Gozal et al. reply

When concentrations of SNO delivered to the dorsocaudal brainstem match the magnitude of peak HVR, the time constants of VSTP after hypoxia and SNO are similar (our unpublished results). Thus VSTP, a principal component of HVR, is critically dependent on SNO formation and its subsequent activity in brainstem structures. These results are supported by data that demonstrate the NOS-dependency of both VSTP and long-term facilitation following hypoxia7.

On the contributions of hypocapnia and ramp-presentation of the hypoxic stimulus to the ventilatory undershoot, following cessation of hypoxic gas administration: isocapnic hypoxia and step presentation of hypoxic stimuli to γ-GT-knockout and wild-type mice confirm the anticipated absence and presence of VSTP, respectively (our unpublished results).

S-nitrosoglutathione may not be the only SNO that signals HVR, because other SNOs are also formed during erythrocyte deoxygenation and NOS activation8,9. We used GSNO as a reporter SNO, but found that other l-isomeric SNOs mimic HVR, as indicated by the large increase in HVR in humans treated with N-acetyl-l-cysteine10.

Peripheral chemoreceptors are therefore important in initiating the ventilatory response to hypoxia, and SNO signalling is crucial in determining the characteristics of the HVR. It is not necessary to assume that these two features are mutually exclusive.