Orientador: Maria Terezinha Serrão Peraçoli / Resumo: Introdução: A pré-eclampsia (PE) destaca-se como uma das principais causas de morbidade e mortalidade tanto materna como fetal, se caracteriza por ativação anormal do sistema imune inato e adaptativo. No plasma de gestantes portadoras de PE encontram-se níveis elevados de estruturas moleculares associadas ao estresse e morte celular, denominados padrões moleculares associados ao dano (DAMPs) como, proteína de choque térmico (Hsp70), HMGB1 (high mobility group box 1), Hialurona (HA) e Ácido Úrico, que parecem contribuir diretamente com a patogênese dessa doença. Essas DAMPs ligam-se a receptores presentes em células da imunidade inata, podendo ativar um complexo intracelular denominado inflamassoma, responsável pelo processamento e liberação de IL-1β e IL-18. Essas citocinas são potentes mediadores inflamatórios e importantes na ativação da resposta imune adaptativa, auxiliando a diferenciação de células T em Th17 e Th1, respectivamente. Na PE, a ocorrência de resposta inflamatória sistêmica parece decorrer da deficiência no controle de células T efetoras por células T reguladoras (Treg). Portanto, o balanço entre células Treg e Th17 pode ser crítico para a tolerância ao feto e para a prevenção da doença. Objetivos: O presente trabalho teve como objetivos: Avaliar o estado de ativação, endógena e induzida pelas DAMPs (urato monossódico (MSU), HA e Hsp70), em monócitos, pela identificação da presença do inflamassoma e associação com a produção de citocinas; Avaliar o envolvime... (Resumo completo, clicar acesso eletrônico abaixo) / Doutor

Inflammation in the absence of infection (sterile inflammation) is a crucial host defence response to tissue injury, but is also considered to contribute to the pathogenesis of many diverse disease states, including stroke. Sterile inflammation is initiated by damage associated molecular patterns (DAMPs) which are endogenous molecules released from necrotic cells or that are modified during disease. The pro-inflammatory cytokines IL-1α and IL-1β are key mediators of inflammation. IL-1β release is controlled by caspase-1 which, in turn, is regulated by the inflammasome. The NOD-, LRR-, pyrin domain-containing 3 (NLRP3) inflammasome is most typically associated with sterile inflammation and the recognition of DAMPs. Thus, understanding the mechanisms of NLRP3-activating DAMP-induced inflammation may lead to the identification of novel therapeutic targets with which to treat inflammatory diseases. This thesis sought to determine how NLRP3-activating DAMPs affect the pro-inflammatory response of glia, the immune cells of the brain. Experimental models in vitro typically use a pathogen associated molecular pattern (PAMP) such as LPS to prime cells before observing their response to NLRP3-activating DAMPs. As the brain is protected by the blood brain barrier (BBB), it is unlikely glia would be exposed to PAMP priming. However it remains unclear as to how glia respond to NLRP3-activating DAMPs in the absence of priming, or what the source of endogenous priming is. Therefore, the initial hypothesis was to investigate the pro-inflammatory response of mixed glia in vitro to NLRP3-activating DAMPs in the absence of PAMP priming. It is shown here for the first time that NLRP3-activating DAMPs can initiate an IL-1-NLRP3-independent inflammatory response in mixed glia in the absence of PAMP priming. Moreover, it is shown that the acute phase protein serum amyloid A is elevated in plasma after stroke and may act as an endogenous priming signal to allow IL-1β-dependent inflammation to contribute to the damage after breakdown of the BBB.Inflammation following acute sterile injury such as stroke is augmented by persisting cell death. It was therefore hypothesised that NLRP3-activating DAMPs released after the initial injury, may initiate a form of programmed cell death that continues to drive inflammation. Using inhibitors of specific types of cell death, it was identified that NLRP3-activating DAMP induced cell death is likely to be necrosis and not programmed cell death. Further investigation into the biological importance of DAMP-induced IL-1-independent inflammation and the specific contribution of acute phase proteins to brain pathology may aid the identification of new therapeutic targets.

[French below]TLRs are the sentinels of our cells, they are located at the cell surface and alert the whole immune system of the presence of viruses or bacteria. They detect pathogens by recognizing their molecular patterns; this recognition is specific in order to avoid self-recognition, but they need some degree of promiscuity to remedy to pathogen heterogeneity or mutations. Promiscuity is generally defined as an indiscriminate association with molecules regardless their structure and is the contrary of specificity proper of the classic paradigm of key-lock receptor activation. My thesis results demonstrate that TLR4 and TLR2 are more promiscuous than what was believed and that this promiscuity leads to the recognition of cationic lipids and cardiolipins.Cationic lipids lipopolyamines are synthetic molecules nucleic acid nanocarriers proposed to be used for gene therapy, which consists in replacing a gene that is functioning improperly. This thesis demonstrates that lipopolyamines activate TLR2 by forming conserved and/or alternative H-bonds with TLR residues, simulating the recognition of bacterial lipopeptides and inducing pro-inflammatory cytokines secretion; which is deleterious when we aim to use these nanocarriers in the context of gene therapy. We propose the use of unsaturated cationic lipids to avoid TLR2 recognition. TLR activation could be useful instead to prepare one-component vaccine adjuvants, for which both antigen carrier and TLR activation are needed to turn on the immune system and produce antibodies. The second chapter of this thesis investigates the pro-inflammatory properties of other cationic lipids and describes new lipopolyamines able to activate both TLR2 and TLR4. The study of their adjuvanticity properties showed that they are as efficient as the aluminium salts in stimulating antibodies production.The promiscuity of TLR2 and TLR4 raised questions about the presence of endogenous TLR modulators, often countered by contamination concerns. In the third chapter, we investigate the pro- and anti-inflammatory properties of cardiolipin (CL). CL is a tetra-acylated diphosphatidylglycerol located in the inner mitochondrial membrane. Its fatty acid chain length and degree of unsaturation vary depending on species, tissue and pathological conditions. Here we show that unsaturated cardiolipin acts as a competitive TLR4 antagonist by occupying the binding site of LPS and that unsaturations discriminate between TLR4 antagonistic and agonistic activity. Under physiological conditions, mammalian CL chains are unsaturated, whereas there is an increase of saturated CL in patients affected by the Barth Syndrome. Hence we suggest that unsaturated CL could negatively regulate the inflammation during cell damage or death, via TLR4 inhibition. By contrast, saturated CLs may be involved in the inflammatory state associated with the disease. Finally, our study provides new understandings of the mechanism of TLR4 regulation and extends the library of TLR4 agonists and antagonists to molecules of easier synthesis, lower price and higher biocompatibility compared to LPS-based structures. / Resumé en français Les récepteurs Toll-like (TLRs) sont des protéines transmembranaires qui constituent la première barrière de notre système immunitaire inné. Ils détectent la présence de bactéries et virus et alertent l’organisme via la sécrétion de molécules pro-inflammatoires appelés cytokines. Parmi les TLRs, TLR2 and TLR4 reconnaissent respectivement des lipides spécifiques aux bactéries, les lipopeptides et les lipopolysaccharides bactériens LPS. La reconnaissance de motifs moléculaires spécifiques aux pathogènes et absents dans notre organisme est essentiel afin d’éviter une réponse immunitaire venant du soi. Le but de notre thèse était de démontrer que les récepteurs Toll-like possèdent une certaine plasticité et peuvent reconnaître des ligands non identifiés jusqu’ici tels les lipides cationiques et la cardiolipine. Les lipides cationiques sont des molécules synthétiques utilisées comme agents de transfection. Notre travail démontre que les lipides cationiques dont la tête polaire est constituée par des polyamines peuvent mimer les propriétés des ligands naturels et induire la sécrétion de cytokines pro-inflammatoire via l’activation des TLRs. Cette interaction implique des interactions entre la chaine principale de la protéine et les lipides sans intervention des chaines latérales. Cette réaction inflammatoire est contre-indiquée en thérapie génique et nous proposons donc de remplacer les chaines acylées saturées par des chaines insaturées pour la synthèse des nouveaux agents de transfection non-immunogénique. D’autre part, l’activation des TLRs par des agents de transfections active le système immunitaire inné, ce qui permet l’activation du système adaptatif et la production d’anticorps. Nous avons étudié une large gamme des lipides cationiques et identifié des nouveaux activateurs á la fois de TLR2 et de TLR4. L’étude de leurs propriétés adjuvantes a démontré que les lipides cationiques sont des adjuvants comparables aux sels d’aluminium en terme de production d’anticorps. La cardiolipine est un lipide localisé dans la membrane des mitochondries et des bactéries. Le domaine hydrophobe est constitué de quatre chaines acylées qui chez les mammifères sont insaturées. Il a été démontré que la cardiolipine extracellulaire inhibe la sécrétion de cytokines induite par LPS. Notre travail de thèse démontre que cet effet inhibiteur est du à la capacité de la cardiolipine à bloquer le site de liaison du LPS. Le travail démontre aussi que lorsque les chaines acylées sont saturées, c’est le cas dans le Syndrome de Barth, la cardiolipine devient un activateur de TLR4 en interagissant avec TLR4 de façon similaire à LPS. Ce dernier résultat pourrait expliquer l’aspect inflammatoire du Syndrome de Barth et élargie la librairie de ligands de TLR4 á des molécules de structure plus simple et plus aisées à synthétiser que les dérivés de LPS et qui pourraient être utilisés comme adjuvants ou anti-inflammatoires. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

Introduction: Myocardial Infarction (MI) is the leading cause of premature death worldwide. During MI-induced ischemia, the release of heat shock proteins (HSPs), a classic damage-associated molecular pattern (DAMP), by severely injured cells leads to prolonged inflammation through their activation of innate pattern recognition receptors, fibrosis, and subsequent contractile dysfunction. The regulation of HSPs is orchestrated by its master transcription factor, Heat Shock Factor 1 (HSF1). However, it is unknown if HSF1 is a potential integrated functional target to improve MI outcomes. We addressed this question by asking if the coordinated modulation of HSPs via genetic deletion of Hsf1 can be beneficial in MI. Hypothesis: We hypothesized that genetic deletion of Hsf1 can lead to improved survival and left ventricle (LV) remodeling through reduction of pro-inflammatory pathway activation in a murine model of MI-induced coronary artery ligation. Methods and Results: Eleven to thirteen-week-old male Hsf1-/- mice and Hsf1+/+ littermate controls were subjected to MI by left anterior descending (LAD) coronary artery ligation or sham operation. Hsf1-/- mice subjected to induced-MI had a significant higher survival rate (74%) at 28 days than WT mice post-MI in the same time frame (34%, p<0.001). Echocardiography at 3, 7, and 28 days post-MI; however, did not identify any difference in LV function between Hsf1+/+ and Hsf1-/- mice. Masson Trichrome and Picro Sirius Red staining of heart tissue sections following 7 days of sham or MI-operation indicated that MI-operated Hsf1-/- hearts had a significant smaller infarct size than Hsf1+/+ hearts at 19% compared to 32% (p<0.05), respectively; and less collagen deposition when compared to WT littermates. Cardiac expression of heat shock proteins was significantly lowered in the Hsf1-/- hearts compared to Hsf1+/+ hearts following 3 and 7 days of MI. However, no significant difference was observed in number of immune cells, cardiac gene expression of pro-inflammatory cytokines and chemokines, cardiac protein expression of NF-κB and MAPK-ERK1/2 signaling proteins, and serum IL-6 concentration between Hsf1+/+ and Hsf1-/- mice 3 days post-MI. Following 7 days of MI, there is a significant increase in the gene expression of pro-inflammatory cytokines, such as Il1b, and chemokines, such as Ccl2, in Hsf1-/- hearts than Hsf1+/+ hearts. Conclusion & Future Directions: Overall, the loss of Hsf1 improved survival and reduced infarct size following MI. However, its deletion did not affect inflammatory processes until 7 days post-MI or improved cardiac function in our specific murine MI model.

Enterohemorrhagic E. coli (EHEC) infection is a leading cause of acute kidney failure in otherwise healthy children, and a leading cause of foodborne illness with an outsized economic impact from outbreaks. EHEC secrete two Shiga-like toxins (Stx1 and Stx2) which are AB5 holotoxins that inhibit protein synthesis in cells expressing the toxin receptor Gb3. Infection with EHEC typically begins with a diarrheal prodrome that can progress in 5-15% of cases to hemolytic uremic syndrome (HUS), a clinical diagnosis characterized by thrombocytopenia, hemolytic anemia, and thrombotic microangiopathy. Historically, strains of EHEC expressing Stx2 have been associated with more severe disease. We hypothesized that tissue injury due to the toxins leads to the release of damage-associated molecular patterns (DAMPs), which act through inflammatory receptors to promote the endothelial dysfunction that drives this disease alongside the inciting Shiga toxins. Here we demonstrate that two well-characterized DAMPs, extracellular histones and HMGB1, are produced in two different mouse models when Stx2 is present; one model represents challenge with the toxin alone, and the second model introduces toxin through secretion with a lysogenized bacterium, C. rodentium, mimicking EHEC colonization. We investigate whether Stx1, Stx2, or histones affect the endothelial expression of well-characterized members of the protein C pathway, namely the endothelial protein C receptor (ECPR), protease-activated receptor 1 (PAR1), and thrombomodulin (TM), on human aortic (HAEC) and human renal glomerular endothelial cells (HRGEC). We show that Stx and/or histones reduce endothelial expression of these anti-coagulant molecules and histones dramatically increase expression of pro-thrombotic PAR-1. These changes lead to physiologically important decreases in activated protein C (APC), a critical anti-coagulant and cytoprotective molecule. Finally, we demonstrate that histones exacerbate thrombin's barrier-disruptive effects on the endothelium, and prevent APC's protective effects. These data provide novel mechanistic insight into the endothelial dysfunction that characterizes HUS and also provide a new perspective on systemic consequences of the bacterial Shiga toxins that might drive organ injury in susceptible patients.

Le placenta est l’organe central de la grossesse et son bon fonctionnement est essentiel pour mener à une grossesse à terme sans complication. L’inflammation joue un rôle très important dans les différentes étapes de la grossesse, de l’implantation à l’accouchement. Lorsque cette inflammation est débalancée au niveau du placenta, cela peut causer plusieurs altérations de ses fonctions. L’inflammation peut être induite par deux différents stimuli, soient par une intervention externe, donc infectieux (via les Pathogen-Associated Molecular Patterns, PAMPs) ou de façon endogène (via les Damage-Associated Molecular Patterns, DAMPs). L’augmentation de l’inflammation est aussi associée avec les complications de la grossesse, qui touchent 5-12% de toutes les grossesses et qui peuvent mener à des conséquences dévastatrices sur la santé de la mère et du nouveau-né. Ces complications comprennent la prééclampsie (PE), l’accouchement prématuré (AP) et le retard de croissance intra-utérin (RCIU). Malgré que ces pathologies aient des étiologies différentes, elles partagent un facteur commun qui joue un rôle essentiel soit, l’inflammation. Dans le but de mieux comprendre le rôle et les effets de l’inflammation sur le placenta humain et dans les complications de la grossesse, nous avons évalué les effets d’un PAMP (LPS) et d’un DAMP (IL-1) classiques dans un modèle d’explants placentaires humains et nous avons réalisé une étude transcriptomique non biaisée de placentas issus de chaque pathologie. Tout d’abord, nous avons démontré que le LPS induisait une plus grande quantité de cytokines pro-inflammatoire comparativement à l’IL-1 et que l’inhibition de la voie de signalisation de l’IL-1 par son antagoniste (IL-1Ra) diminuait leur expression et leur sécrétion. De plus, le LPS induisait une augmentation de la mort cellulaire dans les explants et la prolifération des cellules de Hofbauer, macrophages placentaires. Par la suite, l’étude du transcriptome de placentas provenant de complications de la grossesse (PE, AP et RCIU) a permis de constater que les pathologies ne forment pas de groupes (clusters) basés sur l’expression globale des gènes comparativement placentas issus de grossesses sans complication qui se regroupaient majoritairement ensemble. Il a été possible d’identifier les gènes significativement différents dans chaque pathologie et l’ontologie génique de ceux-ci. Nous avons identifié 198 gènes communs à toutes les complications de la grossesse qui sont majoritairement associés à des processus inflammatoires tels que l’interaction entre les cellules lymphoïdes et non lymphoïdes, l’activation leucocytaire et la régulation de la production de cytokines. Enfin, en plus de confirmer les modulations connues de certains gènes dans chaque complication de la grossesse, nous avons été en mesure d’identifier de nouveaux gènes qui n’avaient jamais été associés avec les pathologies, tels que FUT9, SLAMF7 et TGM3. En conclusion, les travaux présentés dans cette thèse démontrent que l’inflammation induite par le LPS et l’IL-1 n’a pas les mêmes effets au niveau du placenta. L’inflammation est une composante commune aux différentes complications de la grossesse et une meilleure compréhension des processus inflammatoires modulés dans chaque pathologie pourrait permettre le développement de nouvelles approches thérapeutiques. Des travaux futurs pourraient s’intéresser au potentiel prolifératif des cellules de Hofbauer et à l’investigation des nouveaux gènes identifiés par les résultats transcriptomiques. / The placenta is the central organ of pregnancy and its adequate functioning is essential to ensure term delivery without complications. Inflammation plays a key role in every step of pregnancy, from implantation to delivery. When this inflammation is unbalanced in the placenta, it can alter its functions. Inflammation can be induced by two different stimuli, either by external intervention like an infection (with Pathogen-Associated Molecular Patterns, PAMPs), or endogenously (with Damage-Associated Molecular Patterns, DAMPs). Increased inflammation is also associated with pregnancy complications and they affect 5-12% of all pregnancies and can have deleterious effects on maternal and infant health. These complications include preeclampsia (PE), preterm birth (PTB) and intrauterine growth restriction (IUGR). Even if these pathologies have different etiologies, they share a common factor that plays an essential role, inflammation. In order to better understand the role and the effects of inflammation on the human placenta and in pregnancy complications, we have evaluated the effects of classical PAMP (LPS) and DAMP (IL-1) on human placental explant model and we have realized an unbiased transcriptomic study of the placentas from each pregnancy complication. First of all, we demonstrated that LPS induced a higher production of pro-inflammatory cytokines compared to IL-1 and by inhibiting the IL-1 pathway using its antagonist (IL-1Ra), we decreased their expression and their secretion. Moreover, LPS treatment induced more cell death in the explants and proliferation of Hofbauer cells, macrophages of the placenta. Thereafter, transcriptomic study of placentas from pregnancy complications (PE, PTB and IUGR) allowed us to demonstrate that the pathologies were not clustering together based on their global gene expression compared to placentas from uncomplicated pregnancies which the majority of them were clustering together. It was possible to identify genes that were significantly modulated in each pathology and their gene ontology. We identified 198 genes common to all pregnancy complications and they were mostly related to inflammatory processes like interaction between lymphoid and non-lymphoid cells, leukocyte activation and regulation of cytokine production. Finally, in addition to confirming gene modulations previously known in each pregnancy complications, we were able to identify new genes that have never been associated with the pathologies such as FUT9, SLAMF7 and TGM3. To conclude, the work presented in this thesis show that inflammation induced by LPS and IL-1 did not have the same effect on the placenta. Inflammation is a key component to pregnancy complications and a better understanding of the inflammatory processes modulated in every pathology could help the development of new therapeutic strategies. Future work could investigate the proliferative potential of the Hofbauer cells and explore the new genes identified by the transcriptomic results.

Les dysfonctions placentaires sont fortement associées aux complications de la grossesse et de plus en plus reliées à une augmentation de médiateurs inflammatoires endogènes, appelés alarmines ou motifs moléculaires associés aux dommages (« damage associated molecular patterns » (DAMPs). High-mobility group box 1 (HMGB1), est un DAMP qui a été associé aux grossesses avec PE et accouchement prématuré. HMGB1 est une protéine nucléaire qui peut être sécrétée dans l’espace extracellulaire de façon passive ou active. Une fois dans l’espace extracellulaire, HMGB1 existe sous différents isoformes ayant des actions inflammatoires distinctes. Le rôle de HMGB1 et de ses isoformes à l’interface materno-fœtale est encore peu connu. L’objectif de mes travaux de maitrise était d’investiguer le rôle de HMGB1 à l’interface materno-fœtale, en déterminant sa localisation subcellulaire pendant la syncytialisation, ainsi que les actions pro-inflammatoires sur le placenta. Méthodes : Un modèle d’explants placentaires en conditions physiologiques fut utilisé afin de déterminer et de moduler la localisation subcellulaire de HMGB1. Le même modèle a été traité avec les différentes isoformes de HMGB1 (HMGB1-disulfide: D ou HMGB1-réduit: R) afin de déterminer leurs effets inflammatoires et leurs impacts sur la fonction placentaire. Des placentas de femmes ayant des grossesses sans complications, une prééclampsie (PE) ou une prééclampsie du postpartum (PPPE) ont été étudiés afin de déterminer la distribution de HMGB1 et ses récepteurs (Receptor for advanced glycation end product (RAGE) et Toll like receptor (TLR4)). Résultats : La localisation intracellulaire de HMGB1 est modulée pendant le processus de syncytialisation avec une localisation majoritairement cytoplasmique et extracellulaire par rapport à une localisation généralement nucléaire dans les trophoblastes différenciés. Favoriser l’export nucléaire de HMGB1 avec un inhibiteur d’histone déacétylase (HDAC), le sodium butyrate (NaB) augmente la concentration cytoplasmique de HMGB1 ainsi que la sécrétion de l’hormone chorionique gonadotrope (-hCG), signe de la différentiation des trophoblastes. L’isoforme disulfide de HMGB1 induit la sécrétion de cytokines pro-inflammatoire (IL-1, IL-6 et MCP-1) et a aussi un impact sur la différenciation des trophoblastes, tel qu’observé par une diminution de la sécrétion de -hCG. En conditions pathologiques, l’expression de HMGB1 et ses récepteurs RAGE et TLR4 est augmentée dans des conditions de prééclampsie du post-partum. Pour conclure, la localisation subcellulaire de HMGB1 est modulée pendant la syncytialisation, dans un contexte non-pathologique. L’accumulation cytoplasmique de HMGB1 est la première étape avant la sécrétion dans l’espace extracellulaire. Lorsque dans l’espace extracellulaire, une isoforme spécifique de HMGB1 (HMGB1-D) entraine l’augmentation de la sécrétion de cytokines pro-inflammatoires. L’expression de HMGB1 et ses récepteurs est augmentée et conditions pathologiques démontrant que ce DAMP peut jouer différents rôles tant dans un contexte inflammatoire et de dysfonctions placentaires ainsi que dans un contexte de différenciation des trophoblastes. / Sterile inflammation, caused by endogenous damaged-associated-molecular-patterns (DAMP), at the maternal fetal interface is frequently observed in pregnancy complications and leads to placental inflammation and dysfunction by unknown mechanisms. HMGB1 has been associated to preeclampsia, preterm birth and it can be released in the extracellular space and associated to increased inflammatory actions. Extracellular HMGB1 has two isoforms, (HMGB1-disulfide-D) inducing proinflammatory cytokines whilst the other (HMGB1-reduced-R) acts as a chemoattractant. The role of HMGB1 and its isoform at the maternal-fetal interface is mostly unknown. The objective of my master was to investigate the roles of HMGB1 at the maternal-fetal interface including its subcellular localisation during trophoblast differentiation and pro-inflammatory effects on the placenta. Methods: Term placental explants were used to determine the subcellular localisation of HMGB1 during trophoblast differentiation or treated with specific HMGB1 isoforms (HMGB1-D or HMGB1-R) to determine the impact on inflammation and placental function. Alongside, placentas from women with either normal term pregnancies, PE or PPPE were used to determine the distribution of HMGB1 and its receptor. Results: HMGB1 subcellular localisation is modulated during the syncytialisation process with major cytoplasmic and extracellular localisation to a more nuclear localisation in differentiated trophoblasts. Promoting HMGB1 nuclear export, using the histone deacetylase inhibitor (HDAC) NaB, increased HMGB1 cytoplasmic concentration leading to increase secretion of human chorionic gonadotropin (-hCG) in placental explants. HMGB1-D treatment of explants led to the secretion of pro-inflammatory cytokines (IL-1, IL-6 et MCP-1) and impacted trophoblasts differentiation observed by decreased -hCG secretion. In pathological conditions, HMGB1 and his receptors, RAGE and TLR4, expression is increased in PPPE compared to non-pathological pregnancy. To conclude, we demonstrated changes in the localisation of HMGB1 in association with trophoblast differentiation in uncomplicated pregnancies. Cytoplasmic accumulation of HMGB1 is the first step before its release in the extracellular space. We showed that a specific isoform of HMGB1 (disulfide isoform) induced inflammatory cytokines secretion which suggests a role of this DAMP in placental inflammation and function. Finally, HMGB1 and its receptors are increased in a pathological condition (PPPE) demonstrating that this DAMP may play different role in both inflammatory context and trophoblast differentiation.