OSA's Digital Library

Virtual Journal for Biomedical Optics

Virtual Journal for Biomedical Optics

| EXPLORING THE INTERFACE OF LIGHT AND BIOMEDICINE

  • Editors: Andrew Dunn and Anthony Durkin
  • Vol. 8, Iss. 5 — Jun. 6, 2013
« Show journal navigation

Visualization of brain circuits using two-photon fluorescence micro-optical sectioning tomography

Ting Zheng, Zhongqing Yang, Anan Li, Xiaohua Lv, Zhenqiao Zhou, Xiaojun Wang, Xiaoli Qi, Shiwei Li, Qingming Luo, Hui Gong, and Shaoqun Zeng  »View Author Affiliations


Optics Express, Vol. 21, Issue 8, pp. 9839-9850 (2013)
http://dx.doi.org/10.1364/OE.21.009839


View Full Text Article

Acrobat PDF (5343 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

Neural circuits are fundamental for brain functions. However, obtaining long range continuous projections of neurons in the entire brain is still challenging. Here a two-photon fluorescence micro-optical sectioning tomography (2p-fMOST) method is developed for high-throughput, high-resolution visualization of the brain circuits. Two-photon imaging technology is used to obtain high resolution, and acoustical optical deflector (AOD), an inertia-free beam scanner is used to realize fast and prolonged stable imaging. The combination of these techniques with imaging and then sectioning method of a plastic-embedded mouse brain facilitated the acquisition of a three-dimensional data set of a fluorescent mouse brain with a resolution adequate to resolve the spines. In addition, the brain circuit tracing ability is showed by several neurons projecting across different brain regions. Besides brain imaging, 2p-fMOST could be used in many studies that requires sub-micro resolution or micro resolution imaging of a large sample.

© 2013 OSA

1. Introduction

In a cellular phone, the information flow is present in real functional connections. Similarly, in our brains, information is transmitted within and between neural circuits through functional neuronal connections based on axonal projections. Revealing the axonal projection of a single neuron is essential to understand the brain functions, in terms of health and disease [1

1. V. Marx, “High-throughput anatomy: Charting the brain’s networks,” Nature 490(7419), 293–298 (2012). [CrossRef] [PubMed]

]. The axon, also known as the nerve fiber, is less than 1 µm in diameter, whereas its projection length may exceed 1 cm [2

2. J. W. Lichtman and W. Denk, “The big and the small: challenges of imaging the brain’s circuits,” Science 334(6056), 618–623 (2011). [CrossRef] [PubMed]

]. Consequently, tracing a single axon within the centimeter range in three dimensions (3D) is challenging.

Optical imaging has recently undergone much development in neural circuit research with a good balance between resolution and large volume [12

12. P. J. Keller, A. D. Schmidt, A. Santella, K. Khairy, Z. Bao, J. Wittbrodt, and E. H. Stelzer, “Fast, high-contrast imaging of animal development with scanned light sheet-based structured-illumination microscopy,” Nat. Methods 7(8), 637–642 (2010). [CrossRef] [PubMed]

21

21. H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage 74, 87–98 (2013), doi:. [CrossRef] [PubMed]

]. The micro-optical sectioning tomography (MOST) method [17

17. A. Li, H. Gong, B. Zhang, Q. Wang, C. Yan, J. Wu, Q. Liu, S. Zeng, and Q. Luo, “Micro-optical sectioning tomography to obtain a high-resolution atlas of the mouse brain,” Science 330(6009), 1404–1408 (2010). [CrossRef] [PubMed]

] has significantly improved the imaging precision for large sample and obtained “the most detailed” [18

18. K. Minogue, “Neuroscience. China’s brain mappers zoom in on neural connections,” Science 330(6005), 747 (2010). [CrossRef] [PubMed]

] sub-micron resolution atlas of the entire mouse brain stained with Golgi method. MOST system did not demonstrated fluorescence brain imaging. Considering the importance of fluorescence imaging, the serial two-photon tomography (STP) [19

19. T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods 9(3), 255–258 (2012). [CrossRef] [PubMed]

] achieves high-throughput fluorescence imaging of mouse brain by imaging an optical section in every 50-micron-thickness layer of tissue. With this spatially coarse sampling rate, no continuous neuronal projections are demonstrated accordingly [19

19. T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods 9(3), 255–258 (2012). [CrossRef] [PubMed]

]. Recently, confocal light sheet microscopy (CLSM) [20

20. L. Silvestri, A. Bria, L. Sacconi, G. Iannello, and F. S. Pavone, “Confocal light sheet microscopy: micron-scale neuroanatomy of the entire mouse brain,” Opt. Express 20(18), 20582–20598 (2012). [CrossRef] [PubMed]

] combines the advantages of light sheet illumination and confocal slit detection to increase the image contrast, exhibiting its ability to obtain a neuron anatomy of the entire mouse brain with spatial resolution of 2 µm in lateral and 9 µm in axial. More recently, with the help of confocal detection to remove the background in block imaging, MOST system realizes fluorescence imaging and demonstrated continuous tracing of neuronal circuits with fluorescent mouse [21

21. H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage 74, 87–98 (2013), doi:. [CrossRef] [PubMed]

]. However, methods in obtaining long-range continuous projections of neurons of the entire brain are still challenging.

Here, we developed a two-photon fluorescence micro-optical sectioning tomography (2p-fMOST) method to trace the axonal projections in the whole brain in a more convenient way. We used two-photon imaging technology and high-numerical aperture (NA) objective lens to obtain high spatial resolution, especially axial resolution. To obtain high throughput, we used an acoustical optical deflector (AOD), which is inertia-free and with non-mechanical working mechanism, to realize fast and prolonged stable imaging. Combined with imaging then sectioning of a plastic embedded mouse brain, we acquired a three-dimensional data set of a Thyl-enhanced green fluorescent protein (Thy1-EGFP) transgenic mouse brain with a resolution adequate to resolve the spines. Furthermore, a 2 µm z-spacing offers a better discrimination ability of closely distributed fibers. Based on these features, we traced 8 neural long-distance projections distributed in different brain regions. Our results established that 2p-fMOST is a powerful tool in neuron connection and brain function research as well as in any study that requires sub-micro resolution or micro resolution imaging of a large sample.

