Atipamezole abolishes spike-frequency adaptation and blocks NA-induced GIRK-I in LC neurons.

(A, B) Sample traces of spike trains induced in an LC neuron by injection of current pulses at 75 and 100 pA. Note the spike-frequency adaptation and pulse-afterhyperpolarization (pulse-AHP).

(C) Plotting of the cumulative spike numbers vs the elapsed time from the onset of the current pulses every 50 ms at 75 pA (blue circles) or 100 pA (red circles) (n = 7). Two-way RM ANOVA, *p < 0.001. The saturation level (a) and the half saturation constant (b) were measured by fitting the saturation function defined as y = (a * x) / (b + x) to the data points. The values of a at 75 and 100 pA current pulses were 44.6 ± 16.2 and 52.5 ± 12.3, respectively (paired t-test, p = 0.010). The values of b at 75 and 100 pA current pulses were 2.1 ± 0.7 and 2.1 ± 0.4, respectively (paired t-test, p = 0.823).

(D, E) Sample traces of spike trains induced in an LC neuron evoked by current pulses at 100 pA under control condition and after application of 10 μM atipamezole. Note that atipamezole suppresses spike-frequency adaptation and pulse-AHP (blue and red stars).

(F) Plotting of the cumulative spike numbers vs the elapsed time during the current pulses every 50 ms before and after application of atipamezole (blue and red circles, respectively) (n = 10). Two-way RM ANOVA, *p < 0.001.

(G) Plotting of a against b, which were measured by fitting the saturation function to the data points in F. Both the values of a and b obtained by curve fitting to the data points after application of atipamezole (open and filled red circles) were significantly shifted to larger values than those obtained under control condition (open and filled blue circles) (paired t-test, a and b, ‡p = 0.003 and ‡p = 0.002, respectively), and there was a significant difference in the relationship between a and b (Wilk’s lambda, §p = 0.002) (n = 10).

(H) Top, Voltage command pulse. Lower panel, Superimposed current traces obtained under control condition (1), after application of 100 μM NA (2), and after application of 10 μM atipamezole (Atip) in addition to NA (3). The amplitudes of inward and outward components in NA-induced GIRK-I were –70 ± 22 and 36 ± 11 pA, respectively (paired t-test, ‡p = 0.005, n = 6).

(I) A graph showing the effects of atipamezole on amplitudes of the inward and outward components of NA-induced GIRK-I at –130 mV and –60 mV, obtained by subtraction of currents recorded under control condition from those recorded after application of NA (2–1) and from those recorded after application of NA and atipamezole (3–1). Inward component, paired t-test, ‡p < 0.001; outward component, paired t-test, ‡p < 0.001 (n = 6).

Ca2+-dependent rundown of NA-induced GIRK-I and its suppression by barbadin.

(A) Upper panel, Ramp command pulse. Lower panel, Superimposed current traces obtained under control condition (1) and after application of 100 μM NA for 5 min (2).

(B) Upper panel, A combined command pulse applied every minute; five trains of 20 positive pulses (5 ms duration to 0 mV at 20 Hz) at an interval of 2 s in the presence of extracellular 30 mM TEA and intracellular 0.2 mM EGTA, which were followed by the ramp pulse after an interval of 25 s. Lower panel, representative Ca2+ currents in response to positive pulses in the presence of NA and 30 mM TEA.

(C) Upper panel, Ramp command pulse. Lower panel, Superimposed current traces obtained after application of NA for 5 min (2) and in response to application of the 20th combined pulse in the presence of NA and TEA (3).

(D) The I-V relationship of NA-induced GIRK-I obtained by subtraction of currents recorded under control condition from that recorded after application of NA for 5 min (red trace, 2–1) and that obtained by subtraction of the control current from that recorded in response to application of the 20th combined pulse in the presence of NA and TEA (blue trace, 3–1).

(E) Plotting of amplitudes of inward components at –130 mV (blue circles) and outward components at –60 mV (red circles) against time.

(F) Plotting of normalized amplitudes of inward components at –130 mV and outward components at –60 mV before and 20 min after application of positive pulse trains in the presence of NA and TEA. The amplitudes of inward components at –130 mV and those of outward components at –60 mV were normalized by those recorded in the presence of NA before applying positive pulse trains. Inward component, paired t-test, ‡p < 0.001; outward component, paired t-test, ‡p < 0.001 (n = 8).

(G) Upper panel, Ramp command pulse. Lower panel, Superimposed current traces obtained under control condition (1) and after application of 100 μM NA for 5 min (2) in LC neurons dialyzed with 100 μM barbadin (β-arrestin/AP2 blocker).

