Video-rate volumetric functional imaging of the brain at synaptic resolution

Journal name:
Nature Neuroscience
Volume:
20,
Pages:
620–628
Year published:
DOI:
doi:10.1038/nn.4516
Received
Accepted
Published online

Abstract

Neurons and neural networks often extend hundreds of micrometers in three dimensions. Capturing the calcium transients associated with their activity requires volume imaging methods with subsecond temporal resolution. Such speed is a challenge for conventional two-photon laser-scanning microscopy, because it depends on serial focal scanning in 3D and indicators with limited brightness. Here we present an optical module that is easily integrated into standard two-photon laser-scanning microscopes to generate an axially elongated Bessel focus, which when scanned in 2D turns frame rate into volume rate. We demonstrated the power of this approach in enabling discoveries for neurobiology by imaging the calcium dynamics of volumes of neurons and synapses in fruit flies, zebrafish larvae, mice and ferrets in vivo. Calcium signals in objects as small as dendritic spines could be resolved at video rates, provided that the samples were sparsely labeled to limit overlap in their axially projected images.

At a glance

Figures

  1. Concept, design and performance of Bessel module for in vivo volume imaging.
    Figure 1: Concept, design and performance of Bessel module for in vivo volume imaging.

    (a) Scanning a Gaussian focus in the xy plane images a thin optical section (yellow-shaded region), whereas scanning a Bessel focus in xy images the structures in a 3D volume (red-shaded region). (b) Schematics of Bessel module, composed of an SLM, three lenses (L1, L2, L3) and an annular mask. The module generates an annular illumination pattern on the galvanometers (galvo) and, after conjugation by lenses L4 and L5, at the back pupil of the objective (Obj), resulting in a focus approximating a Bessel beam. M1 and M2, folding mirrors; FM1 and FM2, flip mirrors to switch between Gaussian (yellow with dashed blue outline) and Bessel (red) paths. (ce) Example application of Bessel focus scanning to a volume of GCaMP6s+ neurites in awake mouse cortex at 30 Hz. (c) Mean intensity projection of a 60-μm-thick image stack collected with Gaussian focus scanning at 1-μm z steps, with structures color-coded by depth. (d) Image of the same volume of neurites collected by scanning a Bessel focus with 0.4 NA and 53-μm axial FWHM. Arrows and numbers label individual axonal varicosities (putative boutons, 1, 2) and dendritic spines (3–10). Insets in c and d show zoomed-in views of dendritic spines. Scale bars, 20 μm. (e) Representative calcium transients measured in boutons and spines from four neurons (color-coded by z depth). Objective, Olympus 25× with 1.05 NA; wavelength, 960 nm; postobjective power, 30 mW for Gaussian and 118 mW for Bessel scanning. Representative images from three mice are shown.

  2. High-throughput in vivo characterization of orientation tuning of dendritic spines in mouse V1.
    Figure 2: High-throughput in vivo characterization of orientation tuning of dendritic spines in mouse V1.

    (a) Mean intensity projection of a 3D image stack (102 μm × 102 μm × 60 μm) collected by Gaussian focus scanning of dendritic branches (100–160 μm below pia, color-coded by depth) in V1 of an anesthetized mouse. (b) Image of the same branches collected by scanning a Bessel focus with 0.4 NA and 35-μm axial FWHM. Example spines (arrowheads, s1–s11) and neighboring dendritic shafts (circles, d1–d11) are indicated. (c) Ten-trial-averaged calcium transients of spines (red) and neighboring dendritic shafts (black) in b for 12 grating angles. (d) Tuning curves of spines from c. (e) Gray histogram: distribution of preferred grating angles of all tuned spines; green line and circles: tuning curve of BAP-induced calcium transient in dendritic branches. All four main branches had similar tuning curves. Only one is shown. (f) Locations of all tuned spines, color-coded by their preferred grating angles. Objective, Olympus 25× with 1.05 NA; postobjective power, 33 mW for Gaussian scanning and 74 mW for Bessel scanning; wavelength, 960 nm; scale bars, 20 μm. Data in c, d and e are averages of ten trials with shadow (c) and error bars (d and e) representing s.e.m.

  3. Activity synchrony among GABAergic neural ensembles in V1 of awake mice.
    Figure 3: Activity synchrony among GABAergic neural ensembles in V1 of awake mice.

    (ae) Measurements on Gad2-IRES-Cre mice transfected with AAV-Syn-Flex-GCaMP6s. (a) Mean intensity projections of 3D volumes of neurons (256 μm × 256 μm × 100 μm each; 0–100, 101–200, 201–300, 301–400 μm below pia; 400 two-dimensional images, color-coded by depth) collected by scanning a Gaussian focus in V1. Postobjective power: 5, 5, 11 and 14 mW, respectively. (b) Images of the same neurons collected by scanning a Bessel focus at 0.4 NA and 78 μm axial FWHM. Postobjective power: 21, 25.2, 31.5 and 42 mW, respectively. (c) Pupil size (red) and concurrently measured calcium transients of neurons numbered in b (blue, Pearson's correlation coefficient R < −0.2; black, −0.05 < R < 0.05; magenta, R > 0.4). (d) R-values between pupil size and calcium transients (first row and first column, neurons sorted in order of descending R) and between the calcium transients of all pairs of neurons identifiable in b. (e) Distribution of R within four depth ranges measured from three mice (n, number of R-values; red open histogram, R between pupil size and neuronal calcium transients; gray filled histogram, R between neurons; dashed lines, medians). (fo) Measurements and analysis on (fj) VIP-IRES-Cre and (ko) SST-IRES-Cre mice transfected with AAV-Syn-Flex-GCaMP6s. Histograms in j and o include all VIP+ and SOM+ neurons 1–400 μm below pia. Objective, Olympus 25× with 1.05 NA; wavelength, 960 nm; scale bars, 50 μm.

  4. 30-Hz volumetric imaging of spinal projection neurons in zebrafish larvae.
    Figure 4: 30-Hz volumetric imaging of spinal projection neurons in zebrafish larvae.

    (a) Mean intensity projection of a 3D volume (270 μm × 270 μm × 100 μm, 100 two-dimensional images) of retrogradely labeled reticulospinal neurons (color-coded by depth) collected by scanning a Gaussian focus in the hindbrain of a zebrafish larva. Postobjective power, 27 mW. (b) Image of the same neurons collected by scanning a Bessel beam with 0.4 NA and 64-μm axial FWHM. Postobjective power, 78 mW. (c) Calcium transients of individual neurons numbered in b during nine repeated presentations (trial T1 to T9) of a 4-ms mechanical and acoustic stimulus (tap). Traces color-coded by depth of neuron. Objective, Olympus 25× with 1.05 NA; wavelength, 920 nm; scale bar, 20 μm.

  5. In vivo volumetric functional imaging of fruit fly brains responding to visual stimuli.
    Figure 5: In vivo volumetric functional imaging of fruit fly brains responding to visual stimuli.

    (a) Brains of head-fixed flies were imaged during the presentation of grating stimuli. (b) In vivo image of the brain of a UAS-GCaMP6f;33A12-GAL4 fly (mean intensity projection of a 60-μm volume). (c) Mean intensity projection of a 3D volume (190 μm × 95 μm × 60 μm, 61 two-dimensional images) within the orange rectangle in b of GCaMP6f+ neurites (color-coded by depth) with Gaussian focus scanning. Postobjective power, 14 mW. (d) Image of the same neurites collected by scanning a Bessel beam with 0.4 NA and 35 μm axial FWHM. Postobjective power, 82 mW. (e) Neurites within the left and right lobes of the ellipsoid body (numbered 1 and 2 in d) showed anticorrelated calcium activity. (f) Ten-trial-averaged calcium transients of two neurites (numbered 3 and 4 in d) strongly driven by grating visual stimuli. (g) Mean intensity projection of a 3D volume (307 μm × 216 μm × 100 μm; 101 two-dimensional images, color-coded by depth) of a GCaMP6f;57c10-GAL4 fly with pan-neuronal expression of GCaMP6f with Gaussian focus scanning. Postobjective power, 11 mW. (h) Image of the same brain collected by scanning a Bessel beam with 0.4 NA and 78 μm axial FWHM. Postobjective power, 31.5 mW. (i) Calcium transients (ten-trial-averaged) from areas numbered 5 and 6 in h were evoked by both the appearance of a static grating (data points within purple stripes) and the onset of its drifting movement (data points after purple stripes). (j) Optically sectioned image at z = 65 μm collected by Gaussian focus scanning. Bright structures are the left and right lobes of the mushroom body. (k) Optically sectioned responses (ten-trial-averaged) in mushroom body (areas numbered 7 and 8 in j) shared similar temporal dynamics to the axially averaged responses in i. Objective, Olympus 25× with 1.05 NA; wavelength, 940 nm; scale bars, 50 μm. Shadows in hk represent s.e.m. across trials.

  6. Optimized Bessel foci for in vivo volume imaging.
    Supplementary Fig. 1: Optimized Bessel foci for in vivo volume imaging.

    (a) Images taken by scanning Bessel foci of various NAs, lateral and axial FWHMs: (Left panels) in vivo volume images of YFP+ neurites in cortex of an awake mouse (Thy1-YFP line H); scale bar: 10 μm. (Right panels) Axial point spread functions measured from 0.2-μm-diameter beads; x scale bar: 5 μm; z scale bar: 25 μm. (b) Mean intensity projection of the same neurites (color-coded by depth below pia) imaged by scanning a 1.05-NA Gaussian focus in 3D. Higher NA Bessel foci have stronger side rings, resulting in hazy backgrounds. Objective: Olympus 25×, 1.05 NA.

  7. Axonal varicosities and dendritic spines are resolvable in single frames of 30-Hz Bessel volume scanning.
    Supplementary Fig. 2: Axonal varicosities and dendritic spines are resolvable in single frames of 30-Hz Bessel volume scanning.

    (a) Individual raw image frames during 30-Hz volumetric measurements of calcium transients in a volume extending 60 μm in z. Arrows and numbers label individual axonal varicosities (putative boutons, 1,2) and dendritic spines (3-10) (same as in Fig. 1d), with their zoomed-in view shown in the panels on the right. (b) As comparison, the same synaptic structures in the average of 5 frames (i.e., effective volume rate of 6 Hz). Scale bars: left panels, 20 μm; right panels, 5 μm.

  8. Bessel focus-scanning is robust against axial motion artifacts.
    Supplementary Fig. 3: Bessel focus-scanning is robust against axial motion artifacts.

    (a) Mean axial projection of a 3D image stack (180 μm × 180 μm × 160 μm, 100 2D images) collected by Gaussian focus scanning, of GCaMP6f+ dendrites (0-160 μm below pia) in awake mouse S1 (structures color-coded by depth). (b) Averaged image of a 266-sec time series of the same dendrites imaged with a Bessel focus (0.4 NA and 91 μm axial FWHM) at 30 Hz. (c) Averaged image after registering the time series using TurboReg plugin of ImageJ. (d) Mean axial projection of a 3D image stack (80 μm × 40 μm × 25 μm, 25 2D images) collected by Gaussian focus scanning of YFP+ dendrites (46-70 μm below pia) in awake mouse V1 (Thy1-YFP line H, structures color-coded by depth). (e) Averaged image of a 100-sec 2D-registered time series of dendrites at 53 μm below pia acquired with Gaussian focus scanning at 7.4 Hz. (f) Averaged image of a 100-sec 2D-registered time series of the volume of dendrites in d imaged with a Bessel focus (0.4 NA and 19 μm axial FWHM) at 7.4 Hz. (g,h) Brain motion (upper panels, quantified as the lateral image displacement with time) causes (g) large changes of fluorescence signal from two YFP+ dendrites (ROI1 and ROI2) in Gaussian focus scanning mode, (h) but not Bessel focus scanning mode. Objective: Olympus 25×, 1.05 NA; Post-objective power: (a) 30 mW, (b,c) 103 mW, (d,e) 9 mW, (f) 21 mW; Wavelength: (a-c) 960 nm, (d-f) 940nm. Scale bars: 20 μm.

