Subscriber access provided by Northern Illinois University
Letter
Subcellular Optogenetic Stimulation for Activity-Dependent Myelination of Axons in a Novel Microfluidic Compartmentalized Platform Hae Ung Lee, Sudip Nag, Agata Blasiak, Yan Jin, Nitish V. Thakor, and In Hong Yang ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.6b00157 • Publication Date (Web): 29 Aug 2016 Downloaded from http://pubs.acs.org on August 31, 2016
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 free 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 accessible to all readers and 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.
ACS Chemical Neuroscience 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 19
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 Chemical Neuroscience
1
Subcellular Optogenetic Stimulation for Activity-Dependent Myelination of Axons in a Novel
2
Microfluidic Compartmentalized Platform
3
Hae Ung Lee1, Sudip Nag1, Agata Blasiak1, Yan Jin1, Nitish Thakor1,2, In Hong Yang1,2,*
4
1
Singapore Institute for Neurotechnology, National University of Singapore, Singapore.
5
2
Department of Biomedical Engineering, School of Medicine, Johns Hopkins University, Baltimore,
6
USA.
7
Abstract
8
Myelination is governed by neuron-glia communication, which in turn is modulated by neural
9
activity. The exact mechanisms remain elusive. We developed a novel in vitro optogenetic
10
stimulation platform that facilitates subcellular activity induction in hundreds of neurons
11
simultaneously. The light isolation was achieved by creating a bio-compatible, light-absorbent, black
12
microfluidic device integrated with a programmable, high-power LED array. The system was applied
13
to a compartmentalized culture of primary neurons whose distal axons were interacting with
14
oligodendrocyte precursor cells. Neural activity was induced along whole neurons or was
15
constrained to cell bodies with proximal axons or distal axons only. All three modes of stimulation
16
promoted oligodendrocyte differentiation and the myelination of axons as evidenced by a decrease
17
in the number of oligodendrocyte precursor cells followed by increases in the number of mature
18
oligodendrocytes and myelin sheath fragments. These results demonstrated the potential of our
19
novel optogenetic stimulation system for the global and focal induction of neural activity in vitro for
20
studying axon myelination.
21 22
Keywords: Optogenetic Stimulation, Myelination, Oligodendrocytes, Microfluidics, Black PDMS
23
1 ACS Paragon Plus Environment
ACS Chemical Neuroscience
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
1
Introduction
2
Neuronal activity plays a pivotal role in the survival, proliferation and migration of neurons and glial
3
cells, as well as the myelination of axons.1-3 Multiple findings in cell culture systems showed that
4
neuronal activation by electrical stimulation (ESTIM) improves myelination.4-7 Accordingly, various in
5
vivo models of neuronal activation demonstrated that neural activity influences myelination.8-10
6
However, the application of ESTIM via electrodes has considerable limitations. Firstly, it is infeasible
7
to simultaneously target multiple neurons with high spatial resolution. Secondly, ESTIM requires
8
mechanical proximity and electrical integrity which are laborious and impractical to apply in awake
9
and moving animals.11
10
Optogenetic stimulation (OSTIM) provides a versatile alternative to ESTIM. A light beam can be
11
easily and quickly targeted to stimulate neurons with expression of light-sensitive proteins, such as
12
channelrhodopsin 2 (ChR2).11 OSTIM elicits similar levels of activation in neurons as ESTIM does.12
13
When ChR2 transfected neurons were exposed to light at 10 - 20 Hz frequency, their electrical
14
activity resembled the physiological firing rate of neurons in the brain.13 Even higher frequencies
15
can be achieved with modified ultrafast ChR2 (ChETA).14 Recently, Gibson et al. found that neuronal
16
activation by OSTIM promotes oligodendrogenesis and myelination in the mouse motor cortex.15
17
Their findings demonstrate that OSTIM can be used to study neuronal activity dependent
18
myelination mechanism.
19
Neurons are highly polarized, as a consequence, multiple cellular processes are region specific.
