A combination of clinical study and Network Pharmacology approach to explore the ecacy and mechanism of Buxue Yimu Pills on gynecological anemia

Background Iron supplement is the rst-line treatment for gynecological anemia (GA). However, its performance is limited by common gastrointestinal reactions and some inadequate responses. Traditional Chinese medicine (TCM) has a long history in the treatment of gynecological conditions but has been restricted by limited high-quality research, unknown bioactive ingredients, and mechanisms. These studies aim to compare the clinical ecacy of Buxue Yimu Pills (BYP),ferrous sulfate, and the addition of BYP to ferrous sulfate on GA through oral administration, and to investigate the mechanisms of BYP using network pharmacology approach. Prospective, open-label, comparative, randomized, multicenter clinical trial. Consistent after accompany with signicant changes in iron metabolism markers, typically with elevated SI, Fer, and decreased TIBC (oral iron group). One unexpected nding was that we did not nd a signicant decline of TIBC in the combination group, even though it did show a most signicant increase in Hb. BYP group also exhibited hemoglobin improvement without obvious changes in iron metabolism markers. This result may be explained by the fact that patients with GA are commonly found to have several closely related features: abnormal uterine bleeding (AUB), iron deciency anemia (IDA), chronic inammatory state 19 . On the one hand, an inammatory state can upregulate the level of hepcidin, a liver-specic peptide, increasing the degradation of ferroportin (FPN) at enterocytes, hepatocytes, and macrophages, leading to iron overload in nonhematopoietic cells but the iron deciency in erythroid precursors 20 . In addition, chronic inammation can cause unbalanced blood production of hematopoietic stem cells (HSCs), impaired HSC self-renewal as well as impaired erythroid production 21–23 . The previous study has found a negative correlation between Hb concentration and inammatory markers, especially interleukin 6 (IL-6) 9 . As mentioned before, Buxue Yimu Pills can reinforce qi and nourish the blood, whose main components have previously been observed to have different ecacies of promoting hematopoiesis and anti-inammatory effects 12–14 . After treatment, with decreased inammation and increased hematopoiesis, iron in nonhematopoietic cells can be transported to erythroid precursors more eciently; subsequently, levels of circulating iron decrease. As a consequence, we did observe a slight upward trend of TIBC in the BYP group and an offsetting decline of TIBC in the combination group, while Hb in both groups elevated. to through larger-scale, more prolonged follow-up cohort research with targeted designs, as well as molecular mechanism studies.


Background
Anemia, one of the major public health issues worldwide, affects about one-quarter of the world's population, especially among preschool children and young women 1 . There are many common gynecologic diseases associated with anemia 2 , such as uterine broids, adenomyosis, endometrial polyps, ovulation disorders, ectopic pregnancy, and abortion, affecting especially young women 3 . Besides, cancer-related anemia and therapy-induced anemia are common in patients with gynecological malignancies [4][5][6] . Oral or intravenous iron supplements is the major therapeutic approach, while its common gastrointestinal side effects, including nausea, constipation, and diarrhea, can lower patient compliance. Over 40% of reproductive age women were reported to fail to adhere to medication 7 . Moreover, patients with gynecological anemia (GA) have been observed to be associated with in ammatory states 8 , causing limited iron absorption and utilization 9,10 . Therefore, the exploration of alternative treatment options is encouraged. Traditional Chinese medicine (TCM) has been utilized to treat gynecological diseases and anemia for a long time, showing high safety, tolerability, and compliance even among the patients with malignancy, chronic kidney disease as well as the elderly 11 . Due to the lack of high-quality randomized controlled trial (RCT) evidence and intensive study, the e cacy and therapeutic mechanisms of TCMs have yet to be well understood or appreciated.
To comprehensively assess its clinical effect and herbal pharmacological mechanisms, we performed a randomized clinical trial at three centers, followed by a network pharmacology research.

Materials And Methods
Clinical study Experimental design and setting.
To compare the effects of Buxue Yimu Pills (BYP), ferrous sulfate, and a combination of both treatments on GA, a prospective, randomized controlled trial was conducted between 2017 and 2019 at Gynecology departments of three public hospitals, including Peking Union Medical College Hospital (PUMCH), West China Hospital of Sichuan University, and Hangzhou Maternity and Child Health Care Hospital.
This study was approved by the Ethics Committee of PUMCH (No. ZS-1254) at January 2017, and registered at the United States National Institutes of Health (registration number NCT03232554, www.clinicaltrials.gov).

Patients
Adult females aged 18-50 years with hemoglobin of 70-110 g/L associated with gynecological disorders were enrolled following the informed consent. Exclusion criteria were: (1) patients with uncontrolled gastrointestinal in ammation or bleeding or urological bleeding; (2) patients presented with other diseases leading to iron de ciency; (3) pregnant women; (4) patients complicated with mental abnormalities or other severe diseases; (5) patients with a history of malignancy, chemotherapy or radiotherapy; (6) patients treated with iron supplements, blood transfusion or participating in any other clinical trial within the past month. Eligible participants were randomized into three groups: BYP group received Buxue Yimu Pills (24 g/day); oral iron group received ferrous sulfate tablets (0.9 g/day), and BYP & oral iron group received both drugs above simultaneously. All drugs were provided by ZhuZhou QianJin Pharmaceutical Co., Ltd. The total treatment periods were four weeks.

Assessments
Assessments were scheduled pre-and post-treatment. A basic physical examination including vital signs, weight, height, body mass index (BMI), blood pressure was performed. Venous blood samples were collected to detect complete blood count, including hemoglobin (Hb), mean cellular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red blood cells (RBC), hematocrit (HCT), reticulocyte (RET) count. Iron indexes of serum iron (SI), serum ferritin (SF), and total iron-binding capacity (TIBC) were analyzed. Moreover, the safety assessment, including liver function, renal function, and a blood coagulation index, were also evaluated (data not shown).

Statistical analysis
Statistical analysis was carried out using SPSS version 20.0 (IBM). Data from normally distributed parameters are presented as means plus or minus standard deviation (SD); paired t-tests were used to compare the intragroup difference pre-and post-treatment, and one-way analysis of variance (ANOVA) was applied for intergroup comparisons. Variables not normally distributed are expressed as median plus 25-75 interquartile range (IQR); intragroup and intergroup comparisons were evaluