Single-Cell Photothermal Neuromodulation for Functional Mapping of

5 days ago - 2 groups find new Parkinson's target. Two studies published earlier this month, one from the biotech start-up Yumanity Therapeutics and ...
0 downloads 0 Views 2MB Size
Subscriber access provided by LUNDS UNIV

Article

Single-Cell Photothermal Neuromodulation for Functional Mapping of Neural Networks Sangjin Yoo, Ji-Ho Park, and Yoonkey Nam ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b07277 • Publication Date (Web): 28 Dec 2018 Downloaded from http://pubs.acs.org on January 2, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Single-Cell Photothermal Neuromodulation for Functional Mapping of Neural Networks

Sangjin Yoo1, Ji-Ho Park1,2,*, and Yoonkey Nam1,2,*

1 Department of Bio and Brain Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea 2 KAIST Institute for Health Science and Technology, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea * Corresponding authors: [email protected] and [email protected]

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract

Photothermal neuromodulation is one of the emerging technologies being developed for neuroscience studies because it can provide minimal-invasive control of neural activity in the deep brain with submillimeter precision. However, single cell modulation without genetic modification still remains a challenge, hindering its path to the broad application. Here, we introduce a nanoplasmonic approach to inhibit single neural activity with high temporal resolution. Low intensity of near infrared (NIR) was focused at single cell size on gold nanorod (GNR) integrated microelectrode array platform, generating photothermal effect underneath a target neuron for photothermal stimulation. We found that the photothermal stimulation modulates the spontaneous activity of a target neuron in an inhibitory manner. Single neuron inhibition was fast and highly reliable without thermal damage, and it can induce changes in network firing patterns, potentially suggesting their application for in vivo circuit modulation and functional connectomes.

Keywords:

neuromodulation, photothermal effects, gold nanorod, local surface plasmonic

resonance, neuron

ACS Paragon Plus Environment

Page 2 of 24

Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Neuromodulation technique at single-cell resolution is invaluable for functional brain research because a single neural firing can drive rhythmic patterns of network activities that can switch the global state of the brain or can be translated into behavior.1-4 To realize precise neuromodulation, optical approaches have attracted great attention due to their excellent spatiotemporal accessibility. For example, the modulation of targeted neurons whose light-sensitive channels are genetically embedded on the plasma membrane has been conducted through a wide-field illumination,5 while the precise stimulation of dendrite or a single cell body has been achieved by optical scanning or sectioning technique.6-11 Although these modulation techniques are undoubtedly powerful for neuroscience studies or interrogating the mechanism of neurological disorders,12,13 they require the genetic modification of cell to express the exogenic actuators, and light scattering inside of tissue can cause off-target modulation. One potential alternative is the direct infrared (IR) stimulation to unmodified neurons, in which transient heating of the local medium would give rise to membrane capacitive current, followed by neuron excitation.14 While the IR pulses can excite a small cell population in deep tissues, careful determination of parameters is needed to elicit reliable responses without thermal damage.15,16 Nanomaterial-assisted photothermal stimulation has been highlighted as a promising neuromodulation technique due to its unique ability to absorb specific wavelengths of light and quickly convert the light energy into heat.17 The main advantage of this technique is that it can reduce the laser power required to excite neurons and can increase the stimulation selectivity.18, 19 Functional resolution and thermal damage of cell are the major issues in this technique due to the high-temperature requirement and slow heat diffusion. However, the detailed characterization and the damage threshold at a cellular level have not been investigated, which can be the biggest obstacle for further translational application.20 Recently, we developed a photothermal stimulation technique, wherein membranelocalized photothermal heating via gold nanorods (GNRs) and low-intensity near-infrared (NIR) light reliably inhibits neural network activities, without genetic modification of cells.21 And it was further demonstrated that the laser power required for the neural inhibition and blocking of axon conductance can be reduced by improving the GNR-neural interface.22 This capability, which is not available with the established methods such as optogenetics and pulsed IR, would provide an important complement

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

for studying brain connectivity and functions in both basic and clinical applications. Despite the surge of interest in this technique, the capability to modulate neural activity at the single cell level and thermal safety have yet to be explored, pose significant challenges to the broad use of this technology in neuroscience. Here, we investigate the potential of GNR-assisted photothermal stimulation technique to modulate a single neuron activity. Tight interaction between the neurons and GNRs were achieved by using a GNR integrated multielectrode array (MEA) chip22 and NIR was concentrated to the GNRs underneath a targeted neuron for the photothermal stimulation. We first sought to quantify the laser intensities required to induce the inhibition effect and thermal damage. Then, we aimed to investigate the spatio-temporal resolution of the single cell modulation by examining the contributions of focal area and NIR pulse width, as well as the change of network firing pattern caused by the single cell modulation.

Results and Discussion

To determine whether the spontaneous activity of a single neuron can be modulated by photothermal stimulation, a simple optical and electrical recording platform was designed on a conventional inverted microscope. Because an intimate contact of cells with GNRs is important for efficient photothermal stimulation, we introduced a NIR-sensitive GNR monolayer on the surface of multielectrode array (MEA) chip (GNR-MEA) and neurons were cultured on the top of the chip so that the neurons had direct contact with the GNRs.22 And GNR monolayer converted the focused NIR light into heat, which stimulated the target neuron by localized photothermal heating (Fig 1A). The aspect ratio of GNRs was optimized to absorb the NIR light (785 nm) and the surface of GNRs was coated by PEG polymer to avoid self-clustering and cellular uptake (Fig 1B). Figure 1C shows SEM image of a neuron and GNRs that GNRs were uniformly integrated onto the substrate (142.2±16.3 GNR/μm2, mean±SD, n=15) and neural cell body and neurites directly covered the GNR monolayer (Fig 1C). Collimated NIR light was guided into the microscope and focused through an objective to illuminate the GNRs underneath a target neuron (Fig. 1D). The diameter of NIR light focused on the GNR-

ACS Paragon Plus Environment

Page 4 of 24

Page 5 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

decorated MEA surface was approximately 20 μm, which was sufficient to cover a single neuron. Next, photothermal stimulation to a single neuron was performed by varying laser powers to characterize the neural response. To minimize off-target stimulation, hippocampal neurons were cultured on the GNR-MEA at low-density (100 cells/mm2), which allowed us to stimulate an isolated single neuron coupled with a microelectrode. We first identified a neuron whose spikes were directly sensed by a nearby microelectrode (Fig. 2A left). Photothermal stimulation to the target neuron at a laser power of 4 mW (12.8 W/mm2) showed a significant inhibitory effect on their spontaneous activity, and increasing the laser power up to 10 mW (32W/mm2) increased the inhibition (up to 90.8%, mean, Fig. 2A right and B). Although we used a low-density network model for this characterization, a few neurites were placed within the stimulation spot which could cause the inhibition by blocking synaptic inputs.22 But inhibition was not induced by the background stimulation, indicating the soma was the main source of spikes and synaptic input to the soma was insignificant in this model network. In a control experiment that did not use GNRs, NIR illumination at 20 mW (64 W/mm2) laser power, which was higher than the those used for photothermal stimulation, did not induce any significant change in the firing rates of single neurons, reflecting no modulation effects without GNRs (Fig S1). In addition to the photothermal inhibition of an isolated single cell, a small group of neurons near a microelectrode were selectively targeted for photothermal stimulation, allowing us to investigate which neurons contribute to generating spiking activity (Fig. 2C). Each neuron within 50 μm radius from a microelectrode was sequentially stimulated for 20 sec while the neural activity was recorded.23 We found that large spike units were selectively inhibited when a neuron positioned directly above the microelectrode was stimulated (cell number 3 in Fig. 2C), while the neural activities were not interrupted with the identical stimulation to other cells (Fig. 2D). This result demonstrates that this technique is able to modulate the activity of a single cell intertwined with each other in a network. Taken together, these results demonstrated that the single neuron activity can be precisely modulated by focused photothermal stimulation. As neurons can be damaged by high temperature induced by intensive laser illumination,16 we next investigated whether photothermal stimulation with higher energy can cause the cell damage.

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Spontaneous activity from a target neuron was monitored while the focused NIR light illuminated to GNRs underneath the target (10 sec duration) at the laser powers higher than those (4 – 10 mW) used for photothermal stimulation. Photothermal stimulation at 13 mW (41.6 W/mm2) clearly induced inhibition effects and the firing rate returned to its baseline level after turning off the laser. In contrast, the consecutive photothermal stimulation at 16 mW (51.2 W/mm2) resulted in an abrupt change of neural firing after 5 sec from the onset. Theses spiking behaviors correspond to the results from previous hyperthermia studies,24-26 implying that photothermal damage can be initiated by continuous stimulation with 16 mW (51.2 W/mm2). As expected, the neural firing was abolished at 19 mW (60.8 W/mm2), and was not restored after removing the illumination (Fig. 3A and B). An independent live-dead cell assay was performed to support these observations. Three groups of hippocampal neuronal cultures were prepared on GNR-MEA substrates, and each culture was subjected to single-cell photothermal stimulation with different laser powers (13, 16, 19 mW, duration: 10 sec). At 1 h post-stimulation, survival rates of target neurons were quantified by the viability assay. The survival rate decreased by 75% after photothermal stimulation at 16 mW (51.2 W/mm2), and an additional by 55 % after the stimulation at 19 mW (60.8 W/mm2, Fig. 3C). Taken together, these results suggest that thermal damage of single neurons can be elicited by intensive photothermal stimulation, and the threshold is around 16 mW (51.2 W/mm2). As neuromodulation effect can vary depending on stimulation region within the high-level of networks,6 we investigated the spatial sensitivity of single neuron to the photothermal stimulation. For this purpose, a high-density culture model (600 cells/ mm2) was used to mimic the complex network and to increase synaptic activities. A spiking neuron near a microelectrode was identified and neural activity changes were monitored while an illumination spot moved from high density of background neurites to the target neuron (Fig. 4A). We found that photothermal stimulation to the background neurites (next target cell) induced modest inhibitory effects, presumably due to partial blocking of the conductive inputs (Fig. 4B, 100 (neurite)/0 (soma); -37.3 %).22 As the illumination spot approached the soma of the target neurons from the background neurites, the inhibition effect increased gradually, and complete coverage of the soma with the light spot led to substantial inhibition of neural activity

ACS Paragon Plus Environment

Page 6 of 24

Page 7 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

(Fig 4B, 0 (neurite)/100 (soma); − 87%). Collectively, these results revealed that photothermal stimulation to high-density neurites can also modulate neural activity and cell body stimulation is more effective in inducing the clear inhibition. Next, we investigated the temporal resolution of single-cell photothermal inhibition. Each set of experiments was composed of 200 trials of photothermal stimulations and different pulse widths (500, 100, 50, 20 and 10 ms) with fixed intensity (10 mW) and pulse interval (3 sec) were applied to a target neuron (Fig. 5A). Peri-event histograms and raster plots showed that single-neuron activities were clearly inhibited during the repetitive photothermal stimuli with the pulse width down to 20 ms (500 ms: − 92.9%, 100 ms: − 91.2%, 50 ms: − 90.9%, 20 ms: − 92.3%, mean) and were fully restored after turning off the laser (Fig 5B and C). However, the photothermal inhibition with 10 ms pulses was negligible (− 4.5%, mean). Further analysis of spike rate changes demonstrated that the photothermal inhibition was induced within 10 ms from the laser onset, and the neural firing was completely restored after 100 ms from the laser offset (Fig. 5D and E). In particular, post inhibitory rebound occurred after photothermal stimulations ranging from 500 ms to 50 ms. These firing behaviors would be due to an intrinsic property of neurons in which they try to promote rhythmic electrical activity after neural suppression.27 Importantly, repetitive photothermal stimuli did not induce any significant fluctuation in baseline spiking activity, implying that single-cell photothermal inhibition is highly reversible without thermal damage. These observations thus demonstrate that photothermal stimulation can consistently inhibit the single-neuron activity with high reproducibility, and temporal resolution of the single-neuron photothermal inhibition is > 20 ms. Lastly, we examined whether inhibition of single neural activity can lead to a change in network firing patterns. Hippocampal neurons were plated on the GNR-MEA chip at a relatively high density (600 cells/mm2) and active channels were screened (mean firing rate > 0.1 Hz). Among the active channels, we further selected a target channel with high firing rate (> 1.5 Hz), and identified a spiking neuron near a microelectrode by measuring its inhibitory response to single-cell photothermal stimulation as shown in Figure 2D. The spiking neuron was repetitively stimulated with a 10 mW NIR laser (10 sec in duration) while neural network activity was simultaneously recorded by the

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

microelectrode array (Fig. 6A). We found that neural activities recorded by non-target channels were either inhibited, unchanged or excited while inhibiting the target channel activity (Fig. 6B). These observations indicate that the target neuron was directly or indirectly connected with others, which may elicit different neural responses depending on their interplay between excitatory or inhibitory connections as reported in the slice model.28 In contrast, any specific inhibition or stimulation of neural network activities was not observed in non-target channels in the absence of a target-neuron inhibition (Fig. S2). Each of channels was then statistically screened to find the affected channels during the target inhibition (Fig. 6C, paired T test, p