Laser tweezers and multiphoton microscopes in life sciences

This is the html version of the file http://www.jenlab.de/english/services/laser.pdf.
G o o g l e automatically generates html versions of documents as we crawl the web.
To link to or bookmark this page, use the following url:

http://www.google.com/search?q=cache:mzdeSE0HlMMJ:www.jenlab.de/english/services/laser.pdf+skin+penetration+depth+of+NIR+radiation&hl=en

Google is not affiliated with the authors of this page nor responsible for its content.

These search terms have been highlighted:

skin

penetration

depth

nir

radiation

Page 1

Abstract Near infrared (NIR) laser microscopy enables

optical micromanipulation, piconewton force determina-

tion, and sensitive fluorescence studies by laser twee-

zers. Otherwise, fluorescence images with high spatial

and temporal resolution of living cells and tissues can be

obtained via non-resonant fluorophore excitation with

multiphoton NIR laser scanning microscopes. Further-

more, NIR femtosecond laser pulses at TW/cm

intensi-

ties can be used to realize non-invasive contact-free sur-

gery of nanometer-sized structures within living cells

and tissues. Applications of these novel versatile NIR la-

ser-based tools for the determination of motility forces,

coenzyme and chlorophyll imaging, three-dimensional

multigene detection, non-invasive optical sectioning of

tissues (“optical biopsy”), functional protein imaging,

and nanosurgery of chromosomes are described.

Key words FISH · FRET · Laser tweezers · Multiphoton

microscopes · Nanosurgery

Introduction

Conventional light microscopy in “Life Sciences” in-

cluding confocal laser scanning microscopy is based on

the use of ultraviolet (UV) and visible radiation (Pawley

1995; Greulich 1999). One of the favorite methods in

live cell studies is fluorescence imaging. Around 80% of

applications of light microscopes in “Life Sciences” in-

volve fluorescence microscopy. Fluorophores include en-

dogenous (intrinsic) and exogenous (applied) dyes. Most

of the endogenous fluorophores in cells, such as trypto-

phan, dopamine, serotonin, NAD(P)H, and flavins,

require UV or blue excitation wavelengths. Important

exogenous fluorophores, such as the DNA markers

Hoechst 33342 and 4′,6-diamidino-2-phenylindole

(DAPI) as well as the calcium indicators Fura-2 and

Indo-1 possess absorption bands in the UV only. Other

exogenous fluorophores are designed to absorb at the

visible wavelengths of the laser and mercury lamp exci-

tation radiation. UV microscopes have also been em-

ployed as microsurgery tools (Greulich 1999).

However, it must be noted that the use of UV and

short-wavelength visible radiations in conventional light

microscopy have certain disadvantages mainly because

of low light penetration depth and the potential of severe

photodamage to living cells (see Cunningham et al.

1985; Tyrell and Keyse 1990). It has been shown that

during high-resolution fluorescence microscopy with the

365-nm mercury lamp radiation results within seconds in

failed cellular reproduction, modifications in intracellu-

lar redox state, and DNA strand breaks (König et al.

1996a).

A versatile innovation in live cell microscopy is based

on the application of near infrared (NIR) laser radiation

in the spectral range of 700–1100 nm, the “optical win-

dow” of cells and tissues. Unpigmented cells appear as

nearly transparent objects in this spectral range. There-

fore, these cells can potentially experience extremely

high light intensities up to 100 gigawatt per square centi-

meter (1 GW/cm

=10

W/cm

) without damage. This

enormous intensity corresponds to a 12 orders higher

light intensity than sunlight reaching the surface of the

earth. When using ultrashort NIR laser pulses and the

mean intensity remains below 20 MW/cm

, the cellular

heating at <100 GW/cm

peak intensities is less than 2°C

(Liu et al. 1995; Schönle and Hell 1998).

The combination of NIR lasers and microscopes pro-

vides novel non-contact biomedical tools in “Life Sci-

ences”. These tools include not only imaging devices for

damage-free live cell imaging, but also for optical micro-

manipulation, intracellular photochemistry, and nanosur-

gery.

Robert Feulgen Prize Lecture 2000 presented at the 42nd sympo-

sium of the Society for Histochemistry in Les Diablerets, Switzer-

land, on 23 September 2000

K. König (

✉

)

Laser Microscopy Division, Institute of Anatomy II,

Friedrich Schiller University Jena, Teichgraben 7,

07743 Jena, Germany

e-mail: kkoe@mti-n.uni-jena.de

Tel.: +49-3641-938560, Fax: +49-3641-938552

Histochem Cell Biol (2000) 114:79–92

DOI 10.1007/s004180000179

ROBERT FEULGEN PRIZE LECTURE

Karsten König

Laser tweezers and multiphoton microscopes in life sciences

Accepted: 26 June 2000 / Published online: 19 July 2000

Page 2

When using highly focused continuous wave (cw)

NIR laser radiation, cells and organelles can be optically

confined in the focal volume of a high numerical aper-

ture (NA) objective by the radiation pressure (Ashkin

1970, 1980; Ashkin and Dziedzik 1987, 1989). These

tools are called laser tweezers (optical traps) and can be

used for optical micromanipulation, cell sorting, pico-

newton (pN) force measurements, motile cell diagnos-

tics, and as sources for two-photon excited fluorescence.

The other important NIR tool is the multiphoton laser

scanning microscope for fluorescence imaging with high

spatial and temporal resolution, photoinduced uncaging

of compounds, and nanoprocessing (Denk et al. 1990;

König et al. 1999a). Typically, these microscopes are

based on the use of ultrashort laser pulses in the femto-

second range (1 fs=10

–15

s). Multiphoton excitation

based on the simultaneous absorption of photons was

predicted in 1931 (Göppert-Meyer 1931) and first real-

ized with the availability of lasers in 1961 (Kaiser and

Garret 1961). In 1990, Denk et al. realized two-photon

excitation of fluorophores in living cells.

This paper focuses on applications of laser tweezers

and multiphoton femtosecond laser microscopes in “Life

Sciences” encompassing cell biology, biotechnology, and

medicine.

Laser tweezers

Principle

Laser tweezers, also known as (single beam gradient

force) optical traps, are based on pN force generation

during the interaction of highly focused laser beams with

dielectric particles, including cells and organelles.

Light, as a carrier of momentum, exerts pressure

which can be used to accelerate particles (Ashkin 1980).

This is the reason why the tail of a comet is directed

away from the sun. If a light beam of sufficient radiation

pressure is directed against gravity, particles can be

transduced into “gravity-free” conditions. The radiation

pressure of the sunlight on earth is of the low order of

µN/cm

, however, with the availability of laser sources,

laser light pressure can be used to move small objects.

As a spin-off of Ashkin’s initial studies on optical parti-

cle trapping (Ashkin 1970, 1980), optical cell microma-

nipulation by laser light became possible.

By focusing a single laser beam with a high NA ob-

jective, a gradient field of light intensity is created with

the highest intensity values in the focal volume and the

lowest in the periphery. Interaction with objects of a

higher refractive index than the surrounding results in

the formation of gradient forces. If the interaction is

primarily determined by beam refraction and when the

absorption is negligible, the net force acts toward the

focal volume. The net force (trapping force) of the or-

der of pN can be used to “pull” and to confine micro-

and nanometer-sized objects in the focal volume

(Fig. 1).

When using highly focused cw NIR laser beams (“mi-

crobeams”), pigment-free cells can be optically trapped

and can be manipulated in three dimensions without

physically touching them. In particular, contact-free cell

transport can be performed by moving the foci of the la-

ser beam in the desired direction. Alternatively, the stage

with the sample can be moved without displacement of

the trapped target.

Typical laser sources for optical trapping are the

Nd:YAG laser at 1064 nm and laser diodes. Often, the la-

ser beam is coupled by light fibers or directly to a video

microscope and focused to a diffraction-limited spot by

objectives with NA>1. Motor-driven mirrors in combi-

nation with joy-sticks allow maneuvering the position of

the tweezers.

Optical micromanipulation

In contrast to mechanical manipulation with microma-

nipulators, laser tweezers enable contact-free microma-

nipulation even through the glass windows of a closed

cell chamber. Such a versatile chamber (JenLab, Jena,

Germany) with 170-µm-thick glasses and a silicone gas-

ket is depicted in Fig. 2. Cells and medium can be easily

transferred and changed with standard needles and sy-

ringes.

We used closed cell chambers to trap human sperma-

tozoa of patients with impaired fertility and to study the

influence of the intense NIR beam on the cellular metab-

olism and vitality (König et al. 1995a). The NIR beam

traps spermatozoa by confinement of the head, the DNA-

containing region of high refractive index, in the focal

volume. The goal of the study was to prove if laser twee-

zers are an appropriate optical micromanipulation tool to

realize laser-assisted in vitro fertilization (IVF) by opti-

cal transport of a single sperm cell to the oocyte. This

method was suggested by Tadir et al. (1989) as an alter-

native to mechanical approaches.

Interestingly, laser tweezers at a wavelength of 800 or

1064 nm did not harm the cell even over extended trap-

ping periods of up to 15 min, whereas optical traps at

Fig. 1 Schematic representation of laser tweezers

Page 3

700–800 nm affected the cellular metabolism and led to

cell death within seconds (König et al. 1996a,b). The

reason is the phenomenon of trap-induced two-photon

excitation followed by harmful UV effects as explained

later.

Most optical trapping studies are conducted with the

1064-nm beam of an Nd:YAG laser which appears as a

safe micromanipulation tool even if the cell experiences

enormous energy densities (fluences) of some GJ/cm

When using 100-mW laser tweezers, the trap-induced

temperature increase is between 1 and 2°C. Meanwhile,

the first clinical applications of laser tweezers for laser-

assisted IVF have been reported (see, for example,

Wiedemann and Montag 1994). In addition, laser twee-

zers have been employed as micromanipulation tools in

pharmacology, biotechnology, and cell biology. For ex-

ample, Ashkin and Dziedzic (1989) studied the mechani-

cal properties of the cytoplasm in the interior of living

cells, and Chu’s group measured relaxation curves of

single DNA molecules (Perkins et al. 1994). Zahn and

Seeger (1999) used laser tweezers for drug screening.

Buican et al. (1987) are referred as pioneers of automat-

ed cell sorting by optical traps.

Force measurements

The net force (trapping force, F) depends linearly

on the laser power P and can be represented by:

F=QP/c

where c is the velocity of light in the medium and Q the

trapping efficiency parameter with values between 0 and

2. The parameter Q depends on the optical properties of

the trapped object, such as the refractive indices, as well

as on the beam profile and alignment. The knowledge of

the trapping parameter allows the calculation of the trap-

ping force and the subsequent use of the laser tweezers

as force measure (picotensometer).

Q can be calculated theoretically in the case of spheri-

cal particles with known refractive index. However, for

“real” biological objects with non-uniform morphology

and heterogeneous optical density, the trapping parame-

ter and the trapping forces, respectively, have to be de-

termined experimentally. The most common method is

the “escape force method” where the trapped object ex-

periences a flow of liquid of increasing velocity. The

flow exerts a drag force, F

drag

, which can be calculated

by Stokes’ law. This force depends on the velocity, vis-

cosity, and the target morphology and dimension. At a

certain velocity, the drag force has the same value as the

trapping force and the object can “escape”. It is at this

point that the drag force can be used to calibrate the trap-

ping force.

We determined the trapping parameter and the trap-

ping forces of human spermatozoa confined in an 800-nm

trap (König et al. 1996c). With a mean Q parameter of

0.12, the trapping force was found to be about 50 pN at

100 mW laser power. In the case of healthy spermatozoa,

a mean trapping power of about 80 mW was required to

confine these highly motile cells. Under the assumption

of linear swim motion, the mean intrinsic motility force

of human sperm can therefore be estimated to be about

45 pN (König et al. 1996c).

Trap-induced forces have been extensively used to

study motor proteins (Block et al. 1990; Kuo and

Scheetz 1993) and binding behavior, such as receptor-

mediated interactions of cells with glycoproteins (Zahn

and Seeger 1999). A review of biological applications of

optical forces was published by Svoboda and Block

(1994).

Fluorescence diagnostics of motile cells

Fluorescence studies on single motile cells are difficult

due to pN motility force driven sample movement. Cer-

tain bacteria, algae, and spermatozoa can achieve veloci-

ties higher than 50 µm/s. The problem of sample move-

ment can be overcome when the object of investigation

is confined to the focal volume of the objective by laser

tweezers. We combined sensitive imaging devices, such

as ultrafast time-gated slow-scan CCD cameras, and la-

ser tweezers to be able to image the autofluorescence of

highly motile cells with high spatial (submicron) and

temporal (picosecond, 1 ps=10

–12

s) resolution.

Figure 3 demonstrates an example of autofluores-

cence imaging of a highly motile sperm cell. The endog-

enous fluorophores have been excited with the 365-nm

radiation of a high-pressure mercury lamp and detected

in the blue/green spectral range. The only intracellular

region of autofluorescence of the motile cell was found

to be the midpiece of the cell which is the location of the

mitochondria. The fluorescence of mitochondria is main-

ly based on the reduced coenzyme NADH. Interestingly,

the sperm head including the acrosome region became

brightly fluorescent following exposure to extended

UVA radiation. The changes in the autofluorescence pat-

Fig. 2 Photograph of a sterile cell chamber which can be used in

optical trapping and multiphoton femtosecond laser microscopy

Page 4

tern were accompanied by reduced motility and finally

loss of vitality (König et al. 1996b).

An example of autofluorescence lifetime imaging is

depicted in Fig. 4. The fluorescence lifetime is an intrin-

sic property of the fluorophore and its microenvironment

and independent of concentration. Imaging of fluores-

cence lifetimes (τ mapping) enables spatially resolved

fluorescence lifetime determination and fluorophore sep-

aration. The figure exhibits the autofluorescence pattern

of the biflagellate green microalga Haematococcus plu-

vialis which was trapped by means of multiple traps at

1047 nm. Chlorophyll was excited with a picosecond la-

ser diode at 633 nm and imaged with an ultrafast time-

gated camera with a tunable time-delay (0–20 ns) be-

tween fluorescence excitation and detection (König et al.

1998). A mean fluorescence lifetime in the picosecond

Fig. 3 Autofluorescence imaging of an optically trapped single

motile sperm cell. The fluorescence emanates from the mitochon-

dria-rich region corresponding to the midpiece of the sperm

Fig. 4 Time-resolved autofluorescence imaging of an optically

trapped motile green alga. The time-delay between excitation of chlo-

rophyll and detection of fluorescence was spaced in steps of 500 ps

Page 5

range was determined which increased up to 1.4 ns when

exposed to herbicides, indicating disturbed energy trans-

fer.

Laser tweezers as non-linear diagnostic tools

Fluorescence diagnostics can also be performed with the

trapping beam as fluorescence excitation source. Surpris-

ingly, during trap experiments with dye-labeled sperma-

tozoa in 1994 we found that highly focused cw NIR

beams at 100 mW power are able to excite the visible

fluorescence of the intracellular dyes (König et al.

1995b). Excitation occurred only in a sub-femtoliter fo-

cal volume (1 fl=10

–18

), the region of highest light in-

tensity (Fig. 5). The fluorescence followed a squared re-

lation on laser power indicating a two-photon process. In

this case, the fluorescence is excited by the simultaneous

absorption of two low-energy (NIR) photons. Each of

them provides half the energy required for molecule ex-

citation (Fig. 6). As an example, a fluorophore which is

normally excited at 400 nm (for example red fluorescent

protoporphyrin IX) can therefore be excited with 800-nm

radiation.

Two-photon excitation radiation requires a high pho-

ton concentration in space and time due to the low mo-

lecular two-photon absorption cross-sections (10

–48

Fig. 5 Trap-induced visible fluorescence inside the optically con-

fined head of spermatozoa labeled with SYBR-green

Fig. 6 Principle of two-photon excitation. The simultaneous ab-

sorption of two near infrared (NIR) photons at MW/cm

and

GW/cm

light intensities induces visible fluorescence. At TW/cm

intensities multiphoton ionization leads to plasma-induced abla-

tion which can be used for nanoprocessing. S

, S

represent dif-

ferent energy levels, E energy

–50

s). To achieve such a high photon concentra-

tion, NIR light intensities of at least 20 MW/cm

are re-

quired.

Assuming a 100-mW laser beam at λ=800 nm and an

illumination spot of diameter d=λ/NA=615 nm and area

0.30 µm

, a high intensity of power/area=33 MW/cm

can

be calculated which is sufficient to induce two-photon

effects. Therefore, laser tweezers act as sources of two-

photon excitation and can be used as novel tools for non-

linear fluorescence excitation (the fluorescence intensity

has a squared, not a linear, dependence on power). One of

the interesting features of trap-induced fluorescence is the

possibility to localize the intracellular trapping spot with

high accuracy. Florin et al. (1998) use this effect to create

a novel photonic force microscope.

The two-photon excited fluorescence can also be used

to study possible harmful effects of the laser tweezers on

the cell trapped with the NIR beam. For example, when

injecting the viability indicators SYBR14 (green fluores-

cent live cell indicator) and propidium iodide (red fluo-

rescent dead cell indicator) from Molecular Probes

(Eugene, Ore., USA) into the cell chamber with sperma-

tozoa, the possible onset of trap-induced lethal effects

can be studied without external fluorescence excitation

sources. Using the methods of microspectrofluorometry

and spectrally resolved fluorescence imaging, we were

able to study the damaging effect of short-wavelength

Page 6

(<800 nm) laser tweezers (König 1998). NIR micro-

beams in the spectral range of 700–800 nm are able

to excite endogenous absorbers with electronic transi-

tions in the UV which may lead to harmful UVA effects

(Cunningham et al. 1985; Tyrell and Keyse 1990).

Using the trap-induced fluorescence we also studied

the effect of the laser beam quality and found: (1) that

the damage process is based on a two-photon process

and (2) that certain cw laser beams contain unstable pi-

cosecond laser pulses (“spikes”) which may enhance de-

structive effects (König et al. 1996d).

Multiphoton laser microscopy

Principle and setup of a multiphoton laser

scanning microscope

CW laser microbeams in combination with scanning mi-

croscopes can be used for three-dimensional (3D) imag-

ing of two-photon excited fluorophores (Hänninen et al.

1994; Booth and Hell 1998). In contrast to conventional

confocal laser scanning microscopes, no spatial filter

(pinhole) is required to obtain 3D images. This is be-

cause of the minute (sub-femtoliter) two-photon excita-

tion volume which can be used to scan the sample

(Fig. 7). Due to lack of NIR absorption outside the focal

volume, there is neither out-of-focus photobleaching nor

out-of-focus photodamage.

In two-photon microscopes, the fluorescence intensity

increases quadratically with the excitation intensity and

the power, respectively. However, trapping effects and

photothermal effects limit the use of high power cw

sources for fast fluorescence scanning microscopy. Fast

multiphoton fluorescence imaging is much more effi-

cient when using high repetition pulsed laser systems

with moderate peak power in the W and kW range but

with low mean µW and mW power. The two-photon ex-

cited fluorescence yield, I

, follows the following rela-

tion (Denk et al. 1990):

≈P

α/(τf

)·π

/(hcλ)

where P is the mean power, α the molecular two-photon

absorption coefficient, τ the pulse width, f the repetition

frequency, NA the numerical aperture, h Planck’s con-

stant, c the velocity of light, and λ the wavelength. Be-

cause the fluorescence yield depends on a P

/τ relation,

two-photon microscopy with 1-ps laser pulses requires

threefold higher mean powers than 110 fs pulses. Both,

picosecond as well as femtosecond laser scanning micro-

scopes are now commercially available from leading mi-

croscopy suppliers.

In three-photon fluorescence microscopy (Gryczynski

et al. 1995; Wokosin et al. 1995; Hell et al. 1996; Maiti

et al. 1997), where three photons are absorbed simulta-

neously, I

depends on a P

/τ

relation. In this case, effi-

cient excitation requires the use of femtosecond laser mi-

croscopes. Three-photon excitation has been used to im-

age serotonin in living cells (Maiti et al. 1997).

Our multiphoton microscopes are based on the use of

femtosecond solid state laser sources. We use turn-key,

air-cooled, single-box laser systems. The first one is a ti-

tanium:sapphire laser (Vitesse 800-HP; Coherent, Santa

Clara, USA) with an output of 1 W mean power,

80 MHz repetition frequency, 800 nm wavelength, 80 fs

pulse width, and 55.5×34×18 cm

dimensions. We use

this laser for two-photon and three-photon fluorescence

studies in cells and tissues, and fluorescence in situ hy-

bridization (FISH) studies, as well as for nanosurgery.

The second ultracompact one (Femtolite; IMRA, Ann

Arbor, Mich., USA) is a frequency doubled erbium-

doped fiber laser at 780 nm with 40 mW mean power,

50 MHz repetition frequency, 780 nm wavelength, 180 fs

pulse width, and 19.3×10.9×8.2 cm

dimensions (elec-

tronic controller: 24.9×30.5×7.2 cm

). The power at the

sample of <7 mW is sufficient to excite a variety of flu-

orophores in cell monolayers and in FISH studies.

Fig. 7 Under same focusing conditions with high numerical aper-

ture objectives, two-photon excitation is confined in the minute

focal volume due to the required high intensity. By contrast, one-

photon excitation results in fluorescence along the illumination

cones. hv Energy of a photon

Page 7

The laser beam is expanded by an 1:4 Galilean tele-

scope, coupled to a modified inverted confocal laser

scanning microscope (LSM410; Zeiss, Jena, Germany)

and focused to a diffraction-limited spot by 40×, 63×, or

100× objectives of NA>1.2. Due to optical dispersion

which results in pulse broadening during transmission

through microscope optics, the pulse width at the sample

is about 150–200 fs (König 2000). At 5 mW mean pow-

er, the peak power and the peak intensities reach values

of 0.4 kW and 0.6×10

W/cm

(0.6 TW/cm

), respec-

tively, when assuming a full width half maximum beam

size of λ/2NA≈310 nm. A typical pixel dwell time of the

beam during one scan is 4 µs which results in a frame

rate of 1 s/frame. At zoom 4, 512×512 pixels cover a

sample area of 80×80 µm.

Multiphoton excited fluorescence is typically regis-

tered with a detector at the baseport of the modified mi-

croscope. A 700-nm short pass filter prevents the scat-

tered laser radiation from reaching the detector. We use

different camera systems, photomultiplier tubes (PMTs),

and a spectrometer as detectors.

The described microscope setup can also be used for

conventional one-photon confocal microscopy with an

internal He-Ne laser, pinholes, and the internal PMTs of

the microscope.

Multiphoton multicolor FISH (MM-FISH)

An interesting feature of multiphoton microscopy is the

possibility to excite a variety of fluorophores simulta-

neously at one NIR wavelength due to overlapping

two(three)-photon excitation spectra (Xu et al. 1996). We

have taken advantage of this to detect various genome

regions labeled with multiple fluorescent targets with

different emission wavelengths (König et al. 2000a). La-

beling is based on standard FISH procedure (see Speel

1999 for a review).

In contrast to the conventional detection method,

where different excitation wavelengths in the UV, blue,

and green range are required to induce the visible fluo-

rescence of the most common FISH fluorophores and of

the general DNA stain DAPI, the novel approach uses

NIR light at one excitation wavelength only. Figure 8

shows the emission spectra of FISH fluorophores which

can be non-linearly excited at 780 and 800 nm.

In addition to the advantage of using a single excita-

tion wavelength to realize multicolor FISH, multiphoton

microscopy enables optical sectioning of thick samples

(interphase nuclei, embryos, biopsies) in different

planes. There is no photobleaching of FISH fluorophores

in out-of-focus regions. Reconstruction of 3D fluores-

cence images from the optical sections provides informa-

tion on the genome architecture, such as 3D organization

of chromosomes and their well-defined domains such as

centromeres and telomeres. Such a 3D fluorescence im-

age of centromeric regions of the chromosomes 1, 3, 6,

12, and X in the interphase nucleus of an amniotic fluid

cell is seen in Fig. 9. The 3D image has been recon-

structed from a stack of 20 images which are spaced by

0.75 µm. The five FISH fluorophores have been excited

simultaneously. Also the blue fluorescence of the DNA

counterstain DAPI has been induced with NIR femtosec-

ond laser pulses. Spatially resolved DAPI fluorescence

imaging provides information about the nuclear architec-

ture and enables the determination of the intranuclear lo-

ci of the fluorescent centromeric regions.

Besides multifluorophore excitation and optical sec-

tioning, a third advantage of using excitation radiation in

the 700- to 1200-nm spectral range is the high light pen-

etration depth. We used MM-FISH for the detection of

centromeric regions in multilayer samples, such as hu-

man biopsies. In particular, we imaged two-photon excit-

ed spectrum green-labeled C-4 and spectrum orange-la-

beled C-X probes in thick kidney cryosections with high

spatial resolution. The nuclear area was visualized after

Fig. 8 The depicted FISH fluorophores with emission in the blue,

green, yellow, and red can be simultaneously excited at 780 and

800 nm. DAPI 4′,6-diamidino-2-phenylindole, DAC diethyl-

aminocoumarine, FITC fluorescein isothiocyanate, Cy cyanine

Page 8

counterstaining with ethidium bromide and optical sec-

tioning with NIR laser pulses (König et al. 2000b).