(H) Upper panel, A combined command pulse applied every minute; five trains of 20 positive pulses (5 ms duration to 0 mV at 20 Hz) at an inter-train interval of 2 s in the presence of extracellular 30 mM TEA and intracellular 0.2 mM EGTA, which were followed by the ramp pulse after an interval of 25 s. Lower panel, Superimposed current traces obtained after application of NA for 5 min (2) and in response to application of the 20th combined pulse in the presence of NA and TEA (3) in LC neurons dialyzed with barbadin.

(I) The I-V relationship of NA-induced GIRK-I obtained by subtraction of currents recorded under control condition from that recorded after application of NA for 5 min (red trace, 2–1) and that obtained by subtraction of the control current from that recorded in response to application of the 20th combined pulse in the presence of NA and TEA (blue trace, 3–1).

(J) Plotting of normalized amplitudes (left vertical axis) of inward components at –130 mV (filled blue circles) and outward components at –60 mV (filled red circles) against time (n = 8). Amplitudes of GIRK-I were normalized to the values obtained before applying pulse trains. Right vertical axis refers to the original amplitudes. Open blue and red circles represent the normalized amplitudes of GIRK-I shown in E.

(K) Plotting of normalized amplitudes of inward components at –130 mV and outward components at –60 mV before and 20 min after application of positive pulse trains in the presence of NA and TEA (n = 8). The amplitudes of inward components at –130 mV and those of outward components at –60 mV were normalized to those recorded in the presence of NA before applying positive pulse trains. Open light pink circles represent results obtained under the control condition in F. Inward component, unpaired t-test, *p < 0.001; outward component, unpaired t-test, *p < 0.001.

RS reduces expression of α2A-ARs in LC Neurons.

(A, B) Confocal images of LC neurons showing immunoreactivities for TH and α2A-ARs, together with a merged one in non-stress and 3-day RS mice. Arrowheads indicate the membrane regions of TH-positive neurons, along which α2A-ARs were differentially expressed between the control and the RS mice. Asterisks indicate LC neurons.

(C) Western blotting analyses showing the expression of α2A-ARs in non-stress and 3-day RS mice (n = 6 and 7, respectively). Unpaired t-test, †p < 0.001.

(D, E) Confocal images of LC neurons in non-stress and RS mice. Upper panels from left to right showing the respective immunoreactivity for TH, α2A-ARs (with rabbit anti-α2A-AR) and Na+-K+-pump and an enlarged image of the region enclosed with a rectangle (arrowheads) in its immediate left panel. Lower panels from left to right showing the merged image of TH and α2A-ARs, that of TH, α2A-ARs and Na+-K+-pump, that of α2A-ARs and Na+-K+-pump and an enlarged image of the region enclosed with a rectangle in its immediate left panel.

(F, G) Western blotting analyses showing expressions of α2A-ARs and Na+-K+-pump in membrane fraction and those of α2A-ARs and β-actin in cytosol fraction in non-stress and 5-day RS mice (n = 3 and 3 samples, respectively). Each sample represents the analysis result in the LC tissues obtained from 2-3 mice. Membrane; unpaired t-test, †p = 0.016. Cytosol; unpaired t-test, †p = 0.083.

RS reduces NA-induced GIRK-I in LC neurons.

(A-E) Representative traces of NA-induced GIRK-I obtained from LC neurons in non-stress and 1-day, 2-day, 3-day, and 5-day RS mice.

(F, G) Amplitudes of inward and outward components of NA-induced GIRK-I obtained from LC neurons in non-stress and 1-day, 2-day, 3-day, and 5-day RS mice (n = 8, 7, 5, 6 and 8, respectively) decreased with the increase in the period of RS, in a way that can be described by a saturation function (red interrupted lines). The saturation level (a + b) and the half saturation constant (c) were determined by fitting the saturation function, defined as y = a + (b* x) / (c + x), to the data points. The values of a, b and c for the inward component of GIRK-I were −151.1, 108, and 0.9, respectively, and those for the outward component of GIRK-I were 81.1, –67, and 1.6, respectively. Inward component: one-way ANOVA, p < 0.001, post hoc fisher’s PLSD, 1-day; #p < 0.001 vs Non-stress and *p < 0.05 vs 3-day and 5-day, 2-day; #p < 0.001 vs Non-stress, 3-day; #p < 0.001 vs Non-stress, 5-day; #p < 0.001 vs Non-stress. Outward component: one-way ANOVA, p < 0.001, post hoc fisher’s PLSD, 1-day; p = 0.004 vs Non-stress and *p < 0.05 vs 3-day and 5-day, 2-day; p = 0.015 vs Non-stress, 3-day; p < 0.001 vs Non-stress, 5-day; p < 0.001 vs Non-stress.

