Open Access
Add to BibSonomyAbstract
Nanosecond laser pulses (λ=355 nm) were used to cut mechanosensory neurons in Caenorhabditis elegans and motorneurons in Drosophila melanogaster larvae. A pulse energy range of 0.8–1.2 µJ and <20 pulses in single shot mode were sufficient to generate axonal cuts. Viability post-surgery was >95% for C. elegans and 60% for Drosophila. Cavitation bubble dynamics generated due to laser-induced plasma formation were observed in vivo by time-resolved imaging in both organisms. Bubble oscillations were severely damped in vivo and cavitation dynamics were complete within 100 ns in C. elegans and 800 ns in Drosophila. We report the use of this system to study axonal transport for the first time and discuss advantages of nanosecond lasers compared to femtosecond sources for such procedures.
©2008 Optical Society of America
1. Introduction
Recent reports on using femtosecond (fs) lasers for conducting neuronal axotomy in C. elegans have generated a lot of excitement in the field of neuroscience [1, 2, 3]. The ability to selectively ablate an individual axon in a genetically tractable organism makes it an extremely powerful system to study several aspects of neuronal function with an important one being regeneration. Another fundamental cellular process, viz. transport of cargo, is essential in long polarized cells like neurons since protein synthesis occurs in the cell body and synaptic transmission at the end of the axon. Long term assays conducted over several hours to study transport were introduced in and were till recently limited to larger organisms such as mice [4]. These are usually carried out by tying a suture to compress long nerves such as the sciatic nerve and examining cargo accumulation at later time-points. Attempts to develop similar transport assays in smaller genetic model organisms such as Drosophila have met with some success [5]. Post axotomy imaging of transport in Aplysia neurons in culture has revealed that transport changes occur rapidly and a microtubule-dependent vesicle trap is set up that optimizes growth cone formation [6]. Since growth cone formation is an important step in axon regrowth it is critical to develop assays in vivo that allow determination of changes to transport during this process. We report the development of a transport assay in C. elegans using laser axotomy to physically block transport. The nanosecond (ns) laser used in our studies reproduced earlier results using fs systems for neuronal axotomy. By imaging accumulations of fluorescently labeled synaptic vesicle proteins at the site of axonal microsurgery we can characterize anterograde and retrograde transport. This assay is simple in procedure and provides quick results (1 hour) and could be used to study several aspects of transport in neurons, some of which we discuss later. We also studied the physical nature of the ablation process using time-resolved imaging to capture cavitation bubble dynamics in C. elegans and Drosophila larvae. Time-resolved imaging of the cavitation bubble size provides a direct measure of the amount of damage during axotomy procedures and can also provide estimates of the mechanical forces generated during the process. We compare cavitation dynamics from our study with those reported in a recent study on laser microsurgery in Drosophila embryos using 355 nm nanosecond laser pulses [7]. We also discuss the advantages of nanosecond lasers as compared to femtosecond lasers for such procedures.
2. Materials and methods
The 3rd harmonic output (λ=355 nm, τp=6 ns) of a Nd:YAG laser (Spitlight 600, Innolas, Munich, Germany) was introduced through the back port of an inverted microscope (Olympus iX71) and reflected by a dichroic mirror (Chroma 532/355-zet532nbdc) into the objective back aperture. The microscope has two back ports allowing separate paths for fluorescence excitation using a standard mercury arc lamp and ablation with the Nd:YAG laser. Transgenic C. elegans jsIs821 and jsIs37 expressing GFP::RAB-3 [8] and synaptobrevin::GFP (SNB-1:GFP) [9] respectively in mechanosensory neurons were used in our experiments. Drosophila OK371-Gal4; UAS-membrane GFP targeted to motor neurons were also used [10, 11]. C. elegans L4 larvae were anesthetized using 0.13% (w/v) sodium azide on an agar pad, while Drosophila 1st instar larvae were immobilized using 4°C cooling. Positioning the animal on the coverslip was random for both C. elegans and Drosophila. In C. elegans, focused posterior lateral mechanosensory (PLM) axons are 1 to 3 µm from the coverslip. In Drosophila, we targeted axon bundles 40 to 50 µm away from the ventral ganglion with these bundles being 5 to 10 µm from the coverslip. Axotomy was carried out with a 100×, 1.3 NA objective. Fluorescently labelled axons were brought into focus and irradiated, with the laser being operated in single shot mode. The operator manually delivered single pulses to the desired site till it was visually confirmed that the axon was cut. In all surgeries reported less than 20 pulses (typically 10 to 15) were required for generating a cut. Pulse energy used was 0.8 µJ for C. elegans and 1.2 µJ for Drosophila. Laser pulse energy was measured at the objective back aperture by removing the objective from the turret and allowing the beam to illuminate an energy detector (J3-05, Molectron Inc.). The manufacturer specified transmission of our objective at 355 nm is 40%. All pulse energies reported were calculated after factoring in the objective transmissivity. The probability of plasma formation in vivo was ≈30% at 0.8 µJ.
