4.1. Ventilation-Induced Lung Injury and Systemic Inflammation
Ventilator-associated lung injury (VILI) includes barotrauma, volutrauma, atelectrauma, and biotrauma, among which volutrauma, the overexpansion produced by high tidal volume, is the major cause of VILI [1, 16, 17]. Using histopathological analysis, we found that ventilated rats had lung injuries, with a significantly higher LIS than basal rats. Moreover, the LIS was significantly higher in the HVt group compared with that in the LVt group, indicating that MV induced lung injury, and the extent of damage increased with increasing tidal volume size, which was consistent with previous studies.
Some scholars use the concept of “organ crosstalk” to describe the effects of mechanical ventilation on multiple organs and the interactions between them [4]. However, how the lung-brain crosstalk occurs is unclear. Studies have shown that the ventilation-induced inflammatory response not only aggravates lung injury, but also leads to distal organ damage through circulation [18–21], which is the “biotrauma” mentioned above. Some researchers have suggested that inflammatory mediators in the blood can be sensed by the brain; thus, altering the permeability of the blood-brain barrier might be one possible mechanism of ventilation-related brain injury [22, 23]. However, previous studies focused on the lung and peripheral inflammation, and few addressed the inflammation status of the CNS [18, 19]. We measured IL-6 and TNF-α in the serum and brain and found that MV increased the blood IL-6 and TNF-α significantly, irrespective of the tidal volume level, whereas in the brain, a significant decrease in IL-6 was observed in the HVt group compared with that in the basal group. MV induced lung injury and the release of inflammatory cytokines into the circulation, but this systemic inflammation did not affect the brain, indicating that inflammation is not the only cause of ventilator-related brain injury.
4.2. Effect of MV on the Brain and TASK-1 Expression
Previous studies showed MV might promote brain activation [18] and influence cerebral tissue oxygenation and metabolism [20] and even brain tissue damage and disintegration of the blood-brain barrier [24–26]. TASK-1, is a two-pore (2P) domain potassium channel, which plays key role in the regulation of cell membrane potential [9, 10]. It is expressed widely in the CNS, especially in brainstem respiratory neurons and motor neurons [9, 27], including the pre-Bötzinger complex of the medulla oblongata, which is the generator of the basic respiratory rhythm [9, 11–13]. Therefore, TASK-1 plays a role in respiratory rhythm generation [10, 12, 14]. Inhibition of TASK-1 leads to an increased excitability of brainstem respiratory neurons, through depolarization of the cell membrane [11], thus enhancing the respiratory drive [13, 14]. Our results showed a significant decrease in TASK-1 channel levels in both the HVt and LVt groups, and there was a trend for the HVt group to have a lower level of TASK-1 compared with that in the LVt group. This suggested that MV might have an effect on respiratory centre function via changing the excitability of respiratory neurons, and this effect might correlate with the tidal volume level.
S100B is a calcium binding protein physiologically produced and released predominantly by astrocytes, because their levels may increase in CSF and blood in several brain pathologies including cell death [24–26]; it is considered to be a marker of neuronal damage. Bickenbach et al. [20] investigated the effects of different tidal volume ventilation on cerebral tissue oxygenation and metabolism in pig and found that low tidal volume (6 ml/kg) group significantly improved brain tissue oxygenation and micrometabolism compared to high tidal volume (12 ml/kg) group. In addition, serum S100B protein levels significantly increased in high tidal volume group compared to baseline level, indicating neuronal damage. In pig model of hypoxemia, Fries et al. [28] found increase of S100B in ALI pig serum, while visible damage was observed in hippocampal neurons under microscope. Also, studies showed that S100B over a certain threshold (ng/ml as a unit of measurement) [29, 30] are indicators of prior brain damage and bear clinical significance as predictors of poor outcome. As neurologic injury markers in blood are hampered by many confounding factors, some scholars proposed that biochemical markers in brain might be more accurate reflectors of cerebral pathological changes [24, 30]. But there are few studies comparing the prognostic value of S100B in serum and brain (CSF or brain homogenates) and the result differs [31, 32]. We for the first time measured S100B in both serum and brain during mechanical ventilation and observed that, in serum, S100B increased significantly in HVt group, which suggested neuronal damage in this group. In brain, LVt group had a higher S100B level than basal group, whereas there was a decrease in S100B level in HVt group.
MV can induce apoptosis in distant organs (including the kidney, intestine, heart, and liver) through different pathways [22]. Although many studies have speculated that potassium channels induce neuronal apoptosis [33, 34] or immune inflammatory injury [35], increasing evidence suggests that an important emerging therapeutic mechanism underlying neuroprotection is the activation/opening K2P channels. Inflammatory plaques of human multiple sclerosis patients displayed profoundly lowered expression of TASK isoforms [15]. Rao et al. found that traumatic brain injury led to a downregulation of potassium channels (RK5, TWIK, and X62859) in the injured cortex, leading to decreased posttraumatic axonal conductance and epilepsy [36]. Liu et al. determined that K2P channels (TASK-1, 2, 3) protected cells effectively from cytotoxic stress by preventing activation of apoptotic pathways [37]. Increased membrane excitability by TASK channel inhibition could contribute to increased electrical activity and subsequent neuronal degeneration caused by intracellular sodium and calcium accumulation, which is known as “excitotoxicity,” the most common pathological mechanism leading to neuronal death. Meuth et al. [15] observed hypoxia depolarized central neurons after specific inhibition of TASK-1; they proposed that upregulation of functional TASK channel expression might exert a neuroprotective effect by dampening neuronal excitability. Neurons expressing nitric oxide (NO) synthase (NOS-I), which is upregulated in many human chronic neurodegenerative diseases, are highly susceptible to neurodegeneration. González-Forero et al. [38] showed that the autocrine activation of the neuronal NO/cGMP pathway induced by XIIth nerve injury enhanced excitability of the motor neuron pool and fully suppressed TASK currents via a protein kinase G- (PKG-) dependent mechanism. Finally, they proposed a hypothesis whereby TASK channel inhibition via persistent autocrine activation of the NO/cGMP/PKG cascade could sensitize NOS-expressing neurons to excitotoxic damage in brain neurodegenerative processes via a sustained increase in their excitability. Moreover, Mazzone and Nistri [39] used a validated in vitro model of spinal cord injury induced by kainate-mediated excitotoxicity to explore relation of S100B levels and damage severity and found that S100B represents a useful biomarker of lesion progression as its level is related to the occurrence and severity of neuronal loss due to excitotoxicity. Therefore, MV might cause neuronal damage directly in the brain or cause apoptosis by reducing the protective factor-TASK-1 channel expression. TASK channel inhibition-mediated sensitization of neurons to excitotoxic damage might be the mechanism of the latter [36].
Our study had certain limitations: (1) as anaesthetized animals soon develop respiratory failure, a group of spontaneously breathing rats was not set in this study; thus the results did not allow us to discriminate the role played by surgical procedures. (2) We hypothesized that mechanical ventilation affects TASK-1, thereby affecting the respiratory centre activity, based on the fact that TASK1 is closely related to respiratory neuron excitability. Functional change in neuron physiology was not determined in our study. As leak conductance of TASK-1 channel under different ventilatory settings can not be measured in vivo via patch clamp techniques, a better experimental design is needed in future research to determine functional status of TASK-1 channel during MV. (3) We measured S100B as a biomarker of neuronal damage and found the change of serum S100B was consistent with literature, indicating neuronal damage in HVt, but in brain, we did not observe correlation between S100B levels and tidal volume levels. Ahmed et al. [40] had found the CSF levels of biomarkers including S100B had time-dependent changes, with first peak appearing at 6 h after the injury and second peak at 24 h or 72 h, indicating that determining S100B continuously at different time intervals might reflect neuronal injury of brain more accurately. Moreover, in subsequent studies we should detect the neuronal apoptosis by TUNEL method in frozen slices of brainstem from different groups to reflect neuronal apoptosis directly and verify the significance of serum or brain S100B in predicting neuronal damage during mechanical ventilation. (4) To explore the mechanism through which MV influenced the brain, we measured IL-6 and TNF-a in serum and brain, respectively, to reflect the difference of inflammatory response in peripheral and central nervous system and obtained some different result from previous studies [22, 23]. Taking the fact that proinflammatory cytokines IL-6 and TNF-a could be influenced by many confounding factors and blood-brain barrier as well, we would like to assess immune cells in both serum and brain during MV in subsequent studies to further reflect peripheral and central inflammatory status. Moreover, we would like to differentiate and quantify immune cells in bronchoalveolar lavage fluid of different groups in supplementary experiment to explore the effect of MV on the immunoinflammatory status of lung, peripheral blood, and CNS, respectively.