dynamic multiphoton imaging: a live view from cells to systems › ... › uploads › 2017 › 04...

The theoretical concept of multipho-ton excitation was first proposed byphysicist Maria Göppert-Mayer in1931 (11) but was not experimentallyproven until the invention of the laser30 years later, and even then itrequired decades of further develop-ment in laser technology to becomeof practical utility. The fields of physi-cal chemistry and physics were thefirst to apply this technique (12, 33),and biological research has only morerecently embraced multiphoton exci-tation to image living cells (7, 8).Biological uses of multiphoton micro-scopy have been extensively reviewed(e.g., Refs. 9 and 29). Here, we pro-vide an introduction to its operatingprinciples, discuss its advantages andlimitations, and illustrate some recentapplications that bring to light thestrengths of this technology.

Fluorescence imaging is widely usedin biology for morphological and func-tional studies, but conventional (single-photon) fluorescence microscopyyields poor images, because in-focusstructures are washed out by fluores-cence outside the plane of focus. Thiscan be mitigated by cutting thin sec-tions (equal to or less than the width ofa single cell), but that approach isclearly inapplicable for living tissue.The introduction of confocal microsco-py thus represented an importantadvance, providing an “optical-sec-tioning” effect by using a pinhole aper-ture to reject out-of-focus fluorescence(41). Nevertheless, confocal microsco-py still suffers from drawbacks, includ-ing limited (a few tens of micrometers)imaging depth into scattering biologi-cal tissues, photodamage owing to theuse of short (high-energy) excitationwavelengths, and photobleaching. The

subsequent development of multipho-ton excitation greatly minimized theseproblems while sharing the optical-sec-tioning ability of confocal microscopy.

How Does MultiphotonMicroscopy Work?

The principle of multiphoton excita-tion is elegantly simple. In the mostcommonly used case of two-photonmicroscopy, a fluorophore molecule isexcited by the nearly simultaneousabsorption (within 10–18 s) of two pho-tons, each about twice the wavelength(half the energy) required for single-photon excitation. Likewise, three-photon excitation results from nearlysimultaneous absorption of three pho-tons, each having approximately one-third the energy needed for excitationof the fluorophore. The resulting fluo-rescent emission is then proportionalto the square of the excitation intensi-ty in two-photon absorption (or thirdpower in the case of three-photonexcitation). This nonlinear relationshipprovides the optical-sectioning abilityof multiphoton imaging. Laser lightfocused on the specimen through amicroscope objective has a high pho-ton density in the focal point, but pho-ton density, and therefore fluores-cence excitation, falls rapidly awayfrom this point. Fluorescence is gener-ated only at the focal spot, and, unlikein confocal microscopy, there is essen-tially no out-of-focus light to reject(FIGURE 1A). A remaining problem isthat multiphoton excitation requiresenormously high light intensities that,if continuous, would almost instantlyvaporize the specimen. The trick is touse lasers that produce amazinglybrief (~10–13 s) pulses at a high repeti-

tion rate (typically ~80 MHz), thus gen-erating very high instantaneous ener-gy but low average energy (FIGURE1B). Indeed, the adoption of multipho-ton microscopy by biologists haslargely followed progressive improve-ments in reliability and user friendli-ness of such “femtosecond” lasers.

In terms of optical-sectioning abilityand image resolution, multiphotonmicroscopy has little advantage overconfocal microscopy (6). However,several major advantages accrue sec-ondarily from the use of long excita-tion wavelengths:

1) Light scattering declines steeplywith increasing wavelength, so thatinfrared light (700–1,000 nm) usedfor two-photon imaging penetratesmuch deeper into tissue (~500 �m)than the equivalent blue light(350–500 nm) used for one-photonexcitation (8, 41).

2) Infrared light causes negligiblephotodamage or phototoxicity tocells. This is particularly advan-tageous for fluorophores and en-dogenous proteins that wouldnormally require excitation at short(<300 nM) and highly damagingultraviolet wavelengths but can bemade to fluoresce by three-photonexcitation at infrared wavelengthsthree times longer (18, 44).

3) Photobleaching is restricted to theplane of focus.

4) The two-photon excitation spectra ofmost fluorophores are broader thanfor one-photon excitation, so a singlemultiphoton excitation wavelengthcan be used to simultaneously excitemultiple fluorophores with distinctemission wavelengths.

15

Dynamic Multiphoton Imaging: A LiveView from Cells to Systems

1548-9213/05 8.00 ©2005 Int.Union.Physiol.Sci./Am.Physiol.Soc

Grace E. Stutzmann and Ian ParkerUniversity of California-Irvine, Irvine, California

[email protected]

Leaps in scientific technology often occur at the interface of seemingly disparate dis-

ciplines. This holds true with the recent application of multiphoton microscopy to the

biological sciences, leading to a new generation of imaging-based studies extend-

ing from the tracking of individual molecules within living cells to the observation of

whole organisms.

EMERGING TECHNOLOGIESPHYSIOLOGY 20: 15–21, 2005; 10.1152/physiol.00028.2004

on April 4, 2008

physiologyonline.physiology.orgD

ownloaded from

http://physiologyonline.physiology.org

bleaching (FRAP; Ref. 31). Anotherexample is fluorescence lifetimeimaging (FLIM), which measureschanges in the lifetime of a fluor-ophore probe that depend on its localenvironment (e.g., pH) but areindependent of concentration. Forexample, although DNA and RNAmolecules have a similar affinity fornucleic acid stains, they can readily bedistinguished in living and apoptoticcells by virtue of differences influorescence lifetimes of certain nu-cleic acid dyes when bound to DNAcompared with RNA (39).

FCS and FLIM are both possibleusing conventional (single-photon)microscopy but benefit from theincreased depth penetration and min-imal phototoxicity of multiphotonimaging. Beyond this, the femtosec-ond laser pulses required for multi-photon microscopy enable imagingtechniques that are not possible byother means. Thus second harmonicgeneration (SHG) microscopy usesnonlinear optical properties that

Novel Imaging Modalities

Multiphoton microscopy is most com-monly used to acquire three-dimen-sional views of specimens labeled witha fluorescent probe but can also beused advantageously in conjunctionwith other microscopy techniques.One example is fluorescence cor-relation spectroscopy (FCS), in whichfluctuations in fluorescence intensity ofa small number of molecules within thefocal spot are used to measurediffusion kinetics of individual mol-ecules, protein-protein interactions,and molecular concentrations (15, 32).Applications of multiphoton FCSinclude analysis of aggregated �-amyloid peptides in human cerebro-spinal fluid, multimerization of voltage-sensitive Ca2+ channels, and alterationsin signaling proteins after changes inneuronal plasticity (24). FCS provides aquantitative, molecular assay in livecells and represents a markedadvancement over existing quanti-tative diffusion-based approaches suchas fluorescence recovery after photo-

emerge at the very high energies dur-ing focused laser pulses. However,rather than being based on absorptionof light energy by fluorescent mole-cules, SHG uses inherent properties ofhighly ordered structures that inher-ently function as frequency doublersof the excitation light. For example,infrared (900 nm) femtosecond pulsesresult in deep blue light at exactlyone-half the wavelength (450 nm) trav-eling in the same direction as the inci-dent light. SHG is well suited for imag-ing of ordered structures (e.g., colla-gen and myosin) deep within tissuewithout the need for any exogenousprobes (5), provides information aboutmolecular orientation and symmetries(27), and can be applied simultaneous-ly with multiphoton fluorescenceimaging to provide additional informa-tion about the specific molecularsource of the SHG (27, 45). Recentdevelopments of functional SHGprobes, such as membrane voltagesensors with uniquely high sensitivity(30), hold further promise.

16

EMERGING TECHNOLOGIES

PHYSIOLOGY • Volume 20 • February 2005 • www.physiologyonline.org

Single photon Multiphoton

Focal spot

A Spatial compression of photons by objective lens B Temporal compression of photons duringfemtosecond pulses

Excitation Emission Excitation EmissionTime

Photons

Lens

Time

Continuous laser

Femtosecond-pulsed laser100 fs

10 ns

FIGURE 1. Principles and imaging modalities of multiphoton microscopyA: demonstration of the differences between single-photon vs. multiphoton laser excitation. In both cases, the spatial compression of photons by anobjective lens is greatest in the focal spot, and the photon density rapidly decreases away from this region. However, with conventional single-pho-ton excitation, fluorophore excitation occurs throughout the beam path, resulting in fluorescence emission above and below the plane of focus.Multiphoton excitation restricts emission to the focal spot, owing to the quadratic relationship between excitation intensity and fluorescence emis-sion. Photographs illustrate fluorescence in a cuvette of fluoresceine resulting from focused laser beams using single-photon excitation at 488 nm(left) and two-photon excitation with 780-nm, femtosecond pulses (right). Figure adapted from Ref. 4 with permission (see http://www.nature.com).B: temporal compression of photons into 100-fs packets achieves the high instantaneous power needed for multiphoton excitation, with an averagepower not much greater than a continuous laser beam as would be used for confocal imaging.

on April 4, 2008


ownloaded from


Finally, we note that multiphotonmicroscopy is not limited merely toimaging but can be applied to per-turb biological tissues with very highspatial and temporal specificity.Examples include the localized“uncaging” of biologically activecompounds such as neurotransmittersor intracellular messengers from pho-tolabile precursors to reveal inter- andintracellular ionic communication (34);the three-photon induction of ared/green color change of a red fluo-rescent protein (DsRed) as a way ofoptically “highlighting” individualcells or subcellular regions (FIGURE2A; Ref. 21); and the selective abla-tion of individual organelles, and even

the use of the femtosecond laserbeam as an optical microtome to con-struct three-dimensional images ofentire fixed organs or organisms bysequentially imaging sections sometens of micrometers thick and then(by turning up the laser power) ablat-ing that section to expose fresh tissueunderneath (1, 38).

Biological Applications

Multiphoton imaging is not a panaceafor all microscopy problems, and formany purposes confocal microscopymay be preferable (and lessexpensive), such as in protocols inwhich rapid exchange between ex-

citation wavelengths is needed (e.g.,ratiometric imaging). However, itsinherent optical sectioning, deeperpenetration in scattering tissue, andreduced phototoxicity are particularlyapplicable to dynamic imaging ofliving specimens. Here, we illustrateapplications at levels from thesubcellular to entire organisms thatexemplify the advantages of multi-photon microscopy.

Subcellular/organelles

Neurobiologists were among the firstto harness multiphoton microscopyfor biological studies. A key functionalcomponent in the nervous system isthe dendritic spine, a tiny structure

17PHYSIOLOGY • Volume 20 • February 2005 • www.physiologyonline.org


ng

s

A DsRed-expressing HEK cells

Single photon

L1

50 µM

20%

1 s

L 2/3

231–264 ms231–264 ms 429–561 ms

Multiphoton

B Dendritic arborization in mouse barrel cortex

C Ca2+-dependent fluorescence D Ca2+ transients in apical dendrites

Day 2 Day 3 Day 4 Day 5 Day 6

pre 33–132 ms

FIGURE 2. Multiphoton applications for invivo and in vitro cellular and subcellularimagingA: application of the multiphoton “greening tech-nique” for labeling subcellular regions. Image pairsshow DsRed-expressing human embryonic kidney(HEK) cells imaged by conventional (single-photon)epifluorescence microscopy, before (upper) and after(lower) greening the cytoplasm (left) or the nucleus(right) by local exposure to focused, 760-nm, fem-tosecond laser pulses. The color change arises froma selective three-photon bleaching of the mature,red form of DsRed, resulting in reduced red fluores-cence and (via a reduction in Forster resonanceenergy transfer) enhanced fluorescence of the imma-ture green species. Figure adapted from Ref. 21 withpermission (see http://www.nature.com). B: Long-term two-photon imaging of the same dendriticarborization within the barrel cortex of a mouse.Arrows indicate examples of dendritic spines thatchanged morphology (yellow) or appeared andretracted (blue, red) throughout an 8-day period.Figure adapted from Ref. 37 with permission (seehttp://www.nature.com). C: subcellular distributionof Ca2+-dependent fluorescence in a fura 2-loaded

pyramidal neuron in an in vitro slice preparation imaged at different times following photoreleaseof the intracellular messenger inositol 1,4,5-trisphosphate. Figure adapted from Ref. 36 with per-mission. D: Ca2+ transients elicited by single action potentials along apical dendrites of a neuronvisualized 400 �m into the brain of a living rat using two-photon microscopy. The image shows aside view of resting fluorescence in a neuron loaded with a fluorescent Ca2+ indicator, reconstruct-ed from an image stack acquired at increasing focal distances. Traces show changes in Ca2+-dependent fluorescence recorded at adjacent depths in image in response to single action poten-tials. Figure adapted from Ref. 40 with permission.

on April 4, 2008


ownloaded from


intracellular stores. Although visual-izing cellular activities, such as ion fluxor membrane voltage changes, canbe accomplished by conventionalfluorescence techniques, multipho-ton excitation facilitates functionalimaging of individual cells in thicktissue (FIGURE 2C; Ref. 34) and inintact animals (FIGURE 2D; Ref. 40).Yet a further advantage is nicelyillustrated in recent studies usingmultiphoton excitation in retinal star-burst amacrine cells that are involvedin the processing of visual informationyet are not activated by infraredwavelengths outside of the visiblespectrum (10). Visible wavelengthswere used to evoke patternedstimulation at the retinal photo-receptors, whereas multiphotonexcitation (using infrared light) wasused concurrently to excite afluorescent Ca2+ indicator loaded intoamacrine cells, revealing the func-tional responses to the patternedstimulation without complications dueto photostimulation by the excitationlight itself.

Imaging within intact organs

For many years biological researchhas pursued a reductionistic trend,with an increasing focus on genomicsand proteomics. Multiphoton micro-scopy now offers an important tooltoward a more integrative approach,because its ability to image functionand morphology at cellular, and evenmolecular levels deep within intactorgans and living animals provides ameans to unify molecular, cellular andorganismal studies. This approach hasfar-reaching applications for under-standing how multiple cells and celltypes interact to determine thefunction of an entire organ and canprovide insight into disease mech-anisms and aid development oftherapeutic strategies. An example isthe application of multiphotonimaging to open a new window ontothe immune system (FIGURE 3A; seeRef. 4 for a review). Activation of theimmune response depends on acomplex choreography of interactionsbetween different cell types, yetdespite a wealth of molecular

about a micrometer in size that formsthe postsynaptic element mediatingcommunication between neurons.Much interest focuses on the chemicaland electrical signaling mechanisms inspines and on the plastic changes intheir distribution, morphology, andfunction that may underlie learningand memory. However, it has beendifficult to image these dynamicprocesses in live preparations.Neurons are sensitive to photo-damage, brain tissue is highlyscattering, regions of interest may liehundreds of micrometers below thesurface of the intact brain, andalthough it is possible to maintainviable slices of brain, these must berelatively thick to preserve functionalconnections. The introduction ofmultiphoton microscopy thus offeredenormous advantages in terms ofhigh-resolution, deep, and non-injurious imaging. Recent examples ofmultiphoton imaging in 300- to 400-�m-thick brain slices demonstrate themorphology, unique signaling dynam-ics, and rapid motility of dendriticspine head and neck regions (14, 19).Most spectacularly, it has beenpossible to resolve individual den-drites and spines in living animals bycreating a viewing window into thebrain by removing or thinning theskull of mice expressing a greenfluorescent protein in a smallsubpopulation of cortical neurons(37). These could then be imagedrepeatedly over several weeks,revealing that spines appear anddisappear frequently in the adultcortex and suggesting that thesechanges might underlie adaptiveremodeling of neural circuits inresponse to sensory experience(FIGURE 2B).

Dynamic cellular imaging

In addition to morphological studies,multiphoton imaging of cellularactivity can be accomplished with theuse of various fluorescent probes. Forexample, Ca2+-indicator dyes canprovide detailed spatial and temporalinformation about Ca2+ entry throughvoltage-gated channels in excitablecells or patterns of Ca2+ release from

knowledge, little is known of the cell-cell interactions deep within nativelymphoid tissue. Again, the depthpenetration and minimal photo-damage of multiphoton microscopyare crucial and, in conjunction withuse of established fluorescent cell-tracker dyes to label adoptivelytransferred lymphocytes, permit six-dimensional imaging (x, y, z, time,intensity, and emission wavelength)into both excised lymph nodes andnodes in an anesthetized mouse.Time-lapse movies reveal strikinglydynamic movements and cellularinteractions within the in vivo setting,in marked contrast to the static viewof lymphocytes in culture, and arebeginning to provide mechanisticinsights into the processes by which Tcells search for antigen andsubsequently become activated byrepeated contacts with antigen-presenting cells (2, 23, 26). Theadvantages of multiphoton imagingare further highlighted becauseindividual naïve T cells can befollowed within the lymph node forhours without loss of viability ormotility (25, 26), whereas analogousstudies using (single-photon) confocalimaging concluded that these cellswere essentially immotile (35), a resultthat may reflect photodamage by theshorter-wavelength excitation.

