Introduction

Apolipoprotein E (apoE) is a structural component of lipoproteins and is required for the plasma clearance of lipoproteins through binding with lipoprotein receptors. ApoE-deficient (apoE−/−) mice exhibit severe hyperlipidemia and develop spontaneous atherosclerosis1,2. Dyslipidemia and atherosclerosis can often co-exist with osteoporosis, and lipid-lowering agents have been shown to have positive effects on osteoporosis3,4,5,6,7. However, only a few studies have described the skeletal changes caused by ApoE deficiency. Some studies have reported that young and adult apoE−/− mice exhibit a high bone mass phenotype due to increased bone formation rates8,9,10. Indeed, these mice show increased bone mass compared to that in wild-type (WT) controls despite the potentially adverse effects of spontaneous hypercholesterolemia and atherosclerosis. An in vitro study has further found that ApoE is expressed in primary murine calvaria osteoblasts at late stages of differentiation8. Conversely, when fed a high-fat diet, ApoE-deficient mice show reduced bone formation by stimulating p53-mediated apoptosis of osteoblastic cells11. Another study has found that at 28 weeks of age, the total bone mineral density (tBMD) of apoE−/− mice was significantly higher than that of WT mice, whereas at 40 weeks of age, there was no significant difference in tBMD between apoE−/− and WT mice12. These studies suggest that the skeletal phenotype of apoE−/− mice may be influenced by a variety of factors, including age, high-fat diet, and renal function. Therefore, it is necessary to study the bone phenotype and metabolism of aged apoE−/− mice.

ApoE gene isoforms are related to aging, longevity, and a number of age-related diseases, including atherogenesis and osteoporosis13,14,15,16,17,18,19. ApoE-deficient mice develop many premature degenerative changes linked to different ages20, suggesting that ApoE-deficient mice may serve as a good model to study aging. Another important gene that is correlated with lifespan and age-related diseases belongs to the sirtuin family21,22. Sirtuins are highly conserved NAD+-dependent deacetylases that have been shown to regulate lifespan in lower organisms and to affect diseases related to aging in mammals23,24. Studies have found that Sirt1 plays an important role in the pathogenesis of atherosclerosis25,26. Sirt1 also plays a key role in osteoblast regulation27 and mesenchymal stem cell differentiation28. One study has shown that endothelial cell-specific Sirt1 transgenic ApoE-knockout mice have fewer atherosclerotic lesions than apoE−/− controls. This finding suggests that the observed effects of ApoE knockout on aging might be related to the insufficient expression of Sirt129. A recent study has reported that aged (2.2-year-old) mesenchymal stem cell (MSC)-specific Sirt1 knockout (MSCKO) mice show reduced cortical bone thickness and trabecular volume. Young mice, however, exhibit less-pronounced effects30. To date, no studies have described the levels of Sirt1 expression in the bone tissues of apoE−/− mice.

Therefore, in this study, we determined the skeletal phenotypes of aged apoE−/− mice using micro-computed tomography (micro-CT). To further analyze the possible mechanisms related to this phenotype, we evaluated Sirt1 expression in the bone tissues of apoE−/− mice and examined the role of Sirt1 in osteoblast differentiation from bone marrow-derived mesenchymal stem cells (BMSCs).

Materials and methods

Experimental animals

All animal studies were conducted with the approval of the Committee on Animal Resources at Huadong Hospital, Fudan University. Male apoE−/− mice on a C57Bl/6J background and their WT littermates were purchased from the Nanjing Biomedical Research Institute (Nanjing University, Nanjing, China) at 18 weeks of age. The mice were maintained in groups under a strict 12-h light/dark cycle in a temperature-controlled environment (25 °C) with free access to food and water. All animals were fed a standard mouse diet containing 20.50% protein, 4.68% fat, 1.23% calcium, and 0.91% phosphorus.

Micro-CT analysis

Femurs collected from apoE−/− mice and WT mice at 18 and 72 weeks of age were dissected, cleaned, fixed in 10% Millonig's formalin, and transferred to 100% ethanol. The femurs were then analyzed using a SkyScan-1176 micro-CT instrument (Bruker Corporation, Kontich, Belgium). Scans were performed using PANalytical's Microfocus Tube, with a 8.96-μmol/L voxel size, a 45 kV, 500 μA, and a 0.6° rotation step (180° angular range). For trabecular bone, micro-CT evaluation was performed on a 1-mm region of metaphyseal spongiosa in the distal femur. The regions were located 0.5 mm above the growth plate. The index included tissue bone mineral density (BMD) and directly measured bone volume fraction (BV/TV), thickness (Tb.Th), number (Tb.N), and separation (Tb.Sp). For cortical bone, measurements were performed on a 0.5-mm region of the mid-diaphysis of the femur. The index included total area (Tt.Ar), cortical area (Tt.Cr), thickness (Ct.Th), and porosity. NR Econ software Version 1.6 (Bruker Corporation) was used for the three-dimensional (3D) reconstruction and viewing of images.

Histological analysis

The tibias of each mouse were fixed for 1 d in 4% paraformaldehyde in phosphate-buffered saline (PBS) at 4 °C. Specimens were decalcified in 10% EDTA in 0.05 mol/L Tris buffer for 7 d at 4 °C until the bones were soft and flexible. The samples were then processed and embedded in paraffin. Sections were cut and stained with hematoxylin and eosin (H&E) and histochemically analyzed for alkaline phosphatase (ALP) activity using a BCIP/NBT Kit (Beyotime Biotechnology, Nanjing, China) and analyzed for tartrate-resistant acid phosphatase (TRAP) activity using a Leukocyte Acid Phosphatase Kit for TRAP staining (Sigma, St Louis, MO, USA). Sections were then counterstained with methyl green and mounted in Kaiser's glycerol jelly. Calculations were performed on a minimum of duplicate specimens from five mice in each group.

Serum bone turnover markers and biochemistry

Serum from each mouse was analyzed individually and in duplicate for the bone formation marker osteocalcin (OCN) using enzyme-linked immunosorbent assays (ELISAs) with a Mouse Osteocalcin EIA Kit (Immutopics International, San Clemente, CA, USA), and for the bone resorption marker tartrate-resistant acid phosphatase 5b (TRAP5b) using Mouse TRAP Assays (Immunodiagnostic Systems Inc, Boldon, UK) according to the manufacturer's instructions. The average values of duplicate measurements were obtained for each mouse. Serum total cholesterol (TC) was measured using a Hitachi 917 autoanalyzer (Tokyo, Japan), and serum oxidized low-density lipoprotein (ox-LDL) was measured using a mouse ox-LDL ELISA Kit (Uscn Life Science Inc, Wuhan, China).

Cell preparation and culture

BMSCs were obtained from apoE−/− mice and WT mice (male, 18 weeks old). The femurs and tibias were removed under aseptic conditions. After removal of the surrounding muscle and connective tissue, the ends of the bones were cut to expose the bone marrow cavity. The cavity was washed three times with PBS, and the bone marrow was injected with L-DMEM medium. The collected cells were directly inoculated in culture in a bottle and allowed to attach for 3 d before the removal of non-adherent cells. When cells reached confluence, they were detached with 0.25% trypsin and scraped from the surface of the culture plate. The second generation of cells was seeded at 1×105 cells/well in a six-well culture plate and cultured in L-DMEM containing 10% fetal bovine serum (FBS). Some cells were cultured in experimental culture medium supplemented with 10 μmol/L EX-527 (Sigma) for 12 h and collected. Alternatively, cells were supplemented with 20 μg/mL ox-LDL (Peking Union-Biololgy Co Ltd, Beijing, China); ox-LDL was added every 24 h, and cells were collected at 12, 24, 48, or 72 h after ox-LDL was added.

Real-time reverse transcription polymerase chain reaction (RT-qPCR)

The total RNA was extracted from tissues and cultured cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. Total RNA was reverse transcribed to cDNA using a QuantiTect Rev Transcription Kit (Qiagen, Chatsworth, CA, USA). The number of cDNA molecules in the reverse-transcribed samples was determined by real-time PCR using a modified method with a QuantiTect SYBR Green PCR Kit (Qiagen) on an Mx3000P Real-Time PCR system (Stratagene, La Jolla, CA, USA) according to the manufacturer's instructions. The primers were obtained from SBSgene (www.sbsgene.com) with the following sequences: Sirt1, 5′-GCAGGTTGCAGGAATCCAA-3′ and 5′-GGCAAGATGCTGTTGCAAA-3′ (63 bp, XM_006514342.1); Runt-related transcription factor 2 (Runx2), 5′-TGTTCTCTGATCGCCTCAGTG-3′ and 5′-CCTGGGATCTGTAATCTGACTCT-3′ (146 bp, XM_006523545.1); ALP, 5′-CACGCGATGCAACACTCAGG-3′ and 5′-GCATGTCCCCGGGCTCAAAGA-3′ (479 bp, XM_ 006538500.1); osteocalcin (OCN), 5′-ACCCTGGCTGCGCTCTGTCTCT-3′ and 5′-GATGCGTTTGTAGGCGGTCTTCA-3′ (240 bp, NM_007541.3); and glyceraldehyde -3-phosphate dehydrogenase (GAPDH), 5′-AGCCTCGTCCCGTAGACA-3′ and 5′-CTCGCTCCTGGAAGATGG-3′ (255 bp, NM_008084.3). The conditions for PCR were 12.5 μL of SYBR Green I Master Mix, 0.25 μmol/L each of forward and reverse primer, and 2 μL of sample, and H2O to a final volume of 25 μL. A melting curve was obtained at the end of each run to discriminate specific from nonspecific cDNA products. The cDNA content was normalized by subtracting the cycle number of the GAPDH gene expression from that of the target gene expression (ΔCt=Ct of target gene-Ct of GAPDH), and the gene expression level was calculated using the 2−(ΔCt) method.

Western blot analysis

All samples were lysed using cell lysis buffer (Beyotime, Nanjing, China) plus a 1:100 volume of phenylmethanesulfonyl fluoride (100 mmol/L). Protein was then quantified using BCA protein assays (P0012; Beyotime) with bovine serum albumin (BSA) as a standard. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) sample loading buffer (P0015, Beyotime) was added to the protein samples, and samples were heated for 5 min in a boiling water bath. Aliquots of samples (40 μg) were then subjected to SDS-PAGE using 12% gels under reducing conditions and electroblotted onto polyvinylidene difluoride (PVDF) membranes (Ipvh00010; Millipore, Bedford, MA, USA). The membranes were blocked with 5% fat-free dry milk in TBST (0.1% Tween-20 and 0.1 mol/L NaCl in 0.1 mol/L Tris-HCl, pH 7.5) for 2 h at room temperature and then incubated overnight at 4 °C with the following primary antibodies: rabbit anti-Sirt1 (1:2000; Proteintech, Chicago, IL, USA); goat anti-Runx2 (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA, USA); and mouse anti-GAPDH (1:10 000; Kangcheng, Shanghai, China). The membranes were then incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000; Santa Cruz Biotechnology) at room temperature for 1 h, and proteins were detected by chemiluminescence (P0018; Beyotime). Each incubation step was followed by three washes (10 min each) with TBST. The protein bands were quantitatively analyzed using an image analysis system (QuantityOne software; Bio-Rad, Hercules, CA, USA).

Statistical analysis

The data are presented as the mean±standard deviation (SD). Statistical significance was determined using Student's t-tests or one-way analysis of variance (ANOVA) followed by Dunnett's test. Analyses were performed using SPSS 17.0 (Chicago, IL, USA), and differences with P-values less than 0.05 were considered significant.

Results

Detection of decreased bone volume in aged apoE−/− mice by micro-CT

First, we performed quantitative analyses of the bone phenotype using reconstruction of a micro-CT on the distal metaphyseal region of the femurs from male WT mice and apoE−/− mice that were 18 and 72 weeks of age (Figure 1). Relative to WT mice, femurs from 18-week-old apoE−/− mice exhibited a marked increase in trabecular BMD, BV/TV, and Tb.Th (+30.43%, +39.69%, and +30.68%, respectively; P<0.05; Figure 1A and 1C). No changes in cortical parameters were observed between the femurs of young WT and apoE−/− mice (Figure 1E and 1G). In contrast, in 72-week-old animals, the extent of total bone loss was much greater in apoE−/− mice. The trabecular BMD decreased significantly in apoE−/− mice compared with that in WT mice (−30.68%, P<0.01). This result was also accompanied by decreases in body weight (P<0.05), femur trabecular BV/TV (−48.19%, P<0.01), Tb.Th (−30.58%, P<0.05), and Tb.N (−60.02%, P<0.01; Figure 1B and 1D). Moreover, measurements of cortical bone from aged apoE−/− mice showed decreased cortical thickness (Ct.Th; −40.96%, P<0.05) and increased cortical porosity (+32.14%, P<0.05; Figure 1F and 1H).

Figure 1
figure 1

Bone mass in 18- and 72-week-old apoE−/− mice and their age-matched WT mice, as determined by micro-CT. (A and B) Representative 3D Micro-CT images showing regions of the trabecular bone in the distal femur. (C and D) Graphs represent measurements of total body weight, bone mineral density (BMD), trabecular bone volume (BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular spacing (Tb.Sp) in the femurs. (E and F) Representative 3D micro-CT images showing regions of the cortical bone in the femurs. (G and H) Graphs represent measurements of bone mineral density (BMD), total area (Tt.Ar), cortical area (Ct.Ar), cortical thickness (Ct.Th), and cortical porosity. Statistical analysis was performed using a two-way ANOVA followed by a Dunnett's test. Mean±SD, n=7. bP<0.05, cP<0.01 vs WT.

PowerPoint slide

Full size image

Reduced bone formation in aged apoE−/− mice

To determine whether the differences in bone architecture were accompanied by corresponding changes in markers of bone formation and resorption, we performed histomorphometric analysis of tibias from 18- and 72-week-old WT and apoE−/− mice for markers associated with bone turnover (Figure 2). In 18-week-old mice, the results showed that the ratio of the ALP-positive osteoblast surface relative to the total bone surface (OB.S/BS) was increased (+28.70%, P<0.05) in apoE−/− mice compared with that in WT mice (Figure 2C). Somewhat contradictory to the results in the young apoE−/− mice, the ALP-positive area (OB.S/BS) decreased significantly in 72-week-old apoE−/− mice compared with that in WT mice (−40.7%, P<0.05; Figure 2D). No changes were observed in the osteoclast surface/bone surface (OC.S/BS) between the aged apoE−/− mice and WT mice at both young and old ages (Figure 2E and 2F). Consistently with results from previous studies, serum levels of OCN, a marker of bone formation, increased 1.22-fold in apoE−/− mice compared with that in WT mice at 18 weeks of age (P<0.05, Figure 3C). Serum levels of Trap5b also increased significantly (P<0.05, Figure 3D), whereas in 72-week-old animals, the serum levels of OCN were decreased relative to those of WT mice (−31.57%, P<0.05; Figure 3C). However, the serum levels of Trap5b did not change significantly in aged apoE−/− mice relative to that in WT mice (Figure 3D). The serum levels of TC and ox-LDL were significantly higher in both 18-week- and 72-week-old apoE−/− mice compared with those in age-matched WT mice (P<0.01, Figure 3A and 3B).

Figure 2
figure 2

Changes of histochemistry in tibias of 18- and 72-week-old apoE−/− mice and their age-matched WT mice. (A and B) The H&E stained sections of trabecular bone in tibias. The trabecular bone volumes (BV/TV) were determined by histomorphometric analysis. Alkaline phosphatase (ALP) staining to quantify the osteoblast surface/bone surface (OB.S/BS) from 18-week-old (C) and 72-week-old (D) mice. The tartrate-resistant acid phosphatase (TRAP) staining of sections of trabecular bone in tibias from 18-week-old (E) and 72-week-old (F) mice to quantify the osteoclast surface/bone surface (OC.S/BS). Statistical analysis was performed using a two-way ANOVA followed by a Dunnett's test. Mean±SD, n=5. bP<0.05 vs WT.

PowerPoint slide

Full size image
Figure 3
figure 3

Serum biomarker and bone turnover marker levels in 18- and 72-week-old apoE−/− mice and their age-matched WT mice. Serum levels are shown for total cholesterol (TC, A), oxidized low-density lipoprotein (ox-LDL, B), bone formation marker osteocalcin (OCN, C) and bone resorption marker tartrate resistant acid phosphatase 5b (TRAP5b, D). Statistical analysis was performed using two-way ANOVA followed by Dunnett's test. Mean±SD, n=5. bP<0.05, cP<0.01, vs WT.

PowerPoint slide

Full size image

Sirt1 expression decreased in bone tissues from aged apoE−/− mice

To further analyze the mechanisms mediating the rapid decrease in bone mass and bone formation in aged apoE−/− mice, we measured Sirt1 and Runx2 expression in bone tissue in both young and aged apoE−/− mice. In 18-week-old mice, Sirt1 mRNA levels were 2.12-fold higher in apoE−/− mice than in WT mice (P<0.05). Similarly, the mRNA levels of the bone formation markers Runx2, ALP, and OCN were significantly higher in apoE−/− mice after normalization (2.9-, 1.29-, and 1.73-fold, respectively; P<0.05; Figure 4A). Western blot analysis confirmed these results (Figure 4C). However, in 72-week-old apoE−/− mice, protein and mRNA levels of Sirt1 and Runx2 in femurs were significantly decreased compared with those in age-matched WT animals (P<0.01, Figure 4B and 4D).

Figure 4
figure 4

Sirt1 and relative expression of bone formation genes in 18- and 72-week-old apoE−/− mice and their age-matched WT mice. RT-PCR gene expression is shown for whole femurs of 18-week-old (A) and 72-week-old mice (B) for Sirt1, Runx2, ALP and OCN. Western blotting showing levels of Sirt1 and Runx2 in protein extracts from whole femur of 18-week-old (C) and 72-week-old mice (D). Statistical analysis was performed using a two-way ANOVA followed by a Dunnett's test. Mean±SD, n=4. bP<0.05, cP<0.01 vs WT.

PowerPoint slide

Full size image

Ox-LDL induced a time-dependent decrease in Sirt1 expression during the osteogenic differentiation of BMSCs from apoE−/− mice in vitro

To further investigate whether Sirt1 plays an active role in the differentiation of BMSCs, we treated BMSCs from 18-week-old apoE−/− mice with the Sirt1 inhibitor EX527. After treatment with EX527 (10 μmol/L), the mRNA (Figure 5A) and protein levels (Figure 5B) of Runx2 decreased significantly (−23.57%, P<0.05 and −54.38%, P<0.01, respectively).

Figure 5
figure 5

Sirt1 induced bone formation in the bone mesenchymal stem cells (BMSCs) from apoE−/− mice in vitro. BMSCs from 18-week-old apoE−/− mice were cultured with 10 μmol/L EX-527 and the relative mRNA expression of Sirt1, Runx2, ALP and OCN (A) and protein (Western blotting) expression of Sirt1 and Runx2 (B) were analyzed for the mean change relative to an untreated control. Mean±SD, n=4. bP<0.05, cP<0.01 vs control.

PowerPoint slide

Full size image

Ox-LDL was much higher in aged apoE−/− mice than in WT mice (Figure 3B); therefore, we examined whether Sirt1 was involved in the osteogenic differentiation of BMSCs during long-term exposure of apoE−/− mice to ox-LDL (Figure 6A and 6B). RT-PCR analysis showed that ox-LDL (20 μg/mL) enhanced the expression of Sirt1 mRNA in BMSCs in the first 12 h (3.3-fold; P<0.05). However, after prolonged treatment with ox-LDL, the expression of Sirt1 mRNA decreased at 72 h compared with that at 0 h (−48%; P<0.05), which indicated that the long-term effects of ox-LDL accumulation downregulated Sirt1 expression in the BMSCs of apoE−/− mice. Similar results were observed for Runx2 mRNA and protein levels.

Figure 6
figure 6

Ox-LDL induced a time-dependent decrease in Sirt1 expression during the osteogenic differentiation of BMSCs from apoE−/− mice in vitro. BMSCs from 18-week-old apoE−/− mice were treated with ox-LDL (20 μg/mL). The relative mRNA expression of Sirt1, Runx2, ALP and OCN (A) and protein levels of Sirt1 and Runx2 (Western blotting, B) in BMSCs after exposure to ox-LDL for 0, 12, 24, 48, and 72 h. Mean±SD, n=4. bP<0.05, cP<0.01 vs 0 h.

PowerPoint slide

Full size image

Discussion

In this study, we investigated the changes in bone phenotype and metabolism of aged apoE−/− mice to elucidate the mechanisms involved. Our results showed that aged apoE−/− mice had lower bone mass compared with age-matched WT mice. Moreover, continuous stimulation with ox-LDL decreased Sirt1 and Runx2 expression in BMSCs, providing insight into the mechanisms of reduced bone formation in aged apoE−/− mice.

We found that femur trabecular bone mass and BV/TV were higher in young apoE−/− mice than in WT mice. This was consistent with the results of most previous studies in which the mice were usually 3–8 months old8,10. However, in our study, the trabecular volume of aged apoE−/− mice decreased significantly compared with that of WT mice of the same age. Histological changes in the tibias further confirmed the decreased bone formation in aged apoE−/− mice. In addition, our in vivo study showed that Runx2 expression in bone tissue was decreased in aged apoE−/− mice compared with that in WT mice. Runx2 is an early master transcription factor that initiates the transcriptional program for the osteogenic lineage; this leads to the upregulation of various bone-related genes and markers, such as OCN, a late bone-formation marker responsible for constructing the bone matrix31,32. Interestingly, in our study, we found that serum OCN in aged apoE−/− mice decreased significantly, whereas the levels of the bone resorption marker Trap5b were not different between apoE−/− and WT mice. These results indicated that there were significant age-induced reductions in bone mass and the bone formation rate in aged apoE−/− mice. To the best of our knowledge, our present study is the first to report the bone phenotype and metabolism of aged apoE−/− mice. A similar study by Wang et al has found that although the trabecular BMD (tBMD) of femurs from 28-week-old apoE−/− mice was higher than that of WT mice, there was no significant difference in the tBMD between 40-week-old apoE−/− mice and WT mice. Moreover, the Tb.N and Tb.Th in trabecular bone increased with age and reached a peak at 28 weeks of age, after which these parameters decreased12.

Previous studies have found that ApoE is expressed in bone morphogenetic protein-2 (BMP-2)-treated mesenchymal progenitors and in mineralized primary osteoblast cultures33. ApoE can enhance the uptake of vitamin K-containing chylomicron remnants by human osteoblasts34. Moreover, although ApoE knockout itself directly affects osteogenic lineage cells and promotes bone formation in vitro9,33,34, the final bone phenotype is the result of the combined effects of multiple factors. At a young age, the effects of ApoE deficiency on bones may play a more dominant positive role compared with the negative effects of hyperlipidemia and spontaneous atherosclerosis. However, in animals consuming a high-fat diet, ApoE deficiency has been shown to reduce bone formation through stimulation of p53-mediated apoptosis11. During aging, apoE−/− mice exhibited an accumulation of ox-LDL, accelerated atherosclerosis, increased production of reactive oxygen species, and enhanced cell aging; these features may lead to aggravated bone loss. Hypercholesterolemia-induced oxidant stress could accelerate the aging of hematopoietic stem cells from apoE−/− mice35. Compared with young mice, aged apoE−/− mice exhibit higher production of reactive oxygen species, leading to DNA damage and apoptosis of bone marrow cells36. Therefore, it is not surprising that the aged apoE−/− mice in this study exhibited decreased bone volume in contrast to the high bone mass observed in young mice. Similar contradictory situations were also found in other organs in apoE−/− mice. Aortic rings isolated from both young (6 to 18-week-old) and adult (20 to 35-week-old) male and female apoE−/− mice fed a standard chow diet exhibit a preserved endothelial oxide synthase (NO)-dependent relaxation response to acetylcholine (ACh) compared with that of WT controls. However, aged apoE−/− mice (50–70 weeks old) also show significant endothelial dysfunction, as demonstrated by a reduced aortic relaxation response to ACh37. Furthermore, various bone phenotypes during different stages of aging in the same type of mouse have also been demonstrated in other gene-knockout animal models. For example, in young (2-month-old) MSCKO mice, the effects of Sirt1 loss on bone are less pronounced, affecting only trabecular thickness without changing overall bone volume. However, in aged (2.2-year-old) MSCKO mice, cortical bone thickness and trabecular BV/TV are reduced 25% and 23%, respectively, as a result of reduced numbers of osteoblasts and lower bone formation rates30. Lyer et al38 have reported that the bone formation rate is 40% higher in Foxo1,-3-4f/f and Osx1-Cre mice than in WT mice at 7 weeks of age. However, at 24 months of age, there are no differences in bone formation rates between the two types of mice.

To further analyze the mechanisms mediating this rapid decrease in bone mass and bone formation in aged apoE−/− mice, we first measured Sirt1 expression in bone tissues in young and aged apoE−/− mice. Sirt1 is the founding member of a group of proteins that have been well studied in aging research39,40, and Sirt1 affects a variety of biological functions, including DNA repair, energy metabolism, tumor suppression, and mitochondrial homeostasis41,42. In addition, transgenic mice with approximately 2-fold higher levels of global Sirt1 expression are protected against metabolic decline due to aging43. These mice are also protected against aging-induced bone loss44. Sirt1+/− heterozygous female mice exhibit reduced bone mass characterized by decreased bone formation and increased marrow apoptosis45. The functional association between apoE knockout and Sirt1 has rarely been studied, and existing studies have focused mainly on atherosclerosis. Sirt1 has been shown to be atheroprotective in apoE−/− mice25,46,47, whereas genetic overexpression of Sirt1 in LDL receptor-deficient (Ldlr−/−) mice enhances atherosclerosis44. Stein et al have reported that young hypercholesterolemic apoE−/− Sirt1+/− mice treated with lipopolysaccharide exhibit more pronounced endothelial expression of intercellular cell adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) compared with apoE−/−Sirt1+/+ mice, suggesting that endogenous Sirt1 reduces endothelial activation in apoE−/− mice25. However, the association between apoE knockout and Sirt1 expression in the bone has not been studied before. Unlike the results observed in the arteries, in this study, we found that the expression of Sirt1 mRNA and protein in the bones of young apoE−/− mice was higher than that in the bones of WT mice (see additional data in Figure 7), whereas in aged apoE−/− mice, the expression of Sirt1 decreased significantly compared with that of WT mice. As mentioned above, Sirt1 plays an important positive role in bone metabolism, and the decreased expression of Sirt1 was consistent with the decreased bone mass and reduced bone formation rate in aged apoE−/− mice.

Figure 7
figure 7

The mRNA expression of Sirt1 in different tissues from 18-week-old male apoE−/− mice and their age-matched WT mice. The mRNA expression of Sirt1 in the femur, tibia, ulna, vertebra, heart, liver, spleen, lung, kidney, intestines, muscle and vessels were analyzed for the mean fold change relative to controls. Mean±SD, n=4. bP<0.05, cP<0.01 vs WT.

PowerPoint slide

Full size image

Based on these findings, we hypothesize that Sirt1 may be involved in the age-induced bone loss observed in apoE−/− mice. To test this hypothesis in more detail, we measured the expression of Sirt1 and Runx2 in BMSCs after stimulation with ox-LDL in vitro. Ox-LDL generates reactive oxygen species (ROS), which accelerate the aging of BMSCs48. In our study, after an initial increase (in the first 12 h), Sirt1 protein and mRNA expression in BMSCs from apoE−/− mice decreased more rapidly and became even lower than that at 0 h. These data indicated that continuous incubation with ox-LDL downregulated Sirt1 expression in the BMSCs of the apoE−/− mice. A recent study examining the effects of Sirt1 on the efferocytosis of ox-LDL-induced apoptotic RAW264.7 cells (macrophages) has revealed that the expression levels of Sirt1 and autophagy marker proteins are increased at 24 h and then decreased at 48 h49. Moreover, inhibition of Sirt1 in EX527-treated BMSCs reduces the expression of Runx2, suggesting that Sirt1 is involved in bone osteogenic differentiation in mice. The Sirt1 agonist resveratrol regulates Sirt1/Runx2 expression, thereby promoting the osteogenic differentiation of BMSCs50. Resveratrol promotes the osteogenesis of human mesenchymal stem cells by upregulating Runx2 gene expression via the Sirt1/Foxo3a axis51. This may explain why aged apoE−/− mice exhibited reduced bone formation rates and bone mass.

Together, our data showed that aged apoE−/− mice exhibited low bone mass and low bone formation. Sirt1 may play an important role in age-induced bone loss in aged apoE−/− mice. However, the direct relationship between ApoE knockout and Sirt1 expression in bone is still unclear. Further studies are required to elucidate these mechanisms.

Conclusions

Aged apoE−/− mice had lower bone mass compared with age-matched WT mice. Long-term treatment with ox-LDL decreased the expression of Sirt1 in BMSCs; this mechanism contributed to the low bone formation observed in these mice.

Author contribution

Zhi-bin YE designed the study; Xiao-li ZHANG directed the research; Wei HONG and Xiao-ya XU performed the histomorphometry, RT-PCR, and Western blotting; Zhan-ying WEI performed the micro-CT analysis; Zhao-hui QIU and Li ZHEN contributed to the animal experiments; Jian-jun GAO prepared the figures; Xiao-ya XU and Wei HONG analyzed the data; Wei HONG wrote the paper. All authors have read and approved the final paper.