2. Setup and principles

2.1 2p-fMOST

The schematic diagram of the 2p-fMOST system is shown in Fig. 1(a)
Fig. 1 Optical scheme of 2P-fMOST system. The whole 2p-MOST system was composed of (a) dispersion compensation, (b) fast scanning, (c) microscope, and (d) motion and sectioning modules. The relay optics consists of scan lens (SL, 150 mm) and tube lens (TL, 180 mm) images of the AOD on the back aperture of the objective. (e) Lateral normal fluorescence intensity distribution of the bead, the full width at half maximum (FWHM) value is approximately 0.45 µm. (f) Axial normal fluorescence intensity distribution of the bead, FWHM value is approximately 1.68 μm.
. The 2p-fMOST system consisted of four modules, namely, the dispersion compensation, fast scan, microscopy, and motion and sectioning modules, with the fast scan module as the most important part. This module consists of two elements, an AOD (DTSXY-400-720-920, AA Opto-electronic Inc) and a custom designed cylinder lens. Fast scan in the y-dimension was accomplished by the AOD, and the cylinder lens was used to compensate the astigmatism caused by the AOD [21

21. H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage 74, 87–98 (2013), doi:. [CrossRef] [PubMed]

]. To compensate for the spatial and temporal dispersion caused by the AOD, the dispersion compensation module was placed in front of the fast scan module. Scanned laser entered the microscope and was focused by a high NA oil objective lens (UPlanFLN 40 × oil, NA1.3, Olympus). Fluorescence was detected by a head-on photomultiplier tube (1924A, Hamamatsu) after a dichroic mirror (FF670, Semrock). In addition, the excitation light was blocked by a barrier filter (FF01-680-SP, Semrock) before detection. The motion and sectioning module played an important role in both imaging and sectioning, detailed description of motion can be seen in [17

17. A. Li, H. Gong, B. Zhang, Q. Wang, C. Yan, J. Wu, Q. Liu, S. Zeng, and Q. Luo, “Micro-optical sectioning tomography to obtain a high-resolution atlas of the mouse brain,” Science 330(6009), 1404–1408 (2010). [CrossRef] [PubMed]

]. A sink filled with oil was mounted on the 3D precision motion stage (X axis: ABL20030, Y axis: ALS130, Z axis: AVL125, Aerotech). Plastic embedded mouse brain was fixed in the sink and moved along with the stage. In a single imaging and sectioning process, the specimen was moved along with the stage in 3D first for imaging. Then, the specimen was driven to a tungsten knife (Triangular Tungsten Carbide, DDK) fixed on the marble base of the stage. Consequently, the surface layers that have already been imaged would be cut off along with the motion of the stage. We call this process that includes one cycle of imaging and sectioning as a single imaging and sectioning process. In the application of axon tracing in the whole brain, the thickness of a single imaging and sectioning process was set to 30 µm, and this 30 µm thick block consisted of 15 optical sections with a 2 µm z-spacing when imaging. We also normally set a pixel residual time of 0.54 μs for a better balance between signal-to-noise ratio (SNR) and throughput. Simultaneously, the pixel size was set to 0.5 µm. However, different settings could be changed flexibly for different requirements.

The light source was a high power Ti: sapphire femtosecond pulsed laser (Chameleon UItra II, Coherent). The AOD was driven by a direct digital synthesizer driver (DDSPA-B415b-0, 10-400 MHz and AMPA-B-34, A&A, France) with a RF power of 2W. Imaging and sectioning control were realized by custom written LabVIEW software, the control signal of the AOD mainly relying on a multifunction data acquisition (PXI6363, National Instrument). Data acquisition was accomplished by a 60 MHz 8-channel digitizer (PXI5105, National Instrument).The whole imaging and sectioning process could be fully automated as long as the parameters were set properly.

Spatial resolution of the system was tested with 190 nm fluorescence beads (Bangs Laboratories, Inc) under a 40 × /NA1.3 oil objective lens, and fluorescence images were acquired by scanning in 3D. Scanning in x- and z-directions were accomplished by the motion stage, the scanning in the y-direction was accomplished by AOD. To measure the resolution, the image pixel size was set at 50 nm, and the step size in z-axis was set at 200 nm. In this experiment, the cylinder lens was moved away. The stage moved at a speed of 10 mm/min, and correspondingly, the point dwell time is 1.4 µs. The measured FWHM of the normalized intensity in lateral is 0.45 µm ± 0.05 µm (mean ± SD, n = 4) (Fig. 1(e)) and 1.8 µm ± 0.18 µm (mean ± SD, n = 4) in axial (Fig. 1(f)). Theoretically, diffraction limits of this system should be 0.28 µm laterally and 0.77 µm axially. The spatial resolution was worse mainly because the excitation light did not fully fill the back of the objective for a larger imaging range. However, the spatial resolution is still generally sufficiently high to resolve spines and axons.

2.2 Dispersion compensation

The dispersion compensation results are shown in Fig. 2
Fig. 2 Generation and compensation of astigmatism of 2p-fMOST system (a) AOD was driven with chirp acoustic frequency; the light spot lengthens in the y-direction after AOD. With the compensation of the converged cylindrical lens, the light spot becomes recollimated. (b) Image of pollen without astigmatism compensation. (c) Image of pollen with astigmatism compensation. Scale bar is 10 µm in (b) and (c).
. One disadvantage of the AOD used in two-photon imaging is the temporal and spatial dispersion. Temporal dispersion results in pulse broadening and decreases the excitation efficiency. Spatial dispersion causes beam distortion and ruins the SNR and resolution. In our system, the temporal and spatial dispersion was compensated simultaneously by a prism [22

22. S. Zeng, X. Lv, C. Zhan, W. R. Chen, W. Xiong, S. L. Jacques, and Q. Luo, “Simultaneous compensation for spatial and temporal dispersion of acousto-optical deflectors for two-dimensional scanning with a single prism,” Opt. Lett. 31(8), 1091–1093 (2006). [CrossRef] [PubMed]

, 23

23. D. Li, S. Zeng, X. Lv, J. Liu, R. Du, R. Jiang, W. R. Chen, and Q. Luo, “Dispersion characteristics of acousto-optic deflector for scanning Gaussian laser beam of femtosecond pulses,” Opt. Express 15(8), 4726–4734 (2007). [CrossRef] [PubMed]

]. The central wavelength of laser pulse is 920 nm, and the original pulse width is 140 fs. The pulse width was broadened by the AOD to 750 fs at the front focal plane of the objective lens without compensation. We used a custom-designed prism with a 66° vertex angle and a 75° incident angle to compensate the dispersions. With compensation, the pulse width was compressed to 169 fs at the front focal plane of the objective lens. The pulse width after compensation is still slightly larger than original, which is mainly caused by the residual angle dispersion in the laser and the minimal high level dispersion of the optical components in the system. However, the compressed pulse is extremely close to the original pulse, and the image performance would be nearly the same as the original.

2.3 Astigmatism compensation

The generation and compensation of astigmatism of 2p-fMOST system was described in Fig. 3
Fig. 3 Imaging strategy of 2p-fMOST system. (a) The blurred area represents the area that has not been imaged, the laser scanning in the y-direction is shown by the red line, and the scanning in the x-direction was accomplished by the stage. (b) Different imaging areas were set according to brain profile. The red frame stands for the real imaging area, and the green areas depict the shape of brain in the different z-planes.
. When the AOD works in fast mode, the acoustic signal varies very fast in the AOD crystal, resulting in simultaneous acoustic signals with different frequencies in the aperture. Under this condition, the diffracted laser not only deflects in the lateral, but also converges or diverges in the direction of acoustic signal and spreads as a cylindrical lens [24

24. A. Kaplan, N. Friedman, and N. Davidson, “Acousto-optic lens with very fast focus scanning,” Opt. Lett. 26(14), 1078–1080 (2001). [CrossRef] [PubMed]

] resulting in astigmatism. The focus of the cylindrical lens is calculated as
f=v2λα.
(1)
where f is the equivalent focus of AOD, v is the velocity of acoustic spread in AOD crystal, λ is the central wavelength of the laser, and α is the chirp rate of acoustic frequency (MHz/µs). Scan speed is determined by α, and the bigger α is, the faster the AOD scans. In the 2p-fMOST system, α was −0.173 MHz/μs when the pixel residual time was 0.54 μs, corresponding to a diverged cylindrical lens of 2650 mm, which was placed closely after the AOD to compensate for the astigmatism to recollimate the light (Fig. 2(a)). To test the compensation effect, we took images of the same pollen without the diverged cylinder and with the cylinder lens. The results show an obvious improvement in resolution after compensation (Fig. 2(b) and Fig. 2(c)). In addition, different cylindrical lenses were made to coordinate with different imaging speeds.

2.4 Imaging strategy

The high-throughput property of 2p-fMOST benefits from its imaging strategy. In general, large-sample imaging requires the movement of the sample after imaging of a small area called mosaic, and one x-y optical section consists of numerous mosaics. Mosaic movement includes acceleration, constant speed, and deceleration. Therefore, the movement of the mosaic is an important part in the whole imaging time, especially in prolonged imaging. Moreover, the two-dimensional mosaic splicing mode increases the complexity of control and requires higher image uniformity of the mosaics. Here, we adopted another image formation strategy that can fully utilize the high-speed motion property of the stage. The formation of the image and the movement of the sample were accomplished simultaneously because the x-direction scanning of the image was realized by the stage (Fig. 3(a)). Hence, no time was wasted for moving mosaics in 2p-fMOST. In addition, the results showed the absence of imaging performance reduction using this imaging strategy. Furthermore, the whole x-y optical section image is spliced in one dimension, facilitating the control and processing of the images.

Moreover, we find that the effective imaging areas remarkably differ in distinct z-planes of the brain. Hence, using a stationary scanning mode will generate a number of useless data and increase the whole imaging time, resulting in wastage in both imaging time and data storage. The brain profile varies smoothly along the z-plane, and we need only to change the image area in several key points for a more effective imaging. In actuality, we changed the scan range 11 times during the whole imaging process and decreased the time of imaging by 25%.

Given the effective imaging strategy, we obtained an average imaging time of 2 min and 2 s for each optical section with 0.54 µs pixel residence time and 0.5 µm pixel size, which is nearly 2.5 times faster than those previously reported mosaic imaging strategy using similar optical parameters [19

19. T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods 9(3), 255–258 (2012). [CrossRef] [PubMed]

].

2.5 Sample preparation

Deeply anesthetized Thy1-EGFP M line mice [25

25. G. Feng, R. H. Mellor, M. Bernstein, C. Keller-Peck, Q. T. Nguyen, M. Wallace, J. M. Nerbonne, J. W. Lichtman, and J. R. Sanes, “Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP,” Neuron 28(1), 41–51 (2000). [CrossRef] [PubMed]

] were transcardially perfused with 50 mL phosphate buffered saline (PBS) (0.1 M) and 200 mL 4% paraformaldehyde (PFA, w/w, PFA in PBS). The brains were post-fixed in 4% PFA overnight at 4 °C after being dissected out. Furthermore, the brains were rinsed in PBS thrice (2 h each for the first two times and overnight for the third one). Then, the brains were dehydrated in a graded series of alcohol (50%, 70%, and 95% each for 1 h). Afterward, we infiltrated the brains in a graded series of GMA [26

26. S. Watanabe, A. Punge, G. Hollopeter, K. I. Willig, R. J. Hobson, M. W. Davis, S. W. Hell, and E. M. Jorgensen, “Protein localization in electron micrographs using fluorescence nanoscopy,” Nat. Methods 8(1), 80–84 (2011). [CrossRef] [PubMed]

] (70%, 85%, and 100% GMA for 2 h each, and another 100% GMA infiltration overnight) and prepolymerized GMA (100% GMA heated to 120 °C and then cooled quickly in ice). Finally, the brains were embedded in gelatin capsules and polymerized at 60 °C for 60 h. Every dehydration and infiltration step was controlled at 4 °C. For better fluorescence preservation, we adjusted the pH of GMA to an alkaline environment, and the GMA we used were pre-filtered with basic alumina [27

27. K. Becker, N. Jährling, S. Saghafi, R. Weiler, and H. U. Dodt, “Chemical clearing and dehydration of GFP expressing mouse brains,” PLoS ONE 7(3), e33916 (2012). [CrossRef] [PubMed]

] to remove the polymerization inhibitor (4-methoxyphenol) prior to use. Note that 100% GMA comprise 67% 2-hydroxyethyl methacrylate, 3% water, 29.4% n-butyl methacrylate, and 0.6% benzoyl peroxide.