Time-resolved images of the ablation process were collected using a setup previously optimized in our lab [12]. In brief, the sample was illuminated with a broadband nanosecond pulse at a specified time delay with respect to the ablation laser pulse. The illumination pulse was generated by pumping a dye solution (LDS 698, Exciton Inc., 0.1 µM in methanol) with the 2nd (λ=532 nm) harmonic output of the Nd:YAG laser. The fluorescent emission of the dye was coupled into a fiber optic line with the distal end of the fiber being focussed onto the sample. The length of the fiber optic line decided the delay between the ablation pulse and the illumination pulse arriving at the sample. Images were collected on an intensified CCD (Andor iStar) with an exposure duration of 5 ns and gain of 120. Multiple images were collected at the same time delay to determine average bubble sizes. For each image the laser focus was moved to a new position on the targeted axon, so that residual bubbles from a previous measurement would not affect the image.
Contrast and brightness were manually adjusted for all figures using the levels menu in Adobe Photoshop. Areas of GFP::RAB-3 accumulations quantified in Fig. 7 were measured in ImageJ using images of cut and uncut axons captured with the same exposure time and camera gain. Significance was tested using the student t-test with unequal variance.
3. Results
3.1. Nanosecond laser axotomy
In Fig. 1 we present results of axonal microsurgery using nanosecond laser pulses in C. elegans larvae. A larva in which the PLM neuron was targeted is shown before ((a),(c),(e)) and after ((b),(d),(f)) the surgery. The damaged region or neuronal cut is shown magnified in the inset in (b), (d) and (f). In all cases we observe a break in the axon at the site of laser focus. These particular axotomies have been chosen to exhibit the range of damage that the laser microbeam produced. For instance, in 1(b) and 1(d) we could observe clear breaks in the axon extending over 3–5 µm. In 1(f) however the visible break is barely discernable being <1 µm and the only other evidence for axonal cutting is the displacement of the proximal end due to cavitation forces generated by plasma formation. The average extent of damage in these axons was measured to be 1.95±1.81 µm (n=8). The large variation in extent of damage occurs due to inclusion of all cut axons including those that move significantly due to the cavitation forces.
Fig. 1. Axonal microsurgery of PLM neurons in C. elegans larva with nanosecond laser pulses. The axon is shown prior to (a, c, e) and immediately after surgery (b,d,f). Insets in (b,d,f) show magnified regions around the cut site. Clear breaks can be seen in the axon (b,d). In (f) the extent of damage is extremely small and is only noticeable as a dark band. The proximal end of the axon was also displaced due to the cavitation forces generated during the laser microsurgery. Pulse energy was 0.8 µJ with the laser being operated in single shot mode.
We found that a pulse energy of 0.8 µJ was sufficient to cut axons in C. elegans with minimal collateral damage. The laser was used in single shot mode and the operator manually fired the laser till the axon was cut. The number of pulses varied between individuals but was always <20. In a majority of surgeries one laser pulse was sufficient to effect a complete axonal cut, with the position of the laser spot being the major determinant of success. In a few cases we also observed partial cutting of the axon effected by the first few pulses, following which a succeeding laser pulse would fully cut it. Physical movement of the cut ends due to cavitation forces could also be observed. Thus, the site of laser focus and physical movement influenced the extent of damage observed immediately after the axotomy. Successful axonal cutting resulted in a retraction of the two ends away from the site of laser irradiation. This retraction was observable within 1 hr of microsurgery and has also been noted by other researchers [3, 13]. We measured the extent of damage in C. elegans to be 4.65±1.85 µm (n=17) at 1 hr after laser microsurgery. To determine collateral damage during laser axotomy, we viewed animals 4, 6 and 8 hours post-surgery under DIC and noted scarring in only 8% of the animals (n=24). Animals undergoing laser axotomy showed viability levels >95% after removal of anesthetic and continued to undergo development and become fertile adults. Viability in control animals (sham surgery treated) was 100% (n=20).
