Qizhi Kebitong Formula Ameliorates Streptozocin-Induced Diabetic Osteoporosis through Regulating the PI3K/Akt/NF-κB Pathway

Background Diabetic osteoporosis (DOP) is a progressive osteoblast dysfunction induced by high glucose, which has negative impacts on bone homeostasis. Qizhi Kebitong formula (QKF) is a traditional Chinese medicine (TCM) formula for treating DOP. However, its role in the protection of DOP has not been clarified yet. Here, we aimed to explore the potential mechanisms of QKF on DOP development via in vivo experiment. Methods Network pharmacology was used to detect the key targets and signaling pathways of QKF on DOP. The effects of QKF on DOP were examined by the phenotypic characteristics, micro-CT, and hematoxylin-eosin (H&E) staining. The predicted targets and pathways were validated by a streptozocin- (STZ-) induced mouse model. Subsequently, the levels of the selected genes and proteins were analyzed using qRT-PCR and Western blot. Finally, AutoDock and PyMOL were used for molecular docking. Results In this study, 90 active compounds and 2970 related disease targets have been found through network pharmacology. And QKF could improve the microstructures of femur bone mass, reduce inflammatory cell infiltration, and downregulate the levels of TNF-α, IKBKB, IL-6, and IL-1β. Moreover, the underlying effect of PI3K/Akt/NF-κB pathways was also recommended in the treatment. Conclusion Altogether, our findings suggested that QKF could markedly alleviate osteoblast dysfunction by modulating the key targets and PI3K/Akt/NF-κB signaling pathway.


Introduction
Diabetic osteoporosis (DOP) is a common complication of diabetes, which primarily affects bone metabolism, joints, and kidney [1,2]. DOP is a skeletal disorder characterized by a chronic high glucose, decreased bone mass, and damaged bone tissue [3][4][5]. With the increasing incidence of diabetes, DOP has become a systemic bone disease to increase bone brittleness, fracture risk, and impaired bone healing [6]. However, the pathogenesis of DOP has not been fully clarified. Notably, studies have shown that high glucose is a crucial determinant of DOP [7], especially increased diabetes-related pathological factors [8,9]. Interestingly, inflammation is defined as one of the major pathological factors of DOP, which leads to bone loss [10], destroys the bone microenvironment, and inhibits bone formation [11,12]. However, a series of DM-induced inflammation is often overlooked or underestimated, seriously affecting the quality of people's life in the later period [13]. Therefore, it is an urgent strategy to prevent the development of inflammation and find effective therapies for DOP.
In this study, the potential targets and protective pathways of QKF on DOP were screened via network pharmacology, and the results were verified in the mouse model. Then, we provided some insights with the possible molecular mechanisms of QKF on the clinical application for delaying DOP progression.
2.5. The Hematoxylin/Eosin (H&E) Staining. The exfoliated femurs were fixed using 4% formaldehyde, decalcified in EDTA glycerol solution, and embedded in paraffin. Paraffin sections were cut into the slices at 4 μm thickness and stained with H&E. Images of the sections were captured using light microscopy (Olympus BX51, Japan) at 200x and 400x ratios, respectively.
2.6. Quantitative Real-Time PCR (qRT-PCR) Analysis. Total RNA was extracted from the femur tissues with a total RNA extraction kit (TIANGEN BIOTECH, China). Subsequently, the reverse transcription of 1 μg total RNA into cDNA was conducted with the iScript cDNA synthesis kit (TIANGEN BIOTECH, China). The qRT-PCR assay was performed with a Bio-Rad CFX96 system, and the gene expressions of IKK, IL-1β, IL-6, and TNF-α were normalized to GAPDH. Relative mRNA levels were quantified using the 2 −ΔΔCt method. The mouse primer sequences are shown in Table 2    2.8. Molecular Docking. AutoDock software, version 4.2, was used for molecular docking. The composite targets were verified using the Lamarckian genetic algorithm; proteins and ligands were prepared using the AutoDock tool. The threedimensional structure of the proteins was downloaded from the RCSB-PDB database (http://www.pdb.org), and the hydrogen atoms were added. We calculated the docking binding energy using the Auto tool. The docking diagrams of target proteins and molecules were performed by the PyMOL visualization software.
2.9. Statistical Analysis. All data were analyzed using Graph-Pad Prism 9.0. These data were compared with several groups by one-way ANOVA. For all statistical analysis, p < 0:05 was considered statistically significant.

Screening of the Intersection Targets and Constructing a
Series of Network. With OB ≥ 30% and DL ≥ 0:18 as screening parameters, 90 candidate compounds of QKF were found for further analysis (Table 3). Besides, 2970 potential targets of DOP were obtained from the four authoritative databases (Figure 1(a)). Through taking the intersection of 122 QKF targets and 2,970 DOP targets, 81 potential targets were obtained (Figure 1(b)). Subsequently, the intersection targets were inputted to Cytoscape software to build the network diagram with multicomponent and multitarget (Figure 1(c)). In addition, 81 potential targets were uploaded to the STRING database to construct the PPI network ( Figure 1(d)). Among these nodes, PIK3CG, Akt1, and RELA were screened out with more relevance and biological functions in the PPI network (Figure 1(e)), suggesting that PIK3CG, Akt1, and RELA were the key genes, probably exhibiting therapeutic effect in DOP.

Functional Enrichment Analysis.
To investigate the potential mechanisms, the 1660 biological processes (BP), 24 cellular components (CC), and 104 molecular functions (MF) were performed using the DAVID database. Moreover, the top 15 results were selected with the p value from small to large (Figure 2(a) and Table 4). KEGG enrichment analysis obtained 128 results. Subsequently, we selected the top 50 according to the p value for further analysis (Figure 2(b)).

Effect of QKF on the General Features of STZ-Induced
Mice. In order to determine the effect of QKF on DOP, we established a STZ-induced mouse model and compared disease evolution in groups ( Figure 3). After administration of QKF for 4 months, blood glucose levels of STZ-induced mice were significantly higher (Figure 3(a)), while body weight was significantly lower (Figure 3(b)). The results demonstrated that the blood glucose of mice increased sharply, which consumed a lot of fat in the body. And compared to the Ctrl group, the weight of mice in STZ and QKF groups was decreased significantly. Meanwhile, the trabecular bone at distal femoral metaphysis was assessed by HE staining (Figure 3(c)); obvious bone loss was observed in STZ-induced mice compared with the Ctrl group, which was gradually mitigated with the increasing dose of QKF. The femurs of normal mice scattered pink trabecular bones, and the number of trabecular bones was reduced in STZinduced mice. Furthermore, the profiles of 3D images (Figure 3(d)) clearly exhibited the breakage of cancellous bone of diabetic mice, and the 3D bone biological parameters (Figures 3(e)-3(j)) quantitatively reflected the significant reduction in Conn.D (p < 0:001), BMD (p < 0:001), BV/TV (p < 0:01), BS/TV (p < 0:001), and Tb.Th (p < 0:01) in the STZ group, while Tb.Sp was significantly increased. However, after the treatment of QKF for 4 months, improved bone mass of trabecular bone and reversed changes of biological parameters indicated the potential therapeutic efficacy of QKF on DOP.
3.4. QKF Improves STZ-Induced Mouse Inflammation. DOP is an inflammatory response caused by high blood glucose [15]. To validate that QKF could reduce the inflammatory expression of STZ-induced mice, we used qRT-PCR and Western blot to determine changes in mRNA and protein levels ( Figure 4). The qRT-PCR results indicated that the mRNA levels of TNF-α, IKK, IL-6, and IL-1β were significantly downregulated after administration (Figure 4(a)). Meanwhile, Western blot results demonstrated that QKF had a similar inhibitory effect at the protein levels (Figures 4(b) and 4(c)). Above all, these results indicate that QKF could attenuate inflammation in STZ-induced mice.
3.5. QKF Mediated Inflammation through the PI3K/Akt/ NF-κB Pathway. Based on the network pharmacological analysis, the PI3K/Akt signaling pathway may be predicted as a potential mechanism of QKF for DOP protection. Meanwhile, NF-κB was a key downstream factor of the PI3K/Akt pathway, which was closely related to the regulation of glucose and lipid metabolism [16]. Therefore, we explored the PI3K/Akt/NF-κB signaling pathway as the potential mechanism of QKF for experimental verification. After administration of QKF, the protein levels of p-PI3K/PI3K and p-Akt/Akt were further upregulated compared with the STZ group, while p-NF-κB/NF-κB was downregulated ( Figure 5). The results indicated that PI3K/Akt/NF-κB signaling could regulate the protective effects of QKF on DOP.
3.6. Molecular Docking Analysis. To further explore the effect of the 3 major compounds of QKF on the 7 potential targets, including PI3K, Akt1, RELA, IKBKB, IL-1β, TNFa, and IL-6, the binding energies were determined by molecular docking ( Figure 6). Firstly, kaempferol and baicalein had a strong binding ability with PI3K, so they would be a potential bioactive compound of QKF on DOP (Figure 6(a)). Akt1 had a stronger binding energy with all compounds (Figure 6(b)). The strongest binding energy was as high as -10.2 kcal/mol. Interestingly, quercetin, kaempferol, and baicalein had the same binding power with RELA and IL-6 (Figures 6(c) and 6(g)). It means that RELA and IL-6 have the best binding force with the above components. Meanwhile, IKBKB and baicalein, IL-1β and quercetin, and TNF-α and quercetin have a stronger binding force (Figures 6(d)-6(f)). Above, quercetin, kaempferol, and baicalein played an important role in QKF. Although they have higher binding force with inflammatory factors, the pharmacological effects of these active compounds in regulating key targets needed to be further verified.

