Portal de Periódicos da CAPES

IL-10 overexpression predisposes to invasive aspergillosis by suppressing antifungal immunity

2017; Elsevier BV; Volume: 140; Issue: 3 Linguagem: Inglês

10.1016/j.jaci.2017.02.034

1097-6825

Cristina Cunha, Samuel M. Gonçalves, Cláudio Duarte‐Oliveira, Luís Leite, Katrien Lagrou, António Marques, Carmen B. Lupiañez, Inês Mesquita, Joana Gaifem, Ana Margarida Barbosa, Carlos Pinho Vaz, Rosa T. Branca, Fernando Campilho, Fátima Freitas, Dário Ligeiro, Cornelia Lass‐Flörl, Jürgen Löffler, Manuel Jurado, Margarida Saraiva, Oliver Kurzai, Fernando Rodrigues, António G. Castro, Ricardo Silvestre, Juan Sáinz, Johan Maertens, Egídio Torrado, Ilse D. Jacobsen, João F. Lacerda, António Campos, Agostinho Carvalho,

Infectious Diseases and Mycology

Proinflammatory immune responses are critically required for antimicrobial host defenses; however, excessive inflammation has the potential to damage host tissues thereby paradoxically contributing to the progression of infection. A central negative regulator of inflammatory responses is IL-10, an immunosuppressive cytokine with a wide variety of functions across multiple cell types.1Saraiva M. O'Garra A. The regulation of IL-10 production by immune cells.Nat Rev Immunol. 2010; 10: 170-181Crossref PubMed Scopus (2026) Google Scholar Although the role of IL-10 during infection appears to vary for different microorganisms, a largely detrimental role has been attributed to this cytokine during fungal disease.2Potenza L. Vallerini D. Barozzi P. Riva G. Forghieri F. Beauvais A. et al.Characterization of specific immune responses to different Aspergillus antigens during the course of invasive Aspergillosis in hematologic patients.PLoS One. 2013; 8: e74326Crossref PubMed Scopus (49) Google Scholar Given the variable risk of infection and its outcome among patients with comparable predisposing factors, susceptibility to invasive aspergillosis (IA) is thought to rely largely on genetic predisposition.3Cunha C. Aversa F. Romani L. Carvalho A. Human genetic susceptibility to invasive aspergillosis.PLoS Pathog. 2013; 9: e1003434Crossref PubMed Scopus (54) Google Scholar The initial investigation of genetic variability at the IL10 locus led to the identification of single nucleotide polymorphisms (SNPs) influencing its transcriptional activity; thus, IL-10 may be a reasonable candidate for the genetic regulation of susceptibility to IA in high-risk patients. In a 2-stage, multicenter study involving 413 hematopoietic stem-cell transplantation (HSCT) donor-recipient pairs, we confirmed that SNPs in IL10 are critical regulators of susceptibility to IA. Using a discovery cohort of donor-recipient pairs (see Table E1 in this article's Online Repository at www.jacionline.org), we analyzed 5 haplotype-tagging SNPs in IL10 (see Table E2 in this article's Online Repository at www.jacionline.org) and found that the donor, but not recipient, rs1800896 SNP was associated with an increased risk of IA (Fig 1, A and see Table E3 in this article's Online Repository at www.jacionline.org). The contribution of the GG genotype to the risk of infection was further illustrated on modeling a recessive mode of inheritance (Fig 1, B). In a multivariate model, the donor GG genotype conferred a 2.6-fold increased risk of developing IA after transplantation, and the association test results were further validated in a confirmation case-control study involving patients with similar demographic and clinical characteristics and by a meta-analysis including all enrolled patients (see Table E4 in this article's Online Repository at www.jacionline.org). Furthermore, although no significant differences were observed, the probability of infection-free survival in the discovery set decreased from 88% among patients with the AA genotype to 