Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Nov;158(11):2108-2116.
doi: 10.1097/j.pain.0000000000000968.

Fully Implantable, Battery-Free Wireless Optoelectronic Devices for Spinal Optogenetics

Free PMC article

Fully Implantable, Battery-Free Wireless Optoelectronic Devices for Spinal Optogenetics

Vijay K Samineni et al. Pain. .
Free PMC article


The advent of optogenetic tools has allowed unprecedented insights into the organization of neuronal networks. Although recently developed technologies have enabled implementation of optogenetics for studies of brain function in freely moving, untethered animals, wireless powering and device durability pose challenges in studies of spinal cord circuits where dynamic, multidimensional motions against hard and soft surrounding tissues can lead to device degradation. We demonstrate here a fully implantable optoelectronic device powered by near-field wireless communication technology, with a thin and flexible open architecture that provides excellent mechanical durability, robust sealing against biofluid penetration and fidelity in wireless activation, thereby allowing for long-term optical stimulation of the spinal cord without constraint on the natural behaviors of the animals. The system consists of a double-layer, rectangular-shaped magnetic coil antenna connected to a microscale inorganic light-emitting diode (μ-ILED) on a thin, flexible probe that can be implanted just above the dura of the mouse spinal cord for effective stimulation of light-sensitive proteins expressed in neurons in the dorsal horn. Wireless optogenetic activation of TRPV1-ChR2 afferents with spinal μ-ILEDs causes nocifensive behaviors and robust real-time place aversion with sustained operation in animals over periods of several weeks to months. The relatively low-cost electronics required for control of the systems, together with the biocompatibility and robust operation of these devices will allow broad application of optogenetics in future studies of spinal circuits, as well as various peripheral targets, in awake, freely moving and untethered animals, where existing approaches have limited utility.


Figure 1
Figure 1. Overall design of the flexible wireless optoelectronic system and the anatomy of the system on the spinal cord
(A) Schematic illustration of the overall system design with open architecture and µ-ILED positioned at the end of the probe stem. (B) Diagram demonstrating the anatomy and location of the optoelectronic system as implanted above the spinal cord. (C) Images of intermittent steps during implantation of the optoelectronic system in the mouse before (left) and after (right) the implantation. (D) Representative image of a device in the ‘ON’ state while resting atop an index finger. (E) Representative image of an awake, freely moving mouse with the optoelectronic system implanted on the spinal cord.
Figure 2
Figure 2. Consistency of power transmission throughout a behavior arena and with mechanical deformation, thermal management, and in vivo robustness of the wireless optoelectronic systems
(A) Image of wireless operation of 12 devices in each arm of a V-maze preference arena with the double loop antennas. (B) Simulation (top) and experimental measurement (bottom) of relative output voltage distribution for 27 devices distributed at heights of 2.5 cm, 5 cm, and 7.5 cm from the arena floor. (C) The distribution of the effective strain in the copper coils for the radius of curvatures R = 5 mm. The inset shows a picture of a bent device with a bending radius of 5 mm. (D) The Q factor and resonance frequency obtained by EM simulation for deformed devices with different radii of curvature, from planar to 5 mm. (E) Infrared image of the device surface before and during continuous wireless operation with a power of 50 mW/mm2. (F) Temperature change of µ-ILEDs as a function of pulse width of operation at different output powers of 10, 20, and 50 mW/mm2. (Lines: calculated theoretical values, Symbols: measured experimental values). (G) Blue light attenuation through longitudinal spinal cord slices (250–1000 µm) at 10, 20, and 50 mW/mm2. (H) Mean (± SEM) days of in vivo device survival when encapsulated with a monolayer of PDMS alone (white column) or a bilayer of PIB and PDMS (blue column).
Figure 3
Figure 3. Anatomical and histological characterization of the spinal cord in TRPV1-ChR2 mice
Transverse sections of spinal cord from TRPV1-ChR2 mice at low (A) and high (B) magnification showing ChR2+ (green) terminals in the dorsal horn overlapping with CGRP (red) in lamina I and IB4 (magenta) in lamina II. Low magnification and high magnification confocal images of a transverse section of the lumbar spinal cord from a mouse with spinal nerve ligation. Scale bar is 350 µm for low magnification images and 125 µm for high magnification images. (C, D) and from a mouse implanted with the optoelectronic device (E, F) for three weeks. DAPI (blue), GFAP (green) marks reactive astrocytes and Iba1 (red) marks activated microglia. Scale bar is 125 µm for low magnification images and 80 µm for high magnification images.
Figure 4
Figure 4. Wireless optical activation of ChR2+ TRPV1 spinal afferents induces pain behaviors and place aversion
(A) Implantation of the spinal optoelectronic device has no effect on motor behavior vs. sham animals in the rotarod test (P = 0.64, n = 6 sham, n = 6 device). (B) Mice with spinal implants did not exhibit any difference in total distance travelled in open field test compared to sham mice (P = 0.56, n = 5 sham, n = 5 device). (C) Implanted mice did not exhibit any difference in time spent in center compared to sham implanted mice in open field test (P = 0.71, n = 5 sham, n = 5 device). (D) Wireless activation of the spinal µ-ILED causes increased nocifensive behaviors (flinching, hind paw licking, jumping and vocalization) in TRPV1-ChR2 mice but not in control (88 vs. 0.25 flinches, P < 0.0001 vs. without illumination n = 6 per group). No other statistical comparisons reach significance. (E) Representative heat maps displaying time spent in a V-maze over a 20 min trial with optical stimulation in the “LED On” arm only for control (top) and TRPV1-ChR2 mice (bottom). (F) TRPV1-ChR2 mice spent significantly less time in stimulation-paired chamber compared with the nonstimulation-paired chamber (54.7 vs. 882.2 s; P < 0.0001, n = 6), whereas control mice did not show any preference (504 vs. 548.3 s; P = 0.665, n = 6). Data are presented as mean ± SEM. Statistical comparisons were made using two-way ANOVA. ***P < 0.0001.

Similar articles

See all similar articles

Cited by 10 articles

See all "Cited by" articles

MeSH terms