3. Results

3.1 2p-fMOST imaging of the entire brain

In the present study, we used 2p-fMOST system for whole brain imaging. A Thy1-EGFP-labeled transgenic mouse brain was used for high-resolution visualization of the brain circuits. The brain was from an adult (P41) male mouse and was embedded as described previously. The fluorescence of EGFP mouse was mainly expressed in the hippocampal and cortical pyramidal neurons. To generate a spine-resolved imaging performance, we used a 40 × oil objective to obtain high spatial resolution. For a higher SNR, we set a pixel residence time of 0.54 μs to collect enough signal. To map the pixel residence time, the stage moved at a speed of 150 mm/min in the x-direction, and a 2650 mm cylinder lens was used to compensate for the astigmatism.

To obtain uniform excitation in different depths of the sample, the imaging depth was restricted to 30 µm. Hence, 30 µm thick samples were imaged, and the same thickness surface was cut in a single imaging and sectioning process. During imaging of the 30 µm thick block, the z-spacing was 2 µm, and the block was composed of 15 single x-y optical sections. The whole brain imaging process consists of 422 single imaging and sectioning processes, corresponding to a depth of 12.66 mm axially. After 215 h, we generated an entire brain database with a voxel size of 0.5 × 0.5 × 2 µm3. Based on this data set, we visualized the mouse brain in 3D (Fig. 4(a)
Fig. 4 Imaging performance on the entire brain. (a) 3D volume presentation of an EGFP mouse brain. The superimposed plane refers to coronal sections shown in (b). (b) 100 µm z-plane maximum projection image, coronal view, scale bar: 1 mm. (c) Enlarged view of the neurons signed as red frame in (b), contrast was enhanced 30% for a better visibility, scale bar: 200 µm. (d) Enlarged view signed as red frame in (c), scale bar: 30 µm. (e) Enlarged view of dendrites signed as red frame in (d), scale bar: 5 µm. dendritic spines are as white arrows pointed. (f) Enlarged view of the lower frame in (c), scale bar: 5 µm, axon is as white arrows pointed (See Media 1).
and Media 1), and the reconstructed result represented a clear profile of brain. Furthermore, the precise structures of the cortical pyramidal neurons were enlarged in Fig. 4 such that the spines and axons were clearly resolved. The precise structure of fibers set up a significant foundation for axon tracing.

2.3 No distortion in imaging after sectioning

GMA-embedded samples had very good cutting performance in obtaining unbroken 1 µm sections using only a simple tungsten carbide knife because of its sufficient hardness. Additionally, the mechanical performances are almost similar in the different spatial distributions. This property produces few deformations during the whole sectioning process. Unavoidably, several residue cutting traces are found on the surface of the sample. However, we have tested the thickness of the cutting traces once the imaging depth ran to 15 µm, and no diacritical disparity in structure (few differences in intensity) was found between the images prior to and post cutting.

To demonstrate the sectioning performance of the 2p-fMOST system, we imaged the same area of the brain before and after sectioning. First, we imaged a 60 µm thick block of brain, and the beginning z-depth is 60 µm below the surface of sample. Then, the top 30 µm sample was sectioned. Afterward, we imaged another 60 µm thick block of brain, and the beginning z-depth is 30 µm below the surface of sample. Finally, the two 60 µm z-plane maximum projection images were merged, and the two pictures matched very well in all typical structures, even the thin dendrites were highly superposed (Fig. 5
Fig. 5 No distortion in imaging after sectioning. (a) A 60 µm z-plane maximum projection image of neurons, depth from 60 μm to 120 μm below the surface of sample. (b) A 60 µm z-plane maximum projection image of neurons with z-depth from 30 μm to 90 μm below the surface after a 30 μm thick sectioning. (c) Merged image of (a) and (b), the mapped areas of (a) and (b) shown in yellow. (d), (e), and (f) are the enlarged views of the white frame in (a), (b), and (c) respectively. High superposition in any typical structure was shown, including soma and slim dendrites as white arrows pointed. Scale bar is 1 mm in (a), (b), and (c), and 20 μm in (d), (e), and (f).
). The match proved the absence of distortion in imaging after sectioning and confirmed that tracing a continuous axon in the entire brain is possible.

2.4 Trace long-range axon projections in whole brain

Based on the high spatial resolution and absence of distortion features, we traced several intact axon projections in the whole brain. One of the main difficulties in axon tracing is the axon cross resulting from low axial resolution and low z-spacing. The ability to discriminate fibers is simultaneously restricted by the 3D resolution and the voxel size. The lack of spatial resolution will result in fiber obscurity. Accordingly, the insufficiency of voxel size will cause loss of information. Closely distributed axons could not be resolved when either of these factor is deficient, hindering tracing and conception. The voxel size and the 3D resolutions were all below 2 µm, which can offer a better discrimination ability of closely distributed fibers than STP and CLSM. Given the high resolution and high sample density, we can separate the closely distributed axons to trace the intact projection.

Finally, we illustrated the long-range projections over a centimeter-sized volume. The mouse brain preparation and the imaging parameters were the same as previously described. First, we visualized the restricted data in 3D and then manually traced several long-distance projections in the whole brain using 3D visualizing software (Amira). We also showed 8 long-range projections in 3D (Fig. 6(a)
Fig. 6 Tracing long-range axonal projections in the whole brain. (a) 8 axonal projections were traced in 3D, different colors represent different axonal projections. (b) Coronal view of the projections. (c) Sagittal views of the projections. (d) Horizon view of the projections. Scale bar is 1 mm in (a) to (d) (See Media 2).
and Media 2). Lastly, we demonstrated the traced long-range projections in coronal (Fig. 6(b)), sagittal (Fig. 6(c)), and horizon (Fig. 6(d)) views. The somas of the axons we traced were distributed in different regions of the brain and the longest of these axons was 12.5 mm. The results have demonstrated that 2p-fMOST method is a good tool in studies that trace long-range projections. With the help of automatic tool [28

28. T. Quan, T. Zheng, Z. Yang, W. Ding, S. Li, J. Li, H. Zhou, Q. Luo, H. Gong, and S. Zeng, “NeuroGPS: automated localization of neurons for brain circuits using L1 minimization model,” Sci Rep 3, 1414 (2013), doi:. [CrossRef] [PubMed]

], it will more powerful.