79% and 75% among subjects carrying the AG or GG genotypes, respectively (Fig 1, C). Using sequence data from the 1000 Genomes Project, we identified all SNPs in linkage disequilibrium (LD) with rs1800896 (see Table E5 in this article's Online Repository at www.jacionline.org), but none of these were exonic. LD around this SNP was limited to the IL10 locus (see Fig E1 in this article's Online Repository at www.jacionline.org), implying that noncoding variation is likely to drive the association with IA. To ascertain whether rs1800896, or a variant in strong LD with it, influenced gene transcription, we monitored IL-10 mRNA and protein expression in PBMCs from healthy blood donors subjected to in vitro infection with Aspergillus fumigatus. We observed striking genotype-specific differences, with PBMCs carrying the GG genotype expressing higher transcript and protein levels than those from AA or AG carriers (Fig 2, A). A similarly enhanced IL-10 production was observed in monocyte-derived macrophages from GG carriers (Fig 2, B), and the overexpression phenotype was independent of specific pattern recognition receptor activation (Fig 2, C). Although the ability of macrophages to ingest the conidia remained intact regardless of the genotype (Fig 2, D), cells carrying the IL-10 high-producing genotype displayed a 25% decrease in their ability to clear the fungus, as compared to AA carriers (Fig 2, E). This defect was dependent on IL-10, because inhibiting IL-10-mediated signals with a neutralizing antibody restored the fungicidal ability. Importantly, the donor GG genotype also differentially regulated the levels of IL-10 in hematological patients, with higher levels present in bronchoalveolar lavages from cases of IA carrying the GG genotype than AA carriers (Fig 2, F). In contrast to IL-10, PBMCs carrying the GG genotype at rs1800896 secreted lower amounts of TNF-α than those from AA or AG carriers after infection (Fig 2, G), a finding implying that, under these conditions, GG homozygotes generate lesser inflammatory responses. The dichotomy between IL-10 and TNF-α production according to rs1800896 genotypes was confirmed in human macrophages, in which the same genotype-specific alterations were observed (Fig 2, H) and extended to other proinflammatory cytokines such as IL-6, IL-1β, and IL-8 (Fig 2, I). Strikingly, we observed that the defect in TNF-α production by macrophages from GG carriers was abolished when IL-10 was neutralized (Fig 2, J). In support of this, the addition of IL-10 to cells carrying low-producing genotypes restrained the production of TNF-α in response to infection (Fig 2, K). Likewise, the median concentrations of TNF-α were also decreased among hematological patients carrying the IL-10 high-producing genotype (Fig 2, L). We have identified rs1800896 (or a variant in strong LD with it) as the underlying causal variant within the IL10 locus and disclosed the suppression of immune responses to A fumigatus as the primary mechanism explaining the increased susceptibility to infection. The rs1800896 alleles have been shown to physically interact with the transcription repressor poly(adenosine diphosphate–ribose) polymerase 1 and the specificity protein 1 in an allele-specific manner (see Fig E2 in this article's Online Repository at www.jacionline.org).4Kang X. Kim H.J. Ramirez M. Salameh S. Ma X. The septic shock-associated IL-10 -1082 A > G polymorphism mediates allele-specific transcription via poly(ADP-Ribose) polymerase 1 in macrophages engulfing apoptotic cells.J Immunol. 2010; 184: 3718-3724Crossref PubMed Scopus (34) Google Scholar, 5Larsson L. Rymo L. Berglundh T. Sp1 binds to the G allele of the -1087 polymorphism in the IL-10 promoter and promotes IL-10 mRNA transcription and protein production.Genes Immun. 