Imaging whole organisms/embryos

At even larger scales, multiphotonmicroscopes can be “zoomed out” tocapture an entire embryo duringdevelopment while maintaining high(subcellular) spatial and temporalresolution over long periods (days toweeks). Transparent organisms such asCaenorhabditis elegens, zebrafish,and Drosophila are ideal candidatesfor such studies, and “in toto” imagingof entire embryos has revealedexpression patterns at the cellular andgenomic levels (FIGURE 3B; Ref. 22).Multiphoton imaging enables long-term observation of biological activityby providing the ability to repeatedlytrack and monitor cells deep within agrowing embryo or organism whileminimizing photodamage to cellstructure and DNA.

18



on April 4, 2008


ownloaded from


Future Prospects

Multiphoton excitation is relativelynew in the biological sciences, andfuture developments will likelyimprove its utility and overcome someexisting shortcomings. These includeoptimization of optics for infraredwavelengths, improved imagingspeed, and the introduction offluorescent probes with high multi-photon cross-section.

Fast image acquisition is critical forcapturing rapid biologicalevents.However, almost all currentmultiphoton microscopes work byraster scanning a single laser spot,requiring a second or more to acquireeach x-y image plane. One approachto improving on this involves the useof a resonant mirror to rapidly scanthe laser in the x dimension, permit-ting acquisition speeds of 30 frames/s(28). An altogether different technolo-gy uses simultaneous scanning bymultiple laser spots, thereby propor-tionately speeding the acquisitionrate (3).

The optical requirements for micro-scope objectives are very differentbetween conventional epifluores-cence microscopy and multiphotonmicroscopy. With single-photon epi-fluorescence, aberrations with theshort-wavelength excitation light arenot important but the lens must bewell corrected for the visible-wave-length emission used to form theimage. On the other hand, multipho-ton excitation requires good correc-tions in the infrared so as to focus thelaser beam to a diffraction-limitedspot, but on the emission side theobjective needs only to collect asmuch light as possible, and aberra-tions are inconsequential. At present,multiphoton microscopists oftenmake use of objectives originallydesigned for infrared differentialinterference contrast and that thushave good long-wavelength transmis-sion, but development of new lensesspecific for multiphoton applicationsmay yield further benefits.

A major advantage of multiphotonmicroscopy is its ability to “see” deepwithin intact organs and organisms.

However, that advantage is partly lostdue to the large physical size of theobjective lens. For example, it is diffi-cult to insert a lens into the abdominalcavity of a mouse to image internalorgans. Recent advances mitigate thisproblem, either by using a gradientindex lens as a thin, rodlike probe to

extend the working distance of a con-ventional objective (16) or by usingfiber optics to a create multiphotonendoscopes. These developmentshave enormous potential for medicaltechnology, such as minimally invasiveendoscopic procedures, as well as sci-entific applications such as imaging



Primaryfollicles

Lymphnode

Diffusecortex

A Intact isolated lymph node

B Hindbrain of 36-h zebrafish

72

3 4

55 6

z axis (dorsal-ventral)x axis (left-right)

y ax

is (

ante

rio

r-p

ost

erio

r)y

axis

(d

ors

al-v

entr

al)

1

FIGURE 3. Multidimensional, multiphotonimaging of whole organs and organismsA: two-photon imaging of living and highly motilelymphoid cells deep within an intact isolatedlymph node. 1) Schematic diagram of a lymphnode, illustrating primary follicles and diffuse cor-tex. 2) Image montage constructed from severaltwo-photon image stacks, showing a “side view”(x-z) reconstruction into a node containing carboxyfluorescien diacetate-labeled living T cells

(green) in the diffusecortex and 4-chloromethylbenzoyl-amino tetramethylrho-damine-labeled B cells(red) in follicles. 3–6)Progressively enlargedviews showing the dis-tribution and cellularmorphology of T cellsand B cells. 7) Sideview reconstruction ofa different node, withreticular fibers stainedred and T cells green.Figure adapted fromRef. 26 (Miller et al.Two-photon imagingof lymphocyte motilityand antigen response

in intact lymph node. Science 296: 1873–1876. Copyright 2002 AAAS). B: hindbrain of a live 36-hzebrafish embryo expressing Islet1:EGFP (yellow) and a Histone H2B-mRFP1 fusion protein (red).The x-z, x-y, and y-z sections into the embryo are shown. The ability to noninjuriously image fluo-rescently tagged proteins within a live embryo allows for the mapping of gene expression patternsthroughout development. Figure adapted from Ref. 22 with permission from Elsevier.

on April 4, 2008


ownloaded from


metabolic disorders just by “looking”into the patient with a cellular level ofresolution (45).

Appendix: Constructing YourOwn Multiphoton System

Technological limitations of earliergenerations of femtosecond laserswere originally a great deterrent totheir use by biologists, but currentmodels are compact and require nomanual intervention by the user.However, the considerable expenseof commercial multiphoton systems(~$700,000) remains a hurdle.Because of this, many multiphotonmicroscopes are purchased formultiuser facilities—a job for whichthey are often ill suited—and end upgathering dust! Instead, multiphotonmicroscopes are better employedwithin the laboratory of an individualinvestigator, where they can bededicated for extended experimentsand equipped as needed withancillary systems for superfusion,environmental control, electrophys-iological recording, etc. Several of theapplications described in this reviewwere made possible because theinvestigators constructed their ownmultiphoton microscopes at a con-siderable saving in cost (17, 20, 28)This task is not as daunting as mightbe feared and, beyond the intimateworking knowledge and inherentsatisfaction it provides, has furtheradvantages, such as that the systemcan be customized for specific

cellular activity from awake, behavinganimals (see Ref. 13 for further read-ing).

On the receiving end of multipho-ton excitation, quantum dots havegreat promise as an extremely brightalternative to conventional fluo-rophores. These are nanocrystals ofsemiconductor that have a very hightwo-photon cross-section, show mini-mal photobleaching, and are availablein a range of emission wavelengths.Initial problems with biocompatibilityappear now to be largely solved, andquantum dots are available coatedwith various substrates, allowingthem, for example, to be conjugatedto specific antibodies. (43)

Applications

The key strengths of multiphotonmicroscopy lie in its applications forreal-time imaging of dynamicprocesses occurring within the nativeenvironment of intact live tissues,organs, and organisms. Develop-mental biologists and neurobiologistswere among the first to use thistechnique, and its more recentadoption by immunologists andmolecular biologists indicates itspotential within other disciplines.Future applications for multiphotonmicroscopy are broad and exciting,and they extend beyond basicresearch into the realm of noninvasiveor endoscopic clinical and diagnosticwork, such as detection of skincancers, Alzheimer’s disease, and

applications. FIGURE 4 outlines thebasic working pathway and com-ponents in a multiphoton imagingsystem, and the references providedetailed instructions for constructingone’s own system. �

We thank Dr. Mark Miller for helpful discussions.

Work in the authors’ laboratory is supportedby grant GM-48071 from the National Instituteof General Medical Sciences.

References1. Arrigoni M. Femtosecond laser pulses in biology:

from microscopy to ablation and micromanipula-tion. Biophotonics Int: 48–51, 2004.

2. Bousso P and Robey E. Dynamics of CD8+ T cellpriming by dendritic cells in intact lymph nodes.Nat Immun 4: 579–585, 2003.

3. Buist AH, Muller M, Squier J, and Brakenhoff GJ.Real time two-photon absorption microscopyusing multipoint excitation. J Microsc 192:217–226, 1998.

4. Cahalan MD, Parker I, Wei SH, and Miller M. Two-photon tissue imaging: seeing the immune sys-tem in a fresh light. Nat Rev Immunol 2: 872–800,2002.

5. Campagnola PJ, Mohler W, and Millard AE.Three-dimensional high-resolution second-har-monic generation imaging of endogenous struc-tural proteins in biological tissues. Biophys J 82:493–508, 2002.

6. Cox G and Sheppard CJR. Practical limits of reso-lution in confocal and non-linear microscopy.Microsc Res Tech 63: 18–22, 2004.

7. Denk W, Delaney KR, Gelperin A, Kleinfeld D,Strowbridge BW, Tank DW, and Yuste R.Anatomical and functional imaging of neuronsusing 2-photon laser scanning microscopy. JNeurosci Methods 54: 151–162, 1994.

8. Denk W, Strickler J, and Webb W. Two-photonlaser scanning fluorescence microscopy. Science248: 73–76, 1990.

9. Diaspro Ed A. Two-photon microscopy—Part II.Microsc Res Tech 63: 1–86, 2004.

10. Euler T, Detwiler PB, and Denk W. Directionallyselective calcium signals in dendrites of starburstamacrine cells. Nature 418: 845–852, 2002.

11. Göppert-Mayer M. Über Elementarakte mit zweiQuantensprüngen. Ann Phys (Leipzig) 9: 273–2941931.

12. Hellwarth R and Christensen P. Nonlinear opticalmicroscope examination of structures in polycrys-talline ZnSe. Opt Commun 12: 318–322, 1974.

13. Helmchen F. Miniaturization of fluorescencemicroscopes using fibre optics. Exp Physiol 87:737–745, 2002.

14. Holthoff K, Tsay D, and Yuste R. Calcium dynam-ics of spines depend on their dendritic location.Neuron 33: 425–437, 2002.

15. Kim S and Schwille P. Intracellular applications offluorescence correlation spectroscopy: prospectsfor neuroscience. Curr Opin Neurobiol 13:583–590, 2003.

16. Levene MJ, Dombeck DA, Kasischke KA, MolloyRP, and Webb WW. In vivo multiphoton microsco-py of deep brain tissue. J Neurophysiol 91:1908–1912, 2004.

17. Mainen Z, Maletic-Savatic M, Shi S, Hayashi Y,Malinow R, and Svoboda K. Two-photon imagingin living brain slices. Methods 18: 231–239, 1999.

20



Photomultiplier

IR filter

Objective

Sample

Tube lens

Dichroicmirror

Scan lens

X-Y scanmirrors

Laser

FIGURE 4. Schematic layoutof a multiphoton systemsetupThe excitation laser beam path isshown in red, and the emissionpathway and detection system are

shown in green. The pulsed infraredlaser light is directed into the microscope by orthogonal x-y scan-ning mirrors, the scanning lens serves as a beam expander, and thetube lens directs the laser point into the objective and onto thespecimen. Emitted fluorescent light passes back through the objec-tive and is deflected by the dichroic mirror into a detector systemsuch as a photomultiplier.

on April 4, 2008


ownloaded from


18. Maiti S, Shear J, Williams R, Zipfel W, and WebbW. Measuring serotonin distribution in live cellswith three photon excitation. Science 275:530–532, 1997.

19. Majewska A, Tashiro A, and Yuste R. Regulation ofspine calcium dynamics by rapid spine motility. JNeurosci Methods 20: 8262–8268, 2000.

20. Majewska A, Yiu G, and Yuste R. A custom-madetwo-photon microscope and deconvolution sys-tem. Pflügers Arch 441: 398–408, 2000.

21. Marchant J, Stutzmann G, Leissring M, LaFerla F,and Parker I. Multiphoton-evoked color changeof DsRed as an optical highlighter for cellular andsubcellular labeling. Nat Biotechnol 19: 654–659,2001.

22. Megason SG and Fraser S. Digitizing life at thelevel of the cell: high-performance laser-scanningmicroscopy and image analysis for in toto imag-ing of development. Mech Dev 120: 1407–1420,2003.

23. Mempel TR, Henrickson SE, and Von Andrian U. T-cell priming by dendritic cells in lymph nodesoccurs in three distinct phases. Nature 8:154–159, 2004.

24. Meyer T and Shen K. In and out of the postsy-naptic region: signalling proteins on the move.Trends Cell Biol 10: 238–244, 2000.

25. Miller MJ, Wei SH, Cahalan MD, and Parker I.Autonomous T cell trafficking examined in vivowith intravital two-photon microscopy. Proc NatlAcad Sci USA 100: 2604–2609, 2003.

26. Miller MJ, Wei SH, Parker I, and Cahalan M. Two-photon imaging of lymphocyte motility and anti-gen response in intact lymph node. Science 296:1873–1876, 2002.

