Accumulation of advanced oxidation protein products aggravates bone-fat imbalance during skeletal aging

Abstract

Background: Skeletal aging is characterized by a decrease in bone mass and an increase in marrowfat content. Advanced oxidation protein products (AOPPs) accumulate easily with aging and disrupt redox homeostasis. We examined whether AOPPs accumulation contributes to the bone-fat imbalance during skeletal aging.

Methods: Both young and aged mice were employed to assess the changes of AOPPs levels and its contribution to bone-fat imbalance during skeletal aging. Primary bone marrow mesenchymal stromal cells (MSCs) were used to examine the potential role of AOPPs in age-related switch between osteogenic and adipogenic differentiation. Aged mice were also gavaged by non-selective antioxidant N-acetyl-L-cysteine (NAC), followed by close monitoring of the changes in AOPPs levels and bone-fat metabolism. Furthermore, young mice were chronically exposed to AOPPs and then evaluated for the changes of bone mass and marrow adiposity.

Results: The levels of AOPPs in serum and bone marrow were markedly higher in aged mice than that in young mice. Age-related accumulation of AOPPs was accompanied by reduced bone formation, increased marrow adiposity and deterioration of bone microstructure. Reduced AOPPs accumulation by antioxidant NAC leaded to improvement of the bone-fat imbalance in aged mice. Similarly, the bone-fat imbalance was induced by chronic AOPPs loading in young mice. Compared with MSCs from young mice, MSCs from aged mice tended to differentiate into adipocytes rather than osteoblasts and displayed cellular senescence. Exposure of primary MSCs to AOPPs resulted in the switch from osteogenic to adipogenic lineage and cellular senescence. AOPPs challenge also increased intracellular ROS generation by the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase and mitochondria. The antioxidant NAC, after scavenging ROS, ameliorated the AOPPs-induced lineage switch and senescence in MSCs by inhibiting the PI3K/AKT/mTOR pathway.

Conclusion: Our findings revealed the involvement of AOPPs in age-related switch between osteogenic and adipogenic differentiation, and illuminated a novel potential mechanism underlying bone-fat imbalance during skeletal aging.

The translational potential of this article: Reducing AOPPs accumulation and its cascading effects on MSCs might be an attractive strategy for delaying skeletal aging.

Keywords: Adipogenesis; Advanced oxidation protein products; Mesenchymal stromal cells; Osteogenesis; Skeletal aging.

Graphical abstract

Fig. 1

AOPPs accumulation was accompanied by the bone-fat imbalance during skeletal aging. A Serum AOPPs levels in young mice (3-month-old) and aged mice (18-month-old) (n = 6). B AOPPs relative levels in bone marrow from young and aged mice (n = 6). C Serum ALP level (n = 6). D Serum FABP4 level (n = 6). E, F HE staining of bone trabeculae and adipocytes in distal femora, and the area of fat vacuoles was quantified. Scale bars = 100 μm. G, H IHC staining for OCN and number of osteoblasts in distal femoral trabeculas. Scale bars = 20 μm. I, J IHC staining for FABP4 and number of adipocytes. Scale bars = 20 μm. K, L Micro-CT scanning of osmium tetroxide-stained tibia and MAT content below the growth plate (n = 3). M, N The protein and mRNA expression of OCN and PPARγ in proximal tibial metaphysis. O Representative micro-CT images of distal femora from young and aged mice (n = 6) (Scale bars = 1 mm). P-V Quantitative trabecular and cortical bone parameters data were displayed in Bone mineral density (BMD), Trabecular Bone Volume/Total Volume (Tb.BV/TV), Trabecular Number (Tb.N), Trabecular thickness (Tb.Th), Trabecular Separation (Tb.Sp), Cortical Thickness (Ct.Th) and Cortical Bone Area (Ct.BArea) (n = 6). W-Y The correlation analysis of serum AOPPs and BMD, ALP, FABP4. Data were shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01 vs. young mice group (Student's t test).

Fig. 2

Decreased AOPPs level by antioxidant leaded to higher bone formation and lower marrow adiposity in aged mice. A Serum AOPPs levels in aged and NAC-treated aged mice (n = 6). B AOPPs relative levels in bone marrow from aged and NAC-treated aged mice (n = 6). C Serum ALP level (n = 6). D Serum FABP4 level (n = 6). E, F HE staining of bone trabeculae and adipocytes in distal femora, and the area of fat vacuoles was quantified. Scale bars = 100 μm. G, H IHC staining for OCN and number of osteoblasts in distal femoral trabeculas. Scale bars = 20 μm. I, J IHC staining for FABP4 and number of adipocytes. Scale bars = 20 μm. K, L Micro-CT scanning of osmium tetroxide-stained tibia and MAT content below the growth plate (n = 3). M, N The protein and mRNA expression of OCN and PPARγ in proximal tibial metaphysis. O Representative micro-CT images of distal femora from aged and NAC-treated aged mice (n = 6) (Scale bars = 1 mm). P-V Quantitative trabecular and cortical bone parameters data were displayed in BMD, Tb.BV/TV, Tb.N, Tb.Th, Tb.Sp, Ct.Th and Ct.BArea (n = 6). Data were shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01 vs. aged mice group (Student's t test).

Fig. 3

Chronic AOPPs loading lead to the bone-fat imbalance in young mice. A Serum AOPPs levels in 6-month-old mice and AOPPs-intervened mice (n = 6). B AOPPs relative levels in bone marrow from 6-month-old mice and AOPPs-intervened mice (n = 6). C Serum ALP level (n = 6). D Serum FABP4 level (n = 6). E, F HE staining of bone trabeculae and adipocytes in distal femora, and the area of fat vacuoles was quantified. Scale bars = 100 μm. G, H IHC staining for OCN and number of osteoblasts in distal femoral trabeculae. Scale bars = 20 μm. I, J IHC staining for FABP4 and number of adipocytes. Scale bars = 20 μm. K, L Micro-CT scanning of osmium tetroxide-stained tibia and MAT content below the growth plate (n = 3). M, N The protein and mRNA expression of OCN and PPARγ in proximal tibial metaphysis. O Representative micro-CT images of distal femora from AOPPs-intervened mice (n = 6) (Scale bars = 1 mm). P-V Quantitative trabecular and cortical bone parameters data were displayed in BMD, Tb.BV/TV, Tb.N, Tb.Th, Tb.Sp, Ct.Th and Ct.BArea (n = 6). Data were presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 vs. PBS group (one-way ANOVA).

Fig. 4

MSCs from aged mice tended to differentiate into adipocytes rather than osteoblasts and exhibited cellular senescence. A, B Representative images of Alizarin Red S (ARS) staining and quantitative analysis of matrix mineralization in MSCs from young and aged mice. C, D Representative images of Oil Red O (ORO) staining and quantitative analysis of lipid droplet formation in MSCs from young and aged mice. Scale bar = 100 μm. E, F Representative images of beta-galactosidase staining and quantitative analysis of senescent MSCs. Scale bar = 100 μm. G qRT-PCR analysis of the mRNA level of osteogenic markers (ALP, RUNX2 and OCN) 、adipogenic markers (PPARγ and C/EBPα) and senescent biomarkers (P53, P21 and P16) in MSCs from young and aged mice. H AOPPs relative levels in cellular supernatant of MSCs derived from young and aged mice. n = 3 each group. Data were shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01 vs. young mice group (Student's t test).

Fig. 5

AOPPs induced the switch from osteogenic to adipogenic lineage and senescence in MSCs. A, B ARS staining and quantitative analysis of matrix mineralization in primary MSCs exposed to AOPPs (0–200 μg/ml). C Western blot analysis of the relative protein levels of ALP, RUNX2 and OCN in AOPPs-treated MSCs. D, E ORO staining and quantitative analysis of lipid droplets in primary MSCs exposed to AOPPs (0–200 μg/ml). Scale bar = 100 μm. F Immunoblotting analysis of PPARγ and C/EBPα protein levels in AOPPs-treated MSCs. G qRT-PCR analysis of the relative mRNA levels of ALP, RUNX2, OCN, PPARγ and C/EBPα. H, I Representative images of beta-galactosidase staining and quantitative analysis of senescent MSCs. Scale bar = 100 μm. J-M Protein levels of cellular senescence markers (P53, P21, P16), senescence-associated secretory phenotype (SASP) (MMP3, IL-6, IL-1β), senescence-associated heterochromatin foci (SAHF) (H3K9Me2), DNA damage (γ-H2AX) and multi-differentiation potential associated transcription factors (OCT-4, NANOG) were determined using immunoblotting. n = 3 each group. Data were presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 vs. control group (one-way ANOVA).

Fig. 6

AOPPs induced intracellular ROS generation by NADPH oxidase and mitochondria. A, B DCFH-DA was used to detect AOPPs-induced intracellular ROS generation in MSCs. C The activation of NADPH oxidase was analyzed by co-immunoprecipitation. D Western blot for the NADPH oxidase subunits NOX1, NOX2 and NOX4 protein levels in AOPPs-treated MSCs. E, F Immunoblotting showed the protein levels of mitochondrial respiratory chain complex Ⅰ (MRCC Ⅰ, NDUFS1), mitochondrial respiratory chain complex Ⅲ (MRCC Ⅲ, UQCRC2) and mitochondrial transcription factor A (TFAM) in MSCs treated by AOPPs. G Quantitative analysis of the activities of MRCC Ⅰ and Ⅲ enzyme in AOPPs-treated MSCs. H, I JC-1 staining fluorescence images of mitochondrial membrane potential (ΔΨm) level in AOPPs-induced MSCs. Numerical data were expressed in terms of the ratio of JC-1 monomers to JC-1 aggregates. Scale bar = 50 μm. J Quantitative analysis of ROS generation in AOPPs-induced MSCs after pretreated with Apocynin, Mito-TEMPO and NAC, respectively. n = 3 each group. Data were presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 vs. control group. #p < 0.05 vs. AOPPs group (one-way ANOVA).

Fig. 7

ROS signaling was involved in AOPPs-induced lineage switch and senescence of MSCs. A, B ARS staining and quantitative analysis of matrix mineralization in NAC-pretreated MSCs exposed to AOPPs. C Protein levels of ALP, RUNX2 and OCN induced by exposure of NAC pretreated MSCs to AOPPs. D, E ORO staining and quantitative analysis of lipid droplets. Scale bar = 100 μm. F Protein levels of PPARγ and C/EBPα. G mRNA levels of ALP, RUNX2, OCN, PPARγ and C/EBPα induced by exposure of NAC pretreated MSCs to AOPPs. H, I Representative images of beta-galactosidase staining and quantitative analysis of senescent MSCs. J-L Western blot analysis of the relative levels of P53, P21, P16, MMP3, IL-6, IL-1β, H3K9Me2 and γ-H2AX protein expression in NAC-pretreated MSCs exposed to AOPPs. n = 3 each group. Data were presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 vs. control group. #p < 0.05 vs. AOPPs group (one-way ANOVA).

Fig. 8

AOPPs induced lineage switch and senescence of MSCs by activating PI3K/AKT/mTOR pathway. A, B The protein levels of the PI3K/AKT/mTOR pathway in MSCs treated with AOPPs for varying periods and pretreated with NAC. C, D ARS staining and quantitative analysis of matrix mineralization in LY294002 or NAC-pretreated MSCs exposed to AOPPs. E, F ORO staining and quantitative analysis of lipid droplets. Scale bar = 100 μm. G, H Representative images of beta-galactosidase staining and quantitative analysis of senescent MSCs. n = 3 each group. Data were presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 vs. control group. #p or αp < 0.05 vs. AOPPs group (one-way ANOVA).

Fig. 9

Schematic representation of AOPPs-induced lineage switch and senescence of MSCs via ROS accumulation and activation of PI3K/AKT/mTOR pathway, resulting in bone-fat imbalance during skeletal aging.

figs1

Fig. S1. Bone mass loss and deterioration of bone microarchitecture of L4 vertebral body in aged mice. A Representative μCT images of L4 vertebral body from young and aged mice. B-F Quantitative trabecular parameters data were displayed in Bone Mineral Density (BMD), Vertebral Bone Volume/Total Volume (Vt.BV/TV), Vertebral Trabecular Number (Vt.Tb.N), Vertebral Trabecular Thickness (Vt.Tb.Th) and Vertebral Trabecular Separation (Vt.Tb.Sp) (n = 6). Data were shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01 vs. young mice group.

figs2

Fig. S2. AOPPs accumulated significantly in the elderly. Blood samples were collected from a total of 46 participants, who underwent physical examination in the health examination center of our hospital. ELISA was used to detect serum AOPPs levels. The research was performed in compliance with the World Medical Association Declaration of Helsinki, and all participants signed informed consent. The study was approved by the Ethics Committee of Nanfang Hospital, Southern Medical University (NFEC-2021-155).

figs3

Fig. S3. NAC alleviated the deterioration of bone microarchitecture of L4 vertebral body in aged mice. A Representative μCT images of L4 vertebral body from aged and NAC-treated aged mice. B-F Quantitative trabecular parameters data were displayed in BMD, Vt.BV/TV, Vt.Tb.N, Vt.Tb.Th and Vt.Tb.Sp (n = 6). Data were shown as mean ± SD. ∗P < 0.05, ∗∗P < 0.01 vs. aged mice group.

figs4

Fig. S4. Chronic AOPPs loading accelerated bone mass loss and bone microarchitecture deterioration in young mice. A Representative Micro CT 3D reconstruction and cross-sectional images of L4 vertebral body from AOPPs-treated mice. B-F Quantitative trabecular parameters data were displayed in BMD, Vt.BV/TV, Vt.Tb.N, Vt.Tb.Th and Vt.Tb.Sp (n = 5). Data were presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 vs. PBS group.