Our results show for the first time, that a severe lung inflammatory reaction induced by orotracheal instillation of whole gastric fluid elicits a liver acute phase response, with elevation of acute phase proteins in the blood stream. The main lung antiproteases are part of this response, since mRNA transcription of α1-AT and mainly of α2-MG, is induced early in the liver, preceding elevation of the acute phase proteins in blood. Our results also show that the liver response contributes to the enrichment of the alveolar spaces with acute phase proteins that are important lung antiproteases. We identified a two-phase response in antiprotease enrichment of the alveolar spaces after aspiration, related to an early increase in alveolar-capillary barrier permeability and a later increment in capillary-alveolar antiprotease concentration gradient, due to increased blood concentration of antiproteases by increased liver synthesis.
Biology of the acute phase response
The systemic inflammatory response is a collection of coordinated physiologic changes initiated during early stages of inflammation as part of the early innate defense, triggered by a variety of stimuli with the goal of achieving homeostasis. A prominent feature of this response is synthesis of several proteins mainly by the liver, under the control of cytokines originating at the site of injury [
16], referred to as the liver acute phase response (APR). Changes in blood during the hepatic APR have been studied in rodents in response to stimuli not affecting the lung primarily, such as LPS and turpentine administration and exposition to thermal injury, and a number of APPs have been shown to participate. With these stimuli, the rodent APR starts a few hours after the insult, reaches a peak 24–48 h later and is back to baseline values at day 7 [
17,
18,
33]. In general, it is assumed that elevation of APPs in blood have beneficial effects based largely on the known function of each of the individual proteins involved and on logical speculations as to how these might be useful in inflammation [
16]. However, in the last few years, some negative implications of the hepatic APR have also been proposed, since the increase in blood of proteins with pro-coagulant effects such as fibrinogen, in relation to lung inflammation induced by urban air pollution and cigarette smoking, has been implicated in adverse cardiovascular effects [
34]. In addition, this response has been implicated in the progression of lung damage in patients with COPD [
35]. Unfortunately, studies relating cardiovascular morbidity and mortality during episodes of increased air pollution have focused only on C-reactive protein and APPs with pro-coagulant effects and have not evaluated the full spectrum of proteins that are known to be co-expressed, among them several antiproteases that can reach the lung improving its defense properties.
Acute phase response elicited by gastric fluid-induced lung inflammation
Although the innate immune response to gastric contents aspiration has been extensively studied in the local lung environment, we have very little insights into extra-pulmonary effects elicited by aspiration that could have implications modifying lung inflammation and the prognosis of a single aspiration event. As a matter of fact, there is little information on the cascade of cytokines released in the lung during aspiration-induced ALI that could induce an APR [
2,
5]. It has been proposed that the pattern of APPs produced in response to an injury and the cytokines involved in their regulation depends on the nature of the inflammatory stimulus [
36]. Although it is known that IL-6, IL-1β and TNF-α modulate the synthesis of APPs in adult human hepatocytes, IL-6 is the major cytokine involved in the synthesis of the full spectrum of APPs [
16,
33,
37]. IL-6 responsive regulatory elements have been found in the α1-AT gene that are responsible for both basal and induced expression of α1-AT in different human cell types and in rat hepatocytes [
33,
38,
39]. On the other hand, it has been shown that IL-6 defective mice have a severely compromised APR to turpentine-induced tissue damage, as well as an impaired response following LPS injection [
40,
41] and bacterial pneumonia [
19]. In our model, IL-6 may play a role in the APR, since it was the only pro-inflammatory cytokine found to have increased levels in blood preceding changes in APPs, despite very high levels of other pro-inflammatory cytokines in the lung. It is also possible however, that other regulators could be involved in our model, since for instance in rats, CINC-1 the counterpart of human IL-8, has also been involved in APP production [
42]. In our study, the high levels of IL-6 and TNF-α observed in BALF in comparison with lung tissue homogenate suggest that epithelial and alveolar mononuclear resident cells might be the source of these cytokines. In this regard, Fujii et al. have shown that the interaction between macrophages and epithelial cells has a synergistic effect on the production and release of mediators involved in the systemic inflammatory response [
43]. With regard to the cytokine IL-10, we found that its concentration in BALF increases at the time pro-inflammatory cytokine levels decrease, in agreement with the known role of IL-10 limiting the cascade of pro inflammatory cytokines in lung inflammation [
44].
In the literature, there is no information on the type and kinetics of APP production during gastric juice-induced ALI. We found a significant increase in the three APPs studied, with a time course of a rodent APR [
17,
18,
33]. Additionally, the increment in α1-AT concentration in blood was associated with a proportional increase in its activity. The increment in blood levels of α1-AT and α2-MG was preceded by an increase in their mRNA expression in the liver, with the largest increase being that of α2-MG, which is considered to be the main acute phase antiprotease in the rat [
17]. As a matter of fact, α1-AT, α2-MG and fibrinogen belong to the same group of APPs [
16], responding to the same type of signal, involving activation of intracellular tyrosine kinase JAK and the acute-phase responsive factor, now called STAT3 [
36,
45,
54].
On the basis of timing and magnitude of the changes observed in our model, we speculate that the increase in liver synthesis of fibrinogen, α1-AT and α2-MG could be mediated by the IL-6 signaling, that is known to participate in activation of the JAK/STAT3 cascade. Future studies using IL-6 inhibition are warranted to further evaluate the role of this cytokine in gastric juice-induced APR.
Although α1-AT is the main lung antiprotease in steady-state conditions, exhibiting a broad range of anti-inflammatory and immunoregulatory activities [
37,
46], it is not the main acute phase antiprotease in the rat. It is produced primarily by the liver, with additional sources such as peripheral blood monocytes and alveolar macrophages contributing with a small fraction to total synthesis. Literature shows that similar to the liver, these additional sources are also known to respond to inflammatory stimuli with a small local APR, but their contribution to the increment in capillary-alveolar concentration gradient and to BALF APP levels is small [
19,
47].
It has been shown that α1-AT is produced as part of the liver APR after turpentine injection in guinea pigs [
48] and rats [
17]. Support for a beneficial effect of α1-AT as part of the APR is provided by a study in a rat model of renal ischemia-reperfusion injury, in which the administration of α1-AT, at a dose that results in plasma levels similar to those observed during an APR, reduces inflammation, apoptotic activity and tissue damage [
49]. Interestingly, we found complex formation between α1-AT and elastase in BALF between 4 and 24 h after instillation, providing evidence of the functionality of this antiprotease in our model.
With regard to α2-MG, it inhibits different types of proteases and is also a carrier and regulator of the function of several cytokines. Its large size prevents it from diffusing easily through the normal alveolar-capillary barrier and thus, it is found in very low concentration in normal alveolar spaces [
46,
50]. It reaches the alveolar spaces in significant amounts whenever there is alveolar-capillary barrier derangement. Its role as a marker of alveolar-capillary barrier permeability has been studied in human acute respiratory distress syndrome [
47]. In this condition, α2-MG has been found forming complexes with proteases and IL-8 [
51]. Very little is known about its role as an acute phase protein in humans, however, it is recognized as the main acute phase antiprotease in the rat [
17].
Impact of the hepatic APR on lung antiprotease defense
Containing the lung inflammatory response induced by gastric juice is critically important in order to inhibit progression to a persistent systemic inflammatory response fueled by persistently increased cytokine production. In this regard, liver-newly synthetized antiproteases reaching the alveolar spaces may play an important role limiting lung inflammation. It has recently been shown that the APR has a role improving animal survival in experimental bacterial pneumonia [
19,
52,
53] and although in these studies antiproteases were not evaluated, other APPs that were increased in blood were found increased in BALF [
19]. On the other hand, research using APR-null mice in pneumonia has shown that in the absence of an APR, APPs do not increase in BALF and there is increased animal mortality [
19,
54,
55]. Our results showed a highly significant enrichment in antiprotease content of the alveolar spaces, with a mean 58-fold increase in α1-AT activity and a mean 192-fold increase in α2-MG concentration in BALF. Given the fact that in our model, the acidic component of gastric contents triggers an early derangement of the alveolar-capillary barrier that precedes in several hours the increment in blood APP concentrations, we were able to show a two-phase response in the enrichment of the alveolar spaces with antiproteases.
Two-phase response in the enrichment of the alveolar spaces with antiproteases
The first phase is characterized by an early and large increment in BALF antiprotease concentration and occurs prior to APP elevation in blood, thus not due to the APR, and instead due to derangement of the alveolar - capillary barrier that loses its protein size selectivity, facilitating the passage of both small (α1-AT) and large (α2-MG) molecular weight proteins to the alveolar spaces [
50]. This phase contributes with a 40-fold increase in lung α1-AT bioactivity and a 51-fold increase in α2-MG concentration.
The second phase is contributed by the liver acute phase response that increases the capillary-alveolar concentration gradient for α1-AT and mainly for α2-MG. This phase provides an extra 18 % increase in BALF α1-AT bioactivity and an extra 281 % elevation in BALF α2-MG concentration (above the high levels provided by the first phase of antiprotease enrichment found at 4 h), in the context of significant reduction in total protein content in the same period of time. Interestingly, the antiprotease with the largest change in BALF in this phase is α2-MG that has also the largest change in blood concentration and in liver mRNA expression.
This two-phase response in lung antiprotease enrichment is likely to be unique to this model and not easy to detect in other models of ALI, in which alveolar-capillary barrier derangement is delayed by depending more on the effects of the inflammatory response than on the direct early effect of the acidic component of gastric juice.
It is also possible that a small part of the lung enrichment in antiproteases may come from alveolar mononuclear cells, that are known to respond to inflammatory stimuli with a low grade local lung APR [
19,
47]. However, given the time-course of changes, this factor is unlikely to play a major role in our model, since antiprotease flooding occurs very early, prior to and during PMNn cell infiltration, and prior to mononuclear cell predominance, which occurs after day 4 [
4,
5]. Furthermore it is accepted that the small amount of α1-AT released from monocytes serves more as a microenvironmental front line defense against proteases provided by the same monocytes [
47].
On the basis of our results, local lung effects of gastric fluid damaging the alveolar-capillary barrier act in concert with lung inflammation-induced hepatic acute phase response to favour the arrival of new antiproteases to improve lung defense. Evidence of complex formation between α1-AT and elastase released from the inflammatory cells in the first 24 h provide support for the idea that antiprotease enrichment of the alveolar spaces plays a role limiting lung injury by proteases in this model. Protease inhibition may represent one of many mechanisms involved in the significant capacity of the lung to repair the severe ALI induced by gastric fluid recently reported by our group [
5].
We postulate that antiprotease supplementation during the window of opportunity in which the hepatic APR co-exists with increased permeability, could be useful during ALI, in species without an increment in α1-AT and/or α2-MG in the liver acute phase response. In addition, it could be possible that either liver diseases or polymorphisms in the acute phase response genes may constitute susceptibility factors for lung or other tissue injuries. In this regard, increased lethality from lung damage has been shown by Borzio et al. [
56] in patients with liver cirrhosis. On the other hand, a reduced APR was implicated in the increased lethality observed in experimental cirrhosis [
57]. In the light of our results it is possible that antiproteases as part of APR could be involved in the results of the above mentioned studies.
In addition, since the APPs in blood remain elevated for several days, it is likely that they may impact lung response to repetitive events of aspiration. Future studies of this response with different lung inflammatory stimuli should consider the evaluation of the full range of proteins that are co-expressed, including the antiproteases, in order to have a better appreciation of the net effects of the liver APR in lung defense.