Altered innate function of plasmacytoid dendritic cells restored by enzyme replacement therapy in Gaucher disease

Blood Cells, Molecules and Diseases 50 (2013) 281–288

Contents lists available at SciVerse ScienceDirect

Blood Cells, Molecules and Diseases journal homepage: www.elsevier.com/locate/bcmd

Altered innate function of plasmacytoid dendritic cells restored by enzyme replacement therapy in Gaucher disease Cécile Braudeau a, b, c, 1, Julie Graveleau e, 1, Marie Rimbert a, b, c, Antoine Néel e, Mohamed Hamidou d, e, Bernard Grosbois f, Audrey Besançon a, Stéphanie Giraudet a, Caroline Terrien a, Régis Josien a, b, c, d,⁎, 2, Agathe Masseau e, 2 a

CHU Nantes, Laboratoire d'Immunologie, Centre d'Immunomonitorage Nantes Atlantique (CIMNA), Nantes F-44000, France CHU Nantes, Institut de Transplantation-Urologie-Néphrologie (ITUN), Nantes F-44000, France c INSERM, UMR1064, Nantes F-44000, France d Université de Nantes, Faculté de Médecine, Nantes F-44000, France e CHU Nantes, Service de Médecine Interne, Nantes F-44000, France f CHU Rennes, Service de Médecine Interne, Rennes F-35000, France b

a r t i c l e

i n f o

Article history: Submitted 22 October 2012 Revised 27 December 2012 Available online 26 January 2013 (Communicated by A. Zimran, M.D., 27 December 2012) Keywords: Gaucher disease Dendritic cells Toll-like receptor Memory T cells

a b s t r a c t Background: Gaucher disease (GD) is caused by an autosomal-recessive deficiency of β-glucocerebrosidase leading to an accumulation of glucosylceramide in monocytes/macrophage lineage. We analyzed immune cells and especially the function of dendritic cells to evaluate the potential impact of glucosylceramide accumulation in these cells and its possible role in infections and malignancies usually described in this pathology. These analyses were performed for each patient without and under enzyme replacement therapy. Methods: Seven GD patients were studied and compared with healthy volunteers. Immune cells (B cells, T cells, NK, dendritic cells), were analyzed by flow cytometry directly on whole blood. Cytokine production by blood dendritic cells was assessed after stimulation by toll-like receptor ligands. Cytokines in sera were measured using a multiplex assay. Results: GD patients displayed decreased numbers of NK cells, γδ2 T cells and increased frequency of memory CD4+CD45RO+ T cells, when compared to healthy controls. Numbers of dendritic cells (myeloid (mDC) and plasmacytoid (pDC) dendritic cells) were also decreased. We demonstrated that pDC from GD patients exhibited a decrease in IFNα production after TLR9 stimulation compared to controls. Importantly, enzyme replacement therapy restored pDC function. Finally, we observed an increase of IL-8 and IL-18 in GD patient sera, which were reduced under enzyme replacement therapy. Conclusions: Our data confirm that patients with GD exhibit altered numbers of innate and T lymphocytes and show for the first time that pDC from GD patients exhibit altered responsiveness to TLR9. These alterations could contribute to a decreased response to pathogens and could favor the development of malignancies. © 2013 Elsevier Inc. All rights reserved.

Introduction Gaucher disease (GD) is a rare autosomal recessive disorder that results from the deficiency of the lysosomal enzyme β-glucocerebrosidase. This is the most frequent lysosomal storage disorder that leads to an accumulation of glucosylceramide in monocytes/macrophage lineage systemically in most patients [1]. The clinical pattern of the disease is very

Abbreviations: DC, dendritic cells; mDC, myeloid dendritic cells; pDC, plasmacytoid dendritic cells; ERT, enzyme replacement therapy; GD, Gaucher disease; TLR, toll-like receptor. ⁎ Corresponding author at: INSERM UMR1064, CHU Nantes Hôtel Dieu, 30 bld Jean Monnet, 44093 Nantes Cedex 1, France. E-mail address: [email protected] (R. Josien). 1 Both authors should be considered as first authors. 2 Both authors should be considered as last authors. 1079-9796/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bcmd.2013.01.001

heterogeneous. Type 1 GD (non-neuronopathic) is the most prevalent form of the disease (90%). Accumulation of macrophages overloaded with glucocerebroside (“Gaucher cells”) is responsible for common features of the disease, such as hepato-splenomegaly, cytopenia (hypersplenism, medullary infiltration), and various bone manifestations. Lipid-laden Gaucher cells in organs (spleen, liver, bone marrow) is the pathologic hallmark of the disease. These “mechanical” consequences are now well understood, but the mechanism leading to other clinical features (such as peripheral polyneuropathy) remains to be elucidated. The main treatment of type 1 GD is based on enzyme replacement therapy (ERT), available since 1991. Three molecules are available to date: imiglucerase, velaglucerase and taliglucerase (off label in France). One substrate reduction therapy (miglustat) is also available. It decreases sphingolipid production, and is indicated for moderate GD when ERT is unsuitable.

4 months ERT 15 months 63 months: 41 ERT, 22 SRT / F 7

46

Splenomegaly, thrombopenia 104 g/L Hepato-splenomegaly 54 M 6

PBC: peripheral blood cell count, ACE: angiotensin converting enzyme, ERT: enzyme replacement therapy, SRT: substrate reduction therapy, NA: not available.

NA NA 15 NA

9 months ERT 9 months 45 months SRT NA 2380 NA NA

Hb 12.7 g/dL PLT 159 g/L Hb 14.2 g/dL PLT 155 g/L Bone pains

12 months ERT 9 months 90 months ERT NA 5760 NA Hepatomegaly 23.5 cm, bone infarcts

Biclonal IgG and IgA kappa gammapathy (7.3 and 6.3 g/L) IgG lambda monoclonal gammapathy (11.9 g/L) / Hepato-splenomegaly, thrombopenia 65 g/L M 5

51

Spleen 13.5 cm, no bone involvement Parkinsonism, homozygous mutation H63D (HFE gene) Hepato-splenomegaly, thrombopenia, bone pains M 4

46

/ Splenomegaly, thrombopenia F 3

35

/ F 2

30

Hepato-splenomegaly, thrombopenia 41 g/L, osteopenia Hepato-splenomegaly M 1

54

NA

9 months ERT 4 months 66 months ERT 0.64 7.3 >1000

90

11 months ERT 4 months 111 months ERT NA 11.1

Hb 14.3 g/dL PLT 137 g/L lymphocytes 1.04 g/L Hb 14.5 g/dL PLT 130 g/L lymphocytes 1.28 g/L Hb 13.6 g/dL PLT 281 g/L

NA

1260

12 months ERT 4 months 26 months ERT 1.26 15 162 Hb 13.9 g/dL PLT 177 g/L

1870

0 0 1.97 NA

Spleen 17 cm, severe osteopenia, diffuse medullar infiltration Spleen 15 cm, hepatomegaly, mild osteopenia No bone involvement /

20.2 1256 Hb 11.5 g/dL PLT 55 g/L

Gamma-globulin (g/L) Ferritin (μg/L) PBC

Clinical presentation at inclusion Clinical presentation at diagnosis Age

Patients from Rennes and Nantes centers were included from December 2009 to October 2010. All patients, except one, were treated by ERT at inclusion. Because of restriction in the enzyme production (viral contamination of the tank), which started in August 2009, most GD patients had interruption in their treatment. The European Medicines Agency (EMA) recommended to continue treatment only in severe diseases. Following these recommendations, patients 2 to 7 included in this study, had interruption in their treatment. Patient 1 had never been treated before inclusion. After validation by the Ethical Committee, we decided to perform immune cell analyses in these untreated patients and to analyze the same patients a second time once they would benefit from ERT again. The first analysis was performed after a minimum of four months of interruption, and the second one after returning to treatment for 4 months minimum. Results were compared with a control group of 10 age-matched healthy controls (HC) (Table 1). Venous blood samples were collected in

Sex

Study patients

Table 1 Patient clinical and biological features at diagnosis and inclusion.

Methods

Associated diseases

Biological features at inclusion

Chitotriosidase (nmol/h mL)

ACE (μkat/L)

Cumulated time of treatment at inclusion

Treatment interruption before sample 1 (“non treated”)

Treatment time before sample 2 (“treated”)

Several macrophagic enzymes are elevated during the course of GD and are used as biomarkers of GD: chitotriosidase, serum tartrate resistant acid phosphatase (TRAP), serum angiotensin conversion enzyme (ACE) and ferritin. The functional consequences of this activated state of macrophages are not fully understood. Increased serum levels of proinflammatory cytokines have been reported in GD patients although results are sometimes discordant. Elevated levels of IL-6, IL-1β, and IL-1Rα are quite constantly reported. Barak et al. reported a correlation between these cytokine levels and clinical severity of GD [2]. Several studies suggested that GD patients display a higher susceptibility to systemic infections and myeloma indicating a possible deficiency of the immune system [3–5]. However, a controversy appears concerning the association of Gaucher disease and cancer [6]. Limited data exist about immune monitoring of GD patients with and without ERT. The results of Lacerda et al. showed that nonsplenectomized GD patients presented a decrease in absolute numbers of peripheral blood T lymphocytes, especially the CD4 + T subset. However, when patients were analyzed with respect to the presence of bone disease, the number of CD8 + T lymphocytes was found to be statistically significantly lower in patients presenting bone involvement [7]. It has been reported that GD patients displayed a decrease in the number of blood NK cells which might be a predisposing factor to the increase in malignancies [8]. Other studies described an alteration of anti-bacterial functions of macrophages, monocytes and neutrophils (impaired chemotaxis, decrease of superoxide anion production) [8–10]. Few years ago, Micheva et al. described a decreased frequency of dendritic cells (DC) within PBMC from non-treated GD patients, but normal functional responses of monocyte-derived DC after in vitro stimulation with LPS or TNFα [11]. Dendritic cells (DC) are critical effectors of adaptive immunity, acting both as sentinels that detect the presence of pathogens and as key antigen-presenting cells that prime and regulate the adaptive immune response. The main mechanism of pathogen recognition by DC is via toll-like receptors (TLR), each with specificity for molecular patterns present in bacteria, fungi, viruses and parasites [12,13]. The scarcity and the activation sensitivity of human blood DC have hampered analysis of their function and their role in GD. In this study, we analyzed various circulating immune cell subsets with a particular focus on DC in GD patients compared with and without ERT. We report that GD patients have decreased numbers of NK cells, γδ T cells, mDC and pDC, and increased frequency of memory CD4+ T cells. We also show that pDC have decreased IFNα response to TLR9 triggering and that this defect is corrected by ERT.

7 months ERT

C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288

Patient

282

C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288

LTγδ2 p=0.02

20

100 0 N

T D G

Cytometry and cytokine production by DC and γδ2 T cells Heparinized whole blood (WB) samples were incubated during 3 h and 30 min at 37 °C under 5% CO2 conditions with TLR ligands for DC stimulation or BrHPP for γδ2 T cells stimulation. GolgiPlug for DC and GolgiStop for γδ2 T cells were added during the last 2 h of incubation to inhibit cellular cytokine release. Control conditions included stimulation with medium alone as negative control. WB samples were then incubated with surface mAbs for 15 min followed by erythrocytes lysis (BD Biosciences). Samples were then fixed, permeabilized with Cytofix/Cytoperm Plus and stained with cytokine-directed mAbs. For the identification and numeration of immune cells EDTA whole blood samples were stained with antibodies following with lysis of red blood cells.

EDTA and heparinized vacutainers and processed for analysis within 4 h. Sera were stocked at − 80 °C until used. All patients and healthy controls provided written informed consent.

Antibodies and reagents Absolute counts of CD4 and CD8 T cells, B cells, NK cells were determined with BD Multitest TM CD3/CD8/CD45/CD4 and CD3/CD19/ CD16+56/CD45 in BD TrucountTM Tubes (BD Biosciences, Le Pont de Claix, France). Dendritic cells were identified using the 6 color flow cytometry assay as we described previously [14]. Briefly, whole blood samples were stained with the following antibodies: CD45-Amcyan, Lineage cocktail 1-FITC, HLA-DR-APC-Cy7, CD11c-PECy7 and CD123-PECy5, all from BD Biosciences and BDCA3-PE from MiltenyiBiotec (Paris, France). TNF-APC (BD Biosciences) IL-12-efluor450 (eBiosciences, Paris, France) and IFNα-PE (MiltenyiBiotec) Abs were used to identify intracellular cytokines after stimulation with TLR ligands. CD45-PerCP (BD Biosciences), CD3-FITC (BD Biosciences), CD4-APC (Beckman Coulter, Marseille, France), CD25-PECy7 (BD-Biosciences) and CD127-PE (Beckman Coulter) were used to identify regulatory T cells. CD3-PECy7 (BD Biosciences) and TCR Vdelta2-FITC (Beckman Coulter) were used to characterize γδ2 T cells. IFNγ-PE (BD Biosciences) was used after γδ2 T cell stimulation. CD3-PECy7 (BD Biosciences), CD4-APC-H7 (BD Biosciences), TCR Valpha24-FITC (Beckman Coulter) and TCR Vbeta11-PE (Beckman Coulter) were used to identify NKT cells. Naïve and memory T cells were characterized using CD45RO-

Detection of cytokines in serum of GD patients Using Luminex technology, levels of 38 cytokines, chemokines and growth factors (17-plex and 21-plex, Bio-Rad, France) were investigated in serum collected from GD patients and healthy control (for the 17-plex: IL1β, IL2, IL4, IL5, IL6, IL7, IL8, IL10, IL12, IL13, IL17, G-CSF, GM-CSF, IFNγ, MCP1, MIP1β, and TNFα and for the 21-plex: IL1a, IL2Ra, IL3, IL12p40, IL16, IL18, CTACK, GROα, HGF, IFNα2, LIF, MCP3, M-CSF, MIF, MIG, bNGF, SCF, SCGFb, SDF1a, TNFβ, and TRAIL). Only six patients have been tested using this assay. Statistics Wilcoxon matched pairs non-parametric t-tests were applied to calculate statistical significance of differences between untreated

CD45RO+

60 40 20

p=0.013

50

% CD4+CD45RA+

40 30 20 10

N D G

D G

C

T N

C H

T

0

0

H

% CD4+CD45RO+

80

Naive CCR7+

CD45RA+

p= 0.002

p=0.043

50 40 30 20 10 0

T

G

Fig. 1. Decrease of NK cells and γδ2 T cell numbers in Gaucher disease. Absolute count of NK cells and γδ2 T cells was determined by flow cytometry on whole blood from healthy controls (NK n=10, γδ2 n=6) and untreated Gaucher disease patients (NK n=7, γδ2 n=7).

N

C

D

H

N

H

T

C

0

D

200

40

G

LTγδ2/µl

NK/µl

300

C

60

H

p=0.04

400

PECy7, CD45RA-PE from BD Biosciences and CCR7-APC from R&D Systems. Naïve and memory B cells were identified using CD19-APC, CD27-PE and IgD-FITC from BD Biosciences. Flow cytometry was performed on a BD FacsCanto II analyzer with DIVA software (BD Biosciences). BrHpp was kindly provided by Dr. Marc Bonneville (INSERM UMR, Nantes, France) and used at 100 μM. TLR ligands: HKLM (Heat killed Listeria monocytogenes, TLR2-L, 108 HKLM/mL), Poly(I:C) (TLR3-L, 100 μg/mL), CL097 (imidazoquinoline compound, TLR7/8-L, 2 μg/mL) and CPG ODN2395 (Type C CPG oligonucleotide, TLR9-L, 50 μM) were obtained from Invivogen (Toulouse, France). LPS (lipopolysaccharides from Escherichia coli O26:B6, TLR4-L, 0.1 μg/mL) was purchased from Sigma-Aldrich (St Louis, MI). GolgiStop, GolgiPlug and Cytofix/Cytoperm Plus were obtained from BD Biosciences.

% CD4+CD45RA+CCR7+

NK

283

Fig. 2. Change in the balance naïve/memory T cells in Gaucher disease. Naïve and memory T cells were identified by flow cytometry using CD45, CD3, CD4, CD45RO, CD45RA and CCR7 antibodies on whole blood from healthy controls (n = 10) and untreated Gaucher disease patients (n = 7). Naïve T cells were identified as CD45RA+ or CD45RA+ CCR7+ cells and memory cells as CD45RA−CD45RO+.

284 C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288 Fig. 3. Gating strategy used to identify blood DC subsets and intracellular cytokine production in DCs in whole blood stimulated with TLR ligands. Whole blood samples were incubated with TLR 2, 3, 4, 7/8 or 9 ligands for 3.5 h and then stained for identification of myeloid DC (HLA-DR+, Lin−, CD11c+, CD123−) and plasmacytoid DC (HLA-DR+, lin−, CD11c−, CD123+) together with intracellular cytokine production (TNFα, IL-12, IFNα).

C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288

285

and treated GD patients. The Mann–Whitney non-parametric t-tests were used to determine statistical significance of differences between healthy control and GD patients.

observed a significant reduction in the absolute numbers of both mDC and pDC compared to healthy controls (mDC: 14.69/μl vs 9.06, p = 0.018 and pDC: 11.53 vs 4.76 p = 0.003) (Fig. 4).

Results

Altered function of pDC

Clinical and biological characteristics of patients

We then assessed the innate function of DC in response to TLR stimulation (gating strategy Fig. 3). For this purpose, intracellular production of TNFα, IL12 and IFNα by mDC and pDC was studied after stimulation of whole blood samples with TLR 2, 3, 4, 7/8 and 9 ligands. Responsiveness of both mDC and pDC from GD patients was not altered after stimulation of TLR 2, 3, 4, 7/8 compared to HC (Fig 5A,B). In contrast, an important reduction in the frequency IFNα-producing pDC upon TLR9 triggering was observed in GD patients as compared to HC (respectively 8.31% vs 17.27%, p = 0.037) (Fig. 5B).

Detection of pro-inflammatory cytokines in serum of Gaucher disease patients We finally assessed serum levels of various inflammatory cytokines and chemokines in GD patient before and under ERT. IL-2, IL-4, IL-5, IL-6, IL-13, IL-17, G-CSF, GM-CSF, IFNγ, IL-1a, IL-2Ra, IL-3, IL-12p40, IFNα2, LIF, MCP3, M-CSF, βNGF, SDF1a, and TNFα were not detected in any samples. The levels of IL-8 and IL-18, two pro-inflammatory cytokines produced mainly by macrophages, were significantly increased in serum of untreated patients as compared to HC (IL-8: p = 0.026, IL-18: p = 0.007) (Fig. 7A). IL-18 serum levels were strongly reduced in one patient and moderately slightly reduced in the other patients once on ERT (p = 0.03 but p = 0.062 when

pDC

mDC Abnormalities in naïve/memory T cell frequencies in GD patients

pDC/µl

10

10

5 5

T

H

T N D G

C

0

0

C

Peripheral blood DC subsets were quantified in whole blood by flow cytometry (Fig. 3). Two subsets of DC can be separated on the basis of CD11c and CD123 expression: so-called myeloid or conventional DC (mDC) that are CD11c + CD123 − and plasmacytoid DC (pDC) that are CD11c − CD123 +. In the group of non treated GD, we

p=0.018

H

Decreased numbers of peripheral blood dendritic cells in patients with GD

p=0.003

15

15

mDC/µl

Naïve/memory T cells were studied in whole blood within CD4 and CD8 subsets using CD45RO, CD45RA and CCR7 markers [19]. GD patients exhibited an important increase in memory CD4 + T cells as compared to healthy control (60.9% vs 37.59%, p = 0.002) together with a significant decrease of naïve CD4 + T cells (24.3% vs 38.58%, p = 0.013) (Fig. 2). In contrast, the frequencies of naïve/memory CD8 + T cells were not modified in GD patients.

20

N

GD patients displayed normal numbers of total leucocytes, B cells, CD4 and CD8 T cells compared to healthy controls. The numbers of regulatory T cells, NKT cells and the frequencies of naïve and memory B cells in GD patients were also similar to that of HC. A significant decrease in the number of NK cells (p = 0.043) and in the frequency of γδ2 T lymphocytes (p = 0.022) was observed in non-treated GD patients (Fig. 1). We assessed the production of IFNγ by γδ2 T cells upon stimulation by synthetic agonists (BrHPP) and found no difference between GD and HC.

We then sought to assess whether these alterations in immune cell numbers and functions in GD patients were corrected under ERT. Numbers of leucocytes, B cells, CD4 and CD8 T cells were similar with and without ERT. ERT did not restore normal numbers of mDC, pDC, NK cells, γδ2 T cells or naïve/memory T cells. In contrast, ERT reversed the defect in IFNα production by pDC after TLR9 stimulation (p = 0.01) as the frequency of IFNα-producing pDC in ERT-treated patients returned to normal values (HC: 17.2%, NT-GD: 8.3%, T-GD: 20.7%). In addition, ERT also induced a significant decrease in the frequencies of pDC producing the pro-inflammatory cytokine TNFα upon TLR7/8 triggering (p = 0.03) (Fig. 6).

D

Reduced number of NK cells and γδ2 T lymphocytes in GD patients

Effect of enzyme replacement therapy on immune cell numbers and functions

G

Seven type 1 GD patients have been included (3 females, 4 males). Mean age at inclusion was 45.3 years (30–54 years). Genotype was known in 3 cases (including 2 brothers). Patient clinical and biological features at inclusion are summarized in Table 1. Age at diagnosis was 7 to 54 years old, and age at treatment initiation was 25 to 54 years old. None of the patients were splenectomized. At inclusion, all patients but one, had received ERT for 26 to 111 months. One patient was naive of treatment (P1). Patient 6 (P6) has been treated only by substrate reduction therapy (SRT) for 45 months, then received ERT after the end of drug shortage. The first blood sampling (“non treated patients”) was done after interruption of treatment for at least 4 months (4 to 15 months). The second one was performed after a duration therapy (ERT) for 7 to 12 months (“treated patients”). We also measured ferritinemia before and after treatment in our cohort of patient, as ferritin has been reported as a biomarker for inflammation associated with GD [15,16]. As already described in the literature [17,18], enzyme replacement therapy significantly reduced total hyperferritinemia (p = 0.031) while increasing blood levels of glycosylated ferritin (p = 0.035) (data not shown). During the course of GD, a monoclonal gammapathy involving either IgG or IgA, light chain kappa or lambda, was detected in 2 patients (28.6%). One patient with a disease under control developed Parkinson's disease. This was confirmed by a neurologist and by 18F-fluoro-L-dopa cerebral scintigraphy, showing a severe dopaminergic denervation. L-dopa treatment was ineffective on symptoms. No particular infectious history was reported.

Fig. 4. Reduction of dendritic cell numbers in Gaucher disease. Myeloid and plasmacytoid DC were enumerated by flow cytometry on whole blood samples from healthy controls (n=11) and untreated Gaucher disease patients (n=7).

C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288

TLR2

60

60

60

60

60

40

20

20

H C

G D

N T G D

G D

G D

40

0

0 H C

N T

H C

G D

N T

20

0

0 H C

40

40

40

40

30

30

30

30

% IL12+mDC

10

20

10

T

C

N D G

G

D

N

H

T

C

0 H

% IL12+mDC T N D G

T G

D G

TLR7/8

20

0

D

N

H

T N

H

B

10

0 C

0 C

0

10

20

C

10

20

H

20

% IL12+mDC

30

% IL12+mDC

% IL12+mDC

20

40

H C

20

40

N T

40

% TNFa+mDC

80

% TNFa+mDC

80

% TNFa+mDC

80

40

TLR9

100

100

80

%TNFa+pDC

80

40 20

0

0

D

H G

G D

H

N

T

C

20

T

40

60

N

60

C

25

100

p= 0.037

20

0

0 T N G

D

H

N

5

C

20

T

10

D

40

15

C

60

H

% IFNa+pDC

80

G

% TNFa+pDC

TLR9

TLR7/8 80

0

% IFNa /pDC

TLR4

TLR3

80

% TNFa+mDC

% TNFa+mDC

A

N T

286

Fig. 5. Altered function of plasmacytoid dendritic cells in Gaucher disease. A. Cytokine production in mDC. Percentage of TNFα and IL12-producing mDC after stimulation with TLR ligands (TLRs 2, 3, 4, 7/8, 9) was not different between untreated Gaucher disease patients (NT, n=7) and HC (n=11). B. Cytokine production in pDC. pDC were responsive to TLR7/8 and 9 only, as expected. A decrease in frequency of IFNα-producing DC following stimulation with TLR9 ligand was observed from untreated Gaucher disease patients (n=7) as compared to HC (n=11). In contrast, levels of TNFα in pDC after stimulation with TLR7/8 or TLR9 ligand were not different between Gaucher disease patients and HC.

omitting the patient with strong reduction) whereas IL-8 levels exhibited a tendency to decrease (p = 0.09) (Fig. 7B). Discussion Gaucher disease is the most common glycolipid storage disorder, caused by mutations in the β-glucocerebrosidase gene, leading to decreased enzymatic activity and accumulation of glucocerebroside in macrophages. Previous studies indicated an increased risk of malignancies in GD patients although controversial reports have been published

[6]. However, the impact of storage diseases on immune cells other than macrophages is not fully understood. In addition, the role of ERT in correcting immune dysfunctions has not been addressed. Here we confirmed that GD patients exhibit a decrease in NK cells, as already described by Burstein et al. [8], and showed that γδ2 T cells were also reduced in these patients. In contrast, T and B cell numbers were not altered, pointing to a specific effect of glucocerebroside accumulation on innate lymphocytes. However, increased frequency of total memory CD4 + T cells in GD patients also suggests that adaptive immune response could be altered. Accordingly, a recent study in a murine

C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288

% TNFa+pDC stimulated with CL097

TLR7/8

p=0.01

p=0.03

100

20 10

80 70 60 N

T

T

0

90

T

30

T

40

N

% IFNa+pDC stimulated with ODN2395

TLR9

Fig. 6. Enzyme replacement therapy restores normal TLR-9-induced IFNα production in pDC from Gaucher disease patients. The frequencies of TNF-α and IFN-α-producing pDC after TLR7/8 and 9 stimulation were assessed in GD patients in the absence or under enzyme replacement therapy (n=7). Statistical analysis was performed using a Wilcoxon non parametric t test.

model of GD showed normal frequency of naïve and memory T cells but increased frequencies in Th17 and Th1 cells in peripheral lymphoid organs [20]. We reasoned that DC, which have, like monocytes/macrophages important phagocytic activity, could also exhibit an altered function in GD patients. Accordingly, we observed a decreased number of circulating DC subsets in GD patients. These changes affected both mDC and pDC. These results are in accordance with a previous study which showed a reduced frequency of both mDC and pDC in

A

IL8 20

IL18

p= 0.026

250

15

200

pg/ml

10 5

150 100 50

0

T

C H

G

G D

D

H

N

C

T

0 N

pg/ml

p= 0.007

IL8 500 400

15

300 200

0

0 N

T

100

T

5

T

pg/ml

20

10

p=0.031

N

25

pg/ml

IL18

p=0.09

T

B

Fig. 7. Increased serum levels of IL8 and IL18 in Gaucher disease. A. The concentration of various cytokines was assessed in sera from untreated Gaucher disease patients (n = 6) and HC (n = 11) using a multiplex assay. Among these cytokines, IL-8 and IL-18 serum levels appeared significantly increased in untreated GD patients as compared to HC. B. Levels of serum IL18 were significantly reduced in the same patients after enzyme replacement therapy (n = 6). Statistical analysis was performed using a Wilcoxon non parametric t test.

287

GD blood samples [11]. Whether the low numbers of DC observed in GD patients are related to a reduced DC production related to bone marrow dysfunction or to increased recruitment of blood DC in tissues remains to be determined. In the murine model of GBA1 deficiency which mimics GD disease, an increased frequency of DC was observed in thymus and spleen [21]. This was confirmed in a more recent study [20] which also showed increased expression of costimulatory and MHC II molecules on the surface of conventional DC indicating that DC functions might also be altered because of glycosphingolipid accumulation. A previous study by Micheva et al. has shown that DC could be derived from GD patient's monocytes although with a reduced yield as compared to control. DC could also be derived from patient's CD34 cell with a normal yield. In addition, the allostimulatory functions of these DC were normal as compared to controls. In the present study, we analyzed for the first time the innate function of blood DC in response to various TLR ligands. Surprisingly, we found that pDC but not mDC were defective in GD patients and that this defect was restricted to the capacity of pDC to produce IFNα in response to TLR9 triggering. The fact that TNFα production by pDC was not decreased in response to TLR9 and that response to TLR7/8 ligands was normal suggests a specific defect in TLR9 signaling leading to type I IFN production. Interestingly, endosomal compartmentalization and trafficking are key determinants of TLR9-elicited IFNα production by pDC [22]. Besides, it has been shown in a mouse model of another genetic sphingolipidose (Niemann–Pick disease C), that the accumulation of glycosphingolipid could disrupt endosomal transport [23]. Further, we found that the defect in TLR9-induced IFNα production by pDC was corrected by ERT. Therefore, we hypothesize that this defect could result from a disruption of endosomal signaling pathways downstream of TLR9, that could be directly due to glycosphingolipid accumulation in pDC. This is surprising as pDC are considered as poorly phagocytic [24,25]. A more detailed study on the morphology and in vitro function of isolated pDC from GD patients would be interesting, but will be technically difficult given the very low numbers of circulating pDC in these patients. This apparent decrease capacity of pDCs to secrete IFNα upon TLR9 triggering may decrease their ability to promote efficient antiviral defensive responses and could participate to the increased incidence of malignancies in Gaucher disease. Interestingly, these abnormalities of pDC function were corrected by ERT, suggesting that DC modifications are related to glycolipid accumulation. Moreover, we also observed that TNFα production by pDC upon TLR9 stimulation was decreased with enzyme replacement therapy, suggesting a possible effect of this treatment on inflammation, although untreated GD patients do not display an elevation of TNFα in stimulated DC and in serum. Some of GD symptoms have been attributed to inflammatory cytokines produced by activated macrophages. Accordingly, increased serum levels of pro-inflammatory cytokines (TNFα, IL-6, IL-8, and IL-1β) have been reported in GD [2,26]. However, other studies reported normal levels of IL-6 and TNFα [27]. Serum levels of M-CSF, sCD14 (a macrophage activation marker), and IL-8 were also increased in GD and correlated with the disease severity [27]. Moreover, increased levels of other mediators including CCL18, CD163, chitotriosidase, IL1Ra and CD14 have been reported [2,27–30]. Our data confirm the significant increase of IL-8 in serum of GD patients. Moreover, IL-18, which is a pro-inflammatory cytokine, appears to be also elevated. We report that the serum levels of both of these cytokines decrease with ERT confirming the effect of ERT in reducing macrophage activation. There are several limits to our study. 1) The first one is related to the limited number of patients included (n = 7). However, despite this low number of patients, we still observed significant differences in immune cell numbers and function as compared to healthy controls. It will therefore be important to confirm these data in larger cohorts of patient especially the observed altered innate function of pDC. Moreover, the strength of our study is related to the fact that

288

C. Braudeau et al. / Blood Cells, Molecules and Diseases 50 (2013) 281–288

immune analyses were performed without and under ERT; 2) Our study is rather descriptive; the biological mechanisms leading to altered pDC functions in GD patients remain to be determined. Although such studies will not be feasible in patients given the very low frequency of pDC, murine models of GD could in contrast be helpful; 3) Patients included in this study do not represent the most severe form of the disease; 4) Finally, it remains to determine whether other functions of DC such as antigen capture and presentation are also altered in GD patients. In conclusion, our study confirms abnormalities of blood DC numbers in patients with type I Gaucher disease. More importantly, and unexpectedly we showed that innate pDC function is altered in GD patients, as shown by the defect in their capacity to produce IFNα in response to TLR9 triggering. Whether this dysfunction is directly related to glucocerebrosidase in pDC remains to be determined. Proinflammatory cytokines which are elevated in GD could possibly play a role in pDC dysfunction. It would be interesting to study in more details pDC function in the mouse model of GD. Acknowledgments We thank Dr. Marc Bonneville and Dr. Jacques Lependu, INSERM, UMR892, Nantes, France, for providing reagents. This work was supported by a grant from the Programme Hospitalier de Recherche Clinique Interregional 2009 to AM. We thank Patrick A. Haslett for his critical reading of the manuscript. References [1] R.O. Brady, The enzymatic defect in Gaucher disease, Prog. Clin. Biol. Res. 95 (1982) 309–314. [2] V. Barak, M. Acker, B. Nisman, et al., Cytokines in Gaucher's disease, Eur. Cytokine Netw. 10 (1999) 205–210. [3] C. Fairley, A. Zimran, M. Phillips, et al., Phenotypic heterogeneity of N370S homozygotes with type I Gaucher disease: an analysis of 798 patients from the ICGG Gaucher Registry, J. Inherit. Metab. Dis. 31 (2008) 738–744. [4] L. Marodi, R. Kaposzta, J. Toth, A. Laszlo, Impaired microbicidal capacity of mononuclear phagocytes from patients with type I Gaucher disease: partial correction by enzyme replacement therapy, Blood 86 (1995) 4645–4649. [5] B.E. Rosenbloom, N.J. Weinreb, A. Zimran, et al., Gaucher disease and cancer incidence: a study from the Gaucher Registry, Blood 105 (2005) 4569–4572. [6] F.Y. Choy, T.N. Campbell, Gaucher disease and cancer: concept and controversy, Int. J. Cell Biol. (2011) 150450. [7] L. Lacerda, F.A. Arosa, R. Lacerda, et al., T cell numbers relate to bone involvement in Gaucher disease, Blood Cells Mol. Dis. 25 (1999) 130–138. [8] Y. Burstein, V. Zakuth, G. Rechavi, Z. Spirer, Abnormalities of cellular immunity and natural killer cells in Gaucher's disease, J. Clin. Lab. Immunol. 23 (1987) 149–151.

[9] Y. Liel, A. Rudich, O. Nagauker-Shriker, T. Yermiyahu, R. Levy, Monocyte dysfunction in patients with Gaucher disease: evidence for interference of glucocerebroside with superoxide generation, Blood 83 (1994) 2646–2653. [10] A. Zimran, D. Elstein, A. Abrahamov, et al., Significance of abnormal neutrophil chemotaxis in Gaucher's disease, Blood 84 (1994) 2374–2375. [11] I. Micheva, T. Marinakis, C. Repa, et al., Dendritic cells in patients with type I Gaucher disease are decreased in number but functionally normal, Blood Cells Mol. Dis. 36 (2006) 298–307. [12] V. Hornung, S. Rothenfusser, S. Britsch, et al., Quantitative expression of toll-like receptor 1–10 mRNA in cellular subsets of human peripheral blood mononuclear cells and sensitivity to CpG oligodeoxynucleotides, J. Immunol. 168 (2002) 4531–4537. [13] T. Kawai, S. Akira, Toll-like receptors and their crosstalk with other innate receptors in infection and immunity, Immunity 34 (2011) 637–650. [14] M. Rimbert, M. Hamidou, C. Braudeau, et al., Decreased numbers of blood dendritic cells and defective function of regulatory T cells in antineutrophil cytoplasmic antibody-associated vasculitis, PLoS One 6 (2011) e18734. [15] M.A. Morgan, A.V. Hoffbrand, M. Laulicht, W. Luck, S. Knowles, Serum ferritin concentration in Gaucher's disease, Br. Med. J. (Clin. Res. Ed.) 286 (1983) 1864. [16] P. Stein, H. Yu, D. Jain, P.K. Mistry, Hyperferritinemia and iron overload in type 1 Gaucher disease, Am. J. Hematol. 85 (2010) 472–476. [17] J. Stirnemann, A. Boutten, C. Vincent, et al., Impact of imiglucerase on the serum glycosylated-ferritin level in Gaucher disease, Blood Cells Mol. Dis. 46 (2010) 34–38. [18] A. Mekinian, J. Stirnemann, N. Belmatoug, et al., Ferritinemia during type 1 Gaucher disease: mechanisms and progression under treatment, Blood Cells Mol. Dis. 49 (2012) 53–57. [19] F. Sallusto, D. Lenig, R. Forster, M. Lipp, A. Lanzavecchia, Two subsets of memory T lymphocytes with distinct homing potentials and effector functions, Nature 401 (1999) 708–712. [20] M.K. Pandey, R. Rani, W. Zhang, K. Setchell, G.A. Grabowski, Immunological cell type characterization and Th1–Th17 cytokine production in a mouse model of Gaucher disease, Mol. Genet. Metab. 106 (2012) 310–322. [21] P.K. Mistry, J. Liu, M. Yang, et al., Glucocerebrosidase gene-deficient mouse recapitulates Gaucher disease displaying cellular and molecular dysregulation beyond the macrophage, Proc. Natl. Acad. Sci. U.S.A. 107 (2010) 19473–19478. [22] C. Guiducci, G. Ott, J.H. Chan, et al., Properties regulating the nature of the plasmacytoid dendritic cell response to Toll-like receptor 9 activation, J. Exp. Med. 203 (2006) 1999–2008. [23] D. te Vruchte, E. Lloyd-Evans, R.J. Veldman, et al., Accumulation of glycosphingolipids in Niemann–Pick C disease disrupts endosomal transport, J. Biol. Chem. 279 (2004) 26167–26175. [24] S.P. Robinson, S. Patterson, N. English, et al., Human peripheral blood contains two distinct lineages of dendritic cells, Eur. J. Immunol. 29 (1999) 2769–2778. [25] J.A. Villadangos, L. Young, Antigen-presentation properties of plasmacytoid dendritic cells, Immunity 29 (2008) 352–361. [26] H. Michelakakis, C. Spanou, A. Kondyli, et al., Plasma tumor necrosis factor-a (TNF-a) levels in Gaucher disease, Biochim. Biophys. Acta 1317 (1996) 219–222. [27] C.E. Hollak, L. Evers, J.M. Aerts, M.H. van Oers, Elevated levels of M-CSF, sCD14 and IL8 in type 1 Gaucher disease, Blood Cells Mol. Dis. 23 (1997) 201–212. [28] R.G. Boot, M. Verhoek, M. de Fost, et al., Marked elevation of the chemokine CCL18/PARC in Gaucher disease: a novel surrogate marker for assessing therapeutic intervention, Blood 103 (2004) 33–39. [29] H.J. Moller, M. de Fost, H. Aerts, C. Hollak, S.K. Moestrup, Plasma level of the macrophage-derived soluble CD163 is increased and positively correlates with severity in Gaucher's disease, Eur. J. Haematol. 72 (2004) 135–139. [30] P.B. Deegan, M.T. Moran, I. McFarlane, et al., Clinical evaluation of chemokine and enzymatic biomarkers of Gaucher disease, Blood Cells Mol. Dis. 35 (2005) 259–267.