Supplementary Figure 1

Protective effect of restraint stress: histology

Representative kidney histological sections (H&E stain) illustrating the protective effect of restraint stress against renal ischemia-reperfusion injury (IRI) in DBH-cre mice. IRI induced cast formation, tubule dilation, and/or tubular epithelial denucleation, and these alterations were less severe in DBH-cre mice with restraint stress. Scale bar: 100 μm for main panels and 50 μm for insets.

Nature Neuroscience: doi:10.1038/nn.4526

Supplementary Figure 2

C1 stimulation activates breathing in conscious DBH-cre mice

(a) Rostrocaudal distribution of mCherry-immunoreactive catecholaminergic (TH+) neurons in the ventrolateral medulla oblongata of 25 DBH-cre mice 5-6 weeks after unilateral injection of AAV2–DIO–EF1α–ChR2–mCherry (neurons per section vs. distance from bregma

in mm). (b) Location of fiber optic tips. Distance caudal to bregma in mm shown at lower left of each section. Scale bar 500 m (c)

Original plethysmography recordings (top trace, respiratory flow signal, inspiration downward; bottom trace, respiratory frequency (fR). Optogenetic stimulation of the C1 neurons (5, 10, 15, and 20 Hz; pulse duration, 10 ms; 10 s train at blue bars) activated breathing in a DBH-cre mouse (d-f) Effect of C1 neuron stimulation on fR (d), tidal volume (Vt, e), and minute volume (Ve, f) in 27 DBH-cre mice. Statistics: one-way repeated measure ANOVA with Tukey–Kramer test; [F(3, 104) = 128.1, P < 0.0001] (d), [F(3, 104) = 12.01, P < 0.0001] (e), and [F(3, 104) = 87.49, P < 0.0001] (f). *** P < 0.001 vs. 5 Hz, ††† P < 0.001 vs. 10 Hz, and ‡‡‡ P < 0.001 vs. 20 Hz.

Nature Neuroscience: doi:10.1038/nn.4526

Supplementary Figure 3

Prolonged C1 stimulation regularizes the breathing rate and produces quiescence in conscious DBH-cre mice

(a) Plethysmography record of a DBH-cre mouse prior to and during 10 min unilateral photostimulation of C1 cells (frequency, 5 Hz; pulse duration, 10 ms). Top trace, air flow signal; bottom trace, respiratory frequency (fR). Large amplitude signals in the flow trace denote locomotor activity or other motor behavior (grooming, sniffing). Incidence of such events was much reduced during C1 cell stimulation in DBH-cre mice. Higher resolution respiratory flow trace during active behavior (a1), spontaneous quiescence (a2), and quiescence induced by C1 neuron stimulation (a3). (b) Representative Poincaré plots of breathing frequency (instantaneous frequency of (n+1)th breath vs. previous breath (nth)) before (left) and during C1 neuron stimulation (right panel; 5 Hz; pulse duration, 10 ms, 10 min) in DBH-cre mouse. Standard deviations normal to the line of identity (SD1, sqrt(Σ(fRn+1-fRn)^2 /2(N-1))) and along the line of identity (SD2, sqrt(Σ(fRn+1+fRn-2m)^2 /2(N-1))) were calculated using LabView (National Instruments) and the ellipse area was calculated (SD1 x SD2 x π). (c) Mean ellipse area was significantly reduced by C1 neuron stimulation indicating more frequent episodes of quiescence (10 min; frequency, 5 Hz; pulse duration, 10 ms) (n = 6). Statistical analysis: unpaired t test [t(10) = 3.41, P = 0.0067] ** P < 0.01 vs. Laser(-)

Nature Neuroscience: doi:10.1038/nn.4526

Supplementary Figure 4

Protective effect of C1 stimulation: histology

Representative histological sections (H&E stain) of kidney illustrating the protective effect of C1 neuron stimulation in DBH-cre mice against kidney ischemia-reperfusion injury (IRI). IRI induced cast formation, tubule dilation, and/or tubular epithelial denucleation. These alterations were reduced by C1 photostimulation (ChR2-mcherry expressing vector) but not by sham stimulation (mCherry-expressing vector). Scale bar: 100 μm for main panels and 50 μm for insets. Laser: 5 Hz, 10 min. IR, ischemia-reperfusion.

Nature Neuroscience: doi:10.1038/nn.4526

Supplementary Figure 5

Histological evidence of successful subdiaphragmatic vagotomy

Representative histological picture of the Fluoro-Gold (FG) expression in dorsal motor nucleus of the vagus (DMV). Sham (n = 3) and subdiaphragmatic vagotomized (sVNX, n = 3) mice received i.p. injections of FG (5 mL/kg of 1% solution in sterile water). After 3 days, the mice were deeply anesthetized and perfused. FG expression at DMV is observed in Sham mouse but not in sVNX mouse. Scale bar: 500 μm for main panels and 250 μm for insets.

Nature Neuroscience: doi:10.1038/nn.4526

Supplementary Figure 6

Renal hypoxia-inducible factor 1α mRNA level after 10 min of ischemia vs. 10 min of C1 stimulation

Kidney Hif1α/Gapdh mRNA ratio was significantly elevated after renal ischemia-reperfusion in DBH-cre mice (10 min of complete renal vessel clamping) whereas optogenetic activation of C1 cells (5 Hz for 10 min) produced no change in mRNA level. Kidneys were harvested immediately after 10 min of renal ischemia or 10 min after the cessation of C1 photostimulation. Statistics (5 mice per group): one-way ANOVA with Tukey–Kramer test; [F(2, 20) = 5.653, P = 0.0186]. * P < 0.05 vs. IR(-):Laser(-) and IR(+):Laser(-).

Nature Neuroscience: doi:10.1038/nn.4526

Supplemental Table 1. Experimental groups for C1 stimulation.

But, Butoxamine; Hex, Hexamethonium; Mif, Mifepristone; DMSO, Vehicle of Mif; SPX, Splenectomy; sVNX,

Subdiaphragmatic vagotomy; *1 including sham operation of sVNX; *2 Each number includes the mice for

corticosterone experiment

Supplemental Table 2. Experimental groups for restraint stress

No

restraint Restraint Restraint Restraint

N/A N/A Clozapine

vehicle1 Clozapine2

DBH-cre 24*1 18*1 0 0

No vector

DBH-cre 0 0 6 6

DREADD

DBH-cre 6 6 0 0

Caspase

Total mice 30 24 6 6 1Saline, 2Clozapine N-oxide. *1 Each number includes the mice for corticosterone experiment

No laser Laser Laser Laser Laser Laser No laser Laser Laser

N/A N/A But Hex Mif DMSO SPX SPX sVNX

DBH-

cre 29*1,2 39*1,2 6 6 6 6 6 6 6

ChR2-

mCherry

DBH-

cre 0 6 0 0 0 0 0 0 0

mCherry

Total

mice 29 45 6 6 6 6 6 6 6

Nature Neuroscience: doi:10.1038/nn.4526