Endogenously Bursting, Synchronous iPSC- GRC derived ...€¦ · GRC Excitatory Synapses & Brain...
Transcript of Endogenously Bursting, Synchronous iPSC- GRC derived ...€¦ · GRC Excitatory Synapses & Brain...
4-A
P [
30
0 µ
M]
Coby Carlson, Lisa Harms, Elisabeth Enghofer, Kwi Hye Kim, Christian Kannemeier, Susan DeLaura, Tom Burke, Blake Anson, Arne Thompson, Eugenia Jones and Kile P ManganCellular Dynamics International, Inc., A FUJIFILM Company, Madison, WI USA
AbstractThe mammalian brain works appropriately only when there is a proper balance between excitation and inhibition. An imbalance in the
ratio of excitatory-to-inhibitory neurons (referred to as the E/I ratio) is associated with numerous neurological abnormalities and deficits. Increased E/I ratios, that is higher excitability, leads to prolonged neocortical circuit activity, stimulus hypersensitivity, cognitive impairments, and epilepsy (Hagerman, et al. 2002; Gibson, et al., 2008; Zhang, et al. 2011). Of equal interest and importance, decreases in the E/I ratio, or a stronger inhibitory drive, have been linked to impaired social interactions, autistic behaviors, and mental retardation (Tabuchi, et al. 2007; Dani, et al. 2005). It is well accepted that the E/I ratio changes during neuronal development, with excitation decreasing and inhibition increasing, and that deviations in this process give rise to neurological disorders.
Despite the urgency to advance our understanding of the mechanisms that govern neural network formation and synchronous bursting behaviors, a major challenge in neuroscience research and drug development is access to clinically-relevant cell models. Induced pluripotent stem cell (iPSC) technology can now provide access to previously unattainable human neuronal cell types. We have generated iPSC-derived frontal cortical neurons of an appropriate, brain-similar E/I ratio (~80% excitation), that develop and display network-level, synapse-coupled, coordinated neuronal activity in vitro, evident by periodical, synchronized bursts captured (Poisson and ISI statistics) via MEA. These cultures contain excitatory synapses (HOMER1 and synapsin co-staining) that modulate (blocked via APv and DNQX) the synchronous bursting behaviors observed on the MEA.
In the interest of understanding how excitatory pharmacology alters network-level neuronal interactions, we investigated alterations in the regular bursting properties of iPSC-derived neural cultures after pharmacological addition (0.003 to 300 uM dose curve) of excitatory agents (bicuculline, chlorpromazine, pentylenetetrazol, amoxapine, 4-aminopyridine, glutamic acid), common AEDs (ganaxalone and valproate), as well as vehicle (ethanol, DMSO) and negative controls (acetaminophen). Activity metrics displaying dose-dependent responses with pharmacology include: ‘single-channel’ Poisson bursts (rate, intensity and duration), ‘network-level’ ISI bursts (rate, intensity and duration), and synchrony measures. The presented data illustrates how human iPSC-technology can be leveraged to create an unprecedented investigatory space for understanding the intricacies of how excitatory synapses control network-connected populations of human neurons.
Endogenously Bursting, Synchronous iPSC-derived Neural Cultures Display a SeizurogenicResponse to Excitatory Pharmacology.
Increasing Excitation is Hyper-Synchronizing
iPSC-Derived Neuronal Cell Types
GRCExcitatory Synapses &
Brain Function
Multi-Elecrode Arrays (MEAs) were utilized to assess the activity levels and bursting behaviors of cultures with varying E/I ratios. Representative all-points histogram graphs (top) and example raw data traces (red and black figures from 16 electrodes; 4X4 grid) of single wells on DIV 32 are displayed. The E/I ratios were set by mixing iCell Neurons (majority GABAergic {~80%}) with increasing percentages of iNeurons (primarily glutamatergic {~99%}), resulting in increased excitation from left-to-right across the MEA plate. Monitoring the raw channel recordings demonstrates the unique phenotypes observed on the MEA. Specifically, the expansion of a synchronous network (green boxes), increased intensity (Hz: purple box) of network bursts with the titration of excitation, ultimately leading to the appearance of network seizures in columns 5 and 6 after an optimal threshold of excitability is reached (@ an E/I ratio of approx. 65%). Synchronous network firing essentially stops with too much excitation, underscoring the importance of an optimal E/I balance. Note the addition of iCellAstrocytes (upper right box) stabilizes seizing cultures within 4 DIVs post addition.
≥ 70% Glutamatergic population determined by single cell gene expression
≥90% Pure Neurons
- Easily implemented inexcitotoxic compoundscreening
≥90% CD44 (+) / S100ß (+)
- Expression of key astrocytic markers
- Clearance ofextracellular glutamate
- Stimulation with TNFα, IL-1βand IFNγ upregulates theexpression of the pro-inflammatory cytokine IL-6
MIXED INTO CO-CULTURE!
120k GNCs + 20k Astros
iCe
llG
luta
Ne
uro
ns
iCe
llA
stro
cyte
s
Calcein AM EthD-2
Glutamate Uptake
0 50 100
200
400
0
1
2
3
4
5- inhibitor
+ inhibitor
Replace media ( Phenol red free DMEM with glucose and sodium pyruvate (no glutamine/GlutaMax)) with or without 80µl of 200µM DL-TBOA per well.
Add 20µl of media (untreated) or 5X of Glutamic acid (0.5µM or 1mM for 100 & 200µM ea) to each well and gently tap the plate to mix well.Incubate at 37°C for 1 hour.
Rinse with 1X DPBS once and add 40µl of 1X DPBS per wellAdd 5µl of Inactivation solution per well. Shake the plate for 1 minute and incubate for 5 min.Add 5µl of Neutralization solution and shake.
Prepare Glutamate Assay master mix and add 50µl of the mix per well.Incubate in dark for 1 hour and read the luminescence.
Glutamic Acid (M)
Fo
ld C
han
ge
Glutamate Uptake
0 50 100
200
400
0
1
2
3
4
5- inhibitor
+ inhibitor
Replace media ( Phenol red free DMEM withglucose and sodium pyruvate (no glutamine/GlutaMax)) with or without 80µl of 200µM DL-TBOA per well.
Add 20µl of media (untreated) or 5X of Glutamic acid (0.5µM or 1mM for 100 & 200µM ea) to each well and gently tap the plate to mix well.Incubate at 37°C for 1 hour.
Rinse with1X DPBS once and add 40µl of 1X DPBS per wellAdd 5µl of Inactivation solution per well. Shake the plate for 1 minute and incubate for 5 min.Add 5µl of Neutralization solution and shake.
Prepare Glutamate Assay master mixandadd 50µl of the mixper well.Incubate indark for 1 hour and read the luminescence.
Glutamic Acid (M)
Fo
ld C
han
ge
PEI Coat
48w MEA
Days in Culture -1
Thaw, Mix &
Plate GNCs
+Astros
1
50% Medium Change
(every 48 hours)
Perform Pharmacology Evaluation
by adding
10 µL of 30X sol. to wells
0 ~18
Plate Layout for ALL Plates (µM)
Pharmacology Tested Bicuculline (BIC) Chlorpromazine (CHL) Picrotoxin (PTX) Pentylenetetrazol (PTZ) 4-Aminopyridine (4AP) Glutamatic Acid (GA) Acetaminophen (ACE) Ethanol (EtOH) Methanol (data not shown) DMSO
• Test Compounds• Controls
Assessing Excitatory PharmacologySeizurogenic Assay workflow, analysis and results are presented for all pharmacology tested, including vehicle, control, excitatory (i.e. seizurogenic) and anti-epileptic (AEDs) drugs. iCell GlutaNeurons and iCell Astrocytes were mixed upon thaw and dotted together onto 48-well MEA plates. Evaluation of seizurogenic pharmacology was performed on DIV 19 or 20 by adding 10 µL (30X solutions) to wells containing 300 µL of Brainphys Medium, assessing 6 different concentrations of each drug [0.003, 0.03, 0.3, 3, 30, 300 µM]. Each plate contained positive control (bicuculline [200 µM]) treated wells, as well as untreated wells. ‘Before’ and ‘Treatment’ 8-minute recordings were collected, with pre-incubation periods of 10 minutes preceding each recording. Spike Files (6 SD) were collected and processed for spike and bursting metrics via Axion Neural Metric analysis and by an all-points histogram burst-peak detection suite (CDI NeuroAnalyzer). Differences from baseline were normalized to vehicle control and are presented for all drug concentrations, for each metric. Example analysis flow and raster plots are shown (middle).Far Right: Filled radar graphs (increasing concentration going clock-wise from top) for all pharmacology are presented depicting absolute value changes from baseline of all metrics. *Note control and vehicle display no changes from baseline, while seizurogenic pharmacology alters spike and burst metrics >40 fold.
Vehicle (H2O) BIC (30 µM)
-40
-30
-20
-10
0
10
20
30
40
mn
FRs
Act
Es
NA
BP
M
NA
Co
nn
chan
B
chan
Bsp
k
chan
BD
ur
nB
nB
urs
tPer
c
nB
urs
tDu
r
nEl
ect
Syn
chro
ny
t0003 t003 t03 t3 t30 t300. . .
Ra
tio
No
rma
lized
to
Veh
icle
Bicuculline [µM]
Ratio Normalized to Vehicle Treatment
mnFRs ActEs NA BPM
NA Conn chanB chanBspk
chanBDur nB nBurstPerc
nBurstDur nElect Synchrony
Ganaxolone
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Valproate
0
10
20
30
40t00003
t0003
t003
t03
tp3
t3
4-Aminopyridine
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Chlorpromazine
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Phenothiazine
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Glutamate
0
10
20
30
40t0003
t003
t03
t3
t30
t300
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Acetaminophen 1 Acetaminophen 2
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Bicuculline Bicuculline
1 Day
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Picrotoxin
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Picrotoxin
1 Day
0
10
20
30
40t0003
t003
t03
t3
t30
t300
EtOH
0
10
20
30
40t00625
t0125
t025
t05
t1
t2
0
10
20
30
40t00312
t00625
t0125
t025
t05
t1
DMSO Co
ntro
lsExcitato
ry C
om
po
un
ds
(10
min
)
AED
s
0
10
20
30
40t0003
t003
t03
t3
t30
t300
Glu
tam
ate
[0.3
µM
]
Glu
tam
ate
[3 µ
M] 60 secs
200 Hz
• iCell co-culture bursting behaviors are developed and modulated via excitatory synapse (APv & DNQX).• Excitatory pharmacology dose-dependently induces ‘seizurogenic bursting phenotype’ iCell co-cultures.• Vehicle and negative controls do not alter iCell co-culture bursting behaviors.
Summary and Conclusions• iCell GlutaNeurons cultures exhibit an appropriate E/I ratio to produce robust, rhythmic bursting behaviors.• iCell GlutaNeurons and iCell Astrocytes can be mixed together to generate a purely human neuronal co-culture.• iCell co-cultures develop synchronized, rhythmic and reproducible network-level bursting phenotypes within 2
weeks.
Ch
lorp
rom
azin
e [3
µM
]
Qualifying Compounds: Filled Radar Graphs
Overlay
Synapsin Homer1
Colocalization
IncreasingExcitation
IncreasingInhibition
MEA Plate Layout
Firi
ng
Rat
e (
Hz)
Time (1 min)
Synchrony Spreads Synchrony Speeds
Ele
ctro
de
s (r
aw)
Synchrony Seizes Synchrony Stops
Column 1 – 80:20 Column 2 – 70:30 Column 3 – 60:40 Column 4 – 50:50
Column 5 – 40:60 Column 6 – 31:69 Column 7 – 21:79 Column 8 – 11:89
Increasing ratio of ‘excitatory neurons’
50
0 H
z
Introduction of Astrocytes is Stabilizing
DIV 32
DIV 36: 4 DIV post iCell Astrocyte (20k) Addition – same MEA cultures as below
Multi-Elecrode Arrays (MEAs) were utilized to assess the activity levels and bursting behaviorsof iCell GlutaNeurons (an optimal ~70% E/I ratio) mono-cultures & iCell GlutaNeurons and iCell Astrocytes co-cultures. LEFT: iCell GlutaNeurons mono-cultures (DIV 20) display bursting sensitivity to APv & DNQX, show-casing network-burst behaviors are modulated via excitatorysynapses. RIGHT: iPSC-derived neuronal co-cultures develop synchronized bursting behaviorsbeginning near DIV11 and reach robust, re-producible and reliable bursting levels near DIV16.Co-cultures display a slightly 1) more organized and 2) cleaner bursting phenotypecompared to mono-cultures. Appropriately, co-culture conditions decrease mean Firing Ratebut increase bursting behaviors both at a channel-level (‘Poisson’) as well as at the network-level (ISI) resolution. We utilized Axion’s Neural Metric analysis to assess bursting behaviors,along with an in-house all-points histogram (500 msec bins) algorithm (*) that detects burstingpeak time-points (o).
0
2
4
6
8
10
mn
FR
per
Ele
ct
(Hz)
0
4
8
12
16
Acti
ve E
lectr
od
es [
16]
0
20
40
60
80
100
Bu
rsti
ng
Perc
en
tag
e [
ISI]
0.0
0.2
0.4
0.6
0.8
1.0
Syn
ch
ron
y In
dex [
0-1
]
Synchrony Measures0
20
40
60
80
'Po
isso
n' B
urs
ts (
BP
M)
0
10
20
30
40
50
'Po
isso
n' B
urs
t In
ten
sit
y [
Hz]
0.0
0.5
1.0
1.5
2.0
'Po
isso
n' B
urs
t D
ura
tio
n [
Sec]
Channel ‘Poisson’ Bursting
GNCs+
Astros
0
2
4
6
8
ISI:
Bu
rsti
ng
Rate
(B
PM
)
0
1
2
3
4
ISI:
Bu
rsti
ng
Du
rati
on
(sec)
0
2
4
6
8
Bu
rsti
ng
Rate
[B
PM
]
0
500
1000
1500
2000
2500
Bu
rsti
ng
Peak In
ten
sit
y [
Hz]
Network Bursting
**GNC + Astrocytes
Elec
tro
de
#P
eak
Inte
nsi
ty (
Hz)
Minutes
GNC Mono-culture
Elec
tro
de
#P
eak
Inte
nsi
ty (
Hz)
Minutes
Quantifying Network-Level Synchronization
+dotted onto MEA plates
Above Graph depicts ‘Ratio Normalized to Vehicle’ for each concentration[µM: lighter to darker). Analysis includes variables depicting ‘Spikes’, ‘Channel’,‘Network’ and ‘Synchrony’ measures. ABOVE and TOP RIGHT: Analysis exampleof Bicuculline addition. RIGHT MIDDLE: Mono-culture example raster-plot andall-points histogram before and after Phenothiazine (PTZ) addition. BELOW andBOTTOM RIGHT: Example all-points histogram graphs display changes inbursting behaviors following addition of excitatory pharmacology. *Notevariable changes following pharmacological addition, including increased burstduration, increased bursting frequency and / or complete destruction ofbursting behaviors.
Population Histogram
Ele
ctro
de
s
APv [50 µM] & DNQX [25 µM] AdditionPopulation Histogram
Ele
ctro
de
s
Time (sec)
40 60 10080 120200
Time (sec)
40 60 10080 120200
Time (sec)
40 60 10080 120200
Individual Action Potential Network
Burst
Individual Action Potential’s clustered in ‘Poisson’ Bursts
WASHOUT
40 22060 10080 120 140 160 180 200Time (sec)
40 22060 10080 140 180Time (sec)120 160 200
Baseline PTZ [18 µM], 1 Hr