PERSPECTIVES - Kanazawa Universitybiophys.w3.kanazawa-u.ac.jp/...OpticalMicroscopy-NatureCell-2006.pdf · 944 | DECEMBER 2006 ... Surface-plasmon microscopy (SPM) com- ... NATURE

In the past few years, light microscopy has gone far beyond the qualitative viewing of small objects by contrast enhancement. It is now possible to detect the location, orient-ation, motion and environment of specific biomolecular structures, such as proteins and organelles, that are smaller than the clas-sic resolution limit. Increased sensitivity now makes it possible to take spectroscopic and kinetic measurements of single molecules or of groups of a few of them.

This progress is largely due to improve-ments in the equipment that is used in computer-designed optics, major advances in genetic engineering of specific optical probes, improvements in quantitative and low-light imagers, application of nonlinear optical techniques and probes to biological microscopy, and progress in the information theory of image analysis. But along with these advances, a huge array of techniques has arisen from combining the rich and varied physical properties of photons. Here we demonstrate several examples to show-case the advantages of these combinations and provide a framework in which new innovations might be created.

The fundamental properties of photons can be divided into three categories: photons in isolation; interactions of photons with homogeneous material (in which spatial variations are much larger than the wave-length of light); and interactions of photons with inhomogeneous material (in which spatial variations range from those compara-ble to the wavelength of light down to those of atoms and molecules, at which quantum effects become important). Modern micro-scopy techniques generally use combinations

of at least one property selected from each category, as shown in FIG 1. Advances in mathematical data and digital-image analysis enable the properties in these three categories to be combined.

We discuss in detail five practical tech-niques that comprise different combinations of photon properties, each indicated in FIG. 1. These particular techniques were chosen mainly because of our own research activities, but many other combinations are possible. Some are already in common use and others are waiting to be developed.

Structured illumination TIRFMStructured illumination total internal reflec-tion fluorescence microscopy (TIRFM) combines the properties of structured illumination (see below), evanescent fields and, possibly, saturation (see GLOSSARY; FIG. 1). Its purpose is to achieve ‘super-resolution’ — that is, the capability to distinguish sample features that are separated by a distance of about one-tenth of a light wavelength, which is smaller than the classic resolution of a microscope (about one-half of a wavelength).

Illumination with a pattern. Illumination by any variegated pattern (usually a periodic grid of linear stripes or a checkerboard) is known as structured illumination. Structured illumination can be used in a

non-imaging mode to measure lateral (that is, in the x–y plane) diffusion rates of fluor-escently labelled biomolecules in cellular compartments (membranes, cytoplasm, nuclei and so on)1–3. But, it also can be used for imaging purposes to increase the axial (z) resolution by effectively excluding light that originates from out-of-focus planes above and below the plane of interest in the sample4,5. This ‘optical sectioning’ effect partially overcomes the inherently poor resolution of light microscopes in the axial direction.

With sine-wave periodic structured illumination, effective optical sectioning can be produced by taking three successive exposures, with the periodic illumination pattern in each exposure laterally shifted by one-third of a sine-wave period relative to the other two. With these three images recorded in digital form, a simple mathe-matical computational procedure combines information from the three images and restores some of the sharpness and contrast that is otherwise diminished by the presence of out-of-focus light. Using non-sinusoidal patterns, with more exposures at smaller relative pattern shifts, can further improve out-of-focus exclusion.

Structured illumination also increases lateral (x–y) resolution6, because the one-third-period lateral shift in a high spatial frequency sine-wave pattern can produce large and different relative changes in the fluorescence intensity among closely spaced points in a sample. (Triplet sets of images with differently oriented grids, or checker-board patterns, are needed to improve the resolution in all lateral directions.) The finer the spacing of the grid pattern that is used for structured illumination, the smaller the shift that is necessary to produce these intensity changes, and, con-sequently, the better the final lateral resolu-tion. If the structured illumination pattern is created by projecting an imprinted grid located at a field-diaphragm plane through the microscope objective, then the mini-mum pattern spacing will be limited by the resolution of the microscope. However, laser beams that intersect at the sample plane can produce a finer stripe pattern due to wave interference.

I N N OVAT I O N

Combinatorial microscopyDaniel Axelrod and Geneva M. Omann

Abstract | By taking advantage of combinations of the many rich properties of photons, new forms of optical microscopy can now be used to visualize features of samples beyond thickness and density variations. We are now within reach of viewing the motions, orientations, binding kinetics and specific transient associations of previously ‘submicroscopic’ cellular structures and single molecules.

...light microscopy has gone far beyond the qualitative viewing of small objects by contrast enhancement.

944 | DECEMBER 2006 | VOLUME 7 www.nature.com/reviews/molcellbio

PERSPECTIVES

© 2006 Nature Publishing Group

Particle nature

Intensity (photons per s)

Single-photon detection

a Intrinsic b Within homogeneous materials b Within inhomogeneous materials and quantum systems

Refraction/reflection at dielectrics

Wide-field imaging

Evanescent fields/TIR

Local phase shi�ing

Electronic quantum-state excitation ofatoms/molecules/quantum dots

Spontaneous emission of luminescence

Photobleaching

Lifetime/time resolved

Quenching

Intersystem crossing

Neighbouring quantum systems

Fluorescence resonance energy transfer

Nonlinear effects

Saturated excited state

Multi-photon absorption

Multi-wave mixing

Stimulated emission

Pump-probe microscopy

Coherent anti-Stokes Raman scattering(CARS)

Optical force generation

Force at optical gradients

• Laser tweezers

Photon pressure

Vibrational state excitation

Infrared/Raman emission

Elastic interaction

Raleigh, Mie, Tyndall scattering

Raleigh resonance scattering

Photoionization

Refraction/reflection at metals

Surface plasmons

Quenching

Photoheating

Ablation

Photoacoustics

Photon–photon interaction

Acousto-optic modulation

Sonoluminescence

Wave nature

Interference

• Structured illumination

Diffraction

• Near-field versus far-field

• Confocal detection

Polarization

Linear, circular, radial

Colour

Spectral analysis

Combination with TIRFM. If the laser beams each approach the sample plane at an angle that is greater than a well defined criti-cal angle, such that they undergo total internal reflection (TIR) at that plane, the variegated pattern is further confined as an evanescent field (in liquid, at the sample’s solid–liquid interface) to just a fraction of a wavelength in the direction that is perpendicular to the sample plane (the normal, axial or z direc-tion). As TIR illumination inherently avoids excitation of out-of-focus fluorescence, three shifts of the structured illumination pattern are adequate. But more significantly, the minimum-to-minimum spacing of grid-pattern illumination lines in laser TIR can be as low as ~0.17 µm in the plane of the sample (using a 1.49 numerical aperture objective to set up the TIR), which is smaller than the microscope resolution.

