1
Chronic lithium treatment alters the excitatory/inhibitory balance
1
of synaptic networks and reduces mGluR5-PKC signaling
2
3
Khayachi A
1
., Ase A. R
1
., Liao C
1,2
., Kamesh A
1
., Kuhlmann N
1
., Schorova L
3
., Chaumette
4
B
4,5
., Dion P
1
., Alda M
6
., Séguéla P
1
., Rouleau G.A.
1,2
* & Milnerwood A. J
1
*.
5
6
1
Montreal Neurological Institute, Department of Neurology & Neurosurgery, McGill
7
University, Montréal, Quebec, Canada.
8
2
Department of Human Genetics, McGill University, Montréal, Quebec, Canada.
9
3
McGill University Health Center Research Institute, Montréal, Quebec, Canada.
10
4
Université de Paris, Institut de Psychiatrie et Neuroscience of Paris (IPNP), INSERM U1266,
11
GHU Paris Psychiatrie et Neurosciences, Paris, France.
12
5
Department of Psychiatry, McGill University, Quebec, Canada.
13
6
Department of Psychiatry, Dalhousie University, Halifax, Nova Scotia, Canada.
14
15
16
* Corresponding Authors:
17
Dr Austen J. Milnerwood, PhD & Dr. Guy A. Rouleau, MD, PhD, FRCPC, OQ
18
Department of Neurology and Neurosurgery
19
McGill University
20
Montréal, Québec, Canada H3A 2B4
21
E-mail: austen.milnerwood@mcgill.ca & guy.rouleau@mcgill.ca
22
23
24
25
Running title: Lithium shifts the synaptic network balance toward inhibition.
26
27
28
.CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted September 19, 2020. ; https://doi.org/10.1101/2020.09.18.303578doi: bioRxiv preprint
2
ABSTRACT
29
Bipolar disorder (BD) is characterized by cyclical alternations between mania and depression, often
30
comorbid with psychosis, and suicide. The mood stabilizer lithium, compared to other medications, is
31
the most efficient treatment for prevention of manic and depressive episodes. The pathophysiology
32
of BD, and lithium’s mode of action, are yet to be fully understood. Evidence suggests a change in the
33
balance of excitatory/inhibitory activity, favouring excitation in BD. Here, we sought to establish a
34
holistic appreciation of the neuronal consequences of lithium exposure in mouse cortical neurons
35
and identify underlying mechanisms. We found that chronic (but not acute) lithium treatment
36
significantly reduced intracellular calcium flux, specifically through the activation of the metabotropic
37
glutamatergic receptor mGluR5. This was associated with altered phosphorylation of PKC and GSK3
38
kinases, reduced neuronal excitability, and several alterations to synapse function. Consequently,
39
lithium treatment shifts the excitatory/inhibitory balance in the network toward inhibition. Together,
40
the results revealed how lithium dampens neuronal excitability and glutamatergic network activity,
41
which are predicted to be overactive in the manic phase of BD. Our working model of lithium action
42
enables the development of targeted strategies to restore the balance of overactive networks,
43
mimicking the therapeutic benefits of lithium, but with reduced toxicity.
44
45
.CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted September 19, 2020. ; https://doi.org/10.1101/2020.09.18.303578doi: bioRxiv preprint
3
INTRODUCTION
46
Bipolar disorder (BD) is a major psychiatric illness affecting 1-3% of the population worldwide and
47
one of the top 10 causes of disability (1, 2). BD starts in adolescence and has a life-long course,
48
characterized by frequently disabling episodes of mania and depression, often associated with
49
psychosis and suicide (3). The etiology of BD is complex and unclear, with genetic and environmental
50
factors implicated. Currently, except for very rare cases, no monogenic cause has been consistently
51
identified, and genome-wide association study (GWAS) hits have been found in varied biological
52
processes, including pathways related to intracellular signal transduction, glutamate synaptic
53
function, hormone signaling, and immune system regulation (4-6). The unknown cause and apparent
54
genetic heterogeneity of BD are a challenge to research efforts, especially those aimed at developing
55
appropriate disease models.
56
Lithium is an effective treatment for mania and has been consistently shown to reduce suicide and
57
overall mortality (7). Despite its narrow therapeutic range and potential side effects such as tremor,
58
polyuria, decreased thyroid function, and renal toxicity in a minority of patients (8), lithium remains
59
the first line treatment for prevention of both manic and depressive episodes in BD. For many
60
patients, it is the most effective mood stabilizer (9, 10), and in addition to reducing suicide risk, it
61
often enables patients to regain social and occupational function (7). Despite the widespread use of
62
lithium as a BD treatment for over 60 years, the mode of action needs to be better understood, as
63
does the reason why it is effective in only ~30% of BD patients (11, 12).
64
65
Studies aimed at elucidating lithium’s mode of action have found macroscopic changes in brain
66
structure (13) and alterations at the cellular level (14). For the latter, it has been shown that acute
67
lithium administration increases glutamate signaling (15, 16). In contrast, longer-term chronic
68
treatment over 6-7 days confers protection against glutamate-induced excitotoxicity, by reducing
69
NMDA receptor-dependent calcium flux (17). It has also been shown that lithium alters various
70
intracellular signaling cascades, by decreasing second messenger and calcium signaling, while
71
inhibiting several enzymes and kinases such as glycogen synthesis kinase 3 (GSK3), extracellular-
72
regulated kinase (ERK)/mitogen-activated protein kinase (MAPK) and inositol monophosphate
73
phosphatase (IMPase) (14, 18, 19). The resulting alterations to intracellular signaling likely also
74
converge upon the regulation of gene expression, synaptic transmission and plasticity,
75
neuroprotection, and circadian biology.
76
77
Several genomic studies correlated specific loci with lithium responsiveness (20), suggesting a shared
78
genetic predisposition both to disease and response to treatment. Single nucleotide polymorphisms
79
(SNPs) in the PLCG1 gene have been associated with response to lithium, indicating that the
80
phospholipase C (PLC)-phosphatidylinositol4,5-biphosphate (PIP2)-inositol triphosphate (IP3)
81
signaling pathway may be an important target of lithium. Interestingly, our unpublished data (72)
82
report an association between a SNP in the GRM5 gene encoding the metabotropic glutamate
83
receptor 5 (mGluR5) and response to lithium suggesting that mGluR5 activity and downstream PLC-
84
IP3 signaling is also involved in lithium’s therapeutic action. Other studies have found SNPs
85
associated with BD in the GRIN2A(6) and GRIA2 genes that encode NMDA and AMPA receptor
86
subunits, respectively. Intriguingly, only SNPs in GRIA2 were associated with lithium responsiveness
87
(5, 21), suggesting that lithium alters the regulation of Ca
2+
-permeable AMPA receptors. Beyond
88
genomic association, it remains unclear how lithium interacts with mGluR5, PLCG1 and GluA2
89
signaling to produce the beneficial outcome in lithium-responsive patients.
90
91
.CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted September 19, 2020. ; https://doi.org/10.1101/2020.09.18.303578doi: bioRxiv preprint
4
Here, we sought to establish a holistic appreciation of the neuronal consequences of chronic lithium
92
exposure in mouse cortical neurons and begin to determine the underlying mechanisms. We
93
performed messenger RNA (mRNA) sequencing in neurons treated chronically with lithium and
94
discovered altered transcriptional regulation of genes involved in glutamate receptor trafficking and
95
intracellular calcium signaling. We found that chronic (but not acute) lithium treatment significantly
96
reduced excitatory receptor-mediated intracellular calcium flux, specifically through the mGluR5
97
receptor. This was associated with altered phosphorylation of PKC and GSK3 kinases, reduced
98
neuronal excitability, and several alterations to synapse function. Specifically, chronic lithium
99
exposure reduced excitatory synapse activity and density, while increasing inhibitory synapse activity
100
and density. Consequently, lithium treatment altered the excitatory/inhibitory (E/I) balance in the
101
network, favouring inhibition. Together, the results shed light on how lithium may dampen neuronal
102
excitability and glutamatergic network activity, which are predicted to be overactive in the manic
103
phase of BD (22-24).
104
In addition, this discovery strengthens the potential clinical use of lithium to treat disorder with
105
altered excitatory/inhibitory network activity such as epilepsy and several forms of autism. This study
106
could also help to develop targeted strategies to restore the balance of overactive networks,
107
mimicking the therapeutic benefits of lithium, but with reduced toxicity.
108
109
110
.CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted September 19, 2020. ; https://doi.org/10.1101/2020.09.18.303578doi: bioRxiv preprint
5
MATERIALS AND METHODS
111
112
Primary neuronal cultures and animals
113
Cortical neurons were prepared from wild-type (WT) embryonic (E15.5) C57BL/6 mice as previously
114
described (25). Animals were maintained within the Centre for Neurological Disease Modeling
115
according to the Canadian Council on Animal Care regulations (AUP 2017-7888B). Briefly, cortical
116
neurons were plated in Neurobasal medium (ThermoFisher 21103049) supplemented with 1x B27
117
(ThermoFisher 1750044), 1x glutaMax (ThermoFisher 35050061) on 60-mm dishes or 12-mm glass
118
coverslips (VWR) pre-coated with poly-D-Lysine (0.1 mg mL-1; Sigma). Neurons (600,000 cells per 60-
119
mm dish or 80,000 cells per 12-mm coverslip) were then used at 18- 20 days in vitro (DIV).
120
121
Drug treatment
122
123
The therapeutic range of LiCl (lithium) treatment is between 0.75-1.5mM. In this study, neurons were
124
treated chronically with ~1.5mM LiCl (Sigma L9650) for 7 days starting at 11 DIV post-differentiation.
125
As controls, neurons were treated with ~1.5mM of NaCl (Sigma S5886) to keep the same amount of
126
chloride in the dish as neurons treated with LiCl. Experiments were performed at 18 to 20 DIV.
127
128
Data manipulation and statistical analyses
129
130
Statistical analyses were performed using GraphPad Prism software (GraphPad software, Inc). All
131
data are expressed as mean ± standard error of the mean (s.e.m.). Paired t-tests (Fig 5A,B,C),
132
parametric unpaired t-tests (Figs: 2C-G; 5D-F; Supplementary Figs: 3B; 4C,f; 5A-C) or non-parametric
133
Mann-Whitney tests (Figs: 2H; 3B-E; 4; Supplementary Fig. 2A-C) were used to compare medians of
134
two sets. One-sample t-tests were used with hypothetical value 100 for control (Figs: 3G; 5G, H;
135
Supplementary Fig: 2D, E). Normality for all groups was verified using the Shapiro-Wilk tests and
136
p<0.05 was considered significant.
137
138
Data availability
139
140
All relevant data are in the figures and supplementary figures. Raw data could be requested from
141
the corresponding author.
142
143
A detailed “Materials and Methods” section can be found in the supplementary information.
144
145
.CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted September 19, 2020. ; https://doi.org/10.1101/2020.09.18.303578doi: bioRxiv preprint