Rehmannia Glutinosa Polysaccharide Regulates Bone Marrow Microenvironment via HIF-1α/NF-κB Signaling Pathway in Aplastic Anemia Mice

Abstract Aplastic anemia (AA), a rare disorder, is associated with bone marrow microenvironment (BMM). Presently, AA treatment is of great difficulty. This study aimed to explore the mechanism of action of Rehmannia glutinosa polysaccharide (RGP) in AA. Busulfan was used to induce AA in BALB/c mice; blood cell count and Ray’s Giemsa staining were used to assess the severity of hematopoietic failure; HE was performed to assess the pathological state of the marrow cavity; ELISA was performed to assess IL-4, IL-10, IL-6, IL-12, IL-1β, TNF-α, MCP-1, VEGF, and EPO; and WB was performed to evaluate the effects of RGP on the HIF-1α/NF-κB signaling. Significant downregulation of hemocyte levels in the blood and nucleated cells in the bone marrow was reversed by RGP and Cyclosporine A (CA). Compared with the AA group, dilating blood sinusoids, inflammation, hematopoiesis, decreased bone marrow cells and megakaryocytes were alleviated by RGP and CA, and the HIF-1α/NF-κB signaling was inhibited too. Notably, RGP was more effective when used in combination with CA. In this study, we established a relationship between BMM and the HIF-1α/NF-κB signaling pathway and found that RGP regulates BMM by suppressing the activation of the HIF-1α/NF-κB signaling. Thus, RGP exerts a pharmacological effect on AA.


INTRODUCTION
Aplastic anemia (AA) is a rare disorder that is caused by hematopoietic failure.Bone marrow stem cells and bone marrow microenvironment (BMM) are severely damaged by physical, chemical, biological, and unknown factors (Liu & Liu 2015).Clinically, patients with AA usually have pancytopenia, bleeding, marrow fibrosis, or hypocellular bone marrow (Young 2013, Killick et al. 2016).The pathogenesis of AA is complex, but is still not fully understood.At present, AA pathogenesis might be attributed to immune damage caused by hematopoietic stem cells (HSCs) (Solomou et al. 2007, Young et al. 2010), bone marrow microenvironment failure (Young et al. 2006), impaired hematopoietic cytokine production, and cellular or humoral immunosuppression of the bone marrow (Dasouki et al. 2013).
The bone marrow microenvironment is a complex hematopoietic organ with highly vascular and adipose tissue (Polineni et al. 2020, Usmani et al. 2019).Changes in BMM have been widely described in AA patients where cells such as osteoblasts, osteoclasts, adipocytes, and MSCs, maintain the balance of BMM (Sivasubramaniyan et al. 2012).As reported, when AA occurs, the balance between lipogenic and osteogenic differentiation of bone marrow An Acad Bras Cienc (2023) 95(3) e20220672 2 | 13 HSCs is disrupted (Nuttall & Gimble 2004), the suppressive effect of HSCs on lymphocytes is diminished (Solomou et al. 2007), and HSCs apoptosis is increased (Li et al. 2012).Hypoxiainducible factor 1α (HIF-1α) is a key regulator of nuclear factor kappa-B (NF-κB), and studies have shown that NF-κB is involved in inflammatory processes, oxidative stress, apoptosis, cell proliferation, and differentiation (Feng & Wu 2022).Evidence has reported that NF-κB takes part in the maintenance of BMM (Wang et al. 2016).
Rehmannia glutinosa polysaccharide (RGP) is a critical bioactive compound found in the roots of Rehmannia glutinosa (Wang & Li 1997).As studies have reported, RGP shows positive anticancer (Xu et al. 2017), immunoenhancement (Huang et al. 2019), hypoglycemic, and antilipidemic (Kiho et al. 1992).Rehmannia glutinosa libosch extracts have been shown to promote osteoblast differentiation (Gong et al. 2019).These results suggest that RGP may benefit the BMM of AA.Nevertheless, the function of RGP in BMM has not yet been studied.Considering the role of the HIF-1α/NF-κB signaling pathway in BMM, this study aimed to explore the effects of RGP on AA in BALB/c mice based on the HIF-1α/NF-κB signaling pathway.

Animal model establishment
A total of 42 male BALB/c mice of 7-10 weeks of age were used to establish the AA model.Obtained from the Chengdu Da Shuo Experimental Animal Co., Ltd, China (experimental animal production license number: SYXK (Chuan)2015-030; experimental animal use permit number: SYXK (Chuan)2014-189).The mice were separated into seven groups (n=6): control, AA-A1, AA-A2, AA-A3, AA-B1, AA-B2, and AA-B3.The model was expressed as "AA", after "-", marked "A" when mice were fed 20 mg/kg/d busulfan (Solarbio Co., Ltd, Beijing, China), marked "B" when mice were fed 40 mg/kg/d busulfan; marked "1, 2, or 3" when mice were fed busulfan for "5, 7, or 10 days".All mice were continuously fed for 10 days and saline was used for the control group.When mice were fed with busulfan for less than 10 days, they were fed an equal volume of normal saline in the remaining days.Blood samples were collected on days 1, 5, 7, and 10 and mice were sacrificed on day 10 for the experiments.

Drug treatment
Thirty male BALB/c mice of 7-10 weeks of age were used to establish the AA model.The mice were separated into five groups (n=6): control, AA, AA+RGP, AA+CA, and AA+RGP+CA.Modeling the AA by 20 mg/kg/d busulfan feeding for 10 days, next, mice were fed with 0.1 mL 500 mg/ mL Rehmannia glutinosa polysaccharide (RGP) (Wuhan ChemFaces Biochemical Co., Ltd, Wuhan, China) (AA+RGP), 0.1 mL 5 mg/mL Cyclosporine A (CA) (Solarbio Co., Ltd, Beijing, China) (AA+CA) or both (AA+RGP+CA) for 10 days.Blood samples and femoral bone marrow were collected on day 10 for the experiments, and the mice were sacrificed.

Hematoxylin-eosin (HE) staining
The excised thighbone tissue was fixed with 10% neutral formaldehyde.Then, it was decalcified in 15% EDTA, observed for 3 days/time, and replaced with fresh 15% EDTA until decalcification was completed.The thighbone tissues were washed with PBS, dehydrated in a graded ethanol/xylene series, embedded in paraffin, and cut into 5μm thick slices.The slices were then dewaxed for hematoxylin and eosin.Finally, the cells were examined under a light microscope.

Statistical analysis
GraphPad Prism 9.1.2(GraphPad Software, Inc.) was used to analyze the data and generate the charts in this experiment.All data are presented as the mean ± standard deviation, and all tests were performed at least three times.Unpaired t-test was used to analyze the differences between the two groups.Differences among the three groups were analyzed by one-way analysis of variance (ANOVA) for all experiments.Statistically significance was set at P < 0.05.

RESULTS
The effect of the administration of different doses and methods of busulfan on hemocyte levels in mice The mice in the control group were in a nice psychiatric condition and had smooth coats, while the model mice were depressed and had bowed backs, dull coats, and low physical activity.To discover the best modeling methods, mice were treated with different doses and durations of busulfan treatment.After modeling, the mortality rates in each group were as follows: AA-A1 (16.7%),AA-A2 (33.33%),AA-A3 (50%), AA-B1 (50%), AA-B2 (66.7%), and AA-B3 (83%).As shown in Figure 1, compared with the control group, the AA-B3 group (40 mg/kg/d, 10 days) showed the lowest levels of HGB, WBC, RBC, and PLT on day 10.

The effect of administration of different doses and methods of busulfan on the pathological change of bone marrow
To count nucleated cells, bone marrow smears were stained with Giemsa (Figure 2a).The results indicated that busulfan significantly decreased the number of nucleated cells in the bone marrow in a dose-and time-dependent manner.Hematoxylin and eosin staining was performed to observe pathological changes in mice.As shown in Figure 2b, the number of adipocytes in the femur marrow cavity increased with increasing busulfan dose and treatment time.Meanwhile, there was a reduction in the number  (2023) 95(3) e20220672 5 | 13 of bone marrow cells and megakaryocytes.The phenomenon of dilated and congested blood sinusoids was most serious when the mice were treated with 40 mg/kg/d for 10 days.
Effects of RGP on hemocyte levels of AA mice AA model mice were treated with 20 mg/kg/d busulfan for 10 days.0.1mL of RGP (500 mg/mL), CA (5 mg/mL), or RGP (500 mg/mL) + CA (5 mg/ mL) was administered to AA model mice twice daily for 10d.As shown in Figure 3a-d, compared with the control group, RBC (Figure 3a), PLT (Figure 3b), HGB (Figure 3c), and WBC (Figure 3d) were significantly decreased by busulfan, and this state was reversed by RGP.Additionally, the combination of cyclosporine and RGP resulted in better treatment.

Effects of RGP on the pathological change of the bone marrow of AA mice
To detect the effects of RGP on pathological changes in the bone marrow of AA mice, we performed Giemsa staining of the sternum marrow and HE staining of the right femur of mice after RGP and CA treatment.Figure 3e suggests that when RGP was combined with CA, there was a greater increase in nucleated cells compared with the other single drug treatment groups.Furthermore, as Figure 3f shows, compared with the AA group, three groups (AA+RGP, AA+CA, AA+RGP+CA) displayed an increased number of bone marrow cells and megakaryocytes, a reduced number of adipocytes, and dilated and congested blood sinusoids.The AA+RGP+CA group showed the closest pathological morphology to that of the control group.

