Calcineurin inhibitors reduce NFAT‐dependent expression of antifungal pentraxin‐3 by human monocytes

Abstract Calcineurin (CN) inhibitors are effective clinical immunosuppressants but leave patients vulnerable to potentially fatal fungal infections. This study tested the hypothesis that CN inhibition interferes with antifungal immune defenses mediated by monocytes. We showed that NFAT is expressed by human monocytes, and is activated by exposure to fungal ligands. We confirmed that NFAT translocation potently activated target gene transcription using a human monocytic reporter cell line. Inhibition of CN‐NFAT by cyclosporine A significantly reduced monocyte production of TNF‐α, IL‐10, and MCP‐1 proteins in response to pattern recognition receptor ligands as well as to Aspergillus fumigatus conidia. Moreover, we revealed that human monocytes express the antifungal protein pentraxin‐3 under control of NFAT. In conclusion, clinical CN inhibitors have the potential to interfere with the novel NFAT‐dependent pentraxin‐3 pathway as well as antifungal cytokine production in human monocytes, thereby impeding monocyte‐mediated defenses against fungal infection in immune‐suppressed patients.


INTRODUCTION
Immuno-suppressive drugs, such as cyclosporine A (CsA) and tacrolimus, are effective in preventing the rejection of solid organ transplants, 1 reducing graft-versus-host disease (GvHD) after hematopoietic stem cell transplantation (HSCT), 2,3 and treating autoimmune disorders. 4 However, by down-regulating the immune response these agents also leave patients susceptible to infections, 5 particularly by fungal pathogens. Severe/invasive fungal infections occur in ∼3% of post-transplant patients per year, leading to death in almost 40% of affected patients. [6][7][8] Thus there is an urgent need to understand the interactions of immune-suppressive drugs with key components of the host's immune system, both in terms of how they Abbreviations: CN, calcineurin; CNS, Conserved noncoding sequence; CsA, cyclosporine A; DCs, dendritic cells; GvHD, Graft-versus-host disease; HSCT, hematopoietic stem cell transplantation; PRRs, pattern recognition receptors; PTX-3, pentraxin-3; ROI, region of interest; UTR, untranslated region.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. (DCs) and macrophages, which together cooperate as APCs to initiate, shape, and maintain protective immune responses to pathogens. 9 Initial studies in murine DCs identified NFAT as the key regulator of DC IL-2 expression, 10,11 which is crucial to modulate Th17 responses: mice lacking IL-2 in DCs express higher levels of IL-17 in their lungs, leading to a pathologic Th17 response that results in exacerbated infection by the fungal pathogen Aspergillus fumigatus (A. fumigatus). 12 A similarly important role for IL-2 expressed in myeloid cells was shown as crucial to maintain gut homeostasis. 13 Alongside NFAT regulation of IL-2, genome-wide mapping of NFAT1 binding sites in murine DCs activated by the fungal beta-glucan, curdlan, identified a panel of inflammatory cytokines, including TNF, IL-10, and GM-CSF, that is regulated under the control of the NFAT1 binding promoter. 14 Similarly, murine macrophages impaired in CN exhibit reduced responses to LPS both in vitro and in vivo, producing lower levels of IL-12 and IL-23, 15,16 and defects in fungal killing that were associated with increased mortality from pulmonary aspergillosis. 17 Recent evidence also confirms a clear role for CN-NFAT in human macrophages in intracellular restriction of A. fumigatus infection in vitro. 18,19 NFAT activation in DCs and macrophages can occur following ligation of several pattern recognition receptors (PRRs), including the TLR4-CD14 complex, which detects bacterial LPS, with the crucial role of CD14 in inducing the NFAT activation, 20 TLR9, 21 which binds microbial DNA, and dectin-1, 14,22 which recognizes fungal cell wall glucans.
Of these receptors, dectin-1 is the most crucial during fungal infections: human HSCT patients carrying a dectin-1 polymorphism with lower receptor activity face a higher risk of aspergillosis, which correlates with reduced expression of IFN-, IL-10, and IL-17A by their peripheral blood monocytes. 23 Dectin-1 ligation is known to trigger expression of number of antifungal cytokines including IL-1 , IL-6, IL-23 24 , whereas CN inhibition in murine models has further identified IL-2, 12 IL-10, 25 TNF-21 , and more recently pentraxin-3 (PTX-3) 26 as having notable effects on A. fumigatus and Candida albicans susceptibility. Of particular interest is the emerging role of PTX-3, a soluble PRR with critical functions in antifungal immunity: impaired PTX-3 production directly increases susceptibility to A. fumigatus infection in mouse models 27 and human HSCT patients. 28 Within myeloid cells, PTX-3 is produced by both mononuclear phagocytes responding to inflammatory or infectious stimuli 29 and neutrophils, which produce PTX-3 during granulocytic differentiation and store it in a form of ready-to use cytoplasmic granules. 30 Once released from cells, PTX-3 recognizes and binds fungal conidia, which then activate neutrophil phagocytic activity and complement to clear the infection. 31 PTX-3 expression has previously been associated with several regulatory molecules and transcription factors including Pu1, SP1, NF-B, AP-1, and NF-IL-6. 32 We also recently reported that CN-deficient murine DCs and LysM + myeloid cells exhibit significantly reduced Ptx-3 expression, 26 suggesting a possible link between CN-NFAT signaling and Ptx-3 expression in murine myeloid cells.
Many studies have now been conducted on the roles of NFAT in murine DCs in particular, but there is a dearth of knowledge on the potential roles of the CN-NFAT pathway in other myeloid cell types, and in human cells in particular. Given the emerging data on the significance of the monocytic compartment in determining human susceptibility to aspergillosis, 23 we aimed to establish whether the NFAT pathway is active in human monocytes; the effects of CN inhibition on these cells; and whether there was a direct link between PRR ligands, NFAT, and production of antimicrobial molecules by human monocytes. We reveal that human monocytes do express functional NFAT family members and that inhibition of CN-NFAT signaling profoundly affects the expression of key inflammatory cytokines in these cells. Moreover, PRR ligation as well as stimulation with A. fumigatus conidia induce PTX-3 expression in human monocytes, which is significantly decreased by CN-NFAT inhibition. In the context of these findings the fact that CsA treated monocytes are significantly impaired in PTX-3 expression is of high clinical relevance considering that A. fumigatus infections are serious life-threatening complications in patients undergoing bone marrow transplantations.

Isolation and stimulation of human blood monocytes
Monocytes from healthy donors were isolated from fresh buffy-coats

Immunofluorescence labelling of monocytes
Monocytes or their sorted subsets were seeded into -Slide VI 0.4 (IBIDI) at a concentration of 0.7-1 × 10 6 /ml and incubated for

Evaluation of PTX-3 expression in monocytes
Monocytes labelled for surface marker expression were fixed and intracellularly labelled for PTX-3 (ab125007, Abcam) using an Intracellular Fixation and Permeabilization Buffer Set (eBiosciences) and secondary antibody AF488 donkey-anti-rabbit (Thermo Fisher Scientific). Sample acquisition was performed using a FACS Canto II (BD Biosciences, Franklin Lakes, NJ, USA) and the data were analyzed using FlowJo v.10.

RNA extraction and quantitative real-time PCR
Total cellular RNA was extracted from monocytes or monocyte subsets using Trizol LS reagent (Thermo Fisher Scientific) following manufacturer's recommendations. Samples were then centrifuged at 12,000 ×g for 15 min at 4 • C. The upper aqueous phase was collected and mixed with an equal volume of 70% ethanol and mixture was transferred to RNeasy spin column (Qiagen, Hilden, Germany).
Subsequent RNA purification was performed as recommended by manufacturer. The RNA concentration and quality were determined spectrophotometrically using Nanodrop (Agilent, Santa Clara, CA, USA). RNA was immediately transcribed into cDNA using the highcapacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific).

