Involvement of the Gut Microbiota and Barrier Function in Glucocorticoid‐Induced Osteoporosis (original) (raw)

Journal Article

,

Department of Physiology Michigan State University East Lansing MI USA

Search for other works by this author on:

,

Department of Physiology Michigan State University East Lansing MI USA

Search for other works by this author on:

,

Comparative Medicine and Integrative Biology Program Michigan State University East Lansing MI USA

Department of Physiology Michigan State University East Lansing MI USA

Search for other works by this author on:

,

Department of Physiology Michigan State University East Lansing MI USA

Search for other works by this author on:

,

Department of Molecular Virology and Microbiology Baylor College of Medicine Houston TX USA

Search for other works by this author on:

,

Bone and Joint Center Henry Ford Health System Detroit MI USA

Search for other works by this author on:

,

Department of Biochemistry and Molecular Biology Michigan State University East Lansing MI USA

Search for other works by this author on:

,

Department of Biochemistry and Molecular Biology Michigan State University East Lansing MI USA

Search for other works by this author on:

,

Department of Molecular Virology and Microbiology Baylor College of Medicine Houston TX USA

Search for other works by this author on:

,

Department of Physiology Michigan State University East Lansing MI USA

Search for other works by this author on:

... Show more

Results from this work were presented at: the American Society of Bone and Mineral Research (ASBMR) 2017 Annual Meeting, September 8–11, 2017, in Denver, CO, USA; the ASBMR 2019 Annual Meeting, September 20–23, 2019, in Orlando, FL, USA; and the Experimental Biology 2019 meeting, April 6–9, 2019, in Orlando, FL, USA.

NP and LRM contributed equally to this work and are co‐senior authors.

Author Notes

Received:

03 September 2019

Revision received:

05 December 2019

Accepted:

14 December 2019

Published:

30 December 2019

• PDF
• Split View
• • Article contents
  • Figures & tables
  • Video
  • Audio
  • Supplementary Data
• Cite

Cite

Jonathan D Schepper, Fraser Collins, Naiomy Deliz Rios‐Arce, Ho Jun Kang, Laura Schaefer, Joseph D Gardinier, Ruma Raghuvanshi, Robert A Quinn, Robert Britton, Narayanan Parameswaran, Laura R McCabe, Involvement of the Gut Microbiota and Barrier Function in Glucocorticoid‐Induced Osteoporosis, Journal of Bone and Mineral Research, Volume 35, Issue 4, 1 April 2020, Pages 801–820, https://doi.org/10.1002/jbmr.3947
Close

Navbar Search Filter Mobile Enter search term Search

ABSTRACT

Glucocorticoids (GCs) are potent immune‐modulating drugs with significant side effects, including glucocorticoid‐induced osteoporosis (GIO). GCs directly induce osteoblast and osteocyte apoptosis but also alter intestinal microbiota composition. Although the gut microbiota is known to contribute to the regulation of bone density, its role in GIO has never been examined. To test this, male C57/Bl6J mice were treated for 8 weeks with GC (prednisolone, GC‐Tx) in the presence or absence of broad‐spectrum antibiotic treatment (ABX) to deplete the microbiota. Long‐term ABX prevented GC‐Tx‐induced trabecular bone loss, showing the requirement of gut microbiota for GIO. Treatment of GC‐Tx mice with a probiotic (Lactobacillus reuteri [LR]) prevented trabecular bone loss. Microbiota analyses indicated that GC‐Tx changed the abundance of Verrucomicobiales and Bacteriodales phyla and random forest analyses indicated significant differences in abundance of Porphyromonadaceae and Clostridiales operational taxonomic units (OTUs) between groups. Furthermore, transplantation of GC‐Tx mouse fecal material into recipient naïve, untreated WT mice caused bone loss, supporting a functional role for microbiota in GIO. We also report that GC caused intestinal barrier breaks, as evidenced by increased serum endotoxin level (2.4‐fold), that were prevented by LR and ABX treatments. Enhancement of barrier function with a mucus supplement prevented both GC‐Tx–induced barrier leakage and trabecular GIO. In bone, treatment with ABX, LR or a mucus supplement reduced GC‐Tx–induced osteoblast and osteocyte apoptosis. GC‐Tx suppression of Wnt10b in bone was restored by the LR and high‐molecular‐weight polymer (MDY) treatments as well as microbiota depletion. Finally, we identified that bone‐specific Wnt10b overexpression prevented GIO. Taken together, our data highlight the previously unappreciated involvement of the gut microbiota and intestinal barrier function in trabecular GIO pathogenesis (including Wnt10b suppression and osteoblast and osteocyte apoptosis) and identify the gut as a novel therapeutic target for preventing GIO. © 2019 American Society for Bone and Mineral Research.

DYSBIOSIS, ENDOTOXIN, GLUCOCORTICOID, INTESTINAL PERMEABILITY, LACTOBACILLI, LACTOBACILLUS REUTERI, MDY, MICROBIOTA, OSTEOBLAST, OSTEOCLAST, AND APOPTOSIS, PREDNISOLONE, TRABECULAR BONE

Introduction

Glucocorticoids (GCs) are anti‐inflammatory, immune‐modulating therapeutic drugs used to manage inflammatory diseases including inflammatory bowel disease (IBD), allergic conditions, bronchial asthma, rheumatoid arthritis, ankylosing spondylitis, and chronic renal diseases, as well as some cancers. Although >1.2% of the US population are using GCs as a long‐term therapy,1 this chronic treatment is associated with significant side effects including GC‐induced osteoporosis (GIO).2, 3, 4, 5, 6 Though anti‐resorptive and anabolic drugs can help to reduce osteoporosis due to GCs, GIO remains the most common form of secondary osteoporosis and is a significant risk for fracture.1, 3, 7, 8 Due to a lack of monitoring and unwillingness to take drug therapies for osteoporosis, less than 6% of women under 50 years taking GCs are monitored for bone loss.3

It is well established that GCs promote bone loss by causing rapid bone resorption, followed by prolonged and profound suppression of bone formation.4, 9, 10 The physiologic GC hormone cortisol is released from the adrenal cortex and elevated in response to psychological or psychosocial stress. Physiologic concentrations of GC have been shown to stimulate osteoblast differentiation.11 However, abnormal increases in endogenous cortisol levels associated with aging12, 13 and possibly anorexia,14 as well as pharmacologic doses of exogenous GCs can decrease osteoblast differentiation and viability.11, 15 GC treatment has been shown to induce apoptosis of osteoblasts as well as osteocytes in both mouse5, 16, 17 and human18, 19, 20, 21, 22 bone samples. In addition, GCs have been shown to skew mesenchymal stem cell differentiation toward adipocytes and thus impair osteoblast lineage selection and differentiation.3, 23, 24, 25 Consistent with suppressed osteoblast lineage selection, GC treatment reduces Wnt10b expression (a positive regulator of osteoblastogenesis), cell viability, and bone density.7, 11, 26, 27, 28, 29, 30 GC signaling in osteoblasts involves binding of GC to the cytosolic GC receptor (alpha), which then translocates to the nucleus, binds to GC response elements, and regulates gene expression.4, 15 In addition to the direct effects of GC on bone, GCs also affect other organ systems that can further influence bone health. Especially relevant to the present study, GCs can alter intestinal microbiota composition in animal models.31, 32

The microbiome, defined as the collective genetic material of the microbiota, contains an estimated 3 to 8 million unique genes, >100‐fold larger than the human genome.33, 34 The intestinal microbiota contains bacteria as well as fungi, viruses, and archaea.35 Significant attention is focused on the intestinal bacteria (more than 100 trillion), which include approximately 1000 different species from 29 bacterial phyla.35 Microbiota species diversity is associated with health; in contrast, dysbiosis (altered microbial composition) and reduced diversity are associated with diseases such as obesity, diabetes, and IBD.36, 37, 38, 39, 40 The Firmicutes and Bacteroidetes phyla comprise a significant portion of the human and mouse microbiota, and the ratio of their abundance is often used as a marker of dysbiosis,41 though both phyla contain bacteria that can benefit as well as reduce health.

The microbiota can benefit host health by producing essential nutrients, digesting otherwise indigestible food components, and enhancing maturation of the immune system.42, 43 An unhealthy imbalance in the microbiota community composition, called dysbiosis, is linked to a variety of metabolic, inflammatory, and immunologic diseases.44, 45, 46, 47 Dysbiosis is also linked to increased intestinal permeability/leaky gut.42, 48 A leaky intestinal barrier allows translocation of bacteria and their products into the lamina propria where they can activate immune cells, cause inflammation, and enter the systemic circulation. Thus, barrier leaks, like dysbiosis, are associated with disease.49

It is now clear that changes in the gut microbiota can impact bone density and health.50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69 For example, intestinal infection with pathogenic bacteria can induce bone loss in male mice.70 In contrast, treatment with beneficial bacteria (probiotics) alters gut microbiota composition50, 53, 54, 55, 57, 71, 72, 73, 74 and enhances bone density in healthy male mice,53 post–antibiotic treated male mice,68 ovariectomized mice,50, 54, 55 inflamed intact female mice,75 and type 1 diabetic male mice.76 Given the powerful role of the intestinal microbiota in regulating bone health, we hypothesized that the gut microbiota and barrier function may play a role in the regulation of GIO pathogenesis. Here we report that the GC treatment promotes dysbiosis, which contributes to GIO. We further show that GC treatment causes intestinal barrier leaks and raises serum endotoxin levels. Treatments that target the gut to alter the microbiota (with Lactobacillus reuteri) or enhance intestinal barrier function (with high‐molecular‐weight polymer [MDY]) prevented GC‐mediated osteoblast and osteocyte apoptosis, Wnt10b suppression, and GIO. Correspondingly, we demonstrate that GIO is prevented by upregulation of Wnt10b. Taken together; we identify the gut microbiome and intestinal barrier as key components in GIO pathology that can serve as therapeutic targets for GIO prevention.

Subjects and Methods

Animals and experimental design

Mice

Adult C57BL/6J male mice (15‐week old) were purchased from The Jackson Laboratory (Bar Harbor, ME, USA), co‐housed and allowed to acclimatize to the animal facility for 1 week prior to the start of the experiment. Mice were subsequently weighed and divided into groups based on similar average weights per group. Mice were housed in shoebox cages (at a maximum of five mice/cage), given sterile Teklad 2019 chow (Madison, WI, USA) and sterile water ad libitum, and were maintained on a 12‐hour light/dark cycle at 24°C in specific pathogen‐free facilities. In addition, osteocalcin promoter‐driven Wnt10b transgenic mice (generously provided by Dr. Ormond MacDougald, University of Michigan27) were backcrossed more than 12 generations with C57BL/6J mice and used in studies at 16 weeks of age. Mice were genotyped with genomic DNA isolated from ear tissue samples (DNeasy Blood and Tissue Kit; Qiagen, Valencia, CA, USA). DNA was amplified by RT‐PCR with iQ SYBR Green Supermix (Bio‐Rad Laboratories, Hercules, CA, USA) and primers specific to the transgene.27 All animal procedures were approved by the Michigan State University Institutional Animal Care and Use Committee and complied with NIH guidelines. Animal experiments were repeated at least twice to ensure reproducibility.

Mouse treatments

Mice at 16 weeks of age were anesthetized (isoflurane inhalation) and implanted subcutaneously with either a placebo pellet (control) or a pellet containing 5 mg of prednisolone (60‐day slow release pellet; Innovative Research of America, Sarasota, FL, USA).77 Briefly, a small (10 mm) skin incision was made on the upper back and using a trocar the pellet was inserted into the interscapular region. This treatment corresponds to a daily dose of 0.08 mg prednisolone per day or on average 2.5 mg/kg/day.77

For probiotic and MDY treatment experiments, mice were randomly split into five experimental groups (9–10 mice per group): (i) Control; (ii) prednisolone treatment (GC; 2.5 mg/kg/day via subcutaneous implant) and sterile drinking water (Veh); (iii) GC treatment and sterile drinking water containing 3.3 × 108 CFU/mL L. reuteri 6475 (LR); (iv) GC treatment and sterile drinking water containing 3.3 × 108 CFU/mL Lactobacillus rhamnosus GG (LGG); or (v) GC treatment and sterile drinking water containing 1.25% MDY, a high molecular weight polymer used as a mucus supplement.78, 79 These oral treatments continued for the duration of the 56‐day (8 week) GC experiment.

Microbiota depletion experiment

Male 16‐week‐old mice were split into four experimental treatment groups: (i) control (C); (ii) prednisolone‐treated (GC; 2.5 mg/kg/day via subcutaneous implant); and (iii) continuous broad‐spectrum antibiotics (ABX) or (iv) continuous broad‐spectrum antibiotics with prednisolone treatment (ABX + GC). One week prior to the start of GC treatment, the ABX mouse groups were treated with broad‐spectrum antibiotics ampicillin (1.0 g/L; Sigma‐Aldrich, St. Louis, MO, USA) and neomycin (0.5 g/L; Sigma‐Aldrich) at a dose of 160 and 80 mg/kg/day, respectively.46, 80 These antibiotics are poorly absorbed by the intestine (or unabsorbed in the case of neomycin)81 to allow the successful targeting and depletion of intestinal commensal microbes.46, 68, 82 Water intake was measured to account for any increase in consumption due to GC treatment83 and antibiotic dose was altered to keep dose the same among antibiotic treatment groups.

