Graphical Abstract
mGlu5 receptor function is optimal at resting cell potential. Membrane
depolarization stabilizes an inactive-like conformation of mGlu5
receptor and decreases agonist-induced Gq protein
activation, Ca2+ release from intracellular stores and
gating of NMDA and TRPC6 channels.
Figure 1. Voltage regulates mGlu5 agonist-induced
Ca2+ release. HEK cells stably expressing GCaMP6s
also transiently expressed mGlu5 (B-F) or AT1 (G) receptors. A)Scheme of GCaMP6s fluorescent Ca2+ sensor - adapted
from . GCaMP6s is composed of a circularly permuted GFP, a
calmodulin (CaM) and a peptide from smooth-muscle myosin light-chain
kinase (RS20). B) DHPG (100µM)-induced fluorescence fluctuation
in cells expressing (black and blue) or not (grey) mGlu5 receptor, in
absence (grey and black) or presence (blue) of the mGlu5 NAM, MPEP
(10µM); Representative illustration. C) Peak of fluorescence
fluctuation induced by quisqualate (10µM) in isotonic extracellular
solutions containing increasing KCl concentrations ([KCl] in mM),
normalized to fluorescence fluctuations recorded with the 3mM [KCl]
solution. Data are mean ± SEM of n = 6 triplicates from independent
experiments. Statistics: Kruskal-Wallis test, D) Membrane
potential measured in the current clamp whole-cell configuration with
extracellular solutions containing either 3mM (black) or 100mM (red)
KCl. Data are mean ± SEM of n = 10 cells from 3 independent experiments.
Statistics: Friedman test E) Dose-response curves of
DHPG-induced fluorescence fluctuation with KCl 3mM (black,
Vrest) and 100mM (red, Vdepol),
normalized to Vrest peak of each experiment. Data are
mean ± SEM of n = 3 independent experiments. F and G)Quisqualate (10µM, F) or Angiotensine II (1µM, G) -induced fluorescence
fluctuation in KCl 3mM (Vrest) and 100mM
(Vdepol) solutions. Left, representative illustration;
Right, mean ± SEM of n = 7 to 14 triplicates from independent
experiments; Statistics: one sample Wilcoxon test.
Figure 2. Voltage regulates mGlu5 agonist-induced
Ca2+ oscillations. A) Representative field of GCaMP6s
expressing cells in epifluorescence microscopy - Scale = 10 µm.B) Quisqualate (10µM)-induced fluorescence fluctuation at
Vrest (left, black) and Vdepol (middle,
red), for individual field of view (thin) and average (thick) aligned to
the peak of response, and quantification of the mean ± SEM of n = 14 to
21 independent experiments (right, Statistics: Mann-Whitney test)C) Fluorescence fluctuation recorded in representative
oscillating, single spike and silent cells at Vrest and
Vdepol in response to quisqualate (10µM) with proportion
of cells in each category displayed in a circular diagram, n = 403 cells
for Vrest and 324 cells for VdepolD) Percentage of oscillating cells (> 2
oscillations) at Vrest and Vdepol. n =
324 to 403 cells from 14 to 18 independent experiments. Statistics: N-1
χ² test, p = 0.0001. E) Global frequency of fluorescence
oscillations induced by quisqualate (10µM) application at
Vrest and Vdepol. Data are mean ± SEM of
n = 280 to 408 cells. Statistics: unpaired t-test F)Representative illustrations (left) and violin plots with mean indicated
by a dotted line (right) of fluorescence oscillations frequency induced
by Vrest or Vdepol imposed after
quisqualate (10µM) application (n = 21 to 35 cells - Statistics:
unpaired t-test)
Figure 3. Voltage tunes mGlu5 receptor activation probability .A) Scheme of the TR-FRET sensor of mGlu5 conformation with
SNAP-Lumi4-Tb (purple) and SNAP-fluoroscein (green) linked on
SNAP-tag-mGlu5 VFT domains. The FRET signal is inversely proportional to
the number of receptors in an active-like conformation. B)TR-FRET intensity at Vrest with and without quisqualate
10µM (left) and quisqualate (10µM) effect at Vrest and
Vdepol (right). Data are mean ± SEM of n = 10
independent experiments. Statistics: Wilcoxon test. C)Dose-response curve of LY341495 mGlu antagonist applied at
Vrest or Vdepol. Data are mean ± SEM of
n = 4 independent experiments normalized to the maximum TR-FRET of
Vdepol.
Figure 4. Voltage modulates mGlu5-mediated Gq activation. A)Scheme of EMTA ebBRET sensor of Gq activation . When Gq is
activated by mGlu5, p63-RhoGEF-RlucII is recruited to the membrane where
it interacts with rGFP-CAAX. B and C) Representative
illustrations (B) and mean ± SEM (C) of BRET signals measured before
(CT) and after (DHPG, 100µM) stimulation at Vrest or
Vdepol. N= 86 to 97 cells from more than 3 independent
experiments; statistics: Paired t-test. D) Single cell BRET
measurement of DHPG net effect; statistics: Unpaired t-test.
Figure 5. Voltage controls mGlu5 gating of TRPC6 channels. Aand B) - Whole-cell patch clamp current induced by DHPG (100µM)
on HEK293T cells expressing mGlu5-Venus alone (A) or with TRPC6-tomato
(B). C) mGlu5-Venus/TRPC6-tomato co-transfected cells were held
at -80mV (black) or -20mV (red) during DHPG (100µM) application.
Current-voltage relationship was then recorded at the maximum response
induced by DHPG (100µM) for potentials ranging from -80mV to +60mV in
100ms (bottom). Inset: Mean ± SEM current density at -80mV. Data are
mean ± SEM of n = 9 cells per condition from 4 independent experiments,
Statistics: Mann-Whitney test.
Figure 6. Voltage regulates mGlu5 gating of NMDA receptor in
hippocampal neurons. A) Representative current and mean density induced
by DHPG (50µM, left) or NMDA (30µM, right) at -80mV. B - D)Representative currents (B), mean current density and
INMDA+DHPG/INMDA current ratio (C), and
percentage of DHPG-induced NMDA current potentiation (D) measured in
presence of NMDA (30µM) alone or co-applied with DHPG (50µM) at -80mV
(black) and -40mV (red) holding potential. Measures of
INMDA (blue) and INMDA+DHPG (black) used
for quantification in C and D are indicated by arrows in B. In D, paired
measurement of INMDA + DHPG / INMDA were
performed subsequently at -80mV or -40mV, in a random order, on the same
neuron. C and D data are mean ± SEM of n = 10 to 15 neurons from at
least 4 independent experiments; Statistics: Wilcoxon matched-pairs
signed rank test.