  9. In vivo characterization of dendritic spines in ferret V1 at 30-Hz volume rate.
    Supplementary Fig. 4: In vivo characterization of dendritic spines in ferret V1 at 30-Hz volume rate.

    (a,b) Two examples of applying Bessel focus scanning to the characterization of spines in ferret V1. (Left) Mean intensity projection of 3D image stacks by conventional Gaussian focus imaging in V1 of anesthetized ferrets; (Right) Images collected by scanning a Bessel beam with 0.4 NA and 50 μm axial FWHM. Yellow circles highlight example spines. (c) 10-trial-averaged calcium transients of spines in a and b for 12 grating angles. (d) Tuning curves of individual spines in a and b. Objective: Nikon 16×, NA 0.8; Post-objective power: 55 mW for Gaussian scanning and 84 mW for Bessel scanning; Wavelength: 960 nm; Shadow (c) and error bars (d): s.e.m; Scale bars: 10 μm. Representative images from 2 ferrets.

  10. Removal of neuropil contamination in Bessel focus scanning mode.
    Supplementary Fig. 5: Removal of neuropil contamination in Bessel focus scanning mode.

    (a) (left) Image of a cell body obtained by scanning a Bessel beam with 0.4 NA and 78 μm axial FWHM, (right) overlaid with region of interest (ROI, orange outline) and the area used for calculating neuropil background (red region). (b) The difference between the averaged signal within ROI (F1) and the averaged signal within neuropil mask (Fneuropil) is used to calculate calcium transient of the neuron. After the removal of neuropil contamination, the cell body still exhibited activity correlated with pupil size. (c) and (d) same as a and b, except that the removal of neuropil contamination was applied to a volume of neuropil, resulting in minimal calcium transients. All fluorescence traces were plotted on the same scale.

  11. Activity synchrony among GABAergic neurons persists in the dark.
    Supplementary Fig. 6: Activity synchrony among GABAergic neurons persists in the dark.

    (a) Representative image of GAD2+ neurons expressing GCaMP6s (0-100 μm below pia) collected by scanning a Bessel beam with 0.4 NA and 78 μm axial FWHM. (b) Pupil size (red) and concurrently measured calcium transients of example neurons labeled in a. (c) Pearson’s correlation coefficient (R) between neuronal calcium transients and pupil size (1st row and 1st column, sorted by decreasing R), as well as between the calcium transients of all pairs of neurons identified in a. Objective: Olympus 25×, 1.05 NA; Post-objective power: 21 mW; Wavelength: 960 nm; Scale bar: 50 μm.

  12. Activity synchrony among GABAergic neurons was also observed in conventional Gaussian focus scanning mode.
    Supplementary Fig. 7: Activity synchrony among GABAergic neurons was also observed in conventional Gaussian focus scanning mode.

    (a) GABAergic neurons at three different depths (55, 73, and 93 μm) imaged by scanning a conventional Gaussian focus (1.05 NA, 1.4 μm axial FWHM). (b) Pupil size (red) and concurrently measured calcium transients of example neurons labeled in a. (c) Pearson’s correlation coefficient (R) between neuronal calcium transients and pupil size (1st row and 1st column, sorted by decreasing R), as well as between the calcium transients of all pairs of neurons identified in a. (d) Histogram distribution and (e) scatter plot of R measured from a mouse. Red open histogram: R between pupil size and neuronal calcium transients; Gray filled histogram: R between neurons; Dashed lines: median; n: number of R’s. Objective: Olympus 25×, 1.05 NA; Post-objective power: 18 mW for 55 and 93 μm depth and 14 mW for 73 μm depth; Wavelength: 960 nm; Scale bar: 50 μm.

  13. Engineering Bessel foci with asymmetric axial-intensity distribution with an SLM.
    Supplementary Fig. 8: Engineering Bessel foci with asymmetric axial-intensity distribution with an SLM.

    (a) Overlaying a binary concentric ring pattern on a quadratic phase pattern on the SLM allows the shaping of axial intensity distribution of the Bessel focus. S: period of the concentric rings in units of pixels. (b) Axial images and corresponding intensity profiles of a 2-μm-diameter bead using three different Bessel foci generated with periods S1, S2, S3. (c,d) (From left to right) in vivo images of neurons and neurites measured in V1 of awake mice using Gaussian focus, Bessel foci generated with S1, S2, and S3, respectively. Gaussian images are mean intensity projections with features color-coded by depth. Arrows point to deeper structures that had best image quality when the Bessel focus of increasing intensity with depth (S3 in b) was used. Objective: Olympus 25×, 1.05 NA; Post-objective power: c, 14 mW for Gaussian modality and 32 mW for Bessel modality; d, 21 mW for Gaussian modality and 61 mW for Bessel modality; Wavelength: 940 nm; Scale bars: 20 μm.

  14. Annular apodization mask fine-tunes the axial length and intensity distribution of Bessel foci.
    Supplementary Fig. 9: Annular apodization mask fine-tunes the axial length and intensity distribution of Bessel foci.

    (a-c) and (d-f): two NA-0.4 Bessel foci with different axial lengths and intensity distributions. (a,d) Binary concentric grating patterns (phase values: 0 and π) on SLM. S indicates the period of the gratings in units of pixels. (b,e) (Top) amplitude and (bottom) phase of the electric field of the light at the plane of the annular apodization mask (0.39 mm inner radius and 0.6 mm outer radius for b, 0.478 mm inner radius and 0.6 mm outer radius for e). Electric field within the annulus (areas between red dashed lines) was transmitted. (c,f) (Top) axial (xz) images of 2-μm-diameter fluorescent beads and (bottom) line intensity profiles along the z axis obtained with Bessel foci generated by a and b, d and e, respectively. (g) Simulated axial PSFs and (h) their intensity profiles along z as a function of annulus width of the apodization mask. Incidence beam diameter at SLM: 2.2 mm. Magnification from mask to objective back pupil: 4.7; Objective: Olympus 25×, 1.05 NA; Post-objective power: 14.2 mW and 15.5 mW for c and f, respectively; Wavelength: 900 nm; Scale bar: 10 μm. See Supplementary Technical Notes for more details.