Discussion
In this study, we performed network pharmacology, animal experiments, and molecular docking to explore the active compositions and molecular mechanisms of QKF in the treatment of DOP. The potential targets and enrichment pathways were predicted by network pharmacology. Histopathological staining and micro-CT imaging confirmed the therapeutic effect of QKF on the STZ-induced mouse model. qRT-PCR and Western Blot confirmed that QKF could mediate inflammation through the PI3K/Akt/NF-κB pathway. In summary, this study demonstrated for the first time that QKF mediated inflammation through the PI3K/Akt/ NF-κB pathway, thereby improving bone mass of trabecular bone and reversing the changes of biological parameters in the STZ-induced mouse model.
Based on the TCM theory, seven drugs of QKF were formed for clinical application of DOP-related diseases [17]. Among these drugs, HQ (qi-tonifying), JXT (bloodactivating), HNX (kidney-invigorating), and WLX, SZ, and XXC (dredging collateral) were used for the treatment of DOP [18][19][20]. A large number of reports have focused on bones and kidneys [21]; kidney weakness and blood stasis were the main causes of DOP [22]. Therefore, the kidneynourishing herbs used for the treatment of DOP have aroused concerns [23]. HNX and WLX could tonify the kidney [24], which was deemed as one of the effective methods to alleviate DOP [25]. Furthermore, SZ and XXC had the ability to tonify the kidney and strengthen muscle and bone [26]. Above all, TCM has a series of effects on DOP [27], and it could improve the clinical symptoms of patients, which was worthy of clinical promotion [28]. At the same time, a previous study suggested that quercetin not only promoted the differentiation activity of osteoblasts but also inhibited the absorption activity of osteoclasts, thereby increasing the expression of osteogenic markers [29]. Kaempferol has a significant anti-inflammatory benefits, including promoting osteoblast proliferation, differentiation, and bone formation [30]. Previous studies suggested that baicalin could promote osteogenic differentiation by regulating protein kinases and transcription factors [31]. In sum, the compounds of QKF could provide an alternative strategy to prevent bone loss.
According to reports, trabecular bone loss was one of the common pathological processes occurring in DOP mice. To evaluate the effects of QKF for the treatment of DOP, we assessed trabecular architectural parameters using 3D micro-CT images. The results suggested that QKF could prevent the loss of bone mass induced by DOP and restore the trabecular connectivity by increasing BMD and Conn.D. Moreover, compared with the STZ group, the parameters of Tb.Th, BS/BV, and BV/TV in the QKF group increased significantly, while that of Tb.Sp was inhibited. Treatment of STZ-induced mice with QKF markedly increased trabecular BMD and improved trabecular bone and enhanced trabecular bone area.
In the present study, QKF treatment significantly decreased the mRNA and protein levels of a series of inflammatory factors, including IL-6, TNF-α, IKBKB, and IL-1β in the STZ-induced mouse model, which contributed to the improvement of DOP. However, QKF mediated inflammation through the PI3K/Akt/NF-κB pathway; the relevant key targets were also proven to induce antioxidation, antiinflammation, and immune regulation. Among them, Akt was identified as a unique signaling intermediate in bone homeostasis that controlled the differentiation of osteoblasts and osteoclasts, which was a direct downstream target of PI3K to inhibit the release of inflammatory factors [32][33][34][35][36]. Moreover, NF-κB was also a key downstream factor of the PI3K/Akt pathway, which enhanced the degree of inflammatory response and promoted the differentiation of osteoclast precursors [37,38]. Meanwhile, the PI3K/Akt signaling pathway not only affects inflammatory factors such as NF-κB and TNF-α but also induced the inflammatory reaction in the internal environment of the body. Furthermore, the differentiation of osteoblasts was regulated by TNF-α, which was the earliest inflammatory mediator produced in response to oxidative stress and promoted the production of inflammatory cytokines to promote osteoblast apoptosis [39,40]. In addition, accumulating studies have revealed that the expressions of core targets, including Akt1, TNF-α, IL-6, and RELA, made the vital functions in regulating inflammatory response [41,42]. We have verified that QKF could regulate the key targets and PI3K/Akt/NF-κB signaling pathway to explain the molecular mechanism of QKF treatment on DOP.

Conclusion
In summary, QKF could recuperate the bone loss and improve bone mass of trabecular bone in STZ-induced mouse models by downregulating the expression of IL-6, TNF-α, IKBKB, and IL-1β to alleviate the inflammation. The results might be mediated by the PI3K/Akt/NF-κB pathway based on the prediction from network pharmacology and experiment validation. This study may provide new insights into the molecular mechanisms of QKF in the treatment of DOP.

Data Availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.