(H) Relative expressions of α2A and α2C mRNAs, normalized to GAPDH in LC neurons (n = 5). Paired t-test, ‡p = 0.014.

(I) Relative expressions of GIRK1, GIRK2, and GIRK3 mRNAs, normalized to GAPDH in LC neurons (n = 5). One-way RM ANOVA, p < 0.001, post hoc fisher’s PLSD; *p = 0.008 for GIRK1 vs GIRK2, *p < 0.001 for GIRK1 vs GIRK3, *p = 0.008 for GIRK2 vs GIRK3.

(J, K) Relative expressions of α2A and GIRK1 mRNAs, respectively, normalized to GAPDH in LC neurons in non-stress mice (n = 8), and 3-day RS mice (n = 6). α2A-AR: unpaired t-test, †p = 0.020; GIRK1: unpaired t-test, †p = 0.013.

(L) Normalized relative expressions of GIRK2 mRNA in LC neurons in non-stress mice (n = 8) and 3-day RS mice (n = 6), normalized to the ratio of the mean value of the relative expressions of GIRK1 mRNA to that of GIRK2 mRNA in LC neurons in non-stress mice (I). Unpaired t-test, †p = 0.037.

Expressions of TH, MAO-A, DBH, AEP and tau proteins in LC neurons.

(A) Confocal images of neurons in LC at a low magnification showing the immunoreactivities for TH and MAO-A, together with a merged one. RS appeared to cause slight increases in the expressions of TH and MAO-A. Brightness of background in sample image obtained from the control was corrected to be equal to that obtained from the RS.

(B) Confocal images of LC neurons at a high magnification showing the immunoreactivities for DBH and AEP, together with a merged one. RS caused increases in the expression of AEP.

(C) Western blotting of TH and MAO-A in LC neurons obtained from non-stress and 5-day RS mice.

(D, E) Western blot analysis revealing no significant change in TH expression (n = 10 samples) but a significant increase in MAO-A expression (n = 6 samples) in 5-day RS group compared to non-stress group. Unpaired t-test, †p = 0.029.

(F) Western blotting of pro-AEP and active AEP in LC neurons obtained from non-stress and 5-day RS mice.

(G, H) Western blot analysis revealing significant increases in pro-AEP and active AEP (n = 6 and n = 6 samples, respectively) in 5-day RS group compared to non-stress group. Unpaired t-test, †p = 0.034 (G) and †p = 0.047 (H).

(I) Western blotting of tau and tau N368 fragment in LC neurons obtained from non-stress and 5-day RS mice. Unpaired t-test, †p = 0.025.

(J, K) Western blot analyses revealing a significant increase in tau and tau N368 fragment in LC neurons (n = 8, and n = 6, respectively) in 5-day RS group compared to non-stress group. Each sample represents the analysis result in the LC tissues obtained from 2-3 mice. Unpaired t-test, †p < 0.05.

Presumed cellular mechanisms for the degeneration of LC by stress.

(A, B) Differential free concentrations of NA to be metabolized by MAO-A into DOPEGAL between control and stress conditions: Under the control condition (A), NA in the cytosol is mostly taken up into the cytoplasmic vesicles by vMAT2 (thick arrow 3), rather than being directly metabolized by MAO-A into DOPEGAL (thin interrupted arrow). Vesicular NA is released from cell bodies as autocrine following [Ca2+]i increases caused by action potentials, and the released NA activates α2A-AR-coupled GIRK channels, causing autoinhibition. Subsequently, the autocrine released NA is slowly taken up into the cytosol of LC neurons by NAT after dissociation from α2A-ARs. Under the stress condition (B), an activation of CRF receptors in LC neurons by stress inhibits leak K+ channels and increases firing activities in LC neurons, subsequently causing a larger [Ca2+]i increase together with Ca2+-induced Ca2+ release (CICR). Impairment of autoinhibiton due to Ca2+ dependent internalization of α2A-ARs-coupled GIRK channels leads to the persistent excitation in LC neurons, which enhances autocrine release of NA (thick arrow 1). Subsequently, the excessively autocrine released NA is taken up directly and rapidly by NAT into the cytosol without binding to α2A-ARs (thick arrow 2). Such a facilitation of re-uptake of NA by NAT would increase active NA storage into cytoplasmic vesicles by vMAT2 (thick arrow 3), while the rate of NA leakage from cytoplasmic vesicles would also increase (arrow 4) due to a dynamic equilibrium in cytoplasmic vesicles between active NA storage into cytoplasmic vesicles and passive NA leakage from cytoplasmic vesicles. Subsequently, such an increase in the rate of NA leakage would result in an increase in a MAO-A metabolite, DOPEGAL and AEP, leading to a production of cleaved tau N368 fragment and an impairment of learning/memory.