2010; 11: 181-187Crossref PubMed Scopus (33) Google Scholar Because the DNA-binding activity of these transcription regulators varied in different cell types and experimental conditions, additional factors are likely able to regulate expression at this or other sites containing closely linked variants. For example, E26 domain-containing protein Elk1 was shown to bind specifically to alleles at rs3122605 within the IL10 locus among patients with systemic lupus erythematosus, upregulating circulating IL-10 and correlating with disease activity.6Sakurai D. Zhao J. Deng Y. Kelly J.A. Brown E.E. Harley J.B. et al.Preferential binding to Elk-1 by SLE-associated IL10 risk allele upregulates IL10 expression.PLoS Genet. 2013; 9: e1003870Crossref PubMed Scopus (33) Google Scholar Overall, further research is needed to clarify the cell-specific transcription factor(s) activated during IA and that may regulate the rs1800896-dependent IL-10 expression. The overexpression of IL-10 in macrophages has been reported to promote the autocrine deactivation of these cells, hampering proinflammatory cytokine production and control of pathogen growth.1Saraiva M. O'Garra A. The regulation of IL-10 production by immune cells.Nat Rev Immunol. 2010; 10: 170-181Crossref PubMed Scopus (2026) Google Scholar These immunosuppressive effects were confirmed in mouse macrophages to depend on the host's genetic background via the differential production of type I interferons.7Howes A. Taubert C. Blankley S. Spink N. Wu X. Graham C.M. et al.Differential production of type I IFN determines the reciprocal levels of IL-10 and proinflammatory cytokines produced by C57BL/6 and BALB/c macrophages.J Immunol. 2016; 197: 2838-2853Crossref PubMed Scopus (25) Google Scholar In humans, a SNP in Forkhead box O3 (FOXO3) was found to upregulate IL-10 production, while restraining the inflammatory responses of LPS-stimulated monocytes by regulating TGF-β production,8Lee J.C. Espeli M. Anderson C.A. Linterman M.A. Pocock J.M. Williams N.J. et al.Human SNP links differential outcomes in inflammatory and infectious disease to a FOXO3-regulated pathway.Cell. 2013; 155: 57-69Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar suggesting that several regulatory pathways may occur to explain the immunosuppressive effects of IL-10. In conclusion, our findings may contribute to open new horizons and lay the foundations for risk stratification and preemptive approaches aimed at a more effective management of IA.9Oliveira-Coelho A. Rodrigues F. Campos Jr., A. Lacerda J.F. Carvalho A. Cunha C. Paving the way for predictive diagnostics and personalized treatment of invasive aspergillosis.Front Microbiol. 2015; 6: 411Crossref PubMed Scopus (22) Google Scholar We thank Jean-Paul Latgé (Mycology Department, Institut Pasteur, Paris) for generously providing the fungal strain used in this study. Study approval was obtained from the Ethics Subcommittee for Life and Health Sciences of the University of Minho, Portugal (125/014 and 014/015); the Ethics Committee for Health of the Instituto Português de Oncologia, Porto, Portugal (26/015); the Ethics Committee of the Lisbon Academic Medical Center, Portugal (632/014); the Comité de Ética e Investigación Clínica of the Virgen de las Nieves University Hospital, Spain (02/011); and the National Commission for the Protection of Data, Portugal (1950/015). A total of 216 hematological patients of European descent undergoing allogeneic HSCT at Instituto Português de Oncologia, Porto (Portugal) and at the Virgen de las Nieves University Hospital, Granada (Spain) between 2002 and 2014, and respective donors, were included in the discovery study. Demographic and clinical characteristics of the patients are summarized in Table E1. Fifty-one cases of probable/proven IA were identified according to the revised standard criteria from the European Organization for Research and Treatment of Cancer/Mycology Study Group.E1De Pauw B. Walsh T.J. Donnelly J.P. Stevens D.A. Edwards J.E. Calandra T. et al.Revised definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group.Clin Infect Dis. 2008; 46: 1813-1821Crossref PubMed Scopus (4036) Google Scholar In the confirmation study, a matched case-control analysis was performed using 197 HSCT donors from the Hospital de Santa Maria, Lisbon (Portugal). Sixty-seven cases of probable/proven IA and at least 1 matched control according to donor type and year of transplantation (±5 years) were included. SNPs were selected based on their ability to tag surrounding variants with a pairwise correlation coefficient r2 of at least 0.80 and a minor allele frequency ≥5% using publically available sequencing data from the Pilot 1 of the 1000 Genomes Project for the CEU population.E2Abecasis G.R. Altshuler D. Auton A. Brooks L.D. Durbin R.M. Gibbs R.A. et al.1000 Genomes Project ConsortiumA map of human genome variation from population-scale sequencing.Nature. 2010; 467: 1061-1073Crossref PubMed Scopus (5937) Google Scholar A total of 25 SNPs covering the haplotypic diversity of the IL10 gene was captured using 5 tests (Table E2). Genomic DNA was isolated from whole blood from recipients and donors before transplantation using the QIAcube automated system (Qiagen, Hilden, Germany). Genotyping was performed using KASPar assays (LGC Genomics, Hertfordshire, UK) in an Applied Biosystems 7500 Fast real-time PCR system (Thermo Fisher Scientific, Waltham, Mass). All SNPs were in Hardy-Weinberg equilibrium in the patient population, and allele and genotype frequencies of selected SNPs were in line with those reported in the 1000 Genomes Project.E2Abecasis G.R. Altshuler D. Auton A. Brooks L.D. Durbin R.M. Gibbs R.A. et al.1000 Genomes Project ConsortiumA map of human genome variation from population-scale sequencing.Nature. 2010; 467: 1061-1073Crossref PubMed Scopus (5937) Google Scholar Mean call rate was >98% for all genotyped SNPs. Quality control for the genotyping results was achieved with negative controls, common and rare homozygous controls, and retesting of samples with indeterminate results. Bronchoalveolar lavages (BALs) from hematological patients were collected using standard techniques when infection was clinically suspected (progressive pneumonia under antibiotic therapy). BALs were obtained by instillation of 2 × 20 mL of 0.9% sterile saline solution to the most peripheral bronchus of the most radiologically involved lobe. BAL samples with comparable recovery rates and from patients that were not long-term smokers, without any other relevant lung-associated diseases and undergoing similar drug regimens were used, according to the standardization rules of the European Respiratory Society to measure acellular components.E3Haslam P.L. Baughman R.P. Report of ERS Task Force: guidelines for measurement of acellular components and standardization of BAL.Eur Respir J. 1999; 14: 245-248Crossref PubMed Scopus (209) Google Scholar After careful aspiration, BALs were centrifuged at 3000 rev/min for 5 minutes at 4°C to remove cell debris. All samples were stored at −80°C until use. Buffy coats from healthy donors were obtained after written informed consent at the Hospital de Braga, Braga, Portugal. Briefly, PBMCs were enriched from buffy coats by density gradient using Histopaque-1077 (Sigma-Aldrich, St. Louis, Mo). Cells present in the enriched mononuclear fraction were washed twice in PBS (Sigma) and resuspended in RPMI-1640 culture medium with 2 mmol/L stable glutamine and 2 g/L NaHCO3 supplemented with 10% human serum (Merck Millipore, Billerica, Mass), 10 U/mL penicillin/streptomycin, and 10 mmol/L HEPES (Gibco, Thermo Fisher). Monocytes were then