|
|
||||||||


* Baylor Institute for Immunology Research, Dallas, TX 75204;
University of Texas Southwestern Medical Center, Dallas, TX 75390; and
Baylor College of Medicine, Texas Childrens Cancer Center, Houston, TX 77030
| Abstract |
|---|
|
|
|---|
2-fold) and M-CSF (
4-fold). Higher levels of these cytokines correlated with patients having more extensive disease. Serum levels of FLT3-L and M-CSF were highest in high risk patients with extensive skin and/or multisystem involvement. Finally, patients with bone lesions had relatively higher levels of M-CSF and of stem cell factor. Thus, early hemopoietic cytokines such as FLT3-L, stem cell factor, and M-CSF maybe relevant in LCH pathogenesis and might be considered as novel therapeutic targets. | Introduction |
|---|
|
|
|---|
) infiltrates in the lesions (2).
Numerous recent studies have highlighted the plasticity of monocytes, circulating blood precursors that can be induced to differentiate into M
(with GM-CSF or M-CSF) (3, 4, 5) or dendritic cells (DCs) in response to GM-CSF together with other cytokines such as IL-4 (6, 7, 8), IFN-
(9, 10, 11, 12), TNF (13), and IL-15 (14). However, the blood contains another precursor population, i.e., lineage-negative (lin) HLA-DR+CD11c+CD1a+ cells with multiple differentiation potential (15). Indeed, these cells can give rise to either M
or DCs depending on cytokine environment (16).
Several cytokines, which we collectively call DC-poietins, mobilize DCs in vivo; the cytokines either increase their numbers (G-CSF, GM-CSF, and fms-like tyrosine kinase ligand (FLT3-L) or activate them (IFN-
). Among those, FLT3-L is a growth factor for early hemopoietic progenitor cells (17). FLT3-L has been shown to expand distinct subsets of DCs) in mice in vivo (18, 19). Administration of FLT3-L to healthy human volunteers (HVs) and cancer patients leads to mobilization of both human DC subsets present in the blood, i.e., plasmacytoid DCs (pDCs) and myeloid DCs (mDCs) (15, 20). FLT3-L can be detected in the serum of healthy HVs (21), and it can be elevated in diseases with underlying DC pathology. For example, systemic lupus erythematosus, an autoimmune disease where interplay between distinct DC subsets may play a role (12), is associated with high serum levels of FLT3-L (21). Therefore, we surmised that dysregulation of DC-poietins might also explain LCH-associated pathology.
| Materials and Methods |
|---|
|
|
|---|
Peripheral blood specimens and clinical information from 24 patients with LCH (23 pediatric and 1 adult patient) were obtained after receiving written consent on a protocol approved by the local committee for use of human subjects in research (IRB no. H-6927). All LCH patients were treated at the Texas Childrens Cancer Center and evaluated by K. McClain. The status of their disease and time of specimen relative to treatments are given in Table I. This pilot study was done using archival, retrospective samples. Serum samples from 21 patients and PBMC samples from 10 patients were available for analysis. In some of the patients, blood could be collected at different time points in the course of disease (Table I). Controls were either adult HVs (n = 19) or children visiting the clinic for the reasons other than autoimmunity or infectious disease (n = 15; IRB no. 0199017).
|
PBMCs were isolated by Ficoll and frozen in 10% DMSO. Thawed PBMCs were analyzed in batches by flow cytometry using the following fluorochrome-conjugated anti-human Abs: LINEAGE-FITC mixture (containing CD3, CD14, CD16, CD19, CD20, and CD56), CD123-PE, HLA-DR-PerCP, and CD11c-APC. Samples were acquired on a FACSCalibur and analyzed with CellQuest software (BD Biosciences). mDC and pDC subsets were defined by simultaneous lack of lineage markers, HLA-DR expression, and CD11c or CD123 positivity, respectively.