Effects of RGP on inflammatory cytokines of AA mice
To assess the impact of RGP on the regulation of the bone marrow microenvironment, we performed ELISA to measure inflammatory cytokines in the bone marrow and blood.These cytokines included IL-4, IL-10, IL-6, IL-12, TNF-α, IL-1β, and MCP-1.As shown in Figure 4 (a-i), busulfan-induced expression of IL-6, IL-12, TNF-α, IL-1β, and MCP-1 was inhibited, and IL-4 and IL-10 were promoted by RGP in the blood of mice.Cyclosporin combined with RGP presented the best treatment compared with a single drug.Meanwhile, these cytokines in the bone marrow showed a similar trend (Figure 5a-i).
Effects of RGP on EPO and VEGF of AA mice EPO and VEGF are two important factors associated with AA.Thus, we performed an ELISA to measure the expression of EPO and VEGF.The results indicated that VEGF expression either in the blood or bone marrow was downregulated by busulfan, and RGP reversed this effect (Figure 4h/ Figure 5h).Moreover, EPO expression in both blood and bone marrow was upregulated by busulfan and reversed by RGP (Figure 4i/ Figure 5i).However, RGP was more effective when administered in combination with cyclosporine.

DISCUSSION
Altered BMM is an important cause of AA.The variations in BMM include bone marrow microcirculation, cytokines, and interstitial cells (Xiong et al. 2015).Considering the functions of RGP and busulfan.This study investigated the influence of RGP on busulfan-induced changes in AA and BMM.The results showed that RGP inhibited busulfan-induced AA by suppressing the HIF-1α/NF-κB pathway in BALB/c mice.
Various AA modelling methods have been reported, such as benzene feeding, total-body irradiation, allogeneic lymphocyte infusion combination, heterogeneous lymphocyte implantation, and busulfan feeding (He et al. 2020, Mu et al. 2021, Zheng et al. 2019).Among them, busulfan is a bifunctional alkylating agent that is highly virulent for hematopoietic progenitor cells and HSCs and is commonly used in clinical practice for bone marrow transplantation and anti-tumor treatment (Bartelink et al. 2016).Busulfan-induced aplastic anemia in animals was first studied in 1974 (Santos & Tutschka 1974).AA modeling was further optimized in this study using different doses of busulfan feeding for different durations.We found that feeding mice 20 mg/kg/d busulfan for 10 days was an optimum modeling method.
Changes in the M1/M2 macrophage ratio and IFN-γand TNF-α levels have been reported to lead to AA (Mu et al. 2021).M1-type macrophages are recognized by Toll-like receptors (TLRs) and secrete pro-inflammatory factors, such as IL-6, IL-12, and TNF-α, which promote inflammation progression (Weisser et al. 2013).In contrast, M2-type macrophages are recognized by IL-10, IL-4, or TGF-β and secrete Arginase-1 (Arg-1) to inhibit inflammation progression (Gordon & Martinez 2010).In this study, we found that busulfan induced the expression of cytokines IL-4, IL-10, and VEGF down-regulation significantly, and upregulated the expression of IL-6, IL-12, TNF-α, MCP-1, IL-1β, and EPO.Our findings were consistent with this theory.Meanwhile, the quantities of HGB, WBC, RBC, and PLT were also significantly reduced.Furthermore, busulfan induced lipogenic differentiation and bone marrow sinus dilation.Adipocytes play a negative regulatory role in the differentiation and maturation of HSCs (Sarugaser et al. 2009).The number of adipocytes increases and that of hemocytes decreases whith AA occurrence (Chattopadhyay & Law 2020).This indicates that busulfan causes inflammation and impairs hematopoietic function.These cases were both dose-and time-dependent on busulfan administration.However, these events were reversed by RGP, indicating that RGP could be a potential candidate for AA treatment.
In recent years, except for RGP, other herbal active ingredients have therapeutic effects on AA, such as Dioscorea nipponica Makino and Tripterygium wilfordii polyside (Le 2019), Shenyubuxue prescription (Deng 2018).As Zhang Le reported, D. nipponica Makino inhibits the expression of HIF-1α and NF-kB in the bone marrow, increases the expression of IL-1β, and decreases TNF-α in AA mice (Le 2019).This suggests that the HIF-1α/NF-kB signaling pathway might be a new target for controlling AA.Activation of the NF-κB signaling pathway is controlled by p65 and IκBα phosphorylation (Baldwin 1996).In this study, we found that the phosphorylation of IκBα and p-65 was inhibited by RGP, while busulfan activated it.These results are consistent with those of previous studies.RGP and CA have similar pharmacological effects, and the results demonstrate that RGP provides synergistic effects with CA.In summary, this study provides an excellent reference for the clinical use of AA.

CONCLUSIONS
This study enumerated a traditional Chinese medicine to control AA and illustrated the key role of the HIF-1α/NF-κB signaling pathway in the process of AA.Moreover, RGP suppressed busulfan-induced AA by suppressing the HIF-1α/NF-κB signaling pathway in vivo.This study provides a reference for the treatment of AA and the development of therapeutic drugs.Nevertheless, the effect of RGP on busulfaninduced AA needs to be studied in vivo in the future.