THE PATHOGENESIS OF MALARIA ACUTE RESPIRATORY DISTRESS SYNDROME (MA-ARDS): MODIFICATION OF THE LIPID PROFILE, ANTIOXIDANT DEFENCES AND CYTOKINE CONTENT IN DIFFERENT TISSUES OF MALARIA INFECTED MICE

INTRODUCTION Malaria is a major health problem, with more than 650.000 deaths and 200 million clinical cases each year. Respiratory distress as malaria associated acute respiratory distress syndrome (MA-ARDS) is a common complication. The pathogenesis of MA-ARDS is mainly inflammatory and one of the main observation is the presence of abundant monocytes and macrophages inside the blood capillaries, in the interstitium and also in alveolar spaces. Malaria pigment or haemozoin (Hz) is often seen in these cells reflecting active phagocytosis and leads to the production of cytokines and other inflammatory mediators. Multiple organ dysfunctions are described in MA-ARDS, including liver damages. ARDS in non malarious patients is often associated with disorders of the lung surfactant, which can lead to the increase in surface tension, alveolar collapse and loss of the liquid balance in the lungs. Surfactant is known to reduce the surface tension at the air–liquid interface of lung epithelia and to regulate the local host immune response. It can be separated into a surface active Large Aggregate fractions (LA), representing a reservoir for the surface film located at the air-liquid interface of the alveoli and a less surface active, Small Aggregate fraction (SA). It is not known at present if alterations of the surfactant also exist in MA-ARDS and how they may contribute to the pathology and the development of the inflammatory response. AIM The aim of our studies was to perform a comprehensive analysis of the local and systemic inflammatory response present in MA-ARDS and to analyse the lipid profile of the pulmonary surfactant, the lung and liver tissues and plasma using two different models of murine malaria of similar gravity, but different involvement of lungs or liver. In particular, we studied C57BL/6J mice infected with two different species of Plasmodium: Plasmodium berghei NK65 strain which induces MA-ARDS and Plasmodium chabaudi (PcAS), which does not. The two models allowed us to directly compare the different pathological manifestation of the same infection in order to identify peculiarities which could be exploited for novel therapeutic interventions. RESULTS AND DISCUSSION Macroscopic and functional analysis of lung tissues in the two strains. The lungs of PbNK65 infected mice were swollen, increased in weight and with a dark brown aspect due to micro haemorrhages and to the deposition of Hz clusters in concentrations significantly higher compared to the lungs of mice infected with PcAS. The expression of TNF-α and IFN- was increased only in PbNK65 mice, indicating that these cytokines are induced specifically during MA-ARDS and are not a consequence of malaria infection. This hypothesis was confirmed by the decrease of lung weight and of the CD8+ cells infiltrate, and the reduction /delay in mortality rates seen in PbNK65 mice treated with DEX without a concomitant reduction in parasitaemia. Therefore, DEX seems to ameliorate MA-ARDS, not by inhibiting parasite growth but rather by modulating the immunopathology and the inflammatory response. A significant increase of the total phospholipid (PL) content and cholesterol esters (ChoE) was observed in PbNK65 lungs and was reverted by DEX. Moreover, compared to the control mice (CTR), the fatty acid distribution of lung ChoE was characterized by higher levels of the polyunsaturated fatty acid and an high linoleic/oleic ratio typical of plasma ChoE. All these features confirm a strict correlation between the interstitial oedema and the infiltration of plasma lipoproteins during MA-ARDS. Protein and lipid composition of surfactant and plasma of infected mice. The total bronchoalveolar lavage (BAL) of PbNK65 mice showed a significant increase in the protein levels compared to CTR or PcAS mice, probably due to plasma derived proteins being incorporated into or associated with microstructures in the alveolar hypophase. This event is known to decrease the intrinsic surface activity of surfactant. The total content of PL was not different from CTR, whereas the PL profile of the LA and SA fractions in PbNK65 infected mice showed a significant increase in the amounts of sphingomyelin and a decrease in phosphatidylglycerol. These changes were absent in PcAS mice and may be related to the altered re-uptake and synthesis of PL by injured cells or to PL contamination due to inflammatory cells. The plasma levels of PL and triacylglycerol (TG) were significantly higher in PbNK65 mice than in CTR or PcAS mice. Compared to PcAS or CTR mice in PbNK65 group all the PL classes were significantly increased with the exception of lisophosphatidilcholine (LisoPC) that was decreased. These plasma alterations may be related to an impaired activity of the enzymes involved in the lipoprotein metabolism during infections or inflammatory diseases. The most important observation, both in PbNK65 and PcAS mice, was the significant increase of docosahexahenoic acid (C22:6 n-3, DHA) compared to CTR, which was only partially reverted by DEX treatment, suggesting that the increase of DHA is not related to lung pathology but rather to the malaria infection. Analysis of the liver tissue of infected mice. The hypothesis that the higher PL and TG content of PbNK65 plasma might be due to an enhanced hepatic lipogenesis was confirmed by the higher TG and ChoE content of the liver of PbNK65 mice compared to PcAS and CTR. An increased ratio linoleic (LA)/arachidonic acid (AA) was also present possibly due to the impairment of the elongation/desaturation pathway from LA to AA acid. Higher levels of Hz, compared to PcAS, were present in PbNK65 mice and, in agreement with the Hz capability of stimulating Kupffer cells, we found higher levels of TNF-α. Both Hz and TNF-α can induce lipoperoxidation as confirmed by the elevated levels of malondialdehyde (MDA) in the liver of PbNK65 mice. This finding was paralleled by the lower content of glutathione and of antioxidant enzymes particularly in the late stage of the pathology. CONCLUSIONS This is the first time that a comprehensive analysis of the lipid content and inflammatory response of different organs in a model of murine MA-ARDS has been performed. All together the data suggest that in MA-ARDS as in other severe non infectious pathologies, a pulmonary-liver metabolic interplay exist which may contribute to the pathology. In PbNK65 mice, Hz and the derived inflammation play an important role in the lung pathology inducing changes in the lipid composition of lung and surfactant, cellular infiltration and cytokine production. Lung pathology is associated with liver disorders and alterations in the lipoprotein profile. Hz accumulation may induce macrophages to produce TNF-α and ROS that can interfere with liver functions by inducing lipogenesis and affecting the lipid profile of liver and plasma, which in turn contribute to the altered lipid composition of the lung tissue. These results confirm that severe malaria is a multi-organ dysfunction in which inflammation has an important role in different organs and thus, in addition to antimalarial treatment, adjunct therapies with anti-inflammatory drugs can be envisaged.