Bifidobacterium breve and Lactobacillus rhamnosus treatment is as effective as budesonide at reducing inflammation in a murine model for chronic asthma

Background Asthma is estimated to affect as many as 300 million people worldwide and its incidence and prevalence are rapidly increasing throughout the world, especially in children and within developing countries. Recently, there has been a growing interest in the use of potentially beneficial bacteria for allergic diseases. This study is aimed at exploring the therapeutic effects of long-term treatment with two different beneficial bacterial strains (Bifidobacterium breve M-16 V and Lactobacillus rhamnosus NutRes1) and a glucocorticoid (budesonide), as a reference treatment, on inflammatory response in a murine model for chronic allergic asthma. Methods To mimic the chronic disease in asthmatic patients, we used the murine ovalbumin-induced asthma model combined with prolonged allergen exposure. Airway function; pulmonary airway inflammation; airway remodelling, mRNA expression of pattern recognition receptors, Th-specific cytokines and transcription factors in lung tissue; mast cell degranulation; in vitro T cell activation; and expression of Foxp3 in blood Th cells were examined. Results Lactobacillus rhamnosus reduced lung resistance to a similar extent as budesonide treatment in chronically asthmatic mice. Pulmonary airway inflammation, mast cell degranulation, T cell activation and airway remodelling were suppressed by all treatments. Beneficial bacteria and budesonide differentially modulated the expression of toll-like receptors (TLRs), nod-like receptors (NLRs), cytokines and T cell transcription factors. Bifidobacterium breve induced regulatory T cell responses in the airways by increasing Il10 and Foxp3 transcription in lung tissue as well as systemic by augmenting the mean fluorescence intensity of Foxp3 in blood CD4+ T cells. Conclusion These findings show that Bifidobacterium breve M-16 V and Lactobacillus rhamnosus NutRes1 have strong anti-inflammatory properties that are comparable to budesonide and therefore may be beneficial in the treatment of chronic asthma.


Background
Allergic asthma is a T helper type-2 (Th2) cell-mediated chronic inflammatory disorder of the airways with rapidly increasing incidence and prevalence throughout the world, especially in children and within developing countries [1]. It is characterised by airway inflammation and hyper-responsiveness (AHR) [2]. Structural changes in the airway walls of asthmatic patients, referred to as "airway remodelling", are caused by persistent inflammation and subsequent inadequate repair of damaged airway epithelium [3][4][5].
Various cells are involved in cellular airway inflammation and subsequent airway remodelling, such as eosinophils, neutrophils, T lymphocytes and mast cells [4]. Infiltration of mast cells into the airway smooth muscle cell layer of allergic asthmatics is a key feature of asthma and is thought to be associated with AHR [6,7]. Upon mast cell degranulation various inflammatory mediators, such as vasoactive amines, cytokines and proteases are released [8,9]. Mouse mast cell protease 1 (mMCP-1) was reported to enhance airway narrowing in mice through effects on the epithelium [10,11].
Glucocorticoids (GCs) are by far the most effective treatments for asthma [12]. Besides the undesirable side effects of long-term GC therapy, a significant number of asthmatics is steroid resistant and fails to respond to this therapy [13,14]. These limitations of GCs therapy highlight the need for novel therapeutics with long-term benefits, greater disease control and increased efficacy.
Changes in the microbiota were reported to contribute to the development of allergies and asthma [15,16]. The potential role of beneficial bacteria, such as probiotics, as modulators of the intestinal microbiota and mucosal immune responses has been extensively investigated and discussed in the last few years [17][18][19]. Probiotics are "live microorganisms which, when consumed in adequate amounts, confer a health benefit on the host" [20,21]. Bifidobacteria and Lactobacilli, which are a part of the gut microbiota, were effective in suppressing both allergic and autoimmune responses, reducing allergic symptoms and inhibiting allergic airway response in murine models of acute airway inflammation [22][23][24][25][26].
Because they contain toll-like receptor (TLR) ligands, it is believed that beneficial bacteria can modulate TLRdriven responses and also skew the immune balance towards a Th1-associated response [27]. The TLRs and nod-like receptors (NLRs) are key pattern recognition receptor (PRR) families in the innate immune response. In the human airways, the RRRs are expressed in or on dendritic cells (DCs), epithelial cells, eosinophils, macrophages and mast cells [28,29]. Following TLR and NLR activation in the lung, various chemokines and cytokines are produced by eosinophils and mast cells that attract activated B-lymphocytes and Th lymphocytes to the lung to orchestrate the inflammation in the airways [30].
The function and expression of PRRs were linked to susceptibility towards allergic asthma [31,32]. Functional genetic variations in TLR1, TLR10 and TLR6 genes affecting gene and protein expression were associated with increased mRNA expression of these TLRs and protected against atopic asthma in humans [32]. Genetic variations in TLR2, NOD1 and NOD2 genes that led to either decreased mRNA expression and affected microbial recognition, respectively, were positively associated with disease susceptibility and pathogenesis [25,[33][34][35][36]. Cord blood CD34 (+) cells from high-atopic-risk infants have been reported to have low TLR2, TLR4, and TLR9 expression and the latter was demonstrated to exert protective immunomodulatory effects on asthma [37][38][39]. Tlr3 contributes to asthma exacerbations in mice [40] and a study in a murine macrophage cell line suggested a pro-inflammatory role of Tlr4 and 5 in the disease [41].
This study aimed to explore the therapeutic effects of long-term administration of Bifidobacterium breve M-16 V and Lactobacillus rhamnosus NutRes1 on chronic airway inflammation and remodelling in mice. A glucocorticoid was used as a reference treatment. Findings from this study will contribute to a better understanding of the immunomodulatory and therapeutic effects of beneficial bacteria in chronic allergic asthma.

Animals
Male BALB/c mice (6-8 weeks; Charles River Laboratories, France) were acclimated to their new environment for at least 1 week before the start of the experiment. Mice were housed under standard conditions and had free access to food and water. All in vivo experiments were approved by and were in accordance with the guidelines of the local Dutch Committee of Animal Experimentation.

OVA challenge
A chronic model of asthma was established according to a modification of a model of prolonged allergen-induced airway inflammation described in [2]. Mice were exposed daily to 5% OVA aerosol in saline or saline only using a Pari LC Star nebuliser (PARI GmbH, Starnberg, Germany) in an aerosol cabin for 20 min between days 17 and 23. Control animals were exposed to nebulised saline aerosol only. From day 24 until day 55, the frequency of challenge was reduced to three times a week and mice were exposed to aerosolised OVA (5%) or saline only for 20 min (Figure 1).

Beneficial bacteria treatment
Bifidobacterium breve M-16 V (B. breve, Morinaga Milk Industry, Tokyo, Japan) and Lactobacillus rhamnosus NutRes1 (L. rhamnosus, Danone Research, Wageningen, the Netherlands) were grown in MRS (Oxoid, Basingstoke, UK), supplemented with 0.5 g/L L-cysteine for Bifidobacteria, at pH 6.5 and under anaerobic conditions. Bacteria were harvested in the early stationary phase, washed with phosphate buffered saline (PBS, Lonza Leusden, The Netherlands) and stored with glycerol 20% (w/v), in aliquots at −80°C. Cell counts were determined by plating serial dilutions and fluorescent microscopy by staining with DAPI. The bacteria were resuspended in PBS prior to use.
After development of airway inflammation, mice received 10 9 colony forming units (CFUs) of B. breve or 1.1 × 10 9 CFUs of L. rhamnosus per animal per day. Bacterial strains were suspended in 0.2 mL of PBS and given by oral gavage, 1 h prior to challenge, three times a week from day 22 until day 55 ( Figure 1).

Budesonide treatment
As a reference treatment, mice received 0.5 μg/g of mouse/day of budesonide (Sigma) in PBS. Budesonide was administered to mice by oropharyngeal aspiration after induction of light isoflurane anesthesia as described previously in [47], 1 h prior challenge, three times a week from day 22 until day 55. Control animals received 50 uL of PBS by the same administration route. Mice were rendered asthmatic following the schema presented in Figure 1.

Airway response to methacholine
On day 56, 24 h after the last OVA aerosol challenge, the airway response to increasing doses of methacholine was analysed after insertion of a cannula in the trachea. Lung resistance in anesthetized, mechanically ventilated mice was measured directly using whole-body plethysmography (Emka technologies, Paris, France). Mice were exposed to saline (0 mg/mL methacholine) and increasing doses (0.38 to 25 mg/mL) of aerosolised methacholine (Sigma). Figure 1 Time schedule of the chronic asthma mouse model. Male BALB/c mice were sensitised intraperitoneally to OVA-ImjectAlum on days 0 and 12 and mice were challenged from day 17 until day 23 daily with aerosolised OVA or saline. From day 22 until day 55, mice were treated 3 times a week with either PBS or budesonide by oropharyngeal aspiration or beneficial bacteria (B. breve or L. rhamnosus) by oral gavage. 1 h after treatment, from day 24 until day 55, mice were challenged 3 times/week with aerosolised OVA or saline. Mice were sacrificed on day 56 after pulmonary function measurement.

Bronchoalveolar lavage
After sacrifice, on day 56, lungs were first lavaged through a tracheal cannula with 1 mL saline containing protease inhibitor cocktail (Complete Mini, Roche Diagnostics, Mannheim, Germany), pre-warmed at 37°C. This was followed by 3 additional lavages with 1 mL saline only. Cytospin cell preparations were made by cytospinning the cells onto glass for 5 min (400 g, 4°C) and cytospins were stained by DiffQuick (Merz & Dade AG, Düdingen, Switzerland). Numbers of eosinophils, macrophages, neutrophils and lymphocytes were scored by light microscopy.

RNA isolation and quantitative real-time PCR
After mice were sacrificed on day 56, the lungs were dissected and mRNA was isolated from whole lung tissue. Messenger RNA isolation (n = 6 mice per group) was carried out according to the Qiagen RNeasy Mini Kit protocol (Qiagen Benelux B.V., Venlo, The Netherlands). Reverse transcriptase PCR was performed using an iScript™cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, CA, USA). The reactions were performed in a PTC-100TM Programmable Thermal Controller (M. J. Research Inc., Waltham, Massachusetts, USA) according to manufacturer's protocol.
cDNA was amplified using iQ SYBR Green supermix in a 96-well PCR plate and run in a CFX96 Real-Time PCR Detection System (Bio-Rad). Primers for TLRs, NLRs, ribosomal protein S13 (RPS13, reference gene) and T cell transcription factors were purchased by Isogen (Isogen Life Science, De Meern, The Netherlands). The sequences are listed in Additional file 1: Table S1. For mouse T cell cytokines, RT 2 qPCR Primer Assays (SABiosciences, Venlo, The Netherlands) were used. The protocol used for amplification was 94°C for 3 min, 94°C for 10 sec, specific melt temperature for 45 sec, followed by 39 cycles of 94°C for 10 sec and 95°C for 10 sec.
Normalised gene expression (ΔΔC T ) was calculated using the built-in gene expression analysis module in CFX Manager™ software (CFX Manager™ software version 1.6).

Foxp3 staining and flow cytometry
On days 0 and 56, blood samples were collected from mice by cardiac puncture in tubes containing lithium heparin to prevent coagulation. The blood was then washed in PBS and, after centrifugation; the pellet was subjected to red cell lysis using a buffer containing NH 4 Cl (MERCK, Darmstadt Germany), KHCO 3 (Sigma), ethylenediaminetetraacetic acid (EDTA, MERCK) in demineralised water for 5 min on ice. After several washes with PBA (PBS containing 1% bovine serum albumin (BSA, Roche Diagnostics, Almere, The Netherlands) cells were resuspended in PBA and kept on ice until Foxp3 staining.
The expression of Foxp3 was measured using the Foxp3 Staining Buffer Set (eBioscience, San Diego, CA, USA) using the following protocol: cells were incubated in Fixation/Permeabilization buffer for 30 min on ice. Cells were then washed once with PBA followed by two washes with permeabilization buffer. After a 15 min preincubation in total mouse serum blocking reagent on ice, cells were washed once with permeabilization buffer and then stained with anti-CD4 (FITC, eBioscience), anti-CD25 (PE, eBioscience) and anti-Foxp3 (APC, eBioscience) for 30 min on ice. Cells were washed twice with permeabilization buffer and resuspended in PBA for flow cytometry analysis.
Tregs were defined as CD4 + Foxp3 + CD25hi T cells. The stained cells were analysed on a FACSCanto II flow cytometer (BD Biosciences, USA). Data analysis was performed using BD FACSDiva™ software (BD Biosciences). The gating strategy is illustrated in Additional file 2: Figure S1.

Measurement of mouse mast cell protease 1 levels in serum
To assess mast cell activation, after mice were sacrificed on day 56, blood samples were collected from mice by cardiac puncture. The blood was coagulated for 1 h at room temperature and subsequently centrifuged for 5 min at 17,500 g. Serum samples were stored at −80°C until further analysis. Mouse mast cell protease 1 (mMCP-1) protein expression levels in serum were determined by enzyme-linked immunosorbent assay (ELISA) using the Mouse MCPT-1 (mMCP-1) ELISA Ready-SET-Go!® kit (eBioscience) according to manufacturer's protocol.

Measurement of cytokine production by T cells in thoracic lymph nodes after restimulation with anti-CD3 antibody in vitro
In order to examine specific T cell responses, after mice were sacrificed on day 56, lung-draining lymph nodes were collected from the thorax and transferred to cold sterile PBS. Single cell suspensions of the thoracic lymph nodes (TLNs) were made using a 70 μm nylon cell strainer (BD Biosciences) and rinsed with 15 mL of PBS. The cells were washed and resuspended in RPMI 1640 culture medium without L-glutamine and phenol red (Lonza) supplemented with 10% heat-inactivated fetal calf serum (FCS, Hyclone Laboratories, USA) and 0.1% penicillinstreptomycin solution (pen-strep, Sigma). The total number of cells was determined using a Beckman Z1 coulter® Particle Counter (Beckman, USA). TLN cells (4 × 10 6 cells/mL) were cultured in a Greiner bio-one CellSTAR 96-well U-bottom plate (Greiner Bio-One B.V., Alphen a/d Rijn, The Netherlands) in medium with or without O/N pre-coating of the wells with 50 μg/mL of anti-CD3 antibody (Bioceros BV, Utrecht, The Netherlands). The supernatant was harvested after 5 days of culture at 37°C in 5% CO2 and stored at −20°C until further analysis.
The levels of cytokines in the supernatant were measured by flow cytometry using a BD™ Cytometric Bead Array (CBA) Mouse Th1/Th2/Th17 Cytokine kit (BD Biosciences) according to manufacturer's protocol on a FACSCanto II flow cytometer (BD Biosciences). Data analysis was performed using BD FCAP Array™ v3.0.1 Software (BD Bisosciences).

Histology and immunohistochemistry
After mice were sacrificed on day 56, lungs were fixed with 10% formalin infusion through a tracheal cannula at a constant pressure of 25 cm H 2 O. The lungs were immersed in fixative for at least 24 h, after which the left lung was embedded in paraffin. After paraffin embedding, 5 μm sections were cut and de paraffin sections were first deparaffinised. Endogenous peroxidase activity was blocked with 0.3% H 2 O 2 (Merck) in methanol for 30 min at room temperature and rehydrated in a graded ethanol series to PBS and paraffin section were stained with hematoxylin/eosin (H&E) for inflammation, periodic acidschiff (PAS) for goblet cells, Masson's trichrome for connective tissue, rabbit polyclonal anti-α-smooth muscle actin antibody (Abcam, Cambridge, UK) for smooth muscle cells and rabbit polyclonal anti-Ki76 antibody (Abcam) for proliferating cells according to standard methods. Photomicrographs were taken with an Olympus BX50 microscope equipped with a Leica DFC 320 digital Camera.
Slides were reviewed in blinded fashion by two observers independently and slides were scored on the basis of the percentage of positive stained cells in the following way: −, no positive staining; +/−, less than 25% of cells stained positive; +, 25 to 50% cells stained positive; ++, 50 to 75% cells stained positive.

Statistical analysis
Data analysis was performed using a 1-way analysis of variance (one-way ANOVA) with the Bonferroni's posthoc test. Linear regression analysis was used to calculate correlations. All statistical analyses were performed using GraphPad Prism software program (GraphPad Prism software version 5.03).

L. rhamnosus and budesonide reduce the increased lung resistance in chronically asthmatic mice
To investigate the lung function in the chronic asthmatic mice after treatments, lung resistance was measured (Table 1). OVA-sensitised and challenged control mice (OVA/OVA-PBS) showed a significant increase in the basal lung resistance and lung resistance at all concentrations of methacholine as compared to the sensitised only control group (OVA/Sal-PBS). Treatment of sensitised and challenged mice with L. rhamnosus (OVA/OVA-L. rhamnosus) resulted in a significant decrease in the basal lung resistance and lung resistance at 0, 0.75 and 1.56 mg/mL methacholine as compared to the OVA/ OVA-PBS group. Budesonide treatment significantly decreased the basal lung resistance and lung resistance at 0, 0.38, 0.75 and 1.56 mg/mL methacholine in allergic mice (OVA/OVA-BUD) as compared to OVA/OVA-PBS mice. The basal lung resistance and lung resistance at all concentrations of methacholine remained unchanged in B. breve-treated, sensitised and challenged mice (OVA-OVA-B. breve) as compared to the OVA/OVA-PBS group.
B. breve and L. rhamnosus are as effective as budesonide in reducing pulmonary inflammation in chronically asthmatic mice To gauge the extent of inflammation in the asthmatic mice after treatments, bronchoalveolar lavage (BAL) fluid was examined for leukocyte accumulation ( Figure 2). OVA/OVA-PBS mice showed a significant increase in the total inflammatory cell number (155.25 × 10 4 ± 12.58; n = 6; p < 0.0001) in the BAL fluid, which was due to a relative increase in the number of macrophages, eosinophils and neutrophils as compared to OVA/Sal-PBS mice (29.25 × 10 4 ± 3.12; n = 6). All treatments significantly reduced the total number of inflammatory cells in the BAL fluid of OVA/OVA mice. Budesonide treatment significantly decreased the total inflammatory cell number (83.70 × 10 4 ± 13.66; n = 6; p < 0.0001) and relative number of eosinophils and neutrophils in OVA/OVA-BUD mice. B. breve treatment, however, further decreased the total inflammatory cell number (73.50 × 10 4 ± 6.42, n = 6; p < 0.0001) and significantly decreased the relative number of eosinophils and neutrophils. L. rhamnosus treatment significantly decreased the total inflammatory cell number (79.75 × 10 4 ± 17.49; n = 6; p < 0.0001) and relative number of neutrophils.
B. breve and L. rhamnosus modulate TLR and NLR mRNA expression in lung tissue of chronically asthmatic mice As PRRs in the lung can modulate ongoing chronic inflammation during asthma, the mRNA expression of Tlr1-9 and Nod1-2 were measured ( Figure 3). The mRNA expression of Tlr3 and Nod1 was significantly decreased and Tlr9 was also decreased (p > 0.05) in OVA/OVA-PBS mice as compared to the OVA/Sal-PBS group ( Figure 3A). After B. breve treatment, the OVA/OVA-B. breve mice showed a significant increase in Tlr9 expression compared to the OVA/OVA-PBS group. In non-asthmatic controls, Tlr1 and Tlr2 expression was significantly increased in the OVA-Sal-B. breve group as compared to OVA/Sal-PBS mice, yet, the expression of the other PRRs in non-asthmatic controls remained unchanged. B.breve treatment in asthmatic mice significantly increased Tlr9 expression and decreased Tlr2, Tlr3, Tlr5 and Nod1 expression in the OVA/OVA-B. breve group as compared to the OVA-Sal-B. breve group ( Figure 3B). Mice in the OVA/OVAbudesonide group showed significant increases in the expression of Tlr3 and Nod1 and Tlr9 expression was higher (p > 0.05) as compared to the OVA/OVA-PBS group. The expression of Tlr3 and Tlr5 was significantly decreased in OVA/Sal-budesonide mice as compared to the OVA/Sal-PBS group ( Figure 3C). The OVA/OVA-L. rhamnosus group showed a significant decrease in Tlr4, an increase in Tlr3 and higher Tlr9 expression compared to OVA/OVA-PBS mice. The mRNA expression of Tlr1, Tlr2, Tlr4, Tlr6, Nod1 and Nod2 was significantly decreased in OVA/OVA-L. rhamnosus mice as compared to the OVA/Sal-L. rhamnosus group ( Figure 3D).
Beneficial bacteria and budesonide treatment modulate cytokine mRNA expression in lung tissue of chronically asthmatic mice To determine the extent of inflammation and the Th response in the lung, the mRNA expression of various cytokines was measured ( Figure 4). The mRNA expression of Il1β and Il6 was significantly increased and Il13 and Il17    B. breve skews the immune response away from Th2 and towards Treg in the lungs of chronically asthmatic mice To determine the extent of Th response skewing in the lung, ratios for Gata3/Tbet (Th2/Th1), Foxp3/ Rorγt (Treg/Th17), Foxp3/Gata3 (Treg/Th2) and Foxp3/Tbet (Treg/Th1) mRNA expression were calculated ( Table 2). The OVA/OVA-PBS groups showed a Th2-skewed immune response represented by a significant increase in Gata3/Tbet ratio as compared to OVA/Sal-PBS mice. This ratio was significantly decreased in the OVA/OVA-L. rhamnosus and OVA/ OVA-B. breve groups, but not the OVA/OVA-budesonide group. The Th2-reducing effect of OVA-OVA-B. breve was even more effective in challenged mice than healthy mice as the Gata3/Tbet ratio was decreased in OVA/OVA-B. breve mice as compared to the OVA/Sal-B. breve group. This was not the case for budesonide. Compared to the OVA/Sal-budesonide group, there was a significant increase in the Gata3/ Tbet ratio in the OVA/OVA-budesonide group. Interestingly, the Foxp3/Rorγt and Foxp3/Gata3 ratios were significantly increased in OVA/OVA-B. breve mice as compared to the OVA/OVA-PBS group indicating an increase in Treg-associated responses. The ratio of Foxp3/Tbet did not differ significantly among the different treatment groups. Statistical significance of differences was tested using the post hoc Bonferroni's multiple comparison test after one-way ANOVA. # Statistically significant difference (p < 0.05) compared to the OVA/Sal-PBS group. ¤ Statistically significant difference (p < 0.05) compared to the OVA/OVA-PBS group. ‡ Statistically significant difference (p < 0.05) compared to the OVA/Sal group in each treatment group.

Beneficial bacteria and budesonide increase Foxp3 expression in blood Treg cells of chronically asthmatic mice
In order to determine if the transcriptional changes induced in Foxp3 expression could be found at a protein expression level, flow cytometry was employed to examine Foxp3 expression on Treg cells in the blood ( Figure 6). Treg cells were identified as CD4+ Foxp3 + CD25hi T cells and the magnitude of Foxp3 expression on each cell was determined by examining the mean fluorescence intensity (MFI) of the Foxp3 staining. The percentage of CD4 cells that are Foxp3+ did not differ among the different groups ( Figure 6A). The percentage of CD4 cells that are Foxp3+ and CD25hi was significantly decreased by all treatments as compared to both OVA/Sal-PBS and OVA/OVA-PBS groups ( Figure 6B). However, all treatments increased the  Figure 6C and D). The staining intensity of Foxp3 on day 0 (before sensitisation, challenge and treatment) did not differ among the different groups (data not shown).
B. breve and L. rhamnosus are as effective as budesonide in suppressing mucosal mast cell degranulation in chronically asthmatic mice To examine the effect of the different treatments on mast cells activity, enzyme-linked immunosorbent assay was employed to measure mMCP-1 protein levels in serum (Figure 7). The expression of mMCP-1 was significantly increased in OVA/OVA-PBS mice as compared to the OVA/Sal-PBS group. All treatments were effective in suppressing mucosal mast cell degranulation in OVA/OVA mice, as depicted by decreased protein levels of mMCP-1 in serum when compared to the OVA/OVA-PBS group. mMCP-1 levels in control animals (OVA/Sal) did not change after administration of beneficial bacteria (data not shown).

Beneficial bacteria and budesonide reduce T cell activity in chronically asthmatic mice
To investigate the effect of the treatments on cytokine production by T cells in lung draining lymph nodes, TLNcell cultures were stimulated with anti-CD3, a pan T cell stimulator, and the cytokine levels in the supernatant were measured (Figure 8) B. breve and L. rhamnosus are as effective as budesonide in suppressing airway remodelling features in chronically asthmatic mice In order to examine the effect of the different treatments on airway remodelling features, semi-quantivitive histological and immunhistochemical analyses of lung tissue were performed (Figure 9 and Table 3). In the lung sections of OVA/OVA-PBS mice, increased inflammation score and number of goblet cells, collagenous connective tissue fibers, airway smooth muscle cells and proliferating cells was observed as compared to the OVA/Sal-PBS group. All treatments were effective in reducing the inflammation score and decreasing the number of collagenous connective tissue fibers and proliferating cells in sensitised and challenged mice as compared to the OVA/OVA-PBS group. Yet, the number of airway smooth muscle cells remained unchanged in the different treatment groups. Budesonide treatment was also effective in reducing the number of goblet cells in the

Discussion
The aim of this study was to investigate the therapeutic effects of two different beneficial bacterial strains (B. breve M-16 V and L. rhamnosus NutRes1) and a reference treatment (budesonide) on features of the inflammatory response and airway remodelling in a murine model for chronic asthma. The lung resistance was significantly increased in OVA/OVA-PBS mice at all concentrations of methacholine as compared to the OVA/Sal-PBS group. Additionally, the total inflammatory cell number and individual BAL fluid cell counts, except lymphocytes, were significantly increased in OVA/OVA-PBS mice. Increased inflammation score and number of goblet cells, collagenous connective tissue fibers, airway smooth muscle cells and proliferating cells was observed in lung sections of these mice. Hence, our model mimics the airway inflammation and airway remodelling in chronic asthma [2,48,49]. Treatment with L. rhamnosus or B. breve, hardly influenced the pathophysiological parameters in OVA/Sal-PBS animals. Interestingly, L. rhamnosus and budesonide treatments were effective in suppressing the lung resistance in OVA/OVA mice, significantly decreasing the AHR to methacholine as compared to the OVA/OVA-PBS group. B. breve treatment did not affect the AHR to methacholine. Airway inflammation is a characteristic feature of allergic asthma; however, the degree of airway inflammation is not always correlated with the degree of AHR [50]. In the present study, we demonstrate that L. rhamnosus prevents the ovalbumin-induced increase in basal airway resistance and AHR at lower concentrations of methacholine. In contrast to L. rhamnosus, B. breve was not able to significantly reduce changes in airway resistance. Although all the treatments were able to significantly reduce airway inflammation by 50%, there was no complete reduction up till control levels as shown in the BAL fluid as well as in the histological evaluations. These remaining cells might be responsible for the AHR at higher concentrations of methacholine. There is no clear explanation why L. rhamnosus did suppress the increase in basal airway resistance and B. breve did not. Shifts in cell type such as the reduction in the relative number of macrophages or suppression of specific cytokines such as IFN-γ might explain these observations. The above described results are in accordance with other studies in which beneficial bacteria and budesonide have been individually investigated [1,13,23,24,[51][52][53]. Additionally, B. breve, L. rhamnosus and budesonide treatments reduced the inflammation score and decreased the number of collagenous connective tissue fibers and proliferating cells in ovalbumin-sensitised and challenged mice as compared to the OVA/OVA-PBS group. Budesonide treatment, but not B. breve or L. rhamnosus, was also effective in reducing the number of goblet cells in the lung sections of OVA/ OVA-budesonide mice. The number of airway smooth muscle cells remained unchanged in the different treatment groups as compared to OVA/OVA-mice. To our knowledge, we demonstrate here for the first time that L. rhamnosus NutRes1 is as effective as budesonide in suppressing lung resistance in chronic allergic mice. Additionally, we demonstrate here that both beneficial bacterial strains are as effective as budesonide in reducing chronic allergic inflammation by attenuating the total inflammatory cell number as well as individual BAL fluid cell counts; and airway remodelling.
PRRs are key components of the innate immunity which are also involved in the activation and shaping of adaptive immunity. The function and expression of PRRs was linked to susceptibility towards allergic asthma. The mRNA expression of Tlr3 and Nod1 was significantly decreased in the lungs of OVA/OVA-PBS mice. Previous in vitro studies have demonstrated that upon activation Ratios for Gata3/Tbet (Th2/Th1), Foxp3/Rorγt (Treg/Th17), Foxp3/Gata3 (Treg/ Th2) and Foxp3/Tbet (Treg/Th1) mRNA expression in whole lung tissue are shown for OVA/Sal and OVA/OVA mice for each treatment. Data is represented as mean ratio ± SEM, n = 6 mice/group. The mean ratio was calculated by dividing the individual expression values for the first transcription factor (numerator) by the mean expression value for the second transcription factor (denominator). Statistical significance of differences was tested using post hoc Bonferroni's multiple comparison test after one-way ANOVA. # Statistically significant difference (p < 0.05) compared to the OVA/Sal-PBS group. ¤ Statistically significant difference (p < 0.05) compared to the OVA/OVA-PBS group. ‡ Statistically significant difference (p < 0.05) compared to OVA/Sal mice in each treatment group.
by its natural or endogenous ligands, Tlr3 induces upregulation of other TLRs, various cytokines and chemokines as well as its own expression and thereby contributes to exacerbation of inflammation [40]. Nod1 is an intracellular sensor of pathogenic bacteria. Single nucleotide polymorphisms in Nod1 gene were positively associated with susceptibility towards asthma in farming children and this PRR was reported to be necessary for neutrophil function in mice [25,[33][34][35]. However, no direct associations between Tlr3 and Nod1 expression and function and asthma have been reported yet, and whether the decreased mRNA expression of Tlr3 and Nod1 caused by the chronic inflammatory status of the animals is proinflammatory or anti-inflammatory is also unknown. Moreover, B. breve treatment significantly increased Tlr9 expression. Since TLR9 was shown to exert its immunomodulatory effects on asthma by skewing the increased Th2/Th1 balance towards Th1, this may very well be the situation [37][38][39]. Budesonide treatment significantly increased the expression of Tlr3 and Nod1, restoring them to normal levels, and raised Tlr9 (P > 0.05) expression. This further supports a positive role for Tlr9 and suggests that lowered Tlr3 and Nod1 measured in the asthmatic mice is pro-inflammatory in nature [25,34,40]. L. rhamnosus, besides significantly increasing Tlr3 expression and increasing the expression of Tlr9 expression (P > 0.05), significantly decreased the expression of Tlr4. Asthma patients have been shown to have low expression of TLR4 on their monocytes, lymphocytes and DCs suggesting that this reduction in TLR4 activation might contribute to the disease by reducing the release of Th1 and antiinflammatory cytokines [54]. These findings might suggest that L. rhamnosus treatment in asthmatic mice could not restore the expression of Tlr4 to its normal level. Taken together, we demonstrate here that the mRNA expression of PRRs in mouse lung tissue is differentially regulated by the different treatments. Yet, all treatments, especially B. breve, increased the expression of Tlr9.
Th2 cells play a key role in the pathogenesis of allergic asthma, and asthmatic patients were reported to have Th1/Th2 imbalances as well as disturbed Th17/Treg balances. Th2 dominance was observed in the OVA/ OVA-PBS group represented by a significant decrease in Th1 and Treg transcription factors and high Gata3/Tbet ratio. Hence, our model mimics the Th2-responses found in chronic asthma [1,2]. Importantly, B. breve shifted the immune balance towards Th1 and Treg, with significantly increased Foxp3/Rorγt and Foxp3/Gata3 ratios and a significantly decreased Gata3/Tbet ratio. A different strain of Bifidobacterium, B. animalis, has been already shown to skew the Th1/Th2 balance towards Th1 in a preventative, acute mouse model for respiratory allergy [22]. Budesonide showed only a moderate effect and significantly increased the expression of Foxp3, but did not affect the Gata3/Tbet, Foxp3/Rorγt, Foxp3/ Gata3 and Foxp3/Tbet ratios. These findings are Figure 8 Beneficial bacteria and budesonide suppress cytokine production by T cells in TLNs of chronic asthmatic mice. Thoracic lymph nodes (TLNs) were isolated from mice on day 56 and restimulated in vitro with plate-bound anti-CD3 monoclonal antibody or medium only for 5 days (37°C, 5% CO 2 ). Data is shown as mean of cytokine concentration (ng/ml) ± SEM, n = 6 mice/group. Statistical significance of differences was tested using post hoc Bonferroni's multiple comparison test after one-way ANOVA. # Statistically significant difference (p < 0.05) compared to the OVA/Sal-PBS group. ¤ Statistically significant difference (p < 0.05) compared to the OVA/OVA-PBS group. consistent with results of other studies which have shown that GCs treatment of asthmatic subjects encourages regulatory responses [55]. L. rhamnosus did not influence the mRNA expression of the different Thspecific transcription factors, but significantly decreased the Gata3/Tbet ratio. More importantly, the above observed effects of beneficial bacteria and budesonide on the Th responses in the lung are mirrored by the detection of high Foxp3-expressing Tregs in the blood of treated animals. Tregs play a key role in balancing immune responses and it was demonstrated that an increased expression of Foxp3 in Tregs is directly associated with increased function of these cells [42,56]. No significant differences in the percentage of CD4+ cells that are Foxp3+ were observed. A reduction in the populations of recently activated regulatory T (CD4 + Foxp3 + CD25hi) cells was observed in the treated mice. One possible reason for the loss of these cells may pertain to a reduction in inflammation caused by the treatments. Previous studies have demonstrated that Tregs can   become activated through inflammatory signals in the environment which functions as a natural feedback loop by preventing excessive inflammation. An example of this is through TLR2 stimulation [57]. Expression of the pro-inflammatory cytokines Il1β and Il6 was significantly increased in the OVA/OVA-PBS group and expression of both Il13 and Il17 was also increased but not significantly. These four cytokines were reported to enhance asthma [30]. B. breve treatment, however, significantly increased the expression of the anti-inflammatory cytokine Il10 and the Th2-associated cytokine Il4, yet, the mRNA expression of Gata3 was decreased. Interestingly, the expression of Tbet was tightly correlated with Ifnγ and Il12, while Foxp3 expression was tightly correlated with Il10 expression. It has been reported that IL-12 is involved in the differentiation of Th1 cells and that these cells suppress Th2 cells through the release of IFN-γ. Additionally, Tregs suppress other Th cell effector functions through the release of IL-10 [30]. Budesonide treatment significantly decreased the expression of the pro-inflammatory cytokine Tnfα which is in agreement with previously published studies demonstrating the anti-inflammatory effects of budesonide [58]. L. rhamnosus treatment significantly decreased the expression of Il6, another highly pro-inflammatory cytokine. The mRNA expression of Foxp3 was tightly correlated with Il10 expression.
Mast cells are key effector cells in allergic inflammation and mast cell degranulation is detected in asthmatic lungs. The expression of serum mMCP-1 was significantly increased in OVA/OVA-PBS mice as compared to the OVA/Sal-PBS group indicating pulmonary mast cell degranulation. Hence, previous studies demonstrated that mice undergoing a hypersensitivity reaction in the airways have high serum levels of mMCP-1 of pulmonary origin [59]. All treatments were effective in inhibiting mucosal mast cell degranulation, as shown by the significant decrease in the protein levels of mMCP-1 systemically, in serum, in treated mice as compared to the OVA/OVA-PBS group. The mechanism by which these beneficial bacteria reduced mast cells degranulation still needs to be investigated; yet, these effects might be galectinesmediated. Recently, a combination of Bifidobacterium breve M-16 V with galacto-and fructo-oligosaccharides was shown to induce galectin-9 release from intestinal epithelial cells which in turn systemically suppressed allergic symptoms, including mast cells degranulation, in mice and humans. The same study demonstrated a negative correlation between increased serum galectin-9 levels and serum mMCP-1 levels [60]. The mechanism by which budesonide affects mast cell degranulation still needs further investigation; yet, glucocorticoids were demonstrated to suppress mast cell activation in vitro through the upregulation of inhibitory regulators of these cells [61]. In addition to their effects on mast cells, B. breve, L. rhamnosus and budesonide were also effective at reducing cytokine production by T cells in the TLNs. The levels of IL-2, IL-4, IL-6, IL-17a, TNF-α and IFN-γ were significantly increased in anti-CD3-stimulated TLN cells from OVA/ OVA-PBS mice as compared to the OVA/Sal-PBS group. IL-2, IL-4, IL-6, IL-17a and TNF-α were reported to be involved in asthma. Increased levels of IFN-γ are found in individuals with severe asthma and acute exacerbation [30,62]. The levels of IL-2, IL-4, IL-6, IL-17a and TNF-α were significantly decreased in OVA/OVA mice treated with B. breve, L. rhamnosus or budesonide as compared to the OVA/OVA-PBS group. Levels of IFN-γ were significantly decreased in OVA/OVA-L. rhamnosus and OVA/ OVA-budesonide mice. The level of IL-10 did not differ among the different groups, suggesting that the different treatments could not normalise IL-10 production in chronic allergic mice.

Conclusions
To our knowledge, this is the first report in which the therapeutic effects of long-term treatment with B. breve M-16 V, L. rhamnosus NutRes1 and budesonide on asthma are investigated in a murine ovalbumin-induced chronic allergic asthma model. In this current study we show that B. breve M-16 V and L. rhamnosus NutRes1 are as effective as budesonide in suppressing pulmonary airway inflammation, airway remodelling and inhibiting mast cell degranulation. Additionally, L. rhamnosus, but not B. breve, reduced lung resistance and IFN-γ production by T cells in the TLNs indicating that treatment effects differ between the bacterial strains. mRNA expression of PRRs, T helper-specific cytokines and transcription factors are differentially modulated by the different treatments. Moreover, B. breve induced regulatory responses by increasing Il10 and Foxp3 transcription in lung tissue and augmenting the mean fluorescence intensity of Foxp3 in blood CD4+ T cells. These findings show that B. breve M-16 V and L. rhamnosus NutRes1 may be beneficial in the management of chronic allergic asthma in a therapeutic way.

Additional files
Additional file 1: Table S1. Primers used for quantitative real-time PCR analysis.