Quantitation of serum FLT3-L, M-CSF, and stem cell factor (SCF)
Quantitation was conducted with ELISA using respective immunoassay kits (R&D Systems). Standards and undiluted serum samples were run in duplicate wells for each immunoassay. Serum samples were pipetted into the wells of the microplate (precoated with specific mAbs). After washing, HRP-linked specific polyclonal Abs were added to the wells. After another wash, a substrate solution consisting of both hydrogen peroxide and tetramethylbenzidine was added to the wells. This reaction was then stopped with sulfuric acid. The OD was then measured at 450 nm with a microplate reader (with the correction wavelength set at 540 nm). A four-parameter logistic standard curve was generated from the standards, and mean concentrations were then calculated in picograms per milliliter.
| Results |
|---|
|
|
|---|
Given the predominant LC and M
phenotype in LCH lesions, we surmised that the major alterations in LCH may be at the level of circulating myeloid DC precursors. To this end, we have analyzed the frequency of DC subsets in the blood in a cohort of seven LCH patients. pDCs and mDCs were identified in whole blood as lin HLA-DR+ cells expressing either CD123 or CD11c, respectively (Fig. 1a). We found increased frequency of lin HLA-DR+CD11c+ cells in the majority of patients as compared with age-matched controls (Fig. 1b). There, LCH patients showed a median of 0.56% of lin HLA-DR+CD11c+ cells (range, 0.232; n = 10), pediatric controls showed a median of 0.25% (range, 0.020.92; n = 55), and the median of adult controls was 0.33% (range, 0.210.44; n = 8). These results suggested that LCH sera may contain a factor(s) that mobilizes DC and/or their precursors.
|
We have previously shown that FLT3 ligand (FLT3-L) mobilizes both DC subsets upon repeated administration (15). Thus, increased serum levels of FLT3-L could explain increased numbers of circulating DCs. Indeed, LCH patients showed significantly elevated (p < 0.05) serum levels of FLT3-L as compared with control cohorts, either pediatric or adult HVs (Fig. 2 and Table I). The median serum concentration of FLT3-L in samples from LCH patients was 91 pg/ml (range, 1404; n = 31). Adult and pediatric controls showed a median of 43 pg/ml (range, 2387; n = 19) and 50 pg/ml (range: 0153; n = 15), respectively. When compared with the median of pediatric and/or adult controls, 11 samples from LCH patients showed >2-fold increased serum concentration of FLT3-L. Thus, increased FLT3-L may mobilize blood mDCs leading to increased influx of these cells into LCH lesions.
|
Because LCH lesions are also characterized by the infiltration of M
and eosinophils, we have analyzed serum levels of M-CSF and SCF, two cytokines involved in the mobilization and differentiation of these cells. As shown in Fig. 3 and Table I, LCH patients displayed significantly higher levels of serum M-CSF (p < 0.0001) than did healthy adult controls with median serum concentration of 803 pg/ml (range, 204544; n = 31) and 268 pg/ml (range, 97560; n = 19), respectively. When compared with the median of adult controls, 22 samples from LCH patients showed >2-fold increased serum concentration of M-CSF. SCF concentration was not significantly affected (data not shown). Altogether, these results suggest that early hemopoietic cytokines, such as FLT3-L and M-CSF, which are involved in DC/M
differentiation, may play a role in LCH etiopathogenesis.
|
We next analyzed whether serum cytokines would correlate with clinical status and disease activity. Indeed, in the cohort of patients in whom serial specimens could be analyzed (patients 8, 9, 11, 14, 17, and 21), a decline in levels of all three cytokines was noted when the LCH responded to therapy (Table I). Moreover, the appearance of new lesions was accompanied by elevations of each cytokine (patients 8 and 9). When patients were categorized according to the type of disease (single or multiple bone lesions, bone and skin or lymph node, or multisystem LCH), several patterns of cytokine expression were evident (Fig. 4). Thus, M-CSF and SCF levels were lower in patients with bone disease only, whether in single or multiple lesions. However, M-CSF levels were markedly increased in patients who, in addition to bone disease, also showed skin or lymph node involvement (Fig. 4). These patients also showed markedly increased levels of FLT3-L. Patients with multisystem LCH were the most ill, having large livers and spleens in addition to bone and skin lesions. These patients had highest M-CSF levels, most particularly when they were near death (Fig. 4, patients 7, 8, 11, and 18). SCF levels were not clearly associated with disease status in the multisystem disease group.
|
| Discussion |
|---|
|
|
|---|
20% for all patients and is especially high in children with involvement of the liver, of whom up to 50% die despite treatment. Our results demonstrate significantly elevated levels of DC-poietins, FLT3-L, and M-CSF, most particularly in patients having more extensive disease.
Similarly to FLT3-L and SCF, M-CSF is an early hemopoietic growth factor and shares structural homology with both SCF (27) and FLT3-L (17). M-CSF is a multipotent cytokine that controls macrophage differentiation and function (28), promotes angiogenesis (29), and is essential for osteoclast function (30). Until recently, M-CSF was not considered a DC-poietin and M-CSF-deficient (op/op) mice demonstrate normal LC differentiation (31). However, there is emerging evidence that M-CSF may play a role in LC differentiation upon contact with cutaneous fibroblasts in synergy with TGF-
1 (32, 33). Further, recent studies by Rivollier et al. show that M-CSF together with RANK-L allows in vitro trans-differentiation of immature DCs into osteoclasts (34). Thus, elevated M-CSF could explain osteolytic lesions in LCH patients via promotion of trans-differentiation of immature mDCs mobilized by FLT3-L into osteoclasts. Such model is consistent with the increased expression of M-CSF gene observed in bronchoalveolar lavage cells of a patient with pulmonary LCH (35).
Thus, DC-poietins maybe relevant in LCH pathogenesis and might be considered as novel therapeutic targets. The next question to address is why LCH patients display high serum levels of these two cytokines. Is that related to alterations at genomic level (for example, mutations in gene promoter), or at protein level (for example, alterations in regulatory feedback)? Our results warrant further investigations that may unravel LCH pathogenesis.
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by grants from Baylor Health Care Systems Foundation, The Histiocytosis Association of America (J.B., A.K.P.), and the National Institutes of Health (CA78846 to J.B.). J.B. is the recipient of the Caruth Chair for Transplant Immunology Research. A.K.P. is the recipient of Michael A. Ramsay Chair for Cancer Immunology Research. ![]()
2 Address correspondence and reprint requests to Dr. A. Karolina Palucka, Baylor Institute for Immunology Research, 3434 Live Oak, Dallas, TX 75204 or Dr. Kenneth McClain, Baylor College of Medicine, Texas Childrens Cancer Center, Houston, Texas. E-mail address: karolinp{at}baylorhealth.edu or kmcclain{at}txccc.org ![]()
3 These authors codirected the presented work. ![]()
4 Abbreviations used in this paper: LCH, Langerhans cell histiocytosis; M
, macrophage; DC, dendritic cell; pDC, plasmacytoid DC; mDC, myeloid DC; FLT3-L, fms-like tyrosine kinase ligand; HV, healthy volunteer; SCF, stem cell factor. ![]()
Received for publication August 20, 2004. Accepted for publication December 17, 2004.
| References |
|---|
|
|
|---|
. J. Exp. Med. 179:1109.
and granulocyte-macrophage colony-stimulating factor differentiate peripheral blood monocytes into potent antigen-presenting cells. J. Leukocyte Biol. 64:358.[Abstract]
in systemic lupus erythematosus. Science 294:1540.
1 prevents the noncognate maturation of human dendritic Langerhans cells. J. Immunol. 162:4567.
1 induces the differentiation of CD34+ hematopoietic progenitor cells into Langerhans cells under serum-free conditions without granulocyte-macrophage colony-stimulating factor. J. Invest. Dermatol. 120:256.[Medline]
This article has been cited by other articles:
![]() |
B. Fancke, M. Suter, H. Hochrein, and M. O'Keeffe M-CSF: a novel plasmacytoid and conventional dendritic cell poietin Blood, January 1, 2008; 111(1): 150 - 159. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. S. Lo, P. Gorak-Stolinska, V. Bachy, M. A. Ibrahim, D. M. Kemeny, and J. Maher Modulation of dendritic cell differentiation by colony-stimulating factor-1: role of phosphatidylinositol 3'-kinase and delayed caspase activation J. Leukoc. Biol., December 1, 2007; 82(6): 1446 - 1454. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Gogolak, B. Rethi, I. Szatmari, A. Lanyi, B. Dezso, L. Nagy, and E. Rajnavolgyi Differentiation of CD1a- and CD1a+ monocyte-derived dendritic cells is biased by lipid environment and PPAR{gamma} Blood, January 15, 2007; 109(2): 643 - 652. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Vermaelen and R. Pauwels Pulmonary Dendritic Cells Am. J. Respir. Crit. Care Med., September 1, 2005; 172(5): 530 - 551. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |