doi: 10.1111/j.1365-2222.2007.02792.x
Clinical and Experimental Allergy, 37, 1286–1295 c 2007 The Authors c 2007 Blackwell Publishing Ltd Journal compilation
O R I G I N A L PA P E R
In vivo and in vitro immunomodulation of Der p 1 allergen-specific response by Lactobacillus plantarum bacteria M. Hisbergueswz, M. Magiz, P. Rigauxz, J. Steuvez, L. Garciaz, D. Goudercourt, B. Pot, J. Pestelw1 and A. Jacquetz1 Laboratory of Lactic Acid Bacteria and Mucosal Immunity, Pasteur Institute of Lille, Institute of Biology of Lille, Lille Cedex, France, wPasteur Institute of Lille,
INSERM U416, Lille Cedex, France and zDepartment of Applied Genetics, Free University of Brussels, Gosselies, Belgium
Clinical and Experimental Allergy
Correspondence: Michael Hisbergues, Laboratory of Lactic Acid Bacteria and Mucosal Immunity, Pasteur Institute of Lille, 1, rue du Professeur Calmette, 59019 Lille Cedex, France. E-mail:
[email protected]
Summary Background Lactic acid bacteria (LAB) were reported to reduce some allergic manifestations in mice and humans but their impact on the aeroallergen-dependent immune mechanisms is still debated. Objective The potential capacities of Lactobacillus plantarum NCIMB8826 to reduce the allergic response induced by Der p 1, the major house dust mite allergen of Dermatophagoides pteronyssinus, were evaluated in vivo and in vitro. Methods First, the effect of the intranasal co-administration of LAB and purified Der p 1 allergen before a sensitization protocol was evaluated. The allergen-specific antibody and cellular responses as well as airway inflammation were measured. Second, the impact of LAB on the cytokine profile of spleens cells from Der p 1-sensitized mice was assessed. Third, upon stimulation with LAB, the levels of cytokine produced by dendritic cells derived from the bone marrow (BMDCs) of wild-type, Toll-like receptor 2 (TLR2)-, TLR4- and MyD88-KO mice were compared. Results The co-application of L. plantarum and Der p 1 induced a T-helper type 1 (Th1)-biased allergen-specific IgG response, the absence of specific IgE response and favoured the production of INF-g upon allergen re-stimulation. Moreover, the previous LAB administration reduced the development of bronchoalveolar lavage eosinophilia usually induced by aerosol exposure. Additionally, the studied LAB strain was shown to modify in vitro the cytokine level produced by Der p 1-sensitized spleen cells mainly towards a Th1 profile. Finally, L. plantarum stimulated high IL-12 and moderate IL-10 production in mouse BMDCs notably through the TLR2-, MyD88-dependent and TLR4-independent pathway. Conclusion In vivo co-administration of probiotic LAB with Der p 1 might prevent the development of the mite allergic response. The probiotic L. plantarum was shown to display in vitro therapeutic potentials for the treatment of allergy and to trigger the immune system by a TLR2- and MyD88-dependent signalling pathway. Keywords allergy, Der p 1, house dust mite, lactic acid bacteria, probiotics, T cell polarization, vaccination Submitted 3 January 2007; revised 26 June 2007; accepted 29 June 2007
Introduction The prevalence of allergic diseases has increased dramatically during the last 20 years, especially in developed countries. Among the genetic [1] and environmental factors that contribute to allergy [2], it was suggested that a reduced overall exposure to microbial stimulation in 1
Authors equally contributed to the supervision of this work.
infancy might explain a higher susceptibility to allergens. Recent data on the molecular mechanisms related to the hygiene hypothesis highlight the role of Toll-like receptor (TLR) ligands in the modulation of allergic inflammation and mainly the importance of bacterial products in directing the development of the host’s immune response through activation of TLRs, the main transducers of microbial signals to the innate immune system [3, 4]. Indeed, in human and mice the atopy status
New L. plantarum-based vaccine against house dust mite allergy
was shown to be dependent on the level and exposure frequency of microbial products during pregnancy and the neonatal period [3, 5]. Moreover, the composition of the gut microbiota was reported to be in relation to the development of atopy and asthma [6]. Several studies pointed out a difference in intestinal flora, particularly bifidobacteria, between allergic and non-allergic patients, emphasizing the role of the intestinal flora in allergy development and possible therapeutic intervention via bacterial supplementation [7–9]. Lactic acid bacteria (LAB), which are non-invasive and non-pathogenic Gram-positive commensal microorganisms, are recognized as having health-promoting effects [10]. Together with bifidobacteria, they are among the first bacterial colonizers of the gastrointestinal tract of neonates [10–13]. In mice, recent work highlighted that cell wall components from different bacteria might initiate various immune effects and consequently, via colonic dendritic cells (DC), may induce in the gut different immune responses through the production of immunoregulatory cytokines (notably IL-12 and/or IL-10) [14]. Apart from the protection induced against several infectious diseases, LAB strains were reported to reduce some allergic manifestations in mice and humans [8, 15–18]. In vitro experiments have demonstrated that this anti-allergic effect might be dependent on antigenpresenting cells (APC) and requires IL-12, a key cytokine in the orientation of the immune response towards a T-helper type 1 (Th1) profile [19]. Additionally, LAB can highly reduce the capacity of allergen-specific human T cells to produce Th2 cytokines (IL-4, IL-5), but, in contrast, largely favoured the production of IFN-g [19]. Because mucosa represent the largest area in contact with the external environment and are exposed to a constant load of potentially harmful antigens, attempts to use such LAB in immunotherapy were progressively undertaken. A large body of experiments indicated that administration of antigen vaccine by mucosal route is very efficient [20]. In the context of allergy, nasal administration of different allergens was reported for the induction of immunotherapy as well as for tolerance induction [21–24]. In the present study, we evaluated in mouse the capacity of one LAB strain (Lactobacillus plantarum NCIMB8826) to prevent or modulate allergic immune responses induced by the major mite allergen Der p 1. At first, as LAB vaccine vectors are being extensively studied for mucosal immunization, the potential of the intranasal co-administration of L. plantarum with Der p 1 was studied in a Der p 1-sensitization murine model of allergy. Second, in order to determine the precise part of the mechanism involved in the anti-allergic effect, we analysed the cytokine secretion from LAB-activated murine splenocytes isolated from Der p 1-sensitized mice or naı¨ve bone-marrow DCs. c 2007
1287
Materials and methods Antigens Natural Der p 1 (nDer p 1) were purified as described previously [25]. Endotoxin levels were determined using the limulus assay (Limulus amebocyte lysate QCL 1000, Bio Whittaker, Walkersville, MD, USA). The antigens used for the vaccinations contained o0.2U/10 mg. Animals Female BALB/c mice (7 weeks old) and wild-type (WT) C57BL/6 (B6) (6 weeks old) were obtained from Harlan (AD Horst, the Netherlands). TLR2-, TLR4- and MyD88-KO mice with a B6 background were kindly gifted by Dr F. Trottein (Institut Pasteur de Lille, France), Prof. M. Goldman (IMI, ULB, Belgium) and Dr M. Moser (IBMM, ULB, Belgium), respectively. Bacterial strains and growth conditions L. plantarum NCIMB8826 (National Collection of Industrial and Marine Bacteria, Aberdeen, UK) was isolated from human saliva. The NCIMB8826 strain, carrying the pTG2247 plasmid conferring chloramphenicol resistance [26], was grown at 37 1C in MRS medium (Difco, Becton Dickinson, Sparks, MD, USA) under limited aerobic conditions. Escherichia coli BL21 (DE3), containing the pMW172 plasmid for ampicillin resistance [27] and used as a Gram-negative control strain in the in vitro experiments, was grown at 37 1C in Luria–Bertani medium (Difco, Becton Dickinson) under aeration. Chloramphenicol and ampicillin (Sigma-Aldrich, St Quentin Fallavier, France), when used, were added at 10 and 100 mg/mL, respectively. For in vitro direct stimulation of murine splenocytes, the bacterial strains were grown over-day until the OD600 nm reached 1.8–2 and 1 for Lactobacillus and E. coli strains, respectively. Cells were collected and washed twice with sterile phosphate-buffered saline (PBS), concentrated to 109 colony-forming unit (CFU)/mL before overnight fixation at 4 1C by formalin in phosphate buffer, pH 7.5. Successful inactivation of bacteria was assessed by plating the final suspension on selective agar. Ultimately, the cells were washed again with sterile PBS and resuspended in RPMI complete medium at the desired concentrations. For in vivo pre-treatment, LAB were pre-cultured overnight, re-inoculated and cultured over-day until an OD600 nm of 1.8–2 was reached. Cells were washed twice with PBS and re-suspended to 1011 CFU/mL. Each mouse received this bacterial suspension intranasally with or without purified nDer p 1 in a maximal volume of 30 mL per animal.
The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
1288 M. Hisbergues et al Sensitization × 4
Pre-treatment × 8
Day: 1 2 3 4 7 8 9 10 L. plantarum L. plantarum + Der p 1 Der p 1 PBS
17
24 31 Der p 1 / alum
HDME, house dust mite extract,
Challenge × 7
38
48
killing
54 56 HDME aerosol
, bleeding
Fig. 1. Prophylactic experimental protocol for the evaluation of the immunomodulatory properties of lactic acid bacteria. PBS, phosphate-buffered saline; L. Plantarum, Lactobacillus plantarum.
Preparation of immature bone marrow-derived dendritic cells and cell activation by lactic acid bacteria Bone marrow-derived dendritic cells (BMDCs) were prepared as described [28] with minor modifications. In brief, bone marrow cells, isolated from the femurs and tibias of mice, were cultured at 1.2 106 cells/mL for 8 days in RPMI 1640 supplemented with 10% FBS and 10 ng/mL murine rGMCSF (R&D System, Minneapolis, MN, USA). On day 9, cells were collected, centrifuged and re-suspended in fresh medium. The purity of the DCs was superior to 97% as estimated by CD11c staining. BMDCs from WT, TLR4-, TLR2- or MyD88-KO mice (106 cells/well) were incubated in complete RPMI medium with 107 LAB/well for 16 h at 37 1C. As control, cells were activated with lipopolysaccharide (LPS) (100 ng/mL) (data not shown) or synthetic lipopeptide (1 mg/ mL) N-Palmitoyl-S-[2,3-bis(palmitoyloxy)-(RS)-propyl]-[R]Cys-[S]-Ser1-[S]-Lys(4) trihydrochloride (Pam3CSK4) (InvivoGen, San Diego, CA, USA). After cell stimulation, the supernatants were collected, clarified by centrifugation and stored at 80 1C until cytokine analysis.
In vitro stimulation of spleen cells with bacteria and/or natural Der p 1 5 106 spleen cells from naı¨ve or Der p 1-sensitized female BALB/c mice (intra-peritoneally injected four times at a 1-week interval with 5 mg of purified nDer p 1 adjuvanted with alum) were cultured in a 48-well plate at 37 1C with 5% CO2 (Costar, Cambridge, MA, USA) with 5 105– 5 107 CFU of a fixed L. plantarum strain, as described previously [29]. E. coli was used as a Gram-negative reference strain. For spleen cells from nDer p 1-sensitized mice, 3 mg of nDer p 1 was added per well. The supernatants were collected after 24, 48 and 72 h of incubation and stored at 80 1C until cytokine detections.
facility from the Universit´e Libre de Bruxelles and had free access to tap water and standard mouse chow during the experiments. The pre-treatment, sensitization and aerosol challenge protocols are illustrated in Fig. 1. Pre-treatment. On days 1, 2, 3, 4 and 7, 8, 9 and 10, groups of eight mice were intranasally administered with 30 mL (15 mL per nostril) of LAB containing or not 5 mg of purified nDer p 1. Control groups were intranasally immunized with purified nDer p 1 (5 mg/mouse) alone or saline solution (PBS). Mice were sedated by an intraperitoneal injection of 50 mL of an anaesthetic cocktail [15 mg/mouse Ketamine (Imalge`ne 1000, Merial, Lyon, France), 7.5 mg/ mouse Atropine sulphate (Lab. Aguettant, Lyon, France) and 23.375 mg/mouse Diazepam (Valiums Roche, Roche, Neuilly-sur-Seine, France)] before administration. Sensitization. On days 17, 24, 31 and 38, all groups were intra-peritoneally injected with 5 mg of purified nDer p 1 adjuvanted with alum (Alu-Gel-S, SERVA, Heidelberg, Germany). Mice were bled from the retro-orbital venous plexus on days 1, 17, 38 and 54 and sera were collected. Challenge. To induce airway inflammation, mice were challenged 10 days after the last sensitization by exposures to aerosolized crude Dermatophagoides pteronyssinus extracts as described previously [25]. Measurement of Der p 1-specific immunoglobulin G1, immunoglobulin 2a and immunoglobulin E Sera were assayed for anti-Der p 1 IgG1 and IgG2a and IgE antibodies by ELISA as described [30]. Cytokine assays
Immunization protocols BALB/c mice were housed under specific pathogen-free conditions in micro-isolator cages at the animal care
The levels of IFN-g and IL-5 in the cell culture supernatants upon re-stimulation with the allergen were measured as described before [25]. Mouse IL-10 and IL-12(p40) were measured by ELISA using BD OptEIA kits
c 2007 The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
New L. plantarum-based vaccine against house dust mite allergy
(Becton Dickinson Biosciences, Erembodegen, Belgium) according to the manufacturers’ recommendations. Bronchoalveolar lavage The bronchoalveolar lavages (BAL) as well as the differential cell counts were performed as described previously at day 54 [25]. Statistical analysis The results are expressed as the mean standard deviation. Statistical significance was evaluated for in vivo parameter analysis by one-way ANOVA and a posteriori Dunnett’s test for intergroup comparison when compared with the PBS saline-pre-treated group and followed by a Tukey’s post hoc test for comparison between post-treatment and post-sensitization groups using Analyse-it software. Differences were considered to be significant at P o 0.05 () and either P o 0.01 () and/or P o 0.001 (). For in vitro parameters analysis, the Student test was applied. Results Co-application of lactic acid bacteria and natural Der p 1 induces preferentially a T-helper type 1-biased allergen-specific antibody response The potential adjuvant effect of LAB was assessed in a Der p 1 sensitization model developed in BALB/c mice (Fig. 1). In naı¨ve mice, nDer p 1 alone or co-administered with L. plantarum induced moderate anti-Der p 1 IgG1 responses (Table 1, post-treatment day 17) but strikingly prevented the development of allergen-specific IgE antibodies. Low titres of specific IgG2a were only detected in the group pre-treated with L. plantarum plus nDer p 1 (P o0.01
1289
when compared with PBS saline at day 17). No specific humoral response was observed in animals pre-treated with a saline solution or L. plantarum alone. The subsequent Der p 1 sensitization and aerosol challenge of pre-treated mice led to a significant increase in the specific IgG1 titres [Table 1, post-sensitization, P o 0.05 and P o0.01 when compared with post-treatment (day17)]. At post-sensitization (day 54), a significant induction and boost of the production of specific anti-Der p 1 IgG2a was observed in nDer p 1 or nDer p 11 L. plantarumpre-treated mice (P o0.01 when compared with PBS saline at day 54), respectively (Table 1, post-sensitization). This enhancement of IgG2a titres was significantly different from those observed at post-treatment [P o0.05 and P o 0.01 intra-group comparison at post-treatment (day 17) and post-sensitization (day 54)]. No specific IgG2a was observed in the group treated with L. plantarum alone nor PBS saline solution. As expected, sensitizations with Der p 1 triggered the development of a specific IgE response in groups pretreated by a saline solution (PBS) or L. plantarum only. Strikingly, the absence of specific IgE antibodies still persisted after the injection of alum-adjuvanted Der p 1 and exposure to house dust mite (HDM) extract aerosol in nDer p 1 or nDer p 11L. plantarum-pre-treated mice (P o0.01 when compared with saline solution at day 54). It must be pointed that, for all animal groups, the humoral profiles were not modified after bronchial provocations (data not shown). By taking into account the specific IgG1/IgG2a ratio as well as the absence of allergen-specific IgE in both Der p 1-pre-treated groups, it might be clearly suggested that L. plantarum is of major importance for the development and persistence of a strong specific Th1-biased immune response, even in the presence of a pro-Th2 adjuvant such as alum in the sensitization phase.
Table 1. Modulation of the IgG1, IgG2a and IgE Der p1-specific responses in mice IgG1
IgG2a
IgG1/IgG2a
IgE
Intranasal treatment before sensitization
Posttreatment
Postsensitization
Posttreatment
Postsensitization
Postsensitization
Posttreatment
Postsensitization
Lactobacillus plantarum L. plantarum1nDer p 1 nDer p 1 PBS
0 2953 2774 3919 2980 0
144 375 150 581# 228 500 170 795## 215 275 77 255## 124 805 85 239#
0 449 283 0 0
0 7475 2633,## 705 704# 0
47219 31 305 46240
0 0 0 0
0.11 0.077## 0 0.012## 0.02 0.015,## 0.115 0.079##
Specific anti-Der p 1 IgG1, IgG2a and IgE titres from mice after treatment (post-treatment, day 17) and after sensitization and aerosol challenge (postsensitization, day 54). IgE titres are expressed as OD410 nm values for a 1 : 10 dilution of sera. Values are expressed as means of individual titres (n = 8) SD. ,Significance at P o 0.05 and 0.01 after ANOVA and a posteriori Dunnett’s test when compared with the PBS saline pre-treated group. #,## Significance at P o 0.05 and 0.01 after ANOVA and a posteriori Tukey’s test for intragroup comparison at post-treatment (day17) and post-sensitization (day 54). nDer p 1, natural Der p 1; PBS, phosphate-buffered saline. c 2007
The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
1290 M. Hisbergues et al (a) 250
(b)
**
2500
IFN-γ (pg / mL)
200
IL-5 (pg / mL)
3000
150 100
50
2000 1500 1000 500 0
0 + nDer p 1
nDer p 1
Pre-treatment L. plantarum
PBS solution
+ nDer p 1
nDer p 1
L. plantarum
PBS solution
Fig. 2. IL-5 and IFN-g release from antigen-stimulated splenocytes at killing (day 54). IL-5 (a) and IFN-g (b) production by natural Der p 1 (nDer p 1)restimulated spleen cells, from mice pre-treated by Lactobacillus plantarum1nDer p 1, nDer p 1 alone or phosphate-buffered saline (PBS) solution. Results are expressed by means standard deviations from eight mice. Significance according to ANOVA and a posteriori Dunnett’s test at P o 0.01, respectively, when compared with the PBS solution-pre-treated group; L. plantarum, Lactobacillus plantarum.
(b) 100 90 80 70 60 50 40 30 20 10 0 Cell percentage
Cell number (× 105/mL)
(a) 100 90 80 70 60 50 40 30 20 10 0
*
Macrophages
Lymphocytes
Neutrophils
Eosinophils
Cell type Pre-treatment:
L. plantarum,
L. plantarum + nDer p 1,
nDer p 1,
PBS solution
Fig. 3. Cellular composition in bronchoalveolar lavage fluids after challenge (day 54) with house dust mite aerosols. Total cell counts (a) were performed in a Thoma chamber. Cellular composition (b) is expressed as cell percentage of the four cell populations. Results are expressed as means standard deviations of eight mice. Significance after ANOVA and a posteriori Dunnett’s test at P o0.05 when compared with the phosphate-buffered saline (PBS) solution-pre-treated group. nDer p 1, natural Der p 1; L. plantarum, Lactobacillus plantarum.
Co-application of lactic acid bacteria favoured the production of interferon-g instead of interleukin-5
Lactic acid bacteria administration reduced the bronchoalveolar lavage eosinophilia
The IL-5 production from antigen-induced splenocyte isolated after sensitization and challenge was increased in either nDer p 1-, L. plantarum- or PBS-pre-treated mice, but not in L. plantarum1nDer p 1-pre-treated animals (Fig. 2). Conversely, the specific stimulation of spleen cells increased the secretion of IFN-g only in the L. plantarum-pre-treated groups (L. plantarum or L. plantarum1nDer p 1). However, IFN-g release was higher and significant only in the group pre-treated with L. plantarum1nDer p 1(P o0.01 when compared with PBS saline solution). The IFN-g/IL-5 ratio confirmed that LAB can impair the development of allergen-specific Th2 cells.
Eosinophils were detected in BAL fluids from all mice sensitized with nDer p 1 and challenged with HDM aerosols (Fig. 3). However, mice pre-treated with L. plantarum in the presence or absence of nDer p 1 showed a marked decrease of eosinophilia (47% and 58% decrease, respectively) compared with the PBS solutionpre-treated animals. A weaker decrease of eosinophil content (i.e. 24%) was also observed in the nDer p 1-pre-treated group. However, this reduction was statistically significant for the group pre-treated with L. plantarum alone (P o 0.05, when compared with the PBS solution-pre-treated group). Therefore, our results clearly
c 2007 The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
New L. plantarum-based vaccine against house dust mite allergy
1291
E. coli 100
*** ***
***
*** ***
***
*** ***
60 40
***
L. plantarum
20
*
50
*
** ***
***
100
80
*** ***
IFN-γ (ng/mL)
L. plantarum
***
** ***
200
***
E . coli
Medium
IL-5 (pg/ mL)
250
150
(b) 120
***
Medium
(a) 300
0
0 5×107 5×105
L. plantarum
(d) 6000
0
***
***
*
E. coli 1000
**
***
*** ***
***
***
***
*** ***
** *
2000
2000
Medium
4000
3000
***
E. coli
6000
L. plantarum
4000 ***
8000
IL-12 (pg / mL)
5000
Medium
IL-10 (pg / mL)
10 000
***
12 000
***
(c)
0
Fig. 4. Cytokine expression in Der p 1-sensitized murine spleen cells re-stimulated by the allergen in the presence or not of bacteria. IL-5 (a), IFN-g (b), IL-10 (c) and IL-12 (d) levels are expressed as means standard deviations (n = 3) with murine spleen cells from naı¨ve animals (grey bars) or sensitized ones and re-stimulated with nDer p 1 (hatched bars). , and Significance according to the Student test at P o0.05, P o0.01 and P o0.001, respectively, when compared with medium; L. plantarum, Lactobacillus plantarum; E. coli, Escherichia coli; nDer p 1, natural Der p 1.
indicated that the presence of LAB itself can impair the development of airway eosinophilia.
In vitro immune modulation capacity of lactic acid bacteria on spleen cells Previous in vitro studies demonstrated the capacity of LAB strains to modulate the immune response, e.g. through cytokine stimulation of mouse naı¨ve spleen cells [29, 31]. We further investigated the immunomodulatory properties of L. plantarum NCIMB 8826 using spleen cells from nDer p 1-sensitized mice. These cells were subsequently in vitro restimulated by the allergen in the presence or absence of L. plantarum. An E. coli BL21 (DE3) strain was used as a Gram-negative reference strain. Results were compared with those obtained with spleens from naı¨ve mice (Fig. 4). A typical Th2 cytokine profile characterized by IL-5 production and low secretion of IFN-g, IL-10 and IL-12(p40) was elicited by allergen re-stimulation of splenocytes from nDer p 1-sensitized mice (Fig. 4, medium control upon restimulation). In contrast, L. plantarum was able to shift the Th2 profile towards a Th1 or Treg-biased one as judged by the large amounts of IFN-g, IL-12(p40), IL-10 (P o 0.001) and the very weak IL-5 secretion upon re-stimulation. The Gram-negative reference strain E. coli was shown to be less c 2007
effective than the L. plantarum strain in modulating the Th2 immune response on nDer p 1 re-stimulation, since low secretions of IL-10, IL-12(p40) and intermediate IL-5 level were observed (P o0.001–0.05, when compared with medium control on nDer p 1 re-stimulation). Similar productions of cytokines were obtained with living bacteria (data not shown). Surprisingly, the modulation of cytokine production in nDer p 1-sensitized cells, re-stimulated by the allergen, was shown to be inversely dependent on the LAB cell number whereas a classical LAB dose dependence was observed when the naı¨ve spleen cells were only activated with L. plantarum (Figs 4a and d). Altogether, these results indicated that L. plantarum has the capacity to skew the cytokine profile of an established Th2 allergic response towards Th1 or Treg type.
In vitro immunomodulation capacity of lactic acid bacteria on bone marrow-derived dendritic cells DCs, known as professional APC, are involved in the polarization of the immune response [32, 33]. The adaptative immune response of T cells (i.e. Th1, Th2 or regulatory) depends notably on the chemokines and cytokines generated by DC, in response to a variety of stimuli, mainly the TLR ligands.
The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
1292 M. Hisbergues et al
(a) 1000
200 100 0
$
60 IL-12 (ng/ mL)
IL-10 (pg/ mL)
700 600 500 400 300
(b) 70
$
900 800
$ $
* *
L. plantarum
Medium
Pam3Cys
$ $
50 40 30
* $
20 10
* *
$ $
WT TLR4-KO TLR2-KO MyD88-KO
* $ *
0 L. plantarum
Medium
* * $ Pam3Cys
Fig. 5. Immunomodulation activity of lactic acid bacteria (LAB) on murine bone marrow-derived dendritic cells (BMDCs). IL-10 (a) and IL-12 (p40) (b) expression of BMDCs from mice harbouring null mutations in MyD88 (white bars), Toll-like receptor 2 (TLR2) (grey bars), TLR4 (hatched bars) or no mutation [wild type (WT), black bars] stimulated by LAB, Pam3CSK4 or control (medium), n = 3. $Significance according to the Student test at P o 0.05 when compared with WT and medium, respectively; L. plantarum, Lactobacillus plantarum.
To study the direct activation of DCs by LAB and to consider the impact of TLR signalling on LAB-treated DC, BMDCs prepared from WT, MyD88 / , TLR2 / and TLR4 / mice were incubated with the probiotic L. plantarum strain for 16 h. The production of IL-10 and IL-12(p40) was then analysed in the culture supernatants. As controls, BMDCs were stimulated with LPS (a TLR4 ligand, data not shown) or Pam3CSK4 (a synthetic TLR2 ligand) (Fig. 5). L. plantarum stimulated significantly less IL-10 and IL12(p40) production in BMDCs of TLR2 / and MyD88 / (P o0.05). By contrast, TLR4 / and WT BMDCs displayed a similar cytokine production after stimulation with L. plantarum. As expected, the cytokine secretion triggered either by Pam3CSK4 or LPS was completely abrogated in TLR2 / /MyD88 / (P o 0.05) or TLR4 / /MyD88 / BMDCs, respectively (data not shown). BMDCs from all the mouse strains tested did not release detectable IL-10 and IL-12(p40) after incubation with culture medium alone. Thus, L. plantarum mediates the cytokine production through at least a TLR2- and MyD88-dependent signalling pathway. Discussion Recent reports on the molecular mechanisms related to the hygiene hypothesis highlighted the importance of bacterial products in directing the development of the host’s immune system [5]. Notably, the role of bacterial TLR ligands in modulating allergic inflammation was clearly evidenced [3, 5]. LABs were described as having a beneficial effect on health. Within the last months, few reports underlined their capacity to reduce the allergic manifestations such as eczema [34], airway inflammation [35] or food allergy [36] when applied perinatally (pregnancy and lactation) and/or after birth. Until now, experi-
mental data of their potential impact on defined parameters required for the development of an allergic reaction to a native aeroallergen were limited. In this study, the potential immunomodulatory activities of the L. plantarum NCIMB 8826 strain in the development of the allergic reaction induced by the mite allergen Der p 1 were investigated for the first time. Der p 1 is one of the most clinically relevant allergens world-wide characterized by its proteolytic activity (cysteine protease) recently shown to be crucial for the development of IgE and IgG responses [37]. First of all, using a Der p 1-sensitization murine model of HDM allergy, we clearly demonstrated that co-administration of L. plantarum with the allergen can prevent the development of a specific anti-Der p 1 IgE response. In our experimental conditions, a prophylactic nasal co-instillation of L. plantarum1nDer p 1 induced remarkable levels of specific IgG2a antibodies and no IgE response. Moreover, the subsequent sensitization by multiple injections of nDer p 1 adjuvanted with alum to mice did not lead to the development of an allergic Th2 profile in the co-administrated group, confirming the development of a Th1 profile by this formulation. Groups pretreated either with PBS or L. plantarum alone developed an allergic response. In contrast, the intranasal administration of native Der p 1 in the absence of protecting LAB induced a specific IgG1 response, even if no specific IgE was detected. Thus, the vaccination with nDer p 1 and L. plantarum induced a Der p 1-specific Th1 cell response able to counterbalance a subsequent Th2 immune challenge induced by a strong nDer p 1-alum sensitization. The protection offered by the L. plantarum strain used was evidenced by high titres of specific IgG2a, absence of specific IgE, a high level of IFN-g and weak IL-5 secretion, and confirmed the absolute requirement of L. plantarum
c 2007 The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
New L. plantarum-based vaccine against house dust mite allergy
NCIMB8826 to induce an anti-allergic effect in the mouse model of HDM allergy. Although the absence or the reduction of IgE was previously demonstrated with Betv1[38] or Der p 5- [39] expressing LAB orally or intranasally administrated, to our knowledge, this is the first study that shows such a protection using the full-length major dust mite allergen Der p 1 in mice. The immunomodulatory properties of a recombinant L. plantarum 256 expressing an immunodominant T cell epitope of Der p 1 were reported in C57BL/6 J mice [40]. However, the administration of the recombinant strain led to the inhibition of IFN-g and IL-5 production but not to an alteration of the specific antibody responses. The fact that another L. plantarum strain was used might explain to some extent the different results obtained as the immunomodulation capacity of lactobacilli might also be strain dependent ([41] and M. Hisbergues, unpublished results). The characteristics of the administrated allergen (native protein vs. epitope expressed in LAB) might explain the different observed immune impacts. The second parameter evaluated was the intensity of eosinophilia following allergen aerosol challenge, which represents a key feature of the allergic reaction. Interestingly, mice vaccinated with L. plantarum showed a significant reduction of eosinophilia (58%) vs. the PBS solution-pre-treated group. This eosinophil decrease was in contrast to the observation of Charng et al. [39] with Lactobacillus acidophilus ATCC4356. The presence of neutrophils in BAL is most likely a result of a local acute inflammation as mentioned for the prophylactic model of birch pollen allergy [20]. The neutrophil recruitment observed in our nDer p 1 model might be related to the proinflammatory properties of the LAB strain used, and/or may be associated with chemokine release from airway epithelial cells induced by the proteolytic activity of nDer p 1 [42]. When L. plantarum was co-administrated with the allergen, the reduction of eosinophilia was confirmed (i.e. 47%), suggesting that the protective effect of LAB was maintained even after the allergen challenge, in agreement with the tendency observed with the recombinant Bet v 1-expressing LAB [38]. Third, we assessed the immunomodulatory properties of L. plantarum by measuring the cytokine production in spleen cells from nDer p 1-sensitized mice after in vitro restimulation with the allergen in the presence or absence of the LAB. L. plantarum was found to modify the Th2 cytokine profile following allergen stimulation as low levels of the pro-Th2 cytokine (IL-5) and high amounts of the pro-Th1 cytokine (IFN-g) and the Treg cytokine (IL-10) were produced. These results were obtained with fixed LAB and were inversely dose related. This inverse relationship between the cytokine amount and the bacterial cell number was observed previously for IL-4 release [29] and other cytokines (M. Hisbergues, unpublished results) with LAB-stimulated sensitized spleen cells. In our c 2007
1293
conditions, the higher activity on the cytokine release obtained upon re-stimulation with nDer p 1 was observed at the lowest concentration (5 105 bacteria) (Fig. 4). Apparently, an increased level of bacteria might reduce the proinflammatory effects observed on splenocytes, which could be related to unknown metabolic or enzymatic effects. Other immunological effects cannot be excluded although in the literature there are few (or no) examples of such an inverse dose relationship. However, recent data indicated the importance of the LAB state or their mode of inactivation [43] in cytokine production in human cell lines. Indeed, the proteolytic activity of Der p 1 was reported to decrease IL-12 production by DC [44]. Thus, the potential involvement of a bacterial compound released by the proteolytic action of nDer p 1 on bacteria therefore modifying the cytokine release cannot completely be ruled out. Finally, in an attempt to obtain some hints of the possible mechanism of how LAB can modulate the TH polarization, we further investigated the cytokine production in LAB-activated BMDC from various genetic backgrounds. We clearly demonstrated that the L. plantarum strain used in our experimental model stimulated high IL12 and low IL-10 secretion in BMDCs. Similar results were previously obtained with stimulated human PBMC [45] or monocyte-derived DCs using the L. plantarum strains NCIMB8826 [19]. However, the production of IL-12 and/ or IL-10 was reported as being different from colonic DC incubated with another probiotic, Bifidobacterium longum [14], suggesting that these DC in intestinal lymphoid tissue might behave differently from DC at other sites. Very recently, our team [46] demonstrated that in vitro stimulating BMDCs with different LAB strains allow their classification as pro-inflammatory or non-inflammatory according to the production of cytokines and chemokines, particularly IL-10 and IL-12. This ability to induce IL-10 production by DC was already suggested by Riggby et al. [14] to be a characteristic of certain strains rather than a generalized property in Gram-positive bacteria. Experiments carried out with BMDCs from TLR- or MyD88deficient mice showed that cytokine secretion triggered by L. plantarum was TLR2- and MyD88 dependent but TLR4 independent. To our knowledge, this study provides the first description of the activation of the TLR2, MyD88dependent signalling pathway by L. plantarum. Our previous studies showed that lipoteichoic acids (LTAs) could more likely represent the pathogen-associated molecular pattern responsible for the TLR2 activation by LAB as demonstrated for NCIMB 8826 WT and its LTA mutant [45]. Thus, the IL-12/IL-10 ratio observed after BMDC stimulation underlined the capacity of L. plantarum to induce a Th1 or Treg profile preferentially. More recently, in sensitized humans [47], TLR2 ligands such as LTAs of Gram-positive bacteria were shown to inhibit allergenspecific Th2 responses, supporting strongly the idea that TLR2 ligands such as LAB might be used in pre-clinical
The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
1294 M. Hisbergues et al studies for the potential treatment of allergic diseases. Finally, it was recently suggested that selected probiotics can stimulate DC-regulatory functions by targeting specific pattern-recognition receptors and pathway in an intestinal inflammatory mice model and could be of use in immune intervention therapy for intestinal bowel disease [46]. Taken together, the present data clearly showed that LAB can modulate in vivo as well in vitro the various immune parameters required for the development of an allergic reaction. Thus, the intranasal co-administration of L. plantarum NCIMB8826 and nDer p 1 might represent an efficient vaccine strategy to reduce the deleterious effect related to HDM allergy.
12 13
14
15
16
Acknowledgements This work was supported in part by the Walloon Region (Direction G´en´erale des Technologies, de la Recherche et de l’Energie) of Belgium. We thank Louis Delhaye for excellent technical assistance with the animal facility.
17
18
References 1 Brasch-Andersen C, Haagerup A, Borglum AD, Vestbo J, Kruse TA. Highly significant linkage to chromosome 3q13.31 for rhinitis and related allergic diseases. J Med Genet 2006; 43:e10. 2 Prioult G, Nagler-Anderson C. Mucosal immunity and allergic responses: lack of regulation and/or lack of microbial stimulation? Immunol Rev 2005; 206:204–18. 3 Ege MJ, Bieli C, Frei R et al. Prenatal farm exposure is related to the expression of receptors of the innate immunity and to atopic sensitization in school-age children. J Allergy Clin Immunol 2006; 117:817–23. 4 Vercelli D. Mechanisms of the hygiene hypothesis – molecular and otherwise. Curr Opin Immunol 2006; 18:733–7. 5 Wang Y, McCusker C. Neonatal exposure with LPS and/or allergen prevents experimental allergic airways disease: development of tolerance using environmental antigens. J Allergy Clin Immunol 2006; 118:143–51. 6 Noverr MC, Huffnagle GB. The ‘microflora hypothesis’ of allergic diseases. Clin Exp Allergy 2005; 35:1511–20. 7 Ouwehand AC, Isolauri E, He F, Hashimoto H, Benno Y, Salminen S. Differences in Bifidobacterium flora composition in allergic and healthy infants. J Allergy Clin Immunol 2001; 108:144–5. 8 Kalliomaki M, Isolauri E. Role of intestinal flora in the development of allergy. Curr Opin Allergy Clin Immunol 2003; 3:15–20. 9 Watanabe S, Narisawa Y, Arase S et al. Differences in fecal microflora between patients with atopic dermatitis and healthy control subjects. J Allergy Clin Immunol 2003; 111:587–91. 10 Mercenier A, Pavan S, Pot B. Probiotics as biotherapeutic agents: present knowledge and future prospects. Curr Pharm Des 2003; 9:175–91. 11 Harmsen HJ, Wildeboer-Veloo AC, Raangs GC et al. Analysis of intestinal flora development in breast-fed and formula-fed
19
20
21
22
23
24
25
26
infants by using molecular identification and detection methods. J Pediatr Gastroenterol Nutr 2000; 30:61–7. Salminen S, Isolauri E. Intestinal colonization, microbiota, and probiotics. J Pediatr 2006; 149:S115–20. Hopkins MJ, Macfarlane GT, Furrie E, Fite A, Macfarlane S. Characterisation of intestinal bacteria in infant stools using realtime PCR and northern hybridisation analyses. FEMS Microbiol Ecol 2005; 54:77–85. Rigby RJ, Knight SC, Kamm MA, Stagg AJ. Production of interleukin (IL)-10 and IL-12 by murine colonic dendritic cells in response to microbial stimuli. Clin Exp Immunol 2005; 139:245–56. Fujiwara D, Inoue S, Wakabayashi H, Fujii T. The anti-allergic effects of lactic acid bacteria are strain dependent and mediated by effects on both TH1/TH2 cytokine expression and balance. Int Arch Allergy Immunol 2004; 135:205–15. Prescott SL, Dunstan JA, Hale J et al. Clinical effects of probiotics are associated with increased interferon-gamma responses in very young children with atopic dermatitis. Clin Exp Allergy 2005; 35:1557–64. Ishida Y, Nakamura F, Kanzato H et al. Clinical effects of Lactobacillus acidophilus strain L-92 on perennial allergic rhinitis: a double-blind, placebo-controlled study. J Dairy Sci 2005; 88:527–33. Peng GC, Hsu CH. The efficacy and safety of heat-killed Lactobacillus paracasei for treatment of perennial allergic rhinitis induced by house-dust mite. Pediatr Allergy Immunol 2005; 16: 433–8. Pochard P, Hammad H, Ratajczak C et al. Direct regulatory immune activity of lactic acid bacteria on Der p 1-pulsed dendritic cells from allergic patients. J Allergy Clin Immunol 2005; 116:198–204. Wiedermann U. Mucosal immunity–mucosal tolerance. A strategy for treatment of allergic diseases. Chem Immunol Allergy 2003; 82:11–24. Liu YH, Tsai CF, Kao MC, Lai YL, Tsai JJ. Effectiveness of Dp2 nasal therapy for Dp2-induced airway inflammation in mice: using oral Ganoderma lucidum as an immunomodulator. J Microbiol Immunol Infect 2003; 36:236–42. Palomares O, Batanero E, Canamero M, Villalba M, Rodriguez R. Prophylactic intranasal treatment with fragments of 1,3-betaglucanase olive pollen allergen prevents airway inflammation in a murine model of type I allergy. Int Arch Allergy Immunol 2006; 139:175–80. Winkler B, Bolwig C, Seppala U, Spangfort MD, Ebner C, Wiedermann U. Allergen-specific immunosuppression by mucosal treatment with recombinant Ves v 5, a major allergen of Vespula vulgaris venom, in a murine model of wasp venom allergy. Immunology 2003; 110:376–85. Keller AC, Mucida D, Gomes E et al. Hierarchical suppression of asthma-like responses by mucosal tolerance. J Allergy Clin Immunol 2006; 117:283–90. Magi M, Garcia L, Vandenbranden M, Palmantier R, Jacquet A. Heat denaturation affects the ProDer p 1 IgE reactivity and downregulates the development of the specific allergic response. J Allergy Clin Immunol 2004; 114:545–52. Hols P, Slos P, Dutot P et al. Efficient secretion of the model antigen M6-gp41E in Lactobacillus plantarum NCIMB 8826. Microbiology 1997; 143:2733–41.
c 2007 The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
New L. plantarum-based vaccine against house dust mite allergy
27 Studier FW, Rosenberg AH, Dunn JJ, Dubendorff JW. Use of T7 RNA polymerase to direct expression of cloned genes. Methods Enzymol 1990; 185:60–89. 28 Lutz MB, Kukutsch N, Ogilvie AL et al. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods 1999; 223:77–92. 29 Repa A, Grangette C, Daniel C et al. Mucosal co-application of lactic acid bacteria and allergen induces counter-regulatory immune responses in a murine model of birch pollen allergy. Vaccine 2003; 22:87–95. 30 Jacquet A, Magi M, Haumont M, Jurado M, Garcia L, Bollen A. Absence of immunoglobulin E synthesis and airway eosinophilia by vaccination with plasmid DNA encoding ProDer p 1. Clin Exp Allergy 2003; 33:218–25. 31 Shida K, Makino K, Morishita A et al. Lactobacillus casei inhibits antigen-induced IgE secretion through regulation of cytokine production in murine splenocyte cultures. Int Arch Allergy Immunol 1998; 115:278–87. 32 Neurath MF, Finotto S, Glimcher LH. The role of TH1/TH2 polarization in mucosal immunity. Nat Med 2002; 8:567–73. 33 de Jong EC, Smits HH, Kapsenberg ML. Dendritic cell-mediated T cell polarization. Springer Semin Immunopathol 2005; 26:289–307. 34 Abrahamsson TR, Jakobsson T, Bottcher MF et al. Probiotics in prevention of IgE-associated eczema: a double-blind, randomized, placebo-controlled trial. J Allergy Clin Immunol 2007; 119:1174–80. 35 Blumer N, Sel S, Virna S et al. Perinatal maternal application of Lactobacillus rhamnosus GG suppresses allergic airway inflammation in mouse offspring. Clin Exp Allergy 2007; 37:348–57. 36 Flinterman AE, Knol EF, van Ieperen-van Dijk AG et al. Probiotics have a different immunomodulatory potential in vitro vs. ex vivo upon oral administration in children with food allergy. Int Arch Allergy Immunol 2007; 143:237–44. 37 Kikuchi Y, Takai T, Kuhara T et al. Crucial commitment of proteolytic activity of a purified recombinant major house dust mite allergen Der p1 to sensitization toward IgE and IgG responses. J Immunol 2006; 177:1609–17.
c 2007
1295
38 Daniel C, Repa A, Wild C et al. Modulation of allergic immune responses by mucosal application of recombinant lactic acid bacteria producing the major birch pollen allergen Bet v 1. Allergy 2006; 61:812–9. 39 Charng YC, Lin CC, Hsu CH. Inhibition of allergen-induced airway inflammation and hyperreactivity by recombinant lacticacid bacteria. Vaccine 2006; 24:5931–6. 40 Kruisselbrink A, Heijne Den Bak-Glashouwer MJ, Havenith CE, Thole JE, Janssen R. Recombinant Lactobacillus plantarum inhibits house dust mite-specific T-cell responses. Clin Exp Immunol 2001; 126:2–8. 41 Foligne B, Nutten S, Grangette C et al. Correlation between in vitro and in vivo immunomodulatory properties of lactic acid bacteria. World J Gastroenterol 2007; 13:236–43. 42 Adam E, Hansen KK, Astudillo OF et al. The house dust mite allergen Der p 1, unlike Der p 3, stimulates the expression of interleukin-8 in human airway epithelial cells via a proteinaseactivated receptor-2-independent mechanism. J Biol Chem 2006; 281:6910–23. 43 Wong C, Ustunol Z. Mode of inactivation of probiotic bacteria affects interleukin 6 and interleukin 8 production in human intestinal epithelial-like Caco-2 cells. J Food Prot 2006; 69:2285–8. 44 Ghaemmaghami AM, Gough L, Sewell HF, Shakib F. The proteolytic activity of the major dust mite allergen Der p 1 conditions dendritic cells to produce less interleukin-12: allergen-induced TH2 bias determined at the dendritic cell level. Clin Exp Allergy 2002; 32:1468–75. 45 Grangette C, Nutten S, Palumbo E et al. Enhanced antiinflammatory capacity of a Lactobacillus plantarum mutant synthesizing modified teichoic acids. Proc Natl Acad Sci USA 2005; 102: 10321–6. 46 Foligne B, Zoumpopoulou G, Dewulf J et al. A key role of dendritic cells in probiotic functionality. PLoS ONE 2007; 2:e313. 47 Taylor RC, Richmond P, Upham JW. Toll-like receptor 2 ligands inhibit TH2 response to mite allergen. J Allergy Clin Immunol 2006; 117:1148–54.
The Authors c 2007 Blackwell Publishing Ltd, Clinical and Experimental Allergy, 37 : 1286–1295 Journal compilation
