Langerhans Cells Profile Compared with Epidermal Cell Histiocytosis

The Journal of Immunology

Cell-Speciﬁc Gene Expression in Langerhans Cell Histiocytosis Lesions Reveals a Distinct Proﬁle Compared with Epidermal Langerhans Cells

Carl E. Allen,*,† Liunan Li,* Tricia L. Peters,‡ Hon-chiu Eastwood Leung,*,†,x Alexander Yu,* Tsz-Kwong Man,*,† Sivashankarappa Gurusiddappa,* Michelle T. Phillips,* M. John Hicks,‡ Amos Gaikwad,* Miriam Merad,{ and Kenneth L. McClain*,†

Langerhans cell histiocytosis (LCH) is a rare disease characterized by heterogeneous lesions containing CD207+ Langerhans cells (LCs) and lymphocytes that can arise in almost any tissue and cause significant morbidity and mortality. After decades of research, the cause of LCH remains speculative. A prevailing model suggests that LCH arises from malignant transformation and metastasis of epidermal LCs. In this study, CD207+ cells and CD3+ T cells were isolated from LCH lesions to determine cell-specific gene expression. Compared with control epidermal CD207+ cells, the LCH CD207+ cells yielded 2113 differentially expressed genes (false discovery rate , 0.01). Surprisingly, the expression of many genes previously associated with LCH, including cell-cycle regulators, proinflammatory cytokines, and chemokines, were not significantly different from control LCs in our study. However, several novel genes whose products activate and recruit T cells to sites of inflammation, including SPP1 (osteopontin), were highly overexpressed in LCH CD207+ cells. Furthermore, several genes associated with immature myeloid dendritic cells were overexpressed in LCH CD207+ cells. Compared with the peripheral CD3+ cells from LCH patients, the LCH lesion CD3+ cells yielded only 162 differentially regulated genes (false discovery rate , 0.01), and the expression profile of the LCH lesion CD3+ cells was consistent with an activated regulatory T cell phenotype with increased expression of FOXP3, CTLA4, and SPP1. Results from this study support a model of LCH pathogenesis in which lesions do not arise from epidermal LCs but from accumulation of bone marrow-derived immature myeloid dendritic cells that recruit activated lymphocytes. The Journal of Immunology, 2010, 184: 4557–4567.

angerhans cell histiocytosis (LCH) is a potentially fatal dis- limited organ system involvement have a very good prognosis, and ease characterized by invasive lesions inﬁltrated with multiple patients with multisystem disease have survival rates ∼80% (5). L cell types, including CD1a+/CD207+ cells presumed to be However, survival is poor in patients with high-risk multisystem disease pathologic Langerhans cells (LCs). The incidence of LCH is approxi- who fail to respond to induction therapy (6). Chemotherapy for LCH is

by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. mately five cases per million children and 1/10,000 live births per year based on a lymphoma model of general immune suppression and cy- (1). Approximately 30% as many adults are afflicted, although this totoxicity to rapidly proliferating cells (reviewed in Ref. 7). incidence is likely underestimated (2). LCH includes a spectrum of The etiology of LCH is not known. Scientific debate has focused clinical presentations from single-system involvement in skin or bone to on LCH resulting from malignant transformation or from functional diffuse multisystem involvement of liver, lungs, bone marrow, CNS, proliferation of epidermal LCs in response to external stimuli (8– and other organ systems (reviewed in Refs. 3 and 4). Patients with 12). Regardless of whether clonal proliferation is the initiating factor in LCH, the CD207+ cells from these lesions require multiple interactions with other cell types, including T cells, eosino- *Department of Pediatrics, Texas Children’s Cancer Center and Hematology Service, †Dan L. Duncan Cancer Center, ‡Department of Pathology, and xDepartment of phils, and macrophages (13). LCs will not grow in isolation in vitro http://classic.jimmunol.org Molecular and Cellular Biology, Baylor College of Medicine, Houston, TX 77030; or as xenografts in immune-deficient mice. The mainstay of re- { and Department of Gene and Cell Medicine, Mount Sinai School of Medicine, New search on LCH tissues has been immunohistochemical analysis of York, NY 10029 biopsy samples. Although this approach has been useful, it is also Received for publication July 21, 2009. Accepted for publication February 2, 2010. limited by various difficulties, such as testing multiple Abs si- This work was supported in part by research funding from the Histiocytosis Associ- ation of America (to C.E.A. and K.L.M.), the Histiocytosis Association of Canada (to multaneously, variable sensitivity of Abs, inability to quantita- C.E.A. and K.L.M.), the American Society of Clinical Oncology (to C.E.A.), the tively interpret results, and lack of control tissues. RNA and Thrasher Research Fund (to C.E.A.), the Larry and Helen Hoag Foundation Clinical

Downloaded from protein studies from whole-biopsy samples are also difficult to Translational Research Award (to C.E.A.), and the National Institutes of Health Pediatric Oncology Research Training Grant T32 CA115303-03 (to C.E.A.) and interpret because of the heterogeneous composition of LCH le- R21 CA 114981-01A2 (to K.L.M.). M.T.P. received support from the American sions. To overcome some of the experimental challenges in Cancer Society Institutional Research Grant (2007–08) 93-034-23. studying LCH, we devised a robust procedure to study cell-specific The sequences presented in this article have been submitted to the Gene Expression gene-expression profiling in the cells that most likely contribute to Omnibus Web site under accession number GSE16395. pathology in LCH patients: LCs (CD207+) and T cells (CD3+). Address correspondence and reprint requests to Dr. Carl E. Allen, Texas Children’s Hospital, 6621 Fannin Street, CC1410.00, Houston, TX 77030. E-mail address: [email protected] Materials and Methods The online version of this article contains supplemental material. Subjects Abbreviations used in this paper: DC, dendritic cell; LC, Langerhans cell; LCH, + + Langerhans cell histiocytosis; PI, propidium iodide; Treg, regulatory T cell. LCH diagnosis was established by the presence of CD1a or CD207 histiocytes in clinical biopsy specimens. Samples from the 15 individuals Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 with LCH in this study included patients with relapsed disease and high-

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902336 4558 CELL-SPECIFIC GENE EXPRESSION IN LCH

risk multisystem disease (Supplemental Table IA). Control epidermal LCs GAPDH ratios, as well as signal distribution, were assessed to determine were isolated from discarded skin, primarily elective circumcisions, from the outlier cases. Normalization and probe set summarization was done in patients ,18 y of age. Control tonsil CD3 cells were isolated from dis- BRB-Arraytools (available at http://linus.nci.nih.gov/BRB-ArrayTools. carded samples from elective tonsillectomy in patients ,18 y of age. html) using the Robust Multichip Average algorithm. Hierarchical clus- Studies were performed according to protocols approved by the In- tering was performed using centered correlation and average linkage. The stitutional Review Board of Baylor College of Medicine. Significance Analysis of Microarrays algorithm (16) was used for analysis of differential expression, with a false discovery rate , 0.01. Samples were Isolation of LCs and T cells split into three groups for analysis. Group 1 consisted of 13 CD207 tumor LCH samples. Fifteen fresh LCH biopsy samples were collected. They were samples and 12 CD207 controls obtained from normal skin. Group 2 transported in RPMI 1640 media (Invitrogen, Carlsbad, CA) and processed consisted of seven CD3 LCH lesion samples and seven CD3 matched within 24 h. All samples were processed into single-cell suspension over paired samples obtained from patients’ peripheral blood. Group 3 con- a 70-mM-mesh filter. Cells were washed twice with RPMI 1640 supple- sisted of 12 CD3 LCH lesion samples and 4 CD3 pooled controls obtained mented with 10% FBS and then incubated with conjugated Abs, CD207- from tonsil. Heatmaps and other graphics were created using Multi- PE (Beckman Coulter, Fullerton, CA) and CD3-FITC (BD Biosciences, Experiment Viewer, part of the TM4 Microarray Software Suite (17). San Jose, CA) for 30 min on ice. Cells were washed again and resuspended Quantitative real-time PCR in RPMI 1640/FBS with 2 mg/ml propidium iodide (PI; Molecular Probes, Eugene, OR). Cells were then separated by flow cytometry by gating on 2 Real-time PCR reactions were performed with TaqMan Gene Expression the PI population and Ab-specific fluorescence. Cells were sorted with Assays (Applied Biosystems, Foster City, CA), which include a mix of two a MoFlo Sorter (Beckman Coulter) directly into PicoPure RNA Extraction unlabeled PCR primers (900 nM final concentration) and 1 FAM dye-labeled Buffer (Molecular Devices, Sunnyvale, CA) (Supplemental Table IB). TaqMan MGB Probe (250 nM final concentration). Single-stranded cDNA Select samples were reanalyzed by flow cytometry for purity (Fig. 1). was generated by RNA amplification, as described above, in experiments Control skin LC samples. Control LCs were isolated from 12 skin samples independent from the amplifications used to generate cDNA probe for mi- that were transported in RPMI 1640 media and processed within 24 h. croarray studies. For the CD207+ analysis, one cDNA pool was made from Tissue was incubated in RPMI 1640 with 5 U/ml dispase II (Roche, equal contributions from the 13 LCH CD207+ samples and another from equal Indianapolis, IN) at 4˚C for 8 h prior to separation of the epidermal layer. contributions from the 12 control CD207+ samples. For the CD3+ analysis, The epidermal layer was further treated with 0.25% trypsin-EDTA one cDNA pool was made from equal contributions from the seven peripheral (Invitrogen) for 15 min at 37˚C and then LCs were isolated with CD207- LCH CD3+ samples and another from equal contributions from the seven PE–conjugated Ab (Beckman Coulter), as described above. Patient details matched LCH lesion CD3+ samples. Each reaction included 20 ng cDNA. from the individual skin samples were not available. However, review of TaqMan Fast Universal PCR Master Mix (Applied Biosystems) was used in elective circumcisions performed at Texas Children’s Hospital in an op- 25-ml reactions in 96-well plates on a iQ5 Real-Time PCR Detection System erating room shows a range of ages from 1 mo to 18 y. (Bio-Rad, Hercules, CA). Assays were plated in triplicate in two independent Peripheral T cells. Peripheral T cells were isolated from seven patients with experiments. Thermal cycling conditions were set at 2 min at 50˚C and 10 min active LCH prior to chemotherapy (Supplemental Table I). Peripheral blood at 95˚C and then 40 cycles of 95˚C and 60˚C for 1 min. Relative mRNA levels was collected and stored in EDTA+ tubes and processed within 24 h. were determined using standard DD Ct calculations, with expression levels of PBMCs were isolated after centrifuging blood over a Histopaque-1077 experimental samples normalized to GAPDH (18). TaqMan probe sets (Ap- (Sigma-Aldrich, St. Louis, MO) gradient at 450 3 g for 30 min. PBMCs plied Biosystems) included AFF3, Hs00289335_s1; CCL5, Hs00174575_m1; were washed twice in RPMI 1640 and then T cells were isolated with CD3- CCR1, Hs00174298_m1; CD36, Hs01567188_g1; CD74, Hs00269961_m1; FITC (BD Biosciences)-conjugated Abs, as described above. CDH1, Hs00170423_m1; CEACAM6, Hs00366002_m1; CTLA4, Hs001- 75480_m1; DCAL1, Hs00416849_m1; DUSP4, Hs01027785_m1; FOXP3, Tonsil T cells. T cells were isolated from 20 tonsil samples from children Hs00203958_m1; GAPDH, Hs02758991_g1; HLA-DRA, Hs00219578_m1; who underwent elective tonsillectomy. Tonsils were processed in an HOXB7, Hs00270131_m1; HSPA1A, Hs00359147_s1; IL-17A, Hs00174- identical fashion to the LCH lesions, and CD3+ cells were isolated by flow 383_m1; IL-2, Hs00174114_m1; IL-8, Hs00174103_m1; MMP1, Hs0089- cytometry with CD3-FITC Ab, as described above. by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. 9660_g1; MMP9, Hs00234579_m1; NRP1, Hs00826128_m1; PERP, Hs00- RNA purification and cDNA amplification 751717_s1; S100A7, Hs00162488_m1; S100A8, Hs00374263_m1; SMYD3, Hs00224208_m1; SPP1, Hs00959010_m1; TACSTD1, Hs00158980_m1; Total RNA was processed and cDNA was amplified for 13 LCH CD207 TNFRSF9, Hs00155512_m1; and VNN1, Hs01546812_m1. samples, 12 control CD207 samples, 12 LCH lesion CD3 samples, 7 LCH peripheral CD3 samples, and 4 pooled tonsil CD3 samples (5 individual RNA Immunohistochemistry samples/tonsil pool). Total RNA was processed from the sorted cells, Fresh LCH biopsy tissue, skin samples, and tonsil samples were snap- according to the PicoPure RNA Isolation Kit protocol (Molecular Devices). frozen in embedding medium (Tissue-Tek OCT Compound; Sakura RNA concentration and quality were verified using the 6000 Pico Chip Finetech USA, Torrance, CA) for frozen tissue specimens and stored at 2 (Agilent, Palo Alto, CA) at the Baylor College of Medicine Microarray 80˚C within 8 h from the time of collection. Tissue was cut at a thickness Facility. Any samples with a detectable RNA integrity number (14) ,5 were http://classic.jimmunol.org of 4–8 mm from frozen blocks and adhered to microscope slides. Slides excluded from the study. cDNA amplification was performed with the WT- were stored at 280˚C until use. After removal from storage, slides were Ovation Pico System, according to manufacturer’s protocol (NuGen, San air-dried for 30 min, and tissue was subsequently fixed by incubating Carlos, CA). This is a whole-transcriptome amplification system with which slides in ice-cold acetone for 15 min. After being air-dried for 10 min, the we were able to generate 4–6 mg cDNA from 700 pg to 25 ng of input RNA. tissue was rehydrated by incubation in PBS for 10 min. Immunohisto- Fragmented and biotinylated cDNA samples for the gene chip studies were chemical staining was performed using the Dako EnVision + Dual Link generated using the FL-Ovation cDNA Biotin Module v2 (NuGen). Kit (Dako, Carpinteria, CA). Endogenous peroxidase was blocked using Dako Block Buffer for 10 min. Primary Abs were diluted in Dako Ab Downloaded from Affymetrix GeneChip assays Diluent, and tissue was incubated at room temperature for 3 h or at 4˚C Fragmented and biotinylated cDNA was hybridized to Affymetrix U133 overnight. Abs against CD11b and CD11c were diluted 1:10. All other Plus 2.0 chips (Affymetrix, Santa Clara, CA). cDNA was generated from Abs were diluted 1:50. Mouse and rabbit Abs were detected using the each of the LCH CD207, control CD207, LCH lesion CD3, and peripheral previously diluted HRP-linked secondary Ab included in the kit. Goat Abs CD3 samples and was used to hybridize onto individual chips. Four tonsil were detected using anti-goat/HRP Ab (DakoCytomation) diluted 1:100. CD3 cDNA were generated, each from five pooled cDNA samples. Chips Slides were washed three times for 5 min each in Dako Wash Buffer were hybridized overnight at 45˚C with 60 rpm for 16 h. The chips were between incubations. HRP activity was detected using Dako Liquid DAB processed in the fluidic station using the FS450_004 wash and stain pro- + Substrate Chromagen System. Tissue was then stained with hematox- tocol, according to the manufacturer’s instructions. The stained chips were ylin, coverslipped, and stored at room temperature. Abs used included scanned using the Affymetrix GeneChip scanner 3000. The data files have CD207: mouse monoclonal – ab49730 (Abcam, Cambridge, MA); CD3: been deposited at the Gene Expression Omnibus Web site (www.ncbi.nlm. mouse monoclonal – ab699 (Abcam); HLA-DR+DP+DQ: mouse mono- nih.gov/geo/; accession number GSE16395). clonal – sc59251 (Santa Cruz Biotechnology, Santa Cruz, CA); CTLA4: goat polyclonal – sc1628 (Santa Cruz Biotechnology); Osteopontin: Data analysis mouse monoclonal – sc21742 (Santa Cruz Biotechnology); CD44: rabbit monoclonal – ab51037 (Abcam); Neuropilin1: mouse monoclonal Quality control of 48 Affymetrix U133 Plus 2.0 gene-expression chips was – sc5307 (Santa Cruz Biotechnology); Vanin1: goat polyclonal – sc16776 done using the BioConductor package affyQCReport (15). b-Actin and (Santa Cruz Biotechnology); CEACAM6: rabbit polyclonal – ab56234 The Journal of Immunology 4559

(Abcam); MMP1: rabbit monoclonal – ab52631 (Abcam); MMP9: goat Results polyclonal – sc6840 (Santa Cruz Biotechnology); CD13: goat polyclonal Robustness of the gene-expression profiling procedures – sc6995 (Santa Cruz Biotechnology); CD33: rabbit polyclonal – ab59940 (Abcam); CD1d: mouse monoclonal – ab11076 (Abcam); ICAM1: mouse Validity of the data derived from these gene-expression profiling monoclonal – ab20 (Abcam); CD11b: mouse monoclonal – MAB1699 experiments is predicated on fidelity of the methods used to isolate (R&D Systems, Minneapolis, MN); and CD11c: mouse monoclonal – cells and generate cDNA probe. Cells were sorted using fluorescent MAB1777 (R&D Systems). Images were magnified using the Olympus BX51 microscope, 340 objective. Images were captured using the markers conjugated to CD3 and CD207 Abs using standard Olympus DP71 digital camera with Olympus DP Controller and DP methods (Fig. 1). Cell purity was evaluated in several samples by Manager software. sorting cells into RPMI 1640, then reanalyzing with the MoFlo by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 1. Scatter plot of isolated cells. Tissue samples were prepared as described in Materials and Methods. These images represents typical FACS studies from control tonsil (A), peripheral blood from LCH patients (B), LCH lesions (C, E), and control skin (epidermis, D). The scanner was gated on PI2 cells (living cells) and then CD207+ and CD3+ cells were purified with fluorescent-conjugated Abs. The scatter plots show results (from right to left) with no Ab staining, isotype control, CD3+ and CD207+, and reanalysis of the sorted cells in a purity check. Sorted cell purity ranged from 95.5–99.2%. For the LCH lesions, the CD3/CD207 ratio varied considerably from sample to sample (Supplemental Table IB), but the plot shown in C and E is a typical result and demonstrates the ability to obtain specific cell fractions from LCH lesions. 4560 CELL-SPECIFIC GENE EXPRESSION IN LCH

Sorter. Purity exceeded 95% in all CD3+ and CD207+ samples overall intergroup correlation was .30%. By comparison, the tested (Fig. 1). The NuGen WT-Ovation Pico System was used to control skin CD207 cells showed less overall variability, with an amplify RNA to generate cDNA probe for the gene-expression intergroup correlation .50% (Supplemental Fig. 1) (excluding profiling studies. The yield of cDNA consistently ranged between sample #1, which appears to be an outlier). 4 and 6 mg. Previous studies validated this experimental strategy Expression of 2,113 genes (of .47,000) was significantly dif- (19, 20). In this series of experiments, gene-expression datasets ferent between the LCH CD207 cells and the control CD207 cells generated from technical replicates of amplified probe from 5 ng using stringent statistical criteria (false discovery rate , 0.01). input RNA had a correlation .90% (data not shown). Overall, 747 probes (520 genes) showed increased expression and Cell-specific gene expression from LCH biopsy specimens was 2153 probes (1593 genes) showed decreased expression in the evaluated by comparing hybridization signals from amplified LCH CD207 cells (Supplemental Data 1). Several genes involved cDNA on Affymetrix gene chips (U133A Plus 2.0) (Tables I, II). in cell-cycle regulation, inflammation, and LC development have Three sets of comparisons were performed: 13 LCH CD207+ been evaluated in LCH lesions and implicated in models of LCH samples were compared with 12 control skin LCH CD207+ pathogenesis. The most widely accepted current model of LCH samples; 7 LCH lesion CD3+ samples were compared with 7 can be described as the “Activated-Immature” Model, in which peripheral blood CD3+ samples from the same patients; and 12 immature epidermal CD207+ LCs undergo malignant trans- LCH lesion CD3+ samples were compared with 4 pooled control formation or pathologic immune stimulation, and subsequently tonsil CD3+ samples (5 individual tonsil CD3+ samples/pool). take on some properties of an activated LC, including down- Validity of the data derived from this experimental strategy is regulation of anchoring molecules (E-cadherin) and increased evidenced by cell-specific grouping of the datasets in dendrograms expression of proinflammatory cytokines (TNF-a and IL-1b), constructed with unsupervised clustering (Supplemental Fig. 1). chemokine receptors (CCR6 and CCR7), T cell costimulatory molecules (CD40, CD80, and CD86), and markers of cell pro- Cell-specific gene-expression array analyses liferation (Ki67 and PCNA) (reviewed in Refs. 8, 10, and 21). CD207+cells. Analysis of global gene expression in LCH CD207 Furthermore, several studies described the LCH lesion microen- cells showed a consistent and distinct profile from control skin vironment as a “cytokine storm”, with expression of several CD207 cells, with overall correlation ,0.2. Although the LCH proinflammatory cytokines (10, 21–24). CD207 datasets were grouped with unsupervised clustering, the A comprehensive review of PubMed for genes and proteins associated with LCH pathogenesis was performed, and results were compared with cell-specific gene expression in this study (Fig. 2, Supplemental Table IIA). The results from our study supported Table I. Comparison of real-time PCR and microarray results: LCH some previous observations, including significantly decreased ex- CD207 versus control skin CD207 pression of E-cadherin and relatively increased expression of MMP9 (22, 25–28). However, the lack of differential gene ex- Gene Symbol Real-Time PCR Array pression in many molecules previously associated with LCH was surprising. In some cases, such as TNF-a and IL-1b, genes were SPP1 162.6 22.5–37.0 + CEACAM6 76.7 10.1 expressed at very high levels in control skin CD207 cells and LCH by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. + CDK2NA 73.9 2.1–4.9 lesion CD207 cells. For others, such as IL-2 and IL-17A, we failed JAK3 46.1 1.9–7.1 to find significant levels of expression in control epidermal or LCH VNN1 44.0 5.2–7.7 lesion CD207+ cells. Additionally, some genes, such as CCL20, SMYD3 34.8 9.3–12.4 were not found to be significant, despite high levels of expression in AFF3 24.7 3.1–11.3 + HOXB7 20.7 6.6–6.8 some LCH CD207 samples due to highly variable expression. DUSP4 15.7 3.4–10.3 Specific results of differential expression of genes reportedly as- MMP9 15.5 7.6 sociated with LCH in these experiments are detailed below. MMP1 11.5 7.2 Further review of this dataset identified several genes of interest CCR1 8.6 4.5–5.7 not previously associated with increased expression in LCH, in-

http://classic.jimmunol.org CCL5 8.6 4.6 NRP1 4.8 4.5–9.5 cluding genes involved in the regulation of cell cycle (CDC2A, TNFRSF9 4.6 5.8–6.4 AF4/FMR2 [member 3], SMYD3, and HOXB7), apoptosis (BAX, DCAL1 3.2 5.1–6.6 BCL2L1, and CFLAR), signal transduction (DUSP4, JAK3, 2 2 CD36 28.1 (1.4–14.4) PRKCA, TLR2, TLR4, SOCS3, and JAG2), tumor invasion and S100A8 260.2 2(22.5–26.2) EpCAM 2204.4 238.1 CDH1 2269.6 2(32.2–123.6)

Downloaded from PERP 21,172 2(3.9–227.5) S100A7 214,201 2(1.7–74) Table II. Comparison of real-time PCR and microarray results: LCH IL-2 NA 1.0 lesion CD3 versus LCH peripheral CD3 IL-17 NA 1.1 Gene Symbol Real-Time PCR Array Representative genes identified by CD207 gene-expression array studies are listed in the first column. The fold-change (LCH LC versus control skin LC), as determined SPP1 50,031.3 38.9–69.4 by real-time PCR, is listed in the middle column. The fold-change, as determined by IL-8 572.6 21.6–45.7 microarray, is listed in the right column. There are multiple values for some of the array data as the result of multiple probes for different regions of a single gene on the MMP9 219.9 43.6 array chips, which are represented by ranges. Changes are more pronounced in the HSPA 40.8 15.3–21.8 real-time PCR assays, likely because of the lack of signal compression. In the chip FOXP3 24.7 6.7 studies, even if gene expression is absent, background signal may be interpreted as HLADR 18.9 11.6–22.0 minimal expression. In general, real-time PCR confirms trends of increased and CTLA4 11.4 7.5–13.4 decreased expression in candidate genes identified by the gene chip studies. Simi- DUSP4 10.3 6.7–54.6 larly, real-time PCR fails to identify significant differences in some genes previously CD74 5.7 5.2–6.2 associated with LCH. NA, no PCR product was amplified in the real-time PCR reaction, suggesting the Data are organized as described for Table I for array and real-time PCR experi- absence of target template for IL-2 and IL-17. ments using cDNA generated from purified CD3 cells. The Journal of Immunology 4561 by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd.

FIGURE 2. LCH-associated genes: CD207+ CD207+ (LCH lesion CD207+ versus skin CD207) gene heatmap. Literature was reviewed for genes associated with LCH and tested for LCH CD207+-specific gene expression. Repeated sample names represent genes with multiple probes on the Affymetrix gene chip with significant results in the expression-array experiments. Control skin sample CD207+ cells are listed on the left half of each heatmap. LCH CD207+ samples are listed on the right. Yellow boxes indicate increased expression in an individual sample; blue indicates decreased expression. Some http://classic.jimmunol.org genes were not differentially expressed because expression was consistently high in both sets (e.g., IL-1B). Some genes were not differentially expressed because expression was absent in both sets (e.g., IL-17A). Some genes were not indicated as differentially expressed because of variability within the sets (e.g., CCL20). The categorical gene list indicates genes with significant increased (yellow) or decreased (blue) expression that may be relevant to LCH pathology and represent potential targets for diagnosis, prognosis, and therapy. Genes/proteins previously associated with LCH. LCH CD207+ array: BCL2L1 (29, 30), CASP3 (29), CASP8 (31), CCL2 (10), CCL20/MIP-3a (32), CD44 (33), CCL22 (10, 34, 35), CCR6 (32), CCR7 (36), CD11b/ITGAM (25, 33, 37–39), CD11c/ITGAX (25, 33, 38, 40), CD14 (41), CD1a (reviewed in Refs. 3, 4, and 10), CD2 (25, 33, 39, 42), CD36 (43), CD40 (41, 44), CD49d/ITGA4 (25), CD54/ICAM1 (25, 33), CD58 (25, 33), CD68 (38), CD80 (41), CD83 (41, 65), CD86 (41), cdc2a/p16 (30), CD-SIGN/CD209 (46), Downloaded from CFLAR (31), CLA/SELPLG (47), c-myc (48), CXCL11/I-TAC (32), CXCL8/IL-8 (10), DC-LAMP/CD208 (41), E-cadherin/CDH1 (25, 26), FADD (31), FAS (49), Fascin/FSCN1 (50, 51), FASLG (49), FLT3LG (40), GMCSF/CSF2 (10, 52, 53), GM-CSFR/CSF2RA (53), H-ras (48), hTERT (54, 55), IFN-g (10, 23, 24), IL-10 (10, 56, 41), IL-17A (22), IL-1R1 (57), IL-1a (10, 23, 24), IL-1b (10, 23, 24), IL-2 (10, 24), IL-22 (22), IL2Ra/CD25 (39, 58, 59), IL-3 (10, 24), IL-4 (10, 23, 24), IL-6 (10), Ki67 (30, 42), Langerin/CD207 (reviewed in Refs. 3, 4, and 10), M-CSF/CSF1 (40), MDM2 (30), MIP-1a/CCL3 (10), MIP-1b/CCL4 (10), MMP12 (22, 34), MMP9 (22, 27, 28), Osteoprotegerin/TNFRSF11B (58, 59), p53 (30, 60, 61), RANKL/TNFSF11 (22, 23, 58), RANK/TNFRSF11 (56), RANTES/CCL5 (32), S100A7 (62), S100A8 (62), S100A9 (62), TARC/CCL17 (10, 34), TGF-b (10, 23, 24), TIMP2 (28), and TNFa (10, 24, 25, 63). Genes with significant differential gene expression as determined by statistical analysis are indicated with an asterisk (*). Genes not previously identified with LCH, including those involved in lymphocyte trafficking and myeloid DC-related integrins are indicated with the symbol #.

metastasis/tissue invasion (CEACAM6, MMP1, and TGFb1), in cell–cell adhesion, including TACSTD1 (64). PERP, an apo- myeloid cell maturation (CD1D, CD13, CD14, CD33, ITGA2B, ptosis effector, is one of the genes with the most signiﬁcantly ITGAX, ITGAM, and CD300F), and lymphocyte trafﬁcking decreased expression in LCH CD207 cells (65, 66). TIMP2, (SPP1, VNN1, NRP1, CCR1). A large number of the genes with a metalloproteinase inhibitor, also has reduced expression in LCH decreased or no expression in the LCH CD207 cells are involved CD207 cells (Fig. 2, Supplemental Table IIA). 4562 CELL-SPECIFIC GENE EXPRESSION IN LCH by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. http://classic.jimmunol.org

FIGURE 3. LCH-associated genes: CD3+. A, LCH CD3+ (LCH lesion CD3 versus LCH peripheral CD3) gene heatmap expression profile results of CD3+ cells isolated from LCH lesions (LCH tumor CD3) (right columns) and CD3+ cells isolated from peripheral blood of patients with active LCH (LCH peripheral CD3) (left columns). Genes associated with lymphocyte migration and activation had increased expression in tumor-infiltrating lymphocytes. This suggests that lymphocytes may play an active role in tumor formation and progression in LCH. Increased expression of FOXP3 and CTLA-4 spe- cifically suggests overrepresentation of regulatory CD3+CD4+ T cells, which are thought to arise from Ag-driven activation. Osteopontin (SPP1) showed the Downloaded from highest relative expression in LCH CD207+ cells and LCH lesion CD3+ cells. B, CD3+ (LCH lesion CD3 versus tonsil CD3) gene heatmap expression profile results of CD3+ cells isolated from LCH lesions (LCH tumor CD3) (right columns) and control CD3+ cells isolated from tonsils (left columns). These results suggest that, although LCH lesion T cells may be activated, they have significantly less expression of IL-17 and IL-21, associated with acute inflammation, compared with control tonsil T cells. Genes/proteins previously associated with LCH. LCH CD3+ arrays: CCL20/MIP3a (32), CCL5/ RANTES (32), CCR6 (32), CD40LG (41, 44, 58), CXCR3 (10), FOXP3 (56), GMCSF/CSF2 (10, 52, 53), HLA-DR/DQ (10, 41, 68), IFN-g (10, 23, 24), IL-10 (10, 41, 56), IL-17A (22), IL-1R1 (59), IL-1a (10, 23, 24), IL-1b (10, 23, 24), IL-2 (10, 23, 24), IL-22 (22), IL2Ra/CD25 (56, 58), IL-3 (10, 24), IL-4 (10, 23, 24), IL-5 (10, 24), IL-6 (10), TGF-b (10, 23, 24), TNFRSF11B (58, 59), TNFSF11/RANKL (22, 58), and TNFa (10, 24, 25, 63).

CD3+ cells. Analysis of global gene expression in the LCH lesion marily represent tumor-infiltrating lymphocytes. Differential CD3 cells also showed a consistent and distinct profile from the T cell populations within the LCH lesions may also contribute to peripheral LCH CD3 cells and the tonsil control CD3 cells. the variability between the LCH lesion CD3 datasets (correlation Therefore, although lymphocytes from peripheral blood may factor .20%) compared with the peripheral LCH CD3 datasets (corre- into the variability of the gene-expression profiles, these results lation .70%) or the control tonsil CD3 datasets (correlation suggest that the CD3+ cells isolated from the LCH lesions pri- .70%) (Supplemental Fig. 1). The Journal of Immunology 4563

Using stringent statistical criteria, 162 probe sets (129 genes) an activated Th1 (or Th17) response, including IL-17, IL-21, and showed significant differences in hybridization between cDNA de- CD40L, were significantly underexpressed in LCH lesion T cells rived from CD3 cells isolated from LCH lesions and peripheral CD3 compared with control tonsil T cells (Fig. 3B). cells isolated from peripheral blood from the same patient at the time ofbiopsy.Overall,117probesets(84genes)hadincreasedexpression Validation and 45 probe sets (45 genes) showed decreased expression in LCH We performed real-time PCR on independently amplified cDNA lesion CD3 cells (Supplemental Data 2). Differentially expressed samples to validate the microarray data. Although there were some genes in this group represent several markers of T cell activation, variations in the amplitude of differences in relative gene expression including MHC class II genes, CD58, CD74, and HSP70. In- determined by chip hybridization and real-time PCR in the group 1 terestingly, FOXP3, a gene specific to regulatory T cells (Tregs), was (LCH lesion CD207+ cells versus control CD207+ samples) and overexpressed in the LCH lesion CD3 cells, supporting a similar group 2 (LCH lesion CD3+ cells versus LCH peripheral CD3+ cells) observation reported by other investigators (56). CTLA4, an activa- samples, the trends were consistent. Quantitative differences be- tion marker associated with Tregs and inhibition of inflammation, tween the two methodologies were likely due to signal compression was also overexpressed. LCH lesion-infiltrating T cells also signifi- with the chips at very high levels of expression and baseline fluc- cantly overexpressed several genes involved in leukocyte chemo- tuation of noise at very low levels of expression (Tables I, II). taxis, including SPP1, IL-8, CCL3, CCL16, CCR1, CCR5, vitamin D Immunohistochemistry performed on a sample of LCH-asso- receptor, and plasminogen activator/urokinase receptor. SPP1 had ciated proteins also substantiated the gene-expression profiling data the highest relative expression in the group 1 (LCH CD207 versus identified by the array experiments. Immunohistochemistry was skin CD207) and group 2 (LCH lesion CD3 versus LCH peripheral performed on four LCH biopsy samples as well as normal skin and blood CD3) datasets. DUSP4, a negative regulator of MAPK, was tonsil biopsy samples used as controls for Ab staining. The results also highly overexpressed by CD207 cells and CD3 cells from the were reproducible and are representative of all LCH biopsy LCH lesions (67) (Fig. 3A, Supplemental Table IIB). samples tested. Protein products of all candidate genes identified by Comparing the 12 LCH lesion CD3 samples and the 4 control these experiments as overexpressed in LCH lesions (CD207+ or tonsil CD3 pooled samples, only 81 probe sets (70 genes) were CD3+ cells) were validated by immunohistochemistry (Fig. 4). significantly differentially expressed, all of which had decreased The results from this study call into question a current model of relative expression (Supplemental Data 3). Genes associated with LCH pathogenesis in which the epidermal Langerhans cell is the by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 4. Immunohistochemistry of LCH-associated proteins. Immunohistochemistry of hematoxylin-stained tissue with HRP-conjugated secondary Abs was used to identify protein products of genes identiﬁed by the array experiments as overexpressed in LCH lesions. Normal skin and tonsil samples were used as controls for Ab staining. Images were magniﬁed using the Olympus BX51 microscope, 340 objective. Scale bar, 100 mm. Immunohisto- chemistry images for each Ab are from a single LCH biopsy specimen, a single tonsil sample, and a single skin sample and are representative of a series of four LCH lesions, three tonsil samples, and three skin samples that were stained in the same manner. See Materials and Methods for protocol details. 4564 CELL-SPECIFIC GENE EXPRESSION IN LCH

cell of origin. The cell-specific gene expression data support gene-expression results from this study and some previously pub- a model in which CD207+ cells in LCH lesions arise directly from lished observations may be due to the experimental limitations of circulating myeloid DCs rather than from transformed or activated immunohistochemistry, including heterogeneity of LCH lesions, LCs (Fig. 5). variable specificity of Abs, and lack of reliable control tissues. It is also possible that tissue processing altered gene expression relative Discussion to cells that are fixed or frozen in tissue blocks. The concept that LCH arises from epidermal LCs was first proposed This study was designed to identify gene-expression patterns by the landmark observation by Nezelof et al. (11) that the histiocytes common to all LCH lesions. Datasets from all LCH CD207+ in LCH lesions contain Birbeck granules. Subsequently, cell-surface samples clustered in a group distinct from control epidermal LCs, expression of langerin (CD207), a protein associated with Birbeck with no distinction between the profiles from high- and low-risk granules in Ag processing in LCs, was identified as pathognomonic of samples. Additional samples will be required to power a study to normal LCs and pathological LCs in LCH (69, 70). Therefore, models determine whether there are significant cell-specific differences in of LCH have developed around the concept of aberrant activation or gene transcription in biopsies from different patient groups. malignant transformation of the resident epidermal LC, as described In addition to testing previous concepts of LCH and associated above in the “Activated-Immature” Model (Fig. 5) reviewed in Refs. 8, markers, the experimental approach of cell-specific gene expression 10, and 21). However, results from our cell-specific gene-expression revealed some previously unrecognized patterns of gene expression. analysis prompted us to question the origin of the LCH CD207+ cell. LCH CD207+ cells had increased expression of early myeloid First, global expression patterns between control epidermal LCs and markers compared with control epidermal LCs. Epidermal LCs are LCH CD207+ cells showed low overall correlation (Supplemental Fig. derived from myeloid precursors that take up residence in the skin 1). Furthermore, expression levels of specific genes central to the in embryonic life but may be replenished by peripheral monocytes “Activated-Immature” Model were not significantly different between in inflamed skin (reviewed in Ref. 71) (Fig. 5A). Therefore, it has the two groups in our experiments. been difficult to understand how LCs, which are normally restricted This study is the first to establish the gene-expression profile of to the epidermis, could give rise to such a multifocal disorder. We CD207+ cells isolated from patient LCH lesions, rather than from recently identified in mice a population of interstitial CD207+ in vitro models derived from monocytes or monocyte-derived (langerin+) DCs in most tissues, including the lung, liver, and dendritic cells (DCs). LCH-associated genes and proteins were lymph nodes. In contrast to LCs that are confined to the epidermis identified through a comprehensive literature review, and relative and derive from skin-resident hematopoietic precursors, interstitial LCH lesion CD207+ and CD3+ expression in this study are detailed CD207+ DCs are derived from blood-borne precursor cells and are in Figs. 2 and 3 and Supplemental Table II. Differences between the present in most of the tissues affected by LCH (71, 72). Therefore, by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 5. Models of LC function and LCH. A, Epidermal LCs and dermal CD207+ DCs. Two CD207+ DC populations have been described: epidermal LCs and dermal DCs (71). LCs (black star) are derived from self-renewing radioresistant precursors (black circle) that migrate to epidermis (white box) during fetal life (white sphere, dashed arrow). To maintain homeostasis after injury or inﬂammation, monocytes (gray circle) may migrate to the epidermis where they differentiate into LCs. Epidermal LCs are thought to process Ag and then undergo activation via proinﬂammatory cytokines, downregulate adhesion molecules, and migrate to draining lymph nodes, where they present Ag to effector T cells (black/white circles). CD207+ dermal DCs (white star) are a transitory cell population derived from myeloid DC precursors that migrate through the dermis (gray box) and, like LCs, present Ag to effector T cells. B, The “Ac- tivated-Immature” Model of LCH. LCH is thought to arise from pathologic epidermal LCs (black star) that undergo malignant transformation or proliferate as the result of immune dysregulation and form lesions in skin, bone, liver, lung, lymph node, bone marrow, or brain. T cells are present in LCH lesions (black/ white circles) but do not have functional interactions with the pathologic LCs. C, The Misguided Myeloid DC Precursor Model. Based on results from our study, we hypothesize that LCH lesions may arise directly from bone marrow-derived myeloid DC precursors (gray circle). Pathologic DCs acquire CD207Ag expression as they home to sites of disease. The CD207+ LCH cells express factors that attract T cells to the lesions, with enrichment of a Treg population. The Journal of Immunology 4565

langerin expression is inducible and not exclusive to epidermal Tregs are thought to develop as an Ag-specific response. Similarly, LCs. These observations, together with the results of the LCH gene- activation of T cells typically arises through Ag-dependent mech- expression studies presented in this article, suggest that LCH may anisms in vivo. The expression profile of infiltrating CD3 cells not derive from abnormal LCs but rather from abnormal circulating compared with circulating CD3 cells in patients with LCH is con- DC precursors. The observation that plasma M-CSF levels and sistent with an activated phenotype, and also supports enrichment of myeloid DC precursors (lin2HLA-DR+CD11c+) are overabundant Tregs. Therefore, the mechanisms by which T cells are activated in patients with active LCH also supports bone marrow-derived remain to be determined. Another recent study suggested LCH precursors, rather than epidermal LCs, as the source of pathologic arises through expression of IL-17A by pathologic LCs. In our study, CD207+ cells in LCH (40). IL-17A expression was not detectable in CD207 cells, as described We propose a model for LCH pathogenesis in which bone previously (80). Furthermore, relative expression of IL-17A and IL- marrow-derived myeloid DC precursors, rather than epidermal 21 was significantly decreased in LCH lesion T cells compared with LCs, migrate to sites of LCH lesions and differentiate into CD207+ control tonsil T cells, suggesting LCH is not characterized by a Th1/ cells (Fig. 5C). In the “Misguided Myeloid DC Precursor” model, Th17 response. blood-derived myeloid DCs are recruited to sites of disease where In conclusion, increased understanding of the biology is essential to they accumulate and recruit activated T cells. Together, these cells improve the treatment of individuals with LCH. However, research in produce mediators of inflammation and tissue destruction. These LCH is challenged by rare tissue samples and a lack of reliable invitro experiments do not answer the ongoing question of whether LCH or animal models. We are now able to generate cDNA probe for ex- arises from intrinsic defects in a transformed clonal population of pression studies from as little as 1500 cells. The results showed sur- precursor cells or whether it is caused by a functional response to prisinglydistinctgene-expressionprofilesbetweenCD207+ cellsfrom a pathologic stimulus. Lack of differential expression of markers LCH lesions and control epidermal LCs, which suggests that the LCH of proliferation supports a model in which CD207+ LCH cells CD207+ cells are highly differentiated or arise from a distinct cell arise from the accumulation of pathologic cells, which is also population. The relative expression of immature myeloid markers supported by other recent studies (56). from the LCH CD207+ cells further supports a model in which LCH The role of T cells in LCH is not well understood. There is no arises from a bone marrow-derived DC precursor rather than from evidence of a functional interaction between T cells and CD207+ a transformed resident epithelial LC. This study identified several new cells in vivo in LCH (41). The results from this study showed genes that had not been associated with LCH and that may be im- highly variable ratios of CD207/CD3 cells in LCH lesions (Sup- portant diagnostic markers or therapeutic targets. Genes that encode plemental Table IB). The CD3 cells are unlikely to arise primarily osteopontin-1, neuropilin-1, and vanin-1, all of which are involved from contamination of lesions at the time of biopsy because they in lymphocyte trafficking and lymphocyte–DC communication, are have unique gene-expression profiles compared with peripheral highly expressed in LCH CD207+ cells. In summary, we propose blood (Supplemental Fig. 1). Several genes with significant gene a new model in which LCH does not arise from epidermal LCs but expression in LCH CD207+ cells play roles in lymphocyte acti- from misguided myeloid DC precursors. Future studies will de- vation, recruitment, and DC interaction, including osteopontin, termine the definitive reclassification of “histiocytosis X.” neuropilin-1, and vanin-1. Osteopontin, expressed by SPP1, also known as early T lymphocyte activation 1, had the highest dif- by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. ferential expression in the LCH CD207+ versus control LC and Acknowledgments infiltrating LCH CD3+ versus peripheral LCH CD3+ datasets. It is We thank Dr. Cliona Rooney for helpful suggestions throughout this project. We acknowledge Christopher Threeton, Michael Cubbage, and Tatiana a secreted phosphoprotein with putative functions, including T cell Goltsova of the Texas Children’s Cancer Center and Hematology Service activation, Treg differentiation, homing of macrophages and Flow Facility for excellent technical support. Finally, we also thank our T cells, cell survival, and tumor progression (reviewed in Ref. 73). colleagues who helped with collection of the LCH biopsy samples, as well Osteopontin-deficient mice have deficient type 1 immunity to as patients and families for their generous participation in this study. bacterial and viral infections and are unable to form granulomas (74). Osteopontin overexpression has been found in a variety of cancers, including breast cancer, lung cancer, colorectal cancer, Disclosures http://classic.jimmunol.org stomach cancer, ovarian cancer, and melanoma (75). One of the major K.L.M. owns common stock in the Johnson & Johnson Company. Other- osteopontin ligands, CD44, is also significantly overexpressed in wise, the authors have no financial conflicts of interest. LCH CD207 cells compared with control LCs. Neuropilin-1, expressed by NRP1, is a membrane-bound coreceptor to a tyrosine kinase receptor both vascular endothelial growth factor and semaphorin References 1. Carstensen, H., and K. Ornvold. 1993. [Langerhans-cell histiocytosis (histiocy- molecules. Neuropilin-1 is involved in axonal guidance of nerves. In tosis X) in children]. Ugeskr. Laeger 155: 1779–1783. Downloaded from the immune system, it is important in establishing the T cell–DC 2. Arico`, M., M. Girschikofsky, T. Geńe´reau, C. Klersy, K. McClain, N. Grois, synapse, as well as migration and adhesion of thymocytes (76, 77). J. F. Emile, E. Lukina, E. De Juli, and C. Danesino; Report from the International Registry of the Histiocyte Society. 2003. Langerhans cell histiocytosis in adults. Vanin-1, expressed by VNN1, is a GPI-anchored molecule expressed Eur. J. Cancer 39: 2341–2348. by perivascular thymic stromal cells that is essential in the migration 3. Arceci, R. J. 1999. The histiocytoses: the fall of the Tower of Babel. Eur. J. of hematopoietic progenitor cells to the thymus (78). Mice deficient in Cancer 35: 747–767, discussion 767–769. Coxiella bur- 4. McClain, K. L., Y. Natkunam, and S. H. Swerdlow. 2004. Atypical cellular vanin-1 are unable to form granulomas in response to disorders. Hematology (Am Soc Hematol Educ Program) 283–296. netti infection (79). Increased expression of SPP1, NRP1,andVNN1 5. Gadner, H., N. Grois, U. Po¨tschger, M. Minkov, M. Arico`, J. Braier, by LCH CD207+ cells suggests a potentially interesting role for these V. Broadbent, J. Donadieu, J. I. Henter, R. McCarter, and S. Ladisch; Histiocyte genes in CD207+ cell-mediated recruitment of T cells to LCH lesions. Society. 2008. Improved outcome in multisystem Langerhans cell histiocytosis is associated with therapy intensification. Blood 111: 2556–2562. Multiple searches for a common Ag in LCH failed to identify an 6. Gadner, H., N. Grois, M. Arico, V. Broadbent, A. Ceci, A. Jakobson, D. Komp, infectious etiology (reviewed in Ref. 4). A recent study found that J. Michaelis, S. Nicholson, U. Po¨tschger, et al; and Histiocyte Society. 2001. A Tregs were enriched in LCH lesions, as well as in peripheral blood of randomized trial of treatment for multisystem Langerhans’ cell histiocytosis. J. Pediatr. 138: 728–734. patients with active LCH. The same study found polyclonal TCR 7. Allen, C. E., and K. L. McClain. 2007. Langerhans cell histiocytosis: a review of rearrangements in DNA isolated from frozen biopsy samples (56). past, current and future therapies. Drugs Today (Barc) 43: 627–643. 4566 CELL-SPECIFIC GENE EXPRESSION IN LCH

8. Egeler, R. M., N. E. Annels, and P. C. Hogendoorn. 2004. Langerhans cell 36. Fleming, M. D., J. L. Pinkus, M. V. Fournier, S. W. Alexander, C. Tam, M. Loda, histiocytosis: a pathologic combination of oncogenesis and immune dysregula- S. E. Sallan, K. E. Nichols, D. F. Carpentieri, G. S. Pinkus, and B. J. Rollins. 2003. tion. Pediatr. Blood Cancer 42: 401–403. Coincident expression of the chemokine receptors CCR6 and CCR7 by pathologic 9. Fadeel, B., and J. I. Henter. 2003. Langerhans-cell histiocytosis: neoplasia or Langerhans cells in Langerhans cell histiocytosis. Blood 101: 2473–2475. unbridled inflammation? Trends Immunol. 24: 409–410, author reply 410–411. 37. Emile, J. F., S. Fraitag, M. Leborgne, Y. de Prost, and N. Brousse. 1994. 10. Laman, J. D., P. J. Leenen, N. E. Annels, P. C. Hogendoorn, and R. M. Egeler. Langerhans’ cell histiocytosis cells are activated Langerhans’ cells. J. Pathol. 2003. Langerhans-cell histiocytosis ‘insight into DC biology’. Trends Immunol. 174: 71–76. 24: 190–196. 38. Ornvold, K., E. Ralfkiaer, and H. Carstensen. 1990. Immunohistochemical study 11. Nezelof, C., F. Basset, and M. F. Rousseau. 1973. Histiocytosis X histogenic of the abnormal cells in Langerhans cell histiocytosis (histiocytosis x). Virchows arguments for a Langerhans cell origin. Biomedicine 18: 365–371. Arch. A Pathol. Anat. Histopathol. 416: 403–410. 12. Willman, C. L., and K. L. McClain. 1998. An update on clonality, cytokines, and 39. Ruco, L. P., D. Remotti, F. Monardo, S. Uccini, M. L. Cristiani, A. Modesti, and viral etiology in Langerhans cell histiocytosis. Hematol. Oncol. Clin. North Am. C. D. Baroni. 1988. Letterer-Siwe disease: immunohistochemical evidence for 12: 407–416. a proliferative disorder involving immature cells of Langerhans lineage. Virch- 13. Favara, B. E., and A. Steele. 1997. Langerhans cell histiocytosis of lymph nodes: ows Arch. A Pathol. Anat. Histopathol. 413: 239–247. a morphological assessment of 43 biopsies. Pediatr.Pathol. Lab. Med. 17: 769–787. 40. Rolland, A., L. Guyon, M. Gill, Y. H. Cai, J. Banchereau, K. McClain, and 14. Schroeder, A., O. Mueller, S. Stocker, R. Salowsky, M. Leiber, M. Gassmann, A. K. Palucka. 2005. Increased blood myeloid dendritic cells and dendritic cell- S. Lightfoot, W. Menzel, M. Granzow, and T. Ragg. 2006. The RIN: an RNA poietins in Langerhans cell histiocytosis. J. Immunol. 174: 3067–3071. integrity number for assigning integrity values to RNA measurements. BMC 41. Geissmann, F., Y. Lepelletier, S. Fraitag, J. Valladeau, C. Bodemer, M. Debre´, Mol. Biol. 7: 3. M. Leborgne, S. Saeland, and N. Brousse. 2001. Differentiation of Langerhans 15. Gentleman, R. C., V. J. Carey, D. M. Bates, B. Bolstad, M. Dettling, S. Dudoit, cells in Langerhans cell histiocytosis. Blood 97: 1241–1248. B. Ellis, L. Gautier, Y. Ge, J. Gentry, et al. 2004. Bioconductor: open software 42. Hage, C., C. L. Willman, B. E. Favara, and P. G. Isaacson. 1993. Langerhans’ development for computational biology and bioinformatics. Genome Biol. 5: R80. cell histiocytosis (histiocytosis X): immunophenotype and growth fraction. Hum. 16. Tusher, V. G., R. Tibshirani, and G. Chu. 2001. Significance analysis of mi- Pathol. 24: 840–845. croarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. USA 43. Santegoets, S. J., S. Gibbs, K. Kroeze, R. van de Ven, R. J. Scheper, 98: 5116–5121. 17. Saeed, A. I., V. Sharov, J. White, J. Li, W. Liang, N. Bhagabati, J. Braisted, C. A. Borrebaeck, T. D. de Gruijl, and M. Lindstedt. 2008. Transcriptional profiling M. Klapa, T. Currier, M. Thiagarajan, et al. 2003. TM4: a free, open-source of human skin-resident Langerhans cells and CD1a+ dermal dendritic cells: dif- system for microarray data management and analysis. Biotechniques 34: 374–378. ferential activation states suggest distinct functions. J. Leukoc. Biol. 84: 143–151. 18. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression 44. Egeler, R. M., B. E. Favara, J. D. Laman, and E. Claassen. 2000. Abundant data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. expression of CD40 and CD40-ligand (CD154) in paediatric Langerhans cell Methods 25: 402–408. histiocytosis lesions. Eur. J. Cancer 36: 2105–2110. 19. Pradervand, S., A. Paillusson, J. Thomas, J. Weber, P. Wirapati, O. Hagenbu¨chle, 45. Misery, L., N. Rougier, B. Crestani, M. Faure, A. Claudy, D. Schmitt, and and K. Harshman. 2008. Affymetrix Whole-Transcript Human Gene 1.0 STarray is C. Vincent. 1999. Presence of circulating abnormal CD34+ progenitors in adult highly concordant with standard 39 expression arrays. Biotechniques 44: 759–762. Langerhans cell histiocytosis. Clin. Exp. Immunol. 117: 177–182. 20. Singh, R., R. J. Maganti, S. V. Jabba, M. Wang, G. Deng, J. D. Heath, N. Kurn, 46. Soilleux, E. J., and N. Coleman. 2001. Langerhans cells and the cells of and P. Wangemann. 2005. Microarray-based comparison of three amplification Langerhans cell histiocytosis do not express DC-SIGN. Blood 98: 1987–1988. methods for nanogram amounts of total RNA. Am. J. Physiol. Cell Physiol. 288: 47. Simonitsch, I., C. W. Kopp, I. Mosberger, B. Volc-Platzer, and T. Radaszkiewicz. C1179–C1189. 1996. Expression of the monoclonal antibody HECA-452 defined E-selectin li- 21. Bechan, G. I., R. M. Egeler, and R. J. Arceci. 2006. Biology of Langerhans cells gands in Langerhans cell histiocytosis. Virchows Arch. 427: 477–481. and Langerhans cell histiocytosis. Int. Rev. Cytol. 254: 1–43. 48. Abdelatif, O. M., F. W. Chandler, C. G. Pantazis, and B. S. McGuire. 1990. 22. Coury, F., N. Annels, A. Rivollier, S. Olsson, A. Santoro, C. Speziani, O. Azocar, Enhanced expression of c-myc and H-ras oncogenes in Letterer-Siwe disease. A M. Flacher, S. Djebali, J. Tebib, et al. 2008. Langerhans cell histiocytosis reveals sequential study using colorimetric in situ hybridization. Arch. Pathol. Lab. Med. a new IL-17A-dependent pathway of dendritic cell fusion. Nat. Med. 14: 81–87. 114: 1254–1260. 23. de Graaf, J. H., R. Y. Tamminga, A. Dam-Meiring, W. A. Kamps, and 49. Petersen, B. L., P. Rengtved, M. I. Bank, and H. Carstensen. 2003. High ex- W. Timens. 1996. The presence of cytokines in Langerhans’ cell histiocytosis. J. pression of markers of apoptosis in Langerhans cell histiocytosis. Histopathol- Pathol. 180: 400–406. ogy 42: 186–193. 24. Egeler, R. M., B. E. Favara, M. van Meurs, J. D. Laman, and E. Claassen. 1999. 50. Jaffe, R., D. DeVaughn, and E. Langhoff. 1998. Fascin and the differential di- Differential In situ cytokine profiles of Langerhans-like cells and T cells in Pediatr. Dev. Pathol. by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. agnosis of childhood histiocytic lesions. 1: 216–221. Langerhans cell histiocytosis: abundant expression of cytokines relevant to 51. Pinkus, G. S., M. A. Lones, F. Matsumura, S. Yamashiro, J. W. Said, and disease and treatment. Blood 94: 4195–4201. J. L. Pinkus. 2002. Langerhans cell histiocytosis immunohistochemical expres- 25. de Graaf, J. H., R. Y. Tamminga, W. A. Kamps, and W. Timens. 1995. Ex- sion of fascin, a dendritic cell marker. Am. J. Clin. Pathol. 118: 335–343. pression of cellular adhesion molecules in Langerhans cell histiocytosis and 52. Emile, J. F., E. Tartour, L. Brugie`res, J. Donadieu, F. Le Deist, I. Charnoz, A. Fischer, normal Langerhans cells. Am. J. Pathol. 147: 1161–1171. W. H. Fridman, and N. Brousse. 1994. Detection of GM-CSF in the sera of children 26. Geissmann, F., J. F. Emile, P. Andry, C. Thomas, S. Fraitag, Y. De Prost, and with Langerhans’ cell histiocytosis. Pediatr. Allergy Immunol. 5: 162–163. N. Brousse. 1997. Lack of expression of E-cadherin is associated with dissemi- 53. Emile, J. F., S. Fraitag, P. Andry, M. Leborgne, A. Lellouch-Tubiana, and nation of Langerhans’ cell histiocytosis and poor outcome. J. Pathol. 181: 301–304. N. Brousse. 1995. Expression of GM-CSF receptor by Langerhans’ cell histio- 27. da Costa, C. E., N. E. Annels, C. M. Faaij, R. G. Forsyth, P. C. Hogendoorn, and cytosis cells. Virchows Arch. 427: 125–129. R. M. Egeler. 2005. Presence of osteoclast-like multinucleated giant cells in the bone 54. Bechan, G. I., A. K. Meeker, A. M. De Marzo, F. Racke, R. Jaffe, E. Sugar, and and nonostotic lesions of Langerhans cell histiocytosis. J. Exp. Med. 201: 687–693. R. J. Arceci. 2008. Telomere length shortening in Langerhans cell histiocytosis. 28. Hayashi, T., W. L. Rush, W. D. Travis, L. A. Liotta, W. G. Stetler-Stevenson, and http://classic.jimmunol.org Br. J. Haematol. 140: 420–428. V. J. Ferrans. 1997. Immunohistochemical study of matrix metalloproteinases 55. da Costa, C. E., R. M. Egeler, M. Hoogeboom, K. Szuhai, R. G. Forsyth, and their tissue inhibitors in pulmonary Langerhans’ cell granulomatosis. Arch. M. Niesters, R. R. de Krijger, A. Tazi, P. C. Hogendoorn, and N. E. Annels. Pathol. Lab. Med. 121: 930–937. 2007. Differences in telomerase expression by the CD1a+ cells in Langerhans 29. Amir, G., and M. Weintraub. 2008. Association of cell cycle-related gene cell histiocytosis reflect the diverse clinical presentation of the disease. J. Pathol. products and NF-kappaB with clinical parameters in Langerhans cell histiocy- 212: 188–197. tosis. Pediatr. Blood Cancer 50: 304–307. 56. Senechal, B., G. Elain, E. Jeziorski, V. Grondin, N. Patey-Mariaud de Serre, 30. Schouten, B., R. M. Egeler, P. J. Leenen, A. H. Taminiau, L. J. van den Broek, F. Jaubert, K. Beldjord, A. Lellouch, C. Glorion, M. Zerah, et al. 2007. Ex- and P. C. Hogendoorn. 2002. Expression of cell cycle-related gene products in pansion of regulatory T cells in patients with Langerhans cell histiocytosis. PLoS Downloaded from Langerhans cell histiocytosis. J. Pediatr. Hematol. Oncol. 24: 727–732. 31. Bank, M. I., C. Gudbrand, P. Rengtved, H. Carstensen, B. Fadeel, J. I. Henter, and Med. 4: e253. B. L. Petersen. 2005. Immunohistochemical detection of the apoptosis-related 57. Rosso, D. A., M. F. Ripoli, A. Roy, R. A. Diez, M. E. Zelazko, and J. L. Braier. proteins FADD, FLICE, and FLIP in Langerhans cell histiocytosis. J. Pediatr. 2003. Serum levels of interleukin-1 receptor antagonist and tumor necrosis Hematol. Oncol. 27: 301–306. factor-alpha are elevated in children with Langerhans cell histiocytosis. J. Pe- 32. Annels, N. E., C. E. Da Costa, F. A. Prins, A. Willemze, P. C. Hogendoorn, and diatr. Hematol. Oncol. 25: 480–483. R. M. Egeler. 2003. Aberrant chemokine receptor expression and chemokine 58. Ishii, R., A. Morimoto, S. Ikushima, T. Sugimoto, K. Asami, F. Bessho, K. Kudo, production by Langerhans cells underlies the pathogenesis of Langerhans cell Y. Tsunematu, J. Fujimoto, and S. Imashuku. 2006. High serum values of soluble histiocytosis. J. Exp. Med. 197: 1385–1390. CD154, IL-2 receptor, RANKL and osteoprotegerin in Langerhans cell histio- 33. de Graaf, J. H., R. Y. Tamminga, W. A. Kamps, and W. Timens. 1994. Lang- cytosis. Pediatr. Blood Cancer 47: 194–199. erhans’ cell histiocytosis: expression of leukocyte cellular adhesion molecules 59. Rosso, D. A., A. Roy, M. Zelazko, and J. L. Braier. 2002. Prognostic value of suggests abnormal homing and differentiation. Am. J. Pathol. 144: 466–472. soluble interleukin 2 receptor levels in Langerhans cell histiocytosis. Br. J. 34. Rust, R., J. Kluiver, L. Visser, G. Harms, T. Blokzijl, W. Kamps, S. Poppema, Haematol. 117: 54–58. and A. van den Berg. 2006. Gene expression analysis of dendritic/Langerhans 60. Bank, M. I., P. Rengtved, H. Carstensen, and B. L. Petersen. 2002. p53 ex- cells and Langerhans cell histiocytosis. J. Pathol. 209: 474–483. pression in biopsies from children with Langerhans cell histiocytosis. J. Pediatr. 35. Vulcano, M., C. Albanesi, A. Stoppacciaro, R. Bagnati, G. D’Amico, S. Struyf, Hematol. Oncol. 24: 733–736. P. Transidico, R. Bonecchi, A. Del Prete, P. Allavena, et al. 2001. Dendritic cells 61. Weintraub, M., K. G. Bhatia, R. S. Chandra, I. T. Magrath, and S. Ladisch. 1998. as a major source of macrophage-derived chemokine/CCL22 in vitro and in vivo. p53 expression in Langerhans cell histiocytosis. J. Pediatr. Hematol. Oncol. 20: Eur. J. Immunol. 31: 812–822. 12–17. The Journal of Immunology 4567