separated by positive selection using magnetically labelled CD14 MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany) on a MiniMACS separator. Isolated monocytes were resuspended in complete RPMI medium and seeded at a concentration of 1 × 106 cells/mL in 24-well plates (Corning, Corning, NY) for 7 days in the presence of 20 ng/mL recombinant human granulocyte macrophage colony-stimulating factor (Miltenyi Biotec). Acquisition of macrophage morphology was confirmed by visualization in a BX61 microscope (Olympus, Tokyo, Japan). PBMCs and macrophages seeded at a concentration of 1 × 106 cells/mL in 24-well plates (Corning) were either left untreated or stimulated with live conidia of A fumigatus for 24 hours at an effector-to-target ratio of 1:1 or 1:10, respectively, at 37°C in 5% CO2. In some experiments, cells were stimulated with 1 μg/mL LPS, 10 μg/mL poly(I:C), 0.05 μmol/L 5′-C-phosphate-G-3′, 100 ng/mL macrophage-activating lipopeptide-2, and 5 μg/mL β-1,3-(D)-glucan (all from Sigma) for 24 hours. In others, cells were pretreated with 1, 5, or 10 μg/mL of anti-IL-10 (clone JES3-19F1) or a IgG2a isotype control, or with 1, 10, or 100 ng/mL of human recombinant IL-10 (all from BioLegend, San Diego, Calif) for 1 hour before stimulation. Total RNA from human cells was extracted using the PureLink RNA Mini Kit (Ambion, Thermo Fisher) and reverse transcribed with the NZY First-Strand cDNA Synthesis kit (NZYTech, Lisbon, Portugal). Real-time RT-PCR was performed in an Applied Biosystems 7500 Fast PCR system (Thermo Fisher) using PowerUp SYBR Green Master Mix chemistry (Thermo Fisher). PCR primers for IL10 were as follows: sense, 5′-GAACCTGAAGACCCTCAGGC-3′ and antisense, 5′-AGGCATTCTTCACCTGCTCC-3′. Amplification efficiencies were validated and the expression of the target gene value was normalized against the expression of the β-actin gene (ACTB). The thermal profile was 50°C for 2 minutes for uracil-DNA glycosylase activation, 95°C for 3 minutes followed by 40 cycles of denaturation for 3 seconds at 95°C, and an annealing/extension step of 30 seconds at 60°C. Each data point was examined for integrity by analysis of the amplification plot. The levels of cytokines in culture supernatants and BAL samples were quantified using ELISA MAX Deluxe Set kits (BioLegend) and the Human Premixed Multi-Analyte Kit (R&D Systems, Minneapolis, Minn), respectively. The limit of detection for the single assays was as follows: IL-10 (3.9 pg/mL), TNF-α (7.8 pg/mL), IL-6 (7.8 pg/mL), IL-1β (2 pg/mL), and IL-8 (15.6 pg/mL). To measure the conidiacidal ability, 1 × 105 macrophages adhered to 96-well plates (Corning) were stimulated with live A fumigatus conidia at a 10:1 effector-to-target ratio for 1 hour at 37°C in 5% CO2. The noningested conidia were removed and the macrophages were allowed to kill the internalized conidia for 2 hours at 37°C in 5% CO2. The culture plates were frozen at -80°C and thawed at 37°C to cause cell lysis and release of ingested conidia. Serial dilutions of cell lysates were plated on solid growth media and the number of colony-forming units was enumerated following a 2-day incubation. The percentage of colony-forming units' inhibition—referred to as fungicidal ability—was calculated. To evaluate phagocytosis, 1 × 106 macrophages adhered to chamber slides were infected with fluorescein isothiocyanate–labelled conidia of A fumigatus at a 1:5 effector-to-target ratio. The infection was synchronized for 30 minutes at 4°C and phagocytosis was initiated by shifting the coincubation to 37°C at 5% CO2 for 1 hour. Phagocytosis was stopped by washing with ice-cold PBS and macrophages were labelled with a phycoerythrin-conjugated anti-CD14 antibody (BioLegend) and extracellular conidia stained with 0.25 mg/mL Calcofluor White (Sigma) for 30 minutes at 4°C to avoid further ingestion. After washing with PBS, cells were fixed with 3.7% (vol/vol) formaldehyde/PBS for 15 minutes. The number of ingested green conidia was enumerated by examining the slides by fluorescence microscopy (Olympus). Counting was performed by 2 different individuals that were blind to the genotype status of the samples. A minimum of 500 cells in different fields was evaluated and data were expressed as percentage of internalized conidia. In the discovery data set, the probability of IA according to IL10 genotypes was determined using the cumulative incidence method and compared using the Gray test.E4Gray R.J. A class of K-sample tests for comparing the cumulative incidence of a competing risk.Ann Stat. 1988; 16: 114-154Crossref Google Scholar Cumulative incidences of infection at 24 months were computed with the cmprsk package for R version 2.10.1,E5Scrucca L. Santucci A. Aversa F. Competing risk analysis using R: an easy guide for clinicians.Bone Marrow Transplant. 2007; 40: 381-387Crossref PubMed Scopus (505) Google Scholar with censoring of data at the date of last follow-up visit and relapse and death as competing events. Infection-free survival was defined as the time from transplantation to death from infection and was obtained by the Kaplan-Meier method and compared using the log-rank. All clinical and genetic variables achieving a P value ≤.15 in the univariate analysis were entered 1 by 1 in a pairwise model together and kept in the final model if they remained significant (P < .05). Multivariate analysis was performed using the subdistribution regression model of Fine and Gray with the cmprsk package for R.E6Scrucca L. Santucci A. Aversa F. Regression modeling of competing risk using R: an in depth guide for clinicians.Bone Marrow Transplant. 2010; 45: 1388-1395Crossref PubMed Scopus (339) Google Scholar Conditional logistic regression was used to analyze genotype data from the confirmation study. Data obtained in functional assays were expressed as mean ± SEM (assayed in triplicate), and from BAL as median ± SEM and interquartile ranges with the dots outside the boxes representing outliers (assayed in duplicate). Unpaired Student t-test with Bonferroni adjustment and 2-tailed Mann-Whitney U test were used to determine statistical significance (P < .05).Fig E2Transcription factor binding to the sequence around rs1800896. The input SNP rs1800896 is indicated in red and the remaining SNPs in strong LD with it are indicated in black. The regions in yellow, purple, and turquoise represent the promoter, exons, and the 3′-untranslated region, respectively. The transcription factors binding to each sequence were computationally predicted using the PROMO software and the version 8.3 of TRANSFACE8Farre D. Roset R. Huerta M. Adsuara J.E. Rosello L. Alba M.M. et al.Identification of patterns in biological sequences at the ALGGEN server: PROMO and MALGEN.Nucleic Acids Res. 2003; 31: 3651-3653Crossref PubMed Scopus (752) Google Scholar and are indicated in blue boxes. The experimentally validated transcription factors were retrieved from the literature and are indicated in green boxes. AP-2α, Activating enhancer binding protein 2 alpha; CEPBβ, CCAAT/enhancer-binding protein β; GR, glucocorticoid receptor; PARP-1, poly(adenosine diphosphate–ribose) polymerase 1; PR-A, progesterone receptor A; PR-B, progesterone receptor B; Sp1, specificity protein 1; STAT4, signal transducer and activator of transcription 4; TSS, transcription start site.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table E1Baseline characteristics of transplant recipients enrolled in the discovery studyVariablesIA (n = 51)No IA (n = 165)P valueAge at transplantation ≤20 y7 (14.0)26 (16.0).01 21-40 y7 (14.0)55 (33.7) >40 y36 (72.0)82 (50.3)Sex Female14 (27.5)66 (40.0).12 Male37 (72.5)99 (60.0)Underlying disease Acute leukemia22 (43.1)87 (53.1).42 Chronic lymphoproliferative diseases18 (35.3)43 (26.2) Chronic myeloproliferative diseases3 (5.9)4 (2.4) Myelodysplastic/myeloproliferative diseases3 (5.9)11 (6.7) Aplastic anemia4 (7.8)9 (5.5) Other1 (2.0)10 (6.1)Transplantation type Matched, related29 (60.4)92 (57.5).51 Matched, unrelated15 (31.3)42 (26.3) Mismatched, related0 (0.0)3 (1.9) Mismatched, unrelated4 (8.3)23 (14.4)Graft source Peripheral blood23 (92.0)95 (81.2).52 Bone marrow2 (8.0)19 (16.2) Cord blood0 (0.0)3 (2.6)Disease stage First complete remission28 (65.1)90 (63.4).95 Second or subsequent remission, or relapse9 (20.9)29 (20.4) Active disease6 (14.0)23 (16.2)Conditioning regimen RIC32 (66.7)112 (70.0).66 Myeloablative16 (33.3)48 (30.0)CMV serostatus of donor and recipient D−/R+ or D+/R+37 (80.4)123 (81.5).75 D−/R− or D+/R−9 (19.6)28 (18.5)Duration of neutropenia, mean (range) (d)∗Neutropenia was defined as ≤0.5 × 109 cells/L.14 (8-26)12 (6-24).01Acute GVHD No GVHD or grades I to II41 (80.4)149 (91.4).03 Grades III to IV10 (19.6)14 (8.6)Antifungal prophylaxis†Other antifungals used in prophylaxis included voriconazole (n = 10), liposomal amphotericin B (n = 5), and micafungin (n = 4). Fluconazole13 (30.2)51 (36.4).26 Posaconazole13 (30.2)47 (33.6) Itraconazole14 (32.6)26 (18.6) Other3 (7.0)16 (11.4)Values are n (%) unless otherwise indicated. P values were calculated by Fisher exact probability t test or Student t test for continuous variables. Variables with P < .15 were included in the multivariate model. Numbers for each variable may not reflect the total patients due to missing clinical information.CMV, Cytomegalovirus; D, donor; GVHD, graft-versus-host-disease; R, recipient; RIC, reduced intensity conditioning.∗ Neutropenia was defined as ≤0.5 × 109 cells/L.† Other antifungals used in prophylaxis included voriconazole (n = 10), liposomal amphotericin B (n = 5), and micafungin (n = 4). Open table in a new tab Table E2Description of haplotype-tagging SNPs in the IL10 geneRefSNPGenome coordinatesGenomic locationAllelesCEU MAFHWESNPs taggedrs1800896chr1:205013520Near gene 5′G>A0.4690.7344rs1800893, rs1878672, rs3024496, rs3024491, rs2222202, rs3024500, rs3024502rs1800872chr1:205013030Near gene 5′C>A0.2080.7221rs1800871, rs3024490, rs1518111, rs1518110, rs1554286rs3021094chr1:205011575IntronA>C0.0620.3420—rs3024509chr1:205009920IntronT>C0.0580.2565rs11119474, rs7548373, rs7519318, rs7512090rs3024498chr1:205008152UTR-3A>G0.2830.8102rs61815632, rs6673928, rs3024492, rs17015767Publically available sequencing data from Pilot 1 of the 1000 Genomes Project (www.1000genomes.org) was used to identify all SNPs tagging surrounding variants with a pairwise correlation coefficient r2 > 0.8 and a MAF ≥5%. Genome coordinates were extracted from the hg18 build.CEU, Utah Residents (CEPH) with Northern and Western Ancestry; MAF, minor allele frequency; HWE, Hardy-Weinberg equilibrium; UTR, untranslated region. Open table in a new tab Table E3Cumulative incidence of IA according to recipient and donor IL10 genotypes in the discovery study and association test resultsRefSNPGenotype(s)Cumulative incidence of IA at 24 mo (%)RecipientP valueDonorP valuers1800896GG20.0.7144.5.009GA13.020.8AA17.417.2rs1800872CC15.6.7828.3.30CA20.023.4AA16.711.8rs3021094AA19.8.8120.3.57AC17.216.4CCn/an/ars3024509TT16.7.5414.3.39TC20.619.2CCn/an/ars3024498AA15.7.7316.3.66AG16.317.6GG21.021.9The P values are for the Gray test using cumulative incidence analysis. After adjustment for multiple testing with the use of the Bonferroni method (5 independent tests), the level for significance was defined as P < .01. Significant P value is indicated in boldface.n/a, No reported cases of IA with the indicated genotype. Open table in a new tab Table E4Multivariate analysis of the