Post surgery, C. elegans larvae were maintained at 16°C and followed in the 1–3 hr time period to assess the time required for accumulation of vesicles and determine cellular out-comes after damage. In Fig. 2(a,b,c) axons in 3 individual larvae are shown at different times after surgery. One hour post surgery (Fig. 2(a)) we could observe robust accumulations of GFP::RAB-3, a known synaptic vesicle marker. The accumulation of GFP::RAB-3 signal results from a block at the distal and proximal cut ends. At later time points we do not see any accumulations. Instead we observed regenerative processes similar to those reported earlier for fs laser surgery in C. elegans. At 3 hours, we could observe close proximity of the distal and proximal ends (Fig. 2(b), inset). This maybe an attempted reconnection of the cut ends and was observed in 46% of the animals (n=13). Regrowth of a new axon from the proximal end occurred in 54% of the animals (n=13). At 9 hours, (Fig. 2(c)), regrowth was seen in all individuals (n=4). Similar putative reconnection events have been observed by others [3, 13].
Fig. 2. Manifestions of axonal recovery at different time-points in C. elegans. In (a) vesicle accumulations are seen at 1 hour, in (b) axonal reconnection is seen at 3 hours and in (c) axonal regrowth at 9 hours. Inset in (a) shows a magnified view of the accumulation zone and inset in (b) shows the magnified reconnection zone. Arrow in (c) indicates the growing proximal end, while arrowhead indicates the distal stump.
3.2. Time-resolved imaging in vivo
To determine the dynamics of the ablation process we conducted time-resolved imaging in vivo. Focused nanosecond laser pulses beyond an intensity threshold (Ith>1010 W/cm2) result in plasma formation during laser microsurgery [14]. The high temperature plasma generates a cavitation bubble whose expansion and collapse leads to collateral damage around the site of the laser irradiation. Our time-resolved imaging could successfully capture these fast bubble dynamics in C. elegans larvae (Fig. 3). This is the first time that the damage process has been visualized at high resolution during laser microsurgery in vivo. In C. elegans, we observed a weak plasma with a lifetime <15 ns. After the plasma was extinguished, the bubble underwent a rapid expansion and reached its maximum size (Rmax) within 20 ns. The bubble collapse occurred within 30 ns, following which recoil could be observed for at least two cycles. In all cases the dynamics were complete within 90 ns i.e. no bubbles could be observed beyond this time at the 0.5 µm spatial resolution of our system. These time-scales are significantly shorter than those reported by other studies [7]. We decided to confirm the short time-scales of bubble collapse in vivo, by conducting time-resolved imaging in Drosophila larvae.
Fig. 3. Time-resolved imaging of cavitation dynamics in C. elegans. The bubble reaches its maximum size within 20 ns of laser irradiation and the first collapse occurs within ≈30 ns. At later times we observe a rebound and collapse. Arrows indicate the cavitation bubble in different frames.
To measure cavitation dynamics in Drosophila, we first developed an axotomy procedure to closely parallel our experiments in C. elegans. In Fig. 4 we demonstrate axotomies in 1st instar Drosophila larvae. We targeted axon bundles about 40 to 50 µm away from the ventral ganglion. In 4(a), two bundles are shown prior to surgery, while in 4(b) we show successful cutting of the 1st bundle followed in 4(c) by cutting the 2nd one. Since we used intact larvae with thick cuticles, a higher pulse energy of 1.2 µJ was necessary to conduct axotomies. In Drosophila, 60% of axotomized animals (n=25) underwent normal development and became flies. Viability in sham surgery treated animals was 100% (n=27).
Fig. 4. Neuronal axotomy in Drosophila 1st larvae. (a) Two motorneuron bundles prior to axotomy and (b,c) after successive cutting of each bundle. Arrows in (b) and (c) point to the first and second axotomies respectively. Ventral ganglion is at lower left in each image (marked with *). Pulse energy was 1.2 µJ and the laser was operated in single shot mode.
In Fig. 5 we show results of time-resolved imaging of cavitation dynamics during motorneuron surgeries in Drosophila. The bubble dynamics were more variable in Drosophila as compared to C. elegans. Within 30 ns the bubble expanded rapidly to an average size of 3.5 µm. The bubble expansion reached a maximum at 112 ns with an average bubble radius of 4.5 µm. A bubble collapse event was observed at 215 ns and all dynamics were complete within 800 ns. The bubble size was measured at each time-delay and the average and standard deviation is plotted in Fig. 6 for C. elegans and Drosophila (n≥3 for each time-point). The maximum bubble radius (Rmax) was 3.4 µm in C. elegans (Fig. 6a) and 4.5 µm in Drosophila (Fig. 6b). We also observed bubble sizes equal to Rmax in the time range of 30 to 40 ns in Drosophila. This indicated that similar to C. elegans, the primary collapse could be within 50 ns. However, it was difficult to conclusively determine the bubble parameters in this time range due to large variability.
Fig. 5. Time-resolved imaging of cavitation dynamics in Drosophila. The Rmax is seen at 112 ns with collapse at 215 ns.
Fig. 6. Cavitation bubble dynamics in (a) C. elegans and (b) Drosophila. Bubble dynamics are complete within 100 ns in C. elegans and 800 ns in Drosophila.
3.3. Transport assay
Our motivation for conducting neuronal axotomy in C. elegans was to develop an assay for studying transport in vivo. By making use of the laser microbeam we could produce a permanent axonal ‘ligature’ in a smaller genetic model organism with its associated ease of handling and genetic manipulation. In the transgenic C. elegans strains that we used, the GFP::RAB-3 and SNB-1::GFP fusion proteins have been shown to be localized to synaptic vesicles. Blockages to transport can be easily viewed as fluorescent accumulations at the proximal and distal ends of the cut axon. The assay involved carrying out laser microsurgery in the animal and imaging the cut axon at 1 hr post-surgery based on our results of axonal recovery (Fig. 2). These images were analyzed in ImageJ to determine areas of fluorescent puncta. Results of the assay are shown in Fig. 7. In Fig. 7(a), we show that the mean area of GFP::RAB-3 accumulation at the proximal and distal ends of the cut (1.08 µm2) (n=15 puncta or individuals) was 6 times larger than in uncut axons (0.18 µm2) (n=130 puncta, 10 individuals). We also found that laser microsurgery did not significantly change the population of vesicles in regions away from the cut (0.24 µm2) compared to uncut animals. The area-wise distribution of fluorescent accumulations with distance is presented in Fig. 7(b). We observe that the area of accumulation near the cut site is much larger compared with GFP::RAB-3 accumulations further away from the cut site. Within the first micron of the retracted distal and proximal ends of the axon, greater than 40% GFP::RAB-3 accumulations are larger than 1 µm2 and more than 20% of the accumulations are between 0.5–1.0 µm2. In uncut axons (Fig. 7(b), inset), fluorescent accumulations with area ≥1 µm2 are not observed while <20% of the puncta are between 0.5–1.0 µm2 in size and the rest are smaller than 0.5 µm2. A second vesicular marker, synaptobrevin::GFP (SNB-1::GFP), was used to confirm the validity of the assay. Areas of SNB-1::GFP accumulations 1 hr post surgery at the proximal end (0.82±0.15 µm2) and distal end (0.95±0.17 µm2) of the cut (n=12 puncta or individuals) were on average 3 times larger than in uncut axons (0.27±0.05 µm2) (n=148 puncta, 24 individuals) (p<10-4). We also axotomized zdIs5 a strain with cytoplasmic GFP marking the axon. Areas of fluorescent puncta in zdIs5 1 hr post surgery were 0.44±0.08 µm2 away from the cut site, 0.25±0.12 µm2 at the proximal end and 0.46±0.16 µm2 at the distal end (n=9). Thus in zdIs5 there was no statistical difference between the size of fluorescent puncta at the cut site and away from it.
Fig. 7. (a) Mean areas of GFP::RAB-3 puncta in axons of uncut and axotomized jsIs821. Axons were cut between 10–20 µm from the cell body. The area in (µm2) of GFP::RAB-3 accumulations at the distal and proximal ends of axotomized PLM axons 1 hour after the cut (1.08) is 6 times larger than in uncut axons (0.18) (p<10-4). In cut axons, the mean area at regions away from the cut (0.24) is not statistically different from the uncut axon. Inset shows an image of a cut axon used for quantification, with arrows pointing to accumulations at the proximal and distal ends. * points to cell body. (b) Size distribution of fluorescent accumulations along the axon length. Within 1 µm of the retracted distal and proximal ends of the axon, ≥40% GFP::RAB-3 accumulations are larger than 1 µm2 and more than 20% of the accumulations are between 0.5–1.0 µm2. In uncut axons (Fig. 7(b), inset), fluorescent accumulations with area ≥1 µm2 are not observed while <20% of the puncta are between 0.5–1.0 µm2 in size and the rest are smaller than 0.5 µm2.
4. Discussion
Femtosecond laser pulses produce extremely localized damage zones as demonstrated previously in single cells [15, 16, 17] which makes them attractive for microsurgery in vivo. However, these laser systems still remain outside the reach of most researchers due to their cost and complex operation. Nanosecond (ns) Nd:YAG pulsed lasers are the workhorses of the bio-medical optics community and their simple operation and robustness combined with relatively cheap cost make them an attractive option to fs laser systems. Laser microsurgery with nanosecond laser pulses has long been used in cell biology with exquisite precision. (See Ref. [18] for a recent review). Nanosecond laser pulses at λ=355 nm have also been used in vivo during removal of the spindle in single-cell C. elegans embyros [19] and ablating cell junctions in the aminoserosa in Drosophila embryos [20]. In the present study we extend the range of microsurgery procedures in vivo by demonstrating neuronal axotomy in C. elegans and Drosophila larvae. Laser microsurgery relies on plasma formation, an intensity based phenomena for inducing damage. The subsequent cavitation dynamics that are associated with plasma formation are dependant upon the total energy provided to create the plasma. By using the 355 nm wavelength we take advantage of higher photon energy to effect a reduction in the pulse energy necessary for plasma formation and hence a reduction in the associated cavitation dynamics. We observe a further reduction in plasma threshold in vivo due to linear absorption by endogenous molecules. Hutson and Ma [7] have hypothesized that the absorber might be NADH, since it is present in large quantities in developing organisms and has an absorption peak at λ=340 nm. Due to the reduction in threshold it was possible to successfully cut axons in C. elegans and Drosophila with limited collateral damage and low pulse numbers. The extent of damage that we observed in C. elegans after microsurgery (Fig. 1) was in a similar range to that noted previously using femtosecond lasers. Specifically, Wu and co-workers report a gap length of 5 µm (Fig. 5d, supplementary information, Ref. [3]) and Bourgeois and Ben-Yakar report a damage range from <0.5 µm (4 nJ/pulse, 100 pulses) to 5.5 µm (12 nJ/pulse, 1000 pulses, Fig. 8, Ref. [21]). Our observed damage length of 1.95±1.81 µm is within the limits provided by these data sets.
In C. elegans at 0.8 µJ pulse energy we noted internal wounding or scarring at the cut site in 2 out of 24 individuals (8 %). Animals with wounds developed into normal egg laying adults. Using a higher pulse energy (1.5 µJ) larger wounds could be observed in all animals (n=26). Wu and co-workers also noted scarring in animals when using 4 ns pulses at λ=440 nm to conduct axotomies in C. elegans, while no scarring resulted during fs laser surgery [3]. This indicates that while ns laser pulses produce axonal damage zones similar to fs laser pulses, in a small number of animals collateral damage is larger with ns pulses.
Axotomy in Drosophila did not prove any different than in C. elegans, even though Drosophila larvae are significantly larger and structures of interest are at least 5–10 µm from the coverslip (Fig. 4). Instead, we found that a small increase in pulse energy was sufficient to perform microsurgies without removing the cuticle. The larger pulse energy caused a higher extent of damage in Drosophila motorneuron bundles (≈10 µm). We also noticed larger physical retraction of the axonal bundle upon cutting and this too contributed to the larger cut lengths. Previously, in experiments studying nerve-muscle interactions, Fernandes and Keshishian had used nanosecond laser pulses (λ=440 nm, τp=3 ns, 30 µJ/pulse, 30 to 60 pulses at 15 Hz) for ablation of Drosophila larval motorneurons [22]. A comparison with this data set is not possible since the authors have not commented on the extent of damage caused by higher pulse energy used during laser ablation, or on viability and development of larvae after ablation. In Drosophila we did not conduct long term studies to determine the occurrence of wounds due to laser irradiation. It is possible that damage due to cavitation effects although unseen could contribute to changes in development programs resulting in the observed lethality prior to adult-hood. However, grossly the development appears to be normal in 60% of larvae and we believe this could be increased by careful selection of laser parameters.
Following microsurgery, axonal recovery was studied by imaging C. elegans larvae at different time-points (Fig. 2). We observed regenerative processes beginning at 3 hours after laser microsurgery in the jsIs821 strain. The time-scale for regeneration was 3–9 hours in our experiments. Interestingly, we observed shorter axonal regeneration time-scales as compared to the 6–24 hour recovery period previously reported for anesthetized animals [3, 13]. The previous data were generated using the zdIs5 or the muIs32 strains that have high levels of soluble GFP marking mechanosensory neurons. Preliminary evidence from our lab using zdIs5 also gave us longer recovery times than the 3–9 hour period observed for jsIs821. Further, recovery times for usIs25 (data not shown), a strain with lower cytoplasmic GFP expression were faster than for zdIs5. This suggests that altered recovery times are due to inherent differences in the strains resulting from varied levels of GFP expression and unlikely to be due to differences in the mechanism of damage or its extent produced by ns as compared to fs pulses.