4. Discussion

Here, we described the 2p-fMOST method for high-throughput and high-resolution visualization of the brain circuits. The excellent performance was achieved by the combination of a number of techniques such as AOD as a fast and steady scanner, two-photon imaging with high numerical aperture oil objective lens for high spatial resolution, effectively compensating dispersions and astigmatism, utilizing efficient imaging strategy to obtain high-throughput, and using plastic embedded sample to eliminate distortion during the entire imaging and sectioning process. Thus, we can trace axon projections in the whole brain with high resolution, during which overcame several difficulties. AOD has been widely used in neuroscience because of its high accuracy, stability, and fast random access feature [21

21. H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage 74, 87–98 (2013), doi:. [CrossRef] [PubMed]

, 29

29. G. Duemani Reddy, K. Kelleher, R. Fink, and P. Saggau, “Three-dimensional random access multiphoton microscopy for functional imaging of neuronal activity,” Nat. Neurosci. 11(6), 713–720 (2008). [CrossRef] [PubMed]

32

32. X. Liu, T. Quan, S. Zeng, and X. Lv, “Identification of the direction of the neural network activation with a cellular resolution by fast two-photon imaging,” J. Biomed. Opt. 16(8), 080506 (2011). [CrossRef] [PubMed]

]. Recently, it has been used to record the calcium signal in dendrite spines at an imaging rate of 1000 frames/s [33

33. X. Chen, U. Leischner, N. L. Rochefort, I. Nelken, and A. Konnerth, “Functional mapping of single spines in cortical neurons in vivo,” Nature 475(7357), 501–505 (2011). [CrossRef] [PubMed]

]. In this study, the AOD enables fast, high precision, and long-standing continuous scanning of more than 10 days to acquire the image of a whole brain without interruption, which ensures consistent fine imaging across the whole mouse brain and enables non-interrupt tracing the brain-wide long distance projection. At the same time, maintaining high resolution when AOD works in scan mode is difficult. The dispersions and astigmatism will significantly ruin the resolution and the SNR. Hence, we compensated the temporal and spatial dispersions with a single customized prism. Moreover, a set of customized cylindrical lens were used to compensate the astigmatism of different scan speeds [24

24. A. Kaplan, N. Friedman, and N. Davidson, “Acousto-optic lens with very fast focus scanning,” Opt. Lett. 26(14), 1078–1080 (2001). [CrossRef] [PubMed]

, 33

33. X. Chen, U. Leischner, N. L. Rochefort, I. Nelken, and A. Konnerth, “Functional mapping of single spines in cortical neurons in vivo,” Nature 475(7357), 501–505 (2011). [CrossRef] [PubMed]

]. To trace long-range axon projections, we adopted a voxel size of 0.5 × 0.5 × 2 µm3 to obtain a high discrimination ability of fibers. However, the high sampling density in 3D requests a higher throughput of 2p-fMOST. The optimization of image formation and the improvement of scan area resulted in a faster technique, nearly 2.5 times faster than previously reported mosaic imaging strategy using similar optical imaging parameters [19

19. T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods 9(3), 255–258 (2012). [CrossRef] [PubMed]

]. Lastly, the plastic embedded sample offered an excellent cutting performance, and as a result, no information was lost and no mismatch of fibers occurred during the entire imaging process.

Compared with other optical methods in neuron connective research, 2p-fMOST is evidently advantageous in long-range axon tracing. Light sheet microscopy-based methods can obtain whole brain data. However, poor resolution restricts their applications. High resolution can be achieved using high NA objective, but this will decrease the size and uniformity of illumination as well as at the cost of imaging time [20

20. L. Silvestri, A. Bria, L. Sacconi, G. Iannello, and F. S. Pavone, “Confocal light sheet microscopy: micron-scale neuroanatomy of the entire mouse brain,” Opt. Express 20(18), 20582–20598 (2012). [CrossRef] [PubMed]

]. Furthermore, the actual imaging performance highly depends on the optical clearing effects, and there will be strong unavoidable remaining scattering when imaging large samples. Although high resolution could be obtained in the STP method for each optical section of 1-3 microns thickness depending on the numerical aperture of the objective and the wavelength, in its demonstration, only an optical section is sampled in every block of 50 µm thickness to ensure high speed for whole mouse brain imaging condition [19

19. T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods 9(3), 255–258 (2012). [CrossRef] [PubMed]

]. Thus vast information is lost. Up to now, no reports on long range axon projection using this method have been published yet. Moreover, once the z spacing decrease to 2 µm, the whole imaging time will increase to 25 days accordingly. It is not clear how STP will perform for prolonged imaging and sectioning using agarose embedded samples, as agarose embedded samples is tender and seriously suffers from the vibration slicer used and the distortion of the brain will become apparent. The fMOST method [21

21. H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage 74, 87–98 (2013), doi:. [CrossRef] [PubMed]

] performed immediate imaging on the sectioned ribbon, the background could be well blocked by mechanic sectioning. However, imaging performance highly relies on the cutting performance and the embedded condition, which casts constrains accordingly. In comparison, 2p-fMOST uses imaging beneath the surface and then sectioning strategy, which significantly decreases the cutting requirements and the embedded condition, thus is more flexible compared to fMOST.

The line scan speed of AOD was 5 k Hz when used in long-range axon projection, which is nearly 10 times away from its highest speed. However, this condition is mainly because the sample is not sufficiently bright. Hence, we reduced the scan speed to obtain enough signal. This condition reveals that 2p-fMOST has a great potential in acquiring high throughput and an entire brain data within a day. The throughput improvement of 2p-fMOST will greatly extend its application in neuron disorders and neural disease, which requires a huge number of samples for the analysis of the effects of different drugs on the neuron network. In addition, the sample was not restricted in GMA embedding, so other embedding methods can also be used in 2p-fMOST.

Although the features of the proposed method were demonstrated in entire brain imaging, different imaging parameter settings can be used to fit different resolution requirements, from spine resolution to cell resolution. Thus, 2p-fMOST could be used not only in neuron connections, but also in any study that requires sub-micro resolution or micro resolution imaging of large samples.

Acknowledgments

We thank C. Yan, Z. Feng, J. Wu, and Y. Zhao of Britton Chance Center for Biomedical Photonics for advices. This work is supported by the National Major Scientific Research Program of China (No. 2011CB910401), the National Nature Science Foundation of China (Grant Nos. 81127002, 30925013, 91232306), Science Fund for Creative Research Group of China (Grant No. 61121004) and 985 project. Correspondence should be addressed to H. Gong at huigong@mail.hust.edu.cn, or S. Zeng at sqzeng@mail.hust.edu.cn.

References and links

1.

V. Marx, “High-throughput anatomy: Charting the brain’s networks,” Nature 490(7419), 293–298 (2012). [CrossRef] [PubMed]

2.

J. W. Lichtman and W. Denk, “The big and the small: challenges of imaging the brain’s circuits,” Science 334(6056), 618–623 (2011). [CrossRef] [PubMed]

3.

W. Denk and H. Horstmann, “Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure,” PLoS Biol. 2(11), e329 (2004). [CrossRef] [PubMed]

4.

K. L. Briggman, M. Helmstaedter, and W. Denk, “Wiring specificity in the direction-selectivity circuit of the retina,” Nature 471(7337), 183–188 (2011). [CrossRef] [PubMed]

5.

L. Luo, E. M. Callaway, and K. Svoboda, “Genetic dissection of neural circuits,” Neuron 57(5), 634–660 (2008). [CrossRef] [PubMed]

6.

K. D. Micheva and S. J. Smith, “Array tomography: a new tool for imaging the molecular architecture and ultrastructure of neural circuits,” Neuron 55(1), 25–36 (2007). [CrossRef] [PubMed]

7.

E. L. Bearer, X. Zhang, and R. E. Jacobs, “Live imaging of neuronal connections by magnetic resonance: Robust transport in the hippocampal-septal memory circuit in a mouse model of Down syndrome,” Neuroimage 37(1), 230–242 (2007). [CrossRef] [PubMed]

8.

S. Canals, M. Beyerlein, A. L. Keller, Y. Murayama, and N. K. Logothetis, “Magnetic resonance imaging of cortical connectivity in vivo,” Neuroimage 40(2), 458–472 (2008). [CrossRef] [PubMed]

9.

S. Mori, K. Oishi, and A. V. Faria, “White matter atlases based on diffusion tensor imaging,” Curr. Opin. Neurol. 22(4), 362–369 (2009). [CrossRef] [PubMed]

10.

L. A. Harsan, D. Paul, S. Schnell, B. W. Kreher, J. Hennig, J. F. Staiger, and D. von Elverfeldt, “In vivo diffusion tensor magnetic resonance imaging and fiber tracking of the mouse brain,” NMR Biomed. 23(7), 884–896 (2010). [CrossRef] [PubMed]

11.

O. Eschenko, H. C. Evrard, R. M. Neves, M. Beyerlein, Y. Murayama, and N. K. Logothetis, “Tracing of noradrenergic projections using manganese-enhanced MRI,” Neuroimage 59(4), 3252–3265 (2012). [CrossRef] [PubMed]

12.

P. J. Keller, A. D. Schmidt, A. Santella, K. Khairy, Z. Bao, J. Wittbrodt, and E. H. Stelzer, “Fast, high-contrast imaging of animal development with scanned light sheet-based structured-illumination microscopy,” Nat. Methods 7(8), 637–642 (2010). [CrossRef] [PubMed]

13.

J. Mertz, “Optical sectioning microscopy with planar or structured illumination,” Nat. Methods 8(10), 811–819 (2011). [CrossRef] [PubMed]

14.

P. S. Tsai, B. Friedman, A. I. Ifarraguerri, B. D. Thompson, V. Lev-Ram, C. B. Schaffer, Q. Xiong, R. Y. Tsien, J. A. Squier, and D. Kleinfeld, “All-optical histology using ultrashort laser pulses,” Neuron 39(1), 27–41 (2003). [CrossRef] [PubMed]

15.

H. U. Dodt, U. Leischner, A. Schierloh, N. Jährling, C. P. Mauch, K. Deininger, J. M. Deussing, M. Eder, W. Zieglgänsberger, and K. Becker, “Ultramicroscopy: three-dimensional visualization of neuronal networks in the whole mouse brain,” Nat. Methods 4(4), 331–336 (2007). [CrossRef] [PubMed]

16.

D. Mayerich, L. Abbott, and B. McCormick, “Knife-edge scanning microscopy for imaging and reconstruction of three-dimensional anatomical structures of the mouse brain,” J. Microsc. 231(1), 134–143 (2008). [CrossRef] [PubMed]

17.

A. Li, H. Gong, B. Zhang, Q. Wang, C. Yan, J. Wu, Q. Liu, S. Zeng, and Q. Luo, “Micro-optical sectioning tomography to obtain a high-resolution atlas of the mouse brain,” Science 330(6009), 1404–1408 (2010). [CrossRef] [PubMed]

18.

K. Minogue, “Neuroscience. China’s brain mappers zoom in on neural connections,” Science 330(6005), 747 (2010). [CrossRef] [PubMed]

19.

T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods 9(3), 255–258 (2012). [CrossRef] [PubMed]

20.

L. Silvestri, A. Bria, L. Sacconi, G. Iannello, and F. S. Pavone, “Confocal light sheet microscopy: micron-scale neuroanatomy of the entire mouse brain,” Opt. Express 20(18), 20582–20598 (2012). [CrossRef] [PubMed]

21.

H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage 74, 87–98 (2013), doi:. [CrossRef] [PubMed]

22.

S. Zeng, X. Lv, C. Zhan, W. R. Chen, W. Xiong, S. L. Jacques, and Q. Luo, “Simultaneous compensation for spatial and temporal dispersion of acousto-optical deflectors for two-dimensional scanning with a single prism,” Opt. Lett. 31(8), 1091–1093 (2006). [CrossRef] [PubMed]

23.

D. Li, S. Zeng, X. Lv, J. Liu, R. Du, R. Jiang, W. R. Chen, and Q. Luo, “Dispersion characteristics of acousto-optic deflector for scanning Gaussian laser beam of femtosecond pulses,” Opt. Express 15(8), 4726–4734 (2007). [CrossRef] [PubMed]

24.

A. Kaplan, N. Friedman, and N. Davidson, “Acousto-optic lens with very fast focus scanning,” Opt. Lett. 26(14), 1078–1080 (2001). [CrossRef] [PubMed]

25.

G. Feng, R. H. Mellor, M. Bernstein, C. Keller-Peck, Q. T. Nguyen, M. Wallace, J. M. Nerbonne, J. W. Lichtman, and J. R. Sanes, “Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP,” Neuron 28(1), 41–51 (2000). [CrossRef] [PubMed]

26.

