoxide-mediated boronate oxidation features activation of afluorophore for detection and release of the anticancer agent,SN-38. The prodrug showed effective antitumor activity in amouse model of metastatic lung disease. This technology isproposed for use in imaging and treatment of metastatictumors, as these characteristically possess high levels of reactiveoxygen species.50Gold quantum dots exhibit potentially favorable optical andmagnetic properties as compared to gold nanoparticles.However, biocompatibility and aqueous stability issues havelimited their use in imaging. New silica encapsulated goldquantum dots have been synthesized that are called “quantumrattles.” They are stable in aqueous solutions, and cytotoxicitystudies are promising. In vivo studies show that they lead toimproved drug delivery and reduced tumor burden viaphotothermal therapy. They couple three imaging modalities:near-infrared fluorescence, photoacoustic, and magnetic resonance imaging.51The RAF serine/threonine kinases regulate cell growththrough the MAPK pathway. It is known that protein multimersplay a role in RAF activation and tumor responses to RAFinhibitors. However, the stoichiometry and cellular location ofspecific protein multimers involved in RAF activation have notbeen reported previously. Nan and co-workers used photoactivated localization microscopy (PALM) along with quantitative spatial analysis to directly visualize protein multimers incells. This study embodies direct confirmation of the existenceof RAF dimers and higher-order multimers as well as theirinvolvement in cell signaling.52The prediction of photophysical properties of fluorophoresfor superresolution microscopy has been addressed via a newstochastic approach. This method allows enumeration ofcalculation and illustrate that the shape of the FRET distribution changes depending on what parameters are included inthe data analysis procedure using single molecule FRET dataobtained on G-quadruplex DNA structures that exhibit largeconformation diversity.Alternating-laser excitation (ALEX) combined with singlemolecule FRET has been found to be a power technique tostudy biological interactions. Hohlbein et al. recently presenteda comprehensive overview of the concept and current applications of ALEX. They discuss how to obtain fully correcteddistance information across the entire FRET range and presentnew ideas for applications of ALEX which they claim will pushthe limits of single molecule FRET-based experiments in termsof temporal and spatial resolution for the study of complexbiological systems.38Finally, Cho et al. have used FRET to devise a rapid, generaland cost-efficient super-resolution imaging method which canbe directly employed using a simple fluorescent imaging systemwith general fluorophores.39 Fluorescent donor molecules thatlabel specific target structures are stochastically quenched bydiffusing acceptor molecules, thereby temporally separatingotherwise spatially overlapped fluorescence signals and allowingsuper-resolution imaging. The authors expect that the newmeth

Specialized Fluorescence Techniques 171 Single Molecule Fluorescence 172 Fluorescence Correlation Spectroscopy 173 Forster Resonance Energy Transfer 173 Imaging and Super-Resolution Imaging (Con-ventional and Lifetime) 174 Instrumentation and Laser Based Fluorescence Techniques 175 Nonlinear Emission Processes in Fluorescence Spectroscopy 176