Inhaled Hydrogen Sulfide: A Rapidly Reversible Inhibitor of Cardiac and Metabolic Function in the Mouse

This study was approved by the Subcommittee on Research Animal Care at the Massachusetts General Hospital, Boston, Massachusetts. We used 12 male C57BL/6 WT mice (Jackson Laboratories, Bar Harbor, ME) for radiotelemetric experiments and 7 male SV129 mice (Taconic Inc., Cambridge, MA) for the echocardiographic studies. The animals were maintained individually in polycarbonate mouse cages on a 12/12-h light–dark schedule at 27°± 1°C ambient temperature (T^a). Mice had free access to food and water, except during exposures to study gas.

To record T^band activity, radiotelemetry devices (model TA10TA-F20; DSI, St. Paul, MN) were surgically implanted in the peritoneal cavity of four mice. During surgery, the animals were anesthetized with ketamine (0.1 mg/g) and xylazine (0.01 mg/g), and they were allowed to recover for a minimum of 2 weeks before commencing experiments.

Radiotelemetry devices (model TA11PA-C10; DSI) equipped with a miniature intraarterial catheter were surgically implanted in the ventral subcutaneous space with a catheter (0.4 mm OD) inserted in the left carotid artery of four mice to record heart rate (HR), mean arterial pressure (MAP), and physical activity.

To record the electrocardiogram during H²S breathing, radiotelemetry devices (model TA10ETA-F20; DSI) were surgically implanted in the peritoneal cavity of three mice. These devices are equipped with two electrocardiographic leads that were implanted within muscle in a lead II configuration (positive lead 1 cm to the left of the xiphoid process, and negative lead in the right pectoral muscle).

Activity was measured in mice with the telemeter based on the signal strength recorded by the receiver (using Dataquest ART 4.0 software; DSI). The signal strength depends on the orientation and the distance of the animal from the receiving antenna. As the mouse moves in its cage or chamber, the relative change in signal strength is converted into activity units. The number of counts or units generated depends on both the distance and the speed of movement.

Respiratory rate (RR) was counted manually and recorded at 30-min intervals during the experiments.

Each mouse was placed in its cage to breathe air for 1 h to record the baseline values of HR, MAP, and RR. To minimize the effects of agitation due to handling and obtain stable hemodynamics, the baseline values were measured while the mice remained in their cages with the telemetric device turned on. After 1 h of recording baseline values in individual cages, each mouse was placed in a 500-ml metabolic chamber (model PLY 3211 V2.1; Buxco Electronics, Inc., Wilmington, NC) in which the mouse breathed air supplied at a rate of 1 l/min at either 27° or 35°C T^a. After 1 h of breathing air in the metabolic chamber, each mouse was exposed to an atmosphere containing 1,000 ppm H²S in nitrogen (Airgas Inc., Radnor, PA), which was mixed with compressed air using a three-tube volumetrically calibrated flowmeter (Cole-Parmer, Vernon Hills, IL), to deliver a final concentration of 80 ppm H²S and 17.5% oxygen (total gas flow delivered at 1 l/min). H²S and oxygen concentrations were continuously measured using a portable gas monitor (ITX Multi-Gas Monitor; Industrial Scientific Corporation, Oakdale, PA). Mice were exposed to this atmosphere continuously for 6 h and then were replaced in their cages to breathe air at 27°C T^afor 3 h. T^b, MAP, HR, and physical activity were measured for 10-s intervals during each minute of the experiment and recorded using Dataquest ART 4.0 software on a computer.

To study the effects of H²S breathing on cardiac function, echocardiography was performed on seven awake SV129 mice. Echocardiography was first performed in mice breathing air at T^a27°C. Body temperature was measured with a rectal probe (FHC, Bowdoin, ME). One week later, echocardiography was repeated after the mice breathed H²S (80 ppm) for 3 h in a chamber at T^a23°C (n = 6). Mice recovered for another week, and echocardiography was performed again after H²S exposure at T^a35°C (n = 5). The mice were kept inside the chamber, breathing H²S throughout the procedure. Echocardiography was performed using a 14-MHz probe (Vivid 7; GE Medical Systems, Milwaukee, WI) in conscious mice. Three to four M-mode acquisitions were obtained at the midpapillary level of a parasternal short axis view as previously described.8Left ventricular internal diameter at end-diastole and left ventricular internal diameter at end-systole were measured (All measures were averaged on a minimum of 5 consecutive beats). HR, shortening fraction, and left ventricular volumes were calculated9; stroke volume (SV) and cardiac output (CO) were derived.

Mice were placed in individual chambers (25 cm long and 5.0 cm in diameter). Animals breathed air (1 l/min total gas flow) for 1 h and then H²S (80 ppm) for 3 h at T^a27°C, followed by a recovery period of 2 h breathing air. A gas sample was continuously withdrawn from the expiratory limb of the circuit to measure H²S concentration, and the electrocardiographic tracing was recorded continuously.

To explore the possible influence of blood gas tensions or electrolyte abnormalities on cardiovascular function, C57BL/6 mice (n = 7) were exposed to room air or H²S (80 ppm) for 2 h, as described above. Each chamber was fitted with a rubber stopper at one end, through which the mouse’s tail protruded. After 2 h of breathing either air or H²S, an arterial blood sample was obtained from the tail artery for analysis of arterial partial pressure of carbon dioxide, arterial partial pressure of oxygen, arterial pH, arterial oxygen saturation, Na⁺, K⁺, Ca²⁺, Mg²⁺, glucose, blood urea nitrogen, and lactic acid levels using a Stat Profile Critical Care Xpress machine (Nova Biomedical, Waltham, MA).

To examine the impact of H²S breathing on carbon dioxide production, C57BL/6 mice (n = 3) implanted with temperature telemeters were exposed to breathing air at either 27° or 35°C T^afor 1 h, as described above. Mice were then exposed for 30 min to H²S (80 ppm at 27°C T^aand 40, 80, or 120 ppm at 35°C T^a), after which they breathed air without H²S for a 30-min recovery period at the same T^a. Carbon dioxide production was measured in the effluent chamber gas airflow once each minute using an infrared carbon dioxide analyzer (LI-820 CO²Gas Analyzer; LI-COR Biosciences, Lincoln, NE).

To study the relation between carbon dioxide production and HR during H²S inhalation, C57BL/6 mice (n = 3) implanted with radiotelemetry devices (model TA11PA-C10; DSI) were first exposed to air at 27°C T^afor 1 h as described above. The animals were then exposed for 30 min to H²S (80 ppm), followed by a 30-min recovery period breathing air at the same T^a. Carbon dioxide production and HR were recorded every minute using the same methods described above.

C57BL/6 mice (n = 3) implanted with temperature telemeters were studied at either 27° or 35°C T^awhile breathing air for 1 h as described above, then breathing air supplemented with H²S (80 ppm) for 30 min, and finally during a 30-min recovery period breathing air without added H²S. The oxygen concentration in the chamber was assayed every 5 min, using a PA-10a oxygen analyzer (Sable Systems, Las Vegas, NV). Oxygen consumption (μl O²· g⁻¹· min⁻¹) was calculated by open circuit respirometry (flow rate 0.5 l/min). The respiratory quotient was calculated as the ratio of carbon dioxide produced/oxygen consumed during H²S breathing in mice at 35°C T^a.

To determine whether inhaled H²S had analgesic effects, we examined the effects of H²S breathing on the response of mice to noxious stimuli as described previously.10Mice were placed in individual chambers (25 cm long and 5.0 cm in diameter). Each chamber was fitted with a rubber stopper at one end through which the mouse’s tail and a rectal temperature probe protruded. Groups of six mice breathed H²S in oxygen (1 l/min total gas flow). A gas sample was continuously drawn from the expiratory limb of the circuit to measure H²S concentration. The rectal temperature of the mice was maintained between 36.5° and 38.5°C using a heat lamp. Mice initially breathed 60 ppm H²S for 60 min, and the H²S concentration was subsequently increased in steps of 20 ppm up to 160 ppm, with 30 min allowed for equilibration after each increment of H²S concentration. A clamp (alligator clip) was applied to the tail for up to 1 min, and the mice were observed for movement in response to the stimulation. In every case, the tail was stimulated proximal to the previous test site. Only the middle third of the tail was used for tail clamping.