Videos

  1. 30 Hz Bessel volumetric imaging of calcium transients in sparsely labeled neurons in V1 of awake mouse.
    Video 1: 30 Hz Bessel volumetric imaging of calcium transients in sparsely labeled neurons in V1 of awake mouse.
    GCaMP6s+ axons and dendrites in a 270 μm ×270 μm×60 μm volume (0-60 μm below pia) imaged at 30 Hz using Bessel focus scanning. Calcium transients were driven by drifting grating stimulation. Structures that appeared blurred were under blood vessels.
  2. Robustness of Bessel scanning against axial brain motion allows easy registration of volume images.
    Video 2: Robustness of Bessel scanning against axial brain motion allows easy registration of volume images.
    Unregistered (left panel, note large lateral motion) and registered (right panel) time-lapse Bessel scanning images of GCaMP6f+ dendritic branches in S1 of an awake mouse. 180 μm × 180 μm × 160 μm volume was imaged at 30 Hz rate. Registration was done with TurboReg plugin of ImageJ.
  3. Bessel focus scanning is more robust against axial motion artifacts than Gaussian focus scanning.
    Video 3: Bessel focus scanning is more robust against axial motion artifacts than Gaussian focus scanning.
    Unregistered (left panels, note large lateral motion) and registered (right panels) time-lapse Gaussian (top panels) and Bessel (bottom panels) focus scanning images of YFP+ dendritic branches in V1 of an awake mouse. Brain motion caused large focal shifts in Gaussian images, whereas dendrites imaged under Bessel mode remained in focus despite brain motion. Frame rate: 7.4 Hz. Bessel image volume: 80 μm × 40 μm × 25 μm. Registration was done with TurboReg plugin of ImageJ.
  4. Video summary of high-throughput volumetric functional mapping of single spines in mouse V1
    Video 4: Video summary of high-throughput volumetric functional mapping of single spines in mouse V1
    A video that summarizes the experiment in Fig. 2 where Bessel scanning was used for high-throughput volumetric functional mapping of single spines. 3D Gaussian scanning revealed single spines on GCaMP6s+ dendrites 100-160 μm below pia in mouse V1, whereas Bessel scanning allowed us to image all these spines in one 2D scan and measure their visually-evoked calcium transients (ten-trial averaged) to 12 different drifting grating stimuli.
  5. Video summary of 30 Hz volumetric functional imaging of spinal projection neurons in zebrafish larvae.
    Video 5: Video summary of 30 Hz volumetric functional imaging of spinal projection neurons in zebrafish larvae.
    A video that summarizes the experiment in Fig. 4 where Bessel scanning was used for functional imaging of spinal projection neurons in zebrafish larvae at 30 Hz. 3D Gaussian scanning recorded positions of spinal projection neurons in a 270 μm ×270 μm×100 μm volume, whereas Bessel scanning allowed us to image at 30 Hz their activity evoked by a mechanical and acoustic stimulus at 5-second mark.
  6. Volumetric functional imaging of a sparsely labeled fruit fly brain.
    Video 6: Volumetric functional imaging of a sparsely labeled fruit fly brain.
    A video that summarizes the experiment in Figs. 5b-f. 3D Gaussian scanning recorded positions of GCaMP6f+ neurites around ellipsoid body in a 190 μm ×95 μm×60 μm volume, whereas Bessel scanning allowed us to image at 2.5 Hz their activity during the presentation of 12 different drifting grating stimuli (ten-trial averaged).
  7. Volumetric functional imaging of a densely labeled fruit fly brain.
    Video 7: Volumetric functional imaging of a densely labeled fruit fly brain.
    A video that summarizes the experiment in Figs. 5g-i. 3D Gaussian scanning recorded fluorescence distribution in a fly brain with pan-neuronal expression of GCaMP6f over a 307 μm × 216 μm × 100 μm volume, whereas Bessel scanning allowed us to image at 3.6 Hz the calcium transients (ten-trial averaged) during the presentation of 12 different drifting grating stimuli. Calcium transients over mushroom bodies were evoked by both the appearance of new stationary gratings and the drifting movement of the gratings.
  8. Functional imaging of a densely labeled fruit fly brain in an optical section through the mushroom bodies.
    Video 8: Functional imaging of a densely labeled fruit fly brain in an optical section through the mushroom bodies.
    A video that summarizes the experiment in Figs. 5j-k. Scanning a Gaussian focus across an optical section that includes the mushroom bodies, we observed in mushroom bodies calcium transients evoked by both the appearance of new stationary gratings and the drifting movement of the gratings.

References

  1. Denk, W., Strickler, J.H. & Webb, W.W. Two-photon laser scanning fluorescence microscopy. Science 248, 7376 (1990).
  2. Welford, W.T. Use of annular apertures to increase focal depth. J. Opt. Soc. Am. 50, 749753 (1960).
  3. Botcherby, E.J., Juškaitis, R. & Wilson, T. Scanning two photon fluorescence microscopy with extended depth of field. Opt. Commun. 268, 253260 (2006).
  4. Thériault, G., De Koninck, Y. & McCarthy, N. Extended depth of field microscopy for rapid volumetric two-photon imaging. Opt. Express 21, 1009510104 (2013).
  5. Thériault, G., Cottet, M., Castonguay, A., McCarthy, N. & De Koninck, Y. Extended two-photon microscopy in live samples with Bessel beams: steadier focus, faster volume scans, and simpler stereoscopic imaging. Front. Cell. Neurosci. 8, 139 (2014).
  6. Dombeck, D.A., Khabbaz, A.N., Collman, F., Adelman, T.L. & Tank, D.W. Imaging large-scale neural activity with cellular resolution in awake, mobile mice. Neuron 56, 4357 (2007).
  7. Andermann, M.L., Kerlin, A.M. & Reid, R.C. Chronic cellular imaging of mouse visual cortex during operant behavior and passive viewing. Front. Cell. Neurosci. 4, 3 (2010).
  8. Thévenaz, P., Ruttimann, U.E. & Unser, M. A pyramid approach to subpixel registration based on intensity. IEEE Trans. Image Process. 7, 2741 (1998).
  9. Jia, H., Rochefort, N.L., Chen, X. & Konnerth, A. Dendritic organization of sensory input to cortical neurons in vivo. Nature 464, 13071312 (2010).
  10. Chen, X., Leischner, U., Rochefort, N.L., Nelken, I. & Konnerth, A. Functional mapping of single spines in cortical neurons in vivo. Nature 475, 501505 (2011).
  11. Chen, T.-W. et al. Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature 499, 295300 (2013).
  12. Wilson, D.E., Whitney, D.E., Scholl, B. & Fitzpatrick, D. Orientation selectivity and the functional clustering of synaptic inputs in primary visual cortex. Nat. Neurosci. 19, 10031009 (2016).
  13. Kubota, Y. Untangling GABAergic wiring in the cortical microcircuit. Curr. Opin. Neurobiol. 26, 714 (2014).
  14. Sohya, K., Kameyama, K., Yanagawa, Y., Obata, K. & Tsumoto, T. GABAergic neurons are less selective to stimulus orientation than excitatory neurons in layer II/III of visual cortex, as revealed by in vivo functional Ca2+ imaging in transgenic mice. J. Neurosci. 27, 21452149 (2007).
  15. Liu, B.H. et al. Visual receptive field structure of cortical inhibitory neurons revealed by two-photon imaging guided recording. J. Neurosci. 29, 1052010532 (2009).
  16. Kameyama, K. et al. Difference in binocularity and ocular dominance plasticity between GABAergic and excitatory cortical neurons. J. Neurosci. 30, 15511559 (2010).
  17. Runyan, C.A. et al. Response features of parvalbumin-expressing interneurons suggest precise roles for subtypes of inhibition in visual cortex. Neuron 67, 847857 (2010).
  18. Taniguchi, H. et al. A resource of Cre driver lines for genetic targeting of GABAergic neurons in cerebral cortex. Neuron 71, 9951013 (2011).
  19. Polack, P.-O., Friedman, J. & Golshani, P. Cellular mechanisms of brain state-dependent gain modulation in visual cortex. Nat. Neurosci. 16, 13311339 (2013).
  20. Fu, Y. et al. A cortical circuit for gain control by behavioral state. Cell 156, 11391152 (2014).
  21. Kerlin, A.M., Andermann, M.L., Berezovskii, V.K. & Reid, R.C. Broadly tuned response properties of diverse inhibitory neuron subtypes in mouse visual cortex. Neuron 67, 858871 (2010).
  22. McGinley, M.J. et al. Waking state: rapid variations modulate neural and behavioral responses. Neuron 87, 11431161 (2015).
  23. Göbel, W. & Helmchen, F. In vivo calcium imaging of neural network function. Physiology (Bethesda) 22, 358365 (2007).
  24. Reimer, J. et al. Pupil fluctuations track fast switching of cortical states during quiet wakefulness. Neuron 84, 355362 (2014).
  25. Hestrin, S. & Galarreta, M. Electrical synapses define networks of neocortical GABAergic neurons. Trends Neurosci. 28, 304309 (2005).
  26. Pakan, J.M.P. et al. Behavioral-state modulation of inhibition is context-dependent and cell type specific in mouse visual cortex. eLife 5, e14985 (2016).
  27. Hu, H., Cavendish, J.Z. & Agmon, A. Not all that glitters is gold: off-target recombination in the somatostatin-IRES-Cre mouse line labels a subset of fast-spiking interneurons. Front. Neural Circuits 7, 195 (2013).
  28. Pfeffer, C.K., Xue, M., He, M., Huang, Z.J. & Scanziani, M. Inhibition of inhibition in visual cortex: the logic of connections between molecularly distinct interneurons. Nat. Neurosci. 16, 10681076 (2013).
  29. Ahrens, M.B., Orger, M.B., Robson, D.N., Li, J.M. & Keller, P.J. Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nat. Methods 10, 413420 (2013).
  30. Panier, T. et al. Fast functional imaging of multiple brain regions in intact zebrafish larvae using selective plane illumination microscopy. Front. Neural Circuits 7, 65 (2013).
  31. Bouchard, M.B. et al. Swept confocally-aligned planar excitation (SCAPE) microscopy for high speed volumetric imaging of behaving organisms. Nat. Photonics 9, 113119 (2015).
  32. Kimmel, C.B., Powell, S.L. & Metcalfe, W.K. Brain neurons which project to the spinal cord in young larvae of the zebrafish. J. Comp. Neurol. 205, 112127 (1982).
  33. O'Malley, D.M., Kao, Y.-H. & Fetcho, J.R. Imaging the functional organization of zebrafish hindbrain segments during escape behaviors. Neuron 17, 11451155 (1996).
  34. Severi, K.E. et al. Neural control and modulation of swimming speed in the larval zebrafish. Neuron 83, 692707 (2014).
  35. Gahtan, E., Sankrithi, N., Campos, J.B. & O'Malley, D.M. Evidence for a widespread brain stem escape network in larval zebrafish. J. Neurophysiol. 87, 608614 (2002).
  36. Gahtan, E. & O'Malley, D.M. Rapid lesioning of large numbers of identified vertebrate neurons: applications in zebrafish. J. Neurosci. Methods 108, 97110 (2001).
  37. Lister, J.A., Robertson, C.P., Lepage, T., Johnson, S.L. & Raible, D.W. nacre encodes a zebrafish microphthalmia-related protein that regulates neural-crest-derived pigment cell fate. Development 126, 37573767 (1999).
  38. Orger, M.B., Kampff, A.R., Severi, K.E., Bollmann, J.H. & Engert, F. Control of visually guided behavior by distinct populations of spinal projection neurons. Nat. Neurosci. 11, 327333 (2008).
  39. Seelig, J.D. et al. Two-photon calcium imaging from head-fixed Drosophila during optomotor walking behavior. Nat. Methods 7, 535540 (2010).
  40. Weir, P.T. & Dickinson, M.H. Functional divisions for visual processing in the central brain of flying Drosophila. Proc. Natl. Acad. Sci. USA 112, E5523E5532 (2015).
  41. Jenett, A. et al. A GAL4-driver line resource for Drosophila neurobiology. Cell Rep. 2, 9911001 (2012).
  42. Vogt, K. et al. Direct neural pathways convey distinct visual information to Drosophila mushroom bodies. eLife 5, e14009 (2016).
  43. Song, A. et al. Volumetric two-photon imaging of neurons using stereoscopy (vTwINS). Preprint at http://www.biorxiv.org/content/early/2016/09/06/073742.
  44. Stamnes, J.J., Heier, H. & Ljunggren, S. Encircled energy for systems with centrally obscured circular pupils. Appl. Opt. 21, 16281633 (1982).
  45. Ji, N., Magee, J.C. & Betzig, E. High-speed, low-photodamage nonlinear imaging using passive pulse splitters. Nat. Methods 5, 197202 (2008).
  46. Podgorski, K. & Ranganathan, G. Brain heating induced by near-infrared lasers during multiphoton microscopy. J. Neurophysiol. 116, 10121023 (2016).
  47. Yang, W. et al. Simultaneous multi-plane imaging of neural circuits. Neuron 89, 269284 (2016).
  48. Pnevmatikakis, E.A. et al. Simultaneous denoising, deconvolution, and demixing of calcium imaging data. Neuron 89, 285299 (2016).
  49. Ji, N., Freeman, J. & Smith, S.L. Technologies for imaging neural activity in large volumes. Nat. Neurosci. 19, 11541164 (2016).
  50. Pologruto, T.A., Sabatini, B.L. & Svoboda, K. ScanImage: flexible software for operating laser scanning microscopes. Biomed. Eng. Online 2, 13 (2003).
  51. Goodman, J.W. Introduction to Fourier Optics (Roberts & Company, 2005).
  52. Sun, W., Tan, Z., Mensh, B.D. & Ji, N. Thalamus provides layer 4 of primary visual cortex with orientation- and direction-tuned inputs. Nat. Neurosci. 19, 308315 (2016).
  53. Drapeau, P., Ali, D.W., Buss, R.R. & Saint-Amant, L. In vivo recording from identifiable neurons of the locomotor network in the developing zebrafish. J. Neurosci. Methods 88, 113 (1999).
  54. Tan, Z., Sun, W., Chen, T.-W., Kim, D. & Ji, N. Neuronal representation of ultraviolet visual stimuli in mouse primary visual cortex. Sci. Rep. 5, 12597 (2015).
  55. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676682 (2012).
  56. Carandini, M. & Ferster, D. Membrane potential and firing rate in cat primary visual cortex. J. Neurosci. 20, 470484 (2000).

Download references

Author information

Affiliations

  1. Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, Virginia, USA.

    • Rongwen Lu,
    • Wenzhi Sun,
    • Yajie Liang,
    • Aaron Kerlin,
    • Johannes D Seelig,
    • Boaz Mohar,
    • Masashi Tanimoto,
    • Minoru Koyama &
    • Na Ji
  2. Champalimaud Neuroscience Programme, Champalimaud Centre for the Unknown, Lisbon, Portugal.

    • Jens Bierfeld &
    • Michael B Orger
  3. Center of Advanced European Studies and Research, Bonn, Germany.

    • Johannes D Seelig
  4. Department of Functional Architecture and Development of Cerebral Cortex, Max Planck Florida Institute for Neuroscience, Jupiter, Florida, USA.

    • Daniel E Wilson,
    • Benjamin Scholl &
    • David Fitzpatrick
  5. Division of Biological Science, Graduate School of Science, Nagoya University, Nagoya, Japan.

    • Masashi Tanimoto

Contributions

N.J. conceived and oversaw the project; R.L. and N.J. designed the Bessel modules; R.L. and N.J. built Bessel modules with help from A.K., W.S., B.M., D.E.W. and B.S.; W.S. and Y.L. prepared mouse samples; R.L., W.S., Y.L., A.K., B.M. and N.J. collected mouse data; M.K. and M.B.O. designed zebrafish experiments; J.B. prepared zebrafish samples with help from M.T.; R.L., A.K., J.B. and N.J. collected zebrafish data; J.D.S., R.L. and N.J. designed Drosophila experiments; J.D.S. prepared Drosophila samples; R.L., J.D.S. and N.J. collected Drosophila data; D.F. and N.J. designed ferret experiments; and D.E.W. and B.S. prepared ferret samples and collected ferret data. All authors contributed to data analysis and presentation. R.L. and N.J. wrote the paper with inputs from all authors.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Optimized Bessel foci for in vivo volume imaging. (702 KB)

    (a) Images taken by scanning Bessel foci of various NAs, lateral and axial FWHMs: (Left panels) in vivo volume images of YFP+ neurites in cortex of an awake mouse (Thy1-YFP line H); scale bar: 10 μm. (Right panels) Axial point spread functions measured from 0.2-μm-diameter beads; x scale bar: 5 μm; z scale bar: 25 μm. (b) Mean intensity projection of the same neurites (color-coded by depth below pia) imaged by scanning a 1.05-NA Gaussian focus in 3D. Higher NA Bessel foci have stronger side rings, resulting in hazy backgrounds. Objective: Olympus 25×, 1.05 NA.

  2. Supplementary Figure 2: Axonal varicosities and dendritic spines are resolvable in single frames of 30-Hz Bessel volume scanning. (1,601 KB)

    (a) Individual raw image frames during 30-Hz volumetric measurements of calcium transients in a volume extending 60 μm in z. Arrows and numbers label individual axonal varicosities (putative boutons, 1,2) and dendritic spines (3-10) (same as in Fig. 1d), with their zoomed-in view shown in the panels on the right. (b) As comparison, the same synaptic structures in the average of 5 frames (i.e., effective volume rate of 6 Hz). Scale bars: left panels, 20 μm; right panels, 5 μm.

  3. Supplementary Figure 3: Bessel focus-scanning is robust against axial motion artifacts. (684 KB)

    (a) Mean axial projection of a 3D image stack (180 μm × 180 μm × 160 μm, 100 2D images) collected by Gaussian focus scanning, of GCaMP6f+ dendrites (0-160 μm below pia) in awake mouse S1 (structures color-coded by depth). (b) Averaged image of a 266-sec time series of the same dendrites imaged with a Bessel focus (0.4 NA and 91 μm axial FWHM) at 30 Hz. (c) Averaged image after registering the time series using TurboReg plugin of ImageJ. (d) Mean axial projection of a 3D image stack (80 μm × 40 μm × 25 μm, 25 2D images) collected by Gaussian focus scanning of YFP+ dendrites (46-70 μm below pia) in awake mouse V1 (Thy1-YFP line H, structures color-coded by depth). (e) Averaged image of a 100-sec 2D-registered time series of dendrites at 53 μm below pia acquired with Gaussian focus scanning at 7.4 Hz. (f) Averaged image of a 100-sec 2D-registered time series of the volume of dendrites in d imaged with a Bessel focus (0.4 NA and 19 μm axial FWHM) at 7.4 Hz. (g,h) Brain motion (upper panels, quantified as the lateral image displacement with time) causes (g) large changes of fluorescence signal from two YFP+ dendrites (ROI1 and ROI2) in Gaussian focus scanning mode, (h) but not Bessel focus scanning mode. Objective: Olympus 25×, 1.05 NA; Post-objective power: (a) 30 mW, (b,c) 103 mW, (d,e) 9 mW, (f) 21 mW; Wavelength: (a-c) 960 nm, (d-f) 940nm. Scale bars: 20 μm.

  4. Supplementary Figure 4: In vivo characterization of dendritic spines in ferret V1 at 30-Hz volume rate. (854 KB)

    (a,b) Two examples of applying Bessel focus scanning to the characterization of spines in ferret V1. (Left) Mean intensity projection of 3D image stacks by conventional Gaussian focus imaging in V1 of anesthetized ferrets; (Right) Images collected by scanning a Bessel beam with 0.4 NA and 50 μm axial FWHM. Yellow circles highlight example spines. (c) 10-trial-averaged calcium transients of spines in a and b for 12 grating angles. (d) Tuning curves of individual spines in a and b. Objective: Nikon 16×, NA 0.8; Post-objective power: 55 mW for Gaussian scanning and 84 mW for Bessel scanning; Wavelength: 960 nm; Shadow (c) and error bars (d): s.e.m; Scale bars: 10 μm. Representative images from 2 ferrets.

  5. Supplementary Figure 5: Removal of neuropil contamination in Bessel focus scanning mode. (250 KB)

    (a) (left) Image of a cell body obtained by scanning a Bessel beam with 0.4 NA and 78 μm axial FWHM, (right) overlaid with region of interest (ROI, orange outline) and the area used for calculating neuropil background (red region). (b) The difference between the averaged signal within ROI (F1) and the averaged signal within neuropil mask (Fneuropil) is used to calculate calcium transient of the neuron. After the removal of neuropil contamination, the cell body still exhibited activity correlated with pupil size. (c) and (d) same as a and b, except that the removal of neuropil contamination was applied to a volume of neuropil, resulting in minimal calcium transients. All fluorescence traces were plotted on the same scale.

  6. Supplementary Figure 6: Activity synchrony among GABAergic neurons persists in the dark. (255 KB)

    (a) Representative image of GAD2+ neurons expressing GCaMP6s (0-100 μm below pia) collected by scanning a Bessel beam with 0.4 NA and 78 μm axial FWHM. (b) Pupil size (red) and concurrently measured calcium transients of example neurons labeled in a. (c) Pearson’s correlation coefficient (R) between neuronal calcium transients and pupil size (1st row and 1st column, sorted by decreasing R), as well as between the calcium transients of all pairs of neurons identified in a. Objective: Olympus 25×, 1.05 NA; Post-objective power: 21 mW; Wavelength: 960 nm; Scale bar: 50 μm.

  7. Supplementary Figure 7: Activity synchrony among GABAergic neurons was also observed in conventional Gaussian focus scanning mode. (402 KB)

    (a) GABAergic neurons at three different depths (55, 73, and 93 μm) imaged by scanning a conventional Gaussian focus (1.05 NA, 1.4 μm axial FWHM). (b) Pupil size (red) and concurrently measured calcium transients of example neurons labeled in a. (c) Pearson’s correlation coefficient (R) between neuronal calcium transients and pupil size (1st row and 1st column, sorted by decreasing R), as well as between the calcium transients of all pairs of neurons identified in a. (d) Histogram distribution and (e) scatter plot of R measured from a mouse. Red open histogram: R between pupil size and neuronal calcium transients; Gray filled histogram: R between neurons; Dashed lines: median; n: number of R’s. Objective: Olympus 25×, 1.05 NA; Post-objective power: 18 mW for 55 and 93 μm depth and 14 mW for 73 μm depth; Wavelength: 960 nm; Scale bar: 50 μm.

  8. Supplementary Figure 8: Engineering Bessel foci with asymmetric axial-intensity distribution with an SLM. (499 KB)

    (a) Overlaying a binary concentric ring pattern on a quadratic phase pattern on the SLM allows the shaping of axial intensity distribution of the Bessel focus. S: period of the concentric rings in units of pixels. (b) Axial images and corresponding intensity profiles of a 2-μm-diameter bead using three different Bessel foci generated with periods S1, S2, S3. (c,d) (From left to right) in vivo images of neurons and neurites measured in V1 of awake mice using Gaussian focus, Bessel foci generated with S1, S2, and S3, respectively. Gaussian images are mean intensity projections with features color-coded by depth. Arrows point to deeper structures that had best image quality when the Bessel focus of increasing intensity with depth (S3 in b) was used. Objective: Olympus 25×, 1.05 NA; Post-objective power: c, 14 mW for Gaussian modality and 32 mW for Bessel modality; d, 21 mW for Gaussian modality and 61 mW for Bessel modality; Wavelength: 940 nm; Scale bars: 20 μm.

  9. Supplementary Figure 9: Annular apodization mask fine-tunes the axial length and intensity distribution of Bessel foci. (520 KB)

    (a-c) and (d-f): two NA-0.4 Bessel foci with different axial lengths and intensity distributions. (a,d) Binary concentric grating patterns (phase values: 0 and π) on SLM. S indicates the period of the gratings in units of pixels. (b,e) (Top) amplitude and (bottom) phase of the electric field of the light at the plane of the annular apodization mask (0.39 mm inner radius and 0.6 mm outer radius for b, 0.478 mm inner radius and 0.6 mm outer radius for e). Electric field within the annulus (areas between red dashed lines) was transmitted. (c,f) (Top) axial (xz) images of 2-μm-diameter fluorescent beads and (bottom) line intensity profiles along the z axis obtained with Bessel foci generated by a and b, d and e, respectively. (g) Simulated axial PSFs and (h) their intensity profiles along z as a function of annulus width of the apodization mask. Incidence beam diameter at SLM: 2.2 mm. Magnification from mask to objective back pupil: 4.7; Objective: Olympus 25×, 1.05 NA; Post-objective power: 14.2 mW and 15.5 mW for c and f, respectively; Wavelength: 900 nm; Scale bar: 10 μm. See Supplementary Technical Notes for more details.

Video

  1. Video 1: 30 Hz Bessel volumetric imaging of calcium transients in sparsely labeled neurons in V1 of awake mouse. (17.4 MB, Download)
    GCaMP6s+ axons and dendrites in a 270 μm ×270 μm×60 μm volume (0-60 μm below pia) imaged at 30 Hz using Bessel focus scanning. Calcium transients were driven by drifting grating stimulation. Structures that appeared blurred were under blood vessels.
  2. Video 2: Robustness of Bessel scanning against axial brain motion allows easy registration of volume images. (44.26 MB, Download)
    Unregistered (left panel, note large lateral motion) and registered (right panel) time-lapse Bessel scanning images of GCaMP6f+ dendritic branches in S1 of an awake mouse. 180 μm × 180 μm × 160 μm volume was imaged at 30 Hz rate. Registration was done with TurboReg plugin of ImageJ.
  3. Video 3: Bessel focus scanning is more robust against axial motion artifacts than Gaussian focus scanning. (46.54 MB, Download)
    Unregistered (left panels, note large lateral motion) and registered (right panels) time-lapse Gaussian (top panels) and Bessel (bottom panels) focus scanning images of YFP+ dendritic branches in V1 of an awake mouse. Brain motion caused large focal shifts in Gaussian images, whereas dendrites imaged under Bessel mode remained in focus despite brain motion. Frame rate: 7.4 Hz. Bessel image volume: 80 μm × 40 μm × 25 μm. Registration was done with TurboReg plugin of ImageJ.
  4. Video 4: Video summary of high-throughput volumetric functional mapping of single spines in mouse V1 (11.79 MB, Download)
    A video that summarizes the experiment in Fig. 2 where Bessel scanning was used for high-throughput volumetric functional mapping of single spines. 3D Gaussian scanning revealed single spines on GCaMP6s+ dendrites 100-160 μm below pia in mouse V1, whereas Bessel scanning allowed us to image all these spines in one 2D scan and measure their visually-evoked calcium transients (ten-trial averaged) to 12 different drifting grating stimuli.
  5. Video 5: Video summary of 30 Hz volumetric functional imaging of spinal projection neurons in zebrafish larvae. (16.48 MB, Download)
    A video that summarizes the experiment in Fig. 4 where Bessel scanning was used for functional imaging of spinal projection neurons in zebrafish larvae at 30 Hz. 3D Gaussian scanning recorded positions of spinal projection neurons in a 270 μm ×270 μm×100 μm volume, whereas Bessel scanning allowed us to image at 30 Hz their activity evoked by a mechanical and acoustic stimulus at 5-second mark.
  6. Video 6: Volumetric functional imaging of a sparsely labeled fruit fly brain. (6.31 MB, Download)
    A video that summarizes the experiment in Figs. 5b-f. 3D Gaussian scanning recorded positions of GCaMP6f+ neurites around ellipsoid body in a 190 μm ×95 μm×60 μm volume, whereas Bessel scanning allowed us to image at 2.5 Hz their activity during the presentation of 12 different drifting grating stimuli (ten-trial averaged).
  7. Video 7: Volumetric functional imaging of a densely labeled fruit fly brain. (5.38 MB, Download)
    A video that summarizes the experiment in Figs. 5g-i. 3D Gaussian scanning recorded fluorescence distribution in a fly brain with pan-neuronal expression of GCaMP6f over a 307 μm × 216 μm × 100 μm volume, whereas Bessel scanning allowed us to image at 3.6 Hz the calcium transients (ten-trial averaged) during the presentation of 12 different drifting grating stimuli. Calcium transients over mushroom bodies were evoked by both the appearance of new stationary gratings and the drifting movement of the gratings.
  8. Video 8: Functional imaging of a densely labeled fruit fly brain in an optical section through the mushroom bodies. (3.34 MB, Download)
    A video that summarizes the experiment in Figs. 5j-k. Scanning a Gaussian focus across an optical section that includes the mushroom bodies, we observed in mushroom bodies calcium transients evoked by both the appearance of new stationary gratings and the drifting movement of the gratings.

PDF files

  1. Supplementary Text and Figures (2,540 KB)

    Supplementary Figures 1–9 and Supplementary Tables 1–4

  2. Supplementary Methods Checklist (408 KB)
  3. Supplementary Technical Note (581 KB)

    Supplementary Technical Note

Zip files

  1. Supplementary Software (12 KB)

    Supplementary Software

Additional data