In the course of the development of a mouse model of human EV71 infection, we demonstrated that EV71 could spread from the gastrointestinal tract or muscle to and accumulate in the CNS, especially in the brainstem of mice after different routes of inoculation . Following neuroinvasion, the animals exhibited paralysis, pulmonary dysfunction, and death, but did not develop PE, the hallmark in human EV71 patients with severe complications . In this study, we demonstrated that EV71-infected mice displayed a cytokine profile that was distinct from that of human patients. The NTS was not the CNS target of EV71 infection as viral titer was extremely low in this area of infected mice, which might explain why PE was absent in the infected animals.
Clinical findings showed that increases in serum IL-1β, IL-6, IL-10, IL-13, TNF-α, and IFN-γ were associated with PE development in EV71 patients [7, 13, 14]. In this study, the EV71-infected mice exhibited only a subtle and transient elevation of serum IL-6, IL-10, and IFN-γ. Thus, the absence of PE in EV71-infected mice might be attributed to inadequate proinflammatory cytokine levels of IL-6, IL-13, and IFN-γ. Indeed, the introduction of exogenous IL-6, IL-13, and IFN-γ in EV71-infected mice exacerbated pulmonary dysfunction and resulted in increases in lung weight/body weight ratio and intra-alveolar exudation. In addition, treatment with IL-6 and IFN-γ was sufficient for provoking the observed effect. Both IL-6 and IFN-γ are related to PE development via their effects on endothelial cells and inflammation , , and both of these cytokines were important for microvascular angiogenesis and leakage in an atopic dermatitis mouse model . Recent animal study indeed demonstrated the pathological role of IL-6 in the induction of tissue damage and mortality in EV71-infected mice . IL-13 alone exacerbated only pulmonary dysfunction in our EV71-infected mice, which is consistent with the biological function of the cytokine. IL-13 is a potent effector cytokine for asthma, as it can directly cause airway hyperreactivity, mucus over-production, increased pulmonary vessel permeability, and smooth muscle hypertrophy, effects that have been shown to contribute to pulmonary dysfunction and respiratory distress . Taken collectively, our data suggest that elevated serum IL-6 and IFN-γ contribute significantly to PE development during EV71 infection in mice.
Based on the distribution of viral antigens and viral genomic sequences, EV71 is likely propagated throughout the CNS via motor pathways . Our previous study indicated a retrograde axonal transport of the virus in neuronal cells , which serves as strong evidence that EV71 preferentially infects certain tissues or cell types in the CNS. Both histopathological  and magnetic resonance imaging studies ,  of EV71 patients with PE showed that the major CNS lesions were in the posterior medulla oblongata, pons, midbrain, dentate nuclei of the cerebellum, and ventral horns of the cervical spinal cord. In addition, EV71 viral antigens and genomic sequences were detected primarily in neurons and neuronal processes under inflammatory conditions . The fact that exogenous IL-6, IL-13, and IFN-γ worked only in EV71-infected, but not HSV-1-infected mice, suggests a prerequisite for a region-specific CNS infection for PE development.
Surprisingly, PE did not developed in adult mice after a direct inoculation of EV71 to the NTS, an area known to contribute to the development of PE in rats upon injury [24, 25]. In our current study, the neurons in the NTS of adult mice seemed to be more resistant to EV71 than those in the ventral medulla. Thus, it is possible that the inability of EV71 to cause PE was due to the lack of NTS damage and perhaps neurogenic cardiopulmonary complications.
Besides an over-stimulated proinflammatory response, sympathetic excitement has been proposed as another mechanism involved in the development PE in EV71 patients . Imbalances in sympathetic functions have been observed to cause systemic or pulmonary hypertension which may contribute to the development of PE due to hemodynamic alterations and hydrostatic overpressure in mice , other animals [26, 27], and human beings [28, 29]. Moreover, neurogenic stimulation may also induce movement of plasma proteins into the airway lumen in rats . The plasma noradrenaline levels were not altered in EV71-infected mice. Thus, it may worthwhile to test the effects of catecholamine solely or in combination with the cytokines with regard to PE development.
Collectively, we demonstrated that IL-6, IL-13, and IFN-γ over-stimulation exacerbated pulmonary dysfunction and clinical symptoms, and provoked a mild PE in EV71-infected mice. A synergistic proinflammatory cytokine response and damage to specific brain regions, or more precisely specific neuronal cells, may be necessary for the development of EV71-induced PE.