Therefore, a combination of structured illumination with laser-based TIR-excited fluorescence6 has the possibility of providing — after computational processing of the images — an unprecedented high, effec-tive lateral resolution that can be used

to distinguish features that are less than 0.10 µm apart. TIR stripes can be produced by splitting a laser beam into two beams, traversing them through opposite sides of a high numerical aperture objective7,8 and intersecting them on the sample3–5. A two-dimensional checkerboard pattern, rather than a grid of linear stripes, can be produced by using four beams that are spaced 90º apart around the objective aperture for improved resolution in both the x and y directions of the sample plane.

Combination with saturation. The lateral resolution of TIR-structured illumination produced by the interference of intersecting laser beams is still limited because the sine-wave intensity pattern varies smoothly, with no sharp edges. If a pattern could be formed with the same inter-stripe spacing but with even sharper gradients between bright and dark regions, then an even higher resolu-tion could be attained. Gustafsson9 recently achieved this by illuminating with light so intense that the bright places on the grid would saturate the capability of fluorophores

to absorb. The dark places would then become ‘spikier’ as the bright peaks broaden out. A final reconstructed-image resolution as low as 0.05 µm was experimentally confirmed9. Some subcellular structures (for example, exocytotic and endocytic vesicles) and many bacteria are only tenths of microns in diameter, and this increase in resolution could significantly enhance their visualization.

Surface-plasmon microscopySurface-plasmon microscopy (SPM) com-bines the fundamental properties of how photons interact with metals with the behav-iour of excited molecules at metal surfaces (FIG. 1). This combination produces several remarkable features, including an enhance-ment of intensity, a possible reduction of the photobleaching rate, and a high sensitivity to refractive-index variations in the sample.

In SPM, the sample is on a thin metal-film-coated standard glass substrate. Qualitatively, the surface plasmon (SP) effect10 arises because the mutual inter action of free electrons in the metal, coupled with their effective inertial mass, gives the

Figure 1 | Combinations of fundamental properties of photons. The properties of photons can be classed in three categories: (a) intrinsic properties of photons ‘in vacuum’ (that is, disregarding the material through which they travel); (b) properties observed in interactions with homogeneous mate-rials (that is, materials with spatial variations that are larger than the wavelength of light); and (c) properties observed in interactions with material containing small inhomogeneities (that is, spatial variations at the size scale of the wavelength of light or smaller). The photon properties that are dis-cussed in most detail in the main text are depicted in light blue. The properties that are used by the five combinations that are discussed in most detail in the main text are connected by coloured lines: nonlinear structured illumination total internal reflection fluorescence microscopy (TIRFM; red); surface-plasmon microscopy (SPM; blue); polarized TIRFM (yellow); TIR–FRAP (total internal reflec-tion–fluorescence recovery after photobleaching) imaging microscopy (purple); and polarized fluorescence resonance energy transfer (pFRET) microscopy (green).

P E R S P E C T I V E S

NATURE REVIEWS | MOLECULAR CELL BIOLOGY VOLUME 7 | DECEMBER 2006 | 945


Bare glass Metal film

Critical anglepeak

SP emission peak

High aperture objective

b Observed emission intensity at BFP

a Theoretical emission intensity versus polar angle

Bare glass Metal film

Critical angle peak SP emission peak

Water

Glass

system of electrons a ‘natural frequency’ of oscillation. If an external driving force (such as an imposed electromagnetic field) is exactly matched to the natural frequency, the electron distribution is driven into a large resonance. In the case of a metal film between water and glass, the resonance of oscillating electrons emits radiation that takes the form of an evanescent field in the water and propagating light into the glass. SPM takes two distinct forms.

Excitation SPM. This form of SPM involves the propagation of excitation light towards the surface through the glass and the cre ation of an evanescent field, much as in TIR. However, as the surface is coated with a thin metal film, the intensity of the evanescent field is generally weak, except for a sharp and intense maximum at a particular incidence angle11. This angle is sensitive to small changes in the local refractive index in the sample, and thereby creates an image-contrast effect on both fluorescent and non-fluorescent samples.

Emission SPM. This form of SPM has the possibility of providing high sensitivity coupled with a low photobleaching rate. Emission SPM involves emission light (rather than excitation light) as it appears within a fraction of a wavelength from the fluorophore (known as the emission near-field). The emission near-field stimulates an SP resonance in the metal, which then re-emits propagating light into the glass12. A remarkable feature of this light is that it emerges into the glass substrate only in a thin hollow cone distribution (FIG. 2a). The cone’s half angle depends on the emission wavelength and the particular type of metal. In many practical situations, this hollow cone can be completely captured by a sufficiently high aperture objective13, as shown in FIG. 2. Even on bare glass, a hollow cone is produced10,14 with a slightly smaller cone angle, but it is not nearly as distinct as the metal-coating-associated SP cone.

The efficiency by which emitted light is channelled into the SP hollow cone is sensi-tively dependent on the distance between a fluorophore and the metal, with the highest efficiency occurring at distances of 20–200 nm. Fluorophores that are closer than 10 nm suffer almost complete quench-ing of their emission in any direction. Even with the presence of the hollow cone, most of the emitted light actually propagates into the water because of reflection from the metal film. A fluorophore that is positioned within 100 nm of the metal film transfers

Figure 2 | Surface-plasmon emission. Demonstration that surface plasmon (SP) emission can be captured by a standard high aperture objective. The figure is a composite (separated by a dashed vertical line) of two different substrates: on the left, the substrate is bare glass (which cannot support SP); and on the right, the substrate is a 20-nm-thick aluminium film on glass (which can support SP). a | A theoretically generated diagram shows a transverse view of light that is emitted from randomly oriented fluorophores (yellow dot) at 20 nm from a surface and is propagating towards the microscope objective10. The radial distance from the yellow dot to the edge of the coloured region at any particu-lar polar angle (measured from a line that is perpendicular to the surface) is proportional to the total expected number of photons that propagate in that angle in the time interval before a typical fluoro-phore is photobleached. b | An experimentally observed view of the intensity patterns produced by a sample of fluorescent glass beads (8.89-μm diameter) resting on a substrate as the emission light traverses the back focal plane (BFP) of a 1.4 numerical aperture microscope objective. The larger the polar angle at which a propagating emission light ray travels through the glass, the larger the radius at which that ray crosses the BFP. The red arrow is the SP emission peak, propagating at the SP angle. This SP emission peak appears only for fluorophores near a metal film. The blue arrow shows the peak of intensity propagating at the critical angle on bare glass, which is slightly smaller than the SP angle on an aluminium film. The critical angle peak also appears (albeit dimly) through a metal film, depicted here by the white arrowhead.




s-pol p-pol Ratio p:s

10 µ

10 µ

1 2 3

4 5 6

diI dipoleorientations

p-polevanescent field

Substrate/TIR surface

Incident beam Totally reflected beam

a

b

c

energy into the metal so rapidly that the average time a fluorophore resides in the excited state near the metal is significantly reduced. The short lifetime of the excited state effectively increases the number of exci-tations that can be experienced by a fluoro-phore before it becomes photobleached. The transfer rate is also dependent on the emitting fluorophore’s dipole orientation. The SP hollow cone is produced almost entirely by fluorophores with dipoles that are oriented perpendicularly to the interface.

Can all these effects that bear on emission intensity, lifetime and direction be corralled together to increase sensitivity with as low a background as possible? Sensitivity is judged best by the number of emission photons that can be captured by a microscope objective from a fluorophore before it suf-fers a photobleaching event. A number of countervailing factors operate. Shortened lifetime increases sensitivity, because it decreases the photobleaching rate. But, the conversion of energy transfer into metal heating reduces the total number of emitted photons. Emission SP increases the number of photons that can be captured and that enter the glass substrate in the hollow cone, but reflection from the metal film increases the overall fraction of the photons that reflect into the water.

Calculations (on the basis of the theo-retical treatment in REF. 10) show that, for some experimental conditions (for example, a 15-nm-thick silver coating, dipoles oriented perpendicularly to the surface at a distance of 150 nm, and a 1.49 numerical aperture objec-tive), viewing the SP hollow cone increases sensitivity by ~30% over that for bare glass. If one ignores the emission SP hollow cone and instead views the emission propagating into the water, simple reflection from the metal film can provide a more than twofold greater sensitivity. Even in this configuration, SP can be involved — evanescent illumination (which provides the advantages of optical sectioning and consequent background reduction) still requires laser incidence at the excitation SP angle to provide sufficient intensity through the metal film.

Another optical option might be to selectively view just the hollow SP cone with appropriate opaque disks to block non-SP light that is emitted at smaller angles. This approach has no overall sensitivity advan-tages over bare glass (in fact, it allows for viewing of <5% of the photons emitted from dipoles), but it will image just those fluoro-phores that are oriented perpendicularly to the surface, and will discriminate against background light.

Figure 3 | Polarized total internal reflection. a | Evanescent field polarization is shown here as a white, vertical, two-headed arrow that is oriented perpendicularly to the substrate that supports the cells. Polarization predominates if the totally internally reflecting incident light beam is p-polarized (p-pol), which indicates that its polarization is in the plane of incidence (that is, the plane that con-tains both the incident and the totally reflected beam). For diI (3,3′-dioctadecylindocarbocyanine), which is incorporated into membranes with its excitation-dipole moment (that is, the direction of light polarization the fluorophore is most likely to absorb) parallel to the membrane, only membrane regions that are non-parallel to the substrate will be excited by p-pol total internal reflection (TIR). Light shading represents the membrane areas that can become excited, and dark shading represents areas that cannot become excited. TIR that is s-polarized (s-pol) (that is, the evanescent electric field is directed perpendicularly to the plane of incidence; not shown here) is used in a separate exposure to normalize against local fluorophore-concentration variations by taking the ratio of p-pol/s-pol (or p/s) images. b | Living macrophages labelled with diI14, illuminated by an s-pol evanescent field, a p-pol evanescent field and the calculated ratio p/s. c | A time-lapse series (12 s intervals) of p/s ratios on another diI-labelled macrophage that was adhered to a specially treated (opsinized) surface reveals small dynamic undulations (in the central region) that appear as transient bright spots (pos-sibly exocytosis or endocytosis sites) as indicated by arrowheads. A flat zone (around the periphery) appears as an unchanging dark region.




a Intensity c Dissociation rate

b

1

2

3

4

5

6

10 µm

Speckle

0.004 0.4s–1

15

10

5

00 80 160

Inte

nsit

y

Inte

nsit

y

Inte

nsit

y

Inte

nsit

y

Inte

nsit

y

Inte

nsit

y

115

10

5

00 80 160

3

15

10

5

00 80 160

5

15

10

5

0

0 80 160

2

15

10

5

0

0 80 160

415

10

5

0

0 80 160

6

Time (s)Time (s)Time (s)

Time (s) Time (s) Time (s)

Pre-bleach Post-bleach

Last, the strong and highly confined quenching effect at metal surfaces can be used to discriminate against background fluores-cence arising from nonspecific adsorption to the surface1 and possibly selectively detect (or photobleach) fluorophores on only one leaflet of a lipid bilayer. Such special optical properties of metal films might improve detection of a single surface-anchored biomolecule as it proceeds through confor-mational and orientational states while tran-siently interacting with other biomolecules.

Polarized TIR microscopyThe combination of excitation polarization with TIRFM (FIG. 1) allows the visualization of submicroscopic indentations and protru-sions in the plasma membrane of cells and the orientations of single molecules15–18.

Oriented membranes. Polarized TIR requires all of the fluorophore molecules to be simil-arly oriented in the membrane (say, with the excitation dipole always parallel to the membrane surface), thereby providing a well defined excitation-polarization preference. One such fluorophore that is used frequently in cell biology is 3,3′-dioctadecylindocarbo-cyanine (diI)19. By using excitation light that is polarized in the direction perpendicular to the surface, only those regions of the plasma membrane that are not parallel to the substrate will be excited (FIG. 3a). For this purpose, a p-polarized (p-pol) excitation beam with a highly oblique incidence angle (such as that used for TIR) is required. To normalize against spatial-intensity variations that arise from actual place-to-place differ-ences in the fluorophore concentration, an

s-polarized (s-pol) TIR beam is also used to acquire an alternative image of each scene. Actual p-pol, s-pol and p-pol/s-pol (p/s)-ratio images of diI-labelled macrophages can be seen in FIG. 3b. Polarized TIR has also been used to quantify the order of a phospholipid-like probe in an artificial phospholipid bilayer20.

Single molecules. If one views single mole-cules that have been labelled with single fluorophores, the requirement for orienta-tional order among fluorophores becomes irrelevant. Polarized TIR can be used to successfully determine the orientation of single molecules, as was done recently for sparsely labelled filamentous actin17. The capability to follow the orientational changes of single molecules as they proceed through

Figure 4 | TIR–FRAP on human neutrophils. Neutrophils were bathed in a 10-nM solution of the chemoattractant N-formyl-methionyl-leucyl-phenyl-alanyl-lysine labelled with AlexaFluor532 (REF. 26). a | TIR illumination excites fluorescence only from ligand-bound receptors at cell–substrate contact regions around the periphery of each cell. ‘Speckles’ move around rapidly over time: they can either represent local increases of binding-site density or local regions that are closer to the substrate and therefore brighter in appearance. To improve the signal-to-noise ratio and to help overcome the jittering motion of the speckles, each of the images in the stack was smoothed by averaging adjacent pixels in a travelling 0.3-µm2 area (the size of which is

depicted as a circled white square in the bottom-right corner). b | Six num-bered graphs that show place-to-place variations in fluorescence intensity versus time. At t = 0 s, fluorescent ligand was added to the solution. Beginning at t = 90 s, a bright photobleaching flash (shown as a grey bar) was adminis-tered. Graph numbers correspond to numbered locations in panel a. c | The post-bleach recovery rates, which in this case range from 0.004 to 0.4 per s, were observed at every location in the image and were plotted with a logarithmic grey-scale. Locations with intensities that were below a thresh-old were not analysed, which accounts for the black regions outside the cell–substrate contact regions.




CCD channels1Un-polarized

2In-planepolarized

3Normalpolarized

Hg arc

Beam-splitting polarizing cubes

Coloured filters

Polarizingtriple-imagesplitter

Sample

Mixed colouremission Microscope

Two-colour orthogonal polarizations

a

b

1 0

Relative fraction of donor

Relative fraction of acceptor

Type of transfection

Relative fraction of FRET pair

Donor alone Acceptor alone FRET pair alone Donor + acceptor Donor + FRET pair Donor + acceptor+ FRET pair

Acceptor + FRET pair

their functional states is a real advantage over ensemble polarization readings, which report only average orientations over a whole group of molecules, and thereby blur over what might be key distinctions between conformational and orientational states.

TIR–FRAP imagingFluorescence recovery after photobleaching (FRAP) is a standard microscopy technique that turns an unfortunate property of excited molecules — photobleaching — into an

advantage. By confining rapid irrevers-ible photobleaching (typically induced by a bright focused flash of light) to a small region of a sample, the subsequent motion of unbleached molecules in the surroundings can be detected as a recovery of local post-bleaching fluorescence in that region. In combination with TIRFM (FIG. 1), the bleaching flash is localized to the region of the evanescent field, so that only fluorophores within a few hundred nanometres of the surface become bleached.

If bleached surface-bound molecules are in reversible equilibrium with molecules that are further away from the solution surface (that is, molecules are continually binding and unbinding), then exchange gives rise to a post-bleach fluorescence recovery. The rate of that recovery provides a quantitative measure of the kinetic dissociation rate, which is simply the reciprocal of a typical time that a molecule resides at the surface. Such measurements can be done even on liv-ing cells at the cell–substrate contact regions

Figure 5 | Polarized fluorescence resonance energy transfer. a | An unpo-larized mercury arc beam is split by a polarizing cube into two beams with vertical and horizontal polarizations, respectively. One beam passes through a blue filter (for donor and fluorescence resonance energy transfer (FRET)-pair excitation). The other passes through a green filter (for acceptor excita-tion). The two differently coloured and polarized beams are re-superimposed in another polarizing cube before entering the microscope, both illuminat-ing the sample simultaneously. The emitted light (from a combination of free donors, free acceptors and FRET pairs) enters a triple-image splitter, which first separates it by colour (shown as blue–green for donors and orange for acceptors), and then the acceptor emission is further separated into two orthogonal polarizations, indicated here as in the plane and normal to the

plane of the figure. The three images are simultaneously recorded onto seg-ments of the digital camera array in one exposure. Because the spectra of donors, acceptors and FRET pairs are not completely spectrally distinct, each of the three images contains a unique linear combination of contribu-tions from the three species. This mix can be separated by a mathematical procedure involving Cramer’s rule27. b | Processed polarized FRET images of COS cells that are transfected with either donor (cyan fluorescent protein (CFP)), acceptor (citrine, a form of yellow fluorescent protein), FRET pair (CFP at one end, citrine at the other, separated by a 16-amino-acid linker), or combinations of these, are shown. The pseudocolour scale indicates the fraction of the total fluorescent protein at each location that is free donor, free acceptor or FRET pair. CCD, charge-coupled device.




in the evanescent field. FIG. 4a shows human neutrophils with fluorescently tagged chemo-attractant molecules that bind reversibly to specific cell-surface receptors for a finite time, as determined by this technique.

Surface-dissociation rates. Before digital cameras were introduced to cell biology, TIR–FRAP was used only in a single-channel non-imaging mode21–25. As applied to whole images26, post-bleach recovery is observed at each pixel. The rates of fluor escence recovery at each pixel provide the basis for a spatial map of dissociation rates over the substrate-proximal portions of the cell. FIG. 4b,c show an example of a TIR–FRAP experiment that results in a spatial map of fluorescent-chemoattractant-dissociation rates from the cell-surface receptors of a human neutrophil. The surprising result is the great variability of recovery rates from place to place on the same cell.

Surface-association rates. At early time points, fluorescence rapidly increases due to the introduction of fluorescent ligand, even-tually reaching a quasi-steady state (FIG. 4b). Sudden concentration change in the bulk solution provides a perturbation that is dif-ferent than a photobleaching flash, and leads to an evolving change in the surface-bound concentration that depends on both the dissociation rate and the association rate (a measure of the rapidity of surface binding).

Therefore, by combining concentration-jump experiments with FRAP experiments on the same TIR-illuminated sample, both the dissociation rate and the association rate can be deduced at each pixel26.

Polarized FRET microscopyThe polarized fluorescence resonance energy transfer (pFRET) technique combines the photon properties of both colour and polarization with the capability of a photon-excited fluorophore to transfer energy to a different fluorophore only nanometres away (FIG. 1). Its purpose is to detect such neigh-bours, even if they are only transiently close to each other, as is often the case in cell biol-ogy (for example, in certain ligand–receptor interactions). Conventional (non-polarized) FRET microscopy requires at least two expo-sures with different excitation colours on the same sample. Unfortunately, spectroscopic features that might reveal the presence of FRET could be present during the first expo-sure and be completely gone in the second (or vice versa). By contrast, pFRET gathers all the required information in only a single camera exposure.

Illumination in pFRET uses a mix of two colours, each ‘encoded’ with it own unique polarization. For example, blue might be polarized in the ‘north–south’ direction on the sample plane, and green might be polarized in the ‘east–west’ direction. A triple-image splitter in the emission-detection sys-

tem is configured to detect the ‘imprint’ of the excitation polarization in the emission, and thereby determine how much of the emission originates from direct excitation of the accep-tor and how much originates from excitation of the donor through FRET27 (FIG. 5). For example, if all of the fluorescent proteins at a particular pixel are of any one species (donor, acceptor or FRET pair), this pixel will appear red (FIG. 5b), which corresponds to a ‘relative fraction’ of one. The results are usu-ally positive for the expected combination of fluorescent proteins; they never show a false positive for a fluorescent protein that has not been transfected into the cells.

Other combinations, present and futureThe field of microscopy has entered a new frontier in cell biology and biochemistry — that of understanding not just the structure of cells, but also the dynamical, morpho logical and biochemical processes that underlie life. The challenges presented at this frontier are formidable: we need greater spatial resolu-tion, chemical specificity, sensitivity and time resolution. And, all of these must be achieved under conditions that will not kill cells.

The introduction of sensitive digital image detectors and genetic methods of specific labelling in the past two decades has brought together the previously separate fields of microscopic imaging (which is based largely on the refractive, reflective and diffractive properties of photons) and optical spectroscopy (which is based on the spectral, polarized and quantum properties of photons and states of matter). New techniques based on combinations of photon properties have increased the effective resolution to distance scales that are smaller than the wavelength of light and have allowed optical spectroscopic events to be observed in both temporal and spatially variegated detail in living cells. Every newly introduced technique raises the possibil-ity of further combinations with existing techniques in a continuing process of inno-vation. BOX 1 lists some of these emerging possibilities.

Combinatorial microscopy techniques are being applied to living cells to under-stand many questions in cell biology and biochemistry28. What are the structure, function and dynamics of cellular mem-branes, including microdomains29,30 , lipid rafts31,32 and secretory vesicles33,34? How is gene expression manifested as subcel-lular localization and quantification of RNA35,36? What mechanisms drive and regulate protein trafficking37? What are the molecular interactions that are important in

Box 1 | Other combinatorial microscopy techniques

In addition to the microscopy techniques discussed in the main text, other current and future techniques that are based on combinations of photon properties are possible.

Polarized fluorescence recovery after photobleaching imaging. By selectively bleaching an oriented subset of fluorophores with polarized light, the rotational mobility of molecules can be measured, even in samples for which the native orientational distribution is completely isotropic42–51.

Multiphoton TIRF. Multiphoton excitation by intense infrared light produces fluorescence that is proportional to the square (or higher order) of the intensity of the incident light52. Unwanted scattering of the evanescent field by sample inhomogeneities is not bright enough to excite much multiphoton fluorescence, so multiphoton total internal reflection fluorescence (TIRF) might produce a scattering, artefact-free image. An even more complex combination — multiphoton with total internal reflection (TIR) with structured illumination with photobleaching — has been proposed to improve lateral diffusion measurements53.

Time-resolved polarization-stimulated emission imaging. This technique combines the properties of polarization, lifetime and stimulated emission. It has been used to produce spatial maps of the rotational mobility of specific molecules in living cells54.

TIRF with atomic force microscopy. By combining TIRF with atomic force microscopy, micromechanical properties can be directly measured in living cells55,56 and labelled molecules can be located easily57.

Optical trapping with TIRF. This approach will produce an evanescent optical-gradient force that should be able to pull subcellular organelles towards a membrane surface where they can interact or fuse. The analogue of a ‘photon pressure’ force might be directed parallel to the substrate in the plane of incidence; it might be cancelled out by the use of oppositely directed TIR beams.




signal-transduction pathways38,39? What are the biochemical and mechanical properties of molecular motors, including those that mediate muscle contraction, cell migration, intracellular organization and gene expres-sion40,41? What are the dynamic details of DNA transcription? To begin to answer these biomolecular questions, we must understand more about subcellular localization, orienta-tion, and lateral and rotational motions of biomolecules, as well as the dynamics of intermolecular interactions. As combinations of photon properties that are identified in FIG. 1 are customized to control the time scale of studies, to focus on particular types of motion (that is, rotational, lateral or axial),

to identify molecular interactions over vary-ing distances, to allow improved selectivity of signal over background, and to facilitate the distinction of individual probes in mixtures, these as-yet untested combinations will probably promote further advances.Daniel Axelrod has been at the Department of Physics

& Biophysics Research Division, University of Michigan, Ann Arbor, Michigan 48109, USA.

Geneva M. Omann has been at the Department of Biological Chemistry & Department of Surgery,

University of Michigan as well as the Veterans Administration Medical Center, Ann Arbor, Michigan

48109, USA.

Correspondence to D.A. e-mail: [email protected]

doi:10.1038/nrm2062

1. Fulbright, R. M. & Axelrod, D. Dynamics of nonspecific adsorption of insulin to erythrocyte membrane. J. Fluor. 3, 1–16 (1993).

2. Abney, J. R., Scalettar, B. A. & Thompson, N. L. Evanescent interference patterns for fluorescence microscopy. Biophys. J. 61, 542–552 (1992).

3. Huang, Z. & Thompson, N. L. Theory for two-photon excitation in pattern photobleaching with evanescent illumination. Biophys. Chem. 47, 241–249 (1993).

4. Neil, M. A. A., Juskaitis, R. & Wilson, T. Method of obtaining optical sectioning by using structured light in a conventional microscope. Optics Lett. 22, 1905–1907 (1997).

5. Schaefer, L. H., Schuster, D. & Schaffer, J. Structured illumination microscopy: artefact analysis and reduction utilizing a parameter optimization approach. J. Microsc. 216, 165–174 (2004). ckwell Publishing, Ltd.

6. Gustafsson, M. G. L., Agard, D. A. & Sedat, J. W. Doubling the lateral resolution of wide-field fluorescence microscopy using structured illumination. Proc. SPIE 3919, 141–150 (2000).

7. Stout, A. L. & Axelrod, D. Evanescent field excitation of fluorescence by epi-illumination microscopy. Appl. Opt. 28, 5237–5242 (1989).

8. Axelrod, D. Selective imaging of surface fluorescence with very high aperture microscope objectives. J. Biomed. Optics 6, 6–13 (2001).

9. Gustafsson, M. G. L. Nonlinear structured-illumination microscopy: wide-field fluorescence imaging with theoretically unlimited resolution. Proc. Natl Acad. Sci. USA 102, 13081–13086 (2005).

10. Hellen, E. H. & Axelrod, D. Fluorescence emission at dielectric and metal-film interfaces. J. Opt. Soc. Am. B 4, 337–350 (1987).

11. Axelrod, D. Total internal reflection fluorescence microscopy in cell biology in Biophotonics, Part B (Methods in Enzymology) Vol. 361 (eds Marriott, G. & Parker, I.) 1–33 (Academic Press, San Diego, 2003).

12. Lakowicz, J. R. Radiative decay engineering 5: metal-enhanced fluorescence and plasmon emission. Anal. Biochem. 337, 171–194 (2005).

13. Mattheyses, A. L. & Axelrod, D. Fluorescence emission patterns near glass and metal-coated surfaces investigated with back focal plane imaging. J. Biomed. Optics 10, 054007 (2005).

14. Burghardt, T. P. & Thompson, N. L. Effect of planar dielectric interfaces on fluorescence emission and detection. Evanescent excitation with high-aperture collection. Biophys. J. 46, 729–737 (1984).

15. Sund, S. E., Swanson, J. A. & Axelrod, D. Cell membrane orientation visualized by polarized total internal reflection fluorescence. Biophys. J. 77, 2266–2283 (1999).

16. Osborne, M. A. Real-time dipole orientational imaging as a probe of ligand–protein interactions. J. Phys. Chem. B 109, 18153–18161 (2005).

17. Forkey, J. N., Quinlan, M. E. & Goldman, Y. E. Measurement of single macromolecule orientation by total internal reflection fluorescence polarization microscopy. Biophys. J. 89, 1261–1271 (2005).

18. Wang, C., Liu, L., Wang, G. Y. & Xu, Z. Z. A new method for determining dipole moment orientation of single molecules. Chinese Phys. Lett. 21, 843–845 (2004).

19. Axelrod, D. Carbocyanine dye orientation in red cell membrane studied by microscopic fluorescence polarization. Biophys. J. 26, 557–574 (1979).

20. Thompson, N. L., McConnell, H. M. & Burghardt, T. P. Order in supported phospholipid monolayers detected by the dichroism of fluorescence excited by polarized evanescent illumination. Biophys. J. 46, 739–747 (1984).

21. Thompson, N. L., Burghardt, T. P. & Axelrod, D. Measuring surface dynamics of biomolecules by total internal reflection with photobleaching recovery or correlation spectroscopy. Biophys. J. 33, 435–454 (1981).

22. Burghardt, T. P. & Axelrod, D. Total internal reflection/fluorescence photobleaching recovery study of serum albumin adsorption dynamics. Biophys. J. 33, 455–468 (1981).

23. Hellen, E. & Axelrod, D. Kinetics of epidermal growth factor/receptor binding on cells measured by total internal reflection/fluorescence recovery after photobleaching. J. Fluor. 1, 113–128 (1991).

24. Fulbright, R. M. & Axelrod, D. Dynamics of nonspecific adsorption of insulin to erythrocyte membrane. J. Fluor. 3, 1–16 (1993).

Glossary

Dipole orientationIncident light of an appropriate colour can be absorbed by a fluorophore, but the absorption is most efficient if the light is polarized along a particular axis that is fixed relative to the fluorophore, known as the absorption dipole orientation. Fluorescence (emitted as the fluorophore returns to the ground state) is also polarized, generally along an axis that is fixed relative to the fluorophore, known as the emission dipole orientation.

InterferenceIf two laser beams from the same laser, split by a mirror or prism, are made to intersect, in some places in the region of intersection the sine-wave crests of one beam will always add to the crests of the other, and likewise for the troughs. These places will experience larger electric field oscillations than either beam separately, and they will have enhanced brightness. Other places will have crests of one beam arriving at the same time as troughs of the other and these places will experience a cancellation of the electric field, leading to relative darkness. The pattern of bright and dark is known as an interference pattern. Only beams of the same polarization will interfere. If the two beams are broad and collimated, the interference pattern will be a series of dark and bright stripes or planes, the spacing of which is determined by the relative angle of intersection. Typically, the spacing is measured by the distance from one minimum (a dark region) to the next.

PhotobleachingA fluorophore in the excited state is typically less chemically stable than one in the ground state. It is more easily attacked by oxygen and also more likely to break on its own. Photochemical reactions that involve excited molecules can generate by-products that can even destroy nearby fluorophores in the ground state. The result is a progressive and permanent loss of viable fluorophores and a dimming of fluorescence, which is known as photobleaching.

Plane of incidenceWhen light encounters a planar interface, some reflects and some refracts (for sub-critical angle incidence) or forms an evanescent field (for super-critical angle incidence). The incident beam, the reflected beam and the refracted beam all lie in the same plane of incidence.

PolarizationA coherent laser beam can be viewed as a travelling sine wave of electric field with the field pointing one way in the

crests of the sine wave and the opposite way in the troughs, and always perpendicular to the direction of propagation. The electric field direction is referred to as the polarization.

p-polarizationIf the incident beam’s polarization lies in the plane of incidence, the reflected beam and refracted beam (or evanescent field) are also polarized in that plane; this is known as p-polarization (p-pol).

ResolutionThis is the minimum distance that two points of light on the sample must be separated so that their two images can be distinguished from each other. The classic Raleigh criterion is usually used: 0.61 multiplied by (the wavelength of light) divided by (the numerical aperture of the objective). Modern image analysis can improve on this and distinguish separate points that are considerably closer together than specified by the Raleigh criterion.

SaturationIf the excitation light is bright, fluorophores will become re-excited shortly after they briefly return to the ground state, thereby spending most of their time in the excited state. This situation, which is known as saturation, depletes the number of fluorophores in the ground state that are available for further excitation.

s-polarizationIf the incident beam’s polarization is perpendicular to the plane of incidence, then so are the reflected and refracted beams (or evanescent field); this is known as s-polarization (s-pol).

Total internal reflection, evanescent field and critical angleA light beam that obliquely (that is, non-perpendicularly) approaches an interface with a less dense medium (such as water) can totally internally reflect if the angle of incidence (measured from a line that is perpendicular to the surface) is larger than a well defined critical angle. However, an evanescent field does penetrate into the less dense medium and propagates along the surface. The evanescent-field intensity decays exponentially in the direction perpendicularly away from the interface, generally with a characteristic distance of tens to hundreds of nanometers, depending on the angle of incidence. Fluorophores that reside in the evanescent field emit photons at a rate that exponentially decays with distance from the interface.




25. Pisarchick, M. L., Gesty, D. & Thompson, N. L. Binding kinetics of an anti-dinitrophenyl monoclonal Fab on supported phospholipid monolayers measured by total internal reflection with fluorescence photobleaching recovery. Biophys. J. 63, 215–223 (1992).

26. Chang, P. S., Axelrod, D., Omann, G. M. & Linderman, J. J. G protein threshold behavior in the human neutrophil oxidant response: measurement of G proteins available for signaling in responding and nonresponding subpopulations. Cell. Signal. 17, 605–614 (2005).

27. Mattheyses, A. L., Hoppe, A. & Axelrod, D. Polarized fluorescence resonance energy transfer microscopy. Biophys. J. 87, 2787–2797 (2004).

28. Lippincott-Schwartz, J., Snapp, E. & Kenworthy, A. Studying protein dynamics in living cells. Nature Rev. Mol. Cell Biol. 2, 444–456 (2001).

29. Gaus, K., Zech, T. & Harder, T. Visualizing membrane microdomains by Laurdan 2-photon microscopy. Mol. Memb. Biol. 23, 41–48 (2006).

30. Lagerholm, B. C., Weinreb, G. E., Jacobson, K. & Thompson, N. L. Detecting microdomains in intact cells. Annu. Rev. Phys. Chem. 56, 309–336 (2005).

31. Kenworthy, A. K., Nichols, B. J., Remmert, C. L., Hendrix, G. M., Kumar, M., Zimmerberg, J. & Lippincott-Schwartz, J. Dynamics of putative raft-associated proteins at the cell surface. J. Cell Biol. 165, 735–746 (2004).

32. Rao, M. & Mayor, S. Use of Forster resonance energy transfer microscopy to study lipid rafts. Biochim. Biophys. Acta 1746, 221–233 (2005).

33. Scalettar, B. A. How neurosecretory vesicles release their cargo. Neuroscientist 12, 164–176 (2006).

34. Allersma, M. W., Bittner, M. A., Axelrod, D. & Holz, R. W. Motion matters: secretory granule motion adjacent to the plasma membrane and exocytosis. Mol. Biol. Cell 17, 2424–2438 (2006).

35. Santangelo, P., Nitin, N. & Bao, G. Nanostructured probes for RNA detection in living cells. Annals Biomed. Eng. 34, 39–50 (2006).

36. Dirks, R. W. & Tanke, H. J. Advances in fluorescent tracking of nucleic acids in living cells. Biotechniques 40, 489–496 (2006).

37. Brown, D. Imaging protein trafficking. Nephron. Exp. Nephrol. 103, e55–e61 (2006).

38. Kiyokawa, E., Hara, S., Nakamura, T. & Matsuda, M. Fluorescence (Forster) resonance energy transfer imaging of oncogene activity in living cells. Cancer Sci. 97, 8–15 (2006).

39. Zaccolo, M., Cesetti, T., Di Benedetto, G., Mongillo, M., Lissandron, V., Terrin, A. & Zamparo, I. Imaging the cAMP-dependent signal transduction pathway. Biochem. Soc. Trans. 33, 1323–1326 (2005).

40. Bai, L., Santangelo, T. J. & Wang, M. D. Single-molecule analysis of RNA polymerase transcription. Annu. Rev. Biophys. Biomol. Struct. 35, 342–360 (2006).

41. Rosenburg, S. A., Quinlan, M. E., Forkey, J. N. & Goldman, Y. E. Rotational motions of macromolecules by single-molecule fluorescence microscopy. Acc. Chem. Res. 38, 583–593 (2005).

42. Smith, L. M., McConnell, H. M., Smith Baron, A. & Parce, J. W. Pattern photobleaching of fluorescent lipid vesicles using polarized laser light. Biophys. J. 33, 139–146 (1981).

43. Yoshida, T. M. & Barisas, B. G. Protein rotational motion in solution measured by polarized fluorescence depletion. Biophys. J. 50, 41–53 (1986).

44. Scalettar, B., Selvin, P., Axelrod, D., Hearst, J. & Klein, M. P. A fluorescence photobleaching study of the microsecond reorientational motions of DNA. Biophys. J. 53, 215–226 (1988).

45. Velez, M. & Axelrod, D. Polarized fluorescence photobleaching recovery for measuring rotational diffusion in solutions and membranes. Biophys. J. 53, 575–591 (1988).

46. Timbs, M. M. & Thompson, N. L. Slow rotational mobilities of antibodies and lipids associated with substrate-supported phospholipid monolayers as measured by polarized fluorescence photobleaching recovery. Biophys. J. 58, 413–428 (1990).

47. Velez, M., Barald, K. F. & Axelrod, D. Rotational diffusion of acetylcholine receptors on cultured rat myotubes. J. Cell Biol. 110, 2049–2059 (1990).

48. Scalettar, B., Selvin. P., Axelrod, D., Hearst, J. & Klein, M. P. Rotational diffusion of DNA in agarose gels. Biochemistry 29, 4790–4798 (1990).

49. Selvin, P., Scalettar, B., Axelrod, D., Langmore, J. P., Hearst, J. & Klein, M. P. Rotational diffusion of DNA in intact nucleii. J. Mol. Biol. 214, 911–922 (1990).

50. Yuan, Y. & Axelrod, D. Subnanosecond polarized fluorescence photobleaching: rotational diffusion of acetylcholine receptors on developing muscle cells. Biophys. J. 69, 690–700 (1995).

51. Abney, J. R., Cutler, B., Fillbach, M. L., Axelrod, D. & Scalettar, B. A. Chromatin dynamics in interphase nucleii and its implications for nuclear structure. J. Cell Biol. 137, 1459–1468 (1997).

52. Oheim, M. & Schapper, F. Non-linear evanescent-field imaging. J. Phys. D Appl. Phys. 38, R185–R197 (2005).

53. Huang, Z & Thompson, N. L. Theory for two-photon excitation in pattern photobleaching with evanescent illumination. Biophys. Chem. 47, 241–249 (1993).

54. Buehler, Ch., Dong, C. Y., So, P. T. C. & Gratton, E. Time-resolved polarization imaging by pump-probe (stimulated emission) fluorescence microscopy. Biophys. J. 79, 536–549 (2000).

55. Mathur, A. B., Truskey, G. A. & Reichert W. M. Atomic force and total internal reflection fluorescence microscopy for the study of force transmission in

endothelial cells. Biophys. J. 78, 1725–1735 (2000).

56. Trache, A. & Meininger, G. A. Atomic force multi-optical imaging integrated microscope for monitoring molecular dynamics in live cells. J. Biomed. Optics 10, 064023 (2005).

57. Yamada, T., Afrin, R., Arakawa, H. & Ikai, A. High sensitivity detection of protein molecules picked up on a probe of atomic force microscope based on the fluore-scence detection by a total internal reflection fluore-scence microscope. FEBS Lett. 569, 59–64 (2004).

Competing interests statementThe authors declare no competing financial interests.

FURTHER INFORMATIONDaniel Axelrod’s homepage: http://www.physics.lsa.umich.edu/department/directory/bio.asp?ID=19Geneva M. Omann’s homepage: http://www.biochem.med.umich.edu/biochem/archive/faculty_research/omann.html Access to this links box is available online.

I N N OVAT I O N

Functional and quantitative proteomics using SILACMatthias Mann

Abstract | Researchers in many biological areas now routinely characterize proteins by mass spectrometry. Among the many formats for quantitative proteomics, stable-isotope labelling by amino acids in cell culture (SILAC) has emerged as a simple and powerful one. SILAC removes false positives in protein-interaction studies, reveals large-scale kinetics of proteomes and — as a quantitative phosphoproteomics technology — directly uncovers important points in the signalling pathways that control cellular decisions.

Proteins are in direct control of almost all cellular processes, and post-genomic biology will not reach its potential until we have tools to study proteins on a large scale. Sadly, proteomics has lagged far behind DNA-based technologies, mainly because there are no protein-analysis methods similar to oligonucleotide hybridization, amplification and sequencing. This lag is diminishing rapidly, however, due to increasingly powerful mass spectrometry (MS)-based technologies1. Mass spectro-meters can sequence thousands of peptides from complex mixtures in an automated manner2. The importance of quantitation in proteomics has recently become appreciated, and technologies such as isotope-encoded affinity tags (ICAT)3 have generated wide-spread interest. Much of this interest is still focused on the determination of the relative levels of expression of, ideally, all proteins between two cell or tissue states (expression proteomics). Expression proteomics has also been the elusive goal of the two-dimensional

gel electrophoresis community and has essentially the same principle at the protein level as microarrays have at the mRNA level (transcriptomics).

However, proteomics — unlike transcrip-tomics — is not limited to measuring whole cell or tissue expression levels. Here, I argue that the most important contribu-tions of quantitative proteomics to biological understanding will come from its unique capability to determine changes in function-ally relevant ‘sub-proteomes’. Although many of the strategies described here can be, and sometimes have been, implemented with other quantitative-proteomics techniques, I also argue that stable-isotope labelling by amino acids in cell culture (SILAC) com-bined with sophisticated mass-spectrometric and bioinformatic technology is particularly well suited to reinvent or enhance bio-chemistry-based approaches. This argument will be made, in part, by describing success-ful applications of SILAC to a wide range of biological problems.




PERSPECTIVES - Kanazawa Universitybiophys.w3.kanazawa-u.ac.jp/...OpticalMicroscopy-NatureCell-2006.pdf · 944 | DECEMBER 2006 ... Surface-plasmon microscopy (SPM) com- ... NATURE

Documents