20
Deciphering the role of each subcellular region requires tools that facilitate stimulation and signal
21
registration with limited interference from other cellular regions. Traditionally used Campenot
22
chamber systems for neuron culture applies a physical constraint and isolates bulky cell bodies from
23
thin axons with a set of parallel microgrooves.16 This simple idea has become wide-spread in labs
24
around the world with the increased availability of polydimethylsiloxane (PDMS)-based devices. The
25
microgrooves have since been replaced with microchannels that connect axonal and somatic
26
chambers. The careful balancing of the volumes, and the consequently affected hydrostatic
27
pressures, in both chambers combined with the resistance from the microchannels provides fluidic
28
isolation. Thus, the axonal and somatic microenvironments can be independently controlled.17, 18
29
Another advantage of using PDMS-based microfluidics is easy integration with other techniques. A
30
PDMS-based optical system benefits from low fluorescence and high chemical stability. However, its
31
application is limited as PDMS causes internal light scattering and reflection generation from the top
32
and bottom surfaces, which may affect the input and output light signals. Furthermore, it impedes
33
isolated stimulation of a chosen area of the device, as the light is unrestrained and can propagate 2 ACS Paragon Plus Environment
Page 2 of 19
Page 3 of 19
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 Chemical Neuroscience
1
through PDMS. One of the potential solutions is changing the property of the material and creating
2
light absorbent PDMS.19 It is crucial that while blocking the light, the PDMS remains biocompatible –
3
the material cannot be toxic and it has to provide sufficient gas permeability to support cellular
4
growth.
5
In this study we present a black PDMS microfluidic device that provides spatial, fluidic, and light
6
isolation of the somatic and axonal chambers without compromising biocompatibility. The devices
7
were used to apply highly controlled, selective optogenetic stimulation to the chosen neuronal
8
subcellular region in a massively parallel fashion and without direct glial cell stimulation. The
9
integration of compartmentalized microfluidic co-culture systems with optogenetics allowed testing
10
the hypothesis that focal induction of membrane potential shift is sufficient to increase
11
oligodendrocyte maturation and myelination of axons.
12
Results and Discussion
13
OSTIM applied to animals is a promising technique to study neural activity-dependent myelination.
14
While in vivo approaches provide invaluable information by incorporating the complexity of the real
15
tissue, they suffer from certain disadvantages, e.g. the myelination process might be affected by glial
16
scarring after the insertion of light stimulation devices in the brain. Furthermore, the complexity of
17
the system with mixed cell types, including multiple neuronal subtypes, different glial cells, etc.,
18
impedes deciphering the mechanism of neural activity-induced myelination. Finally, the subcellular
19
stimulation of multiple neurons is not feasible. An in vitro model system that facilitates precise
20
control over subcellular microenvironments and focal neural activity induction is needed to decipher
21
mechanisms behind axon myelination. We aimed to develop such a model by integrating
22
neurobiology with microfluidics and optogenetics.
23
Spatial control of light exposure
24
A compartmentalized PDMS device
25
with a pattern outlining the positions of the chambers and microchannels was obtained through soft
26
lithography. The pattern was then replicated in a thin layer (approx. 30 μm) of transparent
27
polydimethysiloxane (PDMS) by spin-coating and curing the PDMS on top of the wafer. A thick layer
28
(approx. 7 mm) of PDMS supplemented with 1% carbon black particles (CB-PDMS) was cured on top
29
of the transparent PDMS. Subsequently, the PDMS pad was peeled off, the wells (somatic and axonal
30
chambers) were punched out and the pad was bonded to a glass coverslip via oxygen plasma
31
activation (Fig. 1A, C). A high-power LED set up was developed to stimulate a chosen chamber of the
20
was modified to limit light scattering. Briefly, a silicon wafer
3 ACS Paragon Plus Environment
ACS Chemical Neuroscience
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
1
device: somatic, axonal or both. The device contained an array of blue LEDs that were arranged in a
2
2×3 matrix and individually controlled with a programmable controller (Fig 1D). The controller
3
received commands from an USB-based computer interface. Received commands were decoded
4
and stored inside a non-volatile FLASH memory and utilized for generating real-time stimulation
5
sequence (Fig.1B). The system functioned in stand-alone mode if the PC was not present, while
6
reading the pre-programmed stimulation parameters. The CB-PDMS device containing up to three
7
compartmentalized culture systems was placed on top of the LED array, such that each of the six LED
8
lights was under one of the chambers (Fig.1D). The pre-programming assured the small size of the
9
set up - the LED system fit into a Ø10 cm Petri dish that could be placed in an incubator - without
10
losing versatility of the stimulation modes. The CB-PDMS acted as an optical trap – a 1.5 mm sheet
11
blocked more than 99.99% of the light from the LED set up. Additionally, the bottom of the glass
12
coverslip underneath the unstimulated chamber was covered with aluminum foil and dark tape to
13
prevent the light exposure. As a result, when only one of the chambers was stimulated, the light
14
intensity reaching the opposite chamber was limited to < 1.5%.
15
The CB-PDMS light trap not only provided isolated LED light exposure of a chosen chamber but also
16
limited any internal scattering and reflections from the PDMS. The carbon black particles were of
17
sizes < 1μm, and therefore if the CB-PDMS mixture was directly applied at the master, it could
18
interfere or even block 3 μm-high microchannels. The thin layer of transparent PDMS allowed
19
precise replication of the pattern. The submillimeter thickness of the layer ensured that when
20
subsequently covered with the thick layer of CB-PDMS the presence of the transparent PDMS did
21
not affect the light isolating properties of the device. The strict cleansing steps facilitated the
22
removal of toxic compounds from CB-PDMS and provided its biocompatibility. As such, this material
23
can be used for any fluorescent cell assays that benefit from limiting the spatial light exposure and
24
limited level of light reflections , e.g. from the top surface of closed PDMS chamber. 21
4 ACS Paragon Plus Environment
Page 4 of 19
Page 5 of 19
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 Chemical Neuroscience
1 2 3 4 5 6 7 8 9 10 11 12
Fig. 1. Experimental setup for subcellular OSTIM. A. The preparation of CB-PDMS microfluidic device followed previously published protocols except for addition of CB-PDMS layer on top of the thin layer of transparent PDMS. B. Block diagram of the multichannel optogenetic stimulator. A USB transceiver communicated with the work station computer for incoming commands. The received signal was passed to a low power microcontroller, which decoded and processed the commands. The microcontroller also generated timing and control sequences for six LEDs for optical stimulation. The blue LEDs were interfaced with a driver array that was capable of providing required high current to the LEDs. C. One CB-PDMS device contained up to three sets of independent culture chambers compatible with the alignment of LED array D. The optogenetic device was composed of an operating module and a LED module. E.. The OSTIM platform facilitated light exposure of six CBPDMS chambers (three cultures, each with two chambers) in an independent fashion.
13
Primary neuron/glia co-culture in compartmentalized CB-PDMS platform for optogenetic
14
stimulation
15
Primary dorsal root ganglion (DRG) neurons from embryonic day 13 mice were plated in the somatic
16
chamber of the CB-PDMS devices (Fig. 2A,B) andtransfected with light-sensitive ChR2-EYFP probe
17
under the control of the CaMKII promoter.11 After 6-7 days in vitro (DIV), axons sent by the cell
18
bodies crossed through the 500μm-long microchannels to the axonal chamber and extensively
19
elongated. At 10 DIV the axonal chamber was filled in with axons (Fig.2C). The neuronal growth
20
pattern was not significantly different from that observed in the transparent PDMS devices. Neuron
21
morphology was regular as visualized with the neurofilament stain. The expression of ChR2-EYFP
22
was confirmed along the whole cell as the EYFP signal overlapped with the neurofilament stain (Fig.2 5 ACS Paragon Plus Environment
ACS Chemical Neuroscience
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
1
D-G). Most of the axons in the axonal chamber expressed ChR2-EYFP indicating that the transfection
2
did not affect axonal growth. Oligodendrocyte precursor cells (OPCs) were sourced from postnatal
3
day 1 mouse pups and were purified by a magnetic separation system. The purity of separated OPCs
4
was 85-90% as verified with anti-O4-PE. OPCs seeded in the axonal chamber of CB-PDMS with 10
5
DIV-old axons survived in a co-culture for at least 14 DIV. LED optical pulses were used to stimulate
6
ChR2-expressing DRG neurons along their whole length (WholeSTIM), in the axonal chamber
7
(AxonSTIM) and in the somatic chamber (SomaSTIM) (Fig.2 H-J).
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Fig.2. The optogenetic platform was compatible with primary neuron/glia co-cultures and allowed for neuron subcellular stimulation in three different modes. A. Schematic of the compartmentalized culture system. Somatic and axonal chambers were connected with 5 μm-wide and 500 μm-long microchannels. Cells were plated in the somatic chamber and axons, but not cell bodies, crossed to the axonal chamber. B. Primary DRG neurons transfected with the ChR2-EYFP probe were plated in the somatic chamber of the CB-PDMS device. There was no effect of the CB-PDMS on the neurite growth or morphology as visualized with anti-neurofilament stain. C. At 10 DIV axons passed through the microchannels and extensively invaded the axonal chamber. D. Transfected primary neurons expressed ChR2-EYFP in cell bodies and proximal axons in the somatic chamber as visualized in the YFP channel. E. Axons that crossed to the axonal chamber expressed ChR2-EYFP. F. The merged image indicated that almost all neurons in the somatic chamber (cell bodies and proximal axons) expressed ChR2. G. Almost all axons that crossed to the axonal chamber expressed ChR2 as demonstrated by the co-localization of YFP and neurofilament signals. H-J. Schematics of the stimulation modes: whole cell stimulation (H; WholeSTIM) where both chambers were illuminated 6 ACS Paragon Plus Environment
Page 6 of 19
Page 7 of 19
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 Chemical Neuroscience
1 2
with pulses from blue LEDs; axonal chamber stimulation (I; AxonSTIM) and somatic chamber stimulation (J; SomaSTIM), where blue LEDs only stimulated a single, chosen chamber.
3 4
The compartmentalized system provided spatial and fluidic isolation of distal axons and cell bodies.
5
OPCs were exclusively interacting with the axons in the axonal chamber, such that the presence of
6
cell bodies did not interfere with the myelination process. This arrangement reflects the in vivo
7
condition that distal regions of DRG axons reaching towards the target tissue experience different
8
microenvironments than cell bodies clustered in a posterior root of a spinal nerve. Accordingly, axon
9
myelination in vivo can occur hundreds of micrometers away from the cell bodies, so the cell bodies
10
can affect this process only by changing the state of their axons, e.g., inducing anterograde
11
propagation of an electrical signal.
12
CB-PDMS devices contained up to three individual, compartmentalized cultures (a somatic and an
13
axonal chamber for each culture; Fig1. C, D and E). The LED array with independent control of each
14
light source allowed stimulation of all three cultures simultaneously (Fig1. D and E). This
15
arrangement not only increased the replicates of the experiments, but also allowed a direct
16
comparison of the effects of the three modes of the stimulation (WholeSTIM vs. SomaSTIM vs.
17
AxonSTIM) as they could have been applied simultaneously for reducing experimental variations (e.g.
18
culturing conditions).
19
Several methods have been designed to obtain subcellular optogenetic stimulation, including
20
subcellular channel-rhodopsin-assisted circuit mapping (sCRAM)22 and the two photon activation
21
approach.23 These systems facilitate high spatial resolution of stimulation (around 50 μm for sCRAM
22
and