S. Watanabe, A. Punge, G. Hollopeter, K. I. Willig, R. J. Hobson, M. W. Davis, S. W. Hell, and E. M. Jorgensen, “Protein localization in electron micrographs using fluorescence nanoscopy,” Nat. Methods 8(1), 80–84 (2011). [CrossRef] [PubMed]

27.

K. Becker, N. Jährling, S. Saghafi, R. Weiler, and H. U. Dodt, “Chemical clearing and dehydration of GFP expressing mouse brains,” PLoS ONE 7(3), e33916 (2012). [CrossRef] [PubMed]

28.

T. Quan, T. Zheng, Z. Yang, W. Ding, S. Li, J. Li, H. Zhou, Q. Luo, H. Gong, and S. Zeng, “NeuroGPS: automated localization of neurons for brain circuits using L1 minimization model,” Sci Rep 3, 1414 (2013), doi:. [CrossRef] [PubMed]

29.

G. Duemani Reddy, K. Kelleher, R. Fink, and P. Saggau, “Three-dimensional random access multiphoton microscopy for functional imaging of neuronal activity,” Nat. Neurosci. 11(6), 713–720 (2008). [CrossRef] [PubMed]

30.

B. F. Grewe, D. Langer, H. Kasper, B. M. Kampa, and F. Helmchen, “High-speed in vivo calcium imaging reveals neuronal network activity with near-millisecond precision,” Nat. Methods 7(5), 399–405 (2010). [CrossRef] [PubMed]

31.

G. Katona, G. Szalay, P. Maák, A. Kaszás, M. Veress, D. Hillier, B. Chiovini, E. S. Vizi, B. Roska, and B. Rózsa, “Fast two-photon in vivo imaging with three-dimensional random-access scanning in large tissue volumes,” Nat. Methods 9(2), 201–208 (2012). [CrossRef] [PubMed]

32.

X. Liu, T. Quan, S. Zeng, and X. Lv, “Identification of the direction of the neural network activation with a cellular resolution by fast two-photon imaging,” J. Biomed. Opt. 16(8), 080506 (2011). [CrossRef] [PubMed]

33.

X. Chen, U. Leischner, N. L. Rochefort, I. Nelken, and A. Konnerth, “Functional mapping of single spines in cortical neurons in vivo,” Nature 475(7357), 501–505 (2011). [CrossRef] [PubMed]

OCIS Codes
(170.0180) Medical optics and biotechnology : Microscopy
(170.2520) Medical optics and biotechnology : Fluorescence microscopy
(180.6900) Microscopy : Three-dimensional microscopy
(230.1040) Optical devices : Acousto-optical devices
(180.4315) Microscopy : Nonlinear microscopy

ToC Category:
Medical Optics and Biotechnology

History
Original Manuscript: January 15, 2013
Revised Manuscript: March 28, 2013
Manuscript Accepted: March 31, 2013
Published: April 12, 2013

Virtual Issues
Vol. 8, Iss. 5 Virtual Journal for Biomedical Optics

Citation
Ting Zheng, Zhongqing Yang, Anan Li, Xiaohua Lv, Zhenqiao Zhou, Xiaojun Wang, Xiaoli Qi, Shiwei Li, Qingming Luo, Hui Gong, and Shaoqun Zeng, "Visualization of brain circuits using two-photon fluorescence micro-optical sectioning tomography," Opt. Express 21, 9839-9850 (2013)
http://www.opticsinfobase.org/vjbo/abstract.cfm?URI=oe-21-8-9839


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. V. Marx, “High-throughput anatomy: Charting the brain’s networks,” Nature490(7419), 293–298 (2012). [CrossRef] [PubMed]
  2. J. W. Lichtman and W. Denk, “The big and the small: challenges of imaging the brain’s circuits,” Science334(6056), 618–623 (2011). [CrossRef] [PubMed]
  3. W. Denk and H. Horstmann, “Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure,” PLoS Biol.2(11), e329 (2004). [CrossRef] [PubMed]
  4. K. L. Briggman, M. Helmstaedter, and W. Denk, “Wiring specificity in the direction-selectivity circuit of the retina,” Nature471(7337), 183–188 (2011). [CrossRef] [PubMed]
  5. L. Luo, E. M. Callaway, and K. Svoboda, “Genetic dissection of neural circuits,” Neuron57(5), 634–660 (2008). [CrossRef] [PubMed]
  6. K. D. Micheva and S. J. Smith, “Array tomography: a new tool for imaging the molecular architecture and ultrastructure of neural circuits,” Neuron55(1), 25–36 (2007). [CrossRef] [PubMed]
  7. E. L. Bearer, X. Zhang, and R. E. Jacobs, “Live imaging of neuronal connections by magnetic resonance: Robust transport in the hippocampal-septal memory circuit in a mouse model of Down syndrome,” Neuroimage37(1), 230–242 (2007). [CrossRef] [PubMed]
  8. S. Canals, M. Beyerlein, A. L. Keller, Y. Murayama, and N. K. Logothetis, “Magnetic resonance imaging of cortical connectivity in vivo,” Neuroimage40(2), 458–472 (2008). [CrossRef] [PubMed]
  9. S. Mori, K. Oishi, and A. V. Faria, “White matter atlases based on diffusion tensor imaging,” Curr. Opin. Neurol.22(4), 362–369 (2009). [CrossRef] [PubMed]
  10. L. A. Harsan, D. Paul, S. Schnell, B. W. Kreher, J. Hennig, J. F. Staiger, and D. von Elverfeldt, “In vivo diffusion tensor magnetic resonance imaging and fiber tracking of the mouse brain,” NMR Biomed.23(7), 884–896 (2010). [CrossRef] [PubMed]
  11. O. Eschenko, H. C. Evrard, R. M. Neves, M. Beyerlein, Y. Murayama, and N. K. Logothetis, “Tracing of noradrenergic projections using manganese-enhanced MRI,” Neuroimage59(4), 3252–3265 (2012). [CrossRef] [PubMed]
  12. P. J. Keller, A. D. Schmidt, A. Santella, K. Khairy, Z. Bao, J. Wittbrodt, and E. H. Stelzer, “Fast, high-contrast imaging of animal development with scanned light sheet-based structured-illumination microscopy,” Nat. Methods7(8), 637–642 (2010). [CrossRef] [PubMed]
  13. J. Mertz, “Optical sectioning microscopy with planar or structured illumination,” Nat. Methods8(10), 811–819 (2011). [CrossRef] [PubMed]
  14. P. S. Tsai, B. Friedman, A. I. Ifarraguerri, B. D. Thompson, V. Lev-Ram, C. B. Schaffer, Q. Xiong, R. Y. Tsien, J. A. Squier, and D. Kleinfeld, “All-optical histology using ultrashort laser pulses,” Neuron39(1), 27–41 (2003). [CrossRef] [PubMed]
  15. H. U. Dodt, U. Leischner, A. Schierloh, N. Jährling, C. P. Mauch, K. Deininger, J. M. Deussing, M. Eder, W. Zieglgänsberger, and K. Becker, “Ultramicroscopy: three-dimensional visualization of neuronal networks in the whole mouse brain,” Nat. Methods4(4), 331–336 (2007). [CrossRef] [PubMed]
  16. D. Mayerich, L. Abbott, and B. McCormick, “Knife-edge scanning microscopy for imaging and reconstruction of three-dimensional anatomical structures of the mouse brain,” J. Microsc.231(1), 134–143 (2008). [CrossRef] [PubMed]
  17. A. Li, H. Gong, B. Zhang, Q. Wang, C. Yan, J. Wu, Q. Liu, S. Zeng, and Q. Luo, “Micro-optical sectioning tomography to obtain a high-resolution atlas of the mouse brain,” Science330(6009), 1404–1408 (2010). [CrossRef] [PubMed]
  18. K. Minogue, “Neuroscience. China’s brain mappers zoom in on neural connections,” Science330(6005), 747 (2010). [CrossRef] [PubMed]
  19. T. Ragan, L. R. Kadiri, K. U. Venkataraju, K. Bahlmann, J. Sutin, J. Taranda, I. Arganda-Carreras, Y. Kim, H. S. Seung, and P. Osten, “Serial two-photon tomography for automated ex vivo mouse brain imaging,” Nat. Methods9(3), 255–258 (2012). [CrossRef] [PubMed]
  20. L. Silvestri, A. Bria, L. Sacconi, G. Iannello, and F. S. Pavone, “Confocal light sheet microscopy: micron-scale neuroanatomy of the entire mouse brain,” Opt. Express20(18), 20582–20598 (2012). [CrossRef] [PubMed]
  21. H. Gong, S. Zeng, C. Yan, X. Lv, Z. Yang, T. Xu, Z. Feng, W. Ding, X. Qi, A. Li, J. Wu, and Q. Luo, “Continuously tracing brain-wide long-distance axonal projections in mice at a one-micron voxel resolution,” Neuroimage74, 87–98 (2013), doi:. [CrossRef] [PubMed]
  22. S. Zeng, X. Lv, C. Zhan, W. R. Chen, W. Xiong, S. L. Jacques, and Q. Luo, “Simultaneous compensation for spatial and temporal dispersion of acousto-optical deflectors for two-dimensional scanning with a single prism,” Opt. Lett.31(8), 1091–1093 (2006). [CrossRef] [PubMed]
  23. D. Li, S. Zeng, X. Lv, J. Liu, R. Du, R. Jiang, W. R. Chen, and Q. Luo, “Dispersion characteristics of acousto-optic deflector for scanning Gaussian laser beam of femtosecond pulses,” Opt. Express15(8), 4726–4734 (2007). [CrossRef] [PubMed]
  24. A. Kaplan, N. Friedman, and N. Davidson, “Acousto-optic lens with very fast focus scanning,” Opt. Lett.26(14), 1078–1080 (2001). [CrossRef] [PubMed]
  25. G. Feng, R. H. Mellor, M. Bernstein, C. Keller-Peck, Q. T. Nguyen, M. Wallace, J. M. Nerbonne, J. W. Lichtman, and J. R. Sanes, “Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP,” Neuron28(1), 41–51 (2000). [CrossRef] [PubMed]
  26. S. Watanabe, A. Punge, G. Hollopeter, K. I. Willig, R. J. Hobson, M. W. Davis, S. W. Hell, and E. M. Jorgensen, “Protein localization in electron micrographs using fluorescence nanoscopy,” Nat. Methods8(1), 80–84 (2011). [CrossRef] [PubMed]
  27. K. Becker, N. Jährling, S. Saghafi, R. Weiler, and H. U. Dodt, “Chemical clearing and dehydration of GFP expressing mouse brains,” PLoS ONE7(3), e33916 (2012). [CrossRef] [PubMed]
  28. T. Quan, T. Zheng, Z. Yang, W. Ding, S. Li, J. Li, H. Zhou, Q. Luo, H. Gong, and S. Zeng, “NeuroGPS: automated localization of neurons for brain circuits using L1 minimization model,” Sci Rep3, 1414 (2013), doi:. [CrossRef] [PubMed]
  29. G. Duemani Reddy, K. Kelleher, R. Fink, and P. Saggau, “Three-dimensional random access multiphoton microscopy for functional imaging of neuronal activity,” Nat. Neurosci.11(6), 713–720 (2008). [CrossRef] [PubMed]
  30. B. F. Grewe, D. Langer, H. Kasper, B. M. Kampa, and F. Helmchen, “High-speed in vivo calcium imaging reveals neuronal network activity with near-millisecond precision,” Nat. Methods7(5), 399–405 (2010). [CrossRef] [PubMed]
  31. G. Katona, G. Szalay, P. Maák, A. Kaszás, M. Veress, D. Hillier, B. Chiovini, E. S. Vizi, B. Roska, and B. Rózsa, “Fast two-photon in vivo imaging with three-dimensional random-access scanning in large tissue volumes,” Nat. Methods9(2), 201–208 (2012). [CrossRef] [PubMed]
  32. X. Liu, T. Quan, S. Zeng, and X. Lv, “Identification of the direction of the neural network activation with a cellular resolution by fast two-photon imaging,” J. Biomed. Opt.16(8), 080506 (2011). [CrossRef] [PubMed]
  33. X. Chen, U. Leischner, N. L. Rochefort, I. Nelken, and A. Konnerth, “Functional mapping of single spines in cortical neurons in vivo,” Nature475(7357), 501–505 (2011). [CrossRef] [PubMed]

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.

Supplementary Material


» Media 1: AVI (1054 KB)     
» Media 2: AVI (624 KB)     
» Media 3: AVI (1054 KB)     
» Media 4: AVI (624 KB)     

« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited