Pre- and postnatal administration of Lactobacillus reuteri decreases TLR2 responses in infants

Sixty-one children, of whom 29 received probiotics and 32 placebo (Table 1), were selected based on availability of blood cell samples at multiple time points from a double-blind, randomized, placebo-controlled probiotic trial, in which 188 infants completed the study, ClinicalTrials.gov, NCT01285830 [13]. The infants included in this study had cells collected from at least three time points, i.e. birth, 6, 12 and 24 months. They were not statistically significantly different from the main trial with regards to the variables presented in Table 1. They all had a family history of allergic disease, i.e. at least one family member had eczema, asthma, gastrointestinal allergy, allergic urticaria, and/or allergic rhinoconjunctivitis. The pregnant women either received Lactobacillus reuteri ATCC 55730 or placebo daily from week 36 until delivery, and their infants continued from day 1–3 with the same product daily until 12 months of age. An informed consent was obtained from both parents before inclusion.

The Lactobacillus preparation consisted of freeze-dried L. reuteri suspended in coconut oil and peanut oil containing cryoprotective components, while the placebo preparation only comprised the vehicle. The daily intake corresponded to 1 × 10⁸ colony forming units (CFUs).

The children were monitored by research nurses and a final follow up was done by a pediatrician at 2 years of age, i.e. one year after the termination of treatment. Skin prick tests (SPTs) were done at 6, 12 and 24 months of age. Potentially allergic disease included eczema, recurrent wheeze, rhinoconjunctitivis, urticaria and gastrointestinal allergy. Eczema was classified as a pruritic, chronic, or chronically relapsing non-infectious dermatitis with typical features and distribution. Eczema was classified as IgE-associated if the infant was also sensitized, i.e. had at least one positive SPT and/or detectable circulating IgE antibodies to allergens. Wheeze was defined as an episode with obstructive airway symptoms, and recurrent wheeze was defined as three or more wheezing episodes, at least once verified by a physician. Allergy was defined as having at least one positive SPT and/or detectable circulating IgE (>0.35 kU/L), in addition to clinical symptoms consistent with eczema or recurrent wheeze. The eight children defined as allergic in this study all had IgE-associated eczema (n = 3 L. reuteri/n = 5 placebo). One child also had recurrent wheeze, one child allergic urticaria and one gastrointestinal allergy, while none of the children had allergic rhinoconjunctivitis. Non-sensitized children with symptoms (n = 8) and sensitized children without symptoms (n = 14) were not included in the analysis comparing allergic and non-allergic children. In the main trial, the prevalence of IgE-associated eczema at two years was lower in the L reuteri than the placebo group (8% vs. 21%, p = 0.02) [13].

This study was approved by the Regional Ethics Committee for Human Research at Linköping University (Dnr 99323) and registered at ClinicalTrials.gov (NCT01285830).

Cord blood was collected at birth and venous blood samples were drawn at 6, 12 and 24 months into heparinized vacutainers. Peripheral blood mononuclear cells (PBMC) were collected by Ficoll gradient centrifugation. Briefly, blood was layered on a Ficoll gradient, centrifuged and the PBMC layer was collected with subsequently washing and centrifugation steps. Cells were resuspended in freezing media consisting of 40% IMDM, 10% DMSO and 50% FCS. The cells were then placed in a freezing container at −70°C for 24 hours and thereafter stored in liquid nitrogen, pending analysis [19].

PBMCs were thawed and stimulated with 1 ng/ml lipoteichoic acid (LTA) from S. aureus (InvivoGen, San Diego, USA), 1 ng/mL lipopolysaccharide (LPS) from E. coli K12 (Invivogen) or 1 μg/ml CpG ODN M362 (InvivoGen) in AIM V (Life Technologies AB, Täby, Sweden) with 20 μM mercaptoethanol (Sigma-Aldrich) at a concentration of 1 × 10⁶/ml and incubated for 24 hours at 37°C. Unfortunately, enough cells were not always available for all stimuli. Cell cultures were then centrifuged, supernatants were collected and cell pellets were lysed in RLT-buffer (Qiagen GmbH, Hilden, Germany) and stored at −70°C for later mRNA analysis. The mean viability of the freeze/thawed PBMCs was 85.2%.

Total RNA was extracted from the control/spontaneous RLT-lysates using the RNeasy 96 plate-kit (Qiagen GmbH), according to the manufacturer’s instructions. Real-time PCR was conducted, as described previously [10]. Briefly, 1 μL of cDNA was mixed with 19 μL TaqMan Fast Universal Mastermix (Applied Biosystems, Paisley, UK) together with primers and probe for TLR2 mRNA (HS00152932_m1), TLR4 mRNA (HS01060206_m1), TLR9 mRNA (HS00370913_s1) (Applied Biosystems). The sequences for 18S rRNA primers and probe (Eurogentec S.A, Seraing, Belgium) have been described elsewhere [20].

The levels of pro-inflammatory cytokines and chemokines (IL-1β, IL-6, IL12p70, IFN-a, TNF, CCL4 and CXCL8) and the anti-inflammatory cytokine IL-10 in the cell supernatants were analysed with multiplex assay kits, according to the manufacturer’s instructions (Bio-Rad Laboratories, Hercules, CA, USA). The samples were analysed on a Luminex¹⁰⁰ instrument (Biosource, Nivelles, Belgium) and the data was analysed with the software StarStation 2.0 (Applied Cytometry Systems, Sheffield, UK). The lower detection limit was 2,34 pg/mL for IL-1β, 33 pg/mL for IL-6, 1 pg/mL for IL-10, 0.5 pg/mL for IL12p70, 10 pg/mL for IFN-α, 20 pg/mL for CCL4 and CXCL8 and 3 pg/mL for TNF. Undetectable samples were given the value of half the cut off level. Comparisons of the TLR ligand induced cytokine and chemokine responses were made after the control value, i.e. responses from cells cultured in medium alone, was withdrawn. Detectable levels of IL-6, IL-10, CCL4, CXCL8 and TNF could be measured after all stimulations. After LPS and LTA stimulation, IL-1β levels could also be detected. Detectable IFN-α levels were observed only after stimulation with CpG.

As the data were not normally distributed, non-parametric tests were used. Comparisons between unpaired groups were analyzed with Mann–Whitney U-test, and correlations were analyzed with Spearman’s rank order correlation coefficient test. The student’s t-test and χ²-test was used to compare the background factors between the groups. Probiotic supplementation was included as dependent variable (placebo coded as 1, probiotic as 0) and IgE-associated allergic disease was included as independent variable in a logistic regression model, and the association between cytokine and chemokine secretion and these variables was investigated. P-values <0.05 were considered statistically significant. Calculations were performed with a SPSS statistical package version 19.0; SPSS Inc, Chicago, Ill. Undetectable samples were given the value of half the cutoff.

The levels of the chemokines CCL4 and CXCL8 were lower after stimulation with the TLR2 ligand LTA at 12 months in the probiotic than the placebo group (p = 0.014 and p = 0.043, respectively, Figure 1) and for CCL4 also at 24 months (p = 0.048, Figure 1). Furthermore, probiotic supplementation was associated with low levels of the pro-inflammatory cytokines IL-1β and IL-6 after LTA stimulation at 12 months (p = 0.002 and p = 0.017, respectively) and for IL-1β (p = 0.014) at 24 months. The cytokine and chemokine responses to LTA were statistically significantly lower at several time-points and showed similar trends at all ages in the probiotic compared with the placebo treated infants. Other measured cytokines showed no difference (Additional file 1: Table S1).

Although the probiotic treated infants showed the most clear and consistent differences in TLR2 responsiveness, probiotic supplementation was also associated with low CXCL8 responses to the TLR4 ligand LPS at birth (p = 0.040). No further differences were observed in responsiveness to TLR4 and TLR9 ligation (Additional file 1: Table S1).

Since exposure to a farm environment has been suggested to up-regulate TLR mRNA expression [16‐18], we investigated if probiotic supplementation affected the mRNA expression of TLR2/4 or 9. Neither the TLR2 nor the TLR9 expression differed between the two groups (Additional file 1: Table S1). The mRNA expression of TLR4 was significantly lower in the probiotic group than in the placebo group at 6 months of age, however (p = 0.039, Figure 2).

No major differences in TLR ligand induced cytokine and chemokine responses were found between allergic (n = 3 L. reuteri/n = 5 placebo) and non-allergic children (n = 13 L. reuteri/n = 16 placebo) (Additional file 2: Table S2). Allergic infants secreted higher LPS-induced IL-1β levels at 6 months than non-allergic infants (p = 0.042) and lower CpG-induced TNF levels at 24 months (p = 0.048), however. TLR2 mRNA expression was higher at 12 months in non-allergic than allergic children (p = 0.021, Figure 3).

Logistic regression analyses were performed to evaluate whether the relationship between cytokine responses was dependent on a parallel association with allergic disease. Probiotic supplementation and IgE-associated allergic disease were included in the model in order to control for potential confounding, whereas maternal or double atopic heredity were not, since these variables had no effect. Adjusting for allergy development had no major effect on the association between probiotic treatment and cytokine responses in this model (Additional file 3: Table S3).

The TLR4 mRNA levels correlated with LPS induced cytokine and chemokine levels at the corresponding age in cord blood (CXCL8, p = 0.022, rho = 0.40) and at 24 months (TNF, CCL4, CXCL8, IL-1β, p < 0.001-0.04, rho = 0.35–0.66). At 6 months TLR9 mRNA expression correlated with CpG induced IL-10, IFNα, TNF, CCL4 (p < 0.001–0.01, rho = 0.40–0.51). TLR2 mRNA expression did only correlate at one time point with LTA induced cytokine and chemokine expression, (IL-10, 6 months, p = 0.011, rho = 0.34).