association of donor IL10 genotypes with the risk of IA among transplant recipients in the discovery and confirmation studiesGenetic/clinical variablesDiscovery (n = 216)Confirmation (n = 197)Combined (n = 413)Adjusted HR∗HRs were adjusted for patient age and sex, duration of neutropenia, and development of aGVHD grade III to IV. The genotype combinations AA+AG were the reference category for the GG genotype. (95% CI)P valueAdjusted OR†ORs were adjusted for patient age and sex. (95% CI)P valueAdjusted OR (95% CI)P valueGG at rs18008962.58 (1.44-4.62).0012.32 (1.10-4.90).0312.48 (1.48-4.18)<.001aGVHD III to IV2.24 (1.22-4.10).009——Multivariate analyses were based on the subdistribution regression model of Fine and Gray in the discovery study and on conditional logistic regression in the confirmation study and meta-analysis.aGVHD, Acute graft-versus-host-disease; HR, hazard ratio; OR, odds ratio.∗ HRs were adjusted for patient age and sex, duration of neutropenia, and development of aGVHD grade III to IV. The genotype combinations AA+AG were the reference category for the GG genotype.† ORs were adjusted for patient age and sex. Open table in a new tab Table E5LD around rs1800896RefSNPGenomic coordinatesGenomic locationLocation (relative to rs1800896)Distance from rs1800896 (bps)r2rs3024502chr1:205006933Near gene 3′Downstream6,5870.94rs3024500chr1:205007454Near gene 3′Downstream6,0660.97rs3024496chr1:205008487UTR-3Downstream5,0330.97rs1878672chr1:205010336IntronicDownstream3,1841rs3024491chr1:205011669IntronicDownstream1,8510.97rs2222202chr1:205012004IntronicDownstream1,5160.97rs1800893chr1:205013790Near gene 5′Upstream2701rs1800890chr1:205015988Near gene 5′Upstream2,4680.66rs10494879chr1:205018827Near gene 5′Upstream5,3070.79rs6676671chr1:205019371Near gene 5′Upstream5,8510.71rs6667202chr1:205023715Near gene 5′Upstream10,1950.62rs4072226chr1:205024072Near gene 5′Upstream10,5520.54Publically available sequencing data from the Pilot 1 of the 1000 Genomes Project (www.1000genomes.org) was used to identify all SNPs in LD (r2 > 0.5) with rs1800896. Genome coordinates were extracted from the hg18 build. Open table in a new tab Values are n (%) unless otherwise indicated. P values were calculated by Fisher exact probability t test or Student t test for continuous variables. Variables with P < .15 were included in the multivariate model. Numbers for each variable may not reflect the total patients due to missing clinical information. CMV, Cytomegalovirus; D, donor; GVHD, graft-versus-host-disease; R, recipient; RIC, reduced intensity conditioning. Publically available sequencing data from Pilot 1 of the 1000 Genomes Project (www.1000genomes.org) was used to identify all SNPs tagging surrounding variants with a pairwise correlation coefficient r2 > 0.8 and a MAF ≥5%. Genome coordinates were extracted from the hg18 build. CEU, Utah Residents (CEPH) with Northern and Western Ancestry; MAF, minor allele frequency; HWE, Hardy-Weinberg equilibrium; UTR, untranslated region. The P values are for the Gray test using cumulative incidence analysis. After adjustment for multiple testing with the use of the Bonferroni method (5 independent tests), the level for significance was defined as P < .01. Significant P value is indicated in boldface. n/a, No reported cases of IA with the indicated genotype. Multivariate analyses were based on the subdistribution regression model of Fine and Gray in the discovery study and on conditional logistic regression in the confirmation study and meta-analysis. aGVHD, Acute graft-versus-host-disease; HR, hazard ratio; OR, odds ratio. Publically available sequencing data from the Pilot 1 of the 1000 Genomes Project (www.1000genomes.org) was used to identify all SNPs in LD (r2 > 0.5) with rs1800896. Genome coordinates were extracted from the hg18 build.