Femtosecond nearinfrared laser pulses as a versatile noninvasive tool ...

6 downloads 22015 Views 327KB Size Report
non-invasive tool for intra-tissue nanoprocessing in plants without compromising ... *For correspondence (fax: +49 3641 938552; email: [email protected]). Summary ...... intrinsic bulk damage of wide-gap optical materials. Opt. Eng.
The Plant Journal (2002) 31(3), 365±374

TECHNICAL ADVANCE

Femtosecond near-infrared laser pulses as a versatile non-invasive tool for intra-tissue nanoprocessing in plants without compromising viability Uday K. Tirlapur and Karsten KoÈnig* Laser Microscopy Division, Institute of Anatomy II, Friedrich Schiller University, Teichgraben 7, D-07743 Jena, Germany Received 17 December 2001; revised 25 March 2002; accepted 12 April 2002 *For correspondence (fax: +49 3641 938552; email: [email protected])

Summary In this report, we describe a highly reproducible femtosecond near-infrared (NIR) laser-based nanoprocessing technique that can be used both for non-invasive intra-tissue nanodissection of plant cell walls as well as selective destruction of a single plastid or part thereof without compromising the viability of the cells. The ultra-precise intra-tissue nanoprocessing is achieved by the generation of high light intensity (1012 W cm±2) by diffraction-limited focusing of the radiation of an NIR (l = 740 and 800 nm) femtosecond titanium-sapphire laser to a sub-femtolitre volume and subsequent highly localized instantaneous plasma formation. Following nanosurgery, electron microscopical analysis of the corresponding cellular target areas revealed clean non-staggering lesions across the cell wall with a cut width measuring less than 400 nm. To our knowledge, this is the smallest cut made non-invasively within a plant tissue. Further evidence, including two-photon imaging of chlorophyll ¯uorescence, revealed that a single target chloroplast or part thereof can be completely knocked out using intense ultra-fast NIR pulses without any visible deleterious effect on the adjacent plastids. 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. The potential applications of this technical advance include developmental biology applications, particularly studies addressing spatio-temporal control of ontogenetic events and cell±cell interactions, and gravitational biology applications. Keywords: chloroplasts, Elodea, femtosecond laser, gravitropism, nanoprocessing, near-infrared, multi-photon ionization, plasma formation

intra-tissue,

non-invasive,

Introduction Multi-photon ¯uorescence microscopes based on nonlinear chromophore excitation through contemporaneous absorption of two or more NIR photons in the sub-femtolitre focal volume of a high numerical aperture objective are considered safer than conventional laser scanning microscopes for vital imaging of various biological specimens (Diaspro et al., 1999; KoÈnig, 2000; Squirrell et al., 1999; Tirlapur and KoÈnig, 2001, 2002; Wokosin et al., 1996). Because there are virtually no ef®cient cellular absorbers in the 740±1200 nm spectral NIR region, there is no out-of-focus photo-damage, and, moreover, as the multi-photon effect is spatially ã 2002 Blackwell Science Ltd

con®ned to the minute sub-femtolitre focal volume, there is no out-of-focus photo-bleaching (Diaspro, 1999; KoÈnig, 2000; Tirlapur and KoÈnig, 2002). These advantages, along with increased penetration depth of the NIR photons, have proven to be invaluable for acquisition of serial optical sections of almost complete root tips (Tirlapur and KoÈnig, 1999) and embryos (Squirrell et al., 1999), as well as for in vivo time-lapse imaging of chloroplast division in deeply seated living bundle sheath cells of the leaf tissue in Arabidopsis (Tirlapur and KoÈnig, 2001) without compromising viability. 365

366 Uday K. Tirlapur and Karsten KoÈnig Safe non-destructive multi-photon ¯uorescence imaging can be performed using an 80 MHz Ti:sapphire laser with NIR femtosecond (170 fsec) excitation intensities in the range of MW cm±2 to GW cm±2 at mean powers < 10 mW and microsecond beam dwell time. Nevertheless, above certain higher mean power thresholds where the laser intensities are in the range of TW cm±2, non-linearly induced instantaneous destructive effects based on multi-photon ionization of molecules (Oraevsky et al., 1996) and plasma formation are known to occur in the sub-femtolitre excitation volume (KoÈnig and Tirlapur, 2002; KoÈnig et al., 1999). In this study, we have used these spatially con®ned highly destructive non-linear multi-photon effects of intense NIR femtosecond laser pulses and demonstrate a versatile noninvasive technique for intra-tissue nanodissection of cell walls as well as high-precision knock-out of individual organelles or parts thereof without compromising viability.

densa Planch. consist of two distinct layers (Figure 1a), each one cell thick, with regularly distributed intracellular spaces between the large upper cells and the small lower ones. These features are consistent with those reported by Rascio et al. (1991).

Results

Non-invasive intra-tissue nanodissection of cell walls

Three-dimensional two-photon ¯uorescence imaging revealed that the leaves of the aquatic plant Elodea

As discernible from the electron micrograph (Figure 1b), 740 nm NIR femtosecond laser pulses at a mean power of

Vitality of cells after optical micro-manipulation Analysis of time-lapse two-photon transmission and corresponding ¯uorescence images after knocking out part of a target chloroplast (Figures 2 and 3) revealed virtually no effect on the pattern of cytoplasmic streaming in the cells. Furthermore, there was hardly any in¯ux of propidium iodide from the surrounding medium into the cells after manipulation of chloroplasts as ascertained by the absence of any cytoplasmic propidium iodide ¯uorescence following two-photon excitation. Hence, it is conceivable that the viability of the cells is not compromised after nanoprocessing with a laser beam power of 30±50 mW.

Figure 2. Images of semi- and ultrathin sections depicting the intratissue sites of non-invasive nanodissection of the cell wall. (a) Scanned image of a semi-thin section of an E. densa leaf after emboss ®ltering showing the larger upper epidermal (ue) and smaller lower epidermal (le) cells and the site (lightning symbol) where nanodissections were performed in replicate samples (n = 18). Scale bar = 50 mm. (b) Ultrastructural details depicting the ultraprecise cut (size < 400 nm) in the cell wall made by 740 nm NIR femtosecond pulses at a mean power of 50 mW. Scale bar = 1 mm.

ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

Nanodissection and nanoprocessing with NIR laser beams 30±50 mW can create sub-micrometre sized lesions in the wall of E. densa. These cuts were highly precise, with

Figure 2. Transmission (a,c) and auto¯uorescence images (b,d) of chloroplasts (darts) in the lower epidermal cell of E. densa before (a,b) and after (c,d) selective knock-out of a speci®c chloroplast (lightning symbol) with 740 nm NIR femtosecond laser pulses at a mean power of 30±50 mW. Electron micrograph of plastids in the lower epidermal cell (e) and (f) ultrastructural details following selective disruption of only half of the plastid using 800 nm NIR femtosecond laser at a mean power of 30±50 mW. P, plastid; CW, cell wall. Scale bars = 7 mm (a±d) and 3 mm (e,f).

ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

367

widths < 400 nm. At the increased mean power (30± 50 mW), a peak intensity of the order of ~1k W cm±2 is

368 Uday K. Tirlapur and Karsten KoÈnig

Figure 3. Emboss-®ltered transmission (a,c) and pseudo-colour-coded auto¯uorescence images (b,d) of chloroplasts in the epidermal cell of E. densa before (a,b) and 8 sec after (c,d) selective knock-out of part of a speci®c chloroplast (lightning symbol) with 800 nm NIR femtosecond laser pulses at a mean power of 30±50 mW in the presence of the cell-impermeate ¯uorescent dye propidium iodide (PI). Note the active movement of the chloroplasts in the cortical cytoplasmic region of the target cell (arrowheads) as well as in the adjacent cells (arrows) after nanoprocessing. No PI ¯uorescence is discernible in the cytoplasm of the cells, indicating that the cells remain viable. Distinct cytoplasmic streaming in the cortical region of the cells was invariably present even after 30 min. Scale bar = 50 mm. The inset pseudo-colour-coded bar represents a pixel intensity pro®le between 0 and 255 units.

ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

Nanodissection and nanoprocessing with NIR laser beams attained, evoking precise spatially localized phenomena of optical breakdown and plasma formation (Niemz, 1996; Stern et al., 1989). These instantaneous destructive effects accompanying multi-photon-induced plasma formation may result in ultra-precise removal of the cell wall as well as knock-out of a target organelle (see below). Because the formation of plasma is spatially con®ned to the sub-femtolitre focal volume of a high numerical aperture objective, there is virtually no perturbation of the adjacent cell components. Two-photon imaging and selective intracellular knock-out of chloroplasts Using 740 and 800 nm NIR femtosecond laser pulses at a very low mean power of < 0.2 mW, we could image the chloroplasts in transmission mode (Figure 2a) as well as inducing the auto¯uorescence of chlorophyll associated with the chloroplasts by two-photon excitation (Figure 2b). When a single chloroplast (Figure 2a±c) was exposed to 740 or 800 nm femtosecond laser at a mean power of 30 mW for a very short duration of 0.013 sec, the entire irradiated chloroplast was completely destroyed (Figure 2c), and the associated chlorophyll auto¯uorescence completely disappeared (Figure 2d). Comparison of the auto¯uorescence associated with the immediately adjacent chloroplast following knock-out of the target chloroplast did not show any changes (Figure 2b,d). The ultrastructural features of the plastids in the lower epidermal cells of E. densa (Figure 2e) are comparable to those described by Rascio et al. (1991). Furthermore, when only half of a target chloroplast was exposed to intense 800 nm femtosecond pulses at mean power of 30 mW, only the exposed half of the plastid was completely disrupted (Figure 2f). Similarly, we could also selectively knock-out only a small part (less than half) of a chloroplast (Figure 3a±d) without affecting the cytoplasmic streaming of other chloroplasts in the cortical regions of the target cell or those adjacent to it. Discussion Viability of target cells is not compromised after nanoprocessing The viability of the cells after nanoprocessing was checked on a case-by-case basis. Our approach to this was twofold. The ®rst assessment was based on the signi®cantly different appearance of leaf cells when they are not alive. All living leaf cells of E. densa in phase-contrast transmission microscopy show characteristic cytoplasmic streaming (albeit at a slower rate in dark-adapted plants), while the dead cells hardly show any movement of organelles. After part of the chloroplast in a particular cell had been knocked ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

369

out, we compared the cytoplasmic streaming of chloroplasts in the target cell and the neighbouring cells that had not been irradiated. In all instances, the cells looked alike and normal cytoplasmic streaming in the cortical region was observed even after 30 min. This was true even when a entire chloroplast in a cell was knocked out. A second, more careful approach involved use of the vital stain propidium iodide (PI). It is generally known that as long as a cell is alive, it is able to prevent the dye from entering the cytoplasm. When the cell is dead or cannot maintain its normal functions, the dye will penetrate the cell membrane and the whole cell appears red. After nanoprocessing, transmission and two-photon ¯uorescence images of respective cells obtained at a mean laser power of up to 3 mW did not show accumulation of the cell-impermeate ¯uorescent probe PI from the surrounding medium in the target cell. These tests unequivocally show that the cells remain viable after nanoprocessing and show normal cytoplasmic streaming.

Ablations based on continuous wave and nanosecond pulsed lasers in material processing as well as medical sciences Following the initial report of laser-assisted cutting of biological material (Leppard and Raju, 1965), several laboratories have used high-energy UV laser microbeams either to destroy the walls of plant cells (Greulich and Weber, 1992) or to ablate a small area of the extracellular matrix (Berger, 1998; Henriksen and Assmann, 1997). However, to date, the precise magnitude or dimensions of the ablated areas have never been unequivocally substantiated by ultrastructural analysis of cells/tissue in question. Nevertheless, these laser-mediated cell ablation studies have been of great value in gaining non-enzymatic access to the plasma membrane, and understanding root development as well as cell polarization. UV laser microbeams have also been adopted as non-mechanical tools for microsurgery and knock-out of other organelles such as mitochondria (Amy and Storb, 1965) in living plant and animal cells (Berns et al., 1981, Berns et al., 1991; Greulich, 1999). However, the use of high-intensity UV lasers light has major disadvantages. First, even millisecond exposure of plant cells to UV irradiation has been shown to drastically affect cells by altering calcium homeostasis (Frohnmeyer et al., 1999) and inducing cell death (Danon and Gallois, 1998). Second, the penetration depth of UV light in highly turbid and refractive plant tissue is low and thus it is extremely dif®cult to perform intra-tissue nanoprocessing (Berger, 1998). Moreover, UV radiation is also know to be responsible for oxidative stress, paving the way for numerous cellular perturbations leading to apoptosis.

370 Uday K. Tirlapur and Karsten KoÈnig Third, the pulse duration is long (often nanoseconds) and the beam quality is poor. In addition, almost all of the earlier UV laser-based microsurgical studies on plant and animal cells have been performed using nanosecond laser pulses which are known to induce photo-thermal and photo-mechanical damage to the adjacent areas. Microprocessing/ablation with nanosecond-pulsed UV versus femtosecond-pulsed NIR lasers Using a nanosecond pulse of UV laser at 337 nm with a peak power of 85 kW, Blanca¯or et al. (1998) have reported that, in Arabidopsis roots, they have been unable to selectively ablate a single root cap columella cell without compromising the viability of the neighbouring cells. This is consistent with the expectation that the UV laser has to pass through all of the cells along the z axis, and, moreover, the authors explicitly mention that they have also been unable to consistently visualize a columella cell at a particular depth to ensure alignment in the z axis so as to avoid damage to columella cells directly above and below the target cell. We have made similar observations using the 488 nm laser line of a conventional confocal laser scanning microscope (CLSM) (Tirlapur and KoÈnig, 1999), in which several cells above and below the target cell (in the root cap columella of Arabidopsis) were shown to experience photo and thermal stress. By contrast, using femtosecond NIR laser pulses, it is possible to selectively visualize and knock out a single organelle such as the chloroplast or part thereof within a target cell contained in the leaf tissue (present study) without damaging the integrity and viability of the adjacent cells (see Figure 3). With nanosecond-pulsed UV lasers, in addition to material removal because of plasma shielding and plasma heating, massive temperature gradients in the micron range are generated. In contrast to nanosecond-pulsed UV laser-mediated microsurgery, the pulse duration in NIRbased femtosecond laser nanosurgery is extremely short for heat transfer to occur, and hence spatially con®ned nanodissection is possible without micrometre-sized collateral damage to the immediately adjacent areas. Recently, the atomic force microscope (AFM) has been employed as a new mechanical tool for microdissection of human chromosome 2 (Thalhammer et al., 1997) as well as nanodissection of plasmid DNA (Geisler et al., 2000). Because the high-resolution AFM is mainly adaptable for probing biomolecules attached to ¯at substrates, it is virtually impossible to use it as a micro-manipulative tool for intracellular surgery and deep tissue nanoprocessing of three-dimensionally oriented structures in living plant and/ or animal cells. Hence, the NIR femtosecond laser-based novel technique described here may potentially become the method

of choice for non-invasive intra-tissue nanosurgery and nanoprocessing of intracellular organelles in different cell types. Potential applications of the NIR femtosecond laserbased nanoprocessing technique The unique ultra-precise non-invasive intra-tissue nanoprocessing technique described here that harnesses the destructive non-linear multi-photon effects of intense NIR femtosecond laser pulses can be conveniently employed in numerous experimental studies to resolve some of the intricate processes in plant biology including those discussed below. Spatio-temporal control of ontogenetic events and cell±cell interactions. During the ontogenetic differentiation of speci®c cell types in the plant body, for example the root hair cells (Duckett et al., 1994; Tirlapur and KoÈnig, 1999) and stomatal guard cells (Palevitz and Hepler, 1985), specialized cells are distinguishable from the adjacent cells by their morphological and functional differences. It has been hypothesized that, prior to the attainment of distinct features of the root hair cell and stomatal guard cell, it may be necessary for such cells to be symplastically isolated. This means that cell-to-cell communication through plasmodesmata (the intercellular symplastic junctions) between these cells and the neighbouring cells has to be curtailed and uncoupled. This has been indirectly ascertained by analysis of symplastic (via plasmodesmata) dye coupling patterns during growth and differentiation of root hair (Duckett et al., 1994; Tirlapur and KoÈnig, 1999) and guard cells (Palevitz and Hepler, 1985). It would be interesting to know how, if at all, the process of differentiation is affected by selective re-establishment and potentiation of symplastic communication between differentiating root hair cells and stomatal guard cells and their respective neighbouring cells by non-invasive creation of nanometersized channels in the cell walls using the NIR femtosecond laser-based technique (described in this study). Thus, the technique described for highly precise nanoprocessing of cell walls may be invaluable in determining whether restriction of cell-to-cell communication is indeed required for ®nal cell morphogenetic events to occur during pattern formation. Such an approach could eventually complement or contradict the detailed molecular and functional analysis of structural components of the plasmodesmatal complex implicated in cellular coupling (see Baluska et al., 2001). Role of statoliths in gravitropism. In plants, gravitropism is one of the essential sensory mechanisms that enables them to perceive and respond to gravitational stimulus (see Kiss, 2000). Gravitropism involves three stages, namely perception, transduction and response (Fitzelle ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

Nanodissection and nanoprocessing with NIR laser beams and Kiss, 2001; Salisbury, 1993). The cells of the endodermis are believed to be the sites of graviperception in shoots (Chen et al., 1999; Fukaki et al., 1998), while perception of gravity in roots is hypothesized to occur in the columella cells of the root cap (Sack, 1997; Volkmann and Sievers, 1979). The two principal models for the perception stage of gravitropism include the starch±statolith hypothesis and the protoplast pressure theory (Kiss, 2000). The underlying nature of the graviperception mechanism in plants is as yet unresolved due to the lack of convincing evidence for either of these theories (Barlow, 1995; Fitzelle and Kiss, 2001). In line with the starch±statolith theory, physical displacement of and/or accompanying pressure from the amyloplasts (plastids containing starch) is known to elicit a physico-mechanical signal for perception of gravity (see Kiss et al., 1989). Several studies, including correlational studies, indicate that sedimenting amyloplasts are involved in perception of the gravitational stimulus in plants (Audus, 1979; Kiss et al., 1989; Volkmann and Sievers, 1979). Interestingly, a mutant of Arabidopsis thaliana that is unable to synthesize starch has been shown to be gravity-responsive (Casper and Pickard, 1989), suggesting that starch-containing organelles (amyloplasts) are not necessary for graviperception in the roots of Arabidopsis. This led Casper and Pickard (1989) to suggest that the contribution of starch within the amyloplasts is rather minimal for graviperception. Given the argument addressing the functional signi®cance of amyloplasts with or without starch, it would be interesting to determine the role of plastids (with or without starch) in realising the gravitropic response in plants. Ultra-precise nanoprocessing by way of highly selective knock-out of a single plastid or part thereof has enormous implications in unravelling the role of starch-containing plastids (amyloplasts) in gravitropic responses in higher plants. Laser parameters for nanoprocessing in plant versus animal systems. To date, very few attempts have been made to use femtosecond NIR lasers for nanoprocessing of cellular components in living animal systems (KoÈnig and Tirlapur, 2002; KoÈnig et al., 1999) and virtually none with plant systems. However, it is logical to believe that the NIR laser parameters for nanosurgery and nanoprocessing of higher plant versus animal systems do indeed require optimization, mainly because plants cells have a rigid cell wall (unlike animal cells) and quite often contain numerous chloroplasts; these structures have distinct UVabsorbing and two-photon excitable ¯uorophores, such as fruleic acid associated with the cell walls (Lichtenthaler and Miehe, 1997), and chlorophylls and carotonoids present in the chloroplasts (Tirlapur and KoÈnig, 2002) of green plant tissue. It is furthermore interesting to note that nanoprocessing of chromosomes within living CHO cells (KoÈnig et al., 1999) required a mean NIR (l = 720 nm) laser ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

371

power of 30 mW, while incomplete partial incisions of semi-dried human metaphase chromosomes in in vitro preparations (KoÈnig et al., 2001) required a 800 nm 170 fsec laser at a mean laser power of 40 mW. It is therefore conceivable that optimization of nanoprocessing might be dependent on the NIR laser wavelength used and the physical condition of the target material. In this study, we could precisely create nanometre-sized cuts across the plant cell wall at any desired location using a mean laser power of 50 mW without compromising the viability of the target cell or those adjacent to it. Ultra-precise tissue ablation by multi-photon evoked plasma formation. The underlying mechanism of laser tissue interactions when using femtosecond NIR laser pulses has, as yet, not been extensively investigated. It is generally believed and now acknowledged that, in lowabsorbing tissues, multi-photon-mediated optical breakdown at TW cm±2 light intensities is the only mechanistic means of energy deposition (Oraevsky et al., 1996). Moreover, optical breakdown in water-containing media such as biological tissue remains incompletely understood (Bunkin and Bunkin, 1993; Oraevsky et al., 1996). Nevertheless, it has often been found that sub-picosecond pulses produce substantially lower collateral damage to adjacent cells and tissue components than microsecond and picosecond pulses (Vogel et al., 1999). In addition, irrespective of the mechanical properties of the solid-state dielectrics such as quartz and ceramics, plasma generated by way of multi-photon ionization has been shown to induce effective and precise knock-out of sub-micron regions (Jones et al., 1989; Stuart et al., 1996a; Stuart et al., 1996b). Hence it has been suggested that plasmamediated tissue ablation can be employed as an precise and effective means for almost instantaneous removal of hard tissue components. This has now been realised in this study wherein we unequivocally provide the ®rst direct evidence of plasma-mediated ultra-precise knockout of sub-micron sectors of plant cell wall within an intact cell of an plant organ without compromising its viability or that of the neighbouring cells. Taken together, we have successfully exploited the destructive effects of high power NIR femtosecond laser pulses and developed a ultra-precise non-invasive technique for intra-tissue nanoprocessing of plant cell walls and organelles without perturbing the functional integrity and vitality of the target cells or organelles.

Experimental procedures Plants of Elodea densa Planch. (Aston, 1973) used in this study were grown in glass tanks ®lled with tap water at 22°C and 10 h daylight. Individual plantlets were carefully transferred into an microscope chamber containing 5 mg ml±1 propidium iodide

372 Uday K. Tirlapur and Karsten KoÈnig (Molecular Probes Inc., Eugene, Oregon, USA), and a fully differentiated leaf was randomly chosen for nanoprocessing.

Laser systems and technique for intra-tissue nanoprocessing The NIR laser beam at 740 nm was derived from a tuneable 76 MHz femtosecond titanium:sapphire laser source (Mira model 900-F, Coherent, Santa Clara, California, USA) with a typical width of 170 fsec. The 800 nm laser beam was obtained from either a compact mode locked femtosecond Ti:sapphire laser (Vitesse model, Coherent: operating at 80 MHz pulse frequency, 1 W output power and 90 fsec output pulse duration) or from the ®rst tuneable femtosecond pulsed Ti:sapphire laser source in one box (MaiTai, Spectra-Physics Lasers Inc., Mountain View, California, USA), also operating at 80 MHz pulse frequency and approximately 90 fsec output pulse duration, but at approximately 900 mW output power). The laser beams were coupled to a modi®ed inverted Zeiss CLSM model 410 and focused through a high numerical aperture (NA 1.3) 40 3 Zeiss Neo¯uar objective (Tirlapur and KoÈnig, 2002). Laser powers at the object plane were routinely measured using the Fieldmaster FM power meter (Coherent) as described previously (Tirlapur and KoÈnig, 1999). The two cell layers of the leaf were initially visualized with bright-®eld NIR illumination and the wall area of the lower epidermal cell was brought into focus. In order to selectively target the plastids, two-photon imaging of the chlorophyll ¯uorescence (Tirlapur and KoÈnig, 2001) associated with the chloroplast was performed at a very low mean laser power of about 0.2 mW. To knock out a single plastid or part thereof, we targeted the chloroplasts in the vicinity of the nuclei of the cells in the lower epidermis of the leaf. This was necessary because these chloroplasts are more stably anchored than those in the cortical region of the cell where there is vigorous cytoplasmic streaming. Alternatively, we also used dark-adapted leaves which show hardly any cytoplasmic movement of chloroplasts. In the latter, subsequent exposure of the dark-adapted leaves to white light invariably evoked cytoplasmic streaming and movement of the chloroplasts as described by Forde and Steer (1976). An entire target plastid or part of the plastid were exposed to 740 or 800 nm NIR laser for 0.013 sec at a mean power of 30±50 mW by preselecting a region of interest (ROI) that corresponded to a 5 3 5 mm area or half of it. For nanodissection of the walls, the same experimental set-up was used but at higher mean laser powers of either 10±15 mW or 30±50 mW, and an ROI approximately 1 mm thick was scanned for 0.64 sec at the focused wall region of a cell in the lower epidermis of the leaf.

cent probes used for evaluating membrane barrier dysfunction may be less sensitive for assaying of cell viability in the time scale of a few seconds. This is because certain instances of delayed cell death by apoptosis or necrosis might be initiated with very little if any indication of membrane barrier dysfunction. However, visual observation of the cells in transmission mode revealed normal cytoplasmic streaming in the cortical region even after 30 min after speci®cally knocking out a single chloroplast in the vicinity of the nuclei. This latter ®nding of normal cytoplasmic streaming in the target cell 30 min after nanoprocessing provides convincing evidence for cell viability and thus almost totally excludes the possibility of cellular damage. Nevertheless, we note that it would be more desirable to evaluate the viability of the speci®c cell over several hours or perhaps days after nanoprocessing of cellular components.

Electron microscopy of nanoprocessed cells and organelles Having con®rmed that the cells subsequent to nanoprocessing were still alive, speci®c leaves were carefully separated from the plant body and individually ®xed in primary ®xative containing 2% v/v glutaraldehyde (Fluka, Buchs SG, Switzerland) in 0.1 M sodium cacodylate buffer, pH 7.2, at 5°C for 2 h. The leaves were gently rinsed three times in the buffer and post-®xed in 1% w/v OsO4 in cacodylate buffer, pH 7.2, at 5°C for 2 h. They were then washed, dehydrated in graded ethanol series, and embedded in Epon 812 (Ferak, Berlin, Germany). Serial ultra-thin sections (70 nm) were cut with a diamond knife using the Ultracut S microtome (Leica, Vienna, Austria) and collected on formvarcoated copper grids. All sections were subsequently stained in 1% uranyl acetate and freshly prepared 0.2% lead citrate and examined with EM 902A (Zeiss, Oberkochen, Germany) operating at 80 kV.

Acknowledgements We are grateful to the three anonymous reviewers for their constructive comments on an earlier version of this manuscript. The authors wish to thank Ursula Eschler for excellent technical assistance. This research was supported by grants from the Deutsche Forschungsgemeinschaft (KO1361/10-1), the ThuÈringer Ministerium fuÈr Wissenschaft Forschung und Kunst (B509-00004) and the Bundesministerium fuÈr Bildung und Forschung (01ZZ0105).

References Assessment of cell viability after nanoprocessing To ensure that the cells remained alive after micro-manipulation, we constantly monitored the cytoplasmic streaming which is an in vivo indicator of cell viability in Elodea (Forde and Steer, 1976). In addition, we also used the cell-impermeate dye propidium iodide (PI) which is known to accumulate in dead cells and is excluded from living cells. Following nanoprocessing, ¯uorescence images of respective cells were obtained at a mean laser power of up to 3 mW to ascertain the viability of cells. In no instance did cells show accumulation of the cell-impermeate ¯uorescent probe PI from the surrounding medium, thus con®rming the viability of cells. It is nevertheless pertinent to take into account that such ¯uores-

Amy, R.L. and Storb, R. (1965) Selective mitochondrial damage by a ruby laser microbeam: an electron microscope study. Science, 150, 756±757. Aston, H.I. (1973) Aquatic Plants of Australia. Melbourne, Australia: University Press. Audus, L.J. (1979) Plant geosensors. J. Exp. Bot. 30, 1051±1073. Baluska, F., Cvrckova, F., Kendrick, J. and Volkmann, D. (2001) Sink plasmodesmata as gateways for phloem unloading. Myosin VIII and calreticulin as molecular determinants of sink strength? Plant Physiol. 126, 39±46. Barlow, P.W. (1995) Gravity perception in plants: a multiplicity of systems derived by evolution? Plant Cell Environ. 18, 951± 962. ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

Nanodissection and nanoprocessing with NIR laser beams Berger, F. (1998) Cell ablation studies in plant development. Cell Mol. Biol. 44, 711±719. Berns, M.W., Aist, J., Edwards, J., et al. (1981) Laser microsurgery in cell and developmental biology. Science, 213, 505±513. Berns, M.W., Wright, W.H. and Steubing, R.W. (1991) Laser microbeam as a tool in cell biology. Int. Rev. Cytol. 129, 1± 44. Blanca¯or, E.B., Fasano, J.M. and Gilroy, S. (1998) Mapping the functional roles of cap cells in the response of Arabidopsis primary roots to gravity. Plant Physiol. 116, 213±222. Bunkin, N.F. and Bunkin, F.V. (1993) The concepts in the optical breakdown of transparent liquids. Laser Phys. 3, 63±77. Casper, T. and Pickard, B.G. (1989) Gravitropism in a starchless mutant of Arabidopsis: implications for the starch±statolith theory of gravity sensing. Planta, 177, 185±197. Chen, R., Rosen, E. and Masson, P.H. (1999) Gravitropism in higher plants. Plant Physiol. 120, 343±350. Danon, A. and Gallois, P. (1998) UV radiation induces apoptoticlike changes in Arabidopsis thaliana. FEBS Lett. 437, 131±136. Diaspro, A. (1999) Introduction to two-photon microscopy. Microsc. Res. Tech. 47, 163±164. Diaspro, A., Corosu, M., Ramoino, P. and Robello, M. (1999) Adapting a compact confocal microscope system to a twophoton excitation ¯uorescence imaging architecture. Microsc. Res. Tech. 47, 196±205. Duckett, C.M., Oparka, K.J., Prior, D.A., Dolan, L. and Roberts, K. (1994) Dye-coupling in the root epidermis of Arabidopsis is progressively reduced during development. Development, 120, 3247±3255. Fitzelle, K.J. and Kiss, J.Z. (2001) Restoration of gravitropic sensitivity in starch-de®cient mutants of Arabidopsis by hypergravity. J. Exp. Bot. 52, 265±275. Forde, J. and Steer, M.W. (1976) Cytoplasmic streaming in Elodea. Can. J. Bot. 54, 2688±2694. Frohnmeyer, H., Loyall, L., Blatt, M.R. and Grabov, A. (1999) Millisecond UV-B irradiation evokes prolonged elevation of cytosolic-free Ca2+ and stimulates gene expression in transgenic parsley cell cultures. Plant J. 20, 109±117. Fukaki, H., Wysocka-Diller, J., Kato, T., Fujisawa, H., Benfey, P.N. and Tasaka, M. (1998) Genetic evidence that endodermis is essential for shoot gravitropism in Arabidopsis thaliana. Plant J. 14, 425±430. Geisler, B., Noll, F. and Hamoo, N. (2000) Nanodissection and noncontact imaging of plasmid DNA with atomic force microscope. Scanning, 22, 7±11. Greulich, K.O. (1999) Micromanipulation by Light in Biology and Medicine. Basel, Switzerland: BirkhaÈuser-Verlag. Greulich, K.O. and Weber, G. (1992) The light microscope on its way from an analytical to preparative tool. J. Microsc. 167, 127± 151. Henriksen, G.H. and Assmann, S.M. (1997) Laser-assisted patch clamping: a methodology. Eur. J. Physiol. 433, 832±841. Jones, S.C., Braunlich, P., Casper, R.T., Shen, X.-A. and Kelly, P. (1989) Recent progress on laser-induced modi®cations and intrinsic bulk damage of wide-gap optical materials. Opt. Eng. 28, 1039±1047. Kiss, J.Z. (2000) Mechanisms of the early phases of plant gravitropism. Crit. Rev. Plant Sci. 19, 551±573. Kiss, J.Z., Hertel, R. and Sack. F.D. (1989) Amyloplasts are necessary for full gravitropic sensitivity in roots of Arabidopsis thaliana. Planta, 177, 198±206. KoÈnig, K. (2000) Multiphoton microscopy in life sciences. J. Microsc. 200, 83±104. KoÈnig, K. and Tirlapur, U.K. (2002) Cellular and subcellular ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374

373

perturbations during multiphoton microscopy. In: Confocal and Two-photon Microscopy: Foundations, Applications and Advances (Diaspro, A., ed.). New York: John Wiley & Sons Inc., pp. 191±205. KoÈnig, K., Reimann, I., Fischer, P. and Halbhuber, K.-J. (1999) Intracellular nanosurgery with near infrared femtosecond laser pulses. Cell Mol. Biol. 45, 195±201. KoÈnig, K., Reimann, I. and Fritzsche, W. (2001) Nanodissection of human chromosomes with near-infrared femtosecond laser pulses. Opt. Lett. 26, 819±821. Leppard, G.G. and Raju, M.V.S. (1965) The laser as a tool for isolating single cell. Can. J. Bot. 43, 955±958. Lichtenthaler, H.K. and Miehe, J.A. (1997) Fluorescence imaging as a diagnostic tool for plant stress. Trends Plant Sci. 2, 316± 320. Niemz, M.H. (1996) Laser±Tissue Interactions. Berlin: Springer. Oraevsky, A.A., Da Silva, L.B., Rubenchik, A.M., Freit, M.D., Glinsky, M.E., Perry, M.D., Mammini, B.M., Small, W. and Stuart, B.C. (1996) Plasma mediated ablation of biological tissues with nanosecond-to-femtosecond laser pulses: relative role of linear and non-linear absorption. IEEE J. Selected Topics Quantum Electronics, 2, 801±809. Palevitz, B.A. and Hepler, P.K. (1985) Changes in dye coupling of stomatal cells of Allium and Commelina demonstrated by microinjection of Lucifer yellow. Planta, 164, 473±479. Rascio, N., Mariani, P., Tommasini, E., Bodner, M. and Larcher, W. (1991) Photosynthetic strategies in leaves and stems of Egeria densa. Planta, 185, 297±303. Sack, F.D. (1997) Plastids and gravitropic sensing. Planta, 203, s63±s68. Salisbury, F.B. (1993) Gravitropism: changing ideas. Hort. Rev. 15, 233±278. Squirrell, J.M., Wokosin, D.L., White, J.G. and Bavister, B.D. (1999) Long-term two-photon ¯uorescence imaging of mammalian embryos without compromising viability. Nature Biotechnol. 17, 763±767. Stern, D., Schoenlein, R.W., Pulia®to, C.A., Dobi, E.T., Birngruber, R. and Fujimoto, J.G. (1989) Corneal ablation by nanosecond, picosecond and femtosecond lasers at 532 nm and 625 nm. Arch. Ophthalmol. 107, 587±592. Stuart, B.C., Feit, M.D., Herman, S., Rubenchik, A.M., Shore, B.W. and Perry, M.D. (1996a) Nanosecond-to-picosecond laser-induced breakdown in dielectrics. Physics Rev. 53, 1749±1176. Stuart, B.C., Feit, M.D., Herman, S., Rubenchik, A.M., Shore, B.W. and Perry, M.D. (1996b) Optical ablation by high-power sortpulse lasers. J. Opt. Soc. Am. B, 13, 459±468. Thalhammer, S., Stark, R.W., MuÈller, S., Wienberg, J. and Heckl, W.M. (1997) The atomic force microscope as a new microdissecting tool for the generation of genetic probes. J. Struct. Biol. 119, 232±237. Tirlapur, U.K. and KoÈnig, 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. Tirlapur, U.K. and KoÈnig, K. (2001) Femtosecond near infrared lasers as a novel tool for non-invasive real-time high resolution time-lapse imaging of chloroplast division in living bundle sheath cells of Arabidopsis. Planta, 214, 1±10. Tirlapur, U.K. and KoÈnig, K. (2002) Two-photon near infrared femtosecond laser scanning microscopy in plant biology. In: Confocal and Two-photon Microscopy: Foundations, Applications and Advances (Diaspro, A., ed.). New York: John Wiley & Sons Inc., pp. 449±468.

374 Uday K. Tirlapur and Karsten KoÈnig Vogel, A., Nahen, K., Theisen, D., Birngruber, R., Thomas, R. and Rockwell, B.A. (1999) Energy balance of optical breakdown in water at nanosecond to femtosecond time scales. Appl. Phys. 68, 271±280. Volkmann, D. and Sievers, A. (1979) Graviperception in multicellular organs. In: Encyclopaedia of Plant Physiology.

Physiology of Movements New Series. Volume 7 (Haupt, W. and Feinleib, M.E., eds). Berlin: Springer, pp. 573±600. Wokosin, D.L., Centonze, V.E., Crittenden, S. and White, J. (1996) Three-photon excitation ¯uorescence imaging of biological specimens using an all-solid-state laser. Bioimaging, 4, 208± 214.

ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 365±374