Activated Conventional T-Cells Are Present in Langerhans Cell Histiocytosis Lesions Despite the Presence of Immune Suppressive Cytokines

Abstract

Langerhans cell histiocytosis (LCH) lesions are characterized by neoplastic CD1a⁺/Langerin⁺ histiocytes (LCH-cells) and display many features of chronic inflammation. Cancer cells can escape immune-surveillance through intra-tumoral secretion of immune-suppressive cytokines. We therefore studied by immunohistochemistry the local cytokine milieu and phenotypic characteristics of T-cells and LCH-cells present in LCH lesions collected from 25 therapy naïve patients. LCH biopsies predominantly expressed interleukin-10 (IL-10) (10/25), transforming growth factor-beta (TGF-β) (9/25), or both cytokines (6/25). The absolute number of CD3⁺T-cells and the CD3⁺FOXP3⁻ conventional cell (T-CONV) versus the CD3⁺FOXP3⁺ regulatory T-cell (T-REG) was comparable for each suppressive cytokine profile (5:1). IL-10-expressing lesions contained, however, a higher proportion of T-CONV expressing the activation markers CD25 98% (38%–100%) and inducible costimulatory molecule (ICOS) 86% (47%–100%) than lesions wherein solely TGF-β was detected (CD25⁺ 20% (6%–54%); ICOS⁺ 29% (7%–51%)). Virtually all T-REG expressed CD25 and ICOS in IL-10 lesions, whereas TGF-β⁺ lesions contained a lower proportion of ICOS⁺ T-REG (P=0.05). IL-10⁺ lesions contained more LCH-cells expressing high intensity of ICOS ligand (ICOSL) compared with TGF-β⁺ lesions (P=0.03). ICOS expression by lesion-infiltrating T-CONV and T-REG positively correlated to the extent of ICOSL expression by LCH-cells (P=0.004). Our study points out that the combined detection of interlesional IL-10 and ICOSL expression by LCH-cells is associated with the highest prevalence of activated T-CONV. Immune profiling of LCH-affected tissues obtained at the time of diagnosis may set the stage for the development of new types of therapies, which aim at local boosting of immune cells that recognize and eliminate neoplastic LCH-cells.

Introduction

The broad clinical spectrum of Langerhans cell histiocytosis (LCH) involves isolated or disseminated manifestation in various tissues such as bone, lungs, lymph nodes (LN), and skin (Egeler and D'Angio 1995). LCH lesions are typically characterized by aberrantly differentiated CD1a⁺/Langerin⁺ histiocytes (LCH-cells) (Writing Group of the Histiocyte Society 1987; Geissmann and others 2001). These LCH-cells often carry somatic mutations in intracellular signaling molecules, which collectively belong to the RAS-RAF-MEK-ERK-pathway [known as the mitogen-activated protein kinase (MAPK)-pathway] (Badalian-Very and others 2010; Satoh and others 2012; Berres and others 2014; Brown and others 2014; Chakraborty and others 2014; Nelson and others 2014, 2015). The BRAF^V600E mutation detected in ∼50% of LCH patients is expressed by many different types of cancer (Riesco-Eizaguirre and others 2009; Cantwell-Dorris and others 2011; Sigalotti and others 2011). This particular somatic mutation appears to have an impact on the tumor microenvironment as demonstrated by BRAF^V600E mutation-bearing thyroid cancer cells, which secrete functional transforming growth factor-beta (TGF-β) into the tumor microenvironment. This immunosuppressive cytokine allows the tumor cells to escape from clearance by the immune system [immune evasion] (Riesco-Eizaguirre and others 2009). In addition, different types of cancer can accomplish immune evasion by interfering (in)directly with tumor-associated T-cells or by altering dendritic cell–T-cell interactions in the tumor microenvironment (Martin-Orozco and others 2010; Conrad and Gilliet 2013; Esendagli 2013; Faget and others 2013). The local microenvironment wherein BRAF^V600E tumor cells hide may also have a negative impact on therapy-induced clinical responses and even promote resistance to therapy. The composition of the microenvironment is particularly relevant for combinatorial treatment protocols wherein systemic (chemo)therapy or targeted therapy, such as vemurafenib or debrafenib, is combined with immunotherapy. This treatment is currently tested in patients with advanced BRAF^V600E melanoma (ClinicalTrials.gov identifier: NCT01813214).

The microenvironment of LCH lesions is composed of different types of cytokines and immune cell types like macrophages, eosinophilic granulocytes, multinucleated giant cells, and T-cells (Schmitz and Favara 1998; Egeler and others 1999; Annels and others 2003; da Costa and others 2005; Senechal and others 2007). Different subtypes of lesion-infiltrating T-cells have been detected in these lesions such as CD3⁺FOXP3⁻ conventional T-cells (T-CONV) and CD3⁺FOXP3⁺ regulatory T-cells (T-REG). The T-cells appear to interact with LCH-cells in situ as visualized by immunohistochemistry (Egeler and others 1999; Senechal and others 2007). The impact of this interaction on the activation status and effector function of lesion-infiltrating T-CONV is, however, still unclear.

Lesion-infiltrating T-cells are not naïve but rather appear to function in situ as end-stage differentiated effector type cells as they proliferate, express CD45RO, and produce a variety of cytokines (Egeler and others 1999; Annels and others 2003; Senechal and others 2007). Noteworthy, T-CONV acquire these functions after “cross-talk” with professional antigen-presenting cells (APC) and proper T-cell receptor ligation and co-stimulation. T-CONV only express the inducible co-stimulatory molecule (ICOS) after this initial T-cell activation step has occurred (Hutloff and others 1999; Greenwald and others 2005; Ito and others 2008). The interaction between ICOS⁺ T-cells and ICOS ligand (ICOSL)-expressing professional APC is necessary for both differentiation into cytokine-producing cells (Mesturini and others 2013), immune regulation (Akbari and others 2002; Frey and others 2010), and the induction of antitumor-directed immune responses (Strauss and others 2008; Frey and others 2010; Martin-Orozco and others 2010). This interaction can, however, also lead to immune evasion by tumor cells (Martin-Orozco and others 2010; Conrad and Gilliet 2013; Esendagli 2013; Faget and others 2013). The dual role of ICOS-ICOSL in tumor immunity suggests that the net outcome of ICOS-costimulation is determined by the microenvironment in which the T-cell becomes activated.

A subtype of ICOS-expressing human T-REG has earlier been identified (Ito and others 2008). While ICOS⁺ T-REG mainly act via the production of interleukin-10 (IL-10), ICOS⁻ T-REG exert their regulatory function solely through the release of TGF-β (Ito and others 2008). IL-10 and TGF-β both facilitate T-REG generation and expansion in vitro and are capable of direct suppression of T-CONV function (Rao and others 2005). The local production of these immunosuppressive cytokines by LCH lesion-infiltrating T-cells and macrophages has already been reported. Noteworthy, IL-10 or TGF-β have only been detected in a proportion of LCH lesions studied thus far (de Graaf and others 1996; Egeler and others 1999; Geissmann and others 2001; Senechal and others 2007; Ohnishi and others 2010). Likewise, the in situ presence of T-CONV and T-REG expressing CD25 and FOXP3 has been reported (Egeler and others 1999; Annels and others 2003; Senechal and others 2007).

We investigated the impact of the local cytokine milieu in relation to ICOSL expression by LCH-cells and the composition of the concurrent T-cell infiltrate and whether any of the immune signatures is associated with a better clinical outcome.

Materials and Methods

Patient materials

Formalin-fixed-paraffin-embedded (FFPE) biopsies from 25 therapy-naïve LCH patients were included in this study. Confirmation of LCH diagnosis was established by clinical symptoms and histopathological evaluation of biopsied tissues. All LCH-affected tissue blocks contained >5% CD1a⁺ and Langerin⁺ histiocytes as visualized by routine immunohistochemical staining. Based on the available clinical reports, the patients were stratified by affected site [bone, lymph node (LN) and skin] and by disease manifestation (i.e., mono-ostotic, single-system, poly-ostotic, and multi-system LCH). LCH reactivation was defined as the development of an additional LCH lesion manifesting at another site than the (successfully treated) primary lesion. Clinical follow-up was incomplete for 4 patients, who were excluded from the survival analysis. Patient characteristics are shown in Table 1. This study is part of study protocol P-10.163, which was approved by the Medical Ethics Committee of the Leiden University Medical Center. Research was conducted following the ethical guidelines of the national organization of scientific societies (FEDERA). All participants, or their parents in the case of patients below the age of 18 years, provided verbal or written consent, which was registered in the patient's files and/or in informed consent forms.

Table 1.

Cytokine Profiles and Clinical Characteristics of Langerhans Cell Histiocytosis Patients

UPN	Tissue	Diagnosis	Age of onset	Reactivation	Treatment	Cytokine profile ^a
1	Bone	MO	10	No	Local ST	IL-10
2	Bone	MO	16	No	None	IL-10
3	Bone	MO	3	N.A.	None	IL-10
4	Bone	MO^b	N.A.	N.A.	N.A.	IL-10
5	Bone	PO	31	Yes	Local ST	IL-10
6	LN	SS	10	No^b	RE	IL-10
7	LN	SS	1	No	None	IL-10
8	Skin	MS	2	Yes	RE+ST+CHT	IL-10
9	Skin	SS	17	No	N.A.	IL-10
10	Skin	SS	28	No^b	N.A.	IL-10
11	Bone	MO	10	No	Local ST	TGF-β
12	Bone	PO	5	No	Local ST+ST+CHT	TGF-β
13	Bone	PO	27	Yes	Local ST+RDT	TGF-β
14	Bone	PO	16	No	Local ST	TGF-β
15	Bone	PO	6	Yes	Local ST	TGF-β
16	Bone	PO	18	Yes	RE&Local ST	TGF-β
17	Skin	SS	27	No	N.A.	TGF-β
18	Skin	SS	24	No	N.A.	TGF-β
19	Skin	SS	N.A.	No^b	N.A.	TGF-β
20	Bone	MO	3	No	None	TGF-β+IL-10
21	Bone	PO	1	Yes	Local ST	TGF-β+IL-10
22	Bone	PO	2	Yes	Local ST&RE	TGF-β+IL-10
23	LN	SS	1	N.A.	N.A.	TGF-β+IL-10
24	Skin	MS	1	Yes	ST+CHT	TGF-β+IL-10
25	Skin	SS	N.A.	N.A.	N.A.	TGF-β+IL-10

The mean age of onset of patients displaying bone LCH lesions, LN involvement or skin lesions is 11.4, 4.0, and 16.5 years respectively; male to female ratio is 7:4. Follow-up data were available for 21 patients, which varied from 65 days to 19 years after primary diagnosis, with a median follow-up time of 10 years.

Lesions expressing LAP were defined as TGF-β-positive (Khalil 1999; Annes and others 2003).

Follow-up data was available for less than 1.5 years after diagnosis but no signs of reactivation were observed in this follow-up period.

UPN indicates unique patient number; diagnosis is referred to as single system (SS), multi system (MS), mono ostotic (MO), poly ostotic (PO); ST, steroids; CHT, chemotherapy; RDT, radiotherapy; RE, resection; N.A., not available; IL, interleukin-10; LAP, latency-associated peptide; LCH, Langerhans cell histiocytosis; LN, lymph nodes; TGF-β, transforming growth factor-beta.

Immunohistochemical and immunofluorescent staining of LCH biopsies

Four micrometer FFPE tissue sections deposited on coated glass slides were dried overnight at 37°C and stored at 4°C. Selected slides were preheated for 1 h at 63°C and deparaffinized in xylol. Immunohistochemical (IHC)-designated tissue sections were submerged in methanol/0.03% H₂O₂ for blocking endogenous peroxidase activity where after all sections were rehydrated in ethanol and rinsed in demi water followed by antigen retrieval in boiling citrate buffer pH 6.0 or EDTA buffer pH 8.0 for 12 min. Sections were incubated with/without 10% goat serum for 30 min and incubated overnight with antibodies specific for CD1a, CD3 (Uhlig and others 2006) (DAKO, Glostrup, Denmark), CD25 (Tantibhaedhyangkul and others 2009), CD4 (Tantibhaedhyangkul and others 2009) (Novocastra, Newcastle upon Tyne, UK), ICOS (AF169), latency-associated peptide (LAP) (AB-246-NA) (Dasgupta and others 2006) (R&D Systems, Minneapolis, MN), FOXP3 (22510) (Tzankov and others 2008), IL-10 (ab34843) (Battistella and others 2010; Zhang and others 2011) (Abcam, Cambridge, UK), IL-10 (HP9016) (Hycults Biotech, Uden, The Netherlands), or Langerin/CD207 (929F3.01) (Imgenex, San Diego, CA) all diluted in phosphate-buffered saline (PBS)/1% bovine serum albumin (BSA). Immunofluorescent (IF)-designated tissue sections were incubated for 30 min with secondary antibodies bound to Alexa fluorochromes 488, 546 or 647 (Invitrogen, Grand Island, NY), washed, and mounted with moviol (homemade). Images were recorded by a fluorescent microscope (DM5500B; Leica Microsystems, Rijswijk, The Netherlands) (40× original magnification).

For IHC-based visualization of Langerin-positive cells, alkaline phosphatase-conjugated AffiniPure F(ab′)2 fragment goat-anti-Rat IgG was applied and developed with Vector Blue substrate (Burlingame, CA). Hereafter, sections were again placed in boiling citrate buffer pH 6.0 for 10 min, incubated with ICOSL-specific antibodies (2D3; Lifespan Biosciences, Seattle, WA) diluted in PBS/1% BSA where after Envision Poly-HRP goat-anti-mouse IgG (ImmunoLogic, Duiven, The Netherlands) was applied. LAP and IL-10-stained sections were directly incubated with HRP-labeled Bright-vision anti-rat-rabbit-mouse (ImmunoLogic). Bound HRP was developed with 3,3′-diaminobenzidine substrate (Dako, Glostrup, Denmark). Sections were briefly counterstained with hematoxylin (Kinipath, Duiven, The Netherlands) and mounted with Pertex (Leica Microsystems) or with Vectorshield.

Images were captured on a spectral imaging microscope (DM4000B; Leica Microsystems). Lesional IL-10 expression was visualized with 2 different types of antibodies, which gave identical staining results (data upon request). Control tissue slides were stained in parallel with the same secondary staining reagents and color-developing substrates, but the primary antibody was omitted (negative controls) (Faaij and others 2010; Schrevel and others 2012).

Statistical analysis

Each LCH biopsy was evaluated in a blinded fashion by 2 independent researchers (J.S.K. and W.T.Q.) for the (co)expression of IL-10 and LAP and classified according to the dominantly expressed cytokine: IL-10-predominant, TGF-β predominant, or dual IL-10/TGF-β in the case of comparable LAP and IL-10 expression. Using ImageJ (Javascript 1.44), the numbers of lesion-associated CD3⁺ T-cells were scored in at least 10 representative pictures (magnification 40×) taken from each biopsy. Given that LCH biopsies may differ from a few mm to several cm in size, the total number of each T-cell subtype was scored and the average number of cells per microscopic field and the percentage of T-cells expressing certain (surface) markers were calculated. Although, quantification of cell numbers in a few FFPE-tissue slides may not be as accurate as flowcytometric analysis performed on a single cell suspension prepared from the entire tissue biopsy, we demonstrate that quantifying proportions of lesion-infiltrating T-cells by IHC can be done in a reliable and objective manner. This is substantiated by data from 2 different blinded experiments, in which CD3 and FOXP3 were stained in combination with either CD25 or ICOS (Fig. 2A, B), where mean difference in the percentage of T-REG among the total number of CD3⁺T-cells was only 7.7% (P=0.31).

A combinatorial Ruiter scoring method was applied to semi-quantitatively assess the intensity of ICOSL expression and the percentage of ICOSL⁺ Langerin⁺ co-expressing LCH-cells (Ruiter and others 1998). While the staining intensity was classified as absent (0), weak (1), clear (2), or strong (3), the corresponding percentage of ICOSL⁺ LCH-cells was scored as (1) for 1%–5%, (2) for 6%–25%, (3) for 26%–50%, (4) for 51%–75%, and (5) for 76%–100%. A combination score of the intensity and percentage of ICOSL⁺ Langerin⁺ cells ranging between 2 and 6 was designated as moderate expression; a score of ≥7 was considered as strong expression (Ruiter and others 1998).

Statistical analysis was performed using Prism Graphpad 5.04 (Graphpad Software; La Jolla, CA). The non-parametric Mann–Whitney test was applied to compare the cellular distribution among the different cytokine profiles and among lesions with a different extent of ICOSL expression. Correlations between LCH-reactivation and the cellular subsets or cytokine profiles were analyzed using Kaplan–Meier analysis and logrank test. A P value of <0.05 was considered as statistically significant.

Results

The prototypic immunosuppressive cytokines IL-10 and TGF-β are expressed in LCH lesions

A panel of 25 LCH-affected biopsies was stained with antibodies specific for either IL-10 or LAP. The latter strategy was chosen since LAP protein noncovalently associates with inactive TGF-β1-complexes at the cellular membrane preceding the release of active TGF-β1 from the cell (Khalil 1999; Annes and others 2003). Hence, LAP-specific immunostaining can be seen as a surrogate marker for TGF-β1 production and release.

Various cell types present in LCH lesions stained positive for either cytokine, varying from small lymphoblastic cells embedded in IL-10 enriched areas to large histiocytic cells including multinucleated giant cells, which almost exclusively stained positive for LAP. The microenvironment of LCH lesions contained either predominantly IL-10 (10/25) (Fig. 1, upper and lower left photographs) or predominant TGF-β (9/25) (Fig. 1, upper and lower right photographs). Some lesions were classified as dual IL-10/TGF-β (6/25) as comparable staining for both cytokines was detected (Fig. 1, upper and lower photographs in the centre). None of the 3 cytokine profiles correlated with any particular affected site (P=0.38), type of LCH manifestation (P=0.74) or risk of LCH reactivation (P=0.35).

FIG. 1.

Langerhans cell histiocytosis (LCH) lesions display 3 distinct cytokine profiles. Diagnostic LCH biopsies were examined by standard immunohistochemistry for the expression of interleukin-10 (IL-10) (clone HP9016) (upper row) or active transforming growth factor-beta (TGF-β) using a latency-associated peptide (LAP)-specific antibody (bottom row). The brown color indicates the presence of specific cytokine. Representative examples of distinct cytokine profiles are indicated on top of each vertical row of pictures taken from the same (skin) lesion in the same area (magnification 20×). Insets depicted at the upper left corner of each photograph are a larger magnification of the indicated areas. The microenvironment of LCH lesions contained either predominantly IL-10 characterized by abundant expression of IL-10 with sporadic detection of LAP or predominant TGF-β classified by LAP expression while IL-10 was absent or sporadically detected. Some lesions were classified as dual IL-10/TGF-β as comparable staining for both cytokines was detected.

LCH lesions contain comparable numbers of CD3⁺ T-cells and LCH-cells regardless of the corresponding cytokine profile

The presence of IL-10 and/or TGF-β within the LCH microenvironment raises the question whether these immunosuppressive cytokines affect the number and phenotype of lesion-infiltrating T-cells and LCH-cells. The average number of Langerin⁺CD1a⁺ LCH-cells and CD3⁺ T-cells per microscopic field varied among the biopsies (LCH-cells: median 111; range 31–413; T-cells: median 29; range 8–154). They were, however, not associated with any of the 3 different suppressive cytokine profiles (P=0.27) or with any particular affected site or manifestation form (P=0.62). These data indicate that LCH-cells and T-cells were comparably distributed among the different cytokine profiles.

Activation of infiltrating conventional T-cells is restricted to IL-10 cytokine profiles

The effect of either IL-10 or TGF-β expression on the composition of the corresponding T-cell infiltrate was assessed. As published by others (Senechal and others 2007), discrimination between CD3⁺FOXP3⁻ T-CONV and CD3⁺FOXP3⁺ T-REG was based on a dual IF staining technique, which visualizes membranous CD3 and nuclear FOXP3 expression on the same T-cell (Fig. 2A, B, magnified cells depicted in the lower right of Fig. 2A). Regardless of the type of immune suppressive cytokine present, all lesions contained comparable ratios of T-REG/T-CONV (1:5, Fig. 2D) and comparable average T-REG and T-CONV numbers per microscopic field (P=0.61, data not shown).

FIG. 2.

Detection of different T-cell phenotypes and inducible costimulatory molecule ligand (ICOSL)-expressing LCH-cells within LCH-lesions and correlation to the local cytokine environment. After triple IF staining of LCH lesions using antibodies specific for the pan T-cell marker CD3 in combination with the nuclear transcription factor FOXP3 and 2 different cell surface markers each LCH biopsy was scored and the percentage of T-cells expressing certain (surface) markers was subsequently calculated as represented in the following graphs. LCH lesions were additionally stained with antibodies against the LCH-cell-specific marker Langerin (upper left panel) in combination with ICOSL (upper right panel) to detect LCH-cells expressing ICOSL (A) colocalization of intracellular FOXP3 and CD25 expressed on the surface of CD3⁺ T-cells indicate T-REG results in a purple cell with a green nucleus (boxes, right magnification) whereas purple cells without a green nucleus indicate CD25⁺ T-CONV (arrows, left magnification); solitary red cells represent CD25⁻T-CONV (triangles, centered magnification). (B) Combination staining of CD3 with FOXP3 and ICOS reveals the presence of CD3⁺ ICOS⁺ FOXP3⁺ T-REG cells, which typically express a blue nucleus and a red and green cell membrane (boxes, right magnification), red cells with a blue nucleus indicate ICOS⁻ CD3⁺ FOXP3⁺ T-REG (triangles, left magnification), and ICOS⁻ CD3⁺ FOXP3⁻ T-CONV are visualized by red cells without a blue nucleus (arrows, centered magnification). (C) A representative example of a LCH lesion double immunohistochemically stained with antibodies against the LCH-cell-specific marker Langerin (upper left panel) in combination with ICOSL (upper right panel) showing LCH-cells expressing strong levels of ICOSL. The untouched image (RGB-color) shows cells expressing Langerin (blue) in conjunction with ICOSL (blue/brown). To determine co-localization of ICOSL on LCH-cells the different spectra (upper panels in gray) were separated using spectral microscopy. To this end, Langerin was artificially made green and ICOSL was visualized in red as shown in the spectral overlay (unmixed composite). (D) The ratio of T-CONV per T-REG was comparable between the cytokine profiles. (E) The percentage of CD25-positive cells among CD3⁺FOXP3⁻T-CONV, which is significantly lower in TGF-β-predominant lesions compared with the other 2 profiles. (F) The percentage of ICOS-positive cells among CD3⁺FOXP3⁻T-CONV, which is significant lower in TGF-β-predominant lesions compared with IL-10-predominant lesions. (G) The percentage of ICOS-positive cells among CD3⁺FOXP3⁺T-REG, which is significantly lower in TGF-β-predominant lesions compared with IL-10-predominant lesions. (H) Combination scores for intensity and percentage of Langerin⁺ LCH-cells expressing ICOSL. Scoring of each LCH biopsy was performed in representative areas by a Ruiter score as explained in more detail in the Materials and Methods section. Combination scores ranging from 2 to 6 were considered as moderate expression and scores ≥7 as strong (as indicated by the gray area). Moderate expression was more frequently observed in TGF-β-predominant lesions compared with IL-10-predominant lesions and in lesions expressing both cytokines. (I, J) Graph shows the percentage of T-CONV and T-REG that expressed ICOS (right) in relation to the extent of ICOSL expression by LCH-cells (x-axis). (D–G) solid symbols represent lesions with strong ICOSL expression and open symbols lesions with moderate ICOSL expression. The horizontal lines represent the median; *P<0.05; **P<0.01; ***P<0.001.

The activation status of CD3⁺FOXP3⁻ T-CONV cells was substantiated by the co-staining of the IL-2 receptor, CD25 (blue staining in Fig. 2A). A significant higher proportion of T-CONV expressing CD25 was prevalent in IL-10⁺⁺ lesions (38%–100%) than in TGF-β⁺⁺ lesions (6%–54%, P=0.008, Fig. 2E and Table 2). In contrast, all FOXP3⁺ T-REG expressed CD25 regardless of the corresponding cytokine profile (Table 2). Collectively, these data demonstrate that T-REG and T-CONV were present at comparable frequencies but IL-10⁺ lesions contained higher percentages of CD25⁺ T-CONV.

Table 2.

Phenotypic Characteristics of the T-Cell Infiltrate in Relation to the Three Distinct Cytokine Profiles

Profiles
Unit	Marker	Population	IL-10	Dual IL-10/TGF-β	TGF-β	P	ICOSL^mod	ICOSL^strong	P
%	CD25	T-CONV	98 (38–100)	100 (46–100)	20 (6–54)	^**	39 (6–100)	100 (42–100)	^**
%	CD25	T-REG	100 (100)	100 (100)	100 (100)	—	100 (100)	100 (100)	—
sum	ICOSL	LCH-cells	7 (2–8)	8 (7–8)	5 (2–6)	^*	—	—	—
n	ICOS	T-CONV	22 (3–147)	10 (1–27)	4 (1–21)	^**	6 (1–22)	11 (1–147)	^*
n	ICOS	T-REG	6 (2–24)	4 (1–19)	3 (1–10)	—	3 (1–7)	7 (3–24)	^*
%	ICOS	T-CONV	86 (47–100)	81 (3–100)	29 (7–51)	^***	37 (7–86)	88 (3–100)	^*
%	ICOS	T-REG	100 (45–100)	100 (67–100)	81 (45–100)	^*	85 (45–100)	100 (67–100)	—

Absolute numbers (n) or percentages (%) of cells are defined as the median (range) of CD3⁺ FOXP3⁻ T-CONV or CD3⁺ FOXP3⁺ T-REG in IL-10-predominant lesions, TGF-β-predominant lesions or lesions expressing both cytokines and in lesions with moderate or strong expression (sum) of ICOSL by LCH-cells; P values indicate the difference between IL-10-predominant lesions and TGF-β-predominant lesions or the difference between lesions with strong or moderate ICOSL expression.

P<0.05, ^** P<0.01, or ^*** P<0.001.

The percentage of lesion-infiltrating ICOS⁺T-CONV cells and ICOS⁺T-REG varies according to the corresponding cytokine milieu

Following activation through T-cell receptor ligation, ICOS expression is upregulated by both T-CONV and T-REG (Hutloff and others 1999; Greenwald and others 2005; Ito and others 2008). Importantly, ICOS distinguishes 2 functionally different subtypes of human T-REG (Ito and others 2008). Similar to the expression of CD25, the expression of ICOS on T-CONV and T-REG was assessed for each of the 3 different cytokine profiles (Fig. 2B). Significantly higher numbers of ICOS⁺ T-CONV were observed in IL-10-predominant lesions (range 3–147) compared with TGF-β-predominant lesions (range 1–21) (P=0.006, Table 2). Likewise, the percentage of T-CONV expressing ICOS was higher in IL-10⁺⁺ lesions than in TGF-β⁺⁺lesions (P=0.0008, Fig. 2F and Table 2). While the absolute numbers of ICOS⁺ T-REG were comparable among the 3 different cytokine profiles (Table 2), the percentage of T-REG expressing ICOS was higher in IL-10-lesions than in TGF-β-rich lesions. The latter LCH lesions contained more ICOS⁻ T-REG (P=0.05). In summary, CD25 and ICOS expressing T-CONV and ICOS⁺ T-REG are more prevalent in LCH lesions where IL-10 is secreted in the microenvironment.

ICOS expression by lesion-infiltrating T-cells correlates with the extent of ICOSL expression by LCH-cells

Re-exposure to ICOSL in situ may increase ICOS expression, which leads to enhanced cytokine production by T-CONV at their site of action (Beier and others 2000; Ito and others 2008; Martin-Orozco and others 2010). To determine whether ICOSL is expressed by LCH-cells a 2-step IHC staining with ICOSL and Langerin was applied (Fig. 2C). All LCH biopsies contained ICOSL⁺ Langerin⁺ LCH-cells, but the percentage of LCH-cells expressing ICOSL ranged from 1% to 100% with staining intensities varying from weak to strong expression levels. As explained in detail in the Materials and Methods section, the staining intensity of ICOSL on Langerin⁺ LCH-cells was combined with the percentage of ICOSL-expressing LCH-cells. While IL-10⁺⁺ lesions were generally classified as strongly positive for ICOSL-expression (median Ruiter score of 7), TGF-β⁺⁺ lesions contained moderately ICOSL-expressing LCH-cells (median Ruiter score of 5) (P=0.03, Fig. 2H and Table 2).

ICOSL expression had no impact on the absolute numbers of ICOS⁺ T-CONV and CD25⁺ T-CONV (P=0.22 and P=0.18, Table 2). In contrast, ICOSL^hi LCH lesions contained higher percentages of T-CONV expressing ICOS (P=0.01, Fig. 2I and Table 2) and higher percentages of T-CONV expressing CD25 (P=0.006, Table 2). While the number of CD25⁺ T-REG were comparable (P=0.10, data not shown), these ICOSL^hi lesions contained also significant higher numbers of ICOS⁺ T-REG (P=0.02, Table 2) and higher percentages of T-REG expressing ICOS (P=0.05, Fig. 2J and Table 2).

Taken together, we demonstrate that IL-10-predominant lesions are characterized by many ICOSL^hi LCH-cells and high frequencies of ICOS expressing T-CONV and T-REG. In contrast, TGF-β-predominant lesions typically contain ICOSL^mod LCH-cells and significantly fewer ICOS expressing T-CONV and T-REG.

Discussion

LCH is increasingly considered to be a neoplastic myeloid disorder with immunological features as reflected by the in situ presence of a large number of immune cells, which accompany the genetically aberrant myeloid LCH-cells. Phenotypical aspects of the immune signature of LCH lesions are not incorporated into therapy decisions. This is the first study demonstrating that, independent of tissue site or type of LCH manifestation, LCH lesions can be classified according to distinct expression profiles of 2 potent immune suppressive cytokines. The presence of IL-10 and TGF-β alone or in combination were found to correlate with the phenotypic makeup of LCH-cells and to the composition of the corresponding T-cell infiltrate.

The identification of these 3 distinct cytokine profiles fits well with the results from previous studies in which the authors failed to detect either TGF-β or IL-10 in 100% of the LCH lesions analyzed (de Graaf and others 1996; Egeler and others 1999; Geissmann and others 2001; Senechal and others 2007; Ohnishi and others 2010). Our data are also in accordance with previous studies demonstrating that IL-10 and TGF-β are mainly expressed by macrophages and LCH-cells while sporadically expressed by eosinophils and T-cells (de Graaf and others 1996; Egeler and others 1999; Senechal and others 2007; Ohnishi and others 2010). Indeed, co-localization of either cytokine with FOXP3 was rarely observed (data not shown).

Although BRAF^V600E mutation bearing cancer cells secrete functional TGF-β (Riesco-Eizaguirre and others 2009), the BRAF-mutation status of LCH-cells did not correlate with any of the 3 immunosuppressive cytokine profiles (P=0.99, data not shown). Apparently, there are additional factors that drive intralesional cytokine release by LCH-cells. This study shows an association between the extent of ICOSL expression by LCH-cells, locally produced cytokines, and the activation status of co-present CD25⁺ICOS⁺ T-CONV; the latter cells were most prevalent in IL-10⁺ lesions, which contained ICOSL^hi LCH-cells. It is, however, unknown whether intralesional IL-10 induces ICOSL on LCH-cells. Of note, tumor necrosis factor (TNF) has been shown to enhance ICOSL expression on conventional APC (Witsch and others 2002; Martin-Orozco and others 2010). However, no association between TNF and ICOSL expression on LCH-cells was observed in this sample set (P=0.85, data not shown).

Our data show that intralesional T-cells indeed function in situ as end-stage differentiated effector type T-cells producing a variety of T-cell-associated cytokines as earlier reported (Hutloff and others 1999; Greenwald and others 2005). Lesion-infiltrating T-cells are likely activated through T-cell-LCH-cell “cross-talk” as they have been found in close proximity (Egeler and others 2000; da Costa and others 2005) and LCH-cells express several key costimulatory molecules necessary for optimal T-cell activation (Egeler and others 2000; da Costa and others 2005). Thus, despite the presence of immunosuppressive IL-10, activated T-CONV do end up in these LCH lesions.

The various types of T-REG identified in this study, ie, ICOS⁺FOXP3⁺ and ICOS⁻FOXP3⁺ T-REG, all appear to thrive in either immunosuppressive milieu. We consider FOXP3⁺ T-cells as being bona fide T-REG since the vast majority (95%–100%) of these cells also co-express CD4 (data not shown) and CD25, which is in line with earlier reported T-REG phenotypes (Senechal and others 2007; d'Hennezel and others 2011; Sun and others 2011). Here, we observed comparable percentages of CD25⁺ T-REG among the total T-cell numbers [median 15 (range 1%–60%); mean 18%±12%] as reported previously (mean 18%±7%) (Senechal and others 2007). ICOS⁺ T-REG were more prevalent in IL-10⁺⁺ lesions where ICOSL^hi LCH-cells were detected. Although ICOSL⁺ melanoma cells promote the activation and survival of particularly ICOS⁺ T-REG (Martin-Orozco and others 2010), it is unknown whether ICOSL-expressing LCH-cells exhibit similar effects on ICOS⁺ T-REG.

While the average T-REG:T-CONV ratio was comparable between the different cytokine profiles, relatively more T-REG were observed in lesions containing a high percentage of ICOS⁺ T-CONV (P=0.0031, data not shown). This indicates that, as in most circumstances of immune homeostasis, an increased influx of activated T-CONV is counteracted by T-REG which tightly regulate local expansion of T-CONV to prevent excessive collateral damage (Jonuleit and Schmitt 2003). While immunohistochemistry is an excellent technique for visualizing nuclear FOXP3 expression, we can, however, not exclude the possibility that LCH lesions also contain recently activated T-cells, which transiently express FOXP3 but do not have a suppressive function (Tran and others 2007; Miyara and others 2009). Note that in vitro suppression assays are the test of choice to unequivocally demonstrate the suppressive function of FOXP3-expressing T-cells. Prospective collection of fresh biopsies to acquire sufficient numbers T-REG for functional analysis is, however, practically impossible given the low incidence of this rare disease.

None of the reported cytokine profiles in this study or any particular composition of the corresponding T-cell infiltrate was found to correlate with a favorable disease outcome. In fact, similar clinical outcomes were registered for patients displaying completely opposite lesional cytokine profiles. These observations may be explained by the fact that in our country it is common practice to deposit steroids in the biopsied tissue (Table 1). Given that steroids suppress the immune system in several ways, interactions between T-CONV and other APC present in the lesion are most likely disturbed by this therapeutic intervention. Nonetheless, our study shows that classification of new onset LCH lesions according to different immune signatures is feasible. Extended immune profiling of biopsied tissues may set the stage for the development new types of therapies, which aim at local activation of immune cells.

Footnotes

Acknowledgments

The authors thank Dr. M.W. Schilham from the Department of Pediatrics, Leiden University Medical Center and Dr. S. van Vliet (Department of Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The Netherlands) for critical comments on the article. Prof. Dr. A. von Deimling (Department of Neuropathology, Institute of Pathology, University Heidelberg and Clinical Cooperation Unit Neuropathology German Cancer Research Center, Heidelberg, Germany) kindly analyzed LCH samples for their BRAFV600E-mutation status (Sahm and others 2012). We thank Prof. Dr. B.J. Rollins, Dr. G. Badalian-Very, Dr. D.S. Nelson, and the Center for Cancer Genome Discovery at the Dana-Farber Cancer Institute, Boston, MA, for BRAF sequencing as published (Badalian-Very and others ). This work was supported by the “1000 kaarsjes voor Juultje” Foundation.

Author Disclosure Statement

No competing financial interests exist.

References

Akbari

, Freeman

, Meyer

, Greenfield

, Chang

, Sharpe

, Berry

, DeKruyff

, Umetsu

. 2002. Antigen-specific regulatory T cells develop via the ICOS-ICOS-ligand pathway and inhibit allergen-induced airway hyperreactivity. Nat Med, 8(9):1024–1032.

Annels

, da Costa

, Prins

, Willemze

, Hogendoorn

, Egeler

. 2003. Aberrant chemokine receptor expression and chemokine production by Langerhans cells underlies the pathogenesis of Langerhans cell histiocytosis. J Exp Med, 197(10):1385–1390.

Annes

, Munger

, Rifkin

. 2003. Making sense of latent TGFbeta activation. J Cell Sci, 116(Pt 2):217–224.

Badalian-Very

, Vergilio

, Degar

, Macconaill

, Brandner

, Calicchio

, Kuo

, Ligon

, Stevenson

, Kehoe

, Garraway

, Hahn

, Meyerson

, Fleming

, Rollins

. 2010. Recurrent BRAF mutations in Langerhans cell histiocytosis. Blood, 116(11):1919–1923.

Battistella

, Fraitag

, Teillac

, Brousse

, de

, Bodemer

. 2010. Neonatal and early infantile cutaneous langerhans cell histiocytosis: comparison of self-regressive and non-self-regressive forms. Arch Dermatol, 146(2):149–156.

Beier

, Hutloff

, Dittrich

, Heuck

, Rauch

, Buchner

, Ludewig

, Ochs

, Mages

, Kroczek

. 2000. Induction, binding specificity and function of human ICOS. Eur J Immunol, 30(12):3707–3717.

Berres

, Lim

, Peters

, Price

, Takizawa

, Salmon

, Idoyaga

, Ruzo

, Lupo

, Hicks

, Shih

, Simko

, Abhyankar

, Chakraborty

, Leboeuf

, Beltrao

, Lira

, Heym

, Bigley

, Collin

, Manz

, McClain

, Merad

, Allen

. 2014. BRAF-V600E expression in precursor versus differentiated dendritic cells defines clinically distinct LCH risk groups. J Exp Med, 211(4):669–683.

Brown

, Furtado

, Betz

, Kiel

, Weigelin

, Lim

, Elenitoba-Johnson

. 2014. High prevalence of somatic MAP2K1 mutations in BRAF V600E negative Langerhans cell histiocytosis. Blood, 124(10):1655–1658.

Cantwell-Dorris

, O'Leary

, Sheils

. 2011. BRAFV600E: implications for carcinogenesis and molecular therapy. Mol Cancer Ther, 10(3):385–394.

10.

Chakraborty

, Hampton

, Shen

, Simko

, Shih

, Abhyankar

, Lim

, Covington

, Trevino

, Dewal

, Muzny

, Doddapaneni

, Hu

, Wang

, Lupo

, Hicks

, Bonilla

, Dwyer

, Berres

, Poulikakos

, Merad

, McClain

, Wheeler

, Allen

, Parsons

. 2014. Mutually exclusive recurrent somatic mutations in MAP2K1 and BRAF support a central role for ERK activation in LCH pathogenesis. Blood, 124(19):3007–3015.

11.

Conrad

, Gilliet

. 2013. Plasmacytoid dendritic cells and regulatory T cells in the tumor microenvironment: a dangerous liaison. Oncoimmunology, 2(5):e23887.

12.

d'Hennezel

, Yurchenko

, Sgouroudis

, Hay

, Piccirillo

. 2011. Single-cell analysis of the human T regulatory population uncovers functional heterogeneity and instability within FOXP3+ cells. J Immunol, 186(12):6788–6797.

13.

da Costa

, Annels

, Faaij

, Forsyth

, Hogendoorn

, Egeler

. 2005. Presence of osteoclast-like multinucleated giant cells in the bone and nonostotic lesions of Langerhans cell histiocytosis. J Exp Med, 201(5):687–693.

14.

Dasgupta

, Bhattacharya-Chatterjee

, O'Malley

Jr ., Chatterjee

. 2006. Tumor metastasis in an orthotopic murine model of head and neck cancer: possible role of TGF-beta 1 secreted by the tumor cells. J Cell Biochem, 97(5):1036–1051.

15.

de Graaf

, Tamminga

, Dam-Meiring

, Kamps

, Timens

. 1996. The presence of cytokines in Langerhans' cell histiocytosis. J Pathol, 180(4):400–406.

16.

Egeler

, D'Angio

. 1995. Langerhans cell histiocytosis. J Pediatr, 127(1):1–11.

17.

Egeler

, Favara

, Laman

, Claassen

. 2000. Abundant expression of CD40 and CD40-ligand (CD154) in paediatric Langerhans cell histiocytosis lesions. Eur J Cancer, 36(16):2105–2110.

18.

Egeler

, Favara

, van Meurs

, Laman

, Claassen

. 1999. Differential in situ cytokine profiles of Langerhans-like cells and T cells in Langerhans cell histiocytosis: abundant expression of cytokines relevant to disease and treatment. Blood, 94(12):4195–4201.

19.

Esendagli

. 2013. A co-stimulatory trap set by myeloid leukemia cells. Oncoimmunology, 2(6):e24524.

20.

Faaij

, Willemze

, Revesz

, Balzarolo

, Tensen

, Hoogeboom

, Vermeer

, van Wering

, Zwaan

, Kaspers

, Story

, van Halteren

, Vossen

, Egeler

, van Tol

, Annels

. 2010. Chemokine/chemokine receptor interactions in extramedullary leukaemia of the skin in childhood AML: differential roles for CCR2, CCR5, CXCR4 and CXCR7. Pediatr Blood Cancer, 55(2):344–348.

21.

Faget

, Sisirak

, Blay

, Caux

, Bendriss-Vermare

, Menetrier-Caux

. 2013. ICOS is associated with poor prognosis in breast cancer as it promotes the amplification of immunosuppressive CD4 T cells by plasmacytoid dendritic cells. Oncoimmunology, 2(3):e23185.

22.

Frey

, Meisel

, Hutloff

, Bonhagen

, Bruns

, Kroczek

, Morawietz

, Kamradt

. 2010. Inducible costimulator (ICOS) blockade inhibits accumulation of polyfunctional T helper 1/T helper 17 cells and mitigates autoimmune arthritis. Ann Rheum Dis, 69(8):1495–1501.

23.

Geissmann

, Lepelletier

, Fraitag

, Valladeau

, Bodemer

, Debre

, Leborgne

, Saeland

, Brousse

. 2001. Differentiation of Langerhans cells in Langerhans cell histiocytosis. Blood, 97(5):1241–1248.

24.

Greenwald

, Freeman

, Sharpe

. 2005. The B7 family revisited. Annu Rev Immunol, 23:515–548.

25.

Hutloff

, Dittrich

, Beier

, Eljaschewitsch

, Kraft

, Anagnostopoulos

, Kroczek

. 1999. ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature, 397(6716):263–266.

26.

Ito

, Hanabuchi

, Wang

, Park

, Arima

, Bover

, Qin

, Gilliet

, Liu

. 2008. Two functional subsets of FOXP3+ regulatory T cells in human thymus and periphery. Immunity, 28(6):870–880.

27.

Jonuleit

, Schmitt

. 2003. The regulatory T cell family: distinct subsets and their interrelations. J Immunol, 171(12):6323–6327.

28.

Khalil

. 1999. TGF-beta: from latent to active. Microbes Infect, 1(15):1255–1263.

29.

Martin-Orozco

, Li

, Wang

, Liu

, Hwu

, Liu

, Dong

, Radvanyi

. 2010. Melanoma cells express ICOS ligand to promote the activation and expansion of T-regulatory cells. Cancer Res, 70(23):9581–9590.

30.

Mesturini

, Gigliotti

, Orilieri

, Cappellano

, Soluri

, Boggio

, Woldetsadik

, Dianzani

, Sblattero

, Chiocchetti

, Yagi

, Rojo

, Dianzani

. 2013. Differential induction of IL-17, IL-10, and IL-9 in human T helper cells by B7h and B7.1. Cytokine, 64(1):322–330.

31.

Miyara

, Yoshioka

, Kitoh

, Shima

, Wing

, Niwa

, Parizot

, Taflin

, Heike

, Valeyre

, Mathian

, Nakahata

, Yamaguchi

, Nomura

, Ono

, Amoura

, Gorochov

, Sakaguchi

. 2009. Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor. Immunity, 30(6):899–911.

32.

Nelson

, Quispel

, Badalian-Very

, van Ham

, van den Bos

, Bovee

, Tian

, Van

, Ducar

, Macconaill

, Egeler

, Rollins

. 2014. Somatic activating ARAF mutations in Langerhans cell histiocytosis. Blood, 123(20):3152–3155.

33.

Nelson

, van Halteren

AGS

, Quispel

, van den Bos

, Bovee

, Bhumi

, Badalian-Very

, Hummelen

, Ducar

, Macconaill

, Egeler

, Rollins

. 2015. MAP2K1 and MAP3K1 mutations in Langerhans cell histiocytosis. Genes Chromosomes Cancer, 54(6):361–368.

34.

Ohnishi

, Komohara

, Sakashita

, Iyama

, Murayama

, Takeya

. 2010. Macrophages in Langerhans cell histiocytosis are differentiated toward M2 phenotype: their possible involvement in pathological processes. Pathol Int, 60(1):27–34.

35.

Rao

, Petrone

, Ponath

. 2005. Differentiation and expansion of T cells with regulatory function from human peripheral lymphocytes by stimulation in the presence of TGF-{beta}. J Immunol, 174(3):1446–1455.

36.

Riesco-Eizaguirre

, Rodriguez

, De la Vieja

, Costamagna

, Carrasco

, Nistal

, Santisteban

. 2009. The BRAFV600E oncogene induces transforming growth factor beta secretion leading to sodium iodide symporter repression and increased malignancy in thyroid cancer. Cancer Res, 69(21):8317–8325.

37.

Ruiter

, Ferrier

, van Muijen

, Henzen-Logmans

, Kennedy

, Kramer

, Nielsen

, Schmitt

. 1998. Quality control of immunohistochemical evaluation of tumour-associated plasminogen activators and related components. European BIOMED-1 Concerted Action on Clinical Relevance of Proteases in Tumour Invasion and Metastasis. Eur J Cancer, 34(9):1334–1340.

38.

Sahm

, Capper

, Preusser

, Meyer

, Stenzinger

, Lasitschka

, Berghoff

, Habel

, Schneider

, Kulozik

, Anagnostopoulos

, Mullauer

, Mechtersheimer

, Von

. 2012. BRAFV600E mutant protein is expressed in cells of variable maturation in Langerhans cell histiocytosis. Blood, 120(12):e28–e34.

39.

Satoh

, Smith

, Sarde

, Lu

, Mian

, Trouillet

, Mufti

, Emile

, Fraternali

, Donadieu

, Geissmann

. 2012. B-RAF mutant alleles associated with Langerhans cell histiocytosis, a granulomatous pediatric disease. PLoS One, 7(4):e33891.

40.

Schmitz

, Favara

. 1998. Nosology and pathology of Langerhans cell histiocytosis. Hematol Oncol Clin North Am, 12(2):221–246.

41.

Schrevel

, Karim

, ter Haar

, van der Burg

, Trimbos

, Fleuren

, Gorter

, Jordanova

. 2012. CXCR7 expression is associated with disease-free and disease-specific survival in cervical cancer patients. Br J Cancer, 106(9):1520–1525.

42.

Senechal

, Elain

, Jeziorski

, Grondin

, Patey-Mariaud de

, Jaubert

, Beldjord

, Lellouch

, Glorion

, Zerah

, Mary

, Barkaoui

, Emile

, Boccon-Gibod

, Josset

, Debre

, Fischer

, Donadieu

, Geissmann

. 2007. Expansion of regulatory T cells in patients with Langerhans cell histiocytosis. PLoS Med, 4(8):e253.

43.

Sigalotti

, Fratta

, Parisi

, Coral

, Maio

. 2011. Stability of BRAF V600E mutation in metastatic melanoma: new insights for therapeutic success?. Br J Cancer, 105(2):327–328.

44.

Strauss

, Bergmann

, Szczepanski

, Lang

, Kirkwood

, Whiteside

. 2008. Expression of ICOS on human melanoma-infiltrating CD4+CD25highFoxp3+ T regulatory cells: implications and impact on tumor-mediated immune suppression. J Immunol, 180(5):2967–2980.

45.

Sun

, Yi

, Wu

. 2011. Foxp3 is a key regulator of human natural CD4+CD25+regulatory T cell-mediated suppression in alloimmune response. Transpl Immunol, 26(2–3):71–80.

46.

Tantibhaedhyangkul

, Tatevian

, Gilger

, Major

, Davis

. 2009. Increased esophageal regulatory T cells and eosinophil characteristics in children with eosinophilic esophagitis and gastroesophageal reflux disease. Ann Clin Lab Sci, 39(2):99–107.

47.

Tran

, Ramsey

, Shevach

. 2007. Induction of FOXP3 expression in naive human CD4+FOXP3 T cells by T-cell receptor stimulation is transforming growth factor-beta dependent but does not confer a regulatory phenotype. Blood, 110(8):2983–2990.

48.

Tzankov

, Meier

, Hirschmann

, Went

, Pileri

, Dirnhofer

. 2008. Correlation of high numbers of intratumoral FOXP3+ regulatory T cells with improved survival in germinal center-like diffuse large B-cell lymphoma, follicular lymphoma and classical Hodgkin's lymphoma. Haematologica, 93(2):193–200.

49.

Uhlig

, Coombes

, Mottet

, Izcue

, Thompson

, Fanger

, Tannapfel

, Fontenot

, Ramsdell

, Powrie

. 2006. Characterization of Foxp3+CD4+CD25+ and IL-10-secreting CD4+CD25+ T cells during cure of colitis. J Immunol, 177(9):5852–5860.

50.

Witsch

, Peiser

, Hutloff

, Buchner

, Dorner

, Jonuleit

, Mages

, Kroczek

. 2002. ICOS and CD28 reversely regulate IL-10 on re-activation of human effector T cells with mature dendritic cells. Eur J Immunol, 32(9):2680–2686.

51.

Writing Group of the Histiocyte Society. 1987. Histiocytosis syndromes in children. Lancet, 1(8526):208–209.

52.

Zhang

, Wang

, Fan

, Liu

, Peng

. 2011. Impact of interleukin-10 gene polymorphisms on tacrolimus dosing requirements in Chinese liver transplant patients during the early posttransplantation period. Eur J Clin Pharmacol, 67(8):803–813.

Activated Conventional T-Cells Are Present in Langerhans Cell Histiocytosis Lesions Despite the Presence of Immune Suppressive Cytokines

Abstract

Introduction

Materials and Methods

Patient materials

Immunohistochemical and immunofluorescent staining of LCH biopsies

Statistical analysis

Results

The prototypic immunosuppressive cytokines IL-10 and TGF-β are expressed in LCH lesions

LCH lesions contain comparable numbers of CD3+ T-cells and LCH-cells regardless of the corresponding cytokine profile

Activation of infiltrating conventional T-cells is restricted to IL-10 cytokine profiles

The percentage of lesion-infiltrating ICOS+T-CONV cells and ICOS+T-REG varies according to the corresponding cytokine milieu

ICOS expression by lesion-infiltrating T-cells correlates with the extent of ICOSL expression by LCH-cells

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

LCH lesions contain comparable numbers of CD3⁺ T-cells and LCH-cells regardless of the corresponding cytokine profile

The percentage of lesion-infiltrating ICOS⁺T-CONV cells and ICOS⁺T-REG varies according to the corresponding cytokine milieu