Time-resolved imaging of cavitation dynamics in vivo provided a clear exposition of the spatio-temporal scales of the damage process in our studies (Figs. 3 and 5). Bubble dynamics proved reproducible in C. elegans with Rmax=3.5 µm and collapse within 100 ns. In Drosophila, bubble dynamics were variable, due to the larger animal size and possibly different microenvironments at the site of irradiation. Rmax in Drosophila was 4.5 µm with a collapse time of 215 ns. We also observed bubbles equivalent in size to Rmax in the time range of 30 to 40 ns. However, these were rare events and further studies are needed to conclusively determine the complete bubble dynamics. Considering the large cavitation bubble size it was surprising to note the limited extent of damage that we observed in these animals. It has been noted in vitro that cells can withstand cavitation generated shear stresses several orders of magnitude higher than physiological values due to short time-scales of exposure [23]. This fact coupled with the damping of the bubble in vivo could account for the reduced collateral damage in the form of scarring that was observed at the irradiation site. The time-resolved imaging also highlights the regenerative capabilities of these organisms after facing large mechanical stresses.
Recently, Hutson and Ma also studied the cavitation dynamics during laser microsurgery of Drosophila embryos using nanosecond pulses at 355 nm [7]. The dynamics were measured by recording pressure transients associated with shock wave propagation and bubble collapse. The time of bubble collapse was used to estimate Rmax using the Rayleigh equation. At a pulse energy of 1 µJ, the authors observed an oscillation time (Tosc) slightly greater than 2 µs with an estimated Rmax of ≈11 µm (Fig. 2 from Ref. [7]). In our experiments in C. elegans at 0.8 µJ, we imaged Tcol(=Tosc) to be 30 ns with a Rmax of 3.4 µm. To confirm the time-scales of bubble dynamics in our experiments we also conducted time-resolved imaging in the 100–1000 ns timescale in C.elegans and failed to observe any bubbles. Further, we undertook time-resolved studies in Drosophila (Fig. 5). Due to the variability in bubble sizes at different time points it was difficult to assign an exact collapse time. We only note that all dynamics were complete by 800 ns, indicating that even in Drosophila cavitation was strongly damped. At present we do not have an explanation for the difference in collapse times in our data set and that reported in Ref. [7]. One possible reason could be the dissimilar sample geometries in the two experiments. In our case the bubble expansion occurs in a 3D viscoelastic medium so the bubble experiences a strong damping force from all sides. Earlier work has also reported the drastic reduction in bubble size during expansion in tissue [24], while recent work from our lab has shown that the bubble size is reduced by 54–59% for laser irradiation of rat corneas as compared to the laser-induced lysis of adherent cells [12]. In the case of Hutson and Ma, the site of laser irradiation is the amnioserosa, at the surface of the Drosophila embryo. Therefore, the bubble experiences damping only at one surface, possibly causing it to expand into the medium and collapse at longer time-scales. The above discussion highlights that there are still issues related to cavitation damping in vivo that need to be resolved. While our time-resolved images provide a first glimpse at laser microsurgery in vivo, detailed studies using supplementary methods such as interferometry or hydrophone measurements are necessary.
It is also of interest to compare cavitation dynamics for ns and fs laser microsurgery. Recently, Vogel and co-workers reported cavitation bubble sizes with an accuracy of ±10 nm for laser-induced plasma formation in water with 340 fs laser pulses at λ=347, 520 and 1040 nm [25]. The authors observed that at threshold, the cavitation bubble size was smaller than the diffraction limited focus diameter. Specifically, at λ=347 nm, Rmax increased slowly with pulse energy and ranged from 0.19 µm at 3.95 nJ to 1.8 µm at 5.13 nJ. These bubble sizes would be considerably smaller in vivo due to strong damping. The bubbles are smaller by a factor of 2 to 20 as compared to the bubble sizes generated by ns laser pulses in C. elegans. It is thus surprising to note that axotomy damage zones are comparable for ns and fs surgeries and collateral damage in the form of scarring could be observed in only a small percentage of animals.
Finally, we demonstrated the feasibility of neuronal axotomy for studying transport in vivo using fluorescent proteins targeted to vesicles. Mean areas of fluorescent accumulations in C. elegans at the proximal and distal ends in cut axons were 6 times larger for GFP::RAB-3 and at least 3 times larger for SNB-1::GFP as compared to uncut axons. This indicated that vesicles were transported to the site of the cut and remained trapped there. Such fluorescent accumulations at the same time points were not observed for the zdIs5 strain since GFP is a cytoplasmic marker. We only observed transient accumulation of GFP::RAB-3 (Fig. 2(a) and 7(a),inset). This suggests that accumulation is due to a physical block to transport and unlikely to be due to the formation of a growth cone. The growth cone should exist during the entire regenerative process that has been reported to occur during a 6 to over 24 hour period. However no accumulation of GFP::RAB-3 is seen at 3 hours or later in axotomized jsIs821 when regenerative axon extensions are ongoing. The assay we have developed allows the study of several aspects of transport such as: (1) determining whether a given protein is transported in both directions, i.e. retrograde (away from the synapse) and anterograde (to the synapse), (2) assessing the role of specific transport mutants on longer time scales (1–2 hours, as opposed to a few minutes) and (3) studying the dynamics of transport post complete or partial axotomy as pertaining to regenerative processes. Some of these advantages apply equally to Caenorhabditis elegans and Drosophila melanogaster.
5. Conclusion
We report the successful use of nanosecond laser pulses at λ=355 nm to conduct neuronal axotomy in C. elegans and Drosophila larvae. Axotomy in Drosophila could be used for studies comparing regeneration processes across model systems. A pulse energy in the range of 0.8–1.2 µJ and <20 pulses were sufficient to cut axons with high viability in both organisms. Using time-resolved imaging we ascertained that damage occurs due to laser induced plasma formation. The bubble expansion was significantly damped in C. elegans with bubble collapse occurring within 100 ns. In Drosophila, while cavitation was variable, we observed that all dynamics were complete within 800 ns. The bubble damping due to the viscoelastic medium contributed to the reduced collateral damage that we observed. We also demonstrated the feasibility of using neuronal axotomy as a means of creating a transport block in vivo and using vesicular markers to characterize these blockages. We believe the results of our study opens up this field to a wider community of researchers in neuroscience.
Acknowledgement
The authors acknowledge Nivedita Chatterjee for initiating this work. GNR is supported by a Council of Scientific and Industrial Research (CSIR) fellowship. SSK is supported by a Department of Science and Technology (DST) grant no. SR/SO/AS-67/2006 to SPK. KRR acknowledges support from the National Institutes of Health (USA) via the Fogarty International Research Collaboration Award. We thank CGC for the zdIs5 strain.
References and links
1. M. F. Yanik, H. Cinar, H. N. Cinar, A. D. Chisholm, Y. Jin, and A. Ben-Yakar, “Functional regeneration after laser axotomy,” Nature 432, 882 (2004). [CrossRef]
2. S. H. Chung, D. A. Clark, C. V. Gabel, E. Mazur, and A. D. Samuel,“The role of the AFD neuron in C. elegans thermotaxis analyzed using femtosecond laser ablation,” BMC Neuroscience 7, DOI:10.1186/1471–2202–7–30 (2006). [CrossRef] [PubMed]
3. Z. Wu, A. Ghosh-Roy, M. F. Yanik, J. Z. Zhang, Y. Jin, and A. D. Chisholm, “Caenorhabditis elegans neuronal regeneration is influenced by life stage, ephrin signaling and synaptic branching,” PNAS 104, 15132–15137 (2007). [CrossRef] [PubMed]
4. P. Weiss and H. B. Hiscoe, “Experiments on the mechanism of nerve growth,” J. Exp. Zool. 107315–395 (1948). [CrossRef] [PubMed]
5. R. V. Barkus, O. Klyachko, D. Horiuchi, B. J. Dickson, and W. M. Saxton, “Identification of an Axonal Kinesin-3 Motor for Fast Anterograde Vesicle Transport that Facilitates Retrograde Transport of Neuropeptides,” Mol. Biol. Cell 19, 274–283 (2008). [CrossRef]
6. H. Erez, G. Malkinson, M. Prager-Khoutorsky, C. I. De Zeeuw, C. C. Hoogenraad, and M. E. Spira, “Formation of microtubule-based traps controls the sorting and concentration of vesicles to restricted sites of regenerating neurons after axotomy,” J. Cell Biol. 176, 497–507 (2007). [CrossRef] [PubMed]
7. M. S. Hutson and X. Ma, “Plasma and cavitation dynamics during pulsed laser microsurgery in vivo,” Phys. Rev. Lett. 99, 158104 (2007). [CrossRef] [PubMed]
8. T. R. Mahoney, Q. Liu, T. Itoh, S. Luo, G. Hadwiger, R. Vincent, Z-W. Wang, M. Fukuda, and M. L. Nonet, “Regulation of synaptic transmission by RAB-3 and RAB-27 in Caenorhabditis elegans,” Mol. Biol. Cell 17, 2617–2625 (2006). [CrossRef] [PubMed]
9. M. L. Nonet, “Visualization of synaptic specializations in live C. elegans with synaptic vesicle protein-GFP fusions,” J. Neurosci. Methods 89, 33–40 (1999). [CrossRef] [PubMed]
10. T. Lee and L. Luo, “Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis,” Neuron 22, 451–461 (1999). [CrossRef] [PubMed]
11. A. Mahr and H. Aberle, “The expression pattern of the Drosophila vesicular glutamate transporter: A marker protein for motoneurons and glutamatergic centers in the brain,” Gene Expr. Patterns 6, 299–309 (2006). [CrossRef]
12. A. V. Cherian and K. R. Rau, “Pulsed laser-induced damage in rat corneas: time-resolved imaging of physical effects and acute biological response,” J. Biomed. Opt. 13, 024009 (2008). [CrossRef] [PubMed]
13. M. F. Yanik, H. Cinar, H. N. Cinar, A. Gibby, A. D. Chisholm, Y. Jin, and A. Ben-Yakar, “Nerve regeneration in Caenorhabditis elegans after femtosecond laser axotomy,” IEEE J. Quantum Electron. 12, 1283–1291 (2006). [CrossRef]
14. V. Venugopalan, A. Guerra III, K. Nahen, and A. Vogel, “Role of Laser-Induced Plasma Formation in Pulsed Cellular Microsurgery and Micromanipulation,” Phys. Rev. Lett. 88, 078103 (2002). [CrossRef] [PubMed]
15. U.K. Tirlapur and K. König, “Targeted transfection by femtosecond laser,” Nature 418, 290–291 (2002). [CrossRef] [PubMed]
16. W. Watanabe and N. Arakawa, “Femtosecond laser disruption of subcellular organelles in a living cell,” Opt. Express 12, 4203–4213 (2004). [CrossRef] [PubMed]
17. A. Heisterkamp, I. Zaharieva Maxwell, E. Mazur, J. M. Underwood, J. A. Nickerson, S. Kumar, and D. E. Ingber, “Pulse energy dependence of subcellular dissection by femtosecond laser pulses,” Opt. Express 13, 3690–3696 (2005). [CrossRef] [PubMed]
18. V. Magidson, J. Lončarek, P. Hergert, C. L. Reider, and A. Khodjakov,“Laser Microsurgery in the GFP Era: A Cell Biologist’s Perspective,” Methods Cell Biol. 82, 239–266, (2007). [PubMed]
19. W. Grill, P. Gönczy, E. H. K. Stelzer, and A. A. Hyman “Polarity controls forces governing asymmetric spindle positioning in the Caenorhabditis elegans embryo,” Nature 409, 630–633 (2001). [CrossRef] [PubMed]
20. M.S. Hutson, Y. Tokutake, M.-S. Chang, J.W. Bloor, S. Venakides, D.P. Kiehart, and G.S. Edwards “Forces for morphogenesis investigated with laser-microsurgery and quantitative modeling,” Science 300, 145–149 (2003). [CrossRef] [PubMed]
21. F. Bourgeois and A. Ben-Yakar, “Femtosecond laser nanoaxotomy properties and their effect on axonal recovery in C. elegans,” Opt. Express 15, 8521–8531 (2007). [CrossRef] [PubMed]
22. J. J. Fernandes and H. Keshishian “Nerve-muscle interactions during flight muscle development in Drosophila,” Development , 125, 1769–1779 (1998). [PubMed]
23. K. R. Rau, P. A. Quinto-Su, A. Hellman, and V. Venugopalan, “Pulsed laser microbeam-induced cell lysis: time-resolved imaging and analysis of hydrodynamic effects,” Biophys. J. 91, 317–329 (2006). [CrossRef] [PubMed]
24. A. Vogel, M. R. C. Capon, M. N. Asiyo-Vogel, and R. Birngruber, “Intraocular photodisruption with picosecond and nanosecond laser pulses: Tissue effects in cornea, lens, and retina,” Invest. Ophthalmol. Visual Sci. 35, 3032–3044 (1994).
25. A. Vogel, N. Linz, S. Freidank, and G. Paltauf, “Femtosecond-Laser-Induced Nanocavitation in Water: Implications for Optical Breakdown Threshold and Cell Surgery,” Phys. Rev. Lett. 100, 038102 (2008). [CrossRef] [PubMed]
