Osteoporosis (OP), a prevalent degenerative bone disorder, affects an estimated 50% of women and approximately 33% of men, typically manifesting in individuals in their fifth and sixth decades of life [1,2,3]. OP is characterized by dysregulated bone resorption, leading to decreased bone mineral density, compromised microarchitecture, and structural deterioration, thereby increasing the susceptibility to fractures [4]. Additionally, secondary osteoporotic fractures are a significant contributor to morbidity and mortality among the elderly population [5]. The presence of these fragile fractures is associated with a notable increase in mortality rates and a substantial decline in the overall well-being of those affected [6]. Recent studies have identified a discrepancy in the differentiation of bone mesenchymal stem cells (BMSCs) in individuals with OP, wherein BMSCs within the bone marrow exhibit a preference for adipocyte differentiation over osteocyte differentiation [7, 8]. This abnormal shift towards adipogenesis and reduction in osteoblast numbers within the bone marrow contribute to the development of bone loss [9].

Reactive oxygen species (ROS) within the bone microenvironment are acknowledged as significant regulators and potential targets for the modulation of bone metabolism [10, 11]. The abnormal accumulation of ROS within the inflammatory microenvironment, along with the infiltration of inflammatory factors, plays pivotal roles in bone metabolic processes [12, 13]. Ultimately, cellular dysfunction leads to a decrease in lipid metabolism levels and stimulates the overproduction of ROS [14]. This excessive ROS impedes osteogenic differentiation while promoting lipogenic differentiation in BMSC [15]. Furthermore, the heightened lipogenic differentiation serves to enhance the production of additional ROS [16]. Recent advancements in the field of nanomedicine have introduced novel approaches for addressing OP [17]. Nanozymes, a class of nanomaterials, demonstrate catalytic properties akin to those of endogenous enzymes, providing advantages including cost-effectiveness, ease of large-scale production, exceptional stability, and prolonged shelf-life [18]. The utilization of nanozymes with characteristics akin to superoxide dismutase (SOD) and catalase (CAT) is expected to contribute to the management of OP by eradicating ROS [19]. This process is projected to improve lipid metabolism and promote the differentiation of BMSCs towards osteogenesis [20].

Zeolitic imidazolate frameworks (ZIFs), which are topological isomers synthesized using zinc ions and imidazole linkers [21, 22], are classified within the subset of metal–organic frameworks (MOFs) [23, 24]. In vitro studies have shown that ZIF exhibits stability at a pH of 7.4, a level considered physiological, for a period exceeding 15 d. Moreover, it has been established to possess biocompatibility [25]. Additionally, nanoparticles derived from ZIF demonstrate favorable cellular uptake properties [26, 27]. Consequently, the creation of a ZIF with SOD-like enzyme activity could potentially sustain its ability to scavenge reactive oxygen species over an extended period. Magnesium (Mg) deficiency has been associated with a variety of diseases, such as OP, abnormal blood lipid levels, hypertension, atherosclerosis, arrhythmias, and myocardial infarction. The hypothesis posits that reduced Mg levels in the body play a role in the development of oxidative stress [28]. Initially, hypomagnesemia leads to elevated levels of pro-inflammatory cytokines and neutrophils, as well as macrophage activation. NADPH oxidase activity results in these cells functioning as a source of superoxide anion. Additionally, it has been observed that reduced Mg levels can result in mitochondrial dysfunction and ROS [29]. Moreover, it is theorized that Mg plays a protective role in shielding cell membranes from oxidative damage [30].

In light of these findings, the proposal involves the development of a nanozyme utilizing Mg-ZIF to maintain a sustained capacity for ROS scavenging. This strategy capitalizes on the stability of ZIFs and the documented antioxidative properties of Mg, which can further fortify cells against oxidative stress-induced damage. Ultimately, this approach holds promise as a more effective therapeutic intervention for OP Scheme 1.

Scheme 1

Mg-ZIF regulates osteogenic and adipogenic through lipid metabolism

Experimental sectionMaterials

All commercial reagents and solvents were of A. R. grade and were used without further purification. Mg (NO3)2·6H2O, Zn (NO3)2·6H2O, and N, N-dimethylformamide (DMF) purchased from Aladdin (Shanghai, China). CCK-8 purchased from Abbkine (Wuhan, China). Calcein AM/PI staining, DCFH-DA kit, Alkaline Phosphatase Assay Kit were purchased from Beyotime (Beijing, China). Alizarin Red S Solution obtained from Solarbio (Beijing, China). The Real-time PCR Mix and qPCR master mix were purchased from Vazyme (Nanjing, China). SOD enzyme activity was measured using commercially available kits (Beyotime, Beijing) and CAT enzyme activity was measured using a dissolved oxygen meter.

Characterization

The single-crystal x-ray diffraction method (D8 Advance, Bruker-AXS, Germany) was employed to analyze the crystal structure, while FTIR spectroscopy (670-IR + 610-IR, Varian, USA) was used for the characterization of the chemical composition and structure. Transmission electron microscopy (Tecnai 12, Philips, Netherlands) was used to acquire information on the morphology and composition of the samples. Elemental mapping and contents were analyzed by high power transmission electron microscopy (HT7800, Hitachi, Japan).

Cell culture

BMSCs were obtained from four male C57BL/6 J mice sourced from Jinan University's Experimental Animal Center. Briefly, the mice were killed and sterilized by immersion in 75% alcohol. Next, femurs were acquired by dissecting both lower extremities and the bone marrow was washed with PBS. Following lysis using erythrocyte lysate, the primary BMSCs were fragmented into individual cell suspensions, then cultivate for 7 d and use for subsequent cell experiments after reaching full growth. The BMSCs were identified by LepR immunostaining (Fig. S1).

Synthesis of Mg-ZIF

A typical one-pot method was used to synthesize ZIF and Mg-ZIF [31]. In brief, after dissolving Mg (NO3)2 · 6H2O (1.0 mmol, 0.25641 g) and Zn (NO3)2 · 6H2O (1.0 mmol, 0.29749) in DMF, the mixture is stirred for a duration of 20 min. Subsequently, 10 mL of a N, N-Dimethylformamide (DMF) solution containing dimethylimidazole (6.0 mmol, 0.4926) was incrementally added to the reaction mixture. The resulting mixture was subjected to overnight stirring, followed by centrifugation at 10,000 rpm for 10 min to isolate the precipitate. The precipitate was then subjected to three washes with alcohol and subsequently dried for 12 h in a drying oven set at 60 ℃. With respect to the synthesis of ZIF, Mg (NO3)2 · 6H2O is replaced by 1 mmol of Zn (NO3)2 · 6H2O in the first step, and the rest of the steps remain the same.

Cell proliferation and live/dead staining assay

BMSCs were cultured on 96-well plates at a cell density of 8000 cells per well. The cells were then cultured for 24 h to facilitate cell adhesion. The viability of cells was assessed by employing the Cell Counting Kit 8 (Sigma-Aldrich, Shanghai, China) as per the instructions provided by the manufacturer, following co-incubation with nanoparticles for 1, 3, and 7 d. Likewise, the viability of cells was assessed by employing the live/dead staining kit in accordance with the guidelines provided by the manufacturer. Calcein AM/PI double staining (Beyotime) serves as a cell death assay to quantify the population of live and dead cells. In short, cells were subjected to testing with Calcein AM (for viable cells) and PI (for non-viable cells) solutions at 37 °C for a duration of 30 min in the absence of light. Subsequently, the resulting image was analyzed using a fluorescence microscope.

Assessment of osteogenic activity of BMSCs

Regarding the ALP activity assay (Beyotime, Beijing), BMSCs were cultured with osteogenic differentiation medium for 7 d according to control, H2O2 (200 μM), H2O2 + ZIF, and H2O2 + Mg-ZIF grouping, and ALP activity was assayed using a kit. Next, to confirm mineralization, BMSCs were induced for 21 d, and these cells were fixed using 70% ethanol and then stained for mineralized nodule formation using 2% alizarin red.

Induction of adipogenesis of BMSCs

To stimulate adipogenesis, BMSCs were seeded into plates with a density of 4 × 104 per well. Adipogenesis induction was performed when the cell density reached about 90% according to different groupings. The medium for inducing adipogenesis was split into two solutions, adipogenesis A and B. Adipogenic solution A consisted of a full medium with the addition of 0.5 M dexamethasone, 0.25 mM IBMX, 0.2 mM rosiglitazone, and 10 μg/mL insulin. The solution B for adipogenesis consisted of a fully supplemented medium containing 10 μg/mL of insulin. Solution A was utilized for a duration of 3 d during the initiation of adipogenesis, followed by the utilization of solution B for 1 d. After 21 d of inducing adipogenesis, staining for adipogenesis was conducted.

Analysis using quantitative real-time polymerase chain reaction (qRT-PCR)

First, BMSCs were inoculated at 105 per well in 6-well plates and cultured with different treatment groups for 3 d after 24 h of adherence. Subsequently, total RNA was isolated from the cells through the addition of TRIzol reagent (Beyotime, China), followed by the generation of cDNA using TaKaRa reverse transcription reagent (TaKaRa, Shiga, Japan). Subsequently, a qPCR assay was conducted to analyze the expression of pertinent genes. Real-time PCR was performed using a 20-μL reaction mixture containing SYBR Green qPCR Master Mix, forward and reverse primers, DEPC water, and cDNA template. Femoral head tissue was mechanically disrupted and ground with liquid nitrogen for RNA extraction, followed by cDNA synthesis. Real-time PCR was conducted using a Light Cycler instrument (Roche, Basel, Switzerland). The qPCR conditions were as follows: 1 cycle of initial denaturation at 95 °C for 4 min followed by 40 cycles of denaturation at 95 °C for 15 s, annealing at 60 °C for 30 s, and elongation at 72 °C for 30 s. The 2−ΔΔCt method was applied and gene expression was normalized with Actin. Primers are listed in Table S1 [32,33,34,35,36,37,38,39].

Animal model

Female C57BL/6 J mice, aged 5 weeks, were acclimatized and fed for 1 week before being divided into groups for ovariectomy (OVX) or sham surgery. Following OVX procedure, a single injection of nanozyme was administered into the bone marrow cavity immediately at a dosage of 5 mg/kg, and no further injections were given during the subsequent period. The groups included sham, OVX, OVX + ZIF, and OVX + Mg-ZIF. The femur was taken 6 weeks after treatment for follow-up experiment. Over a period of 6 weeks, animals were housed in cages containing 5 individuals each and provided with food and shelter under standardized conditions, including a temperature of 21 °C and a relative humidity of 55%, as well as a 12 h light/dark cycle. All animal procedures were conducted in accordance with the guidelines set forth by the Institutional Animal Care and Use Committee of Jinan University (Approval number: IACUC-20230617–15) and adhered to the "Guide for the Care and Use of Laboratory Animals" established by the National Institute of Health in China.

Micro-computed tomography reconstructions, immunofluorescence and Trap staining

Following the euthanization of the mice, the femurs were gathered and immersed in 10% neutral buffered formalin for a duration of 24 h. Place the femur vertically in the tube sample holder, and then fix it in the foam to avoid slight movement that may affect the high-resolution scanning. By utilizing the SkyScan1176 micro-CT, one can examine the alterations in the microstructure of the femur and acquire three-dimensional reconstructed visuals. Bone parameters were then analyzed by CTAn on the 3D reconstructed images of the bone. Femurs were decalcified in 10% EDTA for 10 d, embedded, cut into 4 µm sections, and stained for OCN and Trap.

Statistical analysis

Experiments were repeated three times and data were presented as mean ± SD. Statistical analyses were conducted using GraphPad Prism 9 software with significance set at p < 0.05. (* P < 0.05, ** P < 0.01, *** P < 0.001).