27. Mohler W, Millard A, and Campagnola P. Secondharmonic generation imaging of endogenousstructural proteins. Methods 29: 97–109, 2003.

28. Nguyen Q, Callamaras N, and Parker I.Construction of a two-photon microscope forvideo-rate Ca2+ imaging. Cell Calcium 30:383–393, 2001.

29. Niggli E and Egger M. Applications of multi-pho-ton microscopy in cell physiology. Front Biosci 1:1598–1610, 2004.

30. Pons T, Moreaux L, Mongin O, Blanchard-DesceM, and Mertz J. Mechanisms of membranepotential sensing with second-harmonic genera-tion microscopy. J Biomed Opt 8: 428–431, 2003.

31. Reits EA and Neefjes J. From fixed to FRAP:measuring protein mobility and activity in livingcells. Nat Cell Biol 3: E145–E147, 2001.

32. Schwille P, Haupts U, Maiti S, and Webb W.Molecular dynamics in living cells observed by flu-orescence correlation spectroscopy with one- andtwo-photon excitation. Biophys J 77: 2251–2265,1999.

33. Sheppard C and Kompfner R. Resonant scanningoptical microscope. Appl Opt 17: 2879–2882,1978.

34. Soeller C, Jacobs MD, Jones KT, Ellis-Davies GC,Donaldson PJ, and Cannell M. Application oftwo-photon flash photolysis to reveal intercellularcommunication and intracellular Ca2+ move-ments. J Biomed Opt 8: 418–427, 2003.

35. Stoll S, Delon J, Brotz TM, and Germain R.Dynamic imaging of T cell-dendritic cell interac-tions in lymph nodes. Science 296: 1873–1876,2002.

36. Stutzmann GE, LaFerla FM, and Parker I. Ca2+ sig-naling in mouse cortical neurons studied by two-photon imaging and photoreleased inositoltriphosphate. J Neurosci 23: 758–765, 2003.

37. Trachtenberg JT, Chen BE, Knott GW, Feng G,Sanes JR, Welker E, and Svoboda K. Long-term invivo imaging of experience-dependent synapticplasticity in adult cortex. Nature 420: 788–794,2002.

38. Tsai PS, Friedman B, Ifarraguerri AI, ThompsonBD, Lev-Ram V, Schaffer CB, Xiong Q, Tsien RY,Squier JA, and Kleinfeld D. All-optical histologyusing ultrashort laser pulses. Neuron 39: 27–41,2003.

39. Van Zandvoort M, de Grauw C, Gerritsen H,Broers J, oude Egbrink M, Ramaekers F, and SlaafD. Discrimination of DNA and RNA in cells by avital fluorescent probe: lifetime imaging ofSYTO13 in healthy and apoptotic cells. Cytometry47: 226–235, 2002.

40. Waters J, Larkum M, Sakmann B, and Helmchen F.Supralinear Ca2+ influx into dendritic tufts of layer2/3 neocortical pyramidal neurons in vitro and invivo. J Neurosci 23: 8558–8567, 2003.

41. Webb R. Theoretical basis of confocal microsco-py. Methods Enzymol 307: 3–20, 1999.

42. Williams RM, Zipfel WR, and Webb W.Multiphoton microscopy in biological research.Curr Opin Chem Biol 5: 603–608, 2001.

43. Wu X, Liu H, Liu J, Haley KN, Treadway JA, LarsonJP, Ge N, Peale F, and Bruchez MP.Immunofluorescent labeling of cancer markerHer2 and other cellular targets with semiconduc-tor quantum dots. Nat Biotechnol 21: 41–46,2003.

44. Xu C, Zipfel W, Shear JB, Williams RM, and WebbW. Multiphoton fluorescence excitation: newspectral windows for biological nonlinear micro-scopy. Proc Natl Acad Sci USA 93: 10763–10768,1996.

45. Zipfel WR, Williams RM, Christie R, Nikitin AY,Hyman BT, and Webb W. Live tissue intrinsicemission microscopy using multiphoton-excitednative fluorescence and second harmonic gener-ation. Proc Natl Acad Sci USA 100: 7075–7080,2003.



on April 4, 2008


ownloaded from


dynamic multiphoton imaging: a live view from cells to systems › ... › uploads › 2017 › 04...

Documents