Bacterial culture for oral treatment

L. reuteri ATCC PTA 6475 (LR) and Lactobacillus rhamnosus GG (LGG) were initially cultured and kept under anaerobic conditions on deMan, Rgosa, Sharpe media (MRS; DIFCO, Becton Dickinson, Franklin Lakes, NJ, USA) plates at 37°C. To prepare bacteria for oral treatments (via drinking water), LR and LGG were anaerobically cultured in 10 mL of MRS media overnight at 37°C. Bacteria were then subcultured into 250 mL of fresh MRS broth and grown until log phase (optical density 600 nm [OD600] = 0.4), spun down, washed with sterile PBS, then resuspended in 60 mL sterile PBS, and stored at −80°C. Aliquots were taken prior to freezing to determine CFU/mL. Bacteria were resuspended in drinking water at a final concentration of 3 × 108 CFU/mL. Bacteria viability in the drinking water was confirmed by plating aliquots on MRS plates and growing anaerobically overnight at 37°C. Both LR and LGG bacterial identity was confirmed using strain‐specific primers via quantitative real‐time polymerase chain reaction (PCR).84, 85

Bone marrow CD4+ FACS

Total bone marrow (BM), both nucleated and non‐nucleated cells, were isolated from the mouse femur. Specifically, femurs were isolated and extraneous tissue/muscle detached. The femoral head was severed, and the femur placed cut side down into a 0.5‐mL microcentrifuge tube with a small hole in the base. The 0.5‐mL microcentrifuge tube was placed inside a 1.5‐mL microcentrifuge tube and centrifuged at 5000 g for 20 s. Bone marrow was collected in the 1.5‐mL microcentrifuge tube and resuspended in 1 mL of alpha‐MEM (catalog# 12561049; ThermoFishe Scientific, Waltham, MA, USA). Cells (at 2 × 106) were incubated with Fc block (BD Pharmingen, San Jose, CA, USA) for 15 min. Cells were stained with anti‐mouse CD4‐V500 (Clone RM 4–5; BD Bioscience, San Jose, CA, USA) for 30 min at 4°C. Cells were washed three times in assay buffer (PBS, 0.5% bovine serum albumin [BSA], 5mM EDTA) followed by fixation.

Serum measurements

Sterile blood was collected at the time of harvest via cardiac puncture, allowed to clot at room temperature for 5 min, and then centrifuged at 2500 rcf for 10 min. Serum was removed and stored at −80°C. Serum went through no more than two freeze/thaw cycles. Serum bacterial endotoxin levels were detected using the HEK‐Blue LPS Detection Kit (InvivoGen, San Diego, CA, USA) according to the manufacturer's instructions. Standard curves where used for each individual endotoxin assay.

Micro–computed tomography bone analysis

Fixed femurs and vertebrae were scanned using GE Explore Locus micro–computed tomography (μCT) system (GE Healthcare, Piscataway, NJ, USA) using a voxel resolution of 20 μm obtained from 720 views. Angle of increment was 0.5 degrees, and beam strength was set at 80 peak kV and 450 μA. Each run consisted of controls, GC (± treatments), and a calibration phantom to standardize grayscale values and maintain consistency. A fixed threshold of 1100 (determined based on automated and isosurface analyses) was used for all bones. Femoral bone analyses were performed on trabecular bone defined as beginning proximal (a distance of 1% of total bone length) to the growth plate and then extending 10% of total bone length toward the diaphysis, excluding cortical bone. Trabecular bone was also analyzed within the lumbar (L3) vertebrae. Trabecular bone volume fraction, thickness, spacing, and number values were calculated by a GE Healthcare MicroView software application for visualization and analysis of volumetric image data. Values for both bone volume/total volume (BV/TV) and BV/TV/body weight (BV/TV/BW) are given. The latter measure corrects for the influence that changes in body weight can have on the bone. Bones were aligned parallel to the z axis and cortical measurements were performed in a 2.2‐mm × 2‐mm × 2‐mm cube centered midway (1/2 bone length) down the length of the femur. All bone measurements were performed blind.

Mechanical testing

Before testing, the moment of inertia about the anterior/posterior axis (IA/P) and the distance from the neutral axis to the medial surface under tension (c) were determined at the site of fracturing by μCT imaging as described in the previous section. Mechanical properties of the mouse tibias were then determined under four‐point bending using an EnduraTec ELF 3200 Series (Bose, Framingham, MA, USA).86 The base support span was 9 mm with a load span of 3 mm. The tibia was positioned in the loading device, so the medial surface was in tension by placing the most distal portion of the tibia and fibula junction directly over the leftmost support. Each tibia was loaded at 0.01 mm/s until failure while the load and displacement were recorded. The force–deflection curve then used to calculate the structural‐level properties, while tissue‐level properties were estimated using the following beam‐bending equations:

Stress = σ = f·a·c / 2·IA/P.

Strain = ε = 6·c·d / a (3·L – 4·a).

In each equation, f is the applied force, d is the resulting displacement, a is the distance between the inner spans (3 mm), and L is the distance of the outer spans (9 mm). The yield point was determined from the stress–strain relationship using a 20% offset method.87 The analyses were performed blind to the group conditions.

Bone histology and histomorphometry

For dynamic histomorphometric measures of bone formation, mice were injected intraperitoneally with 200 μL of 10 mg/mL calcein (Sigma‐Aldrich) dissolved in sterile saline at 7 and 2 days prior to harvest. Femurs were embedded in paraffin blocks and sectioned in 5‐μm slices.88 Distal femur metaphyseal sections were viewed under a fluorescent Nikon Eclipse E800 microscope (Nikon Instruments Inc, Melville, NY, USA) and four to five digital images were taken. The distance between the calcein lines (mineral apposition rate [MAR]) and the length of the calcein lines (single + ½ double‐labeled surfaces; mineralized surface [MS]) along with the total bone surface (BS) were measured to calculate the bone formation rate (BFR) using Image Pro‐Plus 7.0 (Media Cybernetics, Rockville, MD, USA). Slides were stained for tartrate‐resistant acid phosphatase (TRAP) activity and counterstained with hematoxylin according to the manufacturer's protocol (387A‐1KT; Sigma‐Aldrich). Slides were photographed five images per slide at magnification ×25 for osteoclast and osteoblast counts and at magnification ×10 for adipocytes. Image Pro‐Plus software was used in analysis of slide images. In the distal femoral trabecular region, ranging from the growth plate to 2 mm toward the diaphysis, osteoblast surface area was measured and expressed as a percentage of total bone surface. Adipocytes >30 μm in size were counted in the same trabecular area and expressed as number per μm of marrow area. The identity of sections was not revealed until all measures were obtained. Cell death of osteoblasts and osteocytes was determined using a TACS‐XL Basic In Situ Apoptosis Detection Kit (TUNEL assay) (Trevigen Inc., Gaithersburg, MD, USA) in same trabecular and cortical region. Osteoblasts and osteocytes with positive nuclei were counted and expressed as a percentage of total osteoblasts/osteocytes counted per bone. Five trabecular and cortical regions were examined for each mouse.

RNA analysis

Tibias were collected and cleaned of muscle and connective tissue and immediately flash frozen in liquid nitrogen and stored at −80°C. Frozen tibias were crushed under liquid nitrogen conditions with a Bessman tissue pulverizer (Spectrum Laboratories, Rancho Dominguez, CA, USA). RNA was isolated using TriReagent (Molecular Research Center, Cincinnati, OH, USA) and checked for integrity by agarose‐gel electrophoresis. cDNA was produced by reverse transcription using Superscript II reverse transcriptase kit and oligo dT primers (Invitrogen, Carlsbad, CA, USA). Gene expression levels were amplified by real‐time PCR with iQ SYBR Green supermix (Bio‐Rad Laboratories) and specific gene primers. Hypoxanthine guanine phosphoribosyltransferase (HPRT) mRNA level, which is not affected by treatment groups, was used as housekeeping gene. The PCR protocol included 40 cycles using the iCycler (Bio‐Rad Laboratories) and resulting data was evaluated using iCycler software. Each cycle consists of 95°C for 15 s, 60°C for 30 s, and 72°C for 30 s. Negative controls included primers without cDNA. Primers for mouse genes were as follows: HPRT (Forward, 5′‐AAGCCTAAGATGAGCGCAAG‐3′; Reverse, 5′‐TTACTAGGCAGATGGCCACA), Bax (Forward 5′‐GACAGGGGGCTTTTTGCTA‐3′; Reverse, 5′‐TGTCCACGTCAGCAATCATC‐3′), Bcl‐2 (Forward 5′‐GACAGAAGATCATGCCGTCC‐3′; Reverse, 5′‐GGTACCAATGGCACTTCAAG‐3′), and Wnt10b (Forward 5′‐AATGCGGATCCACAACAACA‐3′; Reverse, 5′‐TTCCATGGCATTTGCACTTC‐3′).

DNA preparation of fecal samples

Fecal samples were transferred to MO BIO Ultra Clean Fecal DNA bead Tubes (MO BIO Laboratories, Inc., Carlsbad, CA, USA) containing 360 μL of buffer ATL (Qiagen) and homogenized for 1 min in a BioSpec Mini‐Beadbeater (BioSpec Products, Inc., Bartlesville, OK, USA). Forty microliters (40 μL) Proteinase K (Qiagen) was added and samples were incubated for 30 min at 55°C, then homogenized again for 1 min and incubated at 55°C for additional 30 min. DNA was extracted with a Qiagen DNeasy Blood and Tissue kit.

DNA extraction from mouse colon and fecal samples, 16S rRNA gene amplification, and sequencing

Immediately following euthanasia, fecal samples were obtained, and the intestines were cleaned of connective tissue and flushed of luminal contents, snap frozen in liquid nitrogen, and stored at −80°C before bacteria sequencing. A random mix of mice housed in different cages was used to avoid measuring cage specific differences. DNA for microbial sequence analysis was extracted from mouse colon and fecal samples by bead‐beating and modified extraction with Qiagen DNeasy Blood and Tissue kits as described.54, 89 Bacterial 16S sequences spanning variable region V4 were amplified by PCR with primers F515/R806 with a dual indexing approach and sequenced by Illumina MiSeq (Illumina, San Diego, CA, USA) as described.90 Twenty‐microliter (20‐μL) PCR reactions were prepared in duplicate and contained 40 ng DNA template, 1× Phusion High‐Fidelity Buffer (New England BioLabs, Ipswich, MA, USA), 200μM dNTPs (Promega [San Luis Obispo, CA, USA] or Invitrogen [Carlsbad, CA, USA]), 10nM primers, 0.2 units of Phu‐sion DNA Polymerase (New England BioLabs), and PCR grade water. Reactions were performed in an Eppendorf Pro thermal cycler with an initial denaturation at 98°C for 30 s, followed by 30 cycles of 10 s at 98°C, 20 s at 51°C, and 1 min at 72°C. Replicates were pooled and purified with Agencourt AMPure XP magnetic beads (Beckman Coulter, Brea, CA, USA). DNA samples were quantified using the QuantIt High Sensitivity DNA assay kit (Invitrogen) and pooled at equimolar ratios. Kit controls were used to control for kit contamination, including primers run in the absence of cDNA. The quality of the pooled sample was evaluated with the Bioanalyzer High Sensitivity DNA Kit (Agilent Technologies, Santa Clara, CA, USA).

Microbial community analysis

Sequence data was processed using the MiSeq pipeline for mothur using software version 1.38.191 (https://mothur.org/wiki/Main_Page) as described.89 In brief, forward and reverse reads were aligned, sequences were quality trimmed and aligned to the Silva 16S rRNA gene reference database formatted for mother, and chimeric sequences were identified and removed using the mothur implementation of UCHIME (https://drive5.com/usearch/manual/uchime_algo.html). Sequences were classified according to the mothur‐formatted Ribosomal Database Project (version 16, February 2016; https://www.mothur.org/wiki/RDP_reference_files) using the Bayesian classifier in mothur, and those sequences classified as Eukarya, Archaea, chloroplast, mitochondria, or unknown were removed. The sequence data were then filtered to remove any sequences present only once in the data set. After building a distance matrix from the remaining sequences with the default parameters in mothur, sequences were clustered into operational taxonomic units (OTUs) with 97% similarity using the average‐neighbor algorithm in mothur. A total of 871 OTUs were identified across all samples with an average rarefaction depth of 54,791 reads per sample. Visualization of microbiome communities were performed with R (R Foundation for Statistical Computing, Vienna, Austria; https://www.r-project.org/), utilizing the phyloseq package.92, 93 Analysis of similarity (ANOSIM) was performed in mothur.

Fecal transplant experimental design

Fecal donor mice, 15‐week‐old male C57BL/6J mice, were split into two experimental treatment groups: (i) control or (ii) GC‐treated. Fresh fecal material was collected from each mouse group weekly. Fecal material was pooled from each mouse group (1 g) and homogenized in 10 mL pre‐reduced sterile PBS; this slurry was used to gavage the recipient mice.89, 94 To deplete their microbiota, recipient mice were treated with antibiotics (broad‐spectrum: ampicillin 1.0 g/L (Sigma‐Aldrich) and neomycin 0.5 g/L (Sigma‐Aldrich) 1 week prior to the first fecal transplant. This antibiotic combination results in minimal CFU in fecal contents by 1 week (Fig. 2 A). Upon cessation of ABX treatment, the recipient mice were gavaged weekly with 300 μL of control or GC‐Tx mice fecal slurry (3 × 105 CFU) for 8 weeks.

Statistical analysis

All measurements are presented as mean ± SE. Power analysis conducted with data from pilot studies, determined using an F‐test that at a 5% level of significance and 80% power to detect differences among the means, that eight mice are needed per group for ANOVA analyses of femur bone outcomes. Each study was repeated at least two times to rule out single environmental effect and to enhance study findings and power. Student's t test and one‐way ANOVA with Tukey post‐test were performed using GraphPad Prism software version 7 (GraphPad, San Diego, CA, USA). A p value ≤.05 was considered significant and <0.01 highly significant. The machine‐learning based algorithm Random Forests (RF) was used to assess how the microbiome data reflected categorical or linear variables related to the treatment or bone measurements. The microbiome data was classified with an RF using 5000 trees according to the treatment groups and the out‐of‐bag error was used to assess model classification accuracy. Variable importance plots from the RF were used to determine which OTUs were most strongly contributing to the classification accuracy and these were plotted as stacked bar graph after ranking the OTUs according to the mean decrease in accuracy of their classification contribution. RF regression models were run in the same manner except the microbiome data was correlated to the BV/TV/BW bone measure and the % of variance explained in the model used to assess how strongly the microbiome reflected that quantitative measure. Variable importance plots were used to determine which microbial OTUs best correlated with BV/TV/BW.

Results

GC treatment alters fecal microbiota composition

Previous studies show GCs as well as gut microbiota to be important regulators of bone health.59, 68, 71, 95, 96, 97, 98, 99 However, whether gut microbiota is necessary for the GC effects on bone is not known. To first assess whether GC treatment (GC‐Tx) alters microbiota, we treated 16‐week‐old male C57BL/6J mice with prednisolone (subcutaneous pellet, 2.5 mg/kg/day) or a placebo for 8 weeks.77 Stool samples were used for 16S rRNA sequencing, which revealed shifts in the relative abundance of bacteria phylum in GC‐Tx compared to control mice (Fig. 1). These shifts include a GC‐Tx–induced decrease in Verrucomicobiales and Bacteriodales and an increase in Clostridiales. Statistical analyses of community structure differences (using ANOSIM) determined that GC‐Tx was significantly different from the control (R = 0.807, p < .0001).

Glucocorticoid treatment causes dysbiosis. Sixteen‐week‐old C57BL/6J male mice were split into two groups: Sterile drinking water (C) or glucocorticoid treatment (GC‐Tx). After 8 weeks of treatment the relative abundances of fecal bacterial communities were determined. n = 5 mice per group.

Depletion of the gut microbiota prevents GC‐induced trabecular bone loss

Knowing that GC treatment significantly alters the mouse gut microbiome, we next tested whether the gut microbiota is necessary for GIO. For this, we treated mice with prednisolone or sham pellet (as in the previous section) with or without continuous oral broad‐spectrum antibiotics (ampicillin [1 g/L] and neomycin [0.5 g/L]) for the duration of the experiment. Control groups included mice that were not treated with prednisolone but treated with antibiotics and mice that did not undergo any treatment. Microbiota depletion in the antibiotic‐treated groups was confirmed by a significant decrease in fecal bacterial CFU (Fig. 2 A). To investigate whether microbiota depletion affected GIO, distal femur metaphyseal and L3 vertebral trabecular bone volume fraction were examined at the 8‐week time point using μCT. Trabecular data was expressed as both BV/TV (Table 1) and BV/TV corrected for body weight (Fig. 2 C,D) because body weight can be altered by GC‐Tx and could contribute to changes in bone volume. As expected, GC‐Tx caused more than a 50% loss of trabecular bone volume in the distal femur (Fig. 2 B,C; Table 1). Remarkably, GC‐Tx of microbiota‐depleted mice (antibiotic‐treated) did not cause femoral bone loss (Fig. 2 B,C). GC‐Tx caused a 20% decrease in vertebral trabecular bone volume, but the change was not statistically significant (Fig. 2 D; Table 1). On the other hand, depletion of the microbiota with antibiotics caused a significant increase in bone density in GC‐Tx mice (Fig. 2 D; GC‐Tx versus ABX + GC‐TX). Microbial depletion for 8 weeks, on its own, had no effect on the bone volume of untreated mice (Fig. 2 B_–_D; control versus ABX). Detailed analyses of trabecular architectural parameters corresponded with changes in femoral bone volume fraction (Fig. 2 E). GC‐Tx did not significantly alter cortical bone parameters (Table 1). These results suggest that the gut microbiota is required for GC‐induced trabecular bone loss, whereas effects on cortical bone changes cannot be determined at this time point.

Depletion of the gut microbiota prevents glucocorticoid induced trabecular bone loss. Sixteen‐week‐old C57BL/6J male mice were split into four experimental groups: Sterile drinking water (C), glucocorticoid treatment (GC‐Tx), continuous broad‐spectrum antibiotics (ABX), or glucocorticoid treatment with continuous broad‐spectrum antibiotics (ABX + GC). (A) Colony forming units (CFU/g) were determined from fecal samples obtained from mice at week 1. (B) Representative iso‐surface images of distal femur metaphyseal trabecular bone. (C, D) μCT analysis of femoral (C) and vertebral (D) trabecular bone volume fraction corrected for body weight. (E) Bone femur microarchitecture analyses. n = 6–10 per group. Box‐plot whiskers represent minimum to maximum levels. Statistical analyses were performed by one‐way ANOVA with Tukey post‐test. ****p < .0001; ***p < .001; **p < .01; *p < .05. BV/TV/BW = bone volume/total volume/body weight; Tb.Sp = trabecular spacing; Tb.Th = trabecular thickness; Tb.N = trabecular number.

ABX Mouse Study: General, Trabecular, and Cortical Parameters

Values are averages ± SE. n = 9–10 per group. Statistical analyses were performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; Ct.Th = cortical thickness; Ct.Ar = cortical area; Ma.Ar = marrow area; Tt.Ar = total area; BMD = bone mineral density; Ec.Pm endocortical perimeter; Ps.Pm = periosteal perimeter; MOI = moment of inertia.

*

p < .05;

****

p < .0001 compared to control (no ABX) mice.

ABX Mouse Study: General, Trabecular, and Cortical Parameters

Values are averages ± SE. n = 9–10 per group. Statistical analyses were performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; Ct.Th = cortical thickness; Ct.Ar = cortical area; Ma.Ar = marrow area; Tt.Ar = total area; BMD = bone mineral density; Ec.Pm endocortical perimeter; Ps.Pm = periosteal perimeter; MOI = moment of inertia.

*

p < .05;

****

p < .0001 compared to control (no ABX) mice.

Probiotic Lactobacillus reuteri 6475 supplementation prevents GC‐induced bone loss

Having established that GC‐Tx alters gut microbiome and that gut microbiota mediates GIO, we next examined if supplementation of gut microbiota with beneficial probiotic bacteria will prevent GIO. For this we chose LR and LGG, which have proven to benefit bone health in humans and mouse models of osteoporosis. Mice were treated with prednisolone or placebo with or without LR or LGG supplementation (3 × 108 CFU/mL drinking water) for the duration of the experiment. Statistical analyses of community structure differences (using ANOSIM) indicated that treatment with the probiotics further shifted the GC‐Tx mouse microbiota to unique compositions (R = 0.944, p < .001; Fig. 3 A). As we found before, μCT imaging of femurs from GC‐Tx mice showed a significant decrease in trabecular bone volume fraction compared to untreated controls when corrected to body weight (Fig. 3 B) or when expressed as BV/TV (Table 2). Supplementation with LR, but not LGG, prevented GC‐Tx–induced bone loss (Fig. 3 B). Like femoral bone volume fraction, femoral trabecular measure (including thickness, number, and spacing) were modulated by GC‐Tx and importantly were prevented with LR supplementation (Fig. 3 C). Vertebral trabecular bone showed a similar response (Fig. 3 D). Control mice (placebo pellets) that were treated with the probiotics did not exhibit any significant bone responses (data not shown). Female mice also benefited from LR supplementation, which prevented GC‐induced trabecular bone loss when corrected for body weight (Fig. 3 E, body weight noted in Fig. 3 F) as well as when expressed as BV/TV (Fig. 3 F).

Probiotic Lactobacillus reuteri 6475 supplementation prevents glucocorticoid‐induced bone loss. Sixteen‐week‐old male C57BL/6J mice were treated with glucocorticoids (prednisolone) for 8 weeks. Mice were either given sterile water (Veh) or supplemented with L. reuteri (LR) or Lactobacillus rhamnosus GG (LGG) in water for duration of experiment. (A) NMDS plot of fecal microbiome, Bray‐Curtis analysis performed (n = 5 mice/group). (B) Representative iso‐surface images of distal femur metaphyseal trabecular bone and quantitation of femoral trabecular bone volume fraction corrected for body weight. (C) Bone femur microarchitecture analyses. (D) Quantitation of vertebral (L3) trabecular bone fraction corrected for body weight. (E) Female femoral and vertebral trabecular bone volume fraction corrected for body weight. (F) Female BV/TV and body weight data. (G) Percentage of CD4+ T lymphocytes within femoral bone marrow. n = 8–10 per group. Box‐plot whiskers represent minimum to maximum levels. Statistical analyses were performed with one‐way ANOVA with Tukey post‐test. ****p < .0001; ***p < .001; **p < .01; *p < .05. BV/TV/BW = bone volume/total volume/body weight; Tb.Sp = trabecular spacing; Tb.Th = trabecular thickness; Tb.N = trabecular number.

Microbiota Modulation Experiment: General and Trabecular Bone Parameter

Values are averages ± SE. n = 9–10 per group. Significant values are bold. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; LR = Lactobacillus reuteri; LGG = Lactobacillus rhamnosus GG.

*p < .05;

**

p < .01;

***p < .001;

****

p < .0001.

Microbiota Modulation Experiment: General and Trabecular Bone Parameter

Values are averages ± SE. n = 9–10 per group. Significant values are bold. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; LR = Lactobacillus reuteri; LGG = Lactobacillus rhamnosus GG.

*p < .05;

**

p < .01;

***p < .001;

****

p < .0001.

To be effective in preventing GIO, a treatment must block GC‐Tx–induced bone loss while maintaining the beneficial immunosuppressive effects of GCs. To examine this in our model, we isolated bone marrow cells from the mice and quantitated the number of CD4+ T‐lymphocytes. Figure 3 G shows that GC‐Tx suppression of CD4+ T‐cells is maintained in all treatment groups. Taken together these data suggest that LR supplementation prevents GC‐induced bone loss without preventing GC‐Tx suppression of CD4+ T‐lymphocytes.

Microbiota composition was significantly different between groups when examined at the level of OTUs and the GC‐Tx bone loss is transferable by fecal transplant

Figure 3 A shows that there are significant differences in microbiota composition “fingerprints” between treatment groups (C, GC‐Tx, GC‐Tx + LR, and GC‐Tx + LGG) as determined by nonmetric multidimensional scaling (NMDS) analyses. Figure 4 A shows differences at the Order level in fecal microbiota composition. In general, as seen in our first study (Fig. 1), we saw that GC‐Tx decreased levels of Verrucomicobiales and Bacteriodales and increased Clostridiales. Using the machine learning, RF classification at the OTU level we found a 100% classification rate into the four separate groups without error, indicating that there were strong signatures of the treatment groups reflected in the microbiome data. The 30 most differential OTUs identified by the RF classification were then plotted according to their taxonomic membership in Supporting Fig. Fig. 1 and noted in Supporting Table 1 (colon) and Supporting Table 2 (fecal). Many of the most varied fecal OTUs were in the Porphyromonadaceae and Clostridiales groups (noted in Fig. 4 B and Supporting Fig. Fig. 1 A, in shades of red and blue, respectively). Colon microbiota analyses also demonstrated strong classification of the treatment groups (Fig. 4 A; Supporting Fig. Fig. 1 A) and differences were primarily in OTUs of the Porphyromonadaceae and Clostridiales groups, with the latter being more abundant in this gastrointestinal (GI) region. To further identify specifically how LR altered the microbiome compared to vehicle‐treated mice, we focused on examining the fecal microbiota for differential OTUs in these groups specifically. Using a variable importance plot, we found that a number of Clostridiales and Porphoromonadaceae OTUs were specifically elevated or suppressed in the fecal microbiota of the LR‐treated group compared to GC treatment alone (Fig. 4 B; Supporting Fig. Fig. 1). Interestingly, LR treatment altered levels of certain Lactobacillaceae OTUs, but at the taxonomy Order level, did not significantly affect the abundance of Lactobacillales (Fig. 4 B).

Microbiota composition differs between treatment groups and promotes bone loss when transplanted to WT mice. (A) Microbiome data profiles at the taxonomic level of order for the different mouse treatment groups. Individual microbiota compositions from five mice per group were averaged and are visualized as stacked bar‐plots of colon or fecal samples. (B) Box‐plots of selected differential OTUs from the fecal samples that were identified by the variable importance plot of the random forest classification and then tested for statistical significance with the Kruskal‐Wallis test and between group significance with the Wilcoxon‐rank sum test. (C, D) Microbiota depleted mice (ABX treated for 1 week) underwent FT by gavaging with fecal microbiota from control (placebo) and GC‐Tx mice (1×/week for 8 weeks). The recipient mice were analyzed for distal femur trabecular bone volume (C) and for distal femur trabecular microarchitecture (D). n = 9–10. Statistical significance determined by t test. *p < .05, **p < .01, ***p < .001. FT = fecal transplant.

To test the functionality of the microbiota, we performed fecal transplants into post–ABX‐treated male C57BL/6J mice. Following 1 week of broad‐spectrum ABX treatment, the mice were gavaged weekly with a pooled fecal slurry that contained feces from control or GC‐Tx mice. This approach allows repopulation of the gut with the fecal microbiota. After 8 weeks, mice were euthanized, and femur trabecular bone analyzed. As shown in Fig. 4 C, the GC‐Tx mouse fecal transplant significantly reduced the BV/TV of the recipient untreated mice when compared to fecal transplant from placebo control donor mice. Trabecular microarchitecture analyses demonstrated a significant increase in trabecular separation in the GC‐Tx fecal transplanted mice (Fig. 4 D). Control fecal transplant mice had bone parameters similar to untouched control mice (data not shown). This finding supports a role for the GC‐Tx dysbiotic microbiota in contributing to GIO.

Probiotics do not affect GC‐Tx–induced changes in cortical and mechanical strength properties

Because long‐term GC‐Tx alters whole‐bone cortical and mechanical strength properties in humans and animal models,13, 30, 79, 80, 81 we investigated whether LR or LGG affected GC‐induced cortical, structural, or tissue‐level properties of mouse bones. Although GC treatment significantly increased marrow area and endocortical perimeter compared to control, this change was not prevented with probiotic supplementation (Table 3). We did not observe a change across treatment groups in moment of inertia (Table 3) or in structural‐level mechanical properties (Fig. 5 A). Analysis of tissue‐level mechanical properties, which estimate material properties of bone, again revealed no significant change among treatment groups (Fig. 5 B). These results suggest that at the time point that we examined, neither GC treatment nor probiotic treatments significantly alter bone mechanical properties.

Microbiota Modulation Experiments: Cortical Bone Parameters

Values are averages ± SE. n = 9–10 per group. Significant values are bold compared to controls. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; LR = Lactobacillus reuteri; LGG = Lactobacillus rhamnosus GG; MDY = high molecular weight polymer; Ct.Th = cortical thickness; Ct.Ar = cortical area; Ma.Ar = marrow area; Tt.Ar = total area; BMD = bone mineral density; Ec.Pm endocortical perimeter; Ps.Pm = periosteal perimeter; MOI = moment of inertia.

*

p < .05;

**

p < .01;

***

p < .001;

****

p < .0001 compared to controls.

Microbiota Modulation Experiments: Cortical Bone Parameters

Values are averages ± SE. n = 9–10 per group. Significant values are bold compared to controls. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; LR = Lactobacillus reuteri; LGG = Lactobacillus rhamnosus GG; MDY = high molecular weight polymer; Ct.Th = cortical thickness; Ct.Ar = cortical area; Ma.Ar = marrow area; Tt.Ar = total area; BMD = bone mineral density; Ec.Pm endocortical perimeter; Ps.Pm = periosteal perimeter; MOI = moment of inertia.

*

p < .05;

**

p < .01;

***

p < .001;

****

p < .0001 compared to controls.

Probiotic treatment did not affect GC‐Tx induced changes in cortical bone mechanical strength properties. Sixteen‐week‐old male C57BL/6J mice were treated with glucocorticoids (prednisolone) for 8 weeks. Mice were either given sterile water (Veh) or supplemented with LR or LGG in water for duration of experiment. Analyses of femoral cortical bone (A) structural and (B) tissue level properties. n = 9–10 per group statistical analysis was performed with one‐way ANOVA. LR = L. reuteri; LGG = Lactobacillus rhamnosus GG.

Barrier dysfunction mediates GC‐induced bone loss

Intestinal barrier disruption and endotoxin leakage into the blood stream are now recognized as important pathogenic events in a number of chronic diseases.100, 101, 102 Importantly, we recently identified intestinal barrier dysfunction as a key mediator of dysbiosis‐induced bone loss.68 Because our results so far indicate that gut microbiota is altered and necessary for GIO, we tested whether GC‐Tx affects barrier function and if strengthening the intestinal barrier can protect mice from GIO. Specifically, mice were treated with or without prednisolone in the presence or absence of MDY,68, 78 a high molecular weight barrier‐enhancing polymer, for the duration of the experiment. As shown in Fig. 6 A, GC‐Tx increased intestinal permeability (barrier leaks) as evidenced by increased levels of endotoxin in the serum. More importantly, MDY treatment effectively prevented the GC‐Tx–induced elevation of serum endotoxin and correspondingly prevented femoral trabecular bone loss induced by GC treatment (BV/TV/BW in Fig. 6 B,C; and BV/TV in Table 4) as well as vertebral bone volume (Table 4), but levels did not reach significance when vertebral bone volume was corrected to body weight (Fig. 6 C). Femoral trabecular architectural parameters corresponded with the femoral bone volume fraction (Fig. 6 D). Pearson's correlation analyses identified a negative correlation between serum endotoxin levels and femoral BV/TV (r = −0.4995, p = .0001; Fig. 6 E). These results suggest that GC‐induced barrier leaks are an important mediator of trabecular bone loss in the GIO model.

Barrier dysfunction mediates glucocorticoid‐induced bone loss. Sixteen‐week‐old male C57Bl/6J mice were treated with glucocorticoids (prednisolone) for 8 weeks. Mice received either sterile water (Veh) or water supplemented with high molecular weight polymer (MDY). (A) Levels of serum endotoxin expressed as fold‐change relative to controls as noted in the Tukey box plot. (B) Representative isosurface images of the metaphyseal trabecular bone of the distal femur. (C,D) μCT analysis of femoral and vertebral trabecular bone volume fraction corrected for body weight and trabeculae parameters. (E) Pearson's correlation plot of mouse serum endotoxin versus BV/TV. Bar graph values (n = 9–10 per group) are average ± SE; μCT box plot whiskers represent minimum to maximum levels. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test. ****p < .0001 ***p < .001; **p < .01; *p < .05.

MDY Study: General and Trabecular Parameters

Values are averages ± SE. n = 9–10 per group. Significant values are bold compared to controls. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; MDY = high molecular weight polymer.

*

p < .05;

****

p < .0001.

MDY Study: General and Trabecular Parameters

Values are averages ± SE. n = 9–10 per group. Significant values are bold compared to controls. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test.

GC‐Tx = glucocorticoid‐treated; MDY = high molecular weight polymer.

*

p < .05;

****

p < .0001.

To identify if chronic antibiotic and LR treatments, from the experiments noted in Figs. 2 and 3, respectively, also enhanced intestinal barrier function we measured serum endotoxin levels. Consistent with the role of barrier leaks in regulating bone health in the GIO model, LR and chronic antibiotic treatments prevented GC‐induced barrier leaks (Fig. 6 A). Together, these data show that the GC‐Tx–altered barrier dysfunction (and therefore barrier leaks) is a key pathogenic event in GIO.

Lactobacillus reuteri and MDY prevent GC suppression of osteoblast activity

To determine whether the probiotic and barrier enhancement treatments affected anabolic and/or catabolic bone parameters, markers of osteoblast and osteoclast activity were measured. Dynamic anabolic bone measures, mineral apposition rate (MAR), and bone formation rate (BFR), were significantly decreased in GC‐Tx mice, whereas both LR and MDY treatments reduced the suppression (Fig. 7 A). Analyses of distal trabecular osteoblast surface did not show significant differences between groups at this time point (Fig. 7 B). Given that osteoblast and adipocyte numbers are often reciprocally related (because they are derived from a common mesenchymal stem cell103), we analyzed marrow adiposity in the bone metaphyseal region. GC‐Tx increased the number of bone marrow adipocytes and both LR and MDY treatments prevented the adiposity (Fig. 7 C). Analyses of catabolic bone parameters, such as osteoclast surface, indicated that GC‐Tx tended to increase osteoclast surface and LR significantly prevented this change (Fig. 7 D). Together, these results suggest that prevention of GIO by LR and MDY is the result of retaining anabolic bone activity and reducing catabolic activity under GC treatment conditions.

LR and MDY prevent glucocorticoid suppression of osteoblast activity. (A) Quantitation of trabecular bone MAR and BFR. (B) Quantification of osteoblast surface/total bone surface in distal trabecular femur region. (C) Representative histological adipocyte images of distal femur at 10× magnification; number of adipocytes in the marrow area of the distal femur. (D) Quantification of osteoclast surface/total bone surface in distal trabecular bone region. n = 5 per group. Box plot whiskers represent minimum to maximum levels. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test. ****p < .0001; ***p < .001; **p < .01; *p < .05. MAR = mineral apposition rate; BFR = bone formation rate.

Role for microbiota in GC‐Tx–induced osteoblast and osteocyte apoptosis

Our data thus far support the role of gut microbiota and barrier dysfunction in mediating GC suppression of osteoblast activity and GIO. Next, we wanted to further understand the cellular mechanisms of intestinal microbiota prevention of GIO. We focused on examining osteoblast and osteocyte apoptosis, which is known to be increased by GC treatment,4, 15, 21, 104, 105 and we tested if chronic antibiotic treatment of GC‐treated mice, described in Fig 2, could alter bone cell death. Tibial RNA analyses showed that the GC induction of the BAX/BCL‐2 expression ratio (pro‐apoptotic/anti‐apoptotic; elevation is an indicator of apoptosis) is prevented by microbiota depletion (Fig. 8 A). To identify specific responses of osteoblast and osteocyte death, femur sections were TUNEL‐stained. Figure 8 B_–_D shows that GC treatment increases both osteoblast and osteocyte (trabecular and cortical) death. Importantly, under microbiota‐depleted conditions GC treatment does not increase apoptosis compared to ABX control groups. These findings support involvement of the microbiota in mediating GC‐induced osteoblast and osteocyte death and ultimately GIO.

Microbiota mediates GC‐induced osteoblast and osteocyte apoptosis. (A) Ratio of pro‐apoptotic (BAX) and anti‐apoptotic (BCL‐2) RNA levels isolated from whole tibia (n = 5 per group, gray dot is an excluded point based on rout outlier test). (B_–_D) Percentage of TUNEL‐positive–stained osteoblast and osteocytes in femoral trabecular bone, as well as femoral cortical osteocytes. Box‐plot whiskers represent minimum to maximum levels. n = 9–10 per group, 5 per group for cortical analyses. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test. ***p < .001; **p < .01; *p < .05.

LR and MDY prevent GC‐induced osteoblast and osteocyte apoptosis

Next, we examined if LR (probiotic) and MDY (barrier enhancer) treatments also prevented GC‐induced osteoblast and osteocyte apoptosis. Consistent with inhibiting bone loss, both LR and MDY prevented GC‐Tx elevation of BAX/BCL‐2 expression ratio (Fig. 9 A). Similarly, both treatments reduced the number of TUNEL‐positive osteoblasts and osteocytes (Fig. 9 B_–_D). Pearson's correlational analyses revealed that TUNEL‐positive trabecular osteoblasts (R = 0.4456, p = .004) and osteocytes (R = 0.4022, p = .01) correlated with serum endotoxin levels (Fig. 9 E). Together, these results indicate that enhanced barrier function as well as supplementation with oral LR reduces osteoblast and osteocyte apoptosis, which likely contributes to preventing GC suppression of bone formation and GIO.

LR and MDY prevent GC‐induced osteoblast and osteocyte apoptosis. (A) Gene expression levels and ratio of pro‐apoptotic (BAX) and anti‐apoptotic (BCL‐2) in whole bone tibia. (B–D) Percentage of TUNEL‐positive–stained osteoblast and osteocytes in femoral trabecular bone, as well as femoral cortical osteocytes. (E) Pearson's correlation plots of serum endotoxin versus TUNEL‐positive trabecular osteoblasts and osteocytes. Box‐plot whiskers represent minimum to maximum levels. n = 5 (MDY treatment); 9 per group. The gray point represents data that was excluded by ROUT outlier analysis. Statistical analyses were performed with one‐way ANOVA with Tukey post‐test. ****p < .0001; **p < .01; *p < .05.

Role for Wnt10b in L. reuteri and MDY prevention of GIO

The Wnt/β‐catenin signaling pathway is a key regulator of anabolic bone activity. Elevated Wnt10b expression promotes osteogenesis as well as osteoblast differentiation and viability.11, 23, 24, 25, 26, 96, 106 Given that GC treatment is known to suppress Wnt10b expression,30, 106 we examined if LR and MDY treatments prevent the negative effects of GC on osteoblasts through regulation of Wnt10b. Indeed, Fig. 10 A shows that both LR and MDY treatment prevent the marked suppression of Wnt10b expression by GC treatment. These data suggest that the suppression of Wnt10b by GC is a critical component in mediating GIO. To test this, mice with targeted osteoblast Wnt10b overexpression (osteocalcin [OC]‐Wnt10b transgenic [TG] mice, created by fusing the Wnt10b gene with the OC promoter to target Wnt10b expression to osteoblasts27) underwent the standard GC treatment (or vehicle) protocol. After 8 weeks, analyses of femoral trabecular bone volume fraction revealed that in contrast to the bone loss seen in GC‐Tx WT mice, Wnt10b overexpression prevents bone loss following GC treatment (Fig. 10 B). No significant differences in body weight were seen (Fig. 10 C). Trabecular architectural measures corresponded with the changes in bone volume (Fig. 10 D). Taken together these data support the key role of Wnt10b suppression in mediating GIO and demonstrate that both LR and MDY are able to prevent GC‐mediated Wnt10b suppression and bone loss.

Role for Wnt10b expression in regulating GIO. (A) Wnt10b expression levels in RNA isolated from whole bone tibia. (B–E) Sixteen‐week‐old male WT and Wnt10b‐TG mice were treated with either placebo or prednisolone pellet for 8 weeks as in previous experiments. (B) Distal femur trabecular bone volume/total volume (n = 5, 5, 6, 6). (C) Mouse body weights at 24 weeks of age. (D) Bone femur microarchitecture analyses. Box‐plot whiskers represent minimum to maximum levels. Statistical analysis was performed with one‐way ANOVA with Tukey post‐test. *p < .05. BV/TV = bone volume/total volume; Tb.Sp = trabecular spacing; Tb.Th = trabecular thickness; Tb.N = trabecular number.

Discussion

GIO is the most common cause of secondary osteoporosis and results in bone fractures in more than 30% of patients undergoing chronic prednisolone treatment.1, 3 GCs reduce Wnt10b expression and increase marrow adiposity and osteoblast and osteocyte apoptosis.15, 23, 24, 104, 107, 108, 109, 110 Our findings are the first to identify the gut microbiota and barrier function as mediators/regulators of GIO. Specifically, depleting the microbiota with chronic antibiotic treatment, modifying the microbiota with oral supplementation of a probiotic (L. reuteri 6475, to alter the microbiota toward a beneficial balance) or directly inhibiting gut barrier leakage with a mucus supplement (MDY) significantly blunts GC‐induced trabecular bone loss. We further demonstrate that the fecal microbiota from the GC‐Tx mice has bone loss activity as demonstrated by its ability to cause bone loss in untreated recipient mice. Finally, we show that manipulation of the gut microbiota or barrier function regulates GC effects on Wnt10b, mesenchymal cell lineage selection, and osteoblast and osteocyte apoptosis.

Our finding that chronic depletion of the microbiota with long‐term treatment of broad‐spectrum antibiotics prevented GIO, provides significant support for the role of the microbiota in mediating GIO. In particular we found that continuous ABX administration prevented the GC‐increase in serum endotoxin levels. This is consistent with other reports showing that ABX reduces microbial load and serum endotoxin levels.111, 112 It should be noted that effects on bone density are markedly different between microbiota depletion by chronic antibiotic treatment (long‐term) versus microbiota repopulation induced by short‐term antibiotic treatment. As we note here, we did not observe any direct effect of chronic ABX treatment on bone density (control versus ABX) similar to Guss and colleagues.69 This is distinct from our recent observation that short‐term ABX treatment followed by natural microbiota repopulation for 4 weeks is associated with significant dysbiosis, elevated serum endotoxin levels, and bone loss.68 Interestingly, in the post‐ABX model as well as in other bone models (including H. hepaticus, L. reuteri, etc.), we observed sex‐specific differences characterized by male mice displaying bone responsiveness to gut changes, but not female mice.54, 68, 70 This was especially evident in the effects of LR, which enhanced bone density in healthy male but not female mice (unless female mice were estrogen deficient or mildly inflamed). In the present GIO model, however, LR was able to prevent trabecular bone loss in both males and females, suggesting distinct mechanisms of actions of LR in different bone models. The finding that estrogen‐deficient mice or intact female mice with inflammation display a response to gut microbiota modulation54, 75 suggests critical roles for estrogen and inflammation in mediating sex differences. How these factors interplay with actions of LR in the GIO model will be addressed in future studies. Furthermore, clinical studies will be necessary to confirm the translatability of these findings.

GCs have been shown to alter the composition of gut microbes, which are linked to changes in gut and brain function.31, 32, 102, 113 For example, stress, which elevates serum corticosteroid levels, alters the microbiome and reduces levels of intestinal lactobacilli and increases the growth of E. coli and pseudomonas.102, 113 Stress also increases intestinal bacterial expression of virulence genes.102 Furthermore, GC treatment can reshape the microbiome in inflammatory bowel disease.31 In the current study, GC treatment was shown to cause a significant overall shift in microbial composition as evidenced through nonmetric multidimensional scaling (NMDS) and RF analyses. Our findings, in multiple experiments (Figs. 1 and 4) showed an overall increase in the relative abundance of Clostridiales and a decrease in Bacteroidales and Verruomicrobiales groups. Our changes are consistent with published studies that examined the microbiota following GC‐Tx in mice31 and more importantly similar changes were observed in GC‐Tx humans.114 The latter indicates the potential for translation of our findings to humans. These studies are particularly interesting because different GC molecules were used (prednisolone, prednisone or dexamethasone), indicating that GC treatment in general may have a reliable effect on fecal bacterial groups and microbial dysbiosis. Most importantly, using fecal transplants we were able to show that the GC‐Tx mouse fecal material functionally reduces trabecular bone volume in naive/untreated recipient mice. In our study we treated recipient mice with broad spectrum antibiotics (poorly absorbed) for 1 week to reduce intestinal bacterial load prior to fecal transplantation. A number of studies have used this approach to assess the functional effects of microbiota/fecal transplants.115, 116, 117 Although other studies have used germ‐free mice as recipient of fecal transplants, we opted to use the antibiotic depletion approach to overcome the dysregulated immune system in the germ‐free mice.

Although the differences between our specific GC treatment groups and the probiotics were not as notable at the Order level, at the OTU level many members of the Clostridiales and Porphyromonadaceae groups were elevated in the different treatments. Specific OTUs from both groups were found to be unique to the LR‐treated GC mouse group, indicating the positive effects on bone density may be mediated by a probiotic mechanism from LR that alters specific members of the Clostridia and the Porphyromonadaceae lineage of the Bacteroidetes phylum. It should be noted that our studies compared the microbiota of age‐matched control versus treated mice. It would also be interesting to examine the microbiota pretreatment versus posttreatment. However, this analysis would require individual housing of mice (to prevent influences of coprophagic mixing of mouse microbiotas), which in itself could increase stress hormone levels in the mice and influence the microbiota composition. Teasing out the specific role of the microbiota composition in GIO is an ongoing focus of our laboratory.

Alterations in the gut microbiota that reduce beneficial bacteria and increase unhealthy bacteria can promote intestinal permeability and increase serum endotoxin; this has been proposed as a mechanism by which the gut microbiome influences bone health.50, 68 Consistent with this proposed mechanism, alteration of the microbiota by GC was accompanied by an increase in serum endotoxin levels, suggesting decreased barrier function. In line with previous studies showing that direct induction of intestinal barrier breaks, through treatment of mice with dextran sodium sulfate (DSS, a detergent), causes bone loss,118 GC treatment and barrier dysfunction in the present study was also associated with bone loss. Importantly, directly enhancing barrier function in GC‐treated mice prevented GIO, showing a strong link between GC's effect on microbiota, barrier function (and therefore endotoxin increase), and bone loss. Consistent with this, serum endotoxin significantly correlates (negatively) with trabecular bone density, further suggesting a strong link between barrier dysfunction and bone health. Beneficial bone effects of enhancing barrier function with MDY have been demonstrated in two other models of bone loss: in broiler chickens infected with intestinal salmonella and in post‐ABX microbial dysbiosis in mice, enhancing barrier function with MDY prevented trabecular bone loss.68, 119 Because MDY is not absorbed, its benefits on bone health in this model are a consequence of its effects on the intestine, thereby underscoring the importance of the gut–bone signaling axis as a therapeutic target for GIO.

Clinical studies have begun to examine the impact of GCs in combination with probiotics on intestinal inflammation. In a study using 83 patients with Crohn's disease, patients were treated with prednisolone and Bifidobacterium Lactobacillus tablets.120 Compared to untreated patients, the treatment group reduced inflammation and improved the intestinal flora (decreasing yeast and increasing lactobacillus). However, the study did not contain a GC‐alone treatment group. In a previous study, prednisolone‐treated patients were split into two groups one receiving BIFICO probiotic tablets (containing a mixture of Enterococcus faecalis, Bifidobacterium longum, and Lactobacillus acidophilus) and the other receiving starch tablets (vehicle). The study suggests that the probiotic treatment elevated IL‐10 and reduced patient relapse.121 In addition to altering the gut microbiome and reducing intestinal inflammation, probiotics are known to promote intestinal barrier function and benefit bone health.50, 53, 54, 55, 56, 57, 63, 67, 68, 71, 73, 75, 76, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132 Consistent with these studies, we found that treatment with LR as well as LGG uniquely altered the microbiota composition, which was significantly different from mice treated with GC alone. We hypothesize that each of these compositions have a different balance of healthy versus disease‐promoting bacteria, which could be playing a role in its differential effects on the intestinal barrier and furthermore bone health. In the present study, only LR had a beneficial effect on barrier function and bone health. LR's ability to strengthen the intestinal barrier and benefit bone could be due to its ability to uniquely modify the microbiota, leading to increased bacterial species that prevent GIO or through its ability to produce biologically active metabolites or proteins that are specifically effective and beneficial in the GIO model.54, 133, 134 LGG is shown to effectively prevent bone loss in another mouse model and has been suggested to increase Clostridia and butyrate levels.71 In our studies, LGG treatment significantly altered the microbiota compared to other treatment groups. We also observed some Clostridiales OTUs that were uniquely elevated in the LGG treatment group. However, these were not linked to bone effects in this model.

The Lactobacilli strains LGG and LR are both used as probiotics to benefit the health of mice and humans.126 Both LR (L. reuteri 6475) and LGG are human‐derived strains that do not (in the case of L. reuteri) or are not expected to (in the case of LGG) colonize the mouse gut. Because we did not disrupt the microbiome prior to addition of the probiotics, it is unlikely that the probiotics colonized the gut. In the case of L. reuteri, human‐derived strains are significantly outcompeted by mouse strains and lack key colonization factors for long‐term colonization of the mouse.135, 136 In contrast to LR, LGG has pili that provide it with mucus adhering properties which may contribute to extend its time in the GI tract.137 Despite reduced colonization, our laboratory and many other laboratories have shown that oral treatment with probiotics which do not readily colonize the gut can have significant effects on gut physiology and health. It is thought that during their transit probiotics secrete/release factors (antimicrobials, metabolites, exopolysaccharides, and unmethylated CpG‐rich DNAs) which modify the microbiota and affect the gut barrier and immune system.137, 138 We have shown that live LR is needed for modulating lymphocyte cytokine production.139 The varied factors produced by probiotics (in response to other bacteria, nutrient availability, oxygen, etc.) likely contribute to the targeting of different signaling pathways and tissues and different probiotic effects. Future studies will examine dosing and diet contributions, which could explain differences between LR and LGG.

Our studies demonstrate that GC treatment has a significant effect on both trabecular and cortical bone parameters: trabecular bone volume and in some groups cortical thickness decreased and marrow area as well as endocortical perimeter and in some cases periosteal perimeter increased. Previous studies also report that GC‐Tx decreases cortical thickness and increases marrow area.140, 141 Reports have also reported increased endocortical perimeter.141 However, reports on GC‐Tx effects on periosteal bone perimeter or formation have been variable,141, 142 in part because of differences in models, the bone analyzed, and animal age. Although our studies show a decrease in bone formation, there may be low‐level formation at the periosteum while there is ongoing significant bone loss at the endosteum. The latter may contribute to the lack of structural strength changes. Modification of the microbiota by LR, ABX, and MDY treatments prevented GC‐induced trabecular bone loss. Interestingly, GC‐induced changes in cortical bone microarchitecture were not altered by the treatments that modified the microbiota or barrier. This could be due to cortical bone being less metabolically active than trabecular bone. The resulting slow remodeling may not respond robustly to the gut changes induced by treatments or may require additional time. Support for the latter is suggested by the observed LR suppression of osteocyte TUNEL staining in cortical bone of GC‐treated mice; one would anticipate cortical bone benefits in a longer experimental time course.

Past studies indicate that endogenous GCs induce bone loss via rapid bone resorption, followed by prolonged and profound suppression of bone formation as a result of decreased osteoblast lineage selection and viability.4, 9, 10 In our studies we saw that the GC reduction in anabolic bone remodeling processes was prevented by treatment with LR and MDY in our 8‐week mouse model. This is consistent with previous reports showing that LR and MDY treatment can increase mineral apposition, bone formation, osteoblast number, and serum osteocalcin.53, 54, 68, 76 For the first time, we show that depletion of the microbiota with ABX, modifying the microbiome with LR treatment, or enhancing barrier function with MDY treatment prevents GC‐induced osteoblast and osteocyte apoptosis. As noted previously, these gut treatments prevented increases to serum endotoxin, which has been shown to induce osteoblast apoptosis.143

Wnt10b is known to be an important regulator of bone density and health.27, 29, 76, 144, 145, 146 We observed that GC‐Tx decreases WNT10b expression and promotes bone cell apoptosis and marrow adipogenesis, consistent with several reports.24, 147, 148 Our studies further show that treatment with LR and MDY prevented GC‐Tx suppression of Wnt10b and prevented GC‐induced bone marrow adiposity. Treatment with LR has been previously shown to block suppression of WNT10b in a type 1 diabetic mouse model where marrow adiposity and bone loss were also prevented76 and LGG was shown to enhance wnt10b expression in healthy mice.71 In our studies, we did not see LGG enhancement of wnt10b expression but did see some benefits of LGG on bone, but they did not reach significance. In further support of a key role for suppression of WNT10b signaling in mediating GIO, we found that OC‐WNT10b TG mice treated with GCs did not display trabecular bone loss. A role for reduced Wnt signaling and osteoblast/osteocyte apoptosis in GIO has also been suggested by other studies including reports demonstrating that enhancing Wnt signaling by suppressing Wnt inhibitors (eg, knockout of SOST or knockdown of dkk) prevents bone loss and/or cell death in response to GC treatment.106, 109, 110

Although it is generally accepted that GCs induce bone loss via direct effects on bone remodeling cells, our studies highlight the importance of the gut microbiome and intestinal barrier function in GIO. Our studies support a model where GC‐Tx causes intestinal dysbiosis, which is linked with intestinal barrier breaks, elevation of serum endotoxin, and subsequently promotes the suppression of Wnt10b to cause marrow adiposity, osteoblast and osteocyte apoptosis, and GIO. By treating the intestinal dysfunction, we were able to suppress GIO. Direct and indirect effects of GC‐Tx on bone are not mutually exclusive and likely work together to determine the final outcome. Similar to our findings, the gut microbiota has been recently demonstrated to be required for PTH anabolic processes149 as well as bone loss in sickle cell disease.150 In summary, our studies using a variety of approaches implicate the gut as a therapeutic target to reduce bone loss resulting from chronic pharmacologic GC treatment and possibly chronic endogenous serum GC elevation associated with aging12 and anorexia.14 Unfortunately, patients treated with chronic GCs are often not monitored for bone loss; in fact, less than 6% of women under 50 years taking glucocorticoids are monitored for bone loss,3 in part because of a lack of GIO risk awareness but also because of an unwillingness to take additional medications to prevent the bone loss. Discovering the connection between the gut microbiota and bone health can accelerate identification for new treatments not only for GIO but also for osteoporosis all together.

Disclosures

LRM and RAB are inventors on a series of patent and patent applications for the use of Lactobacillus reuteri 6475 for bone health, for which MSU has exclusively licensed its rights to BioGaia, Stockholm, Sweden. LRM is an inventor on a series of patents and patent applications for the use of MDY for bone health, for which MSU has exclusively licensed its rights to Midway Pharmaceuticals, Philadelphia, Pennsylvania, USA.

Acknowledgments

These studies were supported by funding from the National Institutes of Health, grants RO1 DK101050 and AT007695, and by Michigan State University (MSU) Foundation. We thank the Investigative Histology Laboratory in the Department of Physiology, Division of Human Pathology and the Biomedical Imaging Center at Michigan State University for their assistance with histology and imaging, respectively. We also thank Dr. Drhuv Sharma, Michigan State University Center for Statistical Training and Consultation, for consultation on statistical approaches and analyses and Ormond MacDougald for his generous contribution of the Wnt10 transgenic mice. We are also grateful to the staff of Campus Animal Resources for the excellent care of our animals. Results were presented at American Society of Bone and Mineral Research 2017151 and 2019 and Experimental Biology 2019 conferences.152

Authors’ roles: JS, FC, JG, HK and NR obtained data used in the manuscript. JG performed strength testing. RR and RQ carried out analyses of microbiota using RF analyses. RB and LS obtained 16S microbiota data. All authors were involved in data analyses and contributed to the writing of the manuscript. NP and LM oversaw the experimental design, data analyses, and writing.

References

Overman

RA

,

Yeh

JY

,

Deal

CL

.

Prevalence of oral glucocorticoid usage in the United States: a general population perspective

.

Arthritis Care Res (Hoboken)

.

2013

;

65

(

2

):

294

–

8

.

Lekamwasam

S

,

Adachi

JD

,

Agnusdei

D

, et al. .

A framework for the development of guidelines for the management of glucocorticoid‐induced osteoporosis

.

Osteoporos Int.

2012

;

23

(

9

):

2257

–

76

.

Fraser

LA

,

Adachi

JD

.

Glucocorticoid‐induced osteoporosis: treatment update and review

.

Ther Adv Musculoskelet Dis

.

2009

;

1

(

2

):

71

–

85

.

Canalis

E

,

Mazziotti

G

,

Giustina

A

,

Bilezikian

JP

.

Glucocorticoid‐induced osteoporosis: pathophysiology and therapy

.

Osteoporos Int

.

2007

;

18

(

10

):

1319

–

28

.

Lane

NE

.

Glucocorticoid‐induced osteoporosis: new insights into the pathophysiology and treatments

.

Curr Osteoporos Rep

.

2019

;

17

(

1

):

1

–

7

.

Compston

J

.

Glucocorticoid‐induced osteoporosis: an update

.

Endocrine

.

2018

;

61

(

1

):

7

–

16

.

Yao

W

,

Cheng

Z

,

Pham

A

, et al. .

Glucocorticoid‐induced bone loss in mice can be reversed by the actions of parathyroid hormone and risedronate on different pathways for bone formation and mineralization

.

Arthritis Rheum

.

2008

;

58

(

11

):

3485

–

97

.

Khosla

S

,

Shane

E

.

A crisis in the treatment of osteoporosis

.

J Bone Miner Res

.

2016

;

31

(

8

):

1485

–

7

.

Hartmann

K

,

Koenen

M

,

Schauer

S

, et al. .

Molecular actions of glucocorticoids in cartilage and bone during Health, disease, and steroid therapy

.

Physiol Rev

.

2016

;

96

(

2

):

409

–

47

.

Hofbauer

LC

,

Rauner

M

.

Minireview: live and let die: molecular effects of glucocorticoids on bone cells

.

Mol Endocrinol

.

2009

;

23

(

10

):

1525

–

31

.

Canalis

E

,

Delany

AM

.

Mechanisms of glucocorticoid action in bone

.

Ann N Y Acad Sci

.

2002

;

966

:

73

–

81

.

Purnell

JQ

,

Brandon

DD

,

Isabelle

LM

,

Loriaux

DL

,

Samuels

MH

.

Association of 24‐hour cortisol production rates, cortisol‐binding globulin, and plasma‐free cortisol levels with body composition, leptin levels, and aging in adult men and women

.

J Clin Endocrinol Metab

.

2004

;

89

(

1

):

281

–

7

.

Lavretsky

H

,

Newhouse

PA

.

Stress, inflammation, and aging

.

Am J Geriatr Psychiatry

.

2012

;

20

(

9

):

729

–

33

.

Lovdel

A

,

Suchacki

KJ

,

Sulston

RJ

, et al. .

Investigating glucocorticoids as mediators of increased bone marrow adiposity during caloric restriction

.

Endocrine Abstracts

.

2018

;

56

:

GP58

. https://doi.org/10.1530/endoabs.56.GP58.

Gruver‐Yates

AL

,

Cidlowski

JA

.

Tissue‐specific actions of glucocorticoids on apoptosis: a double‐edged sword

.

Cells

.

2013

;

2

(

2

):

202

–

23

.

Silvestrini

G

,

Ballanti

P

,

Patacchioli

FR

, et al. .

Evaluation of apoptosis and the glucocorticoid receptor in the cartilage growth plate and metaphyseal bone cells of rats after high‐dose treatment with corticosterone

.

Bone

.

2000

;

26

(

1

):

33

–

42

.

Liu

Y

,

Porta

A

,

Peng

X

, et al. .

Prevention of glucocorticoid‐induced apoptosis in osteocytes and osteoblasts by calbindin‐D28k

.

J Bone Miner Res.

2004

;

19

(

3

):

479

–

90

.

Weinstein

RS

.

Clinical practice. Glucocorticoid‐induced bone disease

.

N Engl J Med.

2011

;

365

(

1

):

62

–

70

.

van Staa

TP

.

The pathogenesis, epidemiology and management of glucocorticoid‐induced osteoporosis

.

Calcif Tissue Int

.

2006

;

79

(

3

):

129

–

37

.

Weinstein

RS

,

Nicholas

RW

,

Manolagas

SC

.

Apoptosis of osteocytes in glucocorticoid‐induced osteonecrosis of the hip

.

J Clin Endocrinol Metab

.

2000

;

85

(

8

):

2907

–

12

.

Plotkin

LI

,

Weinstein

RS

,

Parfitt

AM

,

Roberson

PK

,

Manolagas

SC

,

Bellido

T

.

Prevention of osteocyte and osteoblast apoptosis by bisphosphonates and calcitonin

.

J Clin Invest

.

1999

;

104

(

10

):

1363

–

74

.

Sato

AY

,

Peacock

M

,

Bellido

T

.

Glucocorticoid excess in bone and muscle

.

Clin Rev Bone Miner Metab

.

2018

;

16

(

1

):

33

–

47

.

Pereira

RC

,

Delany

AM

,

Canalis

E

.

Effects of cortisol and bone morphogenetic protein‐2 on stromal cell differentiation: correlation with CCAAT‐enhancer binding protein expression

.

Bone

.

2002

;

30

(

5

):

685

–

91

.

Ito

S

,

Suzuki

N

,

Kato

S

,

Takahashi

T

,

Takagi

M

.

Glucocorticoids induce the differentiation of a mesenchymal progenitor cell line, ROB‐C26 into adipocytes and osteoblasts, but fail to induce terminal osteoblast differentiation

.

Bone

.

2007

;

40

(

1

):

84

–

92

.

Pereira

RC

,

Delany

AM

,

Canalis

E

.

CCAAT/enhancer binding protein homologous protein (DDIT3) induces osteoblastic cell differentiation

.

Endocrinology

.

2004

;

145

(

4

):

1952

–

60

.

Rauch

A

,

Seitz

S

,

Baschant

U

, et al. .

Glucocorticoids suppress bone formation by attenuating osteoblast differentiation via the monomeric glucocorticoid receptor

.

Cell Metab.

2010

;

11

(

6

):

517

–

31

.

Bennett

CN

,

Ouyang

H

,

Ma

YL

, et al. .

Wnt10b increases postnatal bone formation by enhancing osteoblast differentiation

.

J Bone Miner Res.

2007

;

22

(

12

):

1924

–

32

.

Smith

E

,

Redman

RA

,

Logg

CR

,

Coetzee

GA

,

Kasahara

N

,

Frenkel

B

.

Glucocorticoids inhibit developmental stage‐specific osteoblast cell cycle. Dissociation of cyclin A‐cyclin‐dependent kinase 2 from E2F4‐p130 complexes

.

J Biol Chem.

2000

;

275

(

26

):

19992

–

20001

.

Bennett

CN

,

Longo

KA

,

Wright

WS

, et al. .

Regulation of osteoblastogenesis and bone mass by Wnt10b

.

Proc Natl Acad Sci U S A

.

2005

;

102

:

3324

–

9

,

9

.

Ohnaka

K

,

Tanabe

M

,

Kawate

H

,

Nawata

H

,

Takayanagi

R

.

Glucocorticoid suppresses the canonical Wnt signal in cultured human osteoblasts

.

Biochem Biophys Res Commun.

2005

;

329

(

1

):

177

–

81

.

Huang

EY

,

Inoue

T

,

Leone

VA

, et al. .

Using corticosteroids to reshape the gut microbiome: implications for inflammatory bowel diseases

.

Inflamm Bowel Dis

.

2015

;

21

(

5

):

963

–

72

.

Noguera

JC

,

Aira

M

,

Perez‐Losada

M

,

Dominguez

J

,

Velando

A

.

Glucocorticoids modulate gastrointestinal microbiome in a wild bird

.

R Soc Open Sci

.

2018

;

5

(

4

):

171743

.

Human Microbiome Project C

.

Structure, function and diversity of the healthy human microbiome

.

Nature

.

2012

;

486

(

7402

):

207

–

14

.

Qin

J

,

Li

R

,

Raes

J

, et al. .

A human gut microbial gene catalogue established by metagenomic sequencing

.

Nature

.

2010

;

464

(

7285

):

59

–

65

.

Voreades

N

,

Kozil

A

,

Weir

TL

.

Diet and the development of the human intestinal microbiome

.

Front Microbiol

.

2014

;

5

:

494

.

Hviid

A

,

Svanstrom

H

,

Frisch

M

.

Antibiotic use and inflammatory bowel diseases in childhood

.

Gut

.

2011

;

60

(

1

):

49

–

54

.

Blaser

M

.

Antibiotic overuse: stop the killing of beneficial bacteria

.

Nature

.

2011

;

476

(

7361

):

393

–

4

.

Trasande

L

,

Blustein

J

,

Liu

M

,

Corwin

E

,

Cox

LM

,

Blaser

MJ

.

Infant antibiotic exposures and early‐life body mass

.

Int J Obes (Lond)

.

2013

;

37

(

1

):

16

–

23

.

Ray

K

.

Gut microbiota: adding weight to the microbiota's role in obesity: exposure to antibiotics early in life can lead to increased adiposity

.

Nat Rev Gastroenterol Hepatol

.

2012

;

9

(

11

):

615

.

Livanos

AE

,

Greiner

TU

,

Vangay

P

, et al. .

Antibiotic‐mediated gut microbiome perturbation accelerates development of type 1 diabetes in mice

.

Nat Microbiol

.

2016

;

1

(

11

):

16140

.

Mariat

D

,

Firmesse

O

,

Levenez

F

, et al. .

The Firmicutes/Bacteroidetes ratio of the human microbiota changes with age

.

BMC Microbiol Jun

.

2009

;

9

:

123

.

Tremaroli

V

,

Backhed

F

.

Functional interactions between the gut microbiota and host metabolism

.

Nature

.

2012

;

489

(

7415

):

242

–

9

.

Lynch

SV

,

Pedersen

O

.

The human intestinal microbiome in health and disease

.

N Engl J Med.

2016

;

375

(

24

):

2369

–

79

.

Loh

G

,

Blaut

M

.

Role of commensal gut bacteria in inflammatory bowel diseases

.

Gut Microbes

.

2012

;

3

(

6

):

544

–

55

.

Kim

KA

,

Gu

W

,

Lee

IA

,

Joh

EH

,

Kim

DH

.

High fat diet‐induced gut microbiota exacerbates inflammation and obesity in mice via the TLR4 signaling pathway

.

PLoS One.

2012

;

7

(

10

):

e47713

.

Cani

PD

,

Bibiloni

R

,

Knauf

C

, et al. .

Changes in gut microbiota control metabolic endotoxemia‐induced inflammation in high‐fat diet‐induced obesity and diabetes in mice

.

Diabetes

.

2008

;

57

(

6

):

1470

–

81

.

Jackson

MA

,

Verdi

S

,

Maxan

ME

, et al. .

Gut microbiota associations with common diseases and prescription medications in a population‐based cohort

.

Nat Commun.

2018

;

9

(

1

):

2655

.

Tamboli

CP

,

Neut

C

,

Desreumaux

P

,

Colombel

JF

.

Dysbiosis in inflammatory bowel disease

.

Gut

.

2004

;

53

(

1

):

1

–

4

.

Tlaskalova‐Hogenova

H

,

Stepankova

R

,

Kozakova

H

, et al. .

The role of gut microbiota (commensal bacteria) and the mucosal barrier in the pathogenesis of inflammatory and autoimmune diseases and cancer: contribution of germ‐free and gnotobiotic animal models of human diseases

.

Cell Mol Immunol.

2011

;

8

(

2

):

110

–

20

.

Li

JY

,

Chassaing

B

,

Tyagi

AM

, et al. .

Sex steroid deficiency‐associated bone loss is microbiota dependent and prevented by probiotics

.

J Clin Invest

.

2016

;

126

(

6

):

2049

–

63

.

Sjogren

K

,

Engdahl

C

,

Henning

P

, et al. .

The gut microbiota regulates bone mass in mice

.

J Bone Miner Res

.

2012

;

27

(

6

):

1357

–

67

.

Hansen

CH

,

Nielsen

DS

,

Kverka

M

, et al. .

Patterns of early gut colonization shape future immune responses of the host

.

PLoS One.

2012

;

7

(

3

):

e34043

.

McCabe

LR

,

Irwin

R

,

Schaefer

L

,

Britton

RA

.

Probiotic use decreases intestinal inflammation and increases bone density in healthy male but not female mice

.

J Cell Physiol.

2013

;

228

(

8

):

1793

–

8

.

Britton

RA

,

Irwin

R

,

Quach

D

, et al. .

Probiotic L. reuteri treatment prevents bone loss in a menopausal ovariectomized mouse model

.

J Cell Physiol

.

2014

;

229

(

11

):

1822

–

30

.

Ohlsson

C

,

Engdahl

C

,

Fak

F

, et al. .

Probiotics protect mice from ovariectomy‐induced cortical bone loss

.

PLoS One.

2014

;

9

(

3

):

e92368

.

Collins

F

,

Kim

S

,

McCabe

L

,

Weaver

C

.

Intestinal Microbiota and Bone Health: The Role of Prebiotics, Probiotics and Diet

.

New York, NY

:

Springer

;

2015

.

Parvaneh

K

,

Ebrahimi

M

,

Sabran

MR

, et al. .

Probiotics (Bifidobacterium longum) increase bone mass density and upregulate Sparc and Bmp‐2 genes in rats with bone loss resulting from ovariectomy

.

Biomed Res Int

.

2015

;

2015

:

897639

.

Blanton

LV

,

Charbonneau

MR

,

Salih

T

, et al. .

Gut bacteria that prevent growth impairments transmitted by microbiota from malnourished children

.

Science

.

2016

;

351

(

6275

):

aad3311

.

Yan

J

,

Herzog

JW

,

Tsang

K

, et al. .

Gut microbiota induce IGF‐1 and promote bone formation and growth

.

Proc Natl Acad Sci U S A

.

2016

;

113

(

47

):

E7554

–

63

.

Hsu

E

,

Pacifici

R

.

From osteoimmunology to osteomicrobiology: how the microbiota and the immune system regulate bone

.

Calcif Tissue Int

.

2018

;

102

(

5

):

512

–

21

.

Novince

CM

,

Whittow

CR

,

Aartun

JD

, et al. .

Commensal gut microbiota immunomodulatory actions in bone marrow and liver have catabolic effects on skeletal homeostasis in health

.

Sci Rep

.

2017

;

7

(

1

):

5747

.

Quach

D

,

Britton

RA

.

Gut microbiota and bone health

.

Adv Exp Med Biol

.

2017

;

1033

:

47

–

58

.

Rizzoli

R

,

Biver

E

.

Effects of fermented Milk products on bone

.

Calcif Tissue Int

.

2018

;

102

:

489

–

500

.

Sylvester

FA

.

Inflammatory bowel disease: effects on bone and mechanisms

.

Adv Exp Med Biol

.

2017

;

1033

:

133

–

50

.

McCabe

LR

,

Irwin

R

,

Tekalur

A

, et al. .

Exercise prevents high fat diet‐induced bone loss, marrow adiposity and dysbiosis in male mice

.

Bone

.

2019

;

118

:

20

–

31

.

Pan

H

,

Guo

R

,

Ju

Y

, et al. .

A single bacterium restores the microbiome dysbiosis to protect bones from destruction in a rat model of rheumatoid arthritis

.

Microbiome

.

2019

;

7

(

1

):

107

.

Rizzoli

R

.

Nutritional influence on bone: role of gut microbiota

.

Aging Clin Exp Res

.

2019

;

31

(

6

):

743

–

51

.

Schepper

JD

,

Collins

FL

,

Rios‐Arce

ND

, et al. .

Probiotic Lactobacillus reuteri prevents postantibiotic bone loss by reducing intestinal dysbiosis and preventing barrier disruption

.

J Bone Miner Res

.

2019

;

34

(

4

):

681

–

98

.

Guss

JD

,

Horsfield

MW

,

Fontenele

FF

, et al. .

Alterations to the gut microbiome impair bone strength and tissue material properties

.

J Bone Miner Res

.

2017

;

32

(

6

):

1343

–

53

.

Irwin

R

,

Lee

T

,

Young

VB

,

Parameswaran

N

,

McCabe

LR

.

Colitis‐induced bone loss is gender dependent and associated with increased inflammation

.

Inflamm Bowel Dis

.

2013

;

19

(

8

):

1586

–

97

.

Tyagi

AM

,

Yu

M

,

Darby

TM

, et al. .

The microbial metabolite butyrate stimulates bone formation via T regulatory cell‐mediated regulation of WNT10B expression

.

Immunity

.

2018

;

49

(

6

):

1116

–

31.e7

.

Rovensky

J

,

Svik

K

,

Matha

V

, et al. .

Combination treatment of rat adjuvant‐induced arthritis with methotrexate, probiotic bacteria Enterococcus faecium, and selenium

.

Ann N Y Acad Sci

.

2005

;

1051

:

570

–

81

.

Mutus

R

,

Kocabagli

N

,

Alp

M

,

Acar

N

,

Eren

M

,

Gezen

SS

.

The effect of dietary probiotic supplementation on tibial bone characteristics and strength in broilers

.

Poult Sci

.

2006

;

85

(

9

):

1621

–

5

.

de Vrese

M

.

Health benefits of probiotics and prebiotics in women

.

Menopause Int.

2009

;

15

(

1

):

35

–

40

.

Collins

FL

,

Irwin

R

,

Bierhalter

H

, et al. .

Lactobacillus reuteri 6475 increases bone density in intact females only under an inflammatory setting

.

PLoS One

.

2016

;

11

(

4

):

e0153180

.

Zhang

J

,

Motyl

KJ

,

Irwin

R

,

MacDougald

OA

,

Britton

RA

,

McCabe

LR

.

Loss of bone and Wnt10b expression in male type 1 diabetic mice is blocked by the probiotic Lactobacillus reuteri

.

Endocrinology

.

2015

;

156

(

9

):

3169

–

82

.

Thiele

S

,

Baschant

U

,

Rauch

A

,

Rauner

M

.

Instructions for producing a mouse model of glucocorticoid‐induced osteoporosis

.

Bonekey Rep

.

2014

;

3

:

552

.

Valuckaite

V

,

Zaborina

O

,

Long

J

, et al. .

Oral PEG 15‐20 protects the intestine against radiation: role of lipid rafts

.

Am J Physiol Gastrointest Liver Physiol

.

2009

;

297

(

6

):

G1041

–

52

.

Valuckaite

V

,

Seal

J

,

Zaborina

O

,

Tretiakova

M

,

Testa

G

,

Alverdy

JC

.

High molecular weight polyethylene glycol (PEG 15‐20) maintains mucosal microbial barrier function during intestinal graft preservation

.

J Surg Res

.

2013

;

183

(

2

):

869

–

75

.

Ferrier

L

,

Berard

F

,

Debrauwer

L

, et al. .

Impairment of the intestinal barrier by ethanol involves enteric microflora and mast cell activation in rodents

.

Am J Pathol

.

2006

;

168

(

4

):

1148

–

54

.

Tsuji

A

,

Nakashima

E

,

Kagami

I

,

Yamana

T

.

Intestinal absorption mechanism of amphoteric beta‐lactam antibiotics II: Michaelis‐Menten kinetics of cyclacillin absorption and its pharmacokinetic analysis in rats

.

J Pharm Sci

.

1981

;

70

(

7

):

772

–

7

.

Jernberg

C

,

Lofmark

S

,

Edlund

C

,

Jansson

JK

.

Long‐term ecological impacts of antibiotic administration on the human intestinal microbiota

.

ISME J

.

2007

;

1

(

1

):

56

–

66

.

Thunhorst

RL

,

Beltz

TG

,

Johnson

AK

.

Glucocorticoids increase salt appetite by promoting water and sodium excretion

.

Am J Physiol Regul Integr Comp Physiol

.

2007

;

293

(

3

):

R1444

–

51

.

Frese

SA

,

Hutkins

RW

,

Walter

J

.

Comparison of the colonization ability of autochthonous and allochthonous strains of lactobacilli in the human gastrointestinal tract

.

Adv Microbiol

.

2012

;

2

:

399

–

409

.

Ahlroos

T

,

Tynkkynen

S

.

Quantitative strain‐specific detection of lactobacillus rhamnosus GG in human faecal samples by real‐time PCR

.

J Appl Microbiol

.

2009

;

106

(

2

):

506

–

14

.

Gardinier

JD

,

Rostami

N

,

Juliano

L

,

Zhang

C

.

Bone adaptation in response to treadmill exercise in young and adult mice

.

Bone Rep

.

2018

;

8

:

29

–

37

.

Turner

CH

,

Burr

DB

.

Basic biomechanical measurements of bone: a tutorial

.

Bone

.

1993

;

14

(

4

):

595

–

608

.

Porter

A

,

Irwin

R

,

Miller

J

,

Horan

DJ

,

Robling

AG

,

McCabe

LR

.

Quick and inexpensive paraffin‐embedding method for dynamic bone formation analyses

.

Sci Rep

.

2017

;

7

:

42505

.

Quach

D

,

Collins

F

,

Parameswaran

N

,

McCabe

L

,

Britton

R

.

Microbiota reconstitution does not cause bone loss in germ‐free mice

.

mSphere

.

2018

;

3

(

1

):

e00545

–

17

.

Collins

J

,

Auchtung

JM

,

Schaefer

L

,

Eaton

KA

,

Britton

RA

.

Humanized microbiota mice as a model of recurrent Clostridium difficile disease

.

Microbiome

.

2015

;

3

:

35

.

Kozich

JJ

,

Westcott

SL

,

Baxter

NT

,

Highlander

SK

,

Schloss

PD

.

Development of a dual‐index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform

.

Appl Environ Microbiol

.

2013

;

79

(

17

):

5112

–

20

.

Team RC

.

R: A Language and Environment for Statistical Computing

.

Vienna, Austria

:

R Foundation for Statistical Computing

;

2018

.

McMurdie

PJ

,

Holmes

S

.

Phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data

.

PLoS One.

2013

;

8

(

4

):

e61217

.

Turnbaugh

PJ

,

Ridaura

VK

,

Faith

JJ

,

Rey

FE

,

Knight

R

,

Gordon

JI

.

The effect of diet on the human gut microbiome: a metagenomic analysis in humanized gnotobiotic mice

.

Sci Transl Med

.

2009

;

1

(

6

):

6ra14

.

Lane

NE

,

Lukert

B

.

The science and therapy of glucocorticoid‐induced bone loss

.

Endocrinol Metab Clin North Am

.

1998

;

27

(

2

):

465

–

83

.

Briot

K

,

Roux

C

.

Glucocorticoid‐induced osteoporosis

.

RMD Open

.

2015

;

1

(

1

):

e000014

.

Shaker

JL

,

Lukert

BP

.

Osteoporosis associated with excess glucocorticoids

.

Endocrinol Metab Clin North Am

.

2005

;

34

(

2

):

341

–

56

viii–ix.

Weaver

CM

.

Diet, gut microbiome, and bone health

.

Curr Osteoporos Rep

.

2015

;

13

(

2

):

125

–

30

.

Manolides

AS

,

Cullen

DM

,

Akhter

MP

.

Effects of glucocorticoid treatment on bone strength

.

J Bone Miner Metab

.

2010

;

28

(

5

):

532

–

9

.

Turner

JR

.

Intestinal mucosal barrier function in health and disease

.

Nat Rev Immunol

.

2009

;

9

(

11

):

799

–

809

.

Vindigni

SM

,

Zisman

TL

,

Suskind

DL

,

Damman

CJ

.

The intestinal microbiome, barrier function, and immune system in inflammatory bowel disease: a tripartite pathophysiological circuit with implications for new therapeutic directions

.

Therap Adv Gastroenterol

.

2016

;

9

(

4

):

606

–

25

.

Lutgendorff

F

,

Akkermans

LM

,

Soderholm

JD

.

The role of microbiota and probiotics in stress‐induced gastro‐intestinal damage

.

Curr Mol Med

.

2008

;

8

(

4

):

282

–

98

.

Pittenger

MF

,

Mackay

AM

,

Beck

SC

, et al. .

Multilineage potential of adult human mesenchymal stem cells

.

Science

.

1999

;

284

(

5411

):

143

–

7

.

Li

J

,

He

C

,

Tong

W

, et al. .

Tanshinone IIA blocks dexamethasone‐induced apoptosis in osteoblasts through inhibiting Nox4‐derived ROS production

.

Int J Clin Exp Pathol.

2015

;

8

(

10

):

13695

–

706

.

Chen

Z

,

Xue

J

,

Shen

T

,

Ba

G

,

Yu

D

,

Fu

Q

.

Curcumin alleviates glucocorticoid‐induced osteoporosis by protecting osteoblasts from apoptosis in vivo and in vitro

.

Clin Exp Pharmacol Physiol

.

2016

;

43

(

2

):

268

–

76

.

Smith

E

,

Frenkel

B

.

Glucocorticoids inhibit the transcriptional activity of LEF/TCF in differentiating osteoblasts in a glycogen synthase kinase‐3beta‐dependent and ‐independent manner

.

J Biol Chem.

2005

;

280

(

3

):

2388

–

94

.

Sato

AY

,

Tu

X

,

McAndrews

KA

,

Plotkin

LI

,

Bellido

T

.

Prevention of glucocorticoid induced‐apoptosis of osteoblasts and osteocytes by protecting against endoplasmic reticulum (ER) stress in vitro and in vivo in female mice

.

Bone

.

2015

;

73

:

60

–

8

.

Mak

W

,

Shao

X

,

Dunstan

CR

,

Seibel

MJ

,

Zhou

H

.

Biphasic glucocorticoid‐dependent regulation of Wnt expression and its inhibitors in mature osteoblastic cells

.

Calcif Tissue Int

.

2009

;

85

(

6

):

538

–

45

.

Sato

AY

,

Cregor

M

,

Delgado‐Calle

J

, et al. .

Protection from glucocorticoid‐induced osteoporosis by anti‐catabolic signaling in the absence of Sost/Sclerostin

.

J Bone Miner Res

.

2016

;

31

(

10

):

1791

–

802

.

Ueda

S

,

Ichiseki

T

,

Yoshitomi

Y

, et al. .

Osteocytic cell necrosis is caused by a combination of glucocorticoid‐induced Dickkopf‐1 and hypoxia

.

Med Mol Morphol

.

2015

;

48

(

2

):

69

–

75

.

Nitzan

O

,

Elias

M

,

Peretz

A

,

Saliba

W

.

Role of antibiotics for treatment of inflammatory bowel disease

.

World J Gastroenterol

.

2016

;

22

(

3

):

1078

–

87

.

Spiller

RC

.

Role of infection in irritable bowel syndrome

.

J Gastroenterol

.

2007

;

42

(

Suppl 17

):

41

–

7

.

O'Mahony

SM

,

Marchesi

JR

,

Scully

P

, et al. .

Early life stress alters behavior, immunity, and microbiota in rats: implications for irritable bowel syndrome and psychiatric illnesses

.

Biol Psychiatry

.

2009

;

65

(

3

):

263

–

7

.

Qiu

D

,

Xia

Z

,

Deng

J

,

Jiao

X

,

Liu

L

,

Li

J

.

Glucorticoid‐induced obesity individuals have distinct signatures of the gut microbiome

.

Biofactors

.

2019

;

45

(

6

):

892

–

901

.

Smith

AD

,

Foss

ED

,

Zhang

I

, et al. .

Microbiota of MR1 deficient mice confer resistance against Clostridium difficile infection

.

PLoS One

.

2019

;

14

(

9

):

e0223025

.

Kennedy

EA

,

King

KY

,

Baldridge

MT

.

Mouse Microbiota models: comparing germ‐free mice and antibiotics treatment as tools for modifying Gut bacteria

.

Front Physiol

.

2018

;

9

:

1534

.

Le Roy

T

,

Debedat

J

,

Marquet

F

, et al. .

Comparative evaluation of microbiota engraftment following fecal Microbiota transfer in mice models: age, kinetic and microbial status matter

.

Front Microbiol

.

2018

;

9

:

3289

.

Harris

L

,

Senagore

P

,

Young

VB

,

McCabe

LR

.

Inflammatory bowel disease causes reversible suppression of osteoblast and chondrocyte function in mice

.

Am J Physiol Gastrointest Liver Physiol

.

2009

;

296

(

5

):

G1020

–

9

.

Raehtz

S

,

Hargis

BM

,

Kuttappan

VA

,

Pamukcu

R

,

Bielke

LR

,

McCabe

LR

.

High molecular weight polymer promotes bone health and prevents bone loss under salmonella challenge in broiler chickens

.

Front Physiol

.

2018

;

9

:

384

.

Su

H

,

Kang

Q

,

Wang

H

, et al. .

Effects of glucocorticoids combined with probiotics in treating Crohn's disease on inflammatory factors and intestinal microflora

.

Exp Ther Med

.

2018

;

16

(

4

):

2999

–

3003

.

Cui

HH

,

Chen

CL

,

Wang

JD

, et al. .

Effects of probiotic on intestinal mucosa of patients with ulcerative colitis

.

World J Gastroenterol

.

2004

;

10

(

10

):

1521

–

5

.

Yan

FF

,

Murugesan

GR

,

Cheng

HW

.

Effects of probiotic supplementation on performance traits, bone mineralization, cecal microbial composition, cytokines and corticosterone in laying hens

.

Animal

.

2019

;

13

(

1

):

33

–

41

.

Nilsson

AG

,

Sundh

D

,

Backhed

F

,

Lorentzon

M

.

Lactobacillus reuteri reduces bone loss in older women with low bone mineral density: a randomized, placebo‐controlled, double‐blind, clinical trial

.

J Intern Med

.

2018

;

284

(

3

):

307

–

17

.

Montazeri‐Najafabady

N

,

Ghasemi

Y

,

Dabbaghmanesh

MH

,

Talezadeh

P

,

Koohpeyma

F

,

Gholami

A

.

Supportive role of probiotic strains in protecting rats from ovariectomy‐induced cortical bone loss

.

Probiotics Antimicrob Proteins

.

2019

;

11

(

4

):

1145

–

54

.

McCabe

LR

,

Parameswaran

N

.

Advances in probiotic regulation of bone and mineral metabolism

.

Calcif Tissue Int

.

2018

;

102

(

4

):

480

–

8

.

Schepper

JD

,

Irwin

R

,

Kang

J

, et al. .

Probiotics in Gut‐bone signaling

.

Adv Exp Med Biol.

2017

;

1033

:

225

–

47

.

Ricoldi

MST

,

Furlaneto

FAC

,

Oliveira

LFF

, et al. .

Effects of the probiotic Bifidobacterium animalis subsp. lactis on the non‐surgical treatment of periodontitis. a histomorphometric, microtomographic and immunohistochemical study in rats

.

PLoS One

.

2017

;

12

(

6

):

e0179946

.

Pacifici

R

.

Bone remodeling and the microbiome

.

Cold Spring Harb Perspect Med

.

2018

;

8

:

a031203

.

Kobayashi

R

,

Kobayashi

T

,

Sakai

F

,

Hosoya

T

,

Yamamoto

M

,

Kurita‐Ochiai

T

.

Oral administration of Lactobacillus gasseri SBT2055 is effective in preventing _Porphyromonas gingivalis_‐accelerated periodontal disease

.

Sci Rep

.

2017

;

7

(

1

):

545

.

Gatej

SM

,

Marino

V

,

Bright

R

, et al. .

Probiotic Lactobacillus rhamnosus GG prevents alveolar bone loss in a mouse model of experimental periodontitis

.

J Clin Periodontol

.

2018

;

45

(

2

):

204

–

12

.

Collins

FL

,

Rios‐Arce

ND

,

Schepper

JD

,

Parameswaran

N

,

McCabe

LR

.

The potential of probiotics as a therapy for osteoporosis

.

Microbiol Spectr

.

2017

;

5

(

4

) BAD‐0015‐2016.

McCabe

L

,

Britton

RA

,

Parameswaran

N

.

Prebiotic and probiotic regulation of bone health: role of the intestine and its microbiome

.

Curr Osteoporos Rep

.

2015

;

13

(

6

):

363

–

71

.

Thomas

CM

,

Hong

T

,

van Pijkeren

JP

, et al. .

Histamine derived from probiotic Lactobacillus reuteri suppresses TNF via modulation of PKA and ERK signaling

.

PLoS One

.

2012

;

7

(

2

):

e31951

.

Lin

YP

,

Thibodeaux

CH

,

Pena

JA

,

Ferry

GD

,

Versalovic

J

.

Probiotic lactobacillus reuteri suppress proinflammatory cytokines via c‐Jun

.

Inflamm Bowel Dis

.

2008

;

14

(

8

):

1068

–

83

.

Frese

SA

,

Benson

AK

,

Tannock

GW

, et al. .

The evolution of host specialization in the vertebrate gut symbiont Lactobacillus reuteri

.

PLoS Genet

.

2011

;

7

(

2

):

e1001314

.

Oh

PL

,

Benson

AK

,

Peterson

DA

, et al. .

Diversification of the gut symbiont Lactobacillus reuteri as a result of host‐driven evolution

.

ISME J

.

2010

;

4

(

3

):

377

–

87

.

Segers

ME

,

Lebeer

S

.

Towards a better understanding of Lactobacillus rhamnosus GG: host interactions

.

Microb Cell Fact

.

2014

;

13

(

Suppl 1

):

S7

.

Latousakis

D

,

Juge

N

.

How sweet are our gut beneficial bacteria? A focus on protein glycosylation in lactobacillus

.

Int J Mol Sci

.

2018

;

19

(

1

):

136

.

Collins

FL

,

Rios‐Arce

ND

,

Schepper

JD

, et al. .

Beneficial effects of Lactobacillus reuteri 6475 on bone density in male mice is dependent on lymphocytes

.

Sci Rep

.

2019

;

9

(

1

):

14708

.

Ersek

A

,

Santo

AI

,

Vattakuzhi

Y

,

George

S

,

Clark

AR

,

Horwood

NJ

.

Strain dependent differences in glucocorticoid‐induced bone loss between C57BL/6J and CD‐1 mice

.

Sci Rep.

2016

;

6

:

36513

.

Iwamoto

J

,

Seki

A

,

Sato

Y

.

Effect of intermittent administration of hPTH(1–34) on cortical bone geometry in rats treated with high‐dose glucocorticoids

.

Chin J Physiol

.

2014

;

57

(

5

):

231

–

7

.

Klopfenstein Bregger

MD

,

Schawalder

P

,

Rahn

B

,

Eckhardt

C

,

Schneider

E

,

Lill

C

.

Optimization of corticosteroid induced osteoporosis in ovariectomized sheep. A bone histomorphometric study

.

Vet Comp Orthop Traumatol

.

2007

;

20

(

1

):

18

–

23

.

Guo

C

,

Yuan

L

,

Wang

JG

, et al. .

Lipopolysaccharide (LPS) induces the apoptosis and inhibits osteoblast differentiation through JNK pathway in MC3T3‐E1 cells

.

Inflammation

.

2014

;

37

(

2

):

621

–

31

.

Collins

FL

,

Rios‐Arce

ND

,

McCabe

LR

,

Parameswaran

N

.

Cytokine and hormonal regulation of bone marrow immune cell Wnt10b expression

.

PLoS One

.

2017

;

12

(

8

):

e0181979

.

Terauchi

M

,

Li

J‐Y

,

Bedi

B

, et al. .

T lymphocytes amplify the anabolic activity of parathyroid hormone through Wnt10b signaling

.

Cell Metab

.

2009

;

10

(

3

):

229

–

40

.

Kang

S

,

Bennett

CN

,

Gerin

I

,

Rapp

LA

,

Hankenson

KD

,

Macdougald

OA

.

Wnt signaling stimulates osteoblastogenesis of mesenchymal precursors by suppressing CCAAT/enhancer‐binding protein alpha and peroxisome proliferator‐activated receptor gamma

.

J Biol Chem

.

2007

;

282

(

19

):

14515

–

24

.

Yao

W

,

Cheng

Z

,

Busse

C

,

Pham

A

,

Nakamura

MC

,

Lane

NE

.

Glucocorticoid excess in mice results in early activation of osteoclastogenesis and adipogenesis and prolonged suppression of osteogenesis: a longitudinal study of gene expression in bone tissue from glucocorticoid‐treated mice

.

Arthritis Rheum

.

2008

;

58

(

6

):

1674

–

86

.

Almeida

M

,

Han

L

,

Ambrogini

E

,

Weinstein

RS

,

Manolagas

SC

.

Glucocorticoids and tumor necrosis factor alpha increase oxidative stress and suppress Wnt protein signaling in osteoblasts

.

J Biol Chem

.

2011

;

286

(

52

):

44326

–

35

.

Li

J

,

Yu

M

,

Tyagi

AM

,

Adam

J

,

Jones

RM

,

Pacifici

R

.

Microbiota produced butyrate is required for the anabolic effects of PTH in bone

.

ASBMR

.

2019

;

34

:

349

–

60

.

Tavakoli

S

,

Xiao

L

.

Depletion of Intestinal Microbiome partially rescues bone loss in sickle cell disease male mice

.

Sci Rep

.

2019

;

9

(

1

):

8659

.

Schepper

J

,

Collins

F

,

Irwin

R

,

Rios‐Arce

N

,

Parameswaran

N

,

McCabe

LR

.

Glucocorticoid induced osteoporosis is prevented by probiotic lactobacillus reuteri 6475 supplementation

.

J Bone Miner Res

.

2017

;

32

:

S277

.

Schepper

J

,

Rios‐Arce

N

,

Collins

F

,

Parameswaran

N

,

McCabe

L

.

Alterations to the gut microbiome prevent glucocorticoid induced osteoporosis

.

FASEB J

.

2019

;

33

(

1 Suppl

):

589.6

.

Author notes

Results from this work were presented at: the American Society of Bone and Mineral Research (ASBMR) 2017 Annual Meeting, September 8–11, 2017, in Denver, CO, USA; the ASBMR 2019 Annual Meeting, September 20–23, 2019, in Orlando, FL, USA; and the Experimental Biology 2019 meeting, April 6–9, 2019, in Orlando, FL, USA.

NP and LRM contributed equally to this work and are co‐senior authors.

© 2019 American Society for Bone and Mineral Research

Supplementary data

Citations

Views

Altmetric

Metrics

Total Views 393

256 Pageviews

137 PDF Downloads

Since 3/1/2024

Citations

116 Web of Science

×

Email alerts

Email alerts

Citing articles via

• Latest

• Most Read

• Most Cited

More from Oxford Academic