Plasma H²S concentration was measured with a sulfide-sensitive Ag²S electrode (model 9616 BNWP; Orion Research, Beverly, MA) as described previously.11–13In plasma, approximately one third of H²S molecules persist as the undissociated form (H²S) and the remaining two thirds as HS⁻at equilibrium with the H²S.1The Ag²S electrode measures the combined concentrations of both H²S and HS.14The electrode was calibrated against a standard solution produced by dissolving Na²S in distilled, deoxygenated water before each measurement. Fifteen C57BL/6 male mice were anesthetized using ketamine (0.1 mg/g) and xylazine (0.01 mg/g). Mice were intubated via  the trachea, and volume-controlled ventilation (ventilator model 687; Harvard Apparatus, Holliston, MA) was initiated at an RR of 120 breaths/min at an inspired oxygen fraction of 1.0 for the control group (n = 5). A second group of ventilated mice breathed 80 ppm H²S in oxygen (n = 4), and a third group of ventilated mice was exposed to 200 ppm H²S in oxygen (n = 6). After breathing H²S for 30 min, blood samples were collected with a heparinized syringe via  an apical cardiac puncture. Samples were centrifuged for 3 min at 8,000 rpm. The plasma samples were collected and diluted at a 1:1 ratio with a previously prepared solution of sulfide antioxidant buffer (SAOB II; Fisher Scientific, Pittsburgh, PA), according to the manufacturer’s instructions.

Statistical evaluations of multiple group comparisons (echocardiography data analysis) were performed with one-way analysis of variance using a Tukey post hoc  test. A two-way analysis of variance was used to compare data from mice breathing air or H²S at 27° or 35°C T^a. Interactions among variables were studied, and a Holm-Sidak analysis was used as a post hoc  test. If, during the statistical analysis of our data, the parametric assumption was violated (i.e ., normality test failed), a two-way analysis of variance on ranks with a Holm-Sidak post hoc  test was used. Data are presented as mean ± SE. Differences were considered significant if P  was less than 0.05.

Before exposure to H²S at 27°C T^a, the average T^bwas 37.4°± 0.4°C. Initiation of 80 ppm H²S inhalation led to a progressive decrease in T^breaching 33.8°± 1.9°C after 2 h at T^a27°C (fig. 1). After 6 h at H²S breathing, T^breached a steady state of 29.3°± 0.6°C (P < 0.05 compared with baseline). When H²S administration was discontinued, T^brecovered to 35.1°± 0.2°C after 3 h. In contrast, breathing H²S at T^a35°C did not alter T^bafter 6 h of H²S breathing.

In mice breathing 80 ppm H²S at T^a27°C, HR decreased by nearly 50% within 2 h (511 ± 13 beats/min to 265 ± 13 beats/min; P < 0.05; fig. 2A) and remained low as long as mice breathed H²S (measured up to 6 h). In addition, the cardiac rhythm became irregular, an observation that was confirmed by electrocardiogram. After H²S was discontinued, HR increased to 433 ± 34 beats/min after 2 h. Despite the marked reduction of HR, MAP did not change significantly from baseline during H²S inhalation (fig. 2B).

At a T^aof 35°C, breathing H²S reduced the HR approximately 50% after 2 h of exposure (498 ± 24 beats/min to 242 ± 13 beats/min; P < 0.05; fig. 2A), and the bradycardia persisted as long as mice breathed H²S. The effect of H²S breathing on HR did not differ between mice maintained at 27° or 35°C. MAP increased modestly during H²S breathing at 35°C after 6 h and was consistently greater in mice breathing H²S at 35°C than in mice breathing H²S at 27°C (P < 0.05; fig. 2B).

We recorded the electrocardiogram to better characterize the slow and irregular HR observed during H²S inhalation. Breathing H²S induced a sinus bradycardia with sinus arrest, typically starting after 1 h of H²S inhalation and lasting until H²S administration was terminated (fig. 3). After 5 min of breathing air, a rapid recovery to a regular and slow sinus rhythm (HR approximately 390 beats/min) had occurred. A complete recovery of the electrocardiographic tracing was seen after 2 h of breathing air (fig. 3).

An echocardiographic analysis of mice breathing H²S at 23°C T^aconfirmed that 80 ppm H²S inhalation decreased HR from 690 ± 19 beats/min to 233 ± 23 beats/min (P < 0.05; fig. 4A). A slow and irregular HR was noted during H²S breathing, consistent with the observations we obtained with telemetry. H²S breathing decreased CO by 60% (P < 0.05; fig. 4B), although SV remained unaffected (fig. 4C). After breathing H²S for 3 h, T^bwas decreased markedly (35.9°± 0.2°C to 26.5°± 0.2°C; P < 0.05). When mice breathed H²S at a T^aof 35°C, HR also decreased (690 ± 19 beats/min to 340 ± 16 beats/min; P < 0.05; fig. 4A). H²S breathing at 35°C T^aalso decreased CO as compared with baseline CO values (P < 0.05; fig. 4B), whereas SV was not affected (fig. 4C). T^bremained stable during the 3-h period while mice breathed H²S at 35°C T^a.

Breathing H²S reduced the activity level of mice. When mice breathed 80 ppm H²S at 27°C, activity slowly decreased, and after 3 h, mice became inactive and appeared asleep (fig. 5). Interestingly, these mice were able to move when stimulated by tapping on the chamber. When these animals subsequently breathed air without H²S, they regained baseline activity levels within 1 h. Similar reversible effects of H²S breathing on the activity level were observed when we repeated this experiment at 35°C T^a(fig. 5).

Breathing 80 ppm H²S at 27°C T^adecreased RR, reaching a maximal effect after 3 h of inhalation (RR 31 ± 4 breaths/min; P < 0.05). RR remained low as long as mice breathed H²S, rapidly returning to baseline after H²S gas breathing was discontinued. Mice regained a baseline level of RR after 2 h breathing air (115 ± 3 breaths/min). RR was also decreased in a group of mice that breathed H²S at the higher T^a(35°C). After 120 min of H²S breathing at 35°C T^a, the RR was significantly greater than that measured at 27°C T^a(P < 0.05; fig. 6).

We explored the possibility that changes in blood gas tensions or electrolyte concentrations could be responsible for the cardiovascular abnormalities observed in mice breathing H²S. We found that breathing H²S at 80 ppm for 2 h did not change either arterial blood gas tensions or electrolyte concentrations in mice (table 1).

In mice breathing 80 ppm to H²S at T^a27°C, the carbon dioxide production rate was reduced by 46% after 10 min of H²S inhalation (P < 0.05; fig. 7A) and remained at that level during the 30-min exposure to H²S. This reduction in carbon dioxide production was followed by a decrease in T^bof 3.4°C (39°± 1°C to 35°± 1°C; P < 0.05; fig. 7A). After breathing air for 15 min, the carbon dioxide production rate returned to baseline levels (49.8 ± 7 μm/min), and T^breturned to normal levels.

To separate the metabolic effects of H²S inhalation from the reduction of T^b, experiments were conducted at T^a35°C to maintain T^bunchanged during H²S breathing. Breathing 80 ppm H²S at T^a35°C decreased the whole body carbon dioxide production rate by 18% after 10 min (from 60 ± 1 μm/min to 49 ± 1 μm/min; P < 0.05) and by 23% after 30 min of exposure (fig. 7B), whereas T^bdid not change during the experimental period.

