Animals and experimental design
10-Week-old Sprague-Dawley female rats (Da Shuo, China) were housed under specific pathogen-free condition. After acclimatization for 1 week, rats were intraperitoneally anaesthetized by pentobarbital sodium (2%, 40mg/kg), and bilaterally ovariectomized or subjected to sham surgery. Rats were then treated with probiotics or vehicle. Rats were supplemented by oral gavage with 1X107 CFU/day commercially available infant probiotics preparation that contains 10 single strains including Lactobacillus rhamnosu HN001, Bifidobacterium lactis BI-04, Bifidobacterium animals HN019, Lactobacillus fermentium SBS-1, Lactobacillus reuteri 1e1, Bifidobacterium longum BB536, Bifidobacterium breve M16-V, Bifidobacterium infantis Bi-26, Lactobacillus paracasei Lpc-37 (lifespace, Australia, later referred to as lsPro) or single strain probiotics Bifidobacterium longuBL986 and Lactobacillus rhamnosus LRH09 until sacrifice. CP or AP were induced three weeks after ovariectomy/sham operation. The CP rat model was established according to the methods described by Li and Amar with minor modifications[82]. Rats were intraperitoneally anaesthetized by pentobarbital sodium (2%, 40mg/kg) and were ligated with a 5-0 silk suture around the bilateral maxillary first molars to establish experimental periodontitis. P. gingivalis ATCC 33277, which was anaerobically grown and resuspended to a concentration of 1 × 107 CFU/mL in saline, was smeared on the silk suture every 3 days after ligation[41]. The AP rat model was established as reported by Brasil with minor modifications[20]. The tooth pulps of bilateral mandibular first molars were exposed to oral environment through the occlusal surface using micro-round bur in a high-speed motor, leading to a spontaneous development of apical periodontitis.
Specimens collection
Four weeks after ligation or pulp exposure, samples of the rats were collected. Feces were collected in 1.5-ml sterile Eppendorf tubes, and immediately stored at -80℃. Blood samples were collected from the abdominal aorta under intraperitoneal anesthesia. Rats were then sacrificed by cervical dislocation. 2-cm segments of ileum from all rats were immediately excised and submerged into 1 mL of TriZol reagent for RNA isolation. Bilateral maxilla from rats representing periodontitis, bilateral mandibles from rats representing AP and 2-cm segments ileum from all rats were removed and fixed in 4% paraformaldehyde for 24 hours. Femurs were dissected thoroughly free from soft tissue. The tips of the femurs were removed and bone marrow (BM) was harvested by inserting a syringe needle into one end of the bone and flushing with phosphate-buffered saline (PBS).
Micro-CT scanning and analysis
To evaluate the bone destruction and micro-architecture of alveolar bone and femur bone, micro-CT was performed as previously described with minor modifications[41, 83, 84]. Fixed bone specimens were placed in an airtight cylindrical sample holder and scanned with a micro-CT (μCT50; SCANCO). The micro-CT images were imported into CT-Analyser software (version 1.13, Bruker, Kontich, Belgium) to qualitatively depict the alveolar bone loss and perform histomorphometric analysis of trabecular bone. As for alveolar bone, the scanning was performed at 70 kV and 200 mA with 300-millisecond integration time. All samples were scanned in the sagittal position at a voxel resolution of 10μm. As for rats with CP, mesial and distal bone loss of maxilla was quantified by measuring the average distance between the alveolar bone crest and the cemento-enamel junction (CEJ) with sagittal images selected in the middle of the maxillary first molar. 60 continuous sagittal images of the alveolar bone at root furcation, starting at the beginning of trabecular bone without the middle root, were selected for the analysis of trabecular bone at the region of interest (ROI). For each image, the ROI was a rectangular region of 5.2 mm2 defined right below the top of bone septum between mesial and distal roots. The trabecular parameters of bone volume per tissue volume (BV/TV), Trabecular Number (Tb.N), trabecular separation (Tb. Sp) and Trabecular bone pattern factor (Tb. Pf) at the ROI were measured and quantified. As for rats with AP, the volume of bone resorption cavities under the mesial root of mandibular first molar was quantified. Femurs were also scanned at 70 kV and 200 mA with 300-millisecond integration time in the transverse position at a voxel resolution of 12.5μm. 50 continuous images of the site near femoral condylar in horizontal directions were selected for trabecular bone analyses, and the trabecular parameters of BV/TV, Tb. N, Tb. Sp and Tb. Pf at the ROI were measured and quantified. The micro-CT scanning and measurements were performed blindly by one experienced doctor.
Analysis of serum proinflammatory cytokine
Serum was isolated by centrifuging the blood after clotting at 4,000 rpm for 10 min. The serum levels of TNF-α and IL-17A were assayed by ELISA kits (Invitrogen) according to the manufacturer’s directions.
Histologic analyses of alveolar bone and ileum tissue
Histologic analyses were performed with the method previously described with minor modification[41, 84]. Alveolar bone loss at root furcation for periodontitis and apical destruction surrounding the mesial root of mandibular first molar for apical periodontitis was examined by hematoxylin and eosin (H&E) staining.
Osteoclasts in alveolar bone were examined by tartrate-resistant acid phosphatase (TRAP) staining using the Acid Phosphatase Leukocyte kit (Sigma, St. Louis, MO). Osteocalcin+ (OCN) cells, Foxp3+ cells and IL-17A+ cells in the alveolar bone were examined by immunohistochemistry (IHC) using specific primary antibodies (OCN, ab13420, Abcam; Foxp3, ab22510, Abcam; IL-17A, ab136668, Abcam) and anti-mouse HRP-DAB cell&tissue staining kit (CTS002, R&D) or anti-rabbit HRP-DAB cell&tissue staining kit (CTS005, R&D systems). As for the quantification of positively stained cells in the alveolar bone of CP, six high power fields (hpf) (× 400) at ROI were randomly selected, and the positively stained cells were enumerated by Image-Pro Plus (IPP, Media Cybernetics, USA). Data were presented as the number of positively stained cells per square millimeter of bone marrow. As for the quantification of positively stained cells in the alveolar bone of AP, areas selected for cell enumeration were defined as those centered at a fixed distance from the mesial root apical foramen. Positively stained cells from 5 randomly selected areas at the apical region were counted under hpf (×400) magnification, and data were presented as the number of positively stained cells per hpf.
Intestinal barrier integrity was also examined by H&E staining. Images captured at ×100 magnification were randomly selected, and morphologic features of intestinal villi including intestinal villus density (1/mm), intestinal villus height (mm), intestinal crypt depth (mm), and the ratio of villus height to crypt depth (V/C) were evaluated by IPP.
16S rRNA sequencing of gut microbiota
Genomic DNA was extracted and purified from feces (3-5g) with the QIAamp DNA stool mini kit (QIAGEN). The resulting DNA was quantified with Quant-iTTM PicoGreen reagent (Invitrogen). The sequencing of 16S rRNA amplicons (V1-V3 region) was performed by MiSeq 300PE (Illumina MiSeq System) at Majorbio (Shanghai, China). Primers used in present study was 27F (5’-AGAGTTTGATCCTGGCTCAG-3’) and 533R (5’-TTACCGCGGCTGCTGGCAC-3’). A total of 151465 sequence reads of feces were generated from the amplicon library, with an average of 12624 reads per sample. The sequences were clustered into 850 operational taxonomic units (OTU) in feces at a similarity level of 97%. Bioinformatics were performed by Mothur and QIIME2.0 software, including quality control of raw data, taxonomic annotation based on the Silva database, taxonomy-based comparisons at the OTU level, β-diversity analysis including principal component analysis (PCA) and principal coordinates analysis (PCoA), dissimilarity analyses including analysis of similarity (ANOSIM) and non-parametric multivariate analysis of variance (Adonis). The sequencing raw data were deposited in Sequence Read Archive (https://www.ncbi.nlm.nih.gov/Traces/sra; accession nos. SRP285657).
Fecal butyrate quantification
Fresh feces (10 mg per rat) were processed by the ether extraction method. A 20-μL volume of the prepared sample solution was analyzed by a high-performance liquid chromatography (HPLC) system (model 1260; Agilent) with an HPLC column (4.0 mm × 250 mm, 5 μm, InertSustain C18; SHIMADUZ-GL). The mobile phases were 0.2% H3PO4 solution (A phase) and methanol (B phase; Chromatographic Grade; Fisher Scientific), with a flow rate of 0.8 mL/min and a column temperature of 30 °C. HPLC was performed with binary solvent–delivery gradient elution with a detection wavelength of 210 nm.
RNA isolation and quantitative reverse transcription polymerase chain reaction
Intestinal RNA was isolated and purified from an ileum segment with TriZol Reagent (Invitrogen). Reverse transcription of RNA into cDNA was performed with the Primer-Script RT Reagent Kit with gDNA Reaser (RR047A; Takara Bio). The expression of genes encoding intestinal tight junction (TJ) proteins, including occludin, zo-1, claudin1, claudin 2, claudin 3, and Jam3, were quantified with the β-actin as internal control. The primer sequences are presentedin Table 1. Relative quantitative analysis was performed with the 2–ΔΔCT method.
Fecal RNA was isolated and purified with stool RNA kit (R6828-01; OMEGA bio-tec). Reverse transcription of RNA into cDNA was performed with the Primer-Script RT Reagent Kit with gDNA Reaser (RR047A; Takara Bio). The expression levels of butyryl-CoA:acetate CoA transferase (But) and butyrate kinase (Buk) of the gut microbiota were measured with the 16S rRNA gene as internal control. The primer sequences are presented in Table 2. Relative quantitative analysis was performed with the 2–ΔΔCT method. As for the measurement of butyrate producing genera, fecal DNA was isolated and purified with stool DNA kit (QIAamp DNA stool mini kit, QIAGEN). The relative abundance of Clostridium leptum subgroup, Clostridium coccoides subgroup, Fecalibacterium prausnitzii, and Roseburia/E. rectale cluster were measured with the 16s-univ-1 gene as internal control. The primer sequences are presented in Table 3. Relative quantitative analysis was performed with the 2–ΔΔCT method.
Intestinal permeability
After fasting and deprivation of water overnight, rats were gavage-fed with 100 mg/mL of fluorescein isothiocyanate-conjugated dextran (FITC-dextran; 4.4 kDa, catalog FD4, Sigma-Aldrich) at 44 mg/100g of body weight per rat. 4 Hours later, serum was collected from the abdominal aorta under intraperitoneal anesthesia. The concentration of FITC-dextran in serum was analyzed by spectrophotofluorometry with excitation at 485 nm and emission at 528 nm, with reference to a standard of serially diluted FITC-dextran (0, 150, 300, 600, 800, 1,000 μg/mL). Serum lipopolysaccharide (LPS) levels were measured by an ELISA kit (Invitrogen).
Flow cytometry analysis of Th17/Treg cells
Fluorescence-activated cell sorting (FACS) was used to evaluate the frequency (%) of Th17 cells (CD4+IL-17A+ cells) and Treg cells (CD4+CD25+Foxp3+ cells) in the bone marrow of femur.
Th17 cells
Bone marrow cells were incubated at 37°C for 12 h with DMEM cell high-sugar medium mixed with 10% excellent fetal bovine serum and GolgiPulg (1μg/ml). Single-cell suspensions of bone marrow were then prepared in staining buffer, and 1μL of specific FcR blocker was added to block the nonspecific staining mediated by fluorescent antibody FcR receptor. The cells were then stained with fluorescein-isothiocyanate (FITC)-labeled anti-CD4 antibodies to detect surface markers. After membrane rupture, cells were intracellular stained with phycoerythrin (PE)-labeled anti-IL-17A antibodies. Cells were detected using a flow cytometer (Beckman, FC500, USA).
Treg cells
Single-cell suspensions of bone marrow were prepared in staining buffer. The cells were then stained with FITC-labeled anti-CD4 antibodies, followed by PE anti-CD25 antibodies. After the fixation and membrane rupture, cells were stained with allophycocyanin (APC)-labeled anti-Foxp3 antibodies. Cells were detected using a flow cytometer (Beckman, FC500, USA).
Butyrate treatment
To further validate the role of intestinal butyrate in maintaining gut permeability and preventing skewed Th17/Treg-induced bone resorption, additional OVX/sham rats were gavage-fed with sodium butyrate (400 mg/kg) every day for 8 wk until sacrifice. Serum, ileum and bone marrow cells were collected for further analyses with the same methods as described above.
Statistical analysis
All data were statistically analyzed by SPSS v25.0 (Statistical Product and Service Solutions, International Business Machine Inc, USA). Differences between groups were evaluated by one-way analysis of variance with Bonferroni correction for multiple comparisons or by the Kruskal-Wallis H test with post hoc tests applying the Nemenyi test for multiple comparisons. The data were presented as means ± standard deviation (SD), and a 2-tailed p < 0.05 was considered significant.
Table 1. qPCR primer sequences for the TJ protein encoding genes.
Primers
|
FP (5’ to 3’)
|
RP (5’ to 3’)
|
|
Ocln
|
GCACGTTCGACCAATGCTCT
|
AGATGCCCGTTCCATAGGCT
|
|
Cldn 1
|
ACTTTGCAGGCAACCAGAGC
|
TGCTGTGGCCACTAATGTCG
|
|
Cldn2
|
TCCCCAACTGGCGAACAAGT
|
AGCGAGGACATTGCACTGGA
|
|
Cldn 3
|
GGCCAGATGCAGTGCAAGAT
|
CCGAAGGCTGCCAGTAGGAT
|
|
Zo 1
|
TCCCACAAGGAGCCATTCCT
|
GTCACAGTGTGGCAAGCGT
|
|
Jam 3
|
ACGGTCAGACTCAGCCATCT
|
AGAGTGCCTGTCTCCGAGTT
|
|
Table 2. qPCR primer sequences for butyryl-CoA:acetate CoA transferase and butyrate kinase encoding genes.
Primers
|
FP (5’ to 3’)
|
RP (5’ to 3’)
|
|
But
|
GCIGAICATTTCACITGGAAYWSITGGCAYATG
|
CCTGCCTTTGCAATRTCIACRAANGC
|
|
Buk
|
GTATAGATTACTIRYIATHAAYCCNGG
|
CAAGCTCRTCIACIACIACNGGRTCNAC
|
|
Table 3. qPCR primer sequences for the butyrate-producing genera.
Primers
|
FP (5’ to 3’)
|
RP (5’ to 3’)
|
Clostridium leptum subgroup
|
CGTCATCCCCACCTTCCTCC
|
GCAAGACAGTTTCAAGCGCA
|
Clostridium coccoides subgroup
|
AATGCCGCGGTGAATACGTT
|
GCACCTTCCGATACGGCTAC
|
Fecalibacterium prausnitzii
|
GATGGCCTCGCGTCCGATTAG
|
CCGAAGACCTTCTTCCTC
|
Roseburia/E. rectale cluster
|
CKGCAAGTCTGATGTGAAAG
|
GCGGGTCCCCGTCAATTCC
|