ELISA
The commercial DuoSet ELISA (R&D Systems, Minneapolis, MN, USA) was used for detection of PTX-3, TNF-, MCP-1, and IL-6 in supernatants. ELISA were performed exactly as recommended by the manufacturer.
The THP-1 NFAT reporter line was maintained in RPMI 1640 (Lonza) before subsequent stimulation. The cells were pre-treated with CsA to inhibit NFAT translocation (1 g/ml, Cell Signaling Technologies) for at least 30 min before stimulation with ionomycin (1 g/ml, Sigma Aldrich). Luciferase activity was detected 6 h after triggering using the ONE-Glo Luciferase Assay System (Promega, Madison, WI, USA).

NFAT translocation in THP-1 and monocytes and cytospin preparation
THP-1 cells were maintained in RPMI 1640 with 10% FBS (Sigma Aldrich) and peripheral blood monocytes were cultured in X-VIVO 15 (Lonza) without any supplementation at 37 • C in 5% CO 2 atmosphere.

Statistics
Graphpad Prism software v.6 was used for statistical analysis. Data were tested for normal distribution and parametric or nonparametric statistical tests were applied as appropriate. Statistical tests used are specified in the figure legends. P values <0.05 were considered statistically significant.    (Fig. 1A). We then confirmed the expression of the NFAT-activating PRRs, TLR4, and dectin-1 using flow cytometry (Fig. 1B), and the gene expression of the NFAT family member genes (Fig. 1C), with T cells and HELA cells as positive and negative controls for NFAT expression, respectively. Monocyte subsets expressed comparable levels of TLR4 and dectin-1 (Fig. 1B). We also detected transcripts for NFAT1, NFAT2, and NFAT4 in all mono-cyte subsets (Fig. 1C), but transcripts for NFAT3 were only present in T cells. NFAT1 was the most abundantly expressed, and we confirmed its expression at the protein level by fluorescence microscopy (Fig. 1D) and flow cytometry (Fig. 1E).

NFAT expression and translocation in human myeloid cells
The activation of target gene transcription by NFAT occurs downstream of PRR ligation, which induces calcium flux that activates CN. Treatment of human monocytes with either the fungal ligands curdlan, zymosan, and Pam3CSK4 or ionomycin (as a positive control of calcium influx) resulted in increased calcium flux ( Fig. 2A).
Further quantitative (Fig. 2B) and qualitative (Fig. 2C) analyses confirmed the translocation of NFAT1 to the nucleus after stimulation with Pam3CSK4 and zymosan or zymosan with ionomycin. NFAT translocation was specifically inhibited by CsA in both conditions ( Fig. 2B and C).
To directly assess the ability of NFAT to induce gene transcription following nuclear translocation in human myeloid cells, we established a NFAT luciferase reporter using the human monocytic THP-1 leukemia cell line. In this cell line, NFAT activation and translocation leads to initiation of luciferase gene expression through an NFAT binding site in the luciferase promoter. We measured a significant increase in luciferase activity following treatment of THP-1 reporter cells with ionomycin, which was inhibited by CsA treatment (Fig. 3A). We further confirmed this observation by

Expression of PTX-3 in human monocytes is dependent on CN-NFAT signaling
We recently identified decreased PTX-3 expression in murine APCs with impaired CN-NFAT signaling. 26 We thus investigated the regulation of PTX-3 expression in human monocytes and its relationship to NFAT activation. All monocyte subsets expressed intracellular PTX-3 in the steady state (Fig. 5A). We then stimulated total human blood monocytes with LPS or zymosan and assessed PTX-3 expression by flow cytometry and qPCR ( Fig. 5B and C). Although intracellular protein levels of PTX-3 were unaffected by LPS or zymosan stimulation ( Fig. 5B), we detected marked increases in expression at the mRNA level, which were significantly reduced by CsA treatment (Fig. 5C) CsA-dependent PTX-3 down-regulation was calculated in LPS and zymosan treated samples as the ratio between CsA treated and nontreated samples. (D) Secretion of PTX-3 by monocytes stimulated with a variety of PRR ligands, as measured by ELISA. Monocytes were pre-treated or not with CsA and then stimulated with PRR ligands for 4 h. Data are representative of 7-9 donors. Statistical analysis was performed by RM one-way ANOVA followed by Tukey's multiple comparisons test ( * P < 0.05, ** P < 0.01). Data represent the means ± SD and Pam3CSK4 (Fig. 5D). Thus, human monocytes respond to a variety of PRR ligands by increasing transcription and secretion of the antifungal protein PTX-3 in an NFAT-dependent manner.
Because soluble PRR PTX-3 is important in the defense against fungal pathogens, we validated our findings using heat inactivated A. fumigatus conidia. Monocytes responded to A. fumigatus infection by increasing the expression of PTX-3, MCP-1, and TNF-at the mRNA and protein level, which were further reduced by CsA treatment (Fig. 6A and B). Overall, these findings show that an immune response against fungal cell wall components and A.fumigatus conidia trigger PTX-3, MCP-1, and TNF-expression partially in an NFAT-dependent manner.

In silico analysis of PTX-3 sequence homology and NFAT1 binding sites in human and murine PTX-3
The human and mouse PTX-3 gene is localized on chromosome 3 and shares high sequence homology. 41 The promoter region of PTX-3 contains multiple transcription factor binding sites including an NF-B site, binding sites for activator protein 1 (AP-1), AP-2, specificity protein (Sp1), and a gamma IFN activation site 42 ; however, it is unknown whether it also contains a binding site for NFAT. To partially unravel the mechanisms underlying CN-NFAT-regulated PTX-3 expression, we reanalyzed available ChIP-seq data of NFAT1 targets in curdlanactivated murine DCs, 14 revealing four NFAT1 binding sites in the murine Ptx-3 gene (Fig. 7A, ROI1-4). To link these data to our experimental results, we aligned the murine and human sequences of the PTX-3 gene to evaluate sequence homology. Consistently with previously published data, 41 the two sequences shared a high degree of conservation, especially in the region surrounding the PTX-3 5 ′ -UTR (untranslated region; Fig. 7B). Interestingly, this highly conserved region was bound by NFAT1 in curdlan-activated murine DCs (ROI3), 14 suggesting the presence of a putative NFAT1 binding site also in the human sequence. To confirm this hypothesis, we screened both sequences for the presence of a NFAT1 binding site and identified a putative common binding motif in a conserved noncoding sequence (CNS) ∼300 bp upstream PTX-3 5 ′ UTR ( Fig. 7B and C). Interestingly, in the same study 14 Ptx-3 was also differentially expressed in curdlanstimulated murine DCs compared to untreated control; reanalysis of expression data showed that Ptx-3 expression was down-regulated in DCs stimulated with curdlan in the presence of the CN-NFAT inhibitor Tacrolimus/FK506 (Fig. 7D). Indeed, these findings of a putative NFAT factors. 45 Despite the key role of blood monocytes in orchestrating immunity to fungal pathogens, the role of the CN-NFAT pathway has not been studied in these cells.
The importance of understanding the impact of CN inhibitors on susceptibility to fungal infection is clear: pulmonary aspergillosis is a leading cause of mortality in transplant patients. 7 Increased risk of severe/invasive aspergillosis in human HSCT recipients has been linked with impaired dectin-1 signaling 23 and genetic deficiency in PTX-3 28 ; in this study we consider these findings in the context of the blood monocyte compartment. We show that monocytes express the fungal PRR dectin-1, and that ligation of this receptor by fungal cell wall components or heat-inactivated A. fumigatus leads to CN-NFAT dependent expression of multiple cytokines, and increased secretion of PTX-3.
Overall, exposure of monocytes to clinical CN-inhibitor CsA markedly decreased their antifungal responses.
We also noted that human monocytes up-regulated gene expression of PTX-3 in response to the bacterial PRR ligand LPS, and that this process was similarly restricted by CN inhibition. Whereas PTX-3 has been primarily studied for its antifungal role, recent findings have identified it as a diagnostic marker and protective protein during sepsis. 46,47 Immunosuppressive treatment further increases risks of severe sepsis consequences, [48][49][50][51] which is particularly intriguing given the pivotal role of monocytes in sepsis, 45,52  K.B and F.T. contributed equally to this work.