Nature - USA (2020-10-15)

420 | Nature | Vol 586 | 15 October 2020

Article

addition, we found that microglia determine the threshold for stri-

atal neuron activation in response to D1 agonist treatment (Fig. 3d–f).

Acute treatment with a low dose of D1 agonist was sufficient to induce

a substantial and prolonged increase in the frequency of neuronal

events, with no difference in magnitude, in animals lacking microglia

as compared to control mice (Fig. 3d–f, Extended Data Fig. 6c–g).

Neuronal activation is associated with the local release of ATP by

neurons and astrocytes^20 –^23. Microglia can detect synaptic release of

ATP through the surface-expressed purinergic receptor P2RY12^24 ,

which is highly expressed in forebrain microglia^25 and controls ATP/

ADP-dependent chemotaxis and motility of microglia^3 ,^8 ,^24 ,^26 ,^27. Although

extracellular ATP levels in the striatum at baseline (3–5 nM)^23 are

far below concentrations that activate P2RY12 (>100 nM)^24 ,^26 , local

ATP release at the synapses of activated neurons can reach concen-

trations up to 5–500 μM^20 , which is more than sufficient to trigger

microglial P2RY12 signalling^10 ,^24. Using two-photon live imaging of

eGFP-expressing microglia in the cortex, we tracked the positioning

and velocity of microglia protrusions in vivo (Fig. 3g, h, Supplemen-

tary Video 2, Extended Data Fig. 6h–j). Depending on the level of neu-

ronal activity, microglia can display different modes of motility: in the

AAV injection

GRIN lens

1–2 weeks

recovery

a

AAV9.Syn.GCaMP6s

CSF1Ri or control chow

~3 week

habituation

two-photon1 week

live imaging

b Control Microglia-decient

Distance (μm)

Corr

elation

c P < 0.0001

(^010100) 1,000
0.1
0.2
0.3
0.4
0.5
0.6
P < 0.0001
P = 0.0005
g Control h
imaging
Drug
imaging
AAV injection
headplate we4–5eks
VL
Cx3cr1eGFP/+ mice
P = 0.0038
Average velocity per
cell per mouse (
μm min
–1)
SalineCN
O
CNO +
P2 RY12 inhibitorP2RY12 inhibito
r Saline CNO
P = 0.0043
P = 0.015 P = 0.013P = 0.028
Mean intensity
(^0012345678)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Baseline
Cumulative probabilityD1 agonist
f
d
(^0012345678)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Baseline
D1 agonistP < 0.0001
P = 0.005
AAV8.hSyn.hM3D.Gq.mCherry
AAV1.Camk2a.Cre.SV4D
AAV1.CAG.Flex.tdTomato
0
0.5
1.0
1.5
2.0
0
0.2
0.4
0.6
0.8
1.0
0
1
2
3
Events per min (per mouse)
10 min20 min30 min
D1 agonist
10 mi
n
20 mi
n
30 min
Saline
P = 0.0008
P = 0.01
Control
Microglia-
decient
0
1
2
3
0
1
2
(^3) P < 0.0001
Saline
D1 agonist10 min
e
Saline
D1 agonist10 mi
n Event rate (Ca2+ events per min)
2 ΔF/F
25 s
CNO +
P2 RY12 inhibitorP2inhibitorRY12
Events per min (per mouse)Events per min (per mouse)
Event rate (Ca2+ events per min)
P = 0.23
Control
Microglia-decient
Control Microglia-decient Control Microglia-decient
Fig. 3 | Microglia control synchrony and f iring frequency of striatal
neurons in vivo. a, Two-photon calcium imaging of striatal neurons in control
and microglia-deficient mice. b, c, Increased spatiotemporal coding of striatal
neuronal activity in microglia-depleted mice at baseline. b, Representative
neuron traces (n = 10 neurons per mouse); c, Correlation of spatiotemporal
coding of striatal neuron activity (n = 7 control and 9 microglia-deficient mice;
treatment, P < 0.0001; distance, P < 0.0001; interaction, P < 0.0001; two-way
ANOVA with Sidak’s multiple comparison test). d, e, Average frequency of Ca2+
events in control and microglia-deficient mice at baseline and in response to D1
agonist (SKF81297, 3 mg kg−1; n = 7 and 9 mice, respectively). d, Diet, P = 0.48;
treatment, P = 0.0024; interaction, P = 0.08; two-way ANOVA with Tukey’s post
hoc test. e, Paired two-tailed t-test. f, Cumulative probability of Ca2+ events per
neuron per minute in control and microglia-deficient mice at baseline and in
response to D1 agonist (3 mg kg−1; n = 7 and 9 mice, two-sample Kolmogorov–
Smirnov test). g, h, Neuronal ATP-dependent changes in motility and
localization of microglia protrusions. Using live two-photon imaging, eGFP+
microglia protrusions (C x 3 cr1eGFP/+ mice) were tracked in proximity to virally
labelled tdTomato+ hM3Dq+ neurons. h, Microglial protrusion velocity per
mouse (left) and mean intensity of microglial protrusions within 1 μm^2 of
neuronal terminals per mouse (right) at baseline (saline) or upon neuronal
activation (CNO, 5 mg kg−1, i.p.) in the absence or presence of P2RY12 antagonist
(clopidogrel, 100 mg kg−1, i.p.; n = 6, 7, 6, and 3 mice from left to right; left,
P = 0.0012; right, P = 0.01; one-way ANOVA with Tukey’s post hoc test). The data
shown in c–f are combined from two independent imaging cohorts of mice.
Data shown as mean ± s.e.m.