The hypoxic environment differentially reprograms DCs depending on their maturation stage. Identification of TREM-1 as a common hypoxia marker.

1

FACOLTA’ DI MEDICINA E CHIRURGIA

SCUOLA DI SPECIALIZZAZIONE IN PATOLOGIA CLINICA

The

hypoxic

environment

differentially

reprograms

Dendritic Cells depending on their maturation stage.

Identification of TREM-1 as a common hypoxia marker

Relatori: Candidato:

Prof.

ssa

Maria Cristina Mingari Federica Raggi

Dott.

ssa

Maria Carla Bosco

2

INDEX

INTRODUCTION

Pag 5

1. DENDRITIC CELLS

Pag 5

1.1 DCs differentiation and activation

Pag 6

2. HYPOXIA

Pag 9

2.1 Molecular biology of hypoxia

Pag 11

3. DENDRITIC CELLS IN A HYPOXIC

MICROENVIRONMENT

Pag 13

3.1 Regulation of the chemokine/receptor gene expression

profile in DCs by hypoxia

Pag 13

3.2 Trascriptional pathways regulating gene expression

in hypoxic DCs

Pag 16

3.3 Regulation of DCs functional activities by hypoxia

Pag 18

4. JUVANILE IDIOPATIC ARTHRITIS

Pag 21

4.1

JIA classification

Pag 21

 Systemic juvenile idiopathic arthritis Pag 22

 Rheumatoid factor-positive polyarthritis Pag 22

 Oligoarthritis Pag 22

 Rheumatoid factor-negative polyarthritis Pag 23

 Enthesitis-related arthritis Pag 23

 Psoriatic arthritis Pag 23

3

4.2

JIA immunopathogenesis

Pag 24

4.3 Role of hypoxia in JIA

Pag 26

4.4 Role of DCs in JIA

Pag 28

OBJECTIVE OF THE STUDY

Pag 30

MATERIAL AND METHODS

Pag 31

DCs generation and culture

Pag 31

SFMC isolation

Pag 31

Cytokines and antibodies

Pag 32

Flow cytometry

Pag 32

RNA isolation and GeneChip hybridization

Pag 32

Real-time RT-PCR

Pag 33

Cross-linking of TREM-1 cells

Pag 34

Mixed leukocyte reaction

Pag 34

Western blot analysis and immunoprecipitation

Pag 35

ELISA

Pag 35

Statistical analysis

Pag 35

RESULTS

Pag 36

Regulation of DC cell-surface immune-related receptor

repertoire by hypoxia

Pag 36

TREM-1 is expressed in hypoxic iDCs

Pag 37

TREM-1 is expressed in hypoxic mDCs

Pag 42

TREM-1 engagement on H-iDCs stimulates their

Th1/Th17-polarizing proinflammatory activity

Pag 48

TREM-1 cross-linking on H-mDCs promotes

proinflammatory cytokine and chemokine release

Pag 51

TREM-1 is expressed in vivo on H-DCs infiltrating the

4

CONCLUSIONS

Pag 55

5

INTRODUCTION

1. DENDRITIC CELLS

Myeloid dendritic cells (DCs) represent a prominent component of the leukocyte infiltrate in damaged and inflammatory tissues. They belong to the mononuclear phagocyte (MP) system, deriving from a common committed bone marrow (BM) progenitor, and are found in every organ of the body at sites of potential entry of “dangerous factors” [1-6;6-8]. DCs are commonly considered as professional antigen-presenting cells, functioning as sentinels of the immune system and tailoring adaptive immune responses to match environmental cues, thereby serving as a bridge between innate and acquired immunity. They are central in both the orchestration of protective immunity against invading pathogens and the maintenance of tolerance to self-antigens. DC immunostimulatory properties reside in the capacity to migrate and patrol from non-lymphoid peripheral tissues, where they recognize pathogens and danger signals, to T cell areas of secondary lymphoid organs, where they present antigens to naive T cells and trigger T-cell responses.[9-11]. Deregulated DCs responses may result in amplification of inflammation, loss of tolerance, or establishment of immune escape mechanisms [8-12]. DC development and functions are acquired during a complex differentiation and maturation process tightly regulated by a network of inhibitory and activating signals present in the local

microenvironment [1-3;11;13-17]. These cells may experience low pO2 both during

differentiation/maturation in pathologic tissues [18-21] and upon migration to secondary lymphoid organs [20], thus raising the question of the contribution of reduced oxygenation to DC development and acquisition of immunogenic or tolerogenic properties (Figure 1)

6

Fig.1 DC origin and differentiation in pathologic tissues

1.1 DCs differentiation and activation

The DCs population in the body is highly heterogeneous in phenotype and function, associated with a typical anatomical and tissue distribution [1;3;6;8;11;15]. DCs generation involves three functionally and phenotypically distinct stages for which the terms “precursors”, “immature” and “mature” are commonly used [8;11;12]. DCs originate from hematopoietic stem cells in the BM via committed intermediate progenitors, such as the common macrophage/DC precursors (MDPs), which give rise to both the common DCs precursors (CDPs) and blood monocytes. CDPs differenziate into DC-restricted precursors (pre-DCs) and plasmacytoid (p)DCs, although pDCs can also arise from the common lymphoid progenitors (CLPs) [5;8]. In the steady-state, pre-DCs leave the BM to circulate via the bloodstream to reach non-lymphoid and lymphoid peripheral tissues, where they give rise to migratory and lymphoid-tissue resident conventional (c)DCs, respectively. In many cases, pre-DCs differentiate into inflammatory DCs as a consequence of infection or inflammation [8] Peripheral blood monocytes recruited from the circulation to peripheral tissues can also serve as cDC precursors in the steady state, but they mostly give rise to inflammatory DCs in response to microbial or inflammatory stimuli [1-4;6-8;11]. Immature (i)DCs are

7

specialized for antigen capture and processing and have a low T-cell stimulatory activity [9;10]. Under steady-state conditions, iDCs mostly reside at sites of potential pathogen entry within interface tissues, such as the skin and the respiratory or gastrointestinal mucosa, and in lymphoid tissues, where they play crucial roles in maintaining tolerance, not only to self antigens, but also to harmless environmental antigens and commensal organisms [3;11]. However, they continuously scan the surroundings for the presence of pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs), which they recognize through a defined repertoire of pattern-recognition (PRRs) and scavenger cell surface receptors [15;22]. Upon antigen uptake and activation by proinflammatory cytokines and DAMPs released in the context of tissue injury or PAMPs, iDCs undergo phenotypic and functional changes that culminate in their maturation into mature (m)DCs, which have a reduced potential for antigen uptake, but a higher capacity for antigen presentation and T-cell priming [9;11]. mDCs upregulate the expression of MHC antigens, maturation markers and co-stimulatory molecules (such as CD83, CD80, CD86, and CD40), decreasing that of phagocytic/endocytic receptors, and switch their chemokine receptor repertoire, downregulating receptors for inflammatory chemokines (e.g. CCR1, CCR2, CCR5, CCR6, and CXCR1) and upregulating those for homeostatic chemokines required for homing to secondary lymphoid organs, namely CCR7 and CXCR4, where they prime naive T cells triggering specific immune responses (Figure 2). [1;3;17].

8

Fig.2 Chemokine receptors expressed by dendritic cells

Modified from Cytokine and Growth Factor Reviews 2005 Dec,16(6):581-592

DC maturation is also associated with profound changes in the expression profile of both homeostatic (CCL17, CCL22, CCL19, and CCL18) and proinflammatory (CCL2, CCL3, CCL5, CCL20, CXCL8, CX3CL1, CXCL10, and CXCL16) chemokines, and distinct mDC populations and activation states can produce different sets of T cell-polarizing cytokines/chemokines, supporting the activation of inflammatory or regulatory T (Treg) cells, which are associated with different types of adaptive immune responses [1-3;11;13-16]. This classical view of DCs as a separate lineage is being challenged, and DCs are currently considered, like macrophages, a continuum of progeny from a common precursor that occupy defined niches in the body [5;6]. The possibility that monocyte-derived DCs, as opposed to cDCs, represent a differentiation/activation stage of MPs, rather than unique cell populations, must be kept in mind. Models for investigating DC development and functions DC subsets were classified according to their anatomical location, morphological appearance, phenotypic markers, migratory pathway, antigen expression, presumed origin and developmental pathway, functional abilities, and physiologic/pathologic state (“steady-state” or “inflammatory” DCs) [5;8;10-12;15]. Several culture methods can generate in vitro relatively pure populations of DCs using different precursors and

9

various combinations of cytokines/growth factors [2;5;8;11]. These systems represent a valuable source of homogeneous DCs for functional studies and for clinical immunotherapy trials. The most readily available model for DCs in humans uses in vitro-generated DCs, differentiated from CD14+ circulating monocytes or CD34+ hematopoietic stem cells and different combinations of cytokines to skew the in vitro differentiation and maturation of these precursors into DCs with different phenotypes and functions [9;11]. Unfortunately, these in vitro systems showed little correlation, in terms of cytokine requirement or the phenotype of DCs produced, with the most commonly used in vivo system, the mouse spleen, and poorly recapitulate the in vivo development and the full characteristics of their counterparts in the tissues [5;6]. Few

phenotypic markers are currently available to distinguish different DC subsets or

define specific stages of differentiation/ maturation or activation state. A new global

view of DC reprogramming upon development and activation can now beprovided by

gene expression profiling, which is instrumental in identifying new markers to define

DC heterogeneity and understanding DC biology and the relationships between DC

subsets inhuman and mouse [23;24]

2. HYPOXIA

Hypoxia is defined as “a condition of decreased partial oxygen pressure (pO2) relative

to that present in the atmosphere at sea level (≈20% O2)” and is an intrinsic component

of both our physiology and our pathology. O2 levels in the body are quite

heterogeneous. Physiologic pO2 in healthy tissues typically spans between 3% and 9%

depending on vascularization, seldom below 2.5%, as in lymphoid organs. pO2 varies

within a tissue in relationship to the distance from the end of the nearest capillary [19;20]. “Pathologic hypoxia” always develops as an aberrant status in damaged or degenerated tissues as a result of a disorganized or dysfunctional vessel network,

diminished blood supply, or insufficient neovascularization. In these situations, O2

demand is not bilance by O2 supply, and its concentration decreases below the levels

10

0.1%. Pathologic hypoxia is a hallmark of of inflammation, damaged or pathological

tissue and is a leading cause of high frequent pathologies like myocardial ischemia, atherosclerosis, abdominal aortic aneurysm, pulmonary hypertension and systemic hypertension, skin wounds, rheumatoid arthritis, preeclampsia and microbial infection (Figure 3) [18;19;21;25-27]. Cutaneous inflammation (Skin, wounds) HYPOXIA (0-20 mmHg) Growing tumors Cerebral ischemia Chronic obstructive pulmonary disease

Inflammatory arthropaties (RA) Diabetic retinopathy Preeclampsia Microbial infections Aherosclerosis Myocardial ischemia Glucose metabolism Cell proliferation/ survival

Angiogenesis Iron metabolism

Vasomotor control Erytropoiesis

Cell migration

Fig.3 Phatophisiologic hypoxia

Hypoxia is present to some extent in most solid malignant human cancers like prostate, cervix, breast, non-Hodgkin's lymphomas, malignant melanomas, metastatic liver tumors, renal cell cancer and the head and neck, because of an imbalance between the limited oxygen delivery capacity of the abnormal vasculature and the high oxygen consumption of tumor cells. This drives a complex and dynamic compensatory response to enable continued cell survival, including genomic changes leading to selection of hypoxia-adapted cells with a propensity to invade locally, metastasize, and recur following surgery or radiotherapy. There is indisputable clinical evidence from

11

numerous observational and therapeutic studies across a range of tumor types to implicate hypoxia as a key determinant of cancer behavior and treatment outcome.

2.1 Molecular biology of hypoxia

Hypoxia is perceived, at a cellular level, through oxygen sensor relays operating inside the cell, which transduce the signal through second messenger systems leading to the activation of transcription factors. An important and well-characterized ‘‘master

regulator’’ of the adaptive response to alterations in pO2 is the hypoxia-inducible

factor (HIF). Activation of the HIF signaling cascade leads to extensive changes in gene expression, which allows cells to adapt to reduced oxygenation. The downstream effects of hypoxia are mediated through the posttranscriptional stabilization of HIF-1a

and HIF-2a subunits, which is tightly controlled by cellular O2 levels. These proteins

are hydroxylated and acetylated in the presence of O2, bound by the von Hippel–

lindau tumor suppressor protein (pVHL), and rapidly degraded in the proteosome.

With decreasing O2 concentrations, HIFa hydroxylation and acetylation do not occur,

and HIF-a proteins are stabilized, translocate into the nucleus, dimerize with the constitutive HIF-1b/aryl hydrocarbon receptor nuclear translocator (ARNT) subunit, bind to specific HIF binding sites (HBSs) within the hypoxiaresponsive elements (HRE) of most hypoxia-regulated genes, transactivating their expression (Figure 4) [2;18;28;29].

12

Figure 4. HIF1a patway

Several regulatory pathways have been reported to be involved in the control of HIF expression and transcriptional activity. Functional or physical interaction of HIF with various cofactors binding to defined binding sites within HREs is required to drive efficient transcription of HIF target genes. HIF-independent pathways mediating gene transcription activation by hypoxia have also been described, contributing to the complex transcriptional profile activated by hypoxia. In particular, the cross-talk between the HIF/HRE system and the nuclear factor-kB (NF-kB) pathway has been implicated in the regulation of inflammation, highlighting an interdependence of hypoxic responses and innate immunity. A major role for hypoxia has been recognized in regulating development and biological functions of cells involved in innate and adaptive immunity [2;4;7;20;26;27;30]. Furthermore, hypoxia can have both pro- and anti-apoptotic consequences depending on the cellular context [20], inducing cell death [31] or survival of distinct immune cell populations

HIF-1a P P HIF-1a P P OH OH pVHLHIF-1a P P OH OH HIF-1a P P OH OH Ub Ub Ub HIF-1a Proteosomal degradation Ubiquitinatio n NORMOXIA Prolyl -Hydroxylation Ub Ub HYPOXIA Stabilization HIF-1a P P HIF-1a P P HIF-1a P P HIF-1a P P HIF-1a P P HIF-1a P P HIF-1a P P HIF-1a P P Hydroxylation Stabilization

HRE

HIF-1b HIF-1a P P CBP/ p300 Transcription cofactors

HRE

HIF-1b HIF-1a P P CBP/ p300 Transcription cofactors Transcription factors

13

3. DENDRITIC CELLS IN A HYPOXIC MICROENVIRONMENT

The efficacy of an immune response depends on DC functional conditioning by the local tissue microenvironment, which regulate their orientation toward a tolerogenic or

an immunogenic phenotype [12;15]. DCs experience changes in O2 levels during

development at and/or migration to inflammatory sites or secondary lymphoid organs, and adaptation to hypoxia is important for DCs to fulfill their functions in different tissues.

3.1 Regulation of the chemokine/receptor gene expression profile in DCs

by hypoxia

Hypoxia-mediated changes in gene expression have been characterized by DC profiling using the microarray technique [32]. In general, about 500–1000 genes are modulated by hypoxia in different cell types. Pioneer studies by Ricciardi et al demonstrated that iDCs generated from human-monocytes with GM-CSF/IL-4 under chronic hypoxic conditions (H-iDCs) redefine their transcriptome displaying a “hypoxic” gene expression pattern, characterized by upregulated expression of genes belonging to the glycolysis/gluconeogenesis and pentose phosphate pathways, a feature characteristic of the hypoxic status in various cell types [19;33]. Moreover, iDC response to hypoxia involved genes that control cell behavior and function, indicating functional reprogramming by the hypoxic microenvironment. An important feature of the H-iDCs transcriptome was the expression of genes involved in cell migration and motility [32;34], such as those controlling actin cytoskeleton, adherence junction, focal adhesion, and leukocyte transendothelial migration, indicating that H-iDCs have the potential for being mobile cells ready for redistribution within pathologic tissues or migration to the draining lymph nodes [1;3;17]. Furthermore, a dynamic change in the chemokine receptor expression profile occurred in H-iDCs. Upregulation of receptors for inflammatory chemokines, such as CCR2, CCR5, CXCR4, CCR3, and CX3CR1, was observed in the H-iDC transcriptional profile [32;34], and high membrane levels of these receptors were also reported in other

14

studies [35;36], associated with increased responsiveness to specific ligands [32;35]. These data suggest that differentiation under prolonged hypoxia promotes the onset of a migratory phenotype in monocyte-derived iDCs [2]. iDCs not only respond to, but also produce, chemokines, either constitutively or in response to inflammatory stimuli [1;3;14;17]. High levels of chemokines are present in areas of inflammation [7;26;37], and hypoxia modulates the chemotactic network [2;7;33]. In H-iDCs, hypoxia exerted a negative effect on genes coding for chemokines predominantly active on naïve/resting T cells and monocytes or associated with type II polarization, namely CCL18, CCL26 CCL23, CCL24, CCL14, CCL13, and CCL2, while upregulating those recruiting neutrophils and activated/memory Th1 lymphocytes, such as CXCL8,

CXCL1, CXCL10, and CCL20 [32;34;35], similarly to what was reported in hypoxic

monocytes and macrophages [2]. A parallel modulation of protein secretion was reported [32;34;35]. These data indicate that low pO2 may differentially regulate the recruitment of leukocyte subsets in pathological sites by tuning H-iDC production of chemokines. Recent studies defined the gene expression profile of monocytederived mDCs generated under chronic hypoxic conditions (H-mDCs) with GM-CSF/IL-4 and different maturation stimuli and demonstrated inducibility of genes involved in glycolytic metabolism and glucose transport or associated with nonglycolytic metabolism and ion transport, which contribute to the activation of adaptive pathways essential for their servival under hypoxia [34;38]. Furthermore, a set of upregulated genes had immunological relevance falling into immune regulation, inflammatory responses, angiogenesis, cell migration and adhesion pathways. Profound changes were observed in the expression of genes coding for chemokines, with a characteristic dichotomy resulting in the upregulation of those coding for proteins chemotactic for neutrophils, such as CXCL2, CXCL3, CXCL5, CXCL6, and CXCL8, and for activated/memory T lymphocytes, monocytes, and iDCs, e.g. CCL20, CCL3 and CCL5, concomitant with decreased expression of genes coding for naive/resting T

cells chemoattractants, CCL18 and CCL23 [34] Other genes inducible by hypoxia in

15

proliferation, adhesion, and chemotaxis, and on monocyte and T lymphocyte recruitment/activation, including osteopontin, VEGF, MIF IL-23A, and various components of the IL-1 receptor/ligand superfamily [26;37]. The chemokine/cytokine gene expression profile triggered by hypoxia was validated at the protein level [34],suggesting that H-mDCs have an increased ability to induce neo-vascularization and inflammation compared to their normoxic counterpart. A marked increase in the production of proinflammatory/ angiogenic cytokines was also observed in monocyte-derived mDCs previously differentiated under normoxia and induced to maturation under acute hypoxia [36] and in mouse mDCs generated from BM progenitors in the presence of GMCSF/ IL-4 and exposed to short-term hypoxia during maturation with LPS, expanding previous observations made in primary monocytes [33], monocyte-derived macrophages [39] and H-iDCs [32;34;35]. The fine tuning of the chemokine repertoire in H-mDC could be a critical set point for the control of their Th-polarizing activity at sites of inflammation [11;13]. Increased production of CCL3, CCL5, and CCL20 chemokines, which attract CCR5- and CCR6-expressing Th1 and Th17 activated/memory lymphocytes, is compatible with a Th1/Th17 shift of DC responses under chronic hypoxia. Inhibition of CCL18 and CCL23 also supports the notion of H-mDCs driving immune responses toward a Th1/Th17-polarized proinflammatory direction, given the reported role of these chemokines in Th-2 polarization, Treg generation and in maintenance of tolerance [14] . The observation that CCL18 and CCL23 expression is negatively affected by hypoxia not only in H-mDCs but also in primary monocytes and H-iDCs, together with that of other chemokines active on naïve/resting T cells or associated with type-II polarization (e.g. CCL26) [32-35], raises the possibility that inhibition of tolerance is a common feature of monocytic lineage cell adaptation to the hypoxic environment. Increased expression of osteopontin, IL-1, and IL-23, which are involved in the pathway leading to Th1 immunity and Th17 differentiation [11], may represent an additional evidence that

mDC generation under chronic hypoxia is associated with a Th1/Th17-biased

16

modifying their functional attributes and their cytokine/chemokine repertoire, irrespectively of the characteristics of the target population. The experimental conditions and the nature of the maturation stimuli will determine the type of the response to hypoxia, either polarized toward a Th1/Th17 or a Th2/Treg direction (Figure 5) [40].

Fig.5 Role of Hypoxia in the Dc reprogramming toward a Th1/Th17 direction

3.2 Transcriptional pathways regulating gene expression in hypoxic DCs

Immune cells recruited from highly oxygenated blood to pathologic lesions rapidly adapt their metabolism to cope with reduced tissue oxygenation [20]. Hypoxia is perceived through oxygen sensor relays operating inside the cell, which transduce the signal through second messenger systems leading to the activation of transcription factors. Abnormal activation or deregulation of the HIF-1 transcription pathway contributes to the pathogenesis of conditions as diverse as inflammation, vascular

Circulation a) Monocyte trapping Circulation Circulation Circulation ACTIVATION OF TH1/TH17-POLARIZED INFLAMMATORY RESPONSES ↓CC-chemokine receptors ↑Mn-arrest CXC-chemokines c) H-mDC inflammatory phenotype ↑neutrophil-attracting CXC-chemokines b) H-iDC migratory phenotype ↑Chemokine receptors O₂ levels ↓Inflammatory chemokines

↑CC-chemokines active on Mn, iDCs, activated/memory Th1/Th17 cells

↓CC-chemokines active on naive/resting T cells

17

disease, and cancer [18;19;25]. Furthermore, HIF-1α accumulation during DCs differentiation/maturation under chronic hypoxia was reported [35;36]. Similar results were obtained in other studies where iDCs and mDCs were first generated under normoxia and then exposed to acute hypoxia, showing a consistent and time-dependent increase of HIF-1α protein levels, paralleled by transcription of classical downstream target genes, such as VEGF and glucose transporter-1 (GLUT1) [16;38;41]. Interestingly, HIF-1α silencing is associated with a decreased ability of H-mDCs to stimulate allogenic T cells [42] and an increase of IL-12p70 production [34]. Hypoxia-dependent modulation of pro-survival and pro- or anti-apoptotic molecules was also counteracted by HIF-1α silencing in both H-iDCs and H-mDCs, supporting a role for HIF-1α in the activation of programmed cell death [41]. In line with these reports are the results obtained in murine mDCs, showing accumulation of HIF-1α protein in response to hypoxia with consequent induction of glycolytic target genes, enhanced glycolytic activity, and upregulation of T cell costimulatory molecules. These effects were prevented by HIF-1α knockdown, leading to inhibition of maturation and capability to stimulate allogenic T cells. Taken together, these results indicate the importance of 1 α for the regulation of DC functions. HIF-independent regulatory pathways mediating gene transcription by hypoxia were described [29]. The crosstalk between the HIF/HRE system and the NF-kB pathway was implicated in the regulation of inflammation, highlighting the connection between hypoxic responses and innate immunity. Recent evidence suggests the requirement of additional transcription factors, such as NF-kB, Ets-1, CCAAT/enhancer binding protein-α/β (C/EBP alpha/beta, activator-protein-1 (AP-1), and early growth response-1 (Egr- response-1) for hypoxia-dependent transactivation of genes coding for cytokine/chemokine gene in MPs [2;43]. Several inflammatory stimuli induce HIF and the transcription of hypoxia-responsive genes under normoxic conditions. Examples of such stimuli are cytokines (TNF, IL-1), small metabolites (picolinic acid), and microbial products (LPS) [2;27]. These signals are used to induce DC maturation in vitro. Therefore, depending on the in vitro culture conditions, the effects of hypoxia

18

will be superimposed to a variable level of expression of hypoxia-inducible genes generated by the maturation method used. Interaction between LPS and HIF in activating the HRE promoter has been reported [44], and it is conceivable that LPS-dependent or inLPS-dependent maturation protocols will yield a somewhat different mDC hypoxic phenotype.

3.3 Regulation of DC functional activities by hypoxia

Several studies reported that hypoxia decreases the ability of H-iDCs to capture foreign antigens. Elia et al. (2008), Yang et al. (2009), and Ogino et al. (2011) assessed the antigen-uptake ability of monocyte-derived iDCs in the presence of GM-CSF/IL-4 under chronic hypoxic conditions, demonstrating marked inhibition of dextran, BSA, and LPS endocytosis as well as zymosan and necrotic cells phagocytosis, compared to the normoxic counterpart, that was associated with downregulation of C-type lectin receptors, including CD209, CD206, and Dectin-1, and decreased activation of the RhoA/Ezrin-Radixin-Moesin pathway, involved in antigen uptake [34;35]. Consistent with these investigations are the observations by

Jantsch et al. (2008) in mouse DCs, who showed that short-term hypoxia reduced the

phagocytic activity of BM-derived iDCs generated under normoxia. The analysis of T-cell stimulatory capacity of H-DCs is a task complicated by technical and experimental difficulties linked to the hypoxic conditions. DC stimulatory activity on T cells, was measured in MLR by co-culturing T cells with PFA- or mitomycin C pretreated or γ-irradiated normoxic or hypoxic allogenic DCs under normoxia [35;36;45]. The hypoxic response is reversible, and exposure of H-DCs (even if mitomycin C-treated or γ-irradiated) to a normoxic environment during co-culture with T cells will change their phenotype, mimicking a hypoxia–reperfusion situation. H-DC fixation will prevent this phenomenon. Extended co-culture of viable DCs and T cells under hypoxia is an alternative protocol [16;34;42]. Unfortunately, there is no information on the rules regulating the allostimulation/response of hypoxic T cells with respect to costimulatory molecules, cytokine response/production, and

19

biomarkers/function generalization is difficult. Moreover, hypoxia is generally antiproliferative and can cause T cells apoptosis [20;31]. A recent report also demonstrated that exposure to acute hypoxia can activate a cell-death program in monocyte-derived DCs, characterized by upregulation of BNIP- 3 and BAX pro-apoptotic molecules and downregulation of the anti-pro-apoptotic protein Bcl2, associated with enhanced caspase-3 activity and poly (ADP-ribose) polymerase (PARP) cleavage [41]. Cell death generates DAMPS, which, in turn, modulate DC function. Hence, it is very difficult to generate a consensuson the T cell stimulatory activity of hypoxic DCs using in vitro assays. Ogino et al. (2011) showed that H-mDCs generated from monocytic precursors within a chronic hypoxic environment in the presence of GM-CSF/IL-4 followed by LPS stimulation exhibited higher ability to stimulate allogenic T cell proliferation, associated with increased surface expression levels of antigen-presenting receptors and T cell costimulatory molecules, e.g. HLA-DR, CD80, CD86, and CD83, compared to mDCs. Similar results were obtained in H-iDCs, although expression of the activated immunophenotype and ability to stimulate allo-T cell proliferation was reduced compared to H-mDCs, suggesting that hypoxia enhances iDCs phenotypic and functional maturation [32;42]. These data are in line with observations by other groups, who demonstrated that exposure of monocyte derived iDCs generated under normoxia to short-term hypoxia [16] or iDC maturation induced by LPS or other TLR agonists under acute hypoxia promoted overexpression of T cell costimulatory molecules, paralleled by increased T cell stimulatory activity. Accordingly, Jantsch et al. (2008) demonstrated that acute hypoxia amplifies LPS-induced maturation of murine iDC generated from BM progenitors in the presence of GM-CSF/IL-4 under normoxia, leading to a marked increase in the expression of costimulatory and antigen-presenting molecules and the ability to activate allogenic lymphocyte proliferation. In contrast, there are reports of reduced T cell-priming capacity (assessed as T cell proliferation and IFNγ secretion) of mDCs generated and matured by LPS under persistent hypoxia [34], or induced to maturation by LPS or other TLR agonists under acute hypoxia following prior differentiation under

20

normoxia [36]. A parallel decreased expression of costimulatory and

antigen-presenting molecules suggested that O2 availability was necessary to promote full DC

maturation and to prime T cell functions. Impaired phenotypic and functional maturation of BM-derived mouse mDCs generated under hypoxia, that was restored upon cell reoxygenation, was also documented [46]. Recognition of DC reprogramming by hypoxia and translation of this knowledge into clinical medicine will lead to improvement in DC-based immunotherapeutic strategies for tumors, chronic inflammatory diseases, and autoimmune disorders. The in vitro studies on changes in gene expression pointed to pathways triggered by hypoxic conditions, primarily those implicated in DC trafficking and immunoregulatory/inflammatory activities. Validation of these conclusions requires identification of differentially modulated genes that can be translated into biomarkers measurable in the tissue. Great care should be taken in interpreting the results obtained with DCs isolated ex vivo from dissociated hypoxic tissues, because spurious results may be introduced by their in vitro reoxygenation during preparation of single-cell suspensions [2]. A clear conclusion supported by all the studies is that hypoxic DCs are quite different from the normoxic counterpart in that a substantial numbers of genes are differentially modulated and different functions are expressed. We must consider that hypoxia impacts on cell biology not only with changes in gene/marker expression but also with an oxygen poor environment that will inhibit a variety of enzymes (e.g. oxidases) whose levels may not change or even be upregulated, but that cannot perform in the

absence of O2. The impact of hypoxia on enzyme activity will depend on the relative

affinity of the enzyme for molecular O2. A good example is the murine nitric oxide

synthase (iNOS) that is increased at the protein level by hypoxia but that cannot

produce NO in an O2-poor environment [44]. Tissue damage by ischemia–reperfusion

injury portraits the impact of O2 flux in a hypoxic tissue. Therefore, hypoxia-induced

transcription factors and gene expression is only part of a more general hypoxia-driven change in cell biology. An additional level of complexity is given, in vivo, by the interaction of DCs with stromal and other inflammatory cells present in diseased

21

tissues and by the local balance between hypoxia and other microenvironmental alterations, such as glucose depletion, accumulation of lactate and other metabolic byproducts, and low PH, or inflammatory costimuli.

4 JUVANILE IDIOPATHIC ARTRITHIS

Juvenile idiopathic arthritis (JIA) is a group of disorders characterized by chronic arthritis. JIA is the most common chronic rheumatic illness in children and is a significant cause of shortand long-term disability. It is a clinical diagnosis made in a child less than 16 years of age with arthritis (defined as swelling or limitation of motion of the joint accompanied by heat, pain, or tenderness) for at least 6 weeks’ duration with other identifiable causes of arthritis excluded. The incidence of JIA ranges from 1 to 22 per 100,000 with a prevalence of 8 to 150 per 100,000 (Table 1) [47].

Tab.1 Frequency and sex distribution of Juvenile Idiopathic Arthritis

From Lancet 2007,369: 767–78

4.1 JIA classification

Juvenile idiopathic arthritis classification identifies subtypes, many of which seem to represent different diseases characterised by distinct methods of presentation, clinical features, and, in some cases, genetic background.

22

 Systemic juvenile idiopathic arthritis

It is characterized by prominent systemic features, such as fever, rash, and serositis, and is much like adult-onset Still’s disease. Some patients (5-8%), affected by this pathology develop a severe complication as a particular form of hemophagocytic lymphohistiocytosis known as macrophage activation syndrome. The syndrome is characterised by the sudden onset of sustained fever, pancytopenia, hepatosplenomegaly, liver insufficiency, coagulopathy with haemorrhagic manifestations, and neurological symptoms. Laboratory features include raised triglyceride concentrations, low sodium concentrations, and pronounced increate ferritin concentrations. The manifestation of active phagocytosis of haemopoietic cells by macrophages in the bone marrow is common [47]. Gene expression patterns in systemic juvenile idiopathic arthritis are different from those of other subtypes, and include up-regulation of genes related to innate immunity and complement systems, and of a group

of mostly haemopoietic genes

.

 Rheumatoid factor-positive polyarthritis

It is the same disease as adult rheumatoid factor (RF)-positive RA. However, at variance with adult age, it represents only a small proportion (3- 5%) of all JIA cases. For prognosis, it has been reported that children with polyarticular JIA spend the majority of their follow-up with active disease,because children who were RF-positive and with early radiographic evidence of joint damage tended to have the most active disease. Improving outcomes for these subgroups may be an important goal for prospective study [47].

 Oligoarthritis

Oligoarthritis is defined as an arthritis that affects four or fewer joints during the fi rst 6 months of disease. This form, which is typical of children and is not seen in adults, is characterised by asymmetric arthritis, early onset (before 6 years of age), female predilection, high frequency of positive ANAs, and high

23

risk of iridocyclitis (a chronic, nongranulomatous, anterior uveitis that affects the iris and ciliary body and can cause severe visual impairment). The homogeneity of this subgroup of patients is shown by a strong association

with some HLA alleles [47]

.

 Rheumatoid factor-negative polyarthritis

It is a heterogeneous JIA category is defined as an arthritis that affects five or more joints during the fi rst 6 months of disease in the absence of IgM RF This disease is less well defi ned than RF-positive polyarthritis and is probably the most heterogeneous subtype. At least two distinct subsets can be identified: i) a form similar to adult onset RF negative RA and characterized by symmetric synovitis of large and small joints, onset in school age, and negative ANA and ii) a form that resembles oligoarthritis, except for the number of joints affected in the first 6 months of disease [47].

 Enthesitis-related arthritis

This type of JIA is characterized by development of enthesitis in addition to

arthritis. It represents a form of undifferentiated spondyloarthropathy. It has

been shown that patients are HLA-B27-positive and a sizable proportion of them develop the involvement of sacroiliac joints during disease course [47].

 Psoriatic arthritis

The diagnosis of juvenile psoriatic arthritis needs the simultaneous presence of arthritis and a typical psoriatic rash or, if a rash is absent, the presence of arthritis and any two of the following: family history of psoriasis in a first-degree relative,dactylitis (swelling of one or more fi ngers that extends beyond the joint margins),and nail pitting. [47]

 Undifferentiated arthritis

It is by definition heterogeneous as it includes patients who do not fulfill inclusion criteria for any category or fulfill the criteria for more than one category [47]

24

4.2 JIA immunopathogenesis

The pathogenetic mechanisms of JIA are still poorly understood, but seem to be related to several immunological abnormalities, many of which are shared by the different JIA subtypes and are similar to those shown in rheumatoid arthritis of the adult (RA), whereas others are more specific for some disease subtypes [47;48]. All forms of JIA are characterized by abundant and persistent synovial joint inflammation and uncontrolled proliferation of activated fibroblasts, which result in pronounced hyperplasia of the synovial lining layer, leading to pannus formation and progressive proteolytic erosion of cartilage and bone [49], with consequent growth impairment,

joint destruction, and skeletal anomalies [47].The synovium shows pronounced

hyperplasia of the lining layer and an exuberant infiltration of the sub-lining layer with mononuclear cells, including T cells, B cells, macrophages, dendritic cells, and plasma cells [50] The autoreactive immune response in juvenile idiopathic arthritis is assumed to be initially triggered by an adaptive (T cell or B cell) response towards a self-antigen. This assumption is underscored by the fact that joint inflammation in juvenile idiopathic arthritis is characterized by selective accumulation of activated memory T cells in the synovium, which are clustered around DCs in addition to macrophages. However, soon after the initial autoreactive insult, almost all cells belonging to the innate or adaptative immune system are recruited and retained at inflamed synovial sites, taking part in the immune response (Figure 6).

25

The chemokine network in JIA is largely unknown and, chemokine expression was reported to be increased at sites of joint inflammation and in the synovial fluid of JIA patients. High levels of CCL5, CCL3, CXCL10, CCL20 were detected in synovial fluid compared to serum of patients which may contribute to the recruitment of

inflammatory cells expressing CCR5 and CXCR3, predominantly

monocytes/macrophages and memory T cells, to the joint [51]. Continuous stimulation

of T and B lymphocytes by APC cause chronicization of inflammation and release of

proangiogenetic and inflammatory factors. The Endothelil Cells (EC)-specific

mitogen, VEGF and osteopontin (OPN), plays a central role in driving blood vessel neo-formation in the synovium via stimulation of EC survival, proliferation, and

chemotaxis, and in monocytic and T cell recruitment/activation. Secretion of

auto-antibodies and several inflammatory cytokines such as tumor necrosis factor (TNF)-α and interleukin (IL-1, -6, and -8) and IFN-γ also happens after contact with APC and T lymphocytes that become activated and induce the response of B lymphocytes and monocytes/macrophages [47]. This in turn stimulates the activation of fibroblasts-like synoviocytes and chondrocytes and the release of cartilage-degrading enzymes like matrix metalloproteinases [49] promoting tissues disruption. (Figure 7)

26

Fig.7 Schematic representation of the role of chemokines in the joint of JIA patients

Modified from FEBS J. 2008,275(18): 4448-55

4.3 Role of hypoxia in JIA

The reciprocal relationship between inflammation and angiogenesis in the synovial tissue, each potentiating the progression of the other, is crucial for perpetuation of the disease [52]. The inflammatory response persists as a consequence of extensive blood vessel neoformation, which allows the continuous migration and infiltration of mononuclear leukocytes in synovial tissue (ST) and fluid (SF) of affected joints and, in turn, is sustained by the local secretion of various proangiogenic mediators by both inflammatory and stromal cells, resulting in the chronicity of inflammation [48;52]. Several proangiogenic factors, including the vascular endothelial growth factor (VEGF) and the cytokine/extracellular matrix protein, osteopontin (OPN), are markedly upregulated in the ST and SF of inflamed joints from patients affected by different JIA subtypes, compared with healthy specimens, showing a good correlation

27

angiogenesis might lead to the formation of a dysfunctional vascular network due to vigorous proliferation of immature, structurally defective vessels [54], resulting in reduced blood flow and consequent impaired tissue oxygenation. The formation of a hyperplastic inflammatory mass might also impair oxygen perfusion as a result of increased diffusion distances between proliferating synovial cells and their supplying blood vessels, leading to areas of synovial hypoxia, in particular in the lining layer. Oxygen perfusion might be further compromised by synovial swelling and effusion, which result in increased intraarticular pressure exceeding synovial capillary perfusion

pressure, and are exacerbated by movement of the affected joint. Finally, increased O2

consumption and metabolic demand due to the enhanced proliferative activity of synovial fibroblasts and inflammatory cell recruitment, which are not balanced by

sufficient O2 supply, might further aggravate tissue hypoxia. Indeed, two recent

studies documented the hypoxic nature of the JIA inflamed synovium [2;55]. pO2

measurement in the SF of patients belonging to different JIA subtypes, collected at the

time of knee arthrocentesis and intra-articular steroid injection, demonstrated O2 levels

ranging between 7 and 40 mmHg (mean value <30 mmHg), which are in the range of those recorded in the synovium of adult inflammatory arthropathies (mean value <27

mmHg, range 6–50 mmHg), and profoundly reduced compared with physiological O2

concentrations in healthy joints (mean value <74 mmHg, range 69–89 mmHg). Furthermore, strong positivity for both HIF-1a and HIF-2a subunits was detected by immunohistochemistry in the synovial lining and sublining layers of tissue specimens obtained at synovectomy of JIA patients, confirming the presence of a hypoxic environment (Figure 8) [2;55]. These findings raise questions about the consequences of the hypoxic synovial milieu in terms of JIA pathogenesis.

28

Fig.8 Hypoxic tissue in JIA patients

From Journal of Clinical Rheumatology & Musculoskeletal Medicine 2010

4.4 Role of DCs in JIA

Role of DC in JIA it not yet clearly understood even if some studies suggest that these cells can be an important ring in the pathologic chain. Different authors have demonstrated the presence of CD1a+ monocytes-derived DCs in the synovium. This DCs subset has been associated with the presentation of human cartilage glycoprotein 39 (HCgp39, or YKL-40), an RA candidate autoantigen, both in vivo and in vitro [56]. DCs may also be involved in the process of joint destruction in RA. The receptor activator of the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB), RANK/RANK ligand (RANKL) is of particular interest in RA synovial tissue. iDCs CD1a+ expressed both RANK and RANKL,RANK/RANKL interactions could be important for DC-T cell interactions during the inflammatory process [56]. Several studies using in vitro generated RA monocytes -derived DC (CD1a+ DCs) demonstrated that in contrast to normal peripheral blood DCs- precursors and monocytes derived-DC from RA synovial fluid, were resistant to the immunosuppressive effect of IL-10 in vitro. These data imply that synovial CD1a+ DCs maintain their inflammatory potential, contributing in this way to RA synovial

29

inflammation. Interestingly, the existence of DCs progenitors and DCs growth factors in RA synovial fluid supports the concept that this environment may be a reservoir for joint-associated DCs and reveals a compelling mechanism for the amplification and perpetuation of DCs-driven responses in the RA joint, including inflammatory-type Th1 responses. Furthermore, monocytes-derived-mDCs from RA patients showed a markedly increased production of IL-1, IL-6, TNF-α, and IL-10 compared with mDCs from healthy controls. Despite recent data indicating that DC may be over-represented in RA, relatively little is known about the mechanisms promoting differentiation along specific DC pathways within distinct joint microenvironments [57]. It has been reported that human iDCs could differentiate into functional osteoclasts in the presence of M-CSF and RANKL. Importantly, this process was greatly enhanced by RA synovial fluid and involved pro-inflammatory cytokines such as IL-1 or TNF-a, as well as components of the extracellular matrix as hyaluronic acid. In addition, a

unique natural killer (NK) cell subset (CD3-CD56bright) that accumulates in lymph

nodes and chronically inflamed tissues triggered CD14+ monocytes to differentiate into potent Th1-promoting DC. This process required direct contact of monocytes with NK cells and it was mediated by GM-CSF and CD154 derived from NK cells. NK cells might play a role in the maintenance of Th1-mediated inflammatory diseases such as RA by providing a local milieu for monocytes to differentiate into DC. Moreover it has been observed that RA synovial fibroblasts or their soluble factors present in conditioning medium are able to induce the generation of CD1a+ DCs from peripheral blood monocytes [56]. Taking together, these data suggest an important role for the synovial microenvironment in the commitment of monocyte-derived cells and might support the generation of the synovial CD1a+ DCs pool.

30

OBJECTIVE OF THE STUDY

DCs are central to the orchestration of immunity and the maintainance of self-tolerance. DC development and functions are regulated by signals generated in the local microenvironment, and deregulated DC responses may result in amplification of inflammation, loss of tolerance, or immune escape. DC generation from monocytic

precursors recruited at sites of inflammation, tissue damage, or neoplasia occurs under

condition of hypoxia. Understanding DC biology in low O2 environment may open

new therapeutic opportunities for inflammation and cancer. In our laboratory we defined the hypoxic transcriptome of immature (i) and mature (m)DCs generated from human monocytes. We presented data pointing to a role for hypoxia in differentially reprogramming DCs depending on their maturation stage. Specifically, hypoxia promotes the onset of a migratory phenotype in iDCs through the upregulation of chemokine receptors and an inflammatory state in mDCs by increasing production of proinflammatory, Th1-priming chemokines/cytokines. Interestingly, hypoxia induces profound changes in the expression of a significant cluster of genes coding for

immune-related cell surface receptors in both cell subsets. In this study we focus on

IRS receptors, important regulator of DC maturation, immunogenicity, cooperation with T cells, and pro-inflammatory cytokine production; their overexpression is implicated in the pathogenesis of chronic inflammation, allergic inflammatory disorders, and autoimmune diseases. Understanding the DCs hypoxic gene expression pattern of immune-related receptors could be important to identificate potential biomarkers measurable in the tissue.

31

MATERIAL AND METHODS

DC generation and culture

Monocytes were isolated from venous blood of voluntary healthy donors from the blood bank under an Institutional Review Board-approved protocol, by centrifugation over a Ficoll cushion (Histopaque, Sigma), followed by MACS magnetic bead separation (Miltenyi Biotec) at a purity of >93% CD14+. To generate iDCs,

monocytes were plated into six-well culture plates (1.5 × 106 cells/mL) (BD Falcon) in

RPMI 1640 (Euroclone) supplemented with 10% heat-inactivated FCS (HyClone) and

incubated for 4 days under normoxic (20% O2) or hypoxic (1% O2) conditions, in the

presence of GM-CSF and IL-4 (both 100 ng/mL), as detailed [32;35]. A cocktail of proinflammatory mediators containing tumor necrosis factor-α (TNF-α 50 ng/mL), IL-1β (50 ng/mL), IL-6 (10 ng/mL), and prostaglandin E2 (1mM) was added for the last 48 hours to induce DC maturation (mDCs). Hypoxic conditions were obtained by

culturing cells in an anaerobic work-station incubator (Ruskinn InVivoO2) flushed

with a mixture of 1% O2/5% CO2/94% N2.

SFMC isolation

Synovial fluid (SF) samples were obtained at the time of therapeutic knee arthrocentesis from 8 children affected by oligoarticular JIA and collected into

sodium-heparin tubes under vacuum. pO2 levels in SF samples were monitored to

confirm hypoxic conditions. Paired peripheral blood samples and peripheral blood from 5 age-matched control subjects undergoing venipuncture for minor orthopedic procedures were obtained on the occasion of routine venipuncture and collected as for SF. Informed consent obtained according to the procedure approved by the Gaslini’s Ethical Committee. Specimens were centrifuged to prepare cell-free SF and plasma and separated by Ficoll to isolate mononuclear cells (SF mononuclear cells –SFMCs- and peripheral blood mononuclear cells -PBMCs-). SF-derived samples were handled in the anaerobic incubator to prevent cell reoxygenation, as detailed. [32;35]

32

Cytokines and antibodies

Human recombinant GM-CSF, IL-4, TNF-α, IL-1β, and IL-6 were from PeproTech,prostaglandin E2 was from SigmaAldrich. Echinomycin was from Alexis Biochemical. Monoclonal antibodies (mAbs) used for fluorescence-activated cell sorter include: anti-CD83-fluorescein isothiocyanate (FITC) or phycoerythrin (PE)-Cy5,CD86-FITC or -PE (BD Biosciences PharMingen), TREM-1-PE, antiCXCR4FITC or PE (Biolegend), antiCCR7FITC (R&D Systems) or -allophycocyanin (Biolegend), anti-CD1a-FITC or --allophycocyanin (BD), anti- CD141-allophycocyanin (Miltenyi-Biotec) , anti-HLA-DR-PE (BD), and anti-CD40-PE (Immunotech). Proper isotype-matched control Abs (BioLegend, Dako ) were used. Western blot include: mouse anti–human TREM-1 (R&D Systems), mouse anti– human HIF-1_ (BD-Biosciences), rabbit anti–human phospho (p)-ERK)-1, p-Akt, and p-IkBα (Cell Signaling Technology), mouse anti–human DAP12, rabbit anti–human ERK-1, Akt, IkBα, and -actin (Santa Cruz Biotechnology)

Flow cytometry

Flow cytometry was performed as described [32]. Cells resuspended with phosphate-buffered saline supplemented with 0.2% bovine serum albumin, 0.01% NaN3 were incubated with fluorochrome-conjugated mAbs for 30 minutes at 4°C, after blocking nonspecific sites with rabbit IgG (Sigma-Aldrich). Fluorescence was quantitated on a FACSCalibur flow cytometer equipped with CellQuest software Version 2002 (BD Biosciences). Cells were gated according to their light-scatter properties to exclude cell debris.

RNA isolation and GeneChip hybridization

Total RNA was purified from different donor-derived iDCs and mDcs using the QIAGEN RNeasy MiniKit and reverse-transcribed into double-stranded cDNA on a GeneAmp PCR System 2700 thermal cycler (Applied Biosystems) using the one-cycle

33

cDNA synthesis kit (Affymetrix). cDNA derived from 3 donors was purified and biotin labeled with the IVT-expressed kit (Affymetrix), as described.21 Fragmented cRNA was hybridized to Affymetrix HG-U133 plus 2.0 arrays (Genopolis Corporation) containing 54 000 probe sets coding for 38 500 genes,chips were stained with streptavidin-phycoerythrin (Invitrogen) and scanned using an Affymetrix GeneChip Scanner 3000, as described [33] Data capturing was conducted with standard Affymetrix analysis software algorithms (Microarray Suite 5.0). Comparative analysis of hypoxic relative to normoxic expression profiles was carried out with GeneSpring Expression Analysis Software Gx9.0 (Silicon Genetics), and expression data were normalized using “per chip normalization” and “per gene normalization” algorithms. Fold change was calculated as the ratio between the average expression level under hypoxia and normoxia. We selected a modulated gene list of 2-fold induction/inhibition with a false discovery rate of 0%. The significante of gene expression differences between the 2 experimental conditions was calculated using the Mann-Whitney U test. Only genes that passed the test at a confidence level of 95% (P <.05) were considered significant. A complete dataset for each microarray experiment was logged in the Gene Expression Omnibus public repository at National Center for Biotechnology Information (www.ncbi.nlm.nih.gov/geo,accession no. GSE22282). Gene Ontology (GO) data mining for biologic process at level 1 was conducted online using the Database for Annotation, Visualization and Integrated Discovery (DAVID) software Version 2008 (www. david.niaid.nih.gov). HRE consensus elements consisting of a 4-nt core (CGTG) flanked by degenerated sequences ((T G C)(A G)(CGTG)(C G A)(G C T)(G T C) (C T G) were mapped in the promoter regions of genes represented in the chip, as detailed [32].

Real-time RT-PCR

Real-time PCR (qRT-PCR) was performed on a 7500 Real Time PCR System (Applied), using SYBR Green PCR Master Mix and sense/antisense oligonucleotide primers designed using Primer- 3 software from sequences in the GenBank and

34

obtained from TIBMolbiol (Genova) or from Quiagen (RSP18), as detailed [33]. Expression data were normalized on the values obtained in parallel for three reference genes (ARPC1B, RPS18, RPS19), selected among those not affected by hypoxia in the Affymetrix analysis, using the Bestkeeper software, and relative expression values were calculated using Q-gene software, as detailed [58].

Cross-linking of TREM-1 cells

Twelve-well flat-bottom tissue culture plates (Corning Life Sciences) precoated with 10 ug/mL of agonist anti-TREM-1 mAb or control IgG1 were incubated overnight at

37°C before seeding 8X105 H-iDCs or mDCs/ well/mL of RPMI 1640 without

cytokines. Plates were briefly spun at 130g to engage TREM-1. After 24-hour stimulation under hypoxia, supernatants were harvested and tested for cytokine/chemokine content by enzymelinked immunosorbent assay (ELISA) and H-iDCs were used to stimulate allogeneic T cells. In a set of experiments, mDCs were plated in medium without fetal calf serum, and plates were centrifuged and incubated for 20 minutes at 37°C under hypoxic conditions. ERK1, Akt, and IkBα phosphorylation was then assessed byWestern blot.

Mixed leukocyte reaction

T cells were purified by negative selection from peripheral blood mononuclear cells using a PanT kit (Miltenyi Biotec). Total of 1 × 106/mL T cells were cultured with allogeneic H-iDCs previously triggered with anti-TREM-1 mAb or control HLA-I at a 20:1 T:DC ratio. After 4 days, supernatants were collected to measurereleased cytokines by ELISA. To assess proliferation, T cells were pulsed with 1 μCi of 3H-thymidine (Perkin Elmer) for a further 16 h culture, and 3H-3H-thymidine incorporation was quantified using a TopCount microplates scintillation counter (Canberra Packard).

35

Western blot analysis and immunoprecipitation

Total protein extracts were prepared as described [35], subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyvinylidene difluoride membranes (Millipore), and probed with specified Abs. In a few experiments, lysates were subjected to immunoprecipitation with 10 µg/mL anti TREM-1 mAb or control IgG1 ON at 4°C, and protein G-Sepharose 4B (GE Healthcare) for 45 minutes at 4°C. Precipitates were separated by SDS-PAGE and immunoblotted with anti-DAP12 mAb. Chemiluminescence detection was carried out with peroxidase-conjugated goat antirabbit and antimouse Abs using an ECL kit (Pierce Chemical).

ELISA

Conditioned medium (CM) was replaced on day 3 or 4 of mDC generation with fresh medium supplemented with cytokines for 24 hours and tested for soluble TREM-1 (sTREM-1) content by ELISA (R&D Systems) after an additional 24-hour culture. sTREM-1 was also quantified in SF and plasma samples. TNF-α, IL-6, IL-12p70, IL 10, IL-8, CCL4, and CCL5 were measured in CM from mDCs triggered with anti-TREM-1 mAb or control mAb by specific ELISA (R&D Systems). Data were analyzed with the GraphPad Prism-5 Software.

Statistical analysis

Data are the mean plus or minus SE of 3 independent experiments, unless differently specified. The Student t test was used to determine significante of results (P< .05). sTREM-1 concentrations in SF and plasma specimens were evaluated by theWilcoxon rank test (P <.05 statistically significant).

36

RESULTS

Regulation of DC cell-surface immune-related receptor repertoire by hypoxia

DCs can integrate multiple stimulatory and inhibitory signals present in the microenvironment through a defined repertoire of antigen-presenting molecules, pattern recognition and immunoregulatory receptors [11;12;15;22;59]. Immature dendritic cells (iDCs) and mature dendritic cells (mDCs) were generated by culturing human monocytes under normoxic and hypoxic conditions. H-DC transcriptional profile was then assessed by microarray analysis. Pairwise comparison between datasets from normoxic and hypoxic samples revealed differential modulation of a large number of transcripts. H-iDCs expressed higher levels of genes coding for both classical and nonclassical antigen-presenting receptors, including MHC class I and II molecules and tetraspanin family members (CD37, CD53, CD9) that associate with and are implicated in MHC-peptide assembly [60]. We also observed hypoxia-dependent expression of genes coding for immunoregulatory signaling receptors implicated in the regulation of DC maturation/polarization, inflammatory and immune functions [22;60]. The most relevant are: SLAM family member-9 (SLAMF9), low-affinity IgE receptor, FcεRII (CD23A), and IgG receptors, FcγRIIA/B (CD32), CD69, CD58, natural cytotoxicity triggering receptor 3 (LST1), TREM-1, leukocyte Ig-like receptor 9 (LIR9), and leukocyte membrane Ag (CMRF- 35H), whereas expression of CD33 antigen-like 3 (SIGLEC15) and SLAMF1, among others, was downregulated. The hypoxic transcriptome was also characterized by the differential modulation of genes encoding PRR critical to host defense [59] and scavenger receptors implicated in the regulation of fatty acid and/or cholesterol uptake/transport [61], such as CD180, G-protein-coupledreceptor (GPCR) 132 (G2A), TLR 8, CD14, MD-1, and CD36, that were upregulated, and CD163 Ag, TLR5, and MD-2, that were downregulated. In H-mDCS we observed the expression of several pattern recognition receptors critical to host defense, such as CD180, various complement receptor components, Toll-like receptors 1 and 2, C-type lectin receptors CLEC-2D, -2B, -7A, and macrophage receptor with collagenous structure. Of interest is also the upregulation of scavenger

37

receptors implicated in the regulation of fatty acid and/or cholesterol uptake/transport, such as thrombospondin receptor (CD36) and apolipoprotein B48 receptor. A set of genes coding for costimulatory and adhesion/homing molecole was also up-regulated, such as several integrin family members, L1 cell adhesion molecule (CD171), neuropilin-1,ADAM metallopeptidase 8 (CD156), platelet/endothelial cell adhesion molecule (CD31), and lymphocyte adhesion molecule-1. Other hypoxia inducible genes coded for immunoregulatory receptors, the most relevant of which are: Ig-Fc receptors, TREM-1, SLAM family member-9, blood dendritic cell antigen (CD141), leukocyte immunoglobulin-like receptors A1 and A2, and semaphorins 4B and 4D. Interestingly, only a few of the observed hypoxia-induced changes in gene expression were shared between H-iDCs and H-mDCs or monocytic precursors exposed to acute hypoxia [33], whereas most of the genes are differentially expressed in the populations examined. We conclude that hypoxia can selectively modulate the gene expression pattern of immune-related receptors in monocytic lineage cells depending on their differentiation/maturation stage.

TREM-1 is expressed in hypoxic iDCs

Among hypoxia-responsive genes, we identified TREM-1 as a common hypoxia molecular target in iDCs, mDCs, and primary monocytes, pointing to a critical role of this molecule in the MP response to hypoxia. TREM-1 was previously reported to be constitutively expressed in blood monocytes and completely downregulated during monocyte differentiation into DCs under normoxic conditions [62;63]. Hence, we were interested in investigating the functional significance of TREM-1 expression in iDCs generated under hypoxia. To quantify the magnitude of hypoxia effects and address the issue of donor-to-donor variability, we evaluated TREM-1 expression in iDCs generated from seven independent donors under normoxic and hypoxic conditions. As determined by flow cytometry (Table 2), H-iDCs expressed the DC marker, CD1a, and displayed an activated phenotype characterized by higher surface

38

levels of CD80 and CD86 costimulatory molecules and the chemokine receptor, CXCR4, compared to iDCs, in agreement with previous data [35]

Tab.2 Membrane marker expression by iDCs and H-iDCs.

a)The surface expression of all markers was determined by flow cytometry after 4 days’ culture under hypoxic and normoxic conditions.

b)Data are shown as the mean percentage of positive cells ± SEM from six donors.

c)p < 0.05, values significantly different from iDCs. Student’s t-test.

TREM-1 transcript levels were compared in H-iDCs and iDCs by qRT-PCR. Expression of CAXII was ossesse in parallel as an index of response to hypoxia [67]. As depicted in Figure 9, TREM-1 mRNA expression was significantly and consistently higher in H-iDCs than in iDCs from all tested samples, paralleling CAXII induction, although with some differences among individual donors ranging from 10- to 21-fold, thus confirming gene inducibility in H-iDCs. TREM-1 surface expression was then measured by flow cytometry in seven individual samples at day 4 of culture.