Altered respiratory responses to hypoxia in mutant mice deficient in neuronal nitric oxide synthase

Abstract

1. The role of endogenous nitric oxide (NO) generated by neuronal nitric oxide synthase (NOS-1) in the control of respiration during hypoxia and hypercapnia was assessed using mutant mice deficient in NOS-1. 2. Experiments were performed on awake and anaesthetized mutant and wild-type control mice. Respiratory responses to varying levels of inspired oxygen (100, 21 and 12% O2) and carbon dioxide (3 and 5% CO2 balanced oxygen) were analysed. In awake animals, respiration was monitored by body plethysmograph along with oxygen consumption (VO2), CO2 production (VCO2) and body temperature. In anaesthetized, spontaneously breathing mice, integrated efferent phrenic nerve activity was monitored as an index of neural respiration along with arterial blood pressure and blood gases. Cyclic 3',5'-guanosine monophosphate (cGMP) levels in the brainstem were analysed by radioimmunoassay as an index of nitric oxide generation. 3. Unanaesthetized mutant mice exhibited greater respiratory responses during 21 and 12% O2 than the wild-type controls. Respiratory responses were associated with significant decreases in oxygen consumption in both groups of mice, and the magnitude of change was greater in mutant than wild-type mice. Changes in CO2 production and body temperature, however, were comparable between both groups of mice. 4. Similar augmentation of respiratory responses during hypoxia was also observed in anaesthetized mutant mice. In addition, five of the fourteen mutant mice displayed periodic oscillations in respiration (brief episodes of increases in respiratory rate and tidal phrenic nerve activity) while breathing 21 and 12% O2, but not during 100% O2. The time interval between the episodes decreased by reducing inspired oxygen from 21 to 12% O2. 5. Changes in arterial blood pressure and arterial blood gases were comparable at any given level of inspired oxygen between both groups of mice, indicating that changes in these variables do not account for the differences in the response to hypoxia. 6. Respiratory responses to brief hyperoxia (Dejours test) and to cyanide, a potent chemoreceptor stimulant, were more pronounced in mutant mice, suggesting augmented peripheral chemoreceptor sensitivity. 7. cGMP levels were elevated in the brainstem during 21 and 12% O2 in wild-type but not in mutant mice, indicating decreased formation of nitric oxide in mutant mice. 8. The magnitude of respiratory responses to hypercapnia (3 and 5% CO2 balanced oxygen) was comparable in both groups of mice in the awake and anaesthetized conditions. 9. These observations suggest that the hypoxic responses were selectively augmented in mutant mice deficient in NOS-1. Peripheral as well as central mechanisms contributed to the altered responses to hypoxia. These results support the idea that nitric oxide generated by NOS-1 is an important physiological modulator of respiration during hypoxia.

Figure 1. Respiratory responses to varying levels of inspired oxygen in unanaesthetized wild-type and mutant mice

A, representative tracing of respiratory responses to three levels of inspired oxygen in an unanaesthetized, freely moving wild-type and mutant mouse. 100 % O₂, 21 % O₂ and 12 % O₂ indicate inspired oxygen levels. Note the greater increases in respiration in the mutant mouse during 21 and 12 % O₂. B, comparison of respiratory responses during 21 and 12 % O₂ in wild-type and mutant mice. The results are presented as a percentage of 100 % O₂. Data presented as means ±s.e.m. of eight each of wild-type and mutant mice. *P < 0.05 (ANOVA). Note the respiratory responses (RR, V_T and V̇_E) were greater in mutant than wild-type mice during 21 and 12 % O₂.

Figure 2. Comparison of changes in O₂ consumption (V̇_O2), CO₂ production (V̇_CO2) and respiratory quotient (RQ) during 21 and 12 % O₂ in unanaesthetized wild-type and mutant mice

The results are presented as a percentage of 100 % O₂. These data are analysed from the same mice as in Fig. 1 and presented as means ±s.e.m.*P < 0.05 (ANOVA). Note that the decrease in V̇_O2 was significantly greater in mutant than wild-type mice during 21 and 12 % O₂.

Figure 3. Respiratory responses to varying levels of inspired oxygen in anaesthetized wild-type and mutant mice

A, representative tracing of an experiment illustrating respiratory responses during 100, 21 and 12 % inspired oxygen in an anaesthetized wild-type and a mutant mouse deficient in NOS-1. BP (mmHg), arterial blood pressure; Int. Phr., integrated efferent phrenic nerve activity. 100 % O₂, 21 % O₂ and 12 % O₂ indicate inspired oxygen levels. Respiration (respiratory rate and amplitude of tidal phrenic activity) increased in response to 21 % O₂ in both mice. Respiration was depressed in the WT mouse during 12 % O₂, whereas it remained elevated in the mutant mouse. B, comparison of respiratory responses during 21 and 12 % O₂ in anaesthetized wild-type and mutant mice. The results are presented as a percentage of 100 % O₂. Data presented as means ±s.e.m. of wild-type (n = 10) and mutant (n = 14) mice. *P < 0.05 (ANOVA). Note the respiratory rate (RR) and minute neural respiration (Min. Neural. Resp.) were greater in mutant than wild-type mice during 21 and 12 % O₂. Ampl. Phr. Activity, amplitude of phrenic activity.

Figure 4. Respiratory oscillations in mutant mice

A, representative tracing of an experiment illustrating the periodic increases in respiratory rate and tidal phrenic activity during inspired 21 and 12 % O₂ in an anaesthetized, spontaneously breathing NOS-1-deficient mouse. BP (mmHg), arterial blood pressure; Int. Phr., integrated phrenic nerve activity; Raw Phr., action potentials recorded from the efferent phrenic nerve. 100 % O₂, 21 % O₂ and 12 % O₂ indicate inspired oxygen levels. Note the periodic increases in respiratory rate during 21 and 12 % O₂ breathing but not during 100 % O₂. Because of the marked increases in respiratory rate, the integrator was not able to reset back to baseline. The oscillatory changes in respiration could be seen in the absence of oscillations in blood pressure. B, average data of the time interval between oscillations during two levels of inspired oxygen. Data presented are means ±s.e.m. from 5 mutant mice. *P < 0.05 (paired t test).

Figure 5. Respiratory responses to brief hyperoxia in anaesthetized wild-type and mutant mice

A, representative tracing of an experiment illustrating the effect of brief hyperoxia (Dejours test) on efferent phrenic nerve activity in an anaesthetized, spontaneously breathing wild-type and mutant mouse. Int. Phr., integrated efferent phrenic nerve activity. Animals breathed room air. At arrow, 100 % O₂ was added to the inspired air. 100 % O₂ caused prompt decreases in respiration in both types of mice, but the response was more pronounced in mutant mice. B, average data for changes in respiratory responses. Respiratory variables were analysed during the last 15 s of hyperoxia (for details see text). Changes in respiratory variables are presented as a percentage of 21 % O₂ controls. Data presented are means ±s.e.m. from 12 wild-type and 20 mutant mice. *P < 0.05 (ANOVA). Note the greater decreases in respiratory rate by hyperoxia in NOS-1-deficient mice.

Figure 6. Respiratory responses to sodium cyanide in anaesthetized wild-type and mutant mice

A, representative tracing of an experiment illustrating the effect of systemic administration of sodium cyanide on efferent phrenic nerve activity in anaesthetized, spontaneously breathing wild-type and NOS-1 deficient mice. Cyanide (50 μg kg⁻¹i.v.) was injected at the arrow. Raw Phr., action potentials recorded from efferent phrenic nerve activity; Int. Phr., integrated phrenic nerve activity. NaCN caused prompt increases in respiration in both mice, but the response was more pronounced in mutant mice. Because of the marked increases in respiratory rate, the integrator was not able to reset back to baseline. B, average data for changes in respiratory variables. Results are presented as a percentage of pre-injection controls. ‘S’ denotes respiratory changes after saline (vehicle) injection. Data presented as means ±s.e.m. from 7 animals in each group. *P < 0.05 (ANOVA). Note the enhanced responses in respiratory rate and neural minute respiration at 50 and 100 μg kg⁻¹ cyanide in mutant mice.

Figure 7. cGMP and Western blot analysis in wild-type and mutant mice

A, analysis of cGMP levels in brainstem during three levels of inspired oxygen, i.e. 100, 21 and 12 % O₂, in wild-type and mutant mice. Data presented as means ±s.e.m. from 9 animals in each group. Note the increases in cGMP levels during 21 and 12 % O₂ in wild-type (WT) mice (P < 0.05, paired t test), whereas they remained fairly constant in mutant mice (P > 0.05, paired t test). B, representative autoradiogram illustrating the absence of NOS-1 protein in tissue extracts of the brainstem from mutant mice. NOS-1 protein was analysed in the brainstem extracts by Western blot analysis using an antibody directed against NOS-1. The antibody recognizes the 155 kDa NOS-1 protein. Rat pituitary lysate was used as a positive control (Control). Note the presence of the NOS-1 protein in WT mice, but not in mutant mice.

Figure 8. Comparison of hypercapnic (3 and 5 % CO₂ balanced oxygen) respiratory responses in unanaesthetized (A) and anaesthetized (B) wild-type and mutant mice

The results are presented as percentage of 100 % O₂. A, in unanaesthetized mice, data are presented as means ±s.e.m. of 12 wild-type and 9 mutant mice. B, in anaesthetized mice, data are presented as means ±s.e.m. of 7 of each group. Note the magnitude of increase in respiratory variables in unanaesthetized and anaesthetized mice was similar in both groups of mice (P > 0.05, ANOVA).