When mice breathed H²S (80 ppm) at T^a27°C, carbon dioxide production was reduced by approximately 35% after 2 min of inhalation (P < 0.05). The effect of H²S breathing on carbon dioxide production persisted during the 30-min exposure to H²S (fig. 8). Interestingly, during the initial 30 min of H²S breathing, HR did not change. When H²S was stopped after 30 min of inhalation and mice were allowed to breathe air, HR was greater than the baseline for the first 20 min and returned to the baseline level thereafter (P < 0.05; fig. 8). HR and carbon dioxide production rate returned to their baseline levels after 30 min of air breathing recovery.

To more completely characterize the effects of H²S inhalation on metabolism, carbon dioxide production was measured in mice that breathed 40, 80, or 120 ppm H²S for 30 min at a T^aof 35°C (fig. 9). Breathing 40 ppm H²S did not decrease carbon dioxide production after 30 min, but breathing 80 and 120 ppm H²S markedly reduced carbon dioxide production. When H²S breathing ceased, carbon dioxide production returned to baseline levels within 5 min at all three H²S concentrations.

Breathing H²S (80 ppm) at T^a27°C reduced the oxygen consumption rate by 30% after 10 min (25.8 ± 3 vs . 18.1 ± 1 μl O²· g⁻¹· h⁻¹; P < 0.05; fig. 10), and this reduction persisted during the 30-min exposure to H²S. This reduction in oxygen consumption was followed by a decrease in T^bof 1.9°C (37°± 0.4°C to 35.1°± 0.5°C; P = not significant). Within 10 min of discontinuing H²S, the oxygen consumption rate had increased, to a level higher than the baseline. Oxygen consumption returned to baseline levels after 30 min of breathing air (fig. 10).

When oxygen consumption was measured at T^a35°C, H²S breathing reduced the oxygen consumption rate within 20 min, reaching a 25% reduction after 30 min of exposure to the gas (26.4 ± 1 vs . 20.6 ± 1 μl O²· g⁻¹· h⁻¹; P < 0.05). T^bdid not change during the exposure period (37.4°± 0.1°C at baseline to 37.1°± 0.2°C before treatment; P = not significant). When H²S breathing was discontinued, the oxygen consumption rate recovered to baseline levels within 10 min (fig. 10).

The calculated respiratory quotient of mice breathing air at T^a35°C was compared with their respiratory quotient after 30 min of exposure to 80 ppm H²S. We found that breathing H²S at 80 ppm did not significantly alter respiratory quotient (from 1.00 ± 0.01 at baseline to 1.04 ± 0.04 at the end of the 30-min exposure to H²S at 80 ppm).

Although mice breathing H²S at concentrations from 60 to 160 ppm seemed less active, they moved when their tail was noxiously stimulated by tail clamping, suggesting a lack of significant analgesic effects of H²S breathing at these inhaled concentrations. We did not examine the analgesic effects of higher concentrations of H²S because all of the mice we studied died with marked respiratory depression when exposed to higher concentrations of inhaled H²S (data not shown).

The baseline plasma sulfide concentration in C57BL/6 mice was measured at 23.8 ± 0.2 μm. Breathing 80 ppm H²S for 30 min did not alter plasma sulfide levels (23.5 ± 0.2 μm; P = not significant vs . baseline). Plasma sulfide levels modestly increased in mice breathing 200 ppm H²S for 30 min (from 23.8 ± 0.2 to 24.2 ± 0.3 μm; P < 0.05).

The primary aim of the current study was to examine the cardiovascular effects of H²S breathing at a concentration that has been reported to depress metabolism in mice.6In addition, we sought to examine the thermal dependence of the cardiac and metabolic effects of H²S breathing because hypothermia per se  affects both hemodynamics and metabolism. We report that HR and CO, but not MAP and SV, were markedly reduced by H²S breathing, primarily due to the occurrence of a sinus bradycardia independent of body temperature. We also found that when the core body temperature is held constant, breathing H²S markedly and reversibly inhibits metabolism in mice. These observations provide evidence that H²S breathing modifies cardiac and metabolic function independent of its effects on core body temperature.

The most striking hemodynamic change associated with H²S breathing was the profound bradycardia in normothermic as well as hypothermic mice. The thermal independence of the bradycardia produced by H²S was also suggested by the observation that the reduction of HR preceded the reduction of T^bof mice breathing H²S at T^a27°C. This bradycardic effect of H²S breathing was accompanied by a reduction in CO (measured by echocardiography) because H²S breathing did not change the SV. On the other hand, H²S-induced bradycardia was preceded by a marked reduction of the carbon dioxide production rate (fig. 8). The latter observation suggests that the inhibitory effects of H²S on metabolism may also cause the reduction of HR and CO.

Because H²S breathing induces a marked and irregular bradycardia without changing the SV, it is tempting to speculate that H²S breathing adversely affects sinus node function but not myocardial contractility per se . To examine this hypothesis, we obtained a more detailed analysis of the irregular slow HR using telemetric electrocardiographic recordings. They revealed a profound sinus bradycardia with periods of sinus arrest. Although the precise mechanism responsible for the H²S-induced depression of sinus activity is unknown, it was not caused by abnormalities in blood pHa or blood gas tensions or electrolyte concentrations after H²S exposure (table 1). Further studies are needed to elucidate the precise impact of H²S breathing on myocardial function and the cardiac conduction system.

Despite a marked reduction of HR and CO, MAP did not change in mice breathing H²S at T^a27°C, suggesting that a compensatory systemic vasoconstriction occurred. H²S-induced vasoconstriction is unlikely to be due to hypothermia alone, because MAP was also increased in mice breathing H²S with an unchanged body temperature. Previous studies have shown that H²S can produce a potent vasodilation,11,15,16by relaxing smooth muscle cells, presumably by activation of adenosine triphosphate–sensitive K⁺channels.17,18In contrast, Dombkowski et al .19demonstrated that H²S has both vasodilatory and vasoconstrictory properties. A recent study by Ali et al .20also reported the biphasic effects of NaHS on vascular tone: vasoconstriction at lower concentration due to scavenging of endothelial nitric oxide by H²S, and vasodilation at higher concentrations via  activation of adenosine triphosphate–sensitive K⁺channels. It is conceivable that H²S breathing at 80 ppm caused vasoconstriction in our current study because the concentration of H²S in peripheral vascular smooth muscle cells was likely to be low. Alternatively, it is also possible that systemic vasoconstriction (and bradycardia) was triggered by cardiovascular reflexes (likely via  the aortic baroreceptor) activated in response to reduced CO after H²S-induced metabolic depression (i.e ., decreased oxygen consumption).

Mice breathing 80 ppm H²S for 6 h at T^a27°C decreased their T^bmore than 8°C. This reduction of T^band activity is in accord with the results previously reported by Struve et al .21and Blackstone et al .6in rats and mice, respectively, exposed to H²S. It has been reported that H²S (produced from the donor molecule Na²S) inhibits cytochrome c  oxidase activity.22It is possible that mitochondrial thermogenesis may be reduced as a consequence of the inhibition of cytochrome c  oxidase in heat-producing tissues (e.g ., brown fat), decreasing the ability of mice to generate sufficient heat to maintain their body temperature. Our observation that H²S-induced decrements of both carbon dioxide production and oxygen consumption preceded the reduction in T^bwhen mice breathed H²S at T^a27°C suggests that inhibition of metabolism is primary and reduction in T^bis secondary. This hypothesis is further supported by the observation that breathing H²S at T^a35°C markedly reduced carbon dioxide production and oxygen consumption without decreasing T^b.

Inhalation of 80 ppm H²S also decreased the RR in mice exposed to the gas at 27° or 35°C, whereas the magnitude of the reduction was more marked at a lower T^a. It has been suggested that H²S may selectively affect the brain stem4,23and inhibit respiratory drive.1,2The inhibitory effects of H²S on brain stem function may be related to its inhibitory effects on cytochrome c  oxidase and mitochondrial adenosine triphosphate production in the brain.22Interestingly many organisms respond to adverse environmental conditions by reducing their RR. Although the mechanisms responsible for the reduction in RR in adverse conditions are incompletely understood, this response is usually correlated with a decreased metabolic rate24–26and changes of the intrinsic properties of the respiratory rhythm generator in the brain stem.20On the other hand, a previous study showed that inhalation of H²S up to 400 ppm for 3 h did not change sulfide concentration or cytochrome c  oxidase activity in the hindbrain of rats, whereas sulfide concentrations increased and cytochrome c  oxidase activity decreased in lung and nasal respiratory epithelium.5These results suggest that respiratory inhibition by H²S inhalation may be partially mediated via  effects on chemoreceptors in the lung and upper airways. Alternatively, H²S-induced decrements in RR may also reflect the reduced need for minute ventilation due to the marked reduction of carbon dioxide production induced by H²S breathing. The observation that a reduction in carbon dioxide production precedes the decrements of RR supports this hypothesis.

Our observation that H²S breathing at 80 ppm markedly depressed spontaneous activity in mice is consistent with a previous report by Struve et al .21in which rats exposed to higher concentrations of H²S (e.g ., 80–400 ppm) exhibited reduced spontaneous ambulation. Our results do not support a major role for hypothermia in the H²S-induced reduction in spontaneous activity, because mice breathing H²S with stable T^balso exhibited a marked reduction in spontaneous activity. Furthermore, the reduction in activity of mice breathing H²S at T^a27°C preceded their reduction in T^b. These results suggest that hypothermia is not solely responsible for the H²S-induced depression of spontaneous activity. Although H²S can produce a profound dormant state in mice, it does not seem to have profound anesthetic effects in the range of concentrations that we tested (60–160 ppm). Although higher concentrations of inhaled H²S may have produced analgesia, the higher H²S concentrations were not tested in our model due to H²S-induced respiratory depression.

To estimate the amount of systemically absorbed H²S by H²S inhalation, we measured plasma sulfide concentration ([H²S]+[HS⁻]) in mice. The baseline plasma sulfide levels observed in the current study are comparable with values previously reported in mice (approximately 23 μm)27and Sprague-Dawley rats (approximately 46 μm)11as well as the levels reported in human plasma (approximately 34 μm).12Compared with controls, breathing H²S at 200 ppm for 30 min increased plasma sulfide levels in mice, demonstrating that inhaled H²S is systemically absorbed. The absence of an increase in the plasma sulfide concentration after H²S breathing at 80 ppm may suggest that absorbed H²S is converted to other sulfur-containing molecules that are not readily measured by the Ag²S electrode. The major metabolic pathway for H²S is the rapid multistep hepatic oxidation of sulfide to sulfate (SO⁴²⁻) and the subsequent elimination of sulfate in the urine.1,28H²S may also be methylated (e.g ., CH³SCH³), or it can react with cytochrome c  and other metallo- or disulfide-containing proteins. It is likely that the H²S that reacts with cytochrome c  and other proteins has a larger impact on metabolism than H²S in plasma. Alternatively, the failure to detect any increase of plasma sulfide concentration after inhaling H²S at 80 ppm for 30 min may be related to the limitation of our experimental procedures. Although care was taken to avoid the escape of H²S from blood during our sulfide measurements, it is possible that some H²S was lost during sample collection and processing (approximately 10 min elapsed after mice were killed for sampling and measurements).

In summary, inhalation of H²S reversibly and rapidly decreases HR and CO without changing SV and MAP. This was independent of the effects of H²S on core body temperature. H²S breathing markedly and reversibly depresses metabolism in mice, an effect that is also independent of hypothermia. Inhalation of H²S decreases HR and CO without changing SV and MAP, and this effect is also independent of any effects on core body temperature. The rapid inhibition of metabolism produced by H²S breathing may be exploitable, either by gas inhalation or the injection of H²S donors to protect cellular or organ function when the supply of nutrients and oxygen are jeopardized, such as after major trauma or a cardiac arrest. Further studies of the safety of H²S inhalation and the ability of H²S inhalation to be effective in larger species will be required.