Potential applications of this new MM-FISH tech-

nique are in the field of molecular cytogenetics, prenatal

and preimplantation diagnosis, and molecular pathology.

Live cell imaging

In addition to the in vitro studies, imaging of DNA in

living cells and tissues can be performed by two-photon

excited imaging of the DNA fluorophore Hoechst 33342.

The high penetration depth of NIR radiation also allows

the spatially resolved detection of the fluorophore

in deep tissue. NIR radiation at 800 nm has a typical

penetration depth in tissue of several millimeters in

contrast to some micrometers when using one-photon

fluorescence excitation in the UV (Chong et al. 1990).

Figure 10 shows DNA images of tumor tissue in living

mice after the topical application of a Hoechst–DMSO

mixture. The Hoechst-labeled chromatin can be clearly

visualized in the different cell layers. Considering a

120×120×30 µm

=0.4×10

–12

tumor volume, a mean

number of 100 nuclei in such a volume can be counted

from the stack of images. Interestingly, scattering in tur-

bid media such as tissues does not decrease the excellent

lateral resolution of <0.4 µm and the ca. 1 µm axial reso-

lution significantly within the first 100 µm tissue.

Multiphoton microscopy has been widely used in imag-

ing of single living cells, cell monolayers, and embryos

(see König 2000 for a review) including mapping of ion

channels by two-photon photochemical microscopy (Denk

1994) and neuron imaging (Denk and Svoboda 1997).

Fig. 10 Non-invasive optical sectioning in a living anesthetized

mouse. Depicted are Hoechst 33342-labeled tumor tissue layers

Fig. 9 Three-dimensional (dz=0.75 µm) six-color detection of in-

terphase nuclei of amniotic cells with the NIR-excited fluoropho-

res spectrum aqua (centromere chromosome X, blue), spectrum

orange (centromere chromosome 3, red), rhodamine 110 (centro-

mere chromosome 12, yellow), spectrum green (centromere chro-

mosome 6, green), DAC (centromere chromosome 1, white), and

DAPI (light blue). The image is false-color coded and reconstruct-

ed from a stack of 20 optical sections

Page 9

An interesting feature of NIR multiphoton microsco-

py is the possibility to excite endogenous fluorophores

such as the reduced coenzymes NADH and NADPH as

well as flavin coenzymes with NIR radiation in the range

of 700–800 nm (Piston et al. 1995; König et al. 1996d).

Because the oxidized forms NAD(P) do not exhibit fluo-

rescence, the NAD(P)H attributed autofluorescence can

be used to obtain information on the intracellular redox

state, modifications in the respiratory chain, and cellular

metabolism (König and Schneckenburger 1994; König et

al. 1995b).

The two-photon excitation of tissue autofluorescence

may become a useful non-invasive technology to obtain

“optical biopsies” without physical removal of tissue.

Similar to tomography with X-rays, intense NIR micro-

beams can be employed to perform optical sectioning of

the tissue.

Endogenous fluorophores and structures which can be

excited via a two-photon or a three-photon excitation

process include serotonin, dopamine, and tryptophan

with emission in the UV, NAD(P)H, collagen, elastin,

melanin, and flavins with blue/green fluorescence, lipo-

fuscin, Zn-coproporphyrin, and Zn-protoporphyrin with

yellow emission, and coproporphyrin, protoporphyrin,

and chlorophyll with red fluorescence.

The first two-photon excited autofluorescence images

of in vivo human skin have been performed (Masters et

al. 1997; König 2000). Figure 11 depicts a single in vivo

autofluorescence image from a stack of high-resolution

fluorescence images. The image was acquired at a depth

of 30 µm from the palmar surface of the forearm. The

stack was obtained up to 150 µm into the skin with a

depth increment of 5 µm. The stratum corneum with a

typical thickness of 15 µm at the investigated loci was

found to be highly fluorescent when excited at 800 nm.

In particular, the border of the hexagonal-shaped tissue

structures exhibited fluorescence. Below this tissue lay-

er, individual cells could clearly be visualized by the au-

tofluorescence of intracellular structures in the cyto-

plasm. Cell nuclei and cell membranes did not fluoresce.

The autofluorescence was stronger in the innermost layer

of the epidermis, the basal layer, than in the surround-

ings. Epidermis and dermis could be differentiated, and

elastin and collagen fibers could be visualized.

Photodamage due to multiphoton microscopy

Due to the lack of out-of-focus photodamage and photo-

bleaching, multiphoton NIR microscopy appears as a

safe novel tool without impact on cellular metabolism,

reproduction, and vitality. In fact, Chinese hamster ovary

cells can be scanned with an 800-nm femtosecond laser

beam at 2 mW for hours without damage to the exposed

cells and their derivatives (König 2000). Squirrel et al.

(1999) intermittently exposed hamster embryos for 24 h

with ultrashort laser pulses of GW/cm

peak intensity

without impact on embryo development in contrast to

one-photon laser scanning microscopy where the embry-

os did not survive.

However, above certain thresholds photodamage may

occur. As pointed out in the section on laser tweezers, in-

tense cw NIR beams can induce cell damage via a two-

photon effect. The same destructive effects would there-

fore occur when using ultrashort laser pulses at even

higher light intensities.

We found mainly two types of photodamage during

NIR microscopy. The first is a slow process probably

based on two-photon excitation of endogenous absorbers

and subsequent photo-oxidation processes resulting in

the formation of destructive reactive oxygen species

(ROS). This process appears at MW/cm

and GW/cm

intensities and depends strongly on wavelength. The sec-

ond damage process is of immediate effect and requires

high intensities in the TW/cm

range and is based on

multiphoton ionization, optical breakdown phenomena,

and intracellular plasma formation. It results in material

ablation and disruption including complete cell fragmen-

tation.

Studying the first photochemical attributed photo-

damage process we found that the mitochondria and the

Golgi apparatus are the major targets of NIR micro-

beams (Oehring et al. 2000). Above certain light intensi-

ties, laser-exposed cells either fail to divide, become gi-

ant cells, or die. Due to the P

/τ dependence of the slow

damage process, photodamage was found to be more

pronounced at shorter pulses (König et al. 1999b).

