Lactobacillus paracasei KBL382 administration attenuates atopic dermatitis by modulating immune response and gut microbiota

L. rhamnosusKBL365 and L. paracaseiKBL382 were isolated from the feces of healthy Koreans, as described in a previous study.⁴⁶ Briefly, KBL365 and KBL382 strains were cultured in Lactobacilli MRS Agar (BD Difco, Sparks, MD, USA) supplemented with 0.05% L-cysteine-hydrochloride at 37°C under anaerobic conditions maintained with an Anaeropack (Mitsubishi Gas Chemical Company, Inc., Tokyo, Japan) for 24 h. The cells were harvested through centrifugation (1,200 × g) and washed twice with phosphate-buffered saline (PBS) prior to administration to mice. Based on decreases in levels of Th1, Th2, and Th17-type cytokines and an increase in the anti-inflammatory cytokine IL-10 in peripheral blood mononuclear cells in vitro,⁴⁶ we selected KBL382 for our AD-like experiment.

Male 5-week-old NC/Nga mice were purchased from Central Lab Animals Incorporated (Seoul, Korea). All animal experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC: SNU–160928–1–1 and SNU-190919-3-1) of Seoul National University, Korea. The DFE and DNCB (Sigma-Aldrich Corp, St. Louis, MO, USA)-induced in vivomouse AD model was performed as previously described with minor modifications.^5,31 For DFE-induced model experiment, four groups of mice (N = 7–9) were treated with (1) PBS (Control), (2) DFE+PBS, (3) DFE+KBL365, and (4) DFE+KBL382. After removing hair from the ears and dorsal region using electric clippers and hair removal cream, all groups except Control were applied with 150 µL 4% sodium dodecyl sulfate for 3 h to disrupt the skin barrier. Then, they were treated with 100 mg Biostir AD (Biostir, Hiroshima, Japan), an ointment containing DFE extract, once or twice per week for 7 weeks to induce AD. After 3 weeks from the beginning of the experiment, the DFE+KBL365 and DFE+KBL382 groups were administered suspensions of KBL365 and KBL382 (1 × 10⁹ CFU each) in 200 μL PBS and Control and DFE+PBS mice were administered PBS daily via oral gavage for 4 weeks (Figure 1A). For DNCB-induced model experiment, three groups of mice (N = 6–7) were treated with (1) PBS (Control), (2) DNCB+PBS, and (3) DNCB+KBL382. After removing hair from the ears and dorsal region using electric clippers and hair removal cream, all groups except Control were applied with 200 µL of 1% DNCB dissolved in a vehicle (acetone:olive oil = 3:1) and 15 µL to induce skin legion on d 1 and 4. One week later, 0.4% DNCB was treated to challenge the dorsal skin (200 µL) and ears (15 µL) three times a week for 2 weeks. After 1 week from the beginning of the experiment, the DNCB+KBL382 group was administered suspensions of KBL382 (1 × 10⁹ CFU each) prepared in the same as the DFE-induced AD model experiment daily via oral gavage for 2 weeks (Figure S1A). Then, the mice were sacrificed and further analyzed.

NC/Nga mice were anesthetized with isoflurane, and pictures of the skin were taken weekly until the end of the experiment. The severity of dermatitis was evaluated once a week by scoring the symptoms visible on the ear, neck, and dorsal skin. The severity of dermatitis was assessed based on the four aspects of erythema/hemorrhage, scarring/dryness, edema, and excoriation/erosion on a scale of 0 (none), 1 (mild), 2 (moderate), and 3 (severe). A total dermatitis score was evaluated as the sum of the scores for each aspect, for a maximum score of 12.³¹

On the last day of the experiment, TEWL and skin hydration were determined with a Tewameter TM300 and Corneometer CM 825 (Courage and Khazaka, Cologne, Germany), respectively, applied to the dorsal skin under controlled conditions of 21–22°C and 50–55% humidity.⁴⁷ The probe was placed on the skin and stabilized for approximately 30 s, and then the measurement was recorded. This measurement procedure was repeated five times in the same area at the same pressure.

Scratching behavior was observed, as described previously.⁴⁸ Briefly, the mice were placed individually in acrylic cases composed of three equal cells for at least 1 h for acclimation. Scratching behavior was evaluated by measuring the number of scratches and time that mice spent scratching their nose, ears, and dorsal skin for 10 min on the day before sacrifice. As mice make several rapid scratching movements over a period of about 1 s, a series of such movements was counted as one bout of scratching.⁴⁹

Blood was obtained from the eye blood vessel of each mouse. Serum was separated through centrifugation at 1,200 × g for 15 min at 4°C. Total serum concentrations of IgE were measured using an IgE ELISA kit (Komabiotech, Seoul, Korea) according to the manufacturer’s instructions.

To measure cytokine production at the protein level, frozen skin tissues were weighed and homogenized with radioimmunoprecipitation assay buffer and protease inhibitors (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The homogenates were centrifuged at 12,000 × g for 10 min at 4°C, and their supernatants were collected. Levels of IFN-γ, IL-4, IL-5, IL-13, IL-10, and IL-17 were measured in the skin lysates using a customized multiplex magnetic Luminex kit (R&D Systems, Minneapolis, MN, USA). Expression of Foxp3, IL-2, IL-4, IL-5, IL-6, IL-10, IL-31, MDC, TARC, TGF-β, TNF-α, and TSLP in mouse skin tissue at mRNA level was measured using real-time PCR. RNA was isolated from skin using an Easy-spin Total RNA Extraction Kit (Intron, Seoul, Korea) and cDNA was synthesized using the High-Capacity RNA to cDNA Kit (Thermo Fisher Scientific). Real-time PCR was performed in a Rotor-Gene Q (Qiagen, Hilden, Germany) thermal cycler using the QuantiTect SYBR Green PCR kit (Qiagen) with the primers (0.01 mM) listed in Table S1. Each PCR consisted of denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 5 s and 60°C for 10 s. The relative expression level of the target gene was calculated using the 2^–ΔΔCT method, with normalization to the expression level of the reference gene hypoxanthine-guanine phosphoribosyl transferase.⁵⁰

Dorsal skin tissues were fixed in 10% formaldehyde. This tissue was stained with hematoxylin and eosin to evaluate neutrophil infiltration and with Congo red and toluidine blue to detect eosinophil and mast cell infiltration, respectively. The numbers of eosinophils and mast cells were counted using Pannoramic Viewer software (3DHISTECH, Ltd., Budapest, Hungary).

Flow cytometric analyses were carried out as described previously.⁴⁶ Briefly, isolated MLN cells were stained with Fixable Viability Stain 510 (FVS510; BD bioscience, Franklin Lakes, NJ, USA) for live cells, as well as CD3+ fluorescein isothiocyanate (145–2 C11; BD bioscience), CD4+ Percep-Cyanine5.5 (RM4–5; BD bioscience) and CD25+ phycoerythrin (PC61; BD Bioscience) for cell surface staining after blocking with FcγR and permeabilization with fixation/permeabilization buffer (Ebioscience, San Diego, CA, USA) and were also treated with Foxp3+ Alexa Fluor 647 (MF23; BD Bioscience) for intracellular staining. IgG isotypes were used as a control in all flow cytometry experiments. The CD4+ CD25+ Foxp3+ Treg population was analyzed using a BD FACSVerse™ Flow Cytometer (BD Bioscience).

Intestinal microbiota was observed as described previously.⁴⁶ Briefly, the V3-V4 region of the 16S rRNA gene in total genomic DNA collected from cecum was amplified using the barcoded primers 341F and 805R. PCR amplicons were purified using the QIAquick PCR Purification Kit (Qiagen) and sequenced on the MiSeq platform (Illumina, Inc., San Diego, CA, USA).^51,52 Sequence data were processed using the Quantitative Insights into Microbial Ecology 1.8.0 (QIIME) pipeline. The sequences were clustered into operational taxonomic units (OTUs) at the 97% identity level using the Greengenes database (ver. 13_5). Alpha diversity indexes such as chao1 and PD were estimated to identify significant differences among groups. Phylogenetic beta diversity was measured using the UniFrac distance between samples and visualized based on weighted PCoAs. Linear discriminant analyses (LDAs) of effect size (LEfSe) were performed (http://huttenhower.org/galaxy) to identify significantly different phylotypes within the experimental groups. In addition, phylogenetic investigation of the communities was performed by reconstructing unobserved data to identify functional genes in the sampled microbial community based on data in the KEGG pathway database (GenomeNet; https://www.genome.jp/kegg/pathway.html), as described previously.⁵³

Cecum was homogenized with distilled water and centrifuged at 13,000 × g for 5 min. Then, the supernatant was extracted and mixed with an internal standard (benzoic acid for nonvolatile acids and 1% 2-methylpentanoic acid for volatile acids). The supernatant was added to the extraction solvent (chloroform for nonvolatile acids and ethyl ether for volatile acids) and centrifuged at 13,000 × g for 5 min. The organic layer was transferred to a vial (Agilent Technologies, Santa Clara, CA, USA) and measured through gas chromatography. Nitrogen was used as the carrier gas. The oven temperature was set to 170°C. The flame ionization detector and injection port were set to 225°C. The retention times and peak areas of the standard mix were used as references for unknown samples.⁵⁴

Amino acids were analyzed as described previously.⁴⁶ Cecum extracts were transferred to an autosampler vial and derivatized. After incubation for 10 min at 55°C, the derivatized samples were analyzed using Acquity UPLC (Waters Corporation) with a SYNAPT G2-Si mass spectrometer (Waters Corporation). The mass acquisition mode Tof-MRM and positive electrospray ionization mode were used. Capillary voltage was set to 1.5 kV. The sampling cone voltage varied from 20 V to 60 V among amino acids. The desolvation gas and cone gas flow rates were set to 600 L/h and 50 L/h, respectively. The desolvation temperature was 250°C. Amino acids were confirmed through alignment to the analytical standard mixture and internal standard. Data acquisition and quantification were carried out using MassLynx software 4.1 (Waters Corporation).⁵⁵

Data are expressed as mean ± standard error of the mean (SEM). GraphPad Prism 5.04 (GraphPad Software, Inc., La Jolla, CA, USA) was used to visualize and analyze data using the Mann–Whitney U test. When appropriate, false discovery rate (FDR)-correction was not applied in data presentation. Statistical significance is denoted as *P< .05, **P< .01, ***P< .001.