Reply to: Neuroimmune interactions and COVID-19 in lung transplant recipients

We would like to thank Joan Arenas-Prat for his correspondence on our Perspective (Lung innervation in the eye of a cytokine storm: neuroimmune interactions and COVID-19. Nat. Rev. Neurol. 16, 645–652 (2020))¹, which raises some interesting issues (Neuroimmune interactions and COVID-19 in lung transplant recipients. Nat. Rev. Neurol. https://doi.org/10.1038/s41582-021-00484-x (2021))².

In our Perspective, we proposed a novel therapeutic approach to COVID-19 based on a model of neuroimmune interaction in the lungs, in which we suggest that lung inflammation is modulated by lung-innervating nerve fibres. This model implies that targeted modulation of lung innervation, especially vagal fibres, which account for the majority of lung innervation, could help resolve life-threatening lung inflammatory reactions in pathological situations such as COVID-19. As lung transplant surgery results in acute complete denervation of the lungs, such a condition could be of help to better characterize the clinical relevance of our neuroimmune communication model.

In his Correspondence², Arenas-Prat discusses our Perspective in light of a recent publication from Saez-Giménez B. et al.³, stating that “a recent multicentre study showed that, although lung transplant recipients have a clinical presentation of SARS-CoV-2 infection similar to that of the general population, they have a worse prognosis and higher mortality compared with non-transplanted patients.” However, he argues that lung transplantation will also result in the elimination of sensory C-fibres, which would block the release of inflammatory neuropeptides such as substance P and calcitonin gene-related peptide, and lead to the elimination of parasympathetic innervation, which might result in an anti-inflammatory outcome. In addition, he mentions studies in patients who underwent surgical joint denervation, which showed that loss of sensory inputs at the joint relieved osteoarthritic pain and synovial inflammation^4,5. On the basis of these data, the author suggests that sensory denervation could have similar effects in the lungs, which might be at odds with our model whereby vagus innervation is proposed to have a net anti-inflammatory function.

Although we agree with the author that crosstalk between vagal parasympathetic and sensory neuroimmune modulation will be taking place, the net pro-inflammatory or anti-inflammatory outcome is likely to be tissue-specific and will depend on the inflammatory stimulus. The data from Saez-Giménez B. et al.³ provide exciting clinical evidence in support of our model, as acute lung denervation due to transplant surgery was associated with adverse outcomes from COVID-19. Indeed, in line with our proposed model, these data suggest that lack of lung innervation could worsen SARS-CoV-2 infection outcomes, mirroring the poor COVID-19 prognosis observed in elderly individuals⁶, in whom the vagal tone is reduced.

Although these data do not systematically prove our model, they strengthen the idea that the activity of vagal fibres innervating the lungs could be manipulated to prevent maladaptive inflammation in patients affected by COVID-19, and possibly by other respiratory viruses.

Lastly and most importantly, these findings strongly advocate the urgent need for systematic studies to shed light on fibre-specific tissue innervation and its interaction with the immune system in tissue-specific pathophysiological inflammatory conditions, including but not limited to COVID-19.