More recently it has been demonstrated that NIR laser

irradiation at certain higher power levels evoke genera-

Fig. 11 Non-invasive optical biopsy of high spatial resolution

with near infrared femtosecond laser pulses. Depicted is an in vivo

autofluorescence image of human skin showing an epidermal layer

at a depth of 30 µm. Note that individual cells as well as cytoplas-

mic structures are clearly visible

Page 10

Fig. 12A–C Nanosurgery with NIR femtosecond laser pulses.

Force microscopy image of cuts through human chromosomes

(A). A minimum cut size of 110 nm was achieved. Transmission

and fluorescence image of living cells labeled with rhodamine 123

before (B) and after (C) knocking out of a single mitochondrion

(asterisks)

Page 11

tion of ROS that can be cytochemically visualized in vi-

vo using Ni-3,3′-diaminobenzidine (Ni-DAB) as well as

with a recently developed fluorescent probe Jenchrom px

blue from JenLab (Tirlapur et al. submitted). In addition,

the irradiated cells manifest membrane-barrier dysfunc-

tion, drastic alterations in the morphology of the nuclear

envelope, and fragmentation of the DNA. Because the

cytological changes observed in NIR-irradiated cells

bear striking similarities to those seen in cells undergo-

ing programmed cell death (Li and Darzynkiewicz

1999), it has been inferred that intense NIR femtosecond

laser pulses can induce apoptosis-like death (Tirlapur

and König submitted).

Femtosecond laser surgery of cellular nanostructures

When the plasma-mediated photodamage process can be

confined to a tiny intracellular volume by parking the

beam at pixels of interest, the destructive effects of in-

tense NIR laser pulses can be used to cut cellular struc-

tures or to “drill” holes in the target (König et al. 1999a).

Material processing with femtosecond laser pulses has

the advantages of minimal ablation threshold, low trans-

fer of optical energy into destructive mechanical energy,

and the absence of thermal damage to surrounding struc-

tures compared to nanosecond pulses used in conven-

tional microsurgery.

We use the novel surgery tool to perform dissections

of chromosomes and to knock out intracellular structures

at mean powers of 30–50 mW with microsecond and

millisecond beam dwell times (Fig. 12A,B). Because on-

ly the central part of the illumination spot provides suffi-

cient intensity for plasma-induced ablation, laser cuts

and holes can be generated in the target with a size be-

low the diffraction-limited spot size. We realized a mini-

mum cut-size of 110 nm into the human chromosome 1

which is to date likely be the smallest laser cut in biolog-

ical material. The topography of this laser cut was ana-

lyzed and measured with a scanning force microscope.

We were also able to perform chromosome dissections

within round living cells (König et al. 1999a). The use of

femtosecond NIR pulses at TW/cm

light intensities

therefore provide novel non-invasive tools to perform

nanosurgery without destructive influence to the sur-

roundings and enable nanoprocessing within living cells

and tissues.

Further potential applications of NIR microscopes

Four dimensional (4D) microscopy in space and time

An interesting feature of multiphoton laser scanning mi-

croscopes with picosecond and femtosecond pulsed la-

sers is the possibility of simultaneously performing 4D

imaging in space and time. Of particular relevance is mi-

croscopic imaging with ultrafast temporal resolution in

the range of picoseconds and nanoseconds corresponding

to the range of fluorescence lifetimes (see, for example,

Lakowicz 1983; So et al. 1998). Hence with a pulsed la-

ser excitation source it is possible to upgrade a 3D multi-

photon microscope to a versatile 4D imaging device.

Fig. 13 Time-resolved multiphoton microscopy. Fluorescence de-

cay curves after two-photon excitation of intracellular fluoropho-

res along lines of 128 pixels

Page 12

In order to realize 4D two-photon microscopy we

have incorporated a fast PMT in combination with a sin-

gle photon counting unit (SPC 730; Becker and Hickl,

Berlin, Germany). The unit enables the fast registration

of fluorescence decay curves following excitation with

the 80-MHz femtosecond laser at a pixel of interest dur-

ing single-point illumination as well as the registration

of 128×128=16384 decay curves during scanning. Ex-

amples of fluorescence decay curves of a variety of in-

tracellular fluorophores along line scans and an example

of a τ image are presented in Figs. 13 and 14.

Two-photon fluorescence resonance

energy transfer (FRET)

As yet an unexplored interesting potential feature of

multiphoton microscopes with high NA objectives is the

use of the minute sub-femtoliter excitation volume for

studying single molecules and intramolecular interac-

tions. In particular the combination of two-photon NIR

excitation and FRET allows monitoring of the spatially

resolved relationship between two macromolecules with-

in living cells (Fig. 15).

The non-radiative energy transfer (Foerster mecha-

nism) occurs in close proximity of a fluorescence donor

(D) in the excited state and a fluorescence acceptor (A).

The efficiency of the energy transfer decreases with R

–6

where R is the distance between D and A. Therefore,

FRET occurs mainly within intermolecular distances

with R<10 nm. It provides a variety of information on

molecular interactions such as binding behavior, inter-

molecular distance, diffusion kinetics, and association

reactions (Pollak and Heim 1999). In order to realize ef-

ficient two-photon FRET the following conditions are

essential:

1. Efficient two-photon excitation of the donor molecule

2. Spectral overlap between the emission spectrum of

the donor and that of the excitation spectrum of the

one-photon acceptor as well as appropriate relative

orientation of both molecules

3. Separation between acceptor and donor emission

4. Distance less than 10 nm.

Due to the broad two-photon excitation spectrum, a vari-

ety of fluorophores preferably with UV and blue one-

photon absorption maxima can be used as donors. In the

case of probing protein–protein interactions, GFP mu-

tants such as B(lue)FP with 445 nm emission, C(yan)FP

(505 nm), Sapphire (511 nm), eGFP (511 nm) and Y(el-

low)FP (540 nm) can be used. FRET can occur between

two GFPs or one GFP and a second fluorophore. Ac-

cording to the spectral overlap, potential FRET pairs in-

clude BFP–GFP, CFP–YFP, Sapphire–fluorescein, and

GFP–fluorescein molecules.

4D microscopy provides the possibility to perform

time-resolved energy transfer imaging which enables

studies of the static and dynamic mobility of macromole-

Fig. 14 Fluorescence lifetime image of a living Chinese hamster

ovary cell labeled with the DNA probe Hoechst 33342

Fig. 15 Principle of

two-photon fluorescence

resonance energy transfer

Page 13

cules. Often, the fluorescence decays are relatively sensi-

tive to the acceptor-donor distances (Lakowicz 1983).

Non-invasive nanosurgery in tissues

As pointed out earlier, nanosurgery can be performed

within a single living cell. In principle, knocking out of

cellular structures can be performed deep in living tissue

without disturbing surface layers. Such highly precise

NIR laser-based nanoprocessing of cellular structures

without compromising the vitality of cells has numerous

potential applications in cell and developmental biology

particularly in studies addressing spatial-temporal con-

trol of developmental events and functional interactions

between organelles, as well as cell–cell communication.

Initial NIR femtosecond laser experiments have been

performed in living plant tissue. Using 800-nm laser

pulses, laser-mediated dye loading (Tirlapur and König

1999) and intratissue nanosurgery of the cell wall in a

living leaf of Elodea densa has also been successfully

accomplished. Following treatment, the ultrastructural

analysis of the corresponding area revealed a clean non-

staggering cut across the cell wall that measured less

than 350 nm. More interestingly, using the same NIR

femtosecond laser pulses, a single plastid or a part of the

organelle could be completely knocked out without af-

fecting the adjacent organelles or the viability of the cell.

The vitality of the cells after nanoprocessing has been

ascertained by exclusion of propidium iodide from the

cells as well as by the presence of cytoplasmic streaming

(Tirlapur and König submitted).

Potential medical applications include the use of fem-

tosecond laser microscopes in eye- and neurosurgery, tis-

sue engineering, laser-assisted IVF, and gene therapy.

Universal NIR laser-based optical workstations

Currently available laser microscope systems utilize

Nd:YAG lasers or laser diodes for optical trapping, nitro-

gen lasers for microsurgery, and titanium:sapphire lasers

for multiphoton microscopy (Greulich 1999). In contrast

to such an elaborate systems using varied kinds of lasers,

NIR laser workstations in future can be based on a single

laser that combines the possibility of optical trapping,

non-contact cell transport, non-linear fluorescence imag-

ing with high spatial and temporal resolution, and photo-

chemical microscopy, as well as nanosurgery.

Conclusions

NIR laser microscopes can be used as versatile non-inva-

sive biomedical tools for optical micromanipulation, di-

agnostics, photochemistry, and surgery. It is therefore

conceivable that in the following years NIR microscopy

has enormous potential to become a method of choice in

biotechnology, cell biology, and medicine.

Acknowledgements I thank K.-J. Halbhuber, director of the Insti-

tute of Anatomy II, for his constant support and encouragement

and all the members of my research group, including A. Göhlert,

P. Fischer, I. Lemke, H. Oehring, C. Peuckert, I. Riemann, and

U.K. Tirlapur for their help in various ways. Thanks are also due

to M.W. Berns, B. Tromberg, T. Krasieva, Y. Liu, H. Liang, Y.

Tadir (Beckman Laser Institute and Medical Clinic, Irvine, USA),

L. Svaasand (Norwegian Institute of Technology, Trondheim, Nor-

way), and K.O. Greulich (Institute of Molecular Biotechnology,

Jena, Germany) for their support in laser tweezers experiments,

and P. So (MIT, Cambridge, USA) and E. Gratton (University of

Illinois, Urbana-Champaign, USA) for their support in the initial

femtosecond microscopy experiments. The work in the authors

laboratory is supported in part by grants from the Federal Ministry

of Education, Science, Research and Technology (BMBF), the

German Science Foundation (DFG), the Ministry of Science,

Research and Culture of the state of Thuringia (TMWFK), Zeiss,

Jena, and Coherent.

References

Ashkin A (1970) Acceleration and trapping of particles by radia-

tion pressure. Phys Rev Lett 24:156–159

Ashkin A (1980) Application of laser radiation pressure. Science

210:1081–1088

Ashkin A, Dziedzic JM (1987) Optical trapping and manipulation

of viruses and bacteria. Science 235:1517–1520

Ashkin A, Dziedzic JM (1989) Internal cell manipulation using

near infrared laser traps. Proc Nat Acad Sci USA 86:7914–

7918

Block SM, Goldstein LS, Schnapp BJ (1990) Bead movement by

single kinesin molecules studies with optical tweezers. Nature

348:348–352

Booth MJ, Hell SW (1998) Continuous wave excitation two-photon

fluorescence microscopy exemplified with the 647-nm ArKr

laser line. J Microsc 190:298–304

Buican TN, Smyth MJ, Crissman HA, Salzman GC, Stewart CC,

Martin JC (1987) Automated single-cell manipulation and

sorting by light trapping. Appl Opt 26:5311–5316

Chong WF, Prahl SA, Welch AJ. (1990) A review of the optical

properties of biological tissues. IEEE J Quantum Electron

26:2166–2185

Cunningham ML, Johnson JS, Giovanazzi SM, Peak MJ (1985)

Photosensitised production of superoxide anion by monochro-

matic (290–405 nm) ultraviolet irradiation of NADH and

NADPH coenzymes. Photochem Photobiol 42:125–128

Denk W (1994) Two-photon scanning photochemical microscopy:

mapping ligand-gated ion channel distribution. Proc Natl Acad

Sci USA 91:6629–6633

Denk W, Svoboda K (1997) Photon upmanship: why multiphoton

imaging is more than a gimmick. Neuron 18:351–357

Denk W, Strickler JH, Webb WW (1990) Two-photon laser scan-

ning microscope. Science 248:73–76

Florin EL, Pralle A, Stelzer EHK, Hörber JKH (1998) Photonic

force microscope calibration by thermal noise analysis. Appl

Phys A 66:575–578

Göppert-Meyer M (1931) Über Elementarakte mit zwei Quanten-

sprüngen. Göttinger Dissertation. Ann Phys 9:273–294

Greulich KO (1999) Micromanipulation by light in biology and

medicine. Birkhäuser, Basel

Gryczynski I, Szmaczinski H, Lakowicz JR (1995) On the possi-

bility of calcium imaging using Indo-1 with three photon exci-

tation. Photochem Photobiol 62:804–808

Hänninen PE, Soini E, Hell SW (1994) Continuous wave excita-

tion two-photon fluorescence microscopy. J Microsc 176:222–

225

Hell SW, Bahlmann K, Schrader M, Soini A, Malak H, Gryczynski

I, Lakowicz JR (1996) Three-photon excitation in fluorescence

microscopy. J Biomed Opt 1:71–74

Kaiser W, Garret C (1961) Two-photon excitation in CaF

:Eu

Phys Rev Lett 7:229–231

Page 14

König K (1998) Laser tweezers are sources of two-photon excita-

tion. Cell Mol Biol 44:721–734

König K (2000) Invited review: multiphoton microscopy in life

sciences. J Microsc (in press)

König K, Schneckenburger H (1994) Laser-induced autofluores-

cence for medical diagnosis. J Fluorescence 4:17–40

König K, Liang H, Berns MW, Tromberg B (1995a) Cell damage

by near-IR beams. Nature 377:20–21

König K, Liu Y, Sonek GJ, Berns MW, Tromberg BJ (1995b) Au-

tofluorescence spectroscopy of optically-trapped cells during

light stress. Photochem Photobiol 62:830–835

König K, Tadir Y, Patrizio P, Berns MW, Tromberg BJ (1996a)

Effects of ultraviolet exposure and near infrared laser tweezers

on human spermatozoa. Hum Reprod 11:2162–2164

König K, Svaasand L, Liu Y, Sonek G, Patrizio P, Tadir Y, Berns

MW, Tromberg BJ (1996b) Determination of motility forces of

human spermatozoa using an 800 nm optical trap. Cell Mol

Biol 42:501–509

König K, Liang H, Berns MW, Tromberg B (1996c) Cell damage

in near infrared multimode optical traps as a result of multi-

photon absorption. Opt Lett 21:1090–1092

König K, So PTC, Mantulin WW, Tromberg BJ, Gratton E

(1996d) Two-photon excited lifetime imaging of autofluores-

cence in cells during UVA and NIR photostress. J Microsc

183:197–204

König K, Böhme S, Leclerc N, Ahuja R (1998) Time-gated auto-

fluorescence microscopy of motile green microalga in an opti-

cal trap. Cell Mol Biol 44:763–770

König K, Becker TW, Fischer P, Riemann I, Halbhuber KJ

(1999a) Pulse-length dependence of cellular response to in-

tense near-infrared laser pulses in multiphoton microscopes.

Opt Lett 24:113–115

König K, Riemann I, Fischer P, Halbhuber KJ (1999b) Intracellu-

lar nanosurgery with near infrared femtosecond laser pulses.

Cell Mol Biol 45:195–201

König K, Riemann I. Fischer P, Halbhuber KJ (2000a) Multiplex

FISH and three dimensional DNA imaging with near infrared

femtosecond laser pulses. Histochem Cell Biol (in press)

König, K, Göhlert, A, Liehr T, Loncarevic IF, Riemann, I (2000b)

Two-photon multicolour FISH: a versatile technique to detect

specific sequences within single DNA molecules in cells and

tissues. Single Mol 1:41–51

Kuo SC, Scheetz MP (1993) Force of single kinesin molecules

measured with optical tweezers. Science 260:232

Lakowicz JR (1983) Principles of fluorescence spectroscopy. Ple-

num Press, New York

Li X, Darzynkiewicz Z (1999) The Schrödinger’s cat quandary in

cell biology: integration of live functional analysis with mea-

surements of fixed cells in analysis of apoptosis. Exp Cell Res

249:404–412

Liu Y, Cheng D, Sonek GJ, Berns MW, Chapman CF, Tromberg

BJ (1995) Evidence for localised cell heating induced by near

infrared optical tweezers. Biophys J 68:2137–2144

Maiti S, Shear JB, Williams RM, Zipfel WR, Webb WW (1997)

Measuring serotonin in live cells with three-photon excitation.

Science 275:530–532

Masters BR, So PTC, Gratton E (1997) Multiphoton excitation

fluorescence microscopy and spectroscopy of in vivo human

skin. Biophys J 72:2405–2412

Oehring H, Riemann I, Fischer P, Halbhuber KJ, König K (2000)

Ultrastructure and reproduction behaviour of single CHO-K1

cells exposed to near infrared femtosecond laser pulses. Scan-

ning (in press)

Pawley JB (1995) Handbook of biological confocal microscopy,

2nd edn. Plenum Press, New York

Perkins TT, Quake SR, Smith DE, Chu S (1994) Relaxation of sin-

gle DNA molecule observed by optical microscopy. Science

264:822–826

Piston DW, Masters BR, Webb WW (1995) Three-dimensionally

resolved NAD(P)H cellular metabolic redox imaging of the in

situ cornea with two-photon excitation laser scanning micros-

copy. J Microsc 178:20–27

Pollak BA, Heim R (1999) Using GFP in FRET-based applica-

tions. Trends Cell Biol 9:57–60

Schönle A, Hell S (1998) Heating by absorption in the focus of an

objective lens. Opt Lett 23:325–327

So PTC, König K, Berland K, Dong CY, French T, Bühler C,

Ragan T, Gratton E. (1998) Review: new time-resolved tech-

niques in two-photon microscopy. Cell Mol Biol 44:771–793

Speel A (1999) Detection and amplification systems for sensitive

multiple-target DNA and RNA in situ hybridisation: looking

inside cells with a spectrum of colours. Histochem Cell Biol

112:89–113

Squirrel JM, Wokosin DL, White JG, Bavister BD (1999) Long-

term two-photon fluorescence imaging of mammalian embry-

os without compromising viability. Nat Biotechnol 17:763–

767

Svoboda K, Block SM (1994) Biological applications of optical

forces. Annu Rev Biophys Biomol Struct 23:247–285

Tadir Y, Wright WH, Vafa O (1989) Micromanipulation of sperm

by a laser-generated trap. Fertil Steril 52:870–873

Tirlapur UK, König K (1999) Near infrared femtosecond laser

pulses as a novel non-invasive means for dye-permeation and

3D imaging of localised dye-coupling in the Arabidopsis root

meristem. Plant J 20:363–370

Tyrell RM, Keyse SM (1990) The interaction of UVA radiation

with cultured cells. J Photochem Photobiol 4:349–361

Wiedemann R, Montag M (1994) Laser-assisted oocyte microin-

semination and direct transuterine tubal transfer in male infer-

tility. Poster. Intern Symposium on male factor in human infer-

tility. 21–22 April, Paris

Wokosin DL, Centonze, VE, Crittenden A, White JG (1995) Three

photon excitation of blue emitting fluorophores by laser scan-

ning microscopy. Mol Biol Cell 6:113a

Xu C, Zipfel W, Shear JB, Williams RM, Webb WW (1996)

Multiphoton fluorescence excitation: new spectral windows

for biological nonlinear microscopy. Proc Natl Acad Sci USA

93:10763–10768

Zahn M, Seeger S (1999) Optical tweezers in pharmacology. Cell

Mol Biol 44:747–761