of 26
Implantable photonic neural probes for light-sheet
fluorescence brain imaging
Wesley D. Sacher,
a,b,c,d,
Fu-Der Chen,
c,
Homeira Moradi-Chameh,
e,
Xianshu Luo ,
f,
Anton Fomenko ,
e
Prajay T. Shah ,
e
Thomas Lordello,
c
Xinyu Liu,
a
Ilan Felts Almog,
c
John N. Straguzzi,
d
Trevor M. Fowler,
a
Youngho Jung,
c,d
Ting Hu,
g
Junho Jeong,
c
Andres M. Lozano ,
e,h
Patrick Guo-Qiang Lo,
f
Taufik A. Valiante,
c,e,h,i
Laurent C. Moreaux,
a
Joyce K. S. Poon ,
c,d
and Michael L. Roukes
a,b,
a
California Institute of Technology, Division of Physics, Mathematics, and Astronomy,
Pasadena, California, United States
b
Kavli Nanoscience Institute, California Institute of Technology, Pasadena, California,
United States
c
University of Toronto, Department of Electrical and Computer Engineering, Toronto,
Ontario, Canada
d
Max Planck Institute of Microstructure Physics, Halle, Germany
e
University Health Network, Krembil Research Institute, Division of Clinical and
Computational Neuroscience, Toronto, Ontario, Canada
f
Advanced Micro Foundry Pte. Ltd., Singapore
g
Agency for Science Technology and Research (A*STAR), Institute of Microelectronics,
Singapore
h
University of Toronto, Toronto Western Hospital, Division of Neurosurgery,
Department of Surgery, Toronto, Ontario, Canada
i
University of Toronto, Institute of Biomaterials and Biomedical Engineering, Toronto,
Ontario, Canada
Abstract
Significance:
Light-sheet fluorescence microscopy (LSFM) is a powerful technique for high-
speed volumetric functional imaging. However, in typical light-sheet microscopes, the illumi-
nation and collection optics impose significant constraints upon the imaging of non-transparent
brain tissues. We demonstrate that these constraints can be surmounted using a new class of
implantable photonic neural probes.
Aim:
Mass manufacturable, silicon-based light-sheet photonic neural probes can generate planar
patterned illumination at arbitrary depths in brain tissues without any additional micro-optic
components.
Approach:
We develop implantable photonic neural probes that generate light sheets in tissue.
The probes were fabricated in a photonics foundry on 200-mm-diameter silicon wafers. The light
sheets were characterized in fluorescein and in free space. The probe-enabled imaging approach
was tested in fixed,
in vitro
, and
in vivo
mouse brain tissues. Imaging tests were also performed
using fluorescent beads suspended in agarose.
Results:
The probes had 5 to 10 addressable sheets and average sheet thicknesses <
16
μ
m
for
propagation distances up to
300
μ
m
in free space. Imaging areas were as large as
240
μ
m
×
490
μ
m
in brain tissue. Image contrast was enhanced relative to epifluorescence microscopy.
Conclusions:
The neural probes can lead to new variants of LSFM for deep brain imaging and
experiments in freely moving animals.
©
The Authors. Published by SPIE under a Creative Commons Attribution 4.0 Unported License.
Distribution or reproduction of this work in whole or in part requires full attribution of the original pub-
lication, including its DOI.
[DOI:
10.1117/1.NPh.8.2.025003
]
Address all correspondence to Wesley D. Sacher,
wesley.sacher@mpi-halle.mpg.de
; Michael L. Roukes,
roukes@caltech.edu
Equal contribution
Neurophotonics
025003-1
Apr
Jun 2021
Vol. 8(2)
Downloaded From: https://www.spiedigitallibrary.org/journals/Neurophotonics on 23 Apr 2021
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
Keywords:
neurophotonics; integrated optics; functional imaging; microscopy; biophotonics;
light-sheet fluorescence microscopy.
Paper 20060R received Aug. 12, 2020; accepted for publication Mar. 4, 2021; published online
Apr. 19, 2021.
1 Introduction
New methods in optogenetics
1
3
and, especially, the advent of fluorescent reporters of neuronal
activity, have opened many novel approaches for actuating and recording neural activity
en
masse
, through the use of powerful free-space single-photon and multi-photon microscopy
methods.
4
8
However, existing approaches to functional imaging of the brain have significant
limitations. Single-photon (1P) epifluorescence imaging readily lends itself to high frame-rate
wide-field microscopy, but, in its simplest implementations, image contrast is hampered by
out-of-focus background fluorescence, and the depth of imaging is restricted by the optical
attenuation in the tissue. Confocal imaging improves the contrast by optical sectioning, and
out-of-focus light is rejected using a pinhole; however, a laser beam must be scanned across
each point of the tissue and this significantly slows the image acquisition rate.
9
Multi-photon
microscopy is also inherently a point or line scanning method, but because it uses infrared exci-
tation (which provides a longer optical attenuation length
5
), the imaging depth in brain tissue can
be extended to
1mm
and the focus of the light beam can be rastered in three-dimensions to
achieve volumetric imaging.
5
,
10
12
Light-sheet fluorescence microscopy (LSFM), which is also known as selective-plane illu-
mination microscopy, combines the benefits of fast wide-field imaging, volumetric imaging, and
optical sectioning.
13
In conventional LSFM, a thin sheet of excitation light is generated either by
cylindrical focusing elements or digitally scanning a Gaussian or Bessel beam.
14
16
The sheet is
translated in one dimension across the sample; the fluorescence images are then sequentially
collected in the direction perpendicular to the illumination plane to form a volumetric
image.
17
With digitally scanned two-photon (2P) LSFM, it is also possible to increase the optical
penetration depth.
16
Non-digitally scanned 1P-LSFM is inherently faster than point- or line-scan
methods; and since the illumination is restricted to a plane, photobleaching, phototoxicity, and
out-of-focus background fluorescence are reduced compared to epifluorescence microscopy.
However, conventional LSFM requires two orthogonal objective lenses, and appropriately
positioning these largely limits the imaging modality to quasi-transparent organisms (e.g.,
C. elegans
,
Drosophila
embryos, and larval zebrafish), chemically cleared mammalian brains,
17
and brain slices.
18
An LSFM variant called swept confocally aligned planar excitation (SCAPE)
microscopy, which requires only a single objective, removes these constraints.
6
,
19
While
in vivo
calcium neural imaging has been demonstrated using SCAPE in mice,
6
miniaturization of the
system to be compatible with freely moving animal experiments remains challenging due to
the additional optics required.
To make LSFM compatible with non-transparent tissues such as mammalian brains and,
eventually, behavioral experiments with freely moving animals necessitates drastic miniaturiza-
tion of the light-sheet generation and fluorescence imaging compared to today
s archetypical
table-top systems. The feasibility of fluorescence microscopy in small and lightweight form
factors has already been established by way of head-mounted microscopes for 1P and 2P calcium
imaging in mice,
4
,
20
23
though the endoscopic implantation of the requisite gradient index
(GRIN) lenses, with typical diameters of 0.5 to 2 mm, displaces a significant amount of brain
tissue.
On the other hand, it remains a formidable and unsolved challenge to generate light sheets by
implantable elements at arbitrary brain depths, while minimizing tissue displacement and
remaining compatible with a sheet-normal imaging system. For example, in Ref.
24
, to generate
a light sheet perpendicular to the imaging GRIN lens required implantation of a millimeter-scale
prism coupled to a second GRIN lens. In another example, in Ref.
25
, a single light sheet was
produced from a microchip using a grating coupler (GC), a glass spacer block, and a metallic
slit lens. The overall device was
>
100-
μ
m
thick and
>
600-
μ
m
wide, which would displace
a significant amount of tissue upon implantation.
Sacher et al.: Implantable photonic neural probes for light-sheet fluorescence brain imaging
Neurophotonics
025003-2
Apr
Jun 2021
Vol. 8(2)
Downloaded From: https://www.spiedigitallibrary.org/journals/Neurophotonics on 23 Apr 2021
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
Here, we solve these challenges using wafer-scale nanophotonic technology to realize
implantable, silicon-based, light-sheet photonic neural probes that require no additional micro-
optics. They are fully compatible with free-space fluorescence imaging (light collection) outside
the brain, where the axis of collection is oriented perpendicular to the light sheets. These silicon
(Si) probes synthesize light sheets in tissue using sets of nanophotonic GCs integrated onto thin,
implantable, 3-mm-long Si shanks with 50 to
92
μ
m
thickness, widths that taper from 82 to
60
μ
m
along their length, and sharp tips at the distal ends. These prototype photonic neural
probes (Fig.
1
) are capable of generating and sequentially addressing up to five illumination
planes with a pitch of
70
μ
m
. Additionally, the form factor and illumination geometry of the
probes open an avenue toward their integration with GRIN lens endoscopes and miniature micro-
scopes, as shown conceptually in Fig.
2(b)
; offering a singular pathway to rapid, optically sec-
tioned functional imaging at arbitrary depths in the brain.
The probes were fabricated on 200-mm Si wafers in a Si photonics foundry for manufacturing
scalability and mass producibility. Elsewhere, we have used this technology to realize photonic
neural probes that emit dynamically reconfigurable, patterned light with cellular-scale beam
widths
27
and steerable beams without moving parts,
28
adding to a growing number of photonic
Fig. 1
Light-sheet photonic neural probes. (a) Illustration of the light-sheet synthesis method
(adapted from Ref.
26
). A series of simultaneously fed optical waveguides emits light via a row
of GCs designed for large divergences along the sheet-axis and small divergences along the
GC-axis. (b) Optical micrograph of a fabricated neural probe, (inset) scanning electron micrograph
of the tip of a shank. (c) Top-down schematics of the neural probe. (d) and (e) Annotated optical
micrographs of two neural probes with various GC rows emitting light sheets. (d) Neural probe
design with sheets generated from four shanks. (e) Probe design with sheets generated from two
shanks (
half-sheet design
). (f) Optical micrographs showing the routing network from the probe
in (d) guiding light for optical inputs to two different edge couplers. The images in (d) and (e) have
been contrast- and brightness-adjusted to enhance the visibility of the waveguides.
Sacher et al.: Implantable photonic neural probes for light-sheet fluorescence brain imaging
Neurophotonics
025003-3
Apr
Jun 2021
Vol. 8(2)
Downloaded From: https://www.spiedigitallibrary.org/journals/Neurophotonics on 23 Apr 2021
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use