Role of Zinc Homeostasis in the Pathogenesis of Diabetic Osteoporosis in Mice

Abstract

Background: Diabetes induces osteoporosis primarily by impairing osteoblast function. Intracellular zinc homeostasis, which is controlled by zinc transporters, plays a significant role in osteoblast differentiation. In the present study, we aimed to explore the role of zinc homeostasis in the pathogenesis of diabetic bone loss using a diabetic mouse model. Methods: Streptozotocin (STZ)-induced diabetic female mice were used for in vivo experiments. In vitro, the effects of zinc transporter knockdown using small interfering RNA was investigated in MC3T3E1 pre-osteoblastic cells. Results: STZ-induced diabetic mice exhibited severe bone loss and decreased expression of osteogenic genes, as well as a decrease in zinc content and the expression of several zinc transporters localized in the cellular membrane, including Zip6, Zip9, and Zip10 in the tibia. Moreover, the messenger RNA (mRNA) levels of Zip6, Zip9, and Zip10 were positively correlated with trabecular bone mineral density in the tibiae of diabetic mice. This in vitro study, using MC3T3E1 pre-osteoblastic cells, revealed that knockdown of Zip6 reduced the expression of osteogenic genes in pre-osteoblastic cells. Additionally, Zip6 knockdown downregulated protein levels of phosphorylated p38 mitogen-activated protein kinase (p38MAPK) in pre-osteoblastic cells, and this change was observed in the tibiae of diabetic mice. Conclusions: Our data suggest that the downregulation of zinc transporters localized in the cellular membrane, such as Zip6, may be involved in the impairment of osteoblastic differentiation through the inhibition of p38 MAPK signaling, leading to osteoporosis under diabetic conditions. Maintaining zinc homeostasis in bone tissues may be vital for preventing and treating diabetic bone loss, and zinc transporters may serve as novel therapeutic targets for diabetic osteoporosis.

1. Introduction

Recently, it has been widely recognized that osteoporosis is a serious diabetes-related complication [1]. Accumulating clinical evidence has suggested that fracture risk is higher in patients with type 1 and those with type 2 diabetes than in healthy individuals [2,3], and, in particular, patients with type 1 diabetes have significant bone loss, leading to severe bone fragility [4]. The abnormal balance between bone formation and resorption is the direct cause of osteoporosis. Although the influence of type 1 diabetes on bone resorption remains controversial, a decline in bone formation caused by impaired osteoblast differentiation has been observed in diabetic patients and animal models [5], indicating that impaired osteoblast differentiation is considered one of the major factors contributing to diabetic osteoporosis. This impairment has been suggested to result from various detrimental factors induced by diabetes, including hyperglycemia, insulin deficiency, insulin resistance, chronic inflammation, oxidative stress, and increased levels of advanced glycation end products (AGEs) [6,7]. However, the molecular mechanisms underlying type 1 diabetes-induced osteoporosis are not yet fully understood.

Zinc is a micronutrient essential for regulating human metabolism, including bone metabolism [8]. An in vitro study demonstrated that zinc treatment accelerates osteoblast differentiation in human bone marrow-derived mesenchymal stem cells (MSCs) [9]. Conversely, zinc represses the differentiation into osteoclast and its activity in vitro [10]. These actions of zinc in osteoblasts and osteoclasts have also been observed in experimental animal studies [11,12]. Regarding the relationship between zinc and diabetic osteoporosis, human studies have indicated that circulating levels of zinc are lower in diabetic patients than in healthy individuals [13,14], and zinc intake positively correlates with circulating levels of osteocalcin, a marker for bone formation, in patients with type 1 diabetes [15]. Moreover, in vivo studies using diabetic rats have shown the lower levels of serum zinc [16], as well as the preventive effect of zinc supplementation against bone loss in the diabetic condition [16,17,18]. These findings suggest that changes in zinc homeostasis within bone tissues are associated with abnormalities in bone metabolism under diabetic conditions.

Recent studies have revealed that multiple zinc transporters localized in cellular or organelle membranes tightly control intracellular zinc homeostasis [19]. Zinc transporters are divided into the SLC39A/Zip and SLC30A/ZnT families. The Zip family (Zip1-14) increases the levels of zinc ions in the cytoplasm by transporting them from the extracellular matrix or organelles to the cytoplasm, whereas the ZnT family (ZnT1-10) decreases the levels of zinc ions in the cytoplasm by transporting them from the cytoplasm to the extracellular matrix or organelles [19]. Zinc ions transported by these transporters regulate various signaling pathways, including the phosphatidylinositol-3-kinase (PI3K)/Akt and mitogen-activated protein kinase (MAPK) signaling pathways [20,21]. Studies have also indicated that zinc transport mediated by several zinc transporters, including Zip1, Zip14, and ZnT5, positively influences osteoblast differentiation [22,23,24]. Furthermore, Zip13 has been identified as a causative gene for the novel Ehlers–Danlos syndrome in humans, characterized by osteogenesis imperfecta and collagen tissue malformation [25]. This suggests that zinc homeostasis, which is regulated by zinc transporters, plays an important part in maintaining bone health in humans. However, the role of zinc transporters in intracellular zinc homeostasis during the pathogenesis of diabetic osteoporosis remains unclear.

In the present study, we investigated the influence of the diabetic condition on the expressions of zinc transporters in bone tissues using a streptozotocin (STZ)-induced diabetic mouse model to clarify whether changes in zinc dynamics in bone tissue are involved in the pathological mechanism of osteoporosis induced by type 1 diabetes.

2. Materials and Methods

2.1. Materials

Female C57BL/6J mice and a standard diet (MF) were purchased from CLEA Japan, Inc. (Tokyo, Japan) and Oriental Yeast Co., Ltd. (Tokyo, Japan), respectively. STZ and the PI3K inhibitor LY294002 were purchased from Sigma-Aldrich (Tokyo, Japan). Control small interfering RNA (siRNA) and siRNA for Zip6, Zip9, Zip10, and ZnT6 were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Anti-phospho-Akt (Ser473) (#9271), anti-Akt (#9272), phospho-p38 MAPK (Thr180/Tyr182) (#9211), p38 MAPK (#9212), and anti-rabbit IgG HRP-linked antibodies (#7074) were purchased from Cell Signaling Technology Japan (Tokyo, Japan). Anti-Phospho-ERK1 (T202/Y204)/ERK2 (T185/Y187) (#MAB1018) and anti-ERK1/ERK2 (#MAB1576) were purchased from R&D Systems, Inc. (Minneapolis, MN, USA).

2.2. Animal Experiments

Thirteen female C57BL/6J mice at 6 weeks old were purchased from CLEA Japan, Inc. (Tokyo, Japan). Mice were housed in cages in a breeding room controlled at a room temperature of 21 ± 1 °C and humidity of 55±5%. Food and water were available ad libitum. After a week, mice were randomly divided into a control and diabetic group. Type 1 diabetic mice were generated using STZ, a pancreatic β-cell cytotoxin, as previously described with modifications [26]. Briefly, female C57BL/6J mice at 7 weeks old (n = 7) were injected once intraperitoneally with STZ (200 mg/kg body weight in saline) to induce diabetes, whereas control mice (n = 6) were injected once intraperitoneally with saline only. After 4 days, the blood glucose levels (non-fasting state) were measured using a Glutest sensor (Sanwa Kagaku Kenkyusyo, Nagoya, Japan), and mice with elevated blood glucose levels > 300 mg/dL were considered in the diabetic state. No mice were excluded in this study. After 4 weeks, the bone mineral density (BMD) values in their tibiae were estimated using computed tomography (CT). After 6 h of fasting, control and diabetic mice were alternately anesthetized with isoflurane, and then mice were euthanized by blood collection by cardiac puncture, and the tibiae were immediately dissected.

All experiments were conducted in accordance with the National Institutes of Health guidelines and the institutional rules for the use and care of laboratory animals of Kobe Gakuin University (approval number: 18-31).

2.3. Quantitative CT (qCT) Analysis

For the measurement of BMD in the tibiae of mice, qCT analysis was performed under isoflurane anesthesia using a LaTheta LCT-200 experimental animal CT system (Hitachi Aloka Medical, Tokyo, Japan), as previously described with slight modifications [27]. Briefly, the trabecular regions of interest (ROIs) for the assessment of BMD, total volume (TV), bone volume (BV), and trabecular area were located from 96 μm distal to the proximal growth plate to 1500 μm of the diaphysis. Cortical ROIs for evaluating the BMD, thickness, and area were located 2000 μm from the segment in the middle of the tibia. The ratio of BV (mm³) to TV (mm³) was defined as the bone volume fraction. CT was performed with the following settings: voxel size, 48 × 96 μm with a slice thickness of 96 μm; axial field-of-view, 48 mm; tube voltage, 50 kVp; tube current, 500 μA; and integration time, 3.6 ms. LaTheta software (version 3.40) was used for analyzing all bone parameters.

2.4. Blood Analysis

Serum concentrations of zinc and insulin were measured using a Zinc Assay kit LS (Metallogenics Co., Ltd., Chiba, Japan) and an ultrasensitive mouse insulin ELISA kit (Morinaga Institute of Biological Science, Yokohama, Japan), respectively.

2.5. Cell Culture

MC3T3E1 pre-osteoblastic cells (mouse osteoblastic cells, American Type Culture Collection, Manassas, VA, USA) were cultured in α-MEM (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) with 10% fetal bovine serum (Biowest Inc., East Sussex UK), 1% penicillin–streptomycin, and 1% Glutamax (ThermoFisher Scientific, Tokyo, Japan). For pharmacological inhibition of insulin signaling, MC3T3E1 cells that reached 70–80% confluence were treated with the PI3 kinase inhibitor LY294002 (1 mM) for 24 h.

2.6. Transfection with siRNA

MC3T3E1 pre-osteoblastic cells were seeded at a density of 1.5 × 10⁵ cells per well on a six-well plate. After reaching approximately 70% confluence, the transfection of each siRNA (control siRNA; siRNA targeting Zip6, Zip9, Zip10, and ZnT6; and the insulin receptor) at a concentration of 10 nM into cells was performed using Lipofectamine RNAi Max transfection reagent (Life Technologies Japan, Tokyo, Japan). Cells were harvested for 24 or 48 h after siRNA transfection.

2.7. Real-Time Polymerase Chain Reaction (RT-PCR)

After homogenization of tibiae samples (whole bone tissues including bone marrow) using Micro Smash MS100R (TOMY Seiko, Tokyo, Japan), total RNA was isolated from the samples using a RNeasy Mini Kit (Life Technologies Japan). Total RNA of the harvested cells was also isolated using the same kit. cDNA was synthesized from 1 µg of isolated total RNA using a High-Capacity cDNA Reverse Transcription Kit (ThermoFisher Scientific, Tokyo, Japan). A FastStart Essential DNA Green Master Mix and LightCycler 96 (Roche Diagnostics Japan, Tokyo, Japan) were used to perform RT-PCR. The list of primer sequences used for RT-PCR are shown in Table S1. As internal control genes, the mRNA levels of β-actin were measured, and each mRNA level was normalized by the mRNA levels of β-actin.

2.8. Protein Isolation and Western Blotting

Isolation of total protein from the homogenized tibia of mice and cultured cells was performed using cOmplete Lysis-M EDTA-free buffer (Roche Diagnostics Japan) and RIPA buffer (Cell Signaling Japan), with protease and phosphatase inhibitors added, respectively. The protein concentration in the lysate was determined using the Pierce^TM BCA Protein Assay Kit (Thermo Fisher Scientific). Ten milligrams of protein were loaded into each well of a 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis gel and separated by electrophoresis for 1 h at 200 V. Then, the proteins were transferred to a polyvinylidene fluoride membrane for 45 min at 100 V. After blocking the membrane using 3% skim milk in Tris-buffered saline containing 0.05% Tween 20 (TBST) for 1 h at room temperature, the membrane was incubated with the primary antibody overnight at 4 °C, followed by incubation with the secondary antibody at room temperature. The primary antibodies were as follows: phospho-Akt (Ser473; 1:1000), Akt (1:1000), phospho-p38 MAPK (Thr180/Tyr182; 1:1000), p38 MAPK (1:1000), phospho-ERK1 (T202/Y204)/ERK2 (T185/Y187; 1:1000), and ERK1/ERK2 (1:1000). The anti-rabbit IgG HRP-linked antibody (1:1000) was used as the secondary antibody. Blot detection was performed using ECL Prime Western blotting detection reagents (Cytiva, Tokyo, Japan), and chemiluminescent bands were visualized using a Lumino Graph I (ATTO, Tokyo, Japan) and quantified using CSAnalyzer 4 (ATTO).

2.9. Measurement of Zinc Content in Tibia

Mouse tibiae were homogenized in cOmplete Lysis-M EDTA-free buffer (Roche Diagnostics Japan). The zinc content in the lysate was measured using a Zinc Assay kit LS (Metallogenics Co., Ltd.). Zinc levels in the tibiae of mice were normalized to the protein concentration.

2.10. Statistical Analysis

Data are presented as the mean ± standard error of the mean (SEM). Student’s t-test was used to compare data between two groups when the data presented a normal distribution, assessed by the Shapiro–Wilk test. The Mann–Whitney U test was used to compare data between two groups when the data did not present a normal distribution. Pearson’s test was used for correlation analysis. When the p-value was <0.05, differences and correlations were considered statistically significant. GraphPad Prism 8.2.0. was used to perform all statistical analyses.

3. Results

3.1. Zinc Levels in Bone Tissues Were Decreased in Diabetic Mice

A reduction in body weight, a marked elevation in blood glucose levels, and a significant reduction in serum insulin levels were observed in diabetic mice compared with control mice (Figure 1A–C). Although no significant difference was observed in serum zinc levels between control and diabetic mice (Figure 1D), zinc levels in the tibiae of diabetic mice were reduced by 64% compared with those in the tibiae of control mice (Figure 1E).

3.2. Diabetic Mice Showed Severe Bone Loss and Impaired Osteogenesis

Diabetic mice showed a significant reduction in bone parameters, such as BV/TV, trabecular BMD, trabecular area, minimum moment of inertia of the cross-sectional area (bending strength), cortical BMD, and cortical thickness in the tibiae compared with control mice (Figure 2A–F), suggesting severe bone loss in diabetic mice. They also showed a marked reduction in the levels of osteogenic genes, such as Runx2, Osterix, Alp, and Osteocalcin, in the tibiae (Figure 2G), suggesting the impairment of osteogenesis in the tibiae of diabetic mice. The levels of marker genes of bone resorption (receptor activator of nuclear factor-kappa B ligand (Rankl) and osteoprotegerin (Opg)) were decreased in the tibiae of diabetic mice compared with those of control mice (Figure 2H). The levels of adipogenic marker gene Fabp4, but not Pparγ, were significantly elevated in the tibiae of diabetic mice compared with those of control mice (Figure 2I).

3.3. The Expression of Several Zinc Transporters in Bone Tissues Was Decreased in Diabetic Mice

Next, we comprehensively measured the expression levels of zinc transporters (Zip and ZnT families) and metallothioneins in the tibiae of mice to clarify the changes in zinc dynamics in bone tissues. The mRNA levels of several zinc transporters, such as Zip6 (cellular membrane), Zip9 (cellular membrane and Golgi apparatus), Zip10 (cellular membrane), and ZnT6 (Golgi apparatus), were significantly decreased in the tibiae of diabetic mice compared with those of control mice (Figure 3A,B). No difference was observed in the mRNA levels of Metallothionein in the tibiae between diabetic and control mice (Figure 3C).

3.4. Zinc Transporters Localized on the Cellular Membrane Were Positively Correlated with BMD and the Expression of Osteogenic Genes in the Bone Tissues of Diabetic Mice

We next assessed the correlation between bone parameters such as trabecular BMD, the expression of osteogenic genes in the tibiae, and the expressions of zinc transporters, which were decreased in the tibiae of diabetic mice. The mRNA levels of Zip6, Zip10, and Zip9 were positively correlated with trabecular BMD in diabetic mice (Figure 4A–C), whereas there was no correlation between the mRNA levels of ZnT6 and trabecular BMD in diabetic mice (Figure 4D). The mRNA levels of Zip6 and Zip10 were also strongly and positively correlated with the mRNA levels of Runx2, Osterix, and Osteocalcin in the tibiae of diabetic mice, respectively (Figure S1).

3.5. Knockdown of Zip6 Suppresses the Expression of Osteogenic Genes in Pre-Osteoblastic Cells In Vitro

We examined the effect of knocking down zinc transporters (Zip6, Zip9, Zip10, and ZnT6), which were found to be decreased in the tibiae of diabetic mice, in MC3T3E1 pre-osteoblastic cells. Knockdown of Zip6 significantly downregulated the mRNA levels of Runx2, Osterix, and Osteocalcin 24 h after the siRNA transfection, while knockdown of Zip10 showed a similar trend (Figure 5A,B). In contrast, knockdown of Zip9 and ZnT6 did not suppress the levels of osteogenic genes in MC3T3E1 cells 24 h after the siRNA transfection (Figure 5C,D). Moreover, knockdown of Zip6 further decreased the expression of osteogenic genes, such as Osterix, Alp, and osteocalcin in MC3T3E1 cells 48 h after the siRNA transfection (Figure S2).

3.6. The Molecular Pathway by Which Zip6 Knockdown Suppresses Osteogenesis in Pre-Osteoblastic Cells In Vitro

Next, we examined the influence of Zip6 knockdown on osteogenic signaling pathways, including the Akt and MAPK signaling pathways, as well as the expression of growth factors, to elucidate the molecular mechanism by which Zip6 knockdown decreased the expression of osteogenic genes in pre-osteoblastic cells. Knockdown of Zip6 significantly suppressed the phosphorylation of p38MAPK, but not that of Akt and ERK, in MC3T3E1 cells (Figure 6A). Additionally, knockdown of Zip6 significantly decreased the mRNA levels of Igf-1, while no significant changes were observed in the mRNA levels of Fgf-2 and Tgf-β in these cells (Figure 6B).

3.7. Influence of a Diabetic State on the Osteogenic Signaling Pathway in the Tibiae of Mice

Then, we examined whether the signaling pathways altered by Zip6 knockdown in pre-osteoblastic cells were also observed in the tibiae of diabetic mice in vivo. Diabetic mice exhibited a marked decrease in the phosphorylation of p38MAPK, ERK, and Akt in their tibiae compared with control mice (Figure 7A). Although there were no differences in the mRNA levels of Fgf-2 and Tgf-β between the tibiae of control and diabetic mice, the mRNA levels of Igf-1 were decreased in diabetic bone tissues (Figure 7B).

3.8. Inhibition of PI3 Kinase and Knockdown of Insulin Receptor Suppressed the Expressions of Zip6 and Zip10 in Pre-Osteoblastic Cells

Finally, we examined the effect of inhibiting PI3 kinase, which regulates various molecular pathways including insulin signaling, and knockdown of the insulin receptor on the expressions of Zip6 and Zip10 in pre-osteoblastic cells to clarify the mechanism by which diabetes induces changes in zinc dynamics in the bone tissue of diabetic mice. Inhibition of PI3K through treatment with the PI3K inhibitor LY294002 significantly decreased the gene expressions of Zip6 and Zip10 in MC3T3E1 cells (Figure 8A). Moreover, knockdown of the insulin receptor significantly reduced the gene expressions of Zip6 and Zip10 in MC3T3E1 cells (Figure 8B).

4. Discussion

Previous evidence provided by human and animal studies have suggested that diabetes-induced bone loss is caused by a reduction in bone formation due to impaired osteogenesis and an enhancement of adipogenic differentiation from MSCs [6,7,28]. Our previous and current studies demonstrated a reduction in the levels of osteogenic genes and an elevation of the levels of adipogenic gene Fabp4 in bone tissues of type 1 diabetic model mice [26], suggesting impaired osteoblastic differentiation and enhanced adipogenesis in the bone tissue of this experimental model. In contrast, we observed a decrease in the mRNA levels of bone resorption markers in the tibiae of diabetic mice in the present study. Our previous study has also shown that the mRNA levels of RANKL and the number of osteoclasts were decreased in the bone tissue of STZ-treated female mice compared to control female mice [26]. These data suggest that bone resorption was suppressed in type 1 diabetic female mice in the present study. Enhanced bone resorption has been observed in diabetic rat models, while reduced bone resorption has been noted in diabetic mouse models; however, the role of bone resorption in diabetic osteoporosis in humans remains controversial [5]. The present study findings indicate that a major cause of osteoporosis under diabetic conditions is reduced bone turnover due to impaired osteoblast differentiation.

In this study, we investigated the role of zinc dynamics in impaired osteoblast differentiation in type 1 diabetic mice. We observed decreased osteogenic gene expression and bone mass, along with reduced zinc levels in the tibiae of diabetic mice. Similarly, Fushimi et al. showed that a zinc-deficient diet exacerbated osteoporosis in STZ-induced diabetic rats [29]. Moreover, several studies provided evidence for the protective effects of zinc supplementation against diabetic osteoporosis in rats [16,17,18]. For instance, Iitsuka et al. showed that dietary supplementation with zinc increased zinc content and prevented a reduction in osteogenic gene levels and bone loss in the bone tissues of STZ-induced diabetic rodents [18]. In line with this, Bortolin et al. found that zinc supplementation restored serum zinc levels and had a preventive effect by enhancing osteogenesis [16]. Additionally, in vitro experiments using pre-osteoblasts and MSCs revealed that zinc treatment promoted osteogenesis by enhancing the expressions of Runx2 and Osterix [9,30]. These findings suggest that changes in zinc dynamics in bone cells contribute to the regulation of osteogenesis under diabetic conditions.

Here, the expressions of zinc transporters were comprehensively analyzed in bone tissue to elucidate how zinc dynamics are affected by diabetes. Zinc influx into cells is tightly controlled by multiple zinc transporters located in the cell membrane. Regarding the relationship between cell membrane zinc transporters and osteoblast function, Tang et al. showed that the overexpression of Zip1 in the cell membrane promotes the osteoblast differentiation from MSCs [22]. In the present study, although we observed no change in the expression of Zip1 in bone tissue due to diabetes, we found that the expressions of Zip6, Zip9, and Zip10, also localized in cell membranes, were significantly decreased. Importantly, Zip6 facilitates the transport of zinc from the extracellular space to the intracellular space by forming a heteromer with Zip10 [31]. Furthermore, previous reports indicated that the zinc influx through the heteromers of Zip6 and Zip10 is the trigger for mitosis [32] and that the control of intracellular zinc levels by Zip6 and Zip10 plays a significant role in regulating the invasion of cancer cells and epithelial-to-mesenchymal transition [33]. The present study showed that the mRNA levels of Zip6 and Zip10 were significantly and positively correlated with trabecular BMD and the levels of osteogenic genes (Runx2, Osterix, and Osteocalcin) in the bone tissue of diabetic mice. These results suggest that changes in the expressions of these zinc transporters are involved in regulating bone mass and osteoblast differentiation under diabetic conditions. Moreover, knockdown of Zip6 and Zip10 in MC3T3E1 cells led to a significant decrease, or a tendency to decrease, in osteogenic genes such as Runx2, Osterix, and Osteocalcin, respectively, whereas these effects were not observed after Zip9 knockdown. This suggests that Zip6 and Zip10 (particularly Zip6) are directly associated with osteogenic differentiation. Taken together, these findings indicate that the decrease in zinc transporters, such as Zip6 and Zip10, located in the cell membrane contributes to impaired zinc homeostasis in osteoblasts, leading to compromised osteogenesis and bone loss under diabetic conditions.

How does diabetic state downregulate the expression of zinc transporters such as Zip6 and Zip10 in bone tissue? In the present study, we examined the effect of inhibiting PI3 kinase, which regulates various molecular pathways including insulin signaling, and knockdown of insulin receptor on the expression of Zip6 and Zip10 in pre-osteoblastic cells to clarify the mechanisms. We observed that inhibition of the PI3K, through treatment with a PI3K inhibitor, and suppression of insulin receptor expression using siRNA, decreased the expressions of Zip6 and Zip10 in MC3T3E1 cells. These data suggest that insulin deficiency or insufficiency under diabetic conditions might lead to decreased expression of Zip6 and Zip10 in bone tissue.

We also explored the molecular pathways in which changes in Zip6 expression affect osteoblast differentiation. Previous reports have shown that the IGF-1/Akt and MAPK signaling pathways involve osteoblast differentiation [34,35,36,37], and zinc can modulate these pathways [20,21]. Therefore, we examined the influence of Zip6 knockdown on the mRNA levels of IGF-1 and the phosphorylated protein levels of Akt, p38 MAPK, and ERK in pre-osteoblastic cells. We found that the knockdown of Zip6 suppressed p38 MAPK phosphorylation but did not affect ERK phosphorylation. Additionally, Zip6 knockdown did not suppress Akt phosphorylation despite decreasing the levels of IGF-1 mRNA. Our in vivo experiments observed an attenuation of the IGF-1/Akt and p38 MAPK signaling pathways in the tibiae of diabetic mice. These findings suggest that downregulation of Zip6 impairs osteogenesis through the attenuation of p38 MAPK signaling pathways rather than IGF-1/Akt signaling in diabetic bone.

In addition to changes in the expressions of Zip6 and Zip10 localized in the cell membrane caused by diabetes, we found that the mRNA levels of zinc transporters localized in the Golgi apparatus, such as ZnT6 and Zip9, were reduced in the tibiae of diabetic mice. Notably, a recent report showed that the Golgi apparatus has higher levels of zinc than other intracellular organelles [38]. In addition, an in vivo experiment has shown that zinc transport from the Golgi apparatus to the cytoplasm via Zip13 helps to regulate bone formation via activation of the Smad signaling pathway [25]. In this study, we observed that the expression of Zip9, which transports zinc from the Golgi apparatus to the cytoplasm, was positively correlated with trabecular BMD in diabetic mice, whereas ZnT6, which transports zinc from the cytoplasm to the Golgi apparatus, showed no correlation with trabecular BMD in these mice. However, Zip9 knockdown did not reduce the expression of osteogenic genes. These data suggest that changes in the expressions of zinc transporters localized to the Golgi apparatus may not affect bone metabolism under diabetic conditions. However, further studies are needed to clarify the role of changes in zinc dynamics in the Golgi apparatus caused by diabetes in osteoblast function.

The limitations of the present study are as follows. First, although intracellular free zinc (Zn²⁺) is involved various cellular functions, we did not evaluate the influence of changes in zinc transporter expression on Zn²⁺ levels in the bone tissues of diabetic mice and osteoblastic cell lines in the present study. Second, we used a high dose of STZ for generating type 1 diabetic model mice in the present study. A previous report has shown that mice administered a single dose of high-dose STZ exhibit greater suppression of osteogenesis and greater acceleration of adipogenesis in bone tissues compared to mice administered multiple doses of low-dose STZ [39]. However, it has been also suggested that mice administered a high dose of STZ develop other organ damage, such as liver inflammation, and that the low-dose STZ model may more reflective of the human bone phenotype than the high-dose STZ model [39]. Therefore, further experiments using a low dose of STZ may be necessary. Moreover, we only analyzed bone tissues of female mice in the present study since our previous reports have shown that STZ treatment induced severe bone loss and impairment of osteogenesis in female mice compared to male mice [26]. However, further studies including male mice are needed to fully understand the relationships between zinc dynamics and diabetic osteoporosis. Third, the findings were obtained through experiments using mice and mouse-derived pre-osteoblastic cells. Therefore, further evidence is required to demonstrate the significance of zinc transporters in the mechanism of diabetic osteoporosis in humans. However, previous reports have shown that Zip13 is a causative gene for the novel Ehlers–Danlos syndrome in humans, based on the phenotype observed in Zip13 knockout mice [25]. This indicates that the functional analysis of zinc transporters in mouse models may directly lead to an understanding of the mechanisms underlying human metabolic diseases. Moreover, STZ-induced diabetic model mice are currently considered one of the powerful tools for clarifying the molecular mechanisms of human diabetic osteoporosis [40]. Therefore, our findings may provide important insights into the pathogenesis of human diabetic osteoporosis.

5. Conclusions

Our work demonstrated the following. (1) The expressions of zinc transporters localized in the cellular membrane, including Zip6 and Zip10, decreased in bone tissues under diabetic conditions. (2) The levels of Zip6 and Zip10 were correlated with trabecular BMD and the levels of osteogenic genes in the tibiae of diabetic mice. (3) Zip6 knockdown attenuated p38 MAPK signaling and decreased the expression of osteogenic differentiation genes in pre-osteoblastic cells. (4) Attenuation of insulin signaling downregulated Zip6 expression in pre-osteoblastic cells. This evidence suggests that a decrease in Zip6 expression disrupts zinc homeostasis in bone tissues under diabetic conditions and induces the impairment of osteogenic differentiation through the downregulation of p38 MAPK signaling, leading to diabetic osteoporosis. Therefore, maintaining zinc homeostasis in bone tissues may be vital for preventing and treating diabetic bone loss, and zinc transporters may serve as novel therapeutic targets for diabetic osteoporosis.