62. Nakajima, T., S. Watanabe, Y. Sato, Y. Shimosato, M. Motoi, and K. Lennert. 71. Merad, M., F. Ginhoux, and M. Collin. 2008. Origin, homeostasis and function 1982. S-100 protein in Langerhans cells, interdigitating reticulum cells and of Langerhans cells and other langerin-expressing dendritic cells. Nat. Rev. histiocytosis X cells. Gann 73: 429–432. Immunol. 8: 935–947. 63. Wu, W. S., and K. L. McClain. 1997. DNA polymorphisms and mutations of the 72. Ginhoux, F., F. Tacke, V. Angeli, M. Bogunovic, M. Loubeau, X. M. Dai, tumor necrosis factor-alpha (TNF-alpha) promoter in Langerhans cell histiocy- E. R. Stanley, G. J. Randolph, and M. Merad. 2006. Langerhans cells arise from tosis (LCH). J. Interferon Cytokine Res. 17: 631–635. monocytes in vivo. Nat. Immunol. 7: 265–273. 64. Trzpis, M., P. M. McLaughlin, L. M. de Leij, and M. C. Harmsen. 2007. Epi- 73. Wang, K. X., and D. T. Denhardt. 2008. Osteopontin: role in immune regulation thelial cell adhesion molecule: more than a carcinoma marker and adhesion and stress responses. Cytokine Growth Factor Rev. 19: 333–345. molecule. Am. J. Pathol. 171: 386–395. 74. Ashkar, S., G. F. Weber, V. Panoutsakopoulou, M. E. Sanchirico, M. Jansson, 65. Hildebrandt, T., J. Preiherr, N. Tarbe´, S. Klostermann, G. N. Van Muijen, and S. Zawaideh, S. R. Rittling, D. T. Denhardt, M. J. Glimcher, and H. Cantor. 2000. U. H. Weidle. 2000. Identification of THW, a putative new tumor suppressor Eta-1 (osteopontin): an early component of type-1 (cell-mediated) immunity. gene. Anticancer Res. 20(5A): 2801–2809. Science 287: 860–864. 66. Ihrie, R. A., M. R. Marques, B. T. Nguyen, J. S. Horner, C. Papazoglu, 75. Rodrigues, L. R., J. A. Teixeira, F. L. Schmitt, M. Paulsson, and H. Lindmark- R. T. Bronson, A. A. Mills, and L. D. Attardi. 2005. Perp is a p63-regulated gene Ma¨nsson. 2007. The role of osteopontin in tumor progression and metastasis in essential for epithelial integrity. Cell 120: 843–856. breast cancer. Cancer Epidemiol. Biomarkers Prev. 16: 1087–1097. 67. Guan, K. L., and E. Butch. 1995. Isolation and characterization of a novel dual 76. Lepelletier, Y., S. Smaniotto, R. Hadj-Slimane, D. M. Villa-Verde, specific phosphatase, HVH2, which selectively dephosphorylates the mitogen- A. C. Nogueira, M. Dardenne, O. Hermine, and W. Savino. 2007. Control of activated protein kinase. J. Biol. Chem. 270: 7197–7203. human thymocyte migration by Neuropilin-1/Semaphorin-3A-mediated inter- 68. McClain, K. L., P. Laud, W. S. Wu, and M. S. Pollack. 2003. Langerhans cell actions. Proc. Natl. Acad. Sci. USA 104: 5545–5550. histiocytosis patients have HLA Cw7 and DR4 types associated with specific 77. Tordjman, R., Y. Lepelletier, V. Lemarchandel, M. Cambot, P. Gaulard, clinical presentations and no increased frequency in polymorphisms of the tumor O. Hermine, and P. H. Romeó. 2002. A neuronal receptor, neuropilin-1, is essential necrosis factor alpha promoter. Med. Pediatr. Oncol. 41: 502–507. for the initiation of the primary immune response. Nat. Immunol. 3: 477–482. 69. Valladeau, J., O. Ravel, C. Dezutter-Dambuyant, K. Moore, M. Kleijmeer, 78. Aurrand-Lions, M., F. Galland, H. Bazin, V. M. Zakharyev, B. A. Imhof, and Y. Liu, V. Duvert-Frances, C. Vincent, D. Schmitt, J. Davoust, et al. 2000. P. Naquet. 1996. Vanin-1, a novel GPI-linked perivascular molecule involved in Langerin, a novel C-type lectin specific to Langerhans cells, is an endocytic thymus homing. Immunity 5: 391–405. receptor that induces the formation of Birbeck granules. Immunity 12: 71–81. 79. Meghari, S., C. Berruyer, H. Lepidi, F. Galland, P. Naquet, and J. L. Mege. 2007. 70. Valladeau, J., V. Duvert-Frances, J. J. Pin, C. Dezutter-Dambuyant, C. Vincent, Vanin-1 controls granuloma formation and macrophage polarization in Coxiella C. Massacrier, J. Vincent, K. Yoneda, J. Banchereau, C. Caux, et al. 1999. The burnetii infection. Eur. J. Immunol. 37: 24–32. monoclonal antibody DCGM4 recognizes Langerin, a protein specific of 80. Allen, C. E., and K. L. McClain. 2009. Interleukin-17A is not expressed by Langerhans cells, and is rapidly internalized from the cell surface. Eur. J. Im- CD207(+) cells in Langerhans cell histiocytosis lesions. Nat. Med. 15: 483–484, munol. 29: 2695–2704. author reply 484–485. by guest on September 30, 2021. Copyright 2010 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from