Hypoxia effect on genetic regulation and virulence in Acinetobacter baumannii and Pseudomonas aeruginosa, in vitro and in vivo, and on innate immune response in infections caused by both pathogens.

DOCTORAL THESIS Hypoxia effect on genetic regulation and virulence in Acinetobacter baumannii and Pseudomonas aeruginosa, in vitro and in vivo, and on innate immune response in infections caused by both pathogens María Luisa Gil Marqués Seville 2018 Director 1: Dr. Jerónimo Pachón Díaz Director 2: María Eugenia Pachón Ibáñez Director 3: Michael J. McConnell 3 DOCTORAL THESIS FUNDING This work was supported by the Instituto de Salud Carlos III, Subdirección General de Redes y Centros de Investigación Cooperativa, Ministerio de Economía, Industria y Competitividad (PIE13/0004) and by Plan Nacional de I+D+i 2013-2016 and Instituto de Salud Carlos III, Subdirección General de Redes y Centros de Investigación Cooperativa, Ministerio de Economía, Industria y Competitividad, Spanish Network for Research in Infectious Diseases (REIPI RD12/0015/0001; RD12/0015/0012; RD16/0016/0009) - cofinanced by European Development Regional Fund “A way to achieve Europe”, Operative program Intelligent Growth 2014-2020. M.L.G.M. is supported by the program FPU (Formación de Profesorado Universitario; FPU13/04545), Ministerio de Educación, Cultura y Deporte, Spain. M.E.P.I. has a grant by the Ministerio de Economía y Competitividad, Instituto de Salud Carlos III, cofinanced by the European Development Regional Fund (“A way to achieve Europe”), and by the Spanish Network for Research in Infectious Diseases (grant REIPI RD16/0009). M.J.M. is supported by the Subprograma Miguel Servet (CPII16/00061), Instituto de Salud Carlos III, Ministerio de Economía, Industria y Competitividad. 4 SCIENTIFIC PRODUCTION RESULTING FROM THE THESIS Publications 1. ML Gil-Marqués, ME Pachón-Ibáñez, J Pachón, Y Smani. Effect of hypoxia on the pathogenesis of Acinetobacter baumannii and Pseudomonas aeruginosa in vitro and in murine experimental models of infections. Infection and Immunity 2018; Aug 6. pii: IAI.00543-18. doi: 10.1128/IAI.00543-18. [Epub ahead of print] Congress communications 1. ML Gil-Marqués, A Puppo, A Gutiérrez-Pizarraya, A Díaz, Y Smani, M McConnell, J Pachón, J Garnacho, ME Pachón-Ibáñez. Tissue hypoxia, HIF-1α and inflammation biomarkers in patients with septic shock. 26th European Congress of Clinical Microbiology and Infectious Diseases (ECCMID). 9-12 April 2016; Amsterdam, Netherlands. 2. A. Díaz Martín*, ML Gil-Marqués*, A. Puppo Moreno, A. Gutiérrez Pizarraya, Y. Smani, M. McConnell, I. Palacios García, J. Pachón Díaz, J. Garnacho Montero, ME Pachón Ibáñez. Hipoxia Tisular, HIF-1α y Biomarcadores de Inflamación en Pacientes con Shock Séptico. LI Congreso Nacional de la Sociedad Española de Medicina Intensiva y Unidades Coronarias (SEMICYUC). 19-22 June 2016; Valencia, Spain. *The authors contributed equally. 3. Puppo AM*, Gil Marqués ML*, Gutierrez Pizarraya A, Diaz A, Pachón Ibañez ME, Pachón J, Escoresca A, Garnacho J. Factor HIF-1α e Hipoxia Tisular en pacientes con Shock Séptico. LI Congreso Nacional de la Sociedad Española de Medicina Intensiva y Unidades Coronarias (SEMICYUC). 19-22 June 2016; Valencia, Spain. *The authors contributed equally. 5 4. A. Díaz-Martín*, ML Gil-Marqués*, A. Puppo-Moreno, A. Gutiérrez-Pizarraya, Y. Smani, M. McConnell, I. Palacios-García, J. Pachón-Díaz, J. Garnacho-Montero, ME Pachón-Ibáñez. HIF-1α levels and inflammation biomarkers in hypoxemic septic shock patients. 29th European Society of Intensive Care Medicine (ESICM) Annual Congress. 1-5 October 2016; Milan, Italy. *The authors contributed equally. 5. ML Gil-Marqués, A Puppo, Y Smani, A Díaz, J Pachón, J Garnacho, ME Pachón- Ibáñez. Hipoxia tisular, Factor Inducible por Hipoxia 1α (HIF-1α) y biomarcadores de inflamación en pacientes con shock séptico. XVIII Congreso de la Sociedad Andaluza de Enfermedades Infecciosas (SAEI). 24-26 November 2016; Córdoba, Spain. 6. Gil-Marqués ML, Pachón-Ibáñez ME, Pachón J, Smani Y. In vitro effect of hypoxia on the infections caused by Acinetobacter baumannii and Pseudomonas aeruginosa. 27th European Congress of Clinical Microbiology and Infectious Diseases (ECCMID). 22-25 April 2017; Vienna, Austria. 7. ML Gil-Marqués, Y Smani, J Pachón, ME Pachón-Ibáñez. In vivo effect of hypoxia on infections caused by Acinetobacter baumannii and Pseudomonas aeruginosa in different murine models. 27th European Congress of Clinical Microbiology and Infectious Diseases (ECCMID). 22-25 April 2017; Vienna, Austria. 8. ML Gil-Marqués, Y Smani, M McConnell, ME Pachón-Ibáñez, J Pachón. Global transcriptional profiling of Acinetobacter baumannii under microaerobiosis and normoxy conditions. 11th International Symposium on the Biology of Acinetobacter. 20-22 September 2017; Sevilla, Spain. 6 OTHER SCIENTIFIC PRODUCTION ASSOCIATED WITH THE RESEARCH LINE 1. MR Pulido, M García-Quintanilla, ML Gil-Marqués, MJ McConnell. Identifying targets for antibiotic development using omics technologies. Drug Discovery Today 2016; Mar;21(3):465-72. doi: 10.1016/j.drudis.2015.11.014. Epub 2015 Dec 12. 2. ML Gil-Marqués, P Moreno-Martínez, C Costas, J Pachón, J Blázquez, MJ McConnell. Peptidoglycan recycling contributes to intrinsic resistance to fosfomycin in Acinetobacter baumannii. Journal of Antimicrobial Chemotherapy 2018; Aug 13. doi: 10.1093/jac/dky289. [Epub ahead of print] 7 ABBREVIATIONS ....................................................................................................... 10 INTRODUCTION ........................................................................................................ 15 1. Healthcare-acquired infections (HAIs) due to gram-negative bacteria .............. 15 2. Current situation of antimicrobial resistance ........................................................ 16 3. Gram-negative bacilli (GNB): epidemiology, antimicrobial resistance, virulence factors and pathogenesis .............................................................................................. 19 3.1. Acinetobacter baumannii .................................................................................... 20 3.1.1. Epidemiology and antimicrobial resistance .............................................. 20 3.1.2. Virulence factors and pathogenesis ........................................................... 22 3.2. Pseudomonas aeruginosa ................................................................................... 27 3.2.1. Epidemiology and antimicrobial resistance .............................................. 27 3.2.2. Virulence factors and pathogenesis ........................................................... 28 4. Hypoxia influence on infections and immune response ........................................ 33 4.1. Hypoxia regulation: HIF-1α ............................................................................. 33 4.2. Functions of HIF-1 ............................................................................................. 35 4.3. Hypoxia and infection ........................................................................................ 37 4.4. HIF-1α as a pharmacological target for infection treatment ......................... 43 FUNDAMENTS ............................................................................................................ 44 HYPOTHESES ............................................................................................................. 46 OBJECTIVES ............................................................................................................... 47 8 METHODS AND RESULTS ....................................................................................... 49 Chapter I. Article I. .................................................................................................. 51 Chapter II. Article II. ............................................................................................... 95 Chapter III. Article III. .......................................................................................... 175 DISCUSSION .............................................................................................................. 198 Chapter I. Article I. Comparative gene expression profile of Acinetobacter baumannii growing under microaerobiosis and normoxia. ................................ 200 Chapter II. Article II. Effect of hypoxia on the pathogenesis of Acinetobacter baumannii and Pseudomonas aeruginosa in vitro and in murine experimental models of infections. ................................................................................................ 203 Chapter III. Article III. Predictive value of APACHE II, and serum lactate, pyruvate, IL-10 and lysophosphatidylcholine levels of survival in patients with septic shock .............................................................................................................. 208 CONCLUSIONS ......................................................................................................... 211 REFERENCES ........................................................................................................... 213 9 negative pathogens responsible for nosocomial infections are Pseudomonas aeruginosa, Acinetobacter baumannii, Escherichia coli, Klebsiella pneumoniae, and Enterobacter species [11, 12, 13]. 2. Current situation of antimicrobial resistance Antimicrobial resistance in gram-negative pathogens causing nosocomial infections has been a major concern in recent years due to the emergence of multi-drug resistant (MDR) strains and the lack of active antibiotics against them [14]. The 2014 report by the World Health Organization (WHO) warned about this circumstance, and remarked that a post-antibiotic era is a real possibility for the 21st century [15]. These MDR strains have developed effective resistance mechanisms like antibiotic efflux pumps or modifications of the outer membrane proteins, among others, to survive to the antibiotic treatment [16]. The development of new antibiotic resistance mechanisms has led to the emergence of extensively-drug resistant (XDR) and even pan-drug resistant (PDR) organisms [17]. The European Center for Disease Prevention and Control (ECDC) met in 2008 to establish standard definitions for MDR (resistant to more than 1 agent in 3 or more antimicrobial categories), XDR (non-susceptible to more than 1 agent in all but 2 categories), and PDR (resistant to all categories), as well as to begin to establish consistency in the categorization of susceptible and non-susceptible for different organisms and antimicrobial classes [18]. The objective of developing such definitions was for public health and epidemiology purposes. Numerous protocols have been taken to contain multi-drug resistant gram-negative bacilli, such as hand hygiene to prevent patient-to-patient transmission of pathogens or contact isolation precautions including wearing gloves and a gown [1, 19]. However, 16 these infections have still a significant clinical impact. The ECDC reported in 2014 that the percentages of organisms exhibiting antimicrobial resistance, especially resistance to multiple antibiotics, continued to increase in Europe. Data from the European Antimicrobial Resistance Surveillance Network (EARS-Net) shows large variations in percentages of antimicrobial resistance in Europe depending on microorganism, antimicrobial agent and geographical region [2]. The use of antimicrobial drugs has become widespread over several decades, and has been extensively misused in both humans and food-producing animals, where it is not only used for disease treatment, but also for prevention, control, and growth promotion [20]. This has contributed to the selection and spread of resistant bacteria. In 2012, consumption of antibiotics for systemic use in the community ranged from 11.3 to 31.9 defined daily doses (DDD) per 1000 inhabitants and per day. The DDD per 1000 inhabitants and per day was 20.9 in Spain in 2012, while in the European Union/European Economic Area (EU/EEA) population was 21.5 DDD. Among the 29 reporting countries, just one country reported a significant decrease in antibiotic consumption [2]. Therefore, it is necessary to promote prudent use of antimicrobial agents and comprehensive infection control measures to reduce the selection and control the transmission of resistant bacteria. Vasudevan et al. [21] reported a prediction tool for nosocomial multi-drug resistant gram-negative bacilli infections in critically ill patients to help the clinicians to identify critically ill patients who are at risk of antibiotic resistant gram-negative infections. A similar study was reported by Ren et al. [22] and they conclude that the active screening and culture of MDR strains in patients has important value for the controlling and prevention of the infection. This should lead to targeted antibiotic treatment and avoid antibiotic overuse which worsens the vicious cycle of resistance in the ICU. Furthermore, University Hospital Virgen del Rocío 17 (Seville, Spain) launched the Institutional Programme for the Optimization of Antimicrobial Treatment (PRIOAM) in 2011. This programme is coordinated by a multidisciplinary team chosen by the Committee on Infections and Antimicrobials and its aim is to educate, train and promote the knowledge for the proper use of antimicrobials in order to reduce mortality and morbidity in patients with infections and delay the development of resistance [20]. Unfortunately, the increasing problem of multidrug resistance in gram-negative pathogens was not paralleled by the development of novel antimicrobials, as we can see in the Figure 1, which has led to the persistence and spread of these resistant strains [23]. Due to a diminishing antibiotic pipeline and an alarming rise in MDR gramnegative bacteria, clinicians are reintroducing older antibiotics (eg, aminoglycosides, fosfomycin or colistin) and examining new strategies to optimize the treatment with existing classes of antibiotics, such as combination antibiotic therapy [24]. Moreover, small companies are trying to modify compounds in existing classes used in human or animal health to circumvent emerging resistance mechanisms and to improve pharmacokinetics [25]. Furthermore, there are a few new compounds which are being clinically studied in some trials and might be used in the next future, like plazomicin or cefiderocol [26]. 18 Figure 1. Antibiotic discovery and resistance timeline. (Health matters: antimicrobial resistance-GOV.UK). As mentioned above, the most common gram-negative pathogens responsible for nosocomial infections are P. aeruginosa, A. baumannii, E. coli, K. pneumoniae, and Enterobacter species [11, 12, 13]. All of them form the “ESKAPE” group of pathogens, together with Enterococcus faecium and Staphylococcus aureus, which was defined in 2008 [27] and then embraced by the Infectious Diseases Society of America [28]. These are pathogens that cause the majority of HAIs while effectively “escaping” the effects of available antimicrobials [19]. 3. Gram-negative bacilli (GNB): epidemiology, antimicrobial resistance, virulence factors and pathogenesis The emergence of MDR GNB creates a challenge in the treatment of nosocomial infections. GNB are the most common causes of nosocomial infections, especially in the ICU, including most cases of hospital-acquired pneumonia and urinary tract infections and 25% to 30% of bloodstream and surgical site infections [29]. One of the 19 most common causes of HAIs are infections caused by the non-lactose fermenting bacteria A. baumannii and P. aeruginosa. Both bacteria are considered as priority 1 (critical) for research and development of new antibiotics by the WHO in 2017 [30]. 3.1. Acinetobacter baumannii 3.1.1. Epidemiology and antimicrobial resistance Acinetobacter spp. are glucose-non-fermentative, non-motile, non-fastidious, oxidativenegative, catalase-positive, aerobic gram-negative coccobacilli with a DNA G+C content of 39% to 47% [31]. Its ability to survive to desiccation and colonize any type of surface has made this pathogen one of the most common species causing nosocomial outbreaks in hospitals [32]. In the ECDC point prevalence survey of HAIs in European acute care hospitals 2011-2012, Acinetobacter spp. were the 11th most frequently reported microorganisms (3.6%) in microbiologically documented HAIs [2]. Among Acinetobacter species, Acinetobacter baumannii is the most important pathogen associated with HAIs [33]. Most A. baumannii infections occur in critically ill patients in the ICU and account for up to 20% of infections in ICUs worldwide [34]. A. baumannii is a successful pathogen responsible of opportunistic infections of the lower respiratory tract, skin, bloodstream, urinary tract, and other soft tissues [35, 36]. Crude mortality rates of 30-75% have been reported for nosocomial pneumonia caused by this pathogen, and the mortality attributable to A. baumannii infection was found to range from 7.8-43%, with higher levels in patients admitted to ICUs (10-43%) [37]. A. baumannii has a high level of intrinsic resistance to many groups of antimicrobials (e.g. glycopeptides, macrolides, lincosamides, and streptogramins). Moreover, it is able to develop resistance to all classes of antimicrobial agents used in the therapy [38]. For this reason, the resistance of A. baumannii has been highly increased in the last decades, 20 which supposes an important problem for the health system. Some European countries reported that more than half of Acinetobacter spp. isolates were resistant to all antimicrobial categories under surveillance (carbapenems, fluoroquinolones and aminoglycosides) [2]. Multicenter surveillance studies have also reported that the proportion of imipenem-resistant A. baumannii strains is as high as 85% in bloodstream isolates from ICU patients in Greece, and 48% in clinical isolates from hospitalized patients in Spain and Turkey [39]. The rapid emergence of MDR and PDR strains of Acinetobacter highlights the organism’s ability to quickly acclimatize to selective changes in environmental pressures [31] and to acquire antimicrobial resistance [35, 40]. A number of A. baumannii resistance mechanisms are known, including enzymatic degradation of drugs (β-lactamases…), target modifications (PBP2, ArmA…), multidrug efflux pumps (AdeABC, AdeFGH, CmlA…), and permeability defects (OmpA, CarO…) [33, 35]. Figure 2 shows a summary about virulence factors, pathogenesis, antimicrobial resistance and treatment options of this successful pathogen which is important to develop new strategies for combating MDR A. baumannii infections. 21 Figure 2. Biology of Acinetobacter baumannii [35]. 3.1.2. Virulence factors and pathogenesis A. baumannii use their virulence factors to first colonize and then infect the host. Different genomic, transcriptomic and proteomic analyses have helped to identify several virulence factors that participate in the pathogenesis of A. baumannii, including outer membrane porins, proteases, phospholipases, protein secretion systems, lipopolysaccharides (LPS), capsular polysaccharides, and iron-chelating systems [35, 41]. Although recent genomic and phenotypic analyses have identified several virulence factors responsible for its pathogenicity, we still know relatively few virulence factors in A. baumannii, compared to other GNB [42]. Due to the increasing antimicrobial resistance rates and the lack of treatments to combat infections, it is important to identify new virulence factors to characterize the pathogenesis and determine new therapeutic targets that allow the control of this kind of infections. 22 The main identified virulence factors of A. baumannii are: a. Porins: A. baumannii, as the rest of GNB, has a double membrane (outer and inner membrane) which constitutes the first line of contact between bacteria and its external environment (Fig. 3). These two membranes, which differ substantially in their compositions, are separated by the periplasmic space [43]. This double membrane is a selective barrier which acts as a protection mechanism and it allows the entry of nutrients to promote cell survival. Figure 3. Structure of a GNB double membrane. (https://en.wikipedia.org/wiki/Gram-negative_bacteria) The outer membrane has a unique composition and asymmetrical distribution of lipids, with the inner leaflet containing phospholipids, whereas the outer leaflet is composed of lipopolysaccharide (LPS), a highly negatively charged molecule [44]. The outer 23 membrane is mainly composed by proteins, called outer membrane proteins (OMPs) or porins, which are associated with the modulation of cellular permeability [35]. These proteins are involved in nutrient uptake, cell adhesion, cell signaling and waste export [45]. Moreover, for the pathogenic strains, these OMPs also serve as virulence factors for nutrient scavenging and evasion of host defense mechanisms [44]. One of the main OMPs that has been characterized is OmpA. OmpA is a β-barrel porin and one of the most abundant porins in the outer membrane [35]. It plays a considerable role in adherence and invasion of epithelial cells by interacting with fibronectin [46], and binds to factor H in human serum which may allow A. baumannii to resist complement-mediated killing [47]. OmpA is also involved in antibiotic resistance of A. baumannii [48, 49] and in biofilm formation [50]. Omp33-36 is another OMP associated with A. baumannii cytotoxicity [51]. Other porins, such as carbapenem-associated outer membrane protein (CarO) and OprD-like, are also virulence factors associated with attenuated virulence in a mouse model [52]. b. Capsular polysaccharides and lipopolysaccharides (LPS): Capsular polysaccharides are involved in antimicrobial resistance of A. baumannii. It is known that mutants deficient in capsular polysaccharides have lower intrinsic resistance to peptide antibiotics. Moreover, the presence of antibiotics induces hyperproduction of capsular polysaccharides [53]. This increased capsule production depends on the regulation of K locus gene expression by bfmRS two-component regulatory system. bfmR is important for persistence in the lung in a murine pneumonia model [54] and bfmS is involved in biofilm formation, adherence to eukaryotic cells, and resistance to human serum [55]. Capsular polysaccharides also promote survival during periods of desiccation due to their ability to retain water. 24 LPS is the major component of the outer leaflet of the outer membrane in most GNB and is an immunoreactive molecule that induces release of tumor necrosis factor and IL- 8 from macrophages in a Toll-like receptor 4 (TLR4)-dependent manner [35]. LPS is composed of an endotoxic lipid A moiety, an oligosaccharide core, and a repetitive O- antigen and it plays a major role in virulence and survival of A. baumannii [56, 57]. c. Phospholipase: Phospholipase is a lipolytic enzyme essential for phospholipid metabolism and is a virulence factor in A. baumannii [58]. There are three classes of phospholipases: phospholipase A (hydrolyzes fatty acids from the glycerol backbone), phospholipase C (cleaves the phosphorylated head group from the phospholipid), and phospholipase D (transphosphatidylase that only cleaves off the head group). Degradation of phospholipids affects the stability of host cell membranes, and the cleaved head group can interfere with cellular signaling, changing the host immune response [58, 59]. d. Outer Membrane Vesicles (OMVs): OMVs are spherical, 20-200 nm diameter vesicles secreted by various pathogenic GNB. They are composed of LPS, outer membrane and periplasmic proteins, phospholipids, and DNA or RNA, and are recognized as delivery vehicles for bacterial virulence factors to the interior of host cells [60]. These virulence factors include OmpA, proteases and phospholipases [61]. A. baumannii OMVs has also been reported to participate in the spread of antibiotic resistance and induce the horizontal transfer of the OXA-24 carbapenemase gene [62]. 25 the severity of lung disease [106]. Lipid A modifications may be regulated and induced by environmental changes [78]. P. aeruginosa can present simultaneously two types of O-polysaccharide, the A-band and the B-band. A-band elicits a weak antibody response while B-band elicits a strong antibody response [104]. Some strains do not produce any O-polysaccharide and other strains just produce one of them [78]. Many chronic isolates change their proportion of A-band and B-band to evade host adaptive immune response [104]. h. Other virulence factors: Exotoxin A is an ADP-ribosyltransferase that inhibits host elongation factor EF2, so it inhibits protein synthesis and causes cell death. Moreover, it also represses host immune response [107]. Exotoxin A also induces host cell death by apoptosis [108] and strains that produce it show more virulence in a murine model of infection [109]. Phospholipases are also virulent factors of P. aeruginosa because they break down lipids and phospholipids of host cell membranes [92]. Pyocyanin is a blue-green pigment of P. aeruginosa that causes oxidative stress to the host, disrupting its catalase and mitochondrial electron transport [110]. Moreover, it is known that pyocyanin in vitro induces apoptosis in neutrophils and inhibit the phagocytosis of apoptotic bodies by macrophages [110, 111]. Pyocyanin also retards the growth of some other bacteria and thus facilitates colonization by P. aeruginosa [103]. This pathogen also needs iron chelation systems to establish and produce a chronic infection, since host environment has little free iron available. Therefore, P. aeruginosa has several siderophores like pyoverdine or pyochelin. Pyoverdine also acts like a 32 signaling molecule causing the upregulation of exotoxin A, endoprotease and of pyoverdine itself [112]. 4. Hypoxia influence on infections and immune response 4.1. Hypoxia regulation: HIF-1α Hypoxia occurs when cellular oxygen demand exceeds supply. Tissue hypoxia can occur in a wide variety of pathologic conditions including vascular disease, cancer, dermal wounds and respiratory illnesses, and it is common during inflammation associated with acute or chronic bacterial infection [113, 114]. The reasons for the occurrence of hypoxia (or even anoxia) during inflammation is a combination of increased oxygen demand in order to satisfy the requirements of inflamed resident cells, infiltrating inflammatory cells and in some cases multiplying pathogens, along with a decreased perfusion due to vascular dysfunction (during chronic inflammation) [114]. Thus, oxygen levels in the foci of infection are much lower (<1%) than in healthy tissues (2.5-9%) [115]. It is essential that eukaryotic cells, especially immune cells, remain effective under such hypoxic conditions. Therefore, eukaryotic cells have developed the ability to adapt to the hypoxic environment through the transcriptional regulation of multiple genes [116]. One of them is the hypoxia inducible factor 1 (HIF-1), a heterodimeric transcription factor (HIF-1α and HIF-1β) whose expression is regulated at the protein level. Under hypoxia and/or bacterial infection, different oxygen and iron dependent proline hydroxylases (PHDs; PHD1, PHD2 and PHD3) are inhibited, so HIF-1α accumulates and translocates into the nucleus, where it dimerizes with HIF-1β, constitutively expressed [113]. Then, the heterodimeric HIF-1 make a complex with the p300/CBP 33 transcriptional coactivator protein and this active HIF-1 complex binds to the hypoxic response elements (HREs). These bindings induce the expression of multiple genes involved in angiogenesis, glycolysis and cellular stress, among others, as well as the repression of genes involved in a reduction in energy demanding processes (Fig. 4) [117]. In addition, HIF-1 is also activated under iron deprivation because Fe2+ is an essential cofactor of PHDI-3, which promotes HIF-1 degradation [118]. The HIF family also comprises the HIF-2 and HIF-3, but they have been less studied. HIF-2 has similar functions to HIF-1, but HIF-3 seems to inhibit HIF-1 and HIF-2 activities [119]. Under normoxic conditions, oxygen and iron dependent proline hydroxylases are active and hydroxylate the proline residues of HIF-1α. Moreover, the hydroxylation of an asparagine residue by the Factor Inhibiting HIF-1 (FIH) on HIF-1α blocks the interactions between p300/CBP transcriptional coactivator proteins and the α subunit, thereby preventing the transcription of several genes. Finally, the hydroxylated prolines become ubiquitinated by the von Hippel-Lindau tumor suppressor protein and the HIF- 1α undergoes proteosomal degradation. Therefore, HIF-1 gets inhibited and degraded in the presence of oxygen (Fig. 4) [120, 118]. 34 Figure 4. HIF-1α pathway during normoxia and hypoxia [120]. 4.2. Functions of HIF-1 a. Recovery after hypoxia/ischemia: HIF-1 prevent tissue damage induced by ischemia by two fundamental mechanisms: shifting the cellular metabolism to the anaerobic mode, and promoting new vascularization towards the hypoxic areas hence increasing oxygen supply. Several studies show that HIF-1 overexpression induces angiogenesis in hypoxic tissues and it can lead to increased oxygenation of the organ [121]. 35 b. Induction of proinflammatory and antimicrobial responses: After microbial infection, the immune system activates an immune reaction starting with the innate immune response. A key element is the activation of polymorphonuclear leukocytes (PMN-s) which are parts of the myeloid cell family. PMN-s seek out, migrate, identify, phagocytize and eliminate the invading microbes by releasing reactive oxygen species (ROS). ROS are generated in PMN-s in a process called respiratory burst. Recent studies employing HIF-1α gene knock out mice showed that HIF-1α is a regulator of energy metabolism, aggregation, migration and bactericidal activity of PMN-s [122, 123]. In addition, it is known that HIF-1α expression plays a crucial role in the differentiation of myeloid cells into monocytes and macrophages [123]. c. Shifting metabolism toward anaerobic mode: The oxidative phosphorylation is the main source of adenosine triphosphate (ATP) in human cells during normoxia. However, the cellular metabolism needs to be shifted toward anaerobic energy production during hypoxia. HIF-1 is one of the principal molecules to regulate this shift. HIF-1 activates glucose transporters, aldolase A, pyruvate kinase M and a number of glycolytic enzymes [124]. In addition, HIF-1 efficiently downregulates mitochondrial oxygen consumption [125]. Therefore, HIF-1 helps cells to produce energy even in environments of low oxygen concentrations. d. Induction of angiogenesis: HIF-1 promotes angiogenesis in a number of different tissues. During hypoxia, HIF-1α binds the transcription regulatory region of the VEGF gene and induces its transcription and translation [124]. Then the VEGF induces migration of mature endothelial cells toward the hypoxic tissue. These activated endothelial cells start producing new blood vessels which supply the hypoxic areas with more blood and oxygen [126]. 36 e. Promoting tumor progression: When a tumor grows, its quickly dividing cells need an enormous amount of energy, and this induces angiogenesis to meet the increasing needs for blood supply providing oxygen, glucose and other essential molecules. In cancerous tissues HIF-1α activates VEGF inducing angiogenesis, as stated above. Moreover, HIF-1 promotes anaerobic metabolic adaptation of hypoxic metastasizing tumors. It has been found that in vHL- lacking renal carcinoma cells, HIF-1 activation decreases oxygen consumption by inhibiting C-MYC, a transcription factor that regulates mitochondrial oxygen consumption. The down-regulation of C-MYC results in increased glycolysis and suppressed mitochondrial respiration [127, 128]. f. Pro-apoptotic effects: Hypoxia can activate the p53 tumor suppressor via HIF-1, and p53 induces p21 which promotes cell death by apoptosis in embryonic stem cells. A different study in primary alveolar epithelial cells reported that low oxygen levels induce apoptosis via the HIF-1 pathway [129]. The majority of studies describe the HIF-1 as an apoptosis promoting factor, but it has also been reported that in hypoxia-challenged neonatal brain tissues the activation of HIF-1 protected the cells by preventing them from undergoing apoptosis [130]. 4.3. Hypoxia and infection Tissue hypoxia happens in vivo during a range of infections (bacterial, viral and fungal infections). For example, hypoxia occurs within the mucus filled airways of cystic fibrosis patients who are often infected with the pathogen P. aeruginosa [131]. Furthermore, studies have reported that Mycobacterium tuberculosis resides in a hypoxic environment within granulomas in guinea rabbits, pigs, and primates [132]. 37 Pulmonary infection with Aspergillus fumigatus in mice also leads to the establishment of a hypoxic microenvironment in the infected tissue [133]. Sendai virus (also known as a murine parainfluenza virus) responsible for respiratory diseases, shows enhanced replication under hypoxia condition [134]. In summary, tissue hypoxia is a common microenvironmental feature in a different number of infections. HIF-1 is activated during infections of all kind of pathogens, including group A and b Streptococci, Staphylococcus aureus, Salmonella typhimurium Escherichia coli, Chlamydia pneumoniae and P. aeruginosa [135]. This HIF-1 activation occurs due to the low oxygen level because of the presence of a high amount of bacteria in the foci of infection that are consuming it (Fig. 5). It was first reported in some studies using the pathogen Bartonella henselae [136]. In B. henselae infections, enhanced oxygen consumption has been associated with the expression of the outer membrane protein Bartonella Adhesin A (BadA), which is a virulence factor involved in adherence (Fig. 5). Cells infected with a B. henselae mutant in BadA did not get hypoxic upon infection, and no HIF-1 activation was detected. In addition, HIF-1 activation also occurs due to low iron concentrations. Klebsiella pneumoniae siderophores chelate the host cellular iron, causing an inactivation of iron-dependent prolyl hydroxylases and HIF-1α stabilization in lung epithelial cells (Fig. 5). This induces vascular permeability and angiogenesis, so K. pneumoniae can disseminate better under hypoxia than normoxy [137]. Recent studies have shown that bacterial lipopolysaccharide (LPS), a component of the gram negative bacteria, also activates HIF-1α in a TLR4 dependent fashion in macrophages and neutrophils under normoxic conditions due to the activation of p44/42 MAPK and NF-κβ pathways, because HIF-1α gene contains binding sites for this factor (Fig. 5) [138]. 38 Figure 5. Influence of the interaction pathogen-host cells results in HIF-1 activation [118]. Hypoxia and HIF-1 have a protective role in the defense against pathogens, mainly due to NF-κβ, which promote the bactericidal capacity of phagocytes, monocyte/macrophages, dendritic cells, neutrophils and, even, epithelial cells, through the production of inflammatory cytokines and antimicrobial peptides such as cathelicidins (Fig. 6) [115, 139, 140]. Several mouse models have been used to investigate the role of HIF in infections. For example, knockout (KO) mice in HIF-1 are more susceptible to bacterial infection [141]. Moreover, HIF-1α keratinocytes KO mice have shown to develop larger necrotic lesions and had decreased capacity to clear group A Streptococcus because they recruited less neutrophils to the site of infection [142, 143]. Internalization of P. aeruginosa into airway epithelial cells is also lower under hypoxia than normoxia [144] and mice infected with Mycobacterium tuberculosis under hypoxic conditions shown increased survival time compared with those under normoxia 39 [113]. HIF-1α KO mice displayed impaired leucocyte invasion and inflammation using ear and skin inflammation models [122] and BALB/c mice, which are resistant to P. aeruginosa keratitis, suffer severe infection when HIF-1α is knocked down using siRNA [145]. Moreover, another study showed that IL-6 production was lower under hypoxia than normoxia in a keratinocyte culture, but a treatment using a compound which boosts HIF resulted in an increase in IL-6 levels compared even to normoxia. These treated keratinocytes were more efficient in the clearance of S. aureus, P. aeruginosa and A. baumannii. This result suggests that additional changes during hypoxia, and not HIF levels alone, can alter cytokine production and immune response [142, 146]. Figure 6. HIF-1α transcriptional regulation of phagocyte innate immune functions [115]. 40 However, HIF-1 can have a negative role in the course of systemic infection, e.g. in sepsis. A KO of HIF-1α in the myeloid cells showed increased survival rates in a murine LPS-induced sepsis model and decreased serum levels of proinflammatory cytokines [147]. VEGF, another HIF-1 regulated cytokine, is elevated in serum of septicemic patients. Methicillin-resistant S. aureus (MRSA) and Streptococcus pneumoniae infections in meningitis patients induce higher serum VEGF levels [148, 149]. Blocking of VEGF with soluble VEGF receptors increased survival in a murine LPS-sepsis model [150]. This harmful effect of VEGF and HIF-1 might be mediated by enhanced overwhelming immune response and subsequently to vascular leakage, and organ failure, which has already been demonstrated for patients suffering from septic shock (Fig. 7) [151]. HIF-1α is also a determinant of sepsis phenotype through the production of inflammatory cytokines (IL-1, IL-4, IL-6, IL-12 and TNF-α). High cytokines levels may be harmful to the host during early sepsis [152]. Therefore, a potential use of HIF-1 inhibitors in sepsis patients could be an interesting option, especially in times of widespread antibiotic resistances in clinically relevant pathogens, but there is no information from clinical studies available if inhibition of HIF-1 is beneficial for the outcome so far [118]. 41 5. To study the effect of hypoxia on the innate immune response and prognosis in lethal murine models of sepsis by A. baumannii and pneumonia by P. aeruginosa. Objectives of chapter III: 1. To investigate the relationship between tissue hypoxia, HIF-1α levels and immune response in patients with septic shock (20). 2. To determine new biomarkers to predict the outcome of septic shock patients. 48 METHODS AND RESULTS Chapter I. Article I. Comparative gene expression profile of Acinetobacter baumannii growing under microaerobiosis and normoxia. It will be submitted to The Journal of Infectious Diseases. Chapter II. Article II. Effect of hypoxia on the pathogenesis of Acinetobacter baumannii and Pseudomonas aeruginosa in vitro and in murine experimental models of infections. Published in Infection and Immunity 2018. Chapter III. Article III. Predictive value of APACHE II, and serum lactate, pyruvate, IL-10 and lysophosphatidylcholine levels on survival in patients with septic shock. It will be submitted to Critical Care Medicine. 49 50 CHAPTER I. ARTICLE I. COMPARATIVE GENE EXPRESSION PROFILE OF ACINETOBACTER BAUMANNII GROWING UNDER MICROAEROBIOSIS AND NORMOXIA. Acinetobacter baumannii is an aerobic Gram-negative pathogen responsible for healthcare-associated infections of the respiratory tract, skin, bacteremia, urinary tract, and other soft tissues [1]. During the process of infection, bacteria adapt to different environments modifying their gene expression [2]. It is known that tissue hypoxia occurs in the course of inflammation, due to a higher rate of oxygen consumption by immune cells and pathogens together with a reduction of the perfusion caused by vascular dysfunction [3]. This means that oxygen levels in the foci of infection are much lower (<1%) than in healthy tissues [4]. Moreover, there are different common medical conditions that produce hypoxemia and peripheral tissue hypoxia and they are often associated with infection and inflammation [5]. Therefore, bacteria must adapt to a hypoxic environment during infection, which modifies the expression of genes involved in metabolism and virulence [6]. Several virulence factors of A. baumannii have been identified through genomic and phenotypic analyses [7]. However, we still have not a whole picture of its pathogenic mechanisms. Nowadays, techniques that utilize nextgeneration sequencing, such as RNA-seq, have been used for characterization of bacterial genomes under different conditions [8, 9]. Some analyses of the A. baumannii transcriptome have provided important information regarding A. baumannii biological characteristics. RNA-seq has been used to characterize and compare gene expression in A. baumannii under biofilm and planktonic growth conditions [10, 11], to analyze A. baumannii response and resistance mechanisms to different antibiotics and antibacterial agents [12-16], and to examine genetic changes that happen during host infection, 51 bacteremia and treatment [17, 18]. However, no study has investigated the influence of hypoxia/microaerobiosis in A. baumannii growth, which is an important condition that bacteria have to adapt to during infection. The objective of this study was to identify genes of A. baumannii whose expression is regulated by environmental oxygen levels to identify virulence factors. Thus, we compare the transcriptional response in A. baumannii ATCC 17978 under microaerobiosis (0.1-0.3% of oxygen) and normoxia (21% of oxygen) growth conditions. We identified several up and down-regulated genes under microaerobiosis. One of the upregulated genes was A1S_2448 (pstS), which is part of the pst operon (pstA, pstB, pstC, pstS and phoU) encoding a high-affinity phosphate transport system that is activated under low phosphate conditions. This transport system together with the low-affinity transporter Pit are the main phosphate uptake systems of A. baumannii [19]. Materials and Methods Bacterial Strain and Growth curve analysis A. baumannii ATCC 17978 was used in this study. Independent growth in MHB, M9 minimal medium and phosphate-limiting M9 minimal medium was evaluated over time for wild-type strain ATCC 17978 and the pstS mutant. Strains were grown overnight in 20 ml of Mueller-Hinton broth (MHB, Sigma) in static, and then, a 1:10000 dilution was performed to obtain 105 cfu/ml in a 40 ml culture of MHB or minimal medium M9. Minimal medium M9 was supplemented with 0.4% glucose, 2 mM magnesium sulfate and 0.1 mM calcium chloride. The growth under microaerobiosis (0.1-0.3% O2), hypoxia (1% O2), and normoxia (21% O2) were monitored during 24 h at 37 ºC and 160 rpm. We used a hypoxia chamber (Coy Laboratories, Grass Lake, MI, USA) to culture 52 the bacteria under microaerobiosis and hypoxia conditions. After 4 h of growth in MHB, 6 samples (3 of microaerobiosis and 3 of normoxia condition) of bacterial cultures (8·108 cells each) were taken to perform RNA extraction. Bacterial RNA extraction, sequencing and analysis Bacterial RNA was purified from the cultures using the RNeasy Mini Kit (Qiagen). Quality control of the RNA samples (integrity number and concentration) was analyzed prior sequencing. The enrichment of mRNA was carried out by depletion of rRNA with Ribo-Zero rRNA Removal Kit (Illumina). The mRNA enriched fraction was used for library construction of cDNA molecules and the sequencing was performed on Illumina Hiseq2500 platform using 100bp paired-end sequencing reads. Then, the analysis of the generated sequence raw data was performed using CLC Genomics Workbench 8.5.1 (Qiagen) (http://www.clcsupport.com/clcgenomicsworkbench/802/index.php?manual). The bioinformatics analysis started with trimming of raw sequences to generate high quality data. This high-quality sequencing reads were mapped against the A. baumannii ATCC 17978 genome (accession No. NC_009085). The result of mapping against this genome served to determine the gene expression levels based on the Reads per Kilobase of exon model per Million mapped reads (RPKM) method [20]. Finally, a gene differential expression analysis between normoxia and microaerobiosis groups was carried out through an Empirical Analysis of Digital Gene Expression test (DGE-test) [21]. The differentially expressed genes were filtered using standard conditions: False Discovery Rate P-value ≤ 0.05 and Fold change > 2 or < -2 [22, 23]. Genes considered as differentially expressed were used in an enrichment analysis. The R package clusterProfiler, included in Bioconductor [24], was selected with A. baumannii ATCC 17978 (acb) as the organism, and KEGG as the annotation source. All KEGG pathways with a p-value cut-off equal to 1 was taken into account. In addition, genes belonging to 53 the Sulfur metabolism pathway, together with their fold-change values were analyzed with the Pathview web tool to show these genes in their corresponding KEGG pathway graph [25]. Finally, the raw expression for the differentially expressed genes belonging to a KEGG pathway were taken and a heatmap with the 6 different experiments (3 for normoxia, and 3 for microaerobiosis) was built, using the heatmap.2 method from the R gplot package. Quantitative Real-Time PCR (qRT-PCR) verification The RNA was isolated as described in the previous section and using the same conditions. Then, rests of DNA were removed from the RNA samples using the TURBO DNA-free Kit (Invitrogen). The reverse transcription step was carried out using the Thermo-Script RT-PCR kit (Invitrogen), according to the manufacturer’s instructions. The primer3 v.0.4.0 software (http://bioinfo.ut.ee/primer3-0.4.0/) was used to select primers that would amplify a product of approximately 120 bp. We selected two subexpressed genes under microaerobiosis: Glyoxalase/bleomycin resistance protein/dioxygenase A1S_3416 (F: GACCCAAATGGACATCGTTT; R: ATGGAGTAAAACCAAACGCG) and Maleylacetoacetate isomerase A1S_3415 (F: TAGTGGACGGCGATTTAACC; R: AGAAAGAGCCAAAATCCGTG); two overexpressed genes under microaerobiosis: taurine ATP-binding transport system component A1S_1443 (F: GGGTTGTGGCAAAACAACTT; R: TCACGCCTTACTTCCTTGGT) and sulfate transport protein A1S_2531 (F: GCCAGGCGTGGAAATTATTA; R: GGTAACGATGCAAAAGCACA); and two housekeeping genes: gyrB A1S_0004 (F: CAGCTTTGGGAAACCACAAT; R: CGATGATGTTGAACCACGTC) and rpoD A1S_2706 (F: CATGCGTGAAATGGGTACAG; R: TTACTGGCCAAATGCTGTTG). The quantitative real-time PCR assay was performed with SYBR Premix Ex Taq (Takara) in 54 a MxPro 3005p system (Stratagene). Three technical replicates for each sample were included. The amplification conditions were: 95 °C, 30 s, followed by 35 cycles of 95 °C 10 s, 56 °C 25 s, and 72 °C 25 s. The specificity of the reaction was confirmed by a melting curve assay from 95 to 55 °C. Relative quantification of gene expression was analyzed following the Comparative CT Method (Applied Biosystems Guide). Construction of mutant and complemented strains A stable, in-frame deletion mutant strain was constructed in the A. baumannii ATCC 17978 strain by homologous recombination using a described protocol [26]. For construction of the A1S_2448 deletion mutant (ATCC 17978 ΔA1S_2448::Kan), the 500 bp immediately upstream of the genes open reading frame, and the 500 bp immediately downstream were amplified using the primers A1S_2448 Up Forward (CTTGCGGTTTTAGCGATTATG), A1S_2448 Up Reverse (GCCCCAGCTGGCAATTCCGGTCTGTTCTCTCTCATTAATG), A1S_2448 Down Forward (CTAAGGAGGATATTCATATGGTTGGTTTGAATAGGGGCTG) and A1S_2448 Down Reverse (GCGGCACAGACAACAACAGC). The kanamycin resistance gene from the plasmid pKD4 was amplified using the primers Kanamycin Up Forward (CCGGAATTGCCAGCTGGGGC) and Kanamycin Down Reverse (CATATGAATATCCTCCTTAG), resulting in a PCR product with sequences overlapping the PCR fragments containing the sequences upstream and downstream of the genes. The three PCR products were mix in a stitching PCR reaction, and 5 µg of the resulting construct were transformed into the ATCC 17978 strain by electroporation before selection on LB agar plates with 10 mg/L kanamycin. All deletion mutants were confirmed by sequencing and maintained in 30 mg/L kanamycin. 55 In order to complement the obtained mutant, the gene reading frame and 200-400 bp upstream and downstream were amplified using the primers EcoRI A1S_2448 prom. Forward (ACAGAATTCGTGATATTGCGGTTATCTGAC) and XbaI A1S_2448 reg. Reverse (ACATCTAGATATTCTCCACTGTTTTCTCAATTG). This fragment was cloned into the pUCp24 and introduced into the mutant strains by electroporation before selection on LB agar plates containing 10 mg/L of gentamicin to create the complemented strain (ATCC 17978 ∆A1S_2448::Kan/pUCp24-2448). Primers Seq.insert.pUCp24.Forward (TCCCAGTCACGACGTTGTAAAACG) and Seq.insert.pUCp24.Reverse (AATTTCACACAGGAAACAGCTATG) were used to check and sequence the cloned gene. The deletion mutant was also transformed with the empty pUCp24 plasmid for use as controls (ATCC 17978 ∆A1S_2448::Kan/pUCp24). A549 culture and infection Human lungs epithelial cell line A549 was grown in DMEM containing 10% Fetal Bovine Serum (Gibco), 1% HEPES 1M, vancomycin (50 mg/ml), gentamicin (20 mg/ml) and amphotericin B (0.25 mg/ml; Gibco), as previously described [27]. In the case of hypoxia condition studies (1% O2), cells were transferred to a hypoxia chamber with a humidified atmosphere of 1% O2, 5% CO2 and the balance N2 at 37 °C for 6 h prior infection. Cells were seeded (105 cells/well in a 24-well plate) for 30 h in 24-well plates before infection with A. baumannii ATCC 17978 and mutant strains at a multiplicity of infection (MOI) of 500. Immediately before the infection, A549 cells were washed thrice with PBS and incubated in supplemented DMEM. 56 Bactericidal activity, bacterial adherence and bacterial invasion in cell cultures After A549 cells infections with A. baumannii ATCC 17978 and mutant strains under hypoxia and normoxia conditions, extracellular medium was removed and serially diluted to determine bacterial concentration at 2 and 24 h post-infection [27]. Adherence and invasion assays were carried out as previously described [27]. To measure the adherent bacteria number, cells were infected as mentioned before, and, after washing with PBS, 200µl of trypsin-EDTA (Gibco) was added for 5 min at 37 °C. Then, 200 µl of 0.5% Triton X-100 (Sigma) was added for 3 min. The invasion protocol also includes a treatment with tetracycline 256 µg/ml (Sigma) before the addition of trypsin-EDTA. Diluted lysates were counted to determine the attached and internalized bacteria. All assays were performed in triplicate. Biofilm assay Biofilm production was measured based on a previously described method [28]. Strains were cultured in 10 ml MHB overnight at 37 °C, and subsequently diluted to 105 cfu/ml in MHB. Two-hundred µl of the cell suspension were added to each well of a roundbottom 96-well plate and growth overnight at 37 ºC. We washed twice every well to remove non-adherent bacteria and added two-hundred µl of 0.4% crystal violet dye (Sigma). After 10 min incubation, we washed twice and added two-hundred µl of 96% ethanol. After 15 min incubation, biofilm formation was quantified measuring the O.D. at 580 nm (Asys UVM 340 Microplate Reader). All assays were performed in triplicate. Surface motility assay Surface motility was measured based on a previously described method [29]. Overnight cultures of each strain were adjusted to an O.D. at 600 nm of 0.6 in PBS (Lonza). Three 57 Figure 2. A) Clustering of differentially expressed genes assigned to a pathway. All of the genes were grouped by pathways and the expression values were converted to Z- Score for normalization. Note that the same gene can sometimes belong to other of the pathways, but here it is only shown in one of them. B) KEGG pathway for Sulfur metabolism highlighting the differentially expressed genes. In red and green color are highlighted the upregulated and downregulated genes, respectively. 64 Figure 3. A) Growth curves of A. baumannii ATCC 17978 and ΔA1S_2448::Kan in MHB under normoxia, hypoxia and microaerobiosis (21%, 1% and 0.1-0.3% O2, respectively). B) Growth curves of A. baumannii ATCC 17978, ΔA1S_2448::Kan, ATCC 17978/pUCp24, ΔA1S_2448::Kan/pUCp24-2448, and ΔA1S_2448::Kan/pUCp24 in minimal medium M9 (supplemented) under hypoxia. C) Growth curves of A. baumannii ATCC 17978, ΔA1S_2448::Kan, ATCC 17978/pUCp24, ΔA1S_2448::Kan/pUCp24-2448, and ΔA1S_2448::Kan/pUCp24 in minimal medium M9 (supplemented) under normoxia. 65 Figure 4. A) Measurement of bacterial concentration (%) in the extracellular medium after 2 and 24 h of A549 infection by A. baumannii ATCC 17978, ΔA1S_2448::Kan, ATCC 17978/pUCp24, ΔA1S_2448::Kan/pUCp24-2448, and ΔA1S_2448::Kan/pUCp24 under normoxia and hypoxia (1% O2). Bars represent the reduction of bacterial concentration under hypoxia compared to normoxia, being the reduction obtained in the wild-type equal to 100%. ***: P<0.001 Mutant vs. wild-type at 2 or 24 h; **: P<0.01 Mutant vs. wild-type at 2 or 24 h; *: P<0.05 Mutant vs. wild-type at 2 or 24 h. B) Measurement of bacterial adherence (%) after 2 and 24 h of A549 infection by A. baumannii ATCC 17978, ΔA1S_2448::Kan, ATCC 17978/pUCp24, ΔA1S_2448::Kan/pUCp24-2448, and ΔA1S_2448::Kan/pUCp24 under normoxia and hypoxia (1% O2). Bars represent the reduction of bacterial adherence under hypoxia compared to normoxia, being the reduction obtained in the wild-type equal to 100%. **: P<0.01 Mutant vs. wild-type at 2 or 24 h; *: P<0.05 Mutant vs. wild-type at 2 or 24 h. C) Measurement of bacterial internalization (%) after 2 and 24 h of A549 infection by A. baumannii ATCC 17978, ΔA1S_2448::Kan, ATCC 17978/pUCp24, ΔA1S_2448::Kan/pUCp24-2448, and ΔA1S_2448::Kan/pUCp24 under normoxia and hypoxia (1% O2). Bars represent the reduction of bacterial internalization under hypoxia compared to normoxia, being the reduction obtained in the wild-type equal to 100%. ***: P <0.001 Mutant vs. wild-type at 2 or 24 h. 66 Figure 5. A) Effect of PstS loss on biofilm production. Biofilm production was determined for ATCC 17978, ΔA1S_2448::Kan, ATCC 17978/pUCp24, ΔA1S_2448::Kan/pUCp24-2448, and ΔA1S_2448::Kan/pUCp24. Bars represent the average of three separate assays, with error bars representing the SEM. ***: P<0.001 Mutant vs. wild-type; *: P<0.05 Mutant vs. wild-type. B-F) Efect of PstS loss on surface motility (B: ATCC 17978; C: ΔA1S_2448::Kan; D: ATCC 17978/pUCp24; E: ΔA1S_2448::Kan/pUCp24-2448; F:ΔA1S_2448::Kan/pUCp24). 67 Table 1. Genes up-regulated and down-regulated in A. baumannii ATCC 17978 under microaerobiosis (0.1-0.3% O2). Gene ID Protein description Fold Change A1S_0028 FMNH(2)-dependent alkanesulfonate monooxygenase 2,867637691 A1S_0029 ABC-type nitrate/sulfonate/bicarbonate transport system 2,990097961 A1S_0030 Alkanesulfonate transport protein 4,262666192 A1S_0040 Putative oxidoreductase 3,36292177 A1S_0041 Putative linoleoyl-CoA desaturase 3,074780083 A1S_0059 Putative glycosyltransferase -2,934814162 A1S_0067 L-lactate permease 2,867983694 A1S_0068 L-lactate utilization transcriptional repressor (GntR family) 1,740266254 A1S_0069 L-lactate dehydrogenase FMN linked 2,261836192 A1S_0070 D-lactate dehydrogenase NADH independent, FAD-binding domain 1,839352742 A1S_0073 2-methylisocitrate lyase 1,779592074 A1S_0101 Pseudo 3,071015888 A1S_0103 3-hydroxyisobutyrate dehydrogenase 1,909511059 A1S_0119 Phosphopantethiene-protein transferase -2,615172123 A1S_0121 tRNA -6,677734407 A1S_0170 Putative outer membrane copper receptor (OprC) 2,302305958 A1S_0172 Hypothetical protein A1S_0172 28,76221182 A1S_0179 NADPH-dependent FMN reductase 3,316713266 A1S_0180 Putative membrane protein 4,4043089 A1S_0224 Hypothetical protein A1S_0224 1,993084202 A1S_0233 Type 4 fimbriae expression regulatory protein -3,648849006 A1S_0251 Thiamine hydroxymethylpyrimidine moiety synthesis 2,431468999 A1S_0256 High affinity phosphate uptake transcriptional repressor 1,965328674 A1S_0332 tRNA -2,734469972 A1S_0408 Putative glutathione S-transferase -2,848996933 A1S_0427 Aspartate-semialdehyde dehydrogenase NAD(P)-binding 1,723788434 A1S_0463 Putative alkaline phosphatase 2,37093695 A1S_0464 Sec-independent protein translocase protein (TatC) 2,626200932 A1S_0465 Sec-independent protein translocase protein TatB 2,119553888 A1S_0466 Sec-independent protein translocase protein TatA 3,974749135 A1S_0550 Putative VGR-related protein -3,748574485 A1S_0566 Pyridine nucleotide transhydrogenase (proton pump) alpha subunit (part1) -2,398701486 A1S_0567 Pyridine nucleotide transhydrogenase (proton pump) alpha subunit (part2) -2,378531377 A1S_0595 Putative membrane protein -4,298445076 A1S_0617 Hypothetical protein A1S_0617 -2,501352623 A1S_0644 Hypothetical protein A1S_0644 -2,476797404 A1S_0714 tRNA -8,603938479 A1S_0715 tRNA -3,367487475 A1S_0716 tRNA -2,520040074 68 A1S_0717 tRNA -20,90999886 A1S_0736 Hypothetical protein A1S_0736 6,307013814 A1S_0737 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase 2,406360233 A1S_0738 Putative flavoprotein oxidoreductase 1,708012196 A1S_0746 Ribonucleoside-diphosphate reductase beta subunit 1,725080265 A1S_0747 Ribonucleoside diphosphate reductase alpha subunit 1,876363877 A1S_0800 Bacterioferritin -3,193016034 A1S_0801 Putative transport protein (permease) 1,753161686 A1S_0804 Trehalose-6-phosphate phophatase 2,213668666 A1S_0832 tRNA -2,513627493 A1S_0908 RND family multidrug resistance secretion protein 1,771695997 A1S_0909 Putative MFS family drug transporter 1,797570144 A1S_0922 Putative homocysteine S-methyltransferase family protein 2,001867888 A1S_0923 Malate dehydrogenase FAD/NAD(P)-binding domain / Malate dehydrogenase [quinone] 2,08194988 A1S_1000 Sulfate adenylyltransferase subunit 2 2,496503751 A1S_1001 ATP-sulfurylase subunit 1 2,373560166 A1S_1011 Urease accessory protein UreD -3,038251236 A1S_1012 EsvC / Urea amidohydrolase subunit gamma -3,096672296 A1S_1013 Urea amidohydrolase subunit beta -2,661961516 A1S_1049 Hypothetical protein A1S_1049 -2,741753861 A1S_1085 Amino acid transporter -2,539007574 A1S_1088 Hypothetical protein A1S_1088 -2,860965483 A1S_1089 Hypothetical protein A1S_1089 -4,443066817 A1S_1090 Putative transcription regulator (AsnC family) 2,92148878 A1S_1091 Succinylornithine transaminase (carbon starvation protein C) -3,43656727 A1S_1092 Succinylornithine transaminase (carbon starvation protein C) -5,862639426 A1S_1093 Arginine/ornithine N-succinyltransferase beta subunit -5,786008422 A1S_1094 D-serine/D-alanine/glycine transporter -2,998003217 A1S_1216 LysR regulator -2,393257685 A1S_1223 tRNA -9,61263552 A1S_1224 Transposase -4,374194522 A1S_1228 Cold shock protein -4,65134289 A1S_1229 Pyrroline-5-carboxylate reductase -3,387635633 A1S_1255 Lipid A biosynthesis lauroyl acyltransferase -3,996642143 A1S_1266 Putative membrane protein -4,307066915 A1S_1267 Putative lactam utilization protein -3,526606631 A1S_1268 Hypothetical protein A1S_1268 -3,735897828 A1S_1269 Putative allophanate hydrolase subunit 1 and 2 -3,658727334 A1S_1270 Hypothetical protein A1S_1270 -4,433917777 A1S_1288 Putative VGR-related protein -3,288256628 A1S_1320 Transcriptional regulator SoxR -3,066609102 A1S_1335 Phenylacetic acid degradation protein paaN 3,244622056 A1S_1336 Phenylacetate-CoA oxygenase subunit PaaA 1,925774812 A1S_1337 Phenylacetate-CoA oxygenase subunit PaaB 2,151943853 A1S_1338 Hypothetical protein A1S_1338 2,095589207 69 A1S_1339 Phenylacetate-CoA oxygenase PaaJ subunit 2,158694898 A1S_1340 Phenylacetate-CoA oxygenase/reductase PaaK subunit 1,988164111 A1S_1354 (Acyl-carrier protein) phosphodiesterase -4,371305981 A1S_1355 p-hydroxybenzoate hydroxylase transcriptional activator -3,458178368 A1S_1356 p-hydroxybenzoate hydroxylase transcriptional activator 1,909067312 A1S_1372 Hypothetical protein A1S_1372 -3,685048519 A1S_1373 Putative acyl-CoA carboxylase alpha chain protein -3,020431978 A1S_1374 3-methylglutaconyl-CoA hydratase -2,521101799 A1S_1396 ABC-type amino acid transport system 2,514269767 A1S_1398 GlnQ protein 1,839868743 A1S_1435 Hypothetical protein A1S_1435 -2,395374661 A1S_1436 Putative acyl-CoA dehydrogenase 2,136771227 A1S_1437 Putative acyl-CoA dehydrogenase 2,303810193 A1S_1442 Taurine ABC transporter periplasmic taurine-binding protein 9,469009538 A1S_1443 Taurine ATP-binding transport system component 10,64067184 A1S_1444 ABC taurine transporter permease subunit 5,268788099 A1S_1445 Taurine dioxygenase 3,468959789 A1S_1487 Putative Acyl-CoA dehydrogenase 1,812699687 A1S_1488 Putative Acyl-CoA dehydrogenase 2,132746095 A1S_1498 Putative transcriptional regulator (TetR family) -4,003875641 A1S_1505 yyaM 40,4300082 A1S_1665 Putative membrane protein 3,073765337 A1S_1677 Putative porin precursor 2,57721949 A1S_1708 Beta-lactamase-like protein -3,161623735 A1S_1709 Hypothetical protein A1S_1709 0,254873163 A1S_1710 Putative membrane protein -3,257280351 A1S_1726 Aspartate ammonia-lyase (aspartase) 2,378126552 A1S_1731 Acetoacetyl-CoA transferase beta subunit -2,579954492 A1S_1732 Acetoacetyl-CoA transferase alpha subunit -2,890598255 A1S_1735 Hypothetical protein A1S_1735 -2,791113316 A1S_1760 Hypothetical protein A1S_1760 -4,097559303 A1S_1762 Hypothetical protein A1S_1762 1,742620352 A1S_1811 Hypothetical protein A1S_1811 -2,426624706 A1S_1821 Short-chain dehydrogenase/reductase SDR 1,86515805 A1S_1841 Hypothetical protein A1S_1841 -2,862129625 A1S_1926 Putative membrane protein 1,873695091 A1S_1928 Putative signal peptide 4,142089126 A1S_1943 Putative membrane protein -2,884976522 A1S_1984 D-amino acid dehydrogenase small subunit -2,364748955 A1S_2057 Methyl viologen resistance protein (MFS superfamily) -2,451248433 A1S_2090 Hypothetical protein A1S_2090 -2,551398493 A1S_2091 Putative exported protein -2,365873959 A1S_2093 Hypothetical protein A1S_2093 1,859723965 A1S_2122 Transcriptional regulator 1,875285287 A1S_2202 Aspartate racemase 2,402465914 A1S_2218 CsuA/B -3,557110164 70 A1S_2224 Threonine efflux protein 2,836132945 A1S_2230 Hypothetical protein A1S_2230 -3,089609662 A1S_2238 tRNA -4,361194157 A1S_2239 tRNA -2,617275074 A1S_2289 Putative signal peptide -3,75877858 A1S_2296 Putative protease 2,224288367 A1S_2319 Putative membrane protein 2,908415896 A1S_2320 Transcriptional regulator AraC family 1,763689515 A1S_2334 S-adenosyl-L-homocysteine hydrolase 1,931879168 A1S_2347 Hypothetical protein A1S_2347 -3,280881456 A1S_2446 High-affinity phosphate transport protein 2,779460265 A1S_2447 EsvD 3,211746538 A1S_2448 Putative phosphate transporter 4,415621921 A1S_2458 Putative fatty acid desaturase 5,815884329 A1S_2459 Putative oxidoreductase 4,811424151 A1S_2490 UDP-N-acetyl glucosamine-2-epimerase -3,89628889 A1S_2512 Hypothetical protein A1S_2512 -2,702934785 A1S_2531 Sulfate transport protein 14,23755414 A1S_2532 Sulfate transport protein 22,54708807 A1S_2533 Putative esterase 11,70258487 A1S_2534 Sulfate transport protein 5,367772795 A1S_2535 Putative sulfate permease 4,474063585 A1S_2536 Putative ATPase 2,363384072 A1S_2537 Putative LysR-type transcriptional regulator 1,717536577 A1S_2555 Transposition site target selection protein D -2,684175378 A1S_2648 Hypothetical protein A1S_2648 -2,76460688 A1S_2654 Putative periplasmic binding protein of transport/transglycosylase 1,812087761 A1S_2694 Mur ligase middle region -2,5493552 A1S_2695 Hypothetical protein A1S_2695 -2,397710105 A1S_2699 Putative transcriptional regulator 1,732122956 A1S_2710 Hypothetical protein A1S_2710 1,820533836 A1S_2748 Putative ammonium transporter -2,695259213 A1S_2801 tRNA -2,965003472 A1S_2841 Putative type 4 fimbrial biogenesis protein FimT -2,383966682 A1S_2844 Quaternary ammonium compound-resistance protein -2,373154232 A1S_2846 CysI-like sulfite reductase protein 1,795908099 A1S_2863 Putative antioxidant protein 2,866965402 A1S_2909 tRNA -4,931625423 A1S_2911 Uncharacterized membrane protein LemA family -2,40884189 A1S_3010 Hypothetical protein A1S_3010 1,768451868 A1S_3035 Xanthine phosphoribosyltransferase 2,081840296 A1S_3043 Hypothetical protein A1S_3043 -2,706572314 A1S_3044 Hypothetical protein A1S_3044 -2,922101167 A1S_3085 Putative flavohemoprotein 2,283046794 A1S_3135 Putative APC family S-methylmethionine transporter (MmuP) -2,995657413 71 A1S_3166 Pilin like competence factor -2,418282511 A1S_3183 tRNA -2,516094839 A1S_3212 Hypothetical protein A1S_3212 -2,508738565 A1S_3224 Acyl coenzyme A reductase -2,570691211 A1S_3225 Putative sulfate permease -2,594439793 A1S_3231 Putative acetyl-CoA hydrolase/transferase 2,09959443 A1S_3237 Exonuclease putative -2,392405255 A1S_3251 Transporter LysE family -6,89413316 A1S_3305 NADH-dependent FMN reductase 2,176815627 A1S_3350 Hypothetical protein A1S_3350 -3,744907286 A1S_3363 Membrane metalloendopeptidase protein -3,116653492 A1S_3364 Putative VGR-related protein -7,40947075 A1S_3374 Positive pho regulon response regulator 1,92233217 A1S_3375 Positive pho regulon response regulator 2,089499576 A1S_3401 Hypothetical protein A1S_3401 -2,955676254 A1S_3402 Arginase/agmatinase/formimionoglutamate hydrolase -3,028239302 A1S_3403 Imidazolonepropionase -2,661908071 A1S_3404 Proline transport protein (APC family) -2,627117238 A1S_3413 APC family aromatic amino acid transporter -5,539609242 A1S_3414 Fumarylacetoacetase -42,06992163 A1S_3415 Maleylacetoacetate isomerase -54,71945548 A1S_3416 Glyoxalase/bleomycin resistance protein/dioxygenase -26,46556666 A1S_3417 Regulatory proteins IclR -2,362911036 A1S_3418 4-hydroxyphenylpyruvate dioxygenase -23,0065767 A1S_3436 Putative alcohol dehydrogenase 1,794177485 A1S_r01 16S ribosomal RNA -4,406220125 A1S_r08 23S ribosomal RNA -4,297087141 A1S_r12 16S ribosomal RNA 2,13706962 72 Table 2. qRT-PCR data to validate RNA-seq results. Sample Fold difference A1S_3416 relative to microaerobiosis Fold difference A1S_3414 relative to microaerobiosis Fold difference A1S_1443 relative to normoxia Fold difference A1S_2531 relative to normoxia Normoxia 51.15 (36.89 ± 70.93) 166.96 (113.08 ± 246.51) 1 (0.39 ± 2.57) 1 (0.16 ± 6.38) Microaerobiosis 1 (0.38 ± 2.72) 1 (0.43 ± 2.35) 5.38 (3.06 ± 9.46) 15.45 (2.86 ± 83.42) 73 16. Camarena L, Bruno V, Euskirchen G, Poggio S, Snyder M. Molecular mechanisms of ethanol-induced pathogenesis revealed by RNA-sequencing. PLoS Pathog 2010; 6:e1000834. 17. Wright MS, Jacobs MR, Bonomo RA, Adams MD. Transcriptome remodeling of Acinetobacter baumannii during infection and treatment. MBio 2017; 8. pii: e02193-16. 18. Murray GL, Tsyganov K, Kostoulias XP, Bulach DM, Powell D, Creek DJ, Boyce JD, Paulsen IT, Peleg AY. Global Gene Expression Profile of Acinetobacter baumannii During Bacteremia. J Infect Dis 2017; 215(suppl_1):S52-S57. 19. Rosenberg H, Gerdes RG, Chegwidden K. Two systems for the uptake of phosphate in Escherichia coli. J Bacteriol 1977; 131:505-11. 20. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by rna-seq. Nat Methods 2008; 5:621-8. 21. Robinson MD, Smyth GK. Small-sample estimation of negative binomial dispersion, with applications to sage data. Biostatistics 2008; 9:321-32. 22. Raza K, Mishra A. A Novel Anticlustering Filtering Algorithm for the Prediction of Genes as a Drug Target. American Journal of Biomedical Engineering 2012; 2:206-11. 23. Van Iterson M, Boer JM, Menezes RX. Filtering, FDR and power. BMC Bioinformatics 2010; 11:450. 24. Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012; 16:284-7. 80 25. Luo W, Pant G, Bhavnasi YK, Blanchard SG Jr, Brouwer C. Pathview Web: user friendly pathway visualization and data integration. Nucleic Acids Res 2017; 45:W501- 8. 26. Aranda J, Poza M, Pardo BG, Rumbo S, Rumbo C, Parreira JR, Rodríguez-Velo P, Bou G. A rapid and simple method for constructing stable mutants of Acinetobacter baumannii. BMC Microbiology 2010; 10: 279. 27. Gil-Marqués ML, Pachón-Ibáñez ME, Pachón J, Smani Y. Effect of hypoxia on the pathogenesis of Acinetobacter baumannii and Pseudomonas aeruginosa in vitro and in murine experimental models of infections. Infect Immun 2018; pii: IAI.00543-18. 28. O'Toole GA, Pratt LA, Watnick PI, Newman DK, Weaver VB, Kolter R. Genetic approaches to study of biofilms. Methods Enzymol. 1999; 310:91-109. 29. Carretero-Ledesma M, García-Quintanilla M, Martín-Peña R, Pulido MR, Pachón J, McConnell MJ. Phenotypic changes associated with Colistin resistance due to Lipopolysaccharide loss in Acinetobacter baumannii. Virulence 2018; 9:930-42. 30. Wang Z, Gerstein M and Snyder M. RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet 2009; 10:57-63. 31. Hsieh YJ, Wanner BL. Global regulation by the seven-component Pi signaling system. Curr Opin Microbiol 2010; 13:198-203. 32. Chekabab SM, Harel J, Dozois CM. Interplay between genetic regulation of phosphate homeostasis and bacterial virulence. Virulence 2014; 5:786-93. 33. Fang FC, Frawley ER, Tapscott T, Vázquez-Torres A. Bacterial stress responses during host infection. Cell Host Microbe 2016; 20:133-43. 81 34. Lamarche MG, Wanner BL, Crépin S, Harel J. The phosphate regulon and bacterial virulence: a regulatory network connecting phosphate homeostasis and pathogenesis. FEMS Microbiol Rev 2008; 32:461-73. 35. Camarena L, Bruno V, Euskirchen G, Poggio S, Snyder M. Molecular mechanisms of ethanol-induced pathogenesis revealed by RNA-sequencing. PLoS Pathog 2010; 6:e1000834. 36. Neznansky A, Blus-Kadosh I, Yerushalmi G, Banin E, Opatowsky Y. The Pseudomonas aeruginosa phosphate transport protein PstS plays a phosphateindependent role in biofilm formation. FASEB J 2014; 28:5223-33. 37. Pratt JT, Ismail AM, Camilli A. PhoB regulates both environmental and virulence gene expression in Vibrio cholerae. Mol Microbiol 2010; 77:1595-605. 38. Ferreira GM, Spira B. The pst operon of enteropathogenic Escherichia coli enhances bacterial adherence to epithelial cells. Microbiology. 2008; 154:2025-36. 39. Esparza M, Palomares B, García T, Espinosa P, Zenteno E, Mancilla R. PstS-1, the 38-kDa Mycobacterium tuberculosis glycoprotein, is an adhesin, which binds the macrophage mannose receptor and promotes phagocytosis. Scand J Immunol 2015; 81:46-55. 40. Lucas RL, Lostroh CP, DiRusso CC, Spector MP, Wanner BL, Lee CA. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J Bacteriol 2000; 182:1872-82. 41. Davey ME, O’Toole GA. Microbial biofilms: from ecology to molecular genetics. Microbiol Mol Biol Rev 2000; 64:847-67. 82 42. Pratt JT, McDonough E, Camilli A. PhoB regulates motility, biofilms, and cyclic di- GMP in Vibrio cholerae. J Bacteriol 2009; 191:6632-42. 43. Monds RD, Newell PD, Gross RH, O’Toole GA. Phosphate-dependent modulation of c-di-GMP levels regulates Pseudomonas fluorescens Pf0-1 biofilm formation by controlling secretion of the adhesin LapA. Mol Microbiol. 2007; 63:656-79. 44. O’May G, Jacobsen SM, Longwell M, Stoodley P, Mobley HL, Shirtliff ME. The high-affinity phosphate transporter Pst in Proteus mirabilis HI4320 and its importance in biofilm formation. Microbiology 2009; 155:1523-35. 45. Monds RD, Silby MW, Mahanty KH. Expression of the Pho regulon negatively regulates biofilm formation by Pseudomonas aureofaciens PA147-2. Mol Microbiol 2001; 42:415-26. 46. Blus-Kadosh I, Zilka A, Yerushalmi G, Banin E. The effect of pstS and phoB on quorum sensing and swarming motility in Pseudomonas aeruginosa. PLoS One 2013; 8:e74444. 47. Zaborina O, Holbrook C, Chen Y, Long J, Zaborin A, Morozova I, Fernandez H, Wang Y, Turner JR, Alverdy JC. Structure-function aspects of PstS in multi-drug- resistant Pseudomonas aeruginosa. PLoS Pathog 2008; 4:e43. 48. Llop Talaverón JM, Comas Sugrañes D, Badía Tahull MB, Sáez Fernández A, Jódar Masanés R, Gómez Sáez JM. Hypophosphatemia in parenteral nutrition: Prevention and associated risks factors]. Nutr Hosp 2004; 19:362-6. 49. Barak V, Schwartz A, Kalickman I, Nisman B, Gurman G, Shoenfeld Y. Prevalence of hypophosphatemia in sepsis and infection: the role of cytokines. Am J Med 1998; 104:40-7. 83 50. Shor R, Halabe A, Rishver S, Tilis Y, Matas Z, Fux A, Boaz M, Weinstein J. Severe hypophosphatemia in sepsis as a mortality predictor. Ann Clin Lab Sci 2006; 36:67-72. 84 ADDITIONAL DATA We selected five different genes that were overexpressed under microaerobiosis to determine if they were involved in A. baumannii pathogenesis. The genes were A1S_0030, A1S_2532, A1S_0464, A1S_0172 and A1S_2448. We have analyzed the last one in this chapter I. Stable, in-frame deletion mutants were constructed in the A. baumannii ATCC 17978 strain by homologous recombination using the described protocol. For construction of the A1S_0030 and A1S_2532 deletion mutant (ATCC 17978 ΔA1S_0030::Kan, and ATCC 17978 ΔA1S_2532::Kan, respectively), the 500 bp immediately upstream of the genes open reading frame, and the 500 bp immediately downstream were amplified using the primers indicated in Table A1. A more efficient technique was used to construct in-frame deletion mutants of the A1S_0464 and A1S_0172 genes. Gene deletions were performed with an allelic exchange plasmid called pMJG42, which harbors the sacB gene for counterselection via growth on medium containing sucrose as previously described [1]. For construction of the deletion mutants ATCC 17978 ΔA1S_0464 and ATCC 17978 ΔA1S_0172 the 2000 bp immediately upstream of the genes open reading frame, and the 2000 bp immediately downstream were amplified using the primers indicated in Table A1. The plasmid pMJG42 was digested with SpeI and NotI (New England Biolabs). The Up insert was digested with SpeI and XhoI/ BamHI, and the Down insert was digested with XhoI/ BamHI and NotI (XhoI or BamHI depending on the mutant). The plasmid and both inserts were ligated, and the resulting construction was transformed into the Escherichia coli DH5α λpir by electroporation before selection on LB agar plates with Tetracycline 5 mg/L (Sigma). The construction was then transformed into E. coli MFD 85 [2], a Diaminopimelic acid (DAP) auxotrophic strain. MFD donor strain harboring the respective pMJG42-gene(Up/Down) construct and the ATCC 17978 recipient strain were cultured overnight at 37°C in LB (supplemented with tetracycline 5 mg/L and DAP 200 mg/L for the donor strain). Aliquots of 1 mL of the donor strain and 0.5 mL of the recipient strain were washed twice and mix together. The cells were suspended in 30 µl of LB and added on to a sterile 0.45 µm cellulose nitrate filter paper (Millipore) on LB agar plates and incubated for 4 h at 37°C. The cells were washed off from the filter by adding 1 ml of LB and they were plated onto LB agar plates containing tetracycline 5 mg/L. The obtained colonies were grown in LB, plated onto 10% Sucrose plates and incubated overnight at room temperature. All deletion mutants were confirmed by PCR and sequencing. In order to complement the obtained mutants, the genes reading frames and 200-400 bp upstream and downstream were amplified using the primers indicated in Table A1. These fragments were cloned into the pUCp24 and introduced into the mutant strains by electroporation before selection on LB agar plates containing 10 mg/L of gentamicin to create the complemented strains (ATCC 17978 ∆A1S_0030::Kan/pUCp24-0030, ATCC 17978 ∆A1S_2532::Kan/pUCp24-2532, ATCC 17978 ∆A1S_0464/pUCp24-0464, ATCC 17978 ∆A1S_0172/pUCp24-0172). Deletion mutants were also transformed with the empty pUCp24 plasmid for use as controls (ATCC 17978 ∆A1S_0030::Kan/pUCp24, ATCC 17978 ∆A1S_2532::Kan/pUCp24, ATCC 17978 ∆A1S_0464/pUCp24, ATCC 17978 ∆A1S_0172/pUCp24). Growth curves analysis Growth curves in MHB under normoxia (21% O2), hypoxia (1% O2), and microaerobiosis (0.1-0.3% O2) were performed to find out if ATCC 17978 wild-type strain and the mutant strains had a different growth rate (Fig. A1). We showed that there 86 were no significant differences between the wild-type and the mutant strains growth. Complemented strain and strains harboring the empty pUCp24 had the same growth than the other strains under all the conditions (data not shown). Bactericidal activity, bacterial adherence and bacterial invasion in cell cultures We determined if hypoxia affects the bactericidal activity of epithelial cells against the mutant strains more than against the wild-type strain. Bacterial counts of ATCC 17978, ΔA1S_0030::Kan, ΔA1S_2532::Kan, ΔA1S_0464, and ΔA1S_0172 strains found in the extracellular medium of A549 cell line under hypoxia (1% O2) showed a decrease of bacterial concentrations after 2 and 24 h compared to normoxia (data not shown). No significant reductions were found between the mutant strains and the wild-type (Fig. A2). Moreover, bacterial concentration was higher in the case of the mutants ΔA1S_0464 and ΔA1S_0172 at 24 h post-infection. The bacterial adherence of all the mutant strains to A549 cell line was significantly lower at 2 and 24 h post-infection under hypoxia compared to normoxia (data not shown). This decrease was higher in the case of the mutants ΔA1S_0030::Kan, ΔA1S_2532::Kan, ΔA1S_0464, and ΔA1S_0172 at 2 h and ΔA1S_0030::Kan, ΔA1S_2532::Kan, and ΔA1S_0464 at 24 h post-infection compared to the wild-type (Fig. A3). Bacterial counts of the mutant strains ΔA1S_0030::Kan and ΔA1S_2532::Kan, inside epithelial cells showed a higher decrease than the wild-type strain at 24 h post-infection under hypoxia compared to normoxia (Fig. A4). These data indicate that hypoxia affects the adherence and invasion of these mutant strains more than the wild-type. No significant differences were found in the case of the mutants ΔA1S_0464 and ΔA1S_0172. According to these results we chose the mutant strain A1S_2448 to continue with the analysis. 87 Figure A1. A) Growth curves of A. baumannii ATCC 17978 and ΔA1S_0030::Kan in MHB under normoxia, hypoxia and microaerobiosis (21%, 1% and 0.1-0.3% O2, respectively). B) Growth curves of A. baumannii ATCC 17978 and ΔA1S_2532::Kan in MHB under normoxia, hypoxia and microaerobiosis (21, 1 and 0.1-0.3% O2, respectively). C) Growth curves of A. baumannii ATCC 17978 and ΔA1S_0464 in MHB under normoxia and hypoxia (21% and 1% O2, respectively). D) Growth curves of A. baumannii ATCC 17978 and ΔA1S_0172 in MHB under normoxia and hypoxia (21% and 1% O2, respectively). 88 Figure A2. A-D) Measurement of bacterial concentration (%) in the extracellular medium after 2 and 24 h of A549 infection by A. baumannii ATCC 17978, ATCC 17978/pUCp24, ΔA1S_0030::Kan, ΔA1S_2532::Kan, ΔA1S_0464, ΔA1S_0172, and the respective complemented strains and the mutant strains harboring the empty pUCp24 under normoxia and hypoxia (1% O2). Bars represent the reduction of bacterial concentration under hypoxia compared to normoxia, being the reduction obtained in the wild-type equal to 100%. ***: P<0.001 Mutant vs. wild-type at 2 or 24 h; **: P<0.01 Mutant vs. wild-type at 2 or 24 h; *: P<0.05 Mutant vs. wild-type at 2 or 24 h. 89 normoxia [11]. Consecutively, the HIF-1α deletion, but not HIF-1α isoform I.1, in T lymphocytes prevents the antibacterial effect of these cells [15, 16]. During infection, bacteria must adapt to heterogeneous environments [17-19]. The oxygen levels in the foci of infection are much lower (<1%) than in healthy tissues (2.5- 9%) [20] due to a combination of increased oxygen consumption by immune cells and pathogens, along with a decreased perfusion due to vascular dysfunction [21-23]. Therefore, the microenvironment at the area of infection plays a crucial role in determining the outcome of an infection. Hypoxia not only modifies the host cells but also the bacterial metabolism and virulence [5]. In P. aeruginosa and Mycobacterium tuberculosis the expression of virulence factors such as alkaline protease, siderophores and exotoxin A are reduced by hypoxia [24, 25]. However, hypoxia can also increase the production of alginate and the expression of the PA-I lectin/adhesin by P. aeruginosa causing a disruption in intestinal barrier and allowing exotoxin A to cross the epithelium [26, 27]. Exposure to hypoxia also induces antibiotic resistance in P. aeruginosa by an alteration of efflux pumps expression [28]. Together, these studies demonstrate the complexity of HIF-pathogen interactions. The aim of this study was to evaluate the effect of hypoxia on A. baumannii and P. aeruginosa pathogenesis, in vitro, regarding to bactericidal activity and adherence/invasion, and in murine models of infection, regarding to survival, and bacterial load; and the innate immune response in vitro and in vivo. 96 Materials and Methods Bacterial strains and growth condition The wild-type strains A. baumannii ATCC 17978 and P. aeruginosa PAO1 were used. They were cultured at 37ºC overnight (160 rpm) in Mueller Hinton Broth (MHB) (Sigma, Spain). Cultured strains were washed with phosphate-buffered saline (PBS) and suspended in Dulbecco’s modified Eagle’s medium (DMEM) before their use in eukaryotic cell culture experiments (human lungs epithelial cell line A549 and murine macrophage cell line RAW 264.7). Growth curves analysis The growth of A. baumannii ATCC 17978 and P. aeruginosa PAO1 strains under hypoxia (1% and 10% O2) and normoxia (21% O2) in static were monitored during 24 h. Both strains were grown overnight in 20 ml of MHB, and a 1:10000 dilution was performed to obtain, approximately, 105 cfu/ml in a 40 ml culture of MHB (10% and 21% O2 ) or DMEM (1% and 21% O2). Three replicates were performed in different days. A549 and RAW 264.7 culture and infection Human lungs epithelial cell line A549 and murine macrophage cell line RAW 264.7 were grown in DMEM containing 10% Fetal Bovine Serum (Gibco, Spain), 1% HEPES 1M, vancomycin (50 mg/ml), gentamicin (20 mg/ml) and amphotericin B (0.25 mg/ml; Gibco), as previously described [43]. In the case of hypoxia condition studies, cells were transferred to a hypoxia chamber (Coy Laboratories, USA) with a humidified atmosphere of 1% O2, 5% CO2 and the balance N2 at 37°C. Cells were seeded (105 cells/well in a 24-well plate) for 30 h in 24-well plates before infection with A. 97 baumannii ATCC 17978 or P. aeruginosa PAO1 at a multiplicity of infection (MOI) of 500. To mimic hypoxia condition we treated the cells with 0.1 mM Dimethyloxaloylglycine (DMOG) (Sigma, Spain), an inhibitor of prolyl hydroxylases [44], 6 h prior bacterial infection and during infection. Immediately before the infection, A549 cells were washed thrice with PBS and incubated in supplemented DMEM. HIF-1α measurement in cell cultures A549 and RAW 264.7 cells were seeded for 24 h in 6-well plates (106 cells/well). After 6 and 24 h in hypoxia (1% O2) or normoxia condition, cells were washed thrice with PBS, harvested using cell scraper and homogenized in RIPA buffer supplemented with 1 mM phenylmethylsulfonyl fluoride and 10% cocktail of protease inhibitors (Sigma, Spain), and centrifuged at 13000g 4°C for 20 min. The supernatant was removed and the amount of proteins was determined using BCA assay (Promega, Spain). The samples were stored at -80°C. Forty µg of proteins of each sample was used to measure HIF-1α levels with an enzyme-linked immunosorbent assay (ELISA) kit (Thermo Fisher Scientific, Spain). Bactericidal activity, bacterial adherence and bacterial invasion in cell cultures After A549 and RAW 264.7 cells infections with A. baumannii ATCC 17978 and P. aeruginosa PAO1 strains under hypoxia and normoxia conditions, extracellular medium was removed and serially diluted to determine bacterial concentration as previously described [45]. Adherence and invasion assays were performed as previously described [45]. To measure the number of adherent bacteria, cells were infected as mentioned before, and, after washing thrice with PBS, 200 µl of trypsin-EDTA (Gibco, Spain) was added for 5 min at 37°C. Then, 200 µl of 0.5% Triton X-100 (Sigma, Spain) was added for 3 min. 98 The invasion protocol included a treatment with gentamicin 256 µg/ml (Gibco, Spain) before the addition of trypsin-EDTA. Diluted lysates were counted to determine the attached and internalized bacteria by A549 and RAW 264.7 cells. Every assay was performed three times in different days. In the case of invasion assay, four replicates were performed in different days. Cytokine Assay Extracellular medium of infected A549 and RAW 264.7 cells with A. baumannii ATCC 17978 and P. aeruginosa PAO1 strains under hypoxia and normoxia conditions were collected and centrifuged at 5000g for 15 min at 4ºC. The supernatant was stored at - 80ºC until analysis. TNF-α, IL-6 and IL-10 levels were measured in using an ELISA kit (Affymetrix eBioscience, USA), in accordance with the manufacturer’s instructions. Levels of pro- and anti-inflammatory cytokines (IL-6, IL-10 and TNF-α) in mice serum were measured by ELISA assays (Affymetrix eBioscience, USA). iTRAQ assay We analyzed the differential protein expression profile between normoxia and hypoxia conditions in A549 cell infected by A. baumannii ATCC 17978. After 2 h infection, we collected the cells in a lysis buffer composed by 1 M Triethylammonium bicarbonate buffer (Sigma, Spain), 0.05% SDS, 1:100 phosphatase inhibitor cocktail (PhosSTOP EASYpack, Roche, Spain), 1:100 protease inhibitor cocktail (Complete Mini EDTA- free, Roche, Spain), and 0.002% benzonase (Novagen, USA). Pellet was separated from the supernatant and protein concentration was quantified by fluorimetry (Qubit life technology, USA). Samples were treated with 50 mM TCEP (AB Sciex, Spain) to reduce disulfide bonds and 200 mM MMTS (AB Sciex, Spain), and then they were digested with trypsin (Promega, Spain) at a 10:1 substrate:enzyme ratio at 37° 99 overnight. We used an isobaric tag iTRAQ 4 plex (reporters at 114 to 117, AB Sciex, Spain). Samples were analyzed by nano-liquid chromatography (nano LC 100, Thermo Fisher Scientific, USA) and tandem mass spectrometry (Q Exactive Plus Orbitrap, Thermo Electron, USA). Protein identification was performed using Proteome Discoverer 1.4 (Thermo Fisher Scientific, USA). MS/MS fragmentation patterns were mapped against Uniprot database. We considered quantifiable proteins those that were identified through more than 2 peptides with a confidence level ≥ 95%, a P-value ≤ 0.05, and an error factor < 2 with every reference tag. Animals Immunocompetent C57BL/6 male mice, weighing approximately 20 g (Production and Experimentation Animal Center, University of Seville, Seville, Spain) were used; they had a sanitary status of murine pathogen free and were assessed for genetic authenticity. Mice were housed in an individually ventilated cage system under specific pathogenfree conditions, and water and food supplied ad libitum. This study was carried out following the recommendations in the Guide for the Care and Use of Laboratory Animals [46]. This study was carried out in strict accordance with Directive 2010/63/EU on the protection of animals used for scientific purposes. Experiments were approved by the Committee on the Ethics of Animal Experiments of the University Hospital of Virgen del Rocío of Seville, Spain (20-05-14-84). All procedures were performed under sodium thiopental (B. Braun Medical S.A., Spain) anesthesia, and all efforts were made to minimize suffering. Experimental models Both models of infection were carried out under the following conditions: i) hypoxia (10% O2), ii) normoxia and iii) six hours under hypoxia followed by normoxia. The 100 minimum lethal doses (MLD) were calculated for A. baumannii and P. aeruginosa under hypoxia and normoxia conditions. Briefly, groups of 6 mice were inoculated intraperitoneally (ip.) for A. baumannii and intratracheally for P. aeruginosa with increasing concentrations until reaching 100% mortality, of each pathogen and the survival rate was monitored for 7 days. For the hypoxia condition studies, mice were maintained in a hypoxic chamber (Coy Laboratories, USA) with a humidified atmosphere of 10% O2 (standard hypoxic condition) 6 h prior the infection and until the animal death or the end of the experiment. In the experiments in which mice were 6 h under hypoxia followed by normoxia, the animals were maintained in a hypoxic chamber during 6 h prior the infection, and placed outside normoxia until the end of the experiment or the animal death. The same conditions were used with control mice (not infected). To evaluate pneumonia, after 4 h of infection and at the time of death, lungs were aseptically extracted, fixed in 10% formalin and embedded in paraffin wax. Serial sections (3 μm) were cut onto glass slides and stained with hematoxylin and eosin. A N of no more than 5 mice per condition was performed in different weeks to reproduce the experimental models results. (i) Experimental murine model of peritoneal sepsis. A previously characterized murine peritoneal sepsis model by A. baumannii was used [36]. Briefly, animals were inoculated i.p. with 0.5 ml of MLD100, mixed 1:1 with a saline solution containing 10% (wt/vol) mucin from porcine stomach Type II (Sigma, Spain). After 4 h of infection, a group of 34 mice (17 under hypoxia and 17 under normoxia) were sacrificed by i.p. injection of sodium thiopental (200 µl; Braun Medical, USA) and analyzed, and 48 mice (21 under normoxia, 22 under hypoxia and 5 under 6h hypoxia + normoxia) were analyzed at the time of death. Survival rates were recorded under 101 hypoxia and normoxia conditions. Bacteremia was evaluated, both qualitatively and quantitatively after the animal’s death. For qualitative analysis, the blood was inoculated into sterile tubes with 1 ml of MHB and incubated for 24 h at 37ºC, and then 10 µl was plated on sheep blood agar. To evaluate quantitatively the bacteremia (log10 cfu/ml), blood was serially diluted and plated on sheep blood agar. Finally, bacterial load was quantified in spleen and lungs. Briefly, organs were aseptically removed and homogenized (Stomacher 80; Tekmar Co.) in 2 ml of sterile 0.9% NaCl solution. Serial dilutions of the homogenized organs were plated on sheep blood agar for quantitative cultures (log10 cfu/g). Finally, bacterial concentration in peritoneal fluid was also determined by injecting 2 mL of sterile 0.9% NaCl solution i.p. and, after a brief massage on the abdomen, peritoneal lavage was collected and plated on sheep blood agar (log10 cfu/mL). HIF-1α levels in mice serum were measured by ELISA assays (MyBioSource, USA). (ii) Pneumonia model. A previously characterized pneumonia model by P. aeruginosa [47] was used as follows: anesthetized mice (thiopental at 5% [wt/vol], i.p.) were infected by intratracheal instillation, using 50µL of the MLD100 calculated previously. Mice remained in a vertical position for 3 min and then resting at 30º positions until they awakened. After 4 h of infection, 36 mice (18 under normoxia and 18 under hypoxia) were sacrificed (sodium thiopental, Braun Medical, USA) to be analyzed and 46 mice (20 under normoxia, 18 under hypoxia and 8 under 6 h hypoxia + normoxia) were analyzed at the time of death. Survival rates were analyzed for the different conditions. Bacteremia, bacterial load in blood and tissue (spleen and lungs) were performed as described above. HIF-1α levels in mice serum were measured by ELISA assays (MyBioSource, USA). 102 Statistical analysis Statistical analyses were performed using the IBM SPSS Statistics 22 software program. Tests used included ANOVA (bacterial counts in tissues and fluids and mortality time), Chi-square test (bacteremia), and when required Dunnett’s and Tukey post-hoc tests and Student's t-test (bacterial counts in vitro, cytokines and HIF-1α levels). A P-value <0.05 was considered significant. Results Hypoxia increases HIF-1α levels in epithelial and macrophages cells HIF-1α levels in cell lines after 6 and 24 h under hypoxia (1% O2) and normoxy (21% O2) were measured. In epithelial cells, HIF-1α levels were 2.69 times higher after 6 h in hypoxia than in normoxy (2296.98 ± 157.74 pg/mL vs. 853.63 ± 95.47 pg/mL, P<0.001) and were higher than after 24 h (1107.70 ± 96.08 pg/mL vs. 592.27 ± 48.86 pg/mL, P<0.01). In macrophages cells, HIF-1α levels were 1.50 times higher after 6 h in hypoxia than in normoxia (331.64 ± 52.93 pg/mL vs. 220.67 ± 11.87 pg/mL) and were higher than after 24 h under hypoxia (223.59 ± 7.05 pg/mL vs. 235.27 ± 9.31 pg/mL; hypoxia 6 h vs. hypoxia 24 h). No significant differences in HIF-1α levels were observed in normoxia between the different times points analysed. The marked increase of HIF-1α levels after 6 h under hypoxia (1% O2) defined the time of hypoxia condition prior the infection for the in vitro and in vivo experiments. Hypoxia increases bactericidal activity of epithelial and macrophages cells against A. baumannii and P. aeruginosa 103 First, we observed that ATCC 17978 and PAO1 strains growth during 2 and 24 h was indistinguishable between hypoxia (1% O2) and normoxia (Fig. 1A). Next, we determined if hypoxia affects the bactericidal activity of epithelial and macrophages cells. Bacterial counts of ATCC 17978 and PAO1 strains found in the extracellular medium of both cell lines under hypoxia (1% O2) showed a decrease of bacterial concentrations after 2 and 24 h compared to normoxia, (Fig. 1B and 1C). These data support an increase in the bactericidal activity of these cell lines under hypoxia. HIF-1α overexpression increases bactericidal activity of epithelial and macrophages cells against A. baumannii and P. aeruginosa Bacterial counts of ATCC 17978 and PAO1 strains found in the extracellular medium of both cell lines under a 0.1 mM DMOG treatment showed a decrease of bacterial concentrations after 24 h compared to normoxia, (Fig. 1B and 1C). These data support an increase in the bactericidal activity of these cell lines when HIF-1α is overexpressed due to the treatment with DMOG. Hypoxia decreases bacterial adherence and invasion to epithelial and macrophages cells The bacterial adherence of ATCC 17978 and PAO1 strains to both cell lines was significantly lower under hypoxia (1% O2), except in the case of 2 h post-infection by ATCC 17978 strain in the RAW 264.7 cells in which it presented higher bacterial adherence (182.67 ± 11% vs. 100% ± 0%, P<0.001) (Fig. 2A and 2B). Bacterial counts of ATCC 17978 strain inside epithelial and macrophages cells under hypoxia (1% O2) showed an increase of bacterial concentrations 2 h post-infection (150 ± 0% for epithelial cells, and 146.73 ± 5.01% for macrophages cells, P<0.001), and a decrease at 24 h post-infection compared to normoxia (48.55 ± 34.80% for epithelial 104 cells, P<0.001 and 8.69 ± 6.85% for macrophages cells, P<0.001) (Fig. 2C). On the other hand, PAO1 strain counts inside both cell lines under hypoxia (1% O2) showed a decrease of bacterial concentrations after 2 h (P<0.001 for macrophages cells) and 24 h (P<0.001) compared to normoxia (Fig. 2D). These data indicated that hypoxia affects the adherence and invasion of A. baumannii and P. aeruginosa 24 h after bacterial infection. Hypoxia reduces the expression of proteins involved in cell adherence iTRAQ results show that there are 51 down-expressed proteins under hypoxia (Fold Change < 0.6) present in the extracellular medium of a 2 h infection of A549 cells by A. baumannii ATCC 17978 strain (Table S1). Forty-five % are localized in the cytoplasm, 16% are secreted, 19% are in the inner membrane, 10% in the outer membrane, 6% in the periplasm and 4% in the mitochondrion (Fig. S1). The proteins localized in the outer membrane and could be involved in cell adhesion are OmpW, putative ferric siderophore receptor protein A1S_3339, putative ferric siderophore receptor protein A1S_0474, ferric enterobactin receptor A1S_0981, and ferrichrome-iron receptor A1S_1921. Moreover, the secreted uncharacterized protein A1S_3900, which is a protein that presents SH3-like domains, could also be involved in cell adhesion. Hypoxia reduces bacterial load in tissues and fluids in a peritonitis sepsis model by A. baumannii The MLD needed to achieve 100% mortality for ATCC 17978 strain was lower in hypoxia (10% O2) than in normoxia (2.08 vs. 3.20 log10 cfu/mL). For the rest of experiments, we used the MLD calculated in normoxia. The survival time was higher in mice infected under normoxia than hypoxia (10%O2) (36.42 vs. 23.92 h, P<0.001) (Fig. 3A). 105 Table 1. Bacterial load in fluids and tissues in the sepsis model by A. baumannii ATCC 17978 strain. Bacterial load (log 10 CFU per g of tissue or per ml of fluid) in: Time and conditiona Spleen Lungs PFb Blood Bacteremia (%) 4 h N 3.98 ± 0.30 4.07 ± 0.53 4.06 ± 1.29 3.19 ± 0.42 100 H 3.88 ± 0.23 4.07 ± 0.70 3.72 ± 1.15 3.18 ± 0.28 100 Time of death N 8.79 ± 0.56 9.36 ± 0.35 9.31 ± 0.33 8.40 ± 0.56 100 H 8.32 ± 0.71 8.25 ± 0.54e 8.88 ± 0.53c 7.73 ± 0.20d 100 H (6 h) + N 7.90 ± 0.30f 8.32 ± 0.46f 8.75 ± 0.33f 7.85 ± 0.32 100 aN, normoxia; H, hypoxia. bPF, peritoneal fluid. cP ≤ 0.05 (H versus N at the time of death). dP ≤ 0.01 (H versus N at the time of death). eP ≤ 0.001 (H versus N at the time of death). fP ≤ 0.05 [H (6 h) + N versus N at the time of death]. 112 Table 2. Bacterial loads in fluids and tissues in the pneumonia model with P. aeruginosa strain PAO1 Bacterial load (log 10 CFU per g of tissue or per ml of fluid) in: Time and conditiona Spleen Lungs Blood Bacteremia (%) 4 h N 2.64 ± 0.69 7.77 ± 0.61 0.26 ± 0.36 44.44 H 3.10 ± 0.80 7.79 ± 0.42 0.99 ± 0.84 61.11 Time of death N 6.96 ± 0.57 9.81 ± 0.45 7.90 ± 0.67 100 H 5.27 ± 0.60c 9.04 ± 0.58c 5.66 ± 0.78b 100 H (6 h) + N 6.60 ± 0.34 9.79 ± 0.27 6.30 ± 0.46d 100 aN, normoxia; H, hypoxia. bP ≤ 0.01 (H versus N at the time of death). cP ≤ 0.001 (H versus N at the time of death). dP ≤ 0.05 [H (6 h) + N versus N at the time of death]. 113 Discussion To our knowledge, this is the first study that analyses in vitro and in vivo the effect of hypoxia during infection by A. baumannii and P. aeruginosa. We observed that hypoxia in vitro increases bactericidal activity of host cells, and reduce bacterial adherence and invasion. We also found that hypoxia in vivo diminish bacterial load in fluids and tissues, but mice survival time was shorter under hypoxia. We showed that hypoxia doesn’t affect the in vitro growth of A. baumannii and P. aeruginosa. However, it increases the bactericidal activity in epithelial and macrophage cells. The study of Peyssonnaux et al. showed that hypoxia modifies gene regulation in host cells and it increases the LL-37 cathelicidin levels, an antimicrobial peptide involved in the clearance of pathogens [12]. Moreover, we see that hypoxia decreases the bacterial adherence to host cells. This effect might be due to the modification of cell or bacterial membrane under this condition. iTRAQ results confirmed that hypoxia downregulates 51 proteins in A. baumannii ATCC 17978, five of them are localized in the outer membrane which could be involved in cell adherence due to the previous reports of their involvement in the bacterial adherence [29-33]. Regarding to bacterial invasion, we observed differences in behavior between A. baumannii and P. aeruginosa under hypoxia. Our data showed a reduction in the P. aeruginosa internalization into epithelial and macrophages cells, confirming the results obtained in a previous study in which it is demonstrated that hypoxia decreases the P. aeruginosa internalization into A549 cells [34]. However, A. baumannii internalization in both host cells is higher after 2 h under hypoxia, but it is reduced after 24 h. Consequently, hypoxia cannot stop the A. baumannii invasion during the first few hours of infection but it is finally hindered after 24 h. Therefore, we believe that hypoxia 114 confers higher resistance against bacterial invasion to host cells in order to avoid an intracellular replication and the infection evolution. In the in vivo experiments, we observe that a lower bacterial inoculum is needed to cause 100% of mice mortality under hypoxia in the peritoneal sepsis model by A. baumannii. All infected mice presented bacteremia 4 h post-infection for the studied conditions. Moreover, we observe lower bacterial load in blood, PF lungs and spleen under hypoxia. We also show that maintaining animals 6 h under hypoxia before the infection is enough to reduce bacterial load at the time of death. These results are in accordance with a previous study in which the use of the compound AKB-4924, that increases HIF-1α levels, reduced bacterial loads recovered in a S. aureus skin infection model [35]. Moreover, these results are in accordance with the in vitro adherence assays data. The increase of host cells bactericidal activity under hypoxia as well as a reduction of bacterial adherence could allow the immune system to eliminate the infection better. In the pneumonia model of infection by P. aeruginosa, we observed no differences in the inoculum needed to cause 100% of mortality between the studied conditions. Bacteremia observed in mice 4 h post-infection was 44.44% and 61.11% for normoxia and hypoxia, respectively. The difference found in the bacteremia levels between both animal models is because the severity of the sepsis model [36, 37]. As in the sepsis model by A. baumannii, we observed lower bacterial load in fluids and tissues under hypoxia, and under hypoxia followed by normoxia, are in accordance with the in vitro results of adherence and invasion. Again, as in the sepsis animal model, survival time was longer under normoxia. In both animal models, HIF-1α levels were higher after 4 h under hypoxia, being the levels similar at the animal time of death regardless the studied conditions. These results are in accordance with the in vitro studies in which HIF-1α levels increased over time 115 under hypoxia and then decreased 24 h after. We observed that A. baumannii causes an increase of HIF-1α levels compared to the control as what reported in another study in which infection with A. baumannii produced HIF-1α levels increase [4]. In contrast, P. aeruginosa produces HIF-1α levels reduction under hypoxia compared to the control. This result could be explained because 2-alkyl-4-quinolone and Pseudomonas Quinolone Signal triggers the HIF-1α degradation through the 26S-proteasome proteolytic pathway, blocking the HIF-1α effect [38, 39]. As it is well defined in the literature, hypoxia regulates the immune response [20]. In the A. baumannii sepsis model, we observed under hypoxia high IL-10 levels after 4 h infection. Meng et al. indicated that HIF-1α is involved in IL-10 production by B cells [40], and IL-10 is an anti-inflammatory cytokine that suppresses macrophage and dendritic cells function [41]. In the P. aeruginosa pneumonia model, we detected high IL-6 levels 4 h post-infection under hypoxia. It has been showed that HIF-1α increase TNF-α and IL-6 levels [20, 42]. Moreover, IL-10 levels decreased under hypoxia in the pneumonia model by P. aeruginosa at the time of death. However, we did not find differences in cytokines levels between hypoxia and normoxy neither in vitro nor in vivo. Therefore, we found that hypoxia has not a strong impact on cytokine production [20], being more important on the bactericidal activity of host cells and on the reduction of infection in animals. However, this study has some limitations. HIF-1α is a factor involved in multiple cellular pathways and its expression is also regulated by different proteins. Therefore, finding a clear correlation between hypoxia, HIF-1α expression, inflammatory responses and infection is complex, and multiple cellular processes have to be taken into consideration. 116 In conclusion, hypoxia increases the bactericidal activity of host cells. In contrast, mortality in animals under hypoxia is faster even with a lower bacterial load in tissues and fluids. Moreover, we find that hypoxia has not a strong impact on cytokine production by both pathogens. Finally, despite both studied microorganism are close phylogenetically, they present slightly different behavior under hypoxia. 117 References 1. Cramer T, Yamanishi Y, Clausen BE, Förster I, Pawlinski R, Mackman N, Haase VH, Jaenisch R, Corr M, Nizet V, Firestein GS, Gerber HP, Ferrara N, Johnson RS. 2003. HIF-1α is essential for myeloid cell mediated inflammation. Cell 112:645- 657. 2. Peyssonnaux C, Cejudo-Martin P, Doedens A, Zinkernagel AS, Johnson RS, Nizet V. 2007. Cutting edge: essential role of hypoxia inducible factor-1α in development of lipopolysaccharide-induced sepsis. J Immunol 178:7516-7519. 3. Bayele HK, Peyssonnaux C, Giatromanolaki A, Arrais-Silva WW, Mohamed HS, Collins H, Giorgio S, Koukourakis M, Johnson RS, Blackwell JM, Nizet V, Srai SK. 2007. HIF-1 regulates heritable variation and allele expression phenotypes of the macrophage immune response gene SLC11A1 from a Z-DNA forming microsatellite. Blood 110:3039-3048. 4. Werth N, Beerlage C, Rosenberger C, Yazdi AS, Edelmann M, Amr A, Bernhardt W, von Eiff C, Becker K, Schäfer A, Peschel A, Kempf VA. 2010. Activation of hypoxia inducible factor 1 is a general phenomenon in infections with human pathogens. PLoS One 5:e11576. 5. Schaible B, Schaffer K, Taylor CT. 2010. Hypoxia, innate immunity and infection in the lung. Respir Physiol Neurobiol 174:235-243. 6. Rupp J, Gieffers J, Klinger M, van Zandbergen G, Wrase R, Maass M, Solbach W, Deiwick J, Hellwig-Burgel T. 2007. Chlamydia pneumoniae directly interferes with HIF-1a stabilization in human host cells. Cell Microbiol 9:2181-2191. 7. Devraj G, Beerlage C, Brüne B, Kempf VA. 2017. Hypoxia and HIF-1 activation in bacterial infections. Microbes Infect 19:144-156. 118 8. Peyssonnaux C, Datta V, Cramer T, Doedens A, Theodorakis EA, Gallo RL, Hurtado-Ziola N, Nizet V, Johnson RS. 2005. HIF-1α expression regulates the bactericidal capacity of phagocytes. J Clin Invest 115:1806-1815. 9. Berger EA, McClellan SA, Vistisen KS, Hazlett LD. 2013. HIF-1α is essential for effective PMN bacterial killing, antimicrobial peptide production and apoptosis in Pseudomonas aeruginosa keratitis. PLoS Pathog 9:e1003457. 10. Frede S, Stockmann C, Freitag P, Fandrey J. 2006. Bacterial lipopolysaccharide induces HIF-1 activation in human monocytes via p44/42 MAPK and NF-κB. Biochem J 396:517-527. 11. Leire E, Olson J, Isaacs H, Nizet V, Hollands A. 2013. Role of hypoxia inducible factor-1 in keratinocyte inflammatory response and neutrophil recruitment. J Inflamm 10:28. 12. Peyssonnaux C, Boutin AT, Zinkernagel AS, Datta V, Nizet V, Johnson RS. 2008. Critical role of HIF-1 α in keratinocyte defense against bacterial infection. J Invest Dermatol 128:1964-1968. 13. Zinkernagel AS, Peyssonnaux C, Johnson RS, Nizet V. 2008. Pharmacologic augmentation of hypoxia-inducible factor–1α with mimosine boosts the bactericidal capacity of phagocytes. J Infect Dis 197:214-217. 14. Polke M, Seiler F, Lepper PM, Kamyschnikow A, Langer F, Monz D, Herr C, Bals R, Beisswenger C. 2017. Hypoxia and the hypoxia-regulated transcription factor HIF-1a suppress the host defence of airway epithelial cells. Innate Immun 23:373- 380. 15. Thiel M, Caldwell CC, Kreth S, Kuboki S, Chen P, Smith P, Ohta A, Lentsch AB, Lukashev D, Sitkovsky MV. 2007. Targeted deletion of HIF-1alpha gene in T cells 119 prevents their inhibition in hypoxic inflamed tissues and improves septic mice survival. PLoS One 2:e853. 16. Georgiev P, Belikoff BG, Hatfield S, Ohta A, Sitkovsky MV, Lukashev D. 2013. Genetic deletion of the alternative isoform I.1 of HIF-1α in T cells enhances antibacterial immune response and improves survival in the model of bacterial peritonitis in mice. Eur J Immunol 43:655-666. 17. Eichner A, Günther N, Arnold M, Schobert M, Heesemann J, Hogardt M. 2014. Marker genes for the metabolic adaptation of Pseudomonas aeruginosa to the hypoxic cystic fibrosis lung environment. Int J Med Microbiol 304:1050-1061. 18. Legendre C, Mooij MJ, Adams C, O'Gara F. 2011. Impaired expression of hypoxiainducible factor-1α in cystic fibrosis airway epithelial cells – a role for HIF-1 in the pathophysiology of CF. J Cyst Fibros 10:286-290. 19. Cullen L, McClean S. 2015. Bacterial adaptation during chronic respiratory infections. Pathogens 4:66-89. 20. Nizet V, Johnson RS. 2009. Interdependence of hypoxic and innate immune responses. Nat Rev Immunol 9:609-617. 21. Cummins EP, Keogh CE, Crean D, Taylor CT. 2016. The role of HIF in immunity and inflammation. Mol Aspects Med 47-48:24-34. 22. Eltzschig HK, Carmeliet P. 2011. Hypoxia and inflammation. N Engl J Med 364:656-665. 23. Campbell EL, Colgan SP. 2015. Neutrophils and inflammatory metabolism in antimicrobial functions of the mucosa. J Leukoc Biol 98:517-522. 24. Schaible B, Rodriguez J, Garcia A, von Kriegsheim A, McClean S, Hickey C, Keogh CE, Brown E, Schaffer K, Broquet A, Taylor CT. 2017. Hypoxia reduces 120 the pathogenicity of Pseudomonas aeruginosa by decreasing the expression of multiple virulence factors. J Infect Dis 215:1459-1467. 25. Sever JL, Youmans GP. 1957. The relation of oxygen tension to virulence of tubercle bacilli and to acquired resistance in tuberculosis. J Infect Dis 101:193-202. 26. Worlitzsch D, Tarran R, Ulrich M, Schwab U, Cekici A, Meyer KC, Birrer P, Bellon G, Berger J, Weiss T, Botzenhart K, Yankaskas JR, Randell S, Boucher RC, Döring G. 2002. Effects of reduced mucus oxygen concentration in airway Pseudomonas infections of cystic fibrosis patients. J Clin Invest 109:317-325. 27. Patel NJ, Zaborina O, Wu L, Wang Y, Wolfgeher DJ, Valuckaite V, Ciancio MJ, Kohler JE, Shevchenko O, Colgan SP, Chang EB, Turner JR, Alverdy JC. 2007. Recognition of intestinal epithelial HIF-1α activation by Pseudomonas aeruginosa. Am J Physiol Gastrointest Liver Physiol 92:G134-142. 28. Schaible B, Taylor CT, Schaffer K. 2012. Hypoxia increases antibiotic resistance in Pseudomonas aeruginosa through altering the composition of multidrug efflux pumps. Antimicrob Agents Chemother 56:2114-2118. 29. McClean S, Healy ME, Collins C, Carberry S, O'Shaughnessy L, Dennehy R, Adams Á, Kennelly H, Corbett JM, Carty F, Cahill LA, Callaghan M, English K, Mahon BP, Doyle S, Shinoy M. 2016. Linocin and OmpW are involved in attachment of the cystic fibrosis-associated pathogen Burkholderia cepacia complex to lung epithelial cells and protect mice against infection. Infect Immun 84:1424-1437. 30. Kurochkina N. 2015. SH Domains. Springer International Publishing. {Structure- Function Relationship of Bacterial SH3 Domains}. 71-89. 31. Carnielli CM, Artier J, de Oliveira JC, Novo-Mansur MT. 2017. Xanthomonas citri subsp. citri surface proteome by 2D-DIGE: Ferric enterobactin receptor and other 121 (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3541 PE=4 SV=2 - [A7FAX8_ACIBT] A3M5R8 Aldehyde dehydrogenase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1835 PE=4 SV=2 - [A3M5R8_ACIBT] 0,60 Cytoplasm A3M2E9 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0644 PE=4 SV=2 - [A3M2E9_ACIBT] 0,60 Mitochondrion A3M1E1 Alginate biosynthesis protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0260 PE=4 SV=2 - [A3M1E1_ACIBT] 0,60 Inner membrane A7FAX2 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3535 PE=4 SV=2 - [A7FAX2_ACIBT] 0,60 Periplasm A3M6C3 Putative phage integrase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2040 PE=4 SV=2 - [A3M6C3_ACIBT] 0,60 Mitochondrion A7FB06 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3569 PE=4 SV=2 - [A7FB06_ACIBT] 0,60 Cytoplasm 128 Figure S1. Subcellular localization of subexpressed proteins under hypoxia (1%O2). Cytoplasm 45% Secreted 16% Inner membrane 19% Outer membrane 10% Periplasm 6% Mitochondrion 4% Subcellular localization 129 ADDITIONAL DATA Table A1. Identification of the A. baumannii ATCC 17978 overexpressed proteins under hypoxia condition (1%O2). Accession Description Fold Change A3M7U7 23-dihydroxybenzoate-AMP ligase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2574 PE=4 SV=2 - [A3M7U7_ACIBT] 1,49 A3M1P4 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0379 PE=4 SV=2 - [A3M1P4_ACIBT] 1,49 A3M6H4 Putative exported protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2091 PE=4 SV=2 - [A3M6H4_ACIBT] 1,49 A3M4I6 Putative amino acid efflux transmembrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1402 PE=4 SV=2 - [A3M4I6_ACIBT] 1,49 A3M4F4 Oxidoreductase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1370 PE=4 SV=1 - [A3M4F4_ACIBT] 1,49 A3M7E7 50S ribosomal protein L31 OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=rpmE PE=3 SV=1 - [RL31_ACIBT] 1,49 A3M5N6 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1803 PE=4 SV=2 - [A3M5N6_ACIBT] 1,49 A3M194 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0212 PE=4 SV=2 - [A3M194_ACIBT] 1,50 A3M6J0 Putative glutamine amidotransferase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2107 PE=4 SV=2 - [A3M6J0_ACIBT] 1,50 A3M700 RNA-splicing ligase RtcB OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=rtcB PE=3 SV=2 - [A3M700_ACIBT] 1,50 A3M6X5 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2246 PE=4 SV=2 - [A3M6X5_ACIBT] 1,50 A3M343 Putative D-amino acid oxidase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0905 PE=4 SV=2 - [A3M343_ACIBT] 1,51 A3M6Q0 Cytochrome o ubiquinol oxidase subunit I OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2167 PE=3 SV=2 - [A3M6Q0_ACIBT] 1,51 A3M5Z0 Putative peroxidase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1907 PE=4 SV=2 - [A3M5Z0_ACIBT] 1,51 A3M4R7 Methionine import ATP-binding protein MetN OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=metN PE=3 SV=2 - [A3M4R7_ACIBT] 1,51 A3M9I0 Fimbrial protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3177 PE=3 SV=1 - [A3M9I0_ACIBT] 1,52 A3M185 Putative signal peptide OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0202 PE=4 SV=2 1,52 130 - [A3M185_ACIBT] A3M7C9 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2404 PE=4 SV=2 - [A3M7C9_ACIBT] 1,53 A3M8V7 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2946 PE=4 SV=2 - [A3M8V7_ACIBT] 1,54 A3M7Y3 Competence factor involved in DNA uptake OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2610 PE=4 SV=2 - [A3M7Y3_ACIBT] 1,54 A7FBB0 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3673 PE=4 SV=1 - [A7FBB0_ACIBT] 1,54 A3M308 Putative metal-dependent hydrolase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0870 PE=4 SV=2 - [A3M308_ACIBT] 1,54 A3M4Z2 Type 4 fimbrial biogenesis protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1559 PE=4 SV=2 - [A3M4Z2_ACIBT] 1,54 A3M2S3 Putative threonine efflux protein (RhtC) OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0777 PE=4 SV=2 - [A3M2S3_ACIBT] 1,54 A3M1A1 Transcriptional repressor NrdR OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=nrdR PE=3 SV=2 - [NRDR_ACIBT] 1,55 A3M5P7 Putative transporter OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1814 PE=4 SV=2 - [A3M5P7_ACIBT] 1,55 A3M731 ABC Lysine-arginine-ornithine transporter periplasmic ligand binding protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2302 PE=3 SV=2 - [A3M731_ACIBT] 1,55 A3M1E7 Putative membrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0266 PE=4 SV=2 - [A3M1E7_ACIBT] 1,55 A3M3A6 Putative phthalate transporter OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0968 PE=4 SV=2 - [A3M3A6_ACIBT] 1,56 A3M175 Putative transport protein (MFS superfamily) OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0188 PE=4 SV=2 - [A3M175_ACIBT] 1,56 A3M426 Putative ABC family drug transporter OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1242 PE=4 SV=2 - [A3M426_ACIBT] 1,56 A3M1I3 Putative acyl-CoA thioesterase II OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0311 PE=4 SV=2 - [A3M1I3_ACIBT] 1,57 A3M3E0 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1002 PE=4 SV=1 - [A3M3E0_ACIBT] 1,57 A3M2Y3 Putative flavodoxin or tryptophan repressor binding protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0843 PE=4 SV=2 - [A3M2Y3_ACIBT] 1,57 A3M1X9 Non-canonical purine NTP pyrophosphatase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0468 PE=3 SV=2 - [NTPA_ACIBT] 1,57 A3M2V2 Malonyl-[acyl-carrier protein] O-methyltransferase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=bioC PE=3 SV=2 - [A3M2V2_ACIBT] 1,57 A3M3N7 Probable 5-dehydro-4-deoxyglucarate dehydratase OS=Acinetobacter 1,58 131 baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1101 PE=3 SV=2 - [KDGD_ACIBT] A3M9S0 Putative methyltransferase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3275 PE=4 SV=2 - [A3M9S0_ACIBT] 1,58 A3M169 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0182 PE=4 SV=2 - [A3M169_ACIBT] 1,59 A3M4R9 D-methionine transport protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1485 PE=4 SV=2 - [A3M4R9_ACIBT] 1,59 A3M590 Putative siderophore biosynthesis protein putative acetyltransferase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1657 PE=4 SV=2 - [A3M590_ACIBT] 1,60 A3M4M0 Putative acyl-CoA dehydrogenase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1436 PE=4 SV=1 - [A3M4M0_ACIBT] 1,60 A3M998 ATP-dependent DNA helicase RecG OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=recG PE=3 SV=2 - [A3M998_ACIBT] 1,61 A3M5R5 Oxidoreductase FMN-binding OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1832 PE=4 SV=2 - [A3M5R5_ACIBT] 1,61 A3M8U2 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2931 PE=4 SV=2 - [A3M8U2_ACIBT] 1,61 A3M3L8 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1082 PE=4 SV=1 - [A3M3L8_ACIBT] 1,62 A3M684 Molybdenum cofactor guanylyltransferase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=mobA PE=3 SV=2 - [MOBA_ACIBT] 1,62 A3M445 Putative 3-hydroxyacyl-CoA dehydrogenase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1261 PE=4 SV=2 - [A3M445_ACIBT] 1,63 A3M5D4 Dihydrolipoamide acetyltransferase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1701 PE=3 SV=1 - [A3M5D4_ACIBT] 1,63 A3M5K3 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1770 PE=4 SV=2 - [A3M5K3_ACIBT] 1,63 A3M3E7 Putative lipoprotein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1009 PE=4 SV=2 - [A3M3E7_ACIBT] 1,64 A3M5M6 Transcriptional regulator LysR family OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1793 PE=4 SV=2 - [A3M5M6_ACIBT] 1,64 A3M2N6 Putative phage related protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0740 PE=4 SV=2 - [A3M2N6_ACIBT] 1,64 A3M255 Putative membrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0546 PE=4 SV=2 - [A3M255_ACIBT] 1,64 A3M2G3 Transposase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0658 PE=4 SV=2 - [A3M2G3_ACIBT] 1,64 A7FBX4 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3887 PE=4 SV=1 - [A7FBX4_ACIBT] 1,65 A3M9L6 Putative RND family drug transporter OS=Acinetobacter baumannii (strain 1,66 132 ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3219 PE=4 SV=2 - [A3M9L6_ACIBT] A3M5U7 Acyl-CoA dehydrogenase-like protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1864 PE=4 SV=2 - [A3M5U7_ACIBT] 1,66 A3M5A0 Putative ferric hydroxamate siderophore receptor OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1667 PE=3 SV=2 - [A3M5A0_ACIBT] 1,68 A3M6X9 Putative membrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2250 PE=4 SV=2 - [A3M6X9_ACIBT] 1,69 A3M8R5 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2903 PE=4 SV=2 - [A3M8R5_ACIBT] 1,69 A3M8Y0 Putative vanillate O-demethylase oxygenase subunit OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2971 PE=4 SV=2 - [A3M8Y0_ACIBT] 1,70 A3M5F8 Putative ferric siderophore receptor protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1725 PE=3 SV=2 - [A3M5F8_ACIBT] 1,70 A3M2R7 Putative membrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0771 PE=4 SV=2 - [A3M2R7_ACIBT] 1,70 A3M944 Putative membrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3041 PE=4 SV=2 - [A3M944_ACIBT] 1,71 A3M3M5 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1089 PE=3 SV=2 - [A3M3M5_ACIBT] 1,72 A3M5Y0 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1897 PE=4 SV=2 - [A3M5Y0_ACIBT] 1,72 A3M4F1 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1367 PE=4 SV=2 - [A3M4F1_ACIBT] 1,73 A3M9G3 Lipase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3160 PE=4 SV=2 - [A3M9G3_ACIBT] 1,74 A3M5N0 Aldehyde dehydrogenase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1797 PE=4 SV=1 - [A3M5N0_ACIBT] 1,74 A3M236 Putative holo-(Acyl carrier protein) synthase 2 OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0527 PE=4 SV=2 - [A3M236_ACIBT] 1,74 A3MA44 VirP protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3399 PE=4 SV=2 - [A3MA44_ACIBT] 1,75 A3M1N5 General secretion pathway protein F OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0369 PE=3 SV=2 - [A3M1N5_ACIBT] 1,75 A7FB98 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3661 PE=4 SV=1 - [A7FB98_ACIBT] 1,77 A3M3L5 Dichlorophenol hydroxylase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1079 PE=4 SV=2 - [A3M3L5_ACIBT] 1,77 A3M0R6 Anhydro-N-acetylmuramic acid kinase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=anmK PE=3 SV=2 - [A3M0R6_ACIBT] 1,78 A7FBY9 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / 1,81 133 CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3902 PE=4 SV=2 - [A7FBY9_ACIBT] A3M6J9 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2116 PE=4 SV=1 - [A3M6J9_ACIBT] 1,82 A3M5B9 Ribonuclease D OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1686 PE=4 SV=2 - [A3M5B9_ACIBT] 1,83 A3M0Z5 Putative transcriptional regulator (LysR family) OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0100 PE=4 SV=1 - [A3M0Z5_ACIBT] 1,83 A3M754 Putative nitrate transporter transmembrane protein (MFS superfamily) OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2326 PE=4 SV=2 - [A3M754_ACIBT] 1,83 A3M7B7 Putative acinetobactin utilization protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2392 PE=4 SV=2 - [A3M7B7_ACIBT] 1,84 A3M6K4 UPF0756 membrane protein A1S_2121 OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2121 PE=3 SV=2 - [Y2121_ACIBT] 1,86 A3M910 Stringent starvation protein B OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3003 PE=4 SV=2 - [A3M910_ACIBT] 1,86 A3M7V7 MFS family drug transporter OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2584 PE=4 SV=2 - [A3M7V7_ACIBT] 1,86 A3M3A1 Putative transcriptional regulator (AraC family) OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0963 PE=4 SV=2 - [A3M3A1_ACIBT] 1,86 A3M539 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1606 PE=4 SV=1 - [A3M539_ACIBT] 1,89 A3M833 Glycerophosphoryl diester phosphodiesterase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2661 PE=4 SV=2 - [A3M833_ACIBT] 1,89 A3M7T9 Putative ferric siderophore receptor protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2566 PE=3 SV=2 - [A3M7T9_ACIBT] 1,92 A7FBD9 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3702 PE=4 SV=2 - [A7FBD9_ACIBT] 1,92 A3M5T4 Penicillin G amidase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1851 PE=4 SV=2 - [A3M5T4_ACIBT] 1,93 A3M9V2 Putative transcriptional regulator (Lrp-like) OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3307 PE=4 SV=2 - [A3M9V2_ACIBT] 1,94 A3M4S8 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1494 PE=4 SV=2 - [A3M4S8_ACIBT] 1,97 A3M4G2 Putative long chain fatty-acid CoA ligase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1378 PE=4 SV=2 - [A3M4G2_ACIBT] 2,02 A3M1P9 DNA gyrase inhibitor YacG OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=yacG PE=3 SV=1 - [A3M1P9_ACIBT] 2,02 A3M3W0 DNA polymerase V component OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1174 PE=4 SV=1 - [A3M3W0_ACIBT] 2,06 A3M7R9 Secreted trypsin-like serine protease OS=Acinetobacter baumannii (strain 2,07 134 ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2546 PE=4 SV=2 - [A3M7R9_ACIBT] A3M6Y3 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2254 PE=4 SV=2 - [A3M6Y3_ACIBT] 2,08 A3M3Y0 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1194 PE=4 SV=2 - [A3M3Y0_ACIBT] 2,09 A3M6A5 Putative tail fiber OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2022 PE=4 SV=2 - [A3M6A5_ACIBT] 2,10 A3M598 Putative membrane protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1665 PE=3 SV=2 - [A3M598_ACIBT] 2,12 A3M2F8 Ferrous iron transport protein B OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0653 PE=3 SV=1 - [A3M2F8_ACIBT] 2,16 A3M5P4 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1811 PE=4 SV=2 - [A3M5P4_ACIBT] 2,17 A3M5L0 Methylenetetrahydrofolate reductase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1777 PE=3 SV=1 - [A3M5L0_ACIBT] 2,17 A3M1I7 Putative fusaric acid resistance protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0317 PE=4 SV=2 - [A3M1I7_ACIBT] 2,25 A3M6H7 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2094 PE=4 SV=2 - [A3M6H7_ACIBT] 2,28 A7FBR2 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3825 PE=4 SV=2 - [A7FBR2_ACIBT] 2,30 A3M0Y1 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0086 PE=4 SV=2 - [A3M0Y1_ACIBT] 2,32 A3M594 Fructose-26-bisphosphatase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1661 PE=4 SV=2 - [A3M594_ACIBT] 2,32 A3M166 NADPH-dependent FMN reductase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0179 PE=4 SV=2 - [A3M166_ACIBT] 2,37 A3M6U3 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2210 PE=4 SV=2 - [A3M6U3_ACIBT] 2,38 A3M525 Putative Phage head-tail adaptor OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1592 PE=4 SV=1 - [A3M525_ACIBT] 2,40 A3M6S8 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2195 PE=4 SV=1 - [A3M6S8_ACIBT] 2,43 A7FBB9 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3682 PE=4 SV=1 - [A7FBB9_ACIBT] 2,44 A3M5Q5 Putative transcription regulator protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1822 PE=4 SV=2 - [A3M5Q5_ACIBT] 2,48 A3MA53 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3408 PE=4 SV=2 - [A3MA53_ACIBT] 2,62 A3M7D5 Fatty acid desaturase OS=Acinetobacter baumannii (strain ATCC 17978 / 2,75 135 CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2410 PE=4 SV=2 - [A3M7D5_ACIBT] A7FBC0 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3683 PE=4 SV=2 - [A7FBC0_ACIBT] 2,88 A7FAV6 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_3519 PE=4 SV=2 - [A7FAV6_ACIBT] 2,95 A3M1D5 Permease (DMT) superfamily OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0254 PE=4 SV=2 - [A3M1D5_ACIBT] 3,01 A3M8U0 Putative cation efflux system protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2929 PE=4 SV=2 - [A3M8U0_ACIBT] 3,70 A3M4M1 Putative acyl-CoA dehydrogenase OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_1437 PE=4 SV=1 - [A3M4M1_ACIBT] 4,30 A3M8N5 Uncharacterized protein OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_2873 PE=4 SV=1 - [A3M8N5_ACIBT] 5,01 A3M2Q8 NADH dehydrogenase I chain L OS=Acinetobacter baumannii (strain ATCC 17978 / CIP 53.77 / LMG 1025 / NCDC KC755 / 5377) GN=A1S_0762 PE=4 SV=2 - [A3M2Q8_ACIBT] 5,37 136 Table A2. Identification of the A549 subexpressed proteins under hypoxia condition (1%O2). Accession Description Fold Change Q05823 2-5A-dependent ribonuclease OS=Homo sapiens GN=RNASEL PE=1 SV=2 - [RN5A_HUMAN] 0,25 Q9UJT2 Testis-specific serine kinase substrate OS=Homo sapiens GN=TSKS PE=1 SV=3 - [TSKS_HUMAN] 0,36 B7ZVZ4 KIAA1211 protein OS=Homo sapiens GN=KIAA1211 PE=2 SV=1 - [B7ZVZ4_HUMAN] 0,36 Q9Y4A8 Nuclear factor erythroid 2-related factor 3 OS=Homo sapiens GN=NFE2L3 PE=1 SV=1 - [NF2L3_HUMAN] 0,40 B9EIP2 Olfactory receptor, family 51, subfamily A, member 4 OS=Homo sapiens GN=OR51A4 PE=2 SV=1 - [B9EIP2_HUMAN] 0,48 Q99501 GAS2-like protein 1 OS=Homo sapiens GN=GAS2L1 PE=1 SV=2 - [GA2L1_HUMAN] 0,51 P05543 Thyroxine-binding globulin OS=Homo sapiens GN=SERPINA7 PE=1 SV=2 - [THBG_HUMAN] 0,51 Q8N4C6-10 Isoform 3 of Ninein OS=Homo sapiens GN=NIN - [NIN_HUMAN] 0,55 E9PPJ0 Uncharacterized protein OS=Homo sapiens GN=SF3B2 PE=4 SV=1 - [E9PPJ0_HUMAN] 0,55 Q96JB1 Dynein heavy chain 8, axonemal OS=Homo sapiens GN=DNAH8 PE=1 SV=2 - [DYH8_HUMAN] 0,56 Q96LI6 Heat shock transcription factor, Y-linked OS=Homo sapiens GN=HSFY1 PE=1 SV=1 - [HSFY1_HUMAN] 0,56 Q15058 Kinesin-like protein KIF14 OS=Homo sapiens GN=KIF14 PE=1 SV=1 - [KIF14_HUMAN] 0,57 Q9BWV3 Cytidine and dCMP deaminase domain-containing protein 1 OS=Homo sapiens GN=CDADC1 PE=2 SV=1 - [CDAC1_HUMAN] 0,57 O00420 F19541_1 OS=Homo sapiens GN=PRODH2 PE=2 SV=1 - [O00420_HUMAN] 0,57 Q5U086 Serine/threonine-protein phosphatase OS=Homo sapiens PE=2 SV=1 - [Q5U086_HUMAN] 0,57 Q70EL1 Inactive ubiquitin carboxyl-terminal hydrolase 54 OS=Homo sapiens GN=USP54 PE=1 SV=4 - [UBP54_HUMAN] 0,57 Q9Y678 Coatomer subunit gamma OS=Homo sapiens GN=COPG PE=1 SV=1 - [COPG_HUMAN] 0,59 B4E1D5 cDNA FLJ54400, highly similar to Eukaryotic translation initiation factor 3 subunit 8 OS=Homo sapiens PE=2 SV=1 - [B4E1D5_HUMAN] 0,59 B4DMB1 cDNA FLJ53358, highly similar to Heterogeneous nuclear ribonucleoprotein R OS=Homo sapiens PE=2 SV=1 - [B4DMB1_HUMAN] 0,59 P25705 ATP synthase subunit alpha, mitochondrial OS=Homo sapiens GN=ATP5A1 PE=1 SV=1 - [ATPA_HUMAN] 0,59 B7ZM61 PLCE1 protein OS=Homo sapiens GN=PLCE1 PE=2 SV=1 - [B7ZM61_HUMAN] 0,60 F8W6H6 Uncharacterized protein OS=Homo sapiens GN=MYO5A PE=4 SV=1 - [F8W6H6_HUMAN] 0,60 Q8TER5-3 Isoform 3 of Rho guanine nucleotide exchange factor 40 OS=Homo sapiens GN=ARHGEF40 - [ARH40_HUMAN] 0,60 Q562R1 Beta-actin-like protein 2 OS=Homo sapiens GN=ACTBL2 PE=1 SV=2 - [ACTBL_HUMAN] 0,61 Q9P2F8 Signal-induced proliferation-associated 1-like protein 2 OS=Homo sapiens GN=SIPA1L2 PE=1 SV=2 - [SI1L2_HUMAN] 0,61 F5GYZ0 Uncharacterized protein OS=Homo sapiens GN=KLRC1 PE=4 SV=1 - [F5GYZ0_HUMAN] 0,61 B4DWZ4 cDNA FLJ51365, highly similar to Flap endonuclease 1 (EC 3.1.-.-) 0,61 137 69 pathogens, along with a decreased perfusion due to vascular dysfunction (21-23). 70 Therefore, the microenvironment at the area of infection plays a crucial role in 71 determining the outcome of an infection. Hypoxia not only modifies the host cells but 72 also the bacterial metabolism and virulence (5). In P. aeruginosa and Mycobacterium 73 tuberculosis the expression of virulence factors such as alkaline protease, siderophores 74 and exotoxin A are reduced by hypoxia (24,25). However, hypoxia can also increase the 75 production of alginate and the expression of the PA-I lectin/adhesin by P. aeruginosa 76 causing a disruption in intestinal barrier and allowing exotoxin A to cross the epithelium 77 (26,27). Exposure to hypoxia also induces antibiotic resistance in P. aeruginosa by an 78 alteration of efflux pumps expression (28). Together, these studies demonstrate the 79 complexity of HIF-pathogen interactions. 80 The aim of this study was to evaluate the effect of hypoxia on A. baumannii and P. 81 aeruginosa pathogenesis, in vitro, regarding to bactericidal activity and 82 adherence/invasion, and in murine models of infection, regarding to survival, and 83 bacterial load; and the innate immune response in vitro and in vivo. 84 85 RESULTS 86 Hypoxia increases HIF-1α levels in epithelial and macrophages cells 87 HIF-1α levels in cell lines after 6 and 24 h under hypoxia (1% O 2 ) and normoxy (21% 88 O 2 ) were measured. In epithelial cells, HIF-1α levels were 2.69 times higher after 6 h in 89 hypoxia than in normoxy (2296.98 ± 157.74 pg/mL vs. 853.63 ± 95.47 pg/mL, 90 P<0.001) and were higher than after 24 h (1107.70 ± 96.08 pg/mL vs. 592.27 ± 48.86 91 pg/mL, P<0.01). In macrophages cells, HIF-1α levels were 1.50 times higher after 6 h 92 in hypoxia than in normoxia (331.64 ± 52.93 pg/mL vs. 220.67 ± 11.87 pg/mL) and 93 were higher than after 24 h under hypoxia (223.59 ± 7.05 pg/mL vs. 235.27 ± 9.31 Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 144 94 pg/mL; hypoxia 6 h vs. hypoxia 24 h). No significant differences in HIF-1α levels were 95 observed in normoxia between the different times points analysed. 96 The marked increase of HIF-1α levels after 6 h under hypoxia (1% O 2 ) defined the time 97 of hypoxia condition prior the infection for the in vitro and in vivo experiments. 98 99 Hypoxia increases bactericidal activity of epithelial and macrophages cells against 100 A. baumannii and P. aeruginosa 101 First, we observed that ATCC 17978 and PAO1 strains growth during 2 and 24 h was 102 indistinguishable between hypoxia (1% O 2 ) and normoxia (Fig. 1A). Next, we 103 determined if hypoxia affects the bactericidal activity of epithelial and macrophages 104 cells. Bacterial counts of ATCC 17978 and PAO1 strains found in the extracellular 105 medium of both cell lines under hypoxia (1% O 2 ) showed a decrease of bacterial 106 concentrations after 2 and 24 h compared to normoxia, (Fig. 1B and 1C). These data 107 support an increase in the bactericidal activity of these cell lines under hypoxia. 108 109 HIF-1α overexpression increases bactericidal activity of epithelial and 110 macrophages cells against A. baumannii and P. aeruginosa 111 Bacterial counts of ATCC 17978 and PAO1 strains found in the extracellular medium 112 of both cell lines under a 0.1 mM DMOG treatment showed a decrease of bacterial 113 concentrations after 24 h compared to normoxia, (Fig. 1B and 1C). These data support 114 an increase in the bactericidal activity of these cell lines when HIF-1α is overexpressed 115 due to the treatment with DMOG. 116 117 Hypoxia decreases bacterial adherence and invasion to epithelial and macrophages 118 cells Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 145 119 The bacterial adherence of ATCC 17978 and PAO1 strains to both cell lines was 120 significantly lower under hypoxia (1% O2), except in the case of 2 h post-infection by 121 ATCC 17978 strain in the RAW 264.7 cells in which it presented higher bacterial 122 adherence (182.67 ± 11% vs. 100% ± 0%, P<0.001) (Fig. 2A and 2B). 123 Bacterial counts of ATCC 17978 strain inside epithelial and macrophages cells under 124 hypoxia (1% O 2 ) showed an increase of bacterial concentrations 2 h post-infection (150 125 ± 0% for epithelial cells, and 146.73 ± 5.01% for macrophages cells, P<0.001), and a 126 decrease at 24 h post-infection compared to normoxia (48.55 ± 34.80% for epithelial 127 cells, P<0.001 and 8.69 ± 6.85% for macrophages cells, P<0.001) (Fig. 2C). On the 128 other hand, PAO1 strain counts inside both cell lines under hypoxia (1% O 2 ) showed a 129 decrease of bacterial concentrations after 2 h (P<0.001 for macrophages cells) and 24 h 130 (P<0.001) compared to normoxia (Fig. 2D). These data indicated that hypoxia affects 131 the adherence and invasion of A. baumannii and P. aeruginosa 24 h after bacterial 132 infection. 133 134 Hypoxia reduces the expression of proteins involved in cell adherence 135 iTRAQ results show that there are 51 down-expressed proteins under hypoxia (Fold 136 Change < 0.6) present in the extracellular medium of a 2 h infection of A549 cells by A. 137 baumannii ATCC 17978 strain (Table S1). Forty-five % are localized in the cytoplasm, 138 16% are secreted, 19% are in the inner membrane, 10% in the outer membrane, 6% in 139 the periplasm and 4% in the mitochondrion (Fig. S1). The proteins localized in the outer 140 membrane and could be involved in cell adhesion are OmpW, putative ferric 141 siderophore receptor protein A1S_3339, putative ferric siderophore receptor protein 142 A1S_0474, ferric enterobactin receptor A1S_0981, and ferrichrome-iron receptor Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 146 143 A1S_1921. Moreover, the secreted uncharacterized protein A1S_3900, which is a 144 protein that presents SH3-like domains, could also be involved in cell adhesion. 145 146 Hypoxia reduces bacterial load in tissues and fluids in a peritonitis sepsis model by 147 A. baumannii 148 The MLD needed to achieve 100% mortality for ATCC 17978 strain was lower in 149 hypoxia (10% O 2 ) than in normoxia (2.08 vs. 3.20 log 10 cfu/mL). For the rest of 150 experiments, we used the MLD calculated in normoxia. The survival time was higher in 151 mice infected under normoxia than hypoxia (10%O 2 ) (36.42 vs. 23.92 h, P<0.001) (Fig. 152 3A). 153 In the sepsis model by A. baumannii, regardless the studied condition, all mice 154 presented bacteremia after 4 h infection. No differences were found in the bacterial load 155 in tissues (spleen, and lungs) and fluids (PF and blood) between hypoxia and normoxia 156 after 4 h infection (Table 1). However, at the time of death, significant differences 157 between hypoxia and normoxia were found in the bacterial loads in lungs, PF, and 158 blood (Table 1). Moreover, significant differences between animals under normoxia and 159 under hypoxia (6 h) followed by normoxia were found in the bacterial loads at the time 160 of death in spleen, lungs and PF (Table 1). Bacterial loads in spleen, lungs, PF and 161 blood were lower under hypoxia (hypoxia 6 h prior infection, or during the whole 162 experiment) compared to normoxia. 163 HIF-1α levels showed no differences between hypoxia and normoxia in controls 164 animals (not infected). Contradictorily, infected mice under the different studied 165 conditions presented higher HIF-1α levels than controls mice under normoxia at the 166 time of death (Fig. 3B). 167 Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 147 168 Hypoxia reduces bacterial load in tissues and blood in a pneumonia model by P. 169 aeruginosa 170 The MLD calculated for PAO1 strain was the same for both conditions (8.54 log 10 171 cfu/mL). Survival time was significantly higher under normoxia than under hypoxia 172 (10% O 2 ) (P<0.01) or six hours’ hypoxia followed by normoxia (P<0.05) (Fig. 3A). 173 Pathological studies confirmed pneumonia 4 h after infection in all the conditions 174 analyzed, but the symptoms were higher under normoxia (data not shown). 175 After 4 h of infection, no significant differences were found in the bacterial loads in 176 tissues and blood between both conditions (Table 2). Nevertheless, at the mice time of 177 death, significant differences were found in spleen, lungs and blood (between hypoxia 178 and normoxia (Table 2). Similarly, significant differences at the time of death were 179 found in the bacterial loads in blood between mice under 6 h hypoxia prior the infection 180 followed by normoxia and the animals infected in normoxia (Table 2). Bacterial loads 181 in tissues and blood were lower under both hypoxemic conditions than under normoxia. 182 HIF-1α levels were not significant among the studied conditions. Opposing to what 183 happened in the peritoneal sepsis model by A. baumannii; infected mice under the 184 different conditions studied presented lower HIF-1α levels than non-infected mice at the 185 time of death (Fig. 3B). 186 187 In vitro and in vivo cytokines production under hypoxia and normoxia 188 The infection of epithelial cells by ATCC 17978 and PAO1 strains showed that IL-6, 189 TNF-α and IL-10 levels were similar for both conditions at 2 and 24 h post-infection 190 (Fig. 4A). When infecting the RAW 264.7 cells by ATCC 17978 and PAO1 strains, IL- 191 6 and TNF-α levels at 24 h and 2 h post-infection were significantly higher in hypoxia 192 (P<0.05), respectively (Fig. 4B). Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 148 193 In the sepsis model by ATCC 17978 strain, only IL-10 levels were significantly higher 194 after 4 h infection in hypoxia (P<0.05). No differences were found in IL-6 or TNF-α 195 levels, although they were slightly higher under hypoxia. No differences in cytokines 196 levels were found at the animal time of death (Fig. 4C). In the pneumonia model by 197 PAO1 strain, IL-6 levels were significantly higher in hypoxia (P<0.05) after 4 h 198 infection. Again, no differences were found for IL-10 and TNF-α levels although they 199 were rather higher under hypoxia (Fig. 4C). Contradictory to what we observed in the 200 sepsis model, at the mice time of death, we observed lower IL-10 levels under hypoxia 201 (P<0.05). Again, no differences were found in IL-6 or TNF-α levels, although they were 202 slightly lower under hypoxia (Fig. 4C). 203 204 DISCUSSION 205 To our knowledge, this is the first study that analyses in vitro and in vivo the effect of 206 hypoxia during infection by A. baumannii and P. aeruginosa. We observed that hypoxia 207 in vitro increases bactericidal activity of host cells, and reduce bacterial adherence and 208 invasion. We also found that hypoxia in vivo diminish bacterial load in fluids and 209 tissues, but mice survival time was shorter under hypoxia. 210 We showed that hypoxia doesn’t affect the in vitro growth of A. baumannii and P. 211 aeruginosa. However, it increases the bactericidal activity in epithelial and macrophage 212 cells. The study of Peyssonnaux et al. showed that hypoxia modifies gene regulation in 213 host cells and it increases the LL-37 cathelicidin levels, an antimicrobial peptide 214 involved in the clearance of pathogens (12). Moreover, we see that hypoxia decreases 215 the bacterial adherence to host cells. This effect might be due to the modification of cell 216 or bacterial membrane under this condition. iTRAQ results confirmed that hypoxia 217 downregulates 51 proteins in A. baumannii ATCC 17978, five of them are localized in Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 149 218 the outer membrane which could be involved in cell adherence due to the previous 219 reports of their involvement in the bacterial adherence (29-33) 220 Regarding to bacterial invasion, we observed differences in behavior between A. 221 baumannii and P. aeruginosa under hypoxia. Our data showed a reduction in the P. 222 aeruginosa internalization into epithelial and macrophages cells, confirming the results 223 obtained in a previous study in which it is demonstrated that hypoxia decreases the P. 224 aeruginosa internalization into A549 cells (34). However, A. baumannii internalization 225 in both host cells is higher after 2 h under hypoxia, but it is reduced after 24 h. 226 Consequently, hypoxia cannot stop the A. baumannii invasion during the first few hours 227 of infection but it is finally hindered after 24 h. Therefore, we believe that hypoxia 228 confers higher resistance against bacterial invasion to host cells in order to avoid an 229 intracellular replication and the infection evolution. 230 In the in vivo experiments, we observe that a lower bacterial inoculum is needed to 231 cause 100% of mice mortality under hypoxia in the peritoneal sepsis model by A. 232 baumannii. All infected mice presented bacteremia 4 h post-infection for the studied 233 conditions. Moreover, we observe lower bacterial load in blood, PF lungs and spleen 234 under hypoxia. We also show that maintaining animals 6 h under hypoxia before the 235 infection is enough to reduce bacterial load at the time of death. These results are in 236 accordance with a previous study in which the use of the compound AKB-4924, that 237 increases HIF-1α levels, reduced bacterial loads recovered in a S. aureus skin infection 238 model (35). Moreover, these results are in accordance with the in vitro adherence assays 239 data. The increase of host cells bactericidal activity under hypoxia as well as a reduction 240 of bacterial adherence could allow the immune system to eliminate the infection better. 241 In the pneumonia model of infection by P. aeruginosa, we observed no differences in 242 the inoculum needed to cause 100% of mortality between the studied conditions. Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 150 243 Bacteremia observed in mice 4 h post-infection was 44.44% and 61.11% for normoxia 244 and hypoxia, respectively. The difference found in the bacteremia levels between both 245 animal models is because the severity of the sepsis model (36,37). As in the sepsis 246 model by A. baumannii, we observed lower bacterial load in fluids and tissues under 247 hypoxia, and under hypoxia followed by normoxia, are in accordance with the in vitro 248 results of adherence and invasion. Again, as in the sepsis animal model, survival time 249 was longer under normoxia. 250 In both animal models, HIF-1α levels were higher after 4 h under hypoxia, being the 251 levels similar at the animal time of death regardless the studied conditions. These results 252 are in accordance with the in vitro studies in which HIF-1α levels increased over time 253 under hypoxia and then decreased 24 h after. We observed that A. baumannii causes an 254 increase of HIF-1α levels compared to the control as what reported in another study in 255 which infection with A. baumannii produced HIF-1α levels increase (4). In contrast, P. 256 aeruginosa produces HIF-1α levels reduction under hypoxia compared to the control. 257 This result could be explained because 2-alkyl-4-quinolone and Pseudomonas 258 Quinolone Signal triggers the HIF-1α degradation through the 26S-proteasome 259 proteolytic pathway, blocking the HIF-1α effect (38,39). 260 As it is well defined in the literature, hypoxia regulates the immune response (20). In 261 the A. baumannii sepsis model, we observed under hypoxia high IL-10 levels after 4 h 262 infection. Meng et al. indicated that HIF-1α is involved in IL-10 production by B cells 263 (40), and IL-10 is an anti-inflammatory cytokine that suppresses macrophage and 264 dendritic cells function (41). In the P. aeruginosa pneumonia model, we detected high 265 IL-6 levels 4 h post-infection under hypoxia. It has been showed that HIF-1α increase 266 TNF-α and IL-6 levels (20,42). Moreover, IL-10 levels decreased under hypoxia in the 267 pneumonia model by P. aeruginosa at the time of death. However, we did not find Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 151 268 differences in cytokines levels between hypoxia and normoxy neither in vitro nor in 269 vivo. Therefore, we found that hypoxia has not a strong impact on cytokine production 270 (20), being more important on the bactericidal activity of host cells and on the reduction 271 of infection in animals. 272 However, this study has some limitations. HIF-1α is a factor involved in multiple 273 cellular pathways and its expression is also regulated by different proteins. Therefore, 274 finding a clear correlation between hypoxia, HIF-1α expression, inflammatory 275 responses and infection is complex, and multiple cellular processes have to be taken 276 into consideration. 277 In conclusion, hypoxia increases the bactericidal activity of host cells. In contrast, 278 mortality in animals under hypoxia is faster even with a lower bacterial load in tissues 279 and fluids. Moreover, we find that hypoxia has not a strong impact on cytokine 280 production by both pathogens. Finally, despite both studied microorganism are close 281 phylogenetically, they present slightly different behavior under hypoxia. 282 283 MATERIALS AND METHODS 284 Bacterial strains and growth condition 285 The wild-type strains A. baumannii ATCC 17978 and P. aeruginosa PAO1 were used. 286 They were cultured at 37ºC overnight (160 rpm) in Mueller Hinton Broth (MHB) 287 (Sigma, Spain). Cultured strains were washed with phosphate-buffered saline (PBS) and 288 suspended in Dulbecco’s modified Eagle’s medium (DMEM) before their use in 289 eukaryotic cell culture experiments (human lungs epithelial cell line A549 and murine 290 macrophage cell line RAW 264.7). 291 292 Growth curves analysis Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 152 293 The growth of A. baumannii ATCC 17978 and P. aeruginosa PAO1 strains under 294 hypoxia (1% and 10% O 2 ) and normoxia (21% O 2 ) in static were monitored during 24 295 h. Both strains were grown overnight in 20 ml of MHB, and a 1:10000 dilution was 296 performed to obtain, approximately, 10 5 cfu/ml in a 40 ml culture of MHB (10% and 297 21% O 2 ) or DMEM (1% and 21% O 2 ). Three replicates were performed in different 298 days. 299 300 A549 and RAW 264.7 culture and infection 301 Human lungs epithelial cell line A549 and murine macrophage cell line RAW 264.7 302 were grown in DMEM containing 10% Fetal Bovine Serum (Gibco, Spain), 1% HEPES 303 1M, vancomycin (50 mg/ml), gentamicin (20 mg/ml) and amphotericin B (0.25 mg/ml; 304 Gibco), as previously described (43). In the case of hypoxia condition studies, cells 305 were transferred to a hypoxia chamber (Coy Laboratories, USA) with a humidified 306 atmosphere of 1% O 2 , 5% CO 2 and the balance N 2 at 37°C. Cells were seeded (10 5 307 cells/well in a 24-well plate) for 30 h in 24-well plates before infection with A. 308 baumannii ATCC 17978 or P. aeruginosa PAO1 at a multiplicity of infection (MOI) of 309 500. To mimic hypoxia condition we treated the cells with 0.1 mM 310 Dimethyloxaloylglycine (DMOG) (Sigma, Spain), an inhibitor of prolyl hydroxylases 311 (44), 6 h prior bacterial infection and during infection. Immediately before the infection, 312 A549 cells were washed thrice with PBS and incubated in supplemented DMEM. 313 314 HIF-1α measurement in cell cultures 315 A549 and RAW 264.7 cells were seeded for 24 h in 6-well plates (10 6 cells/well). After 316 6 and 24 h in hypoxia (1% O 2 ) or normoxia condition, cells were washed thrice with 317 PBS, harvested using cell scraper and homogenized in RIPA buffer supplemented with Downloaded from http://iai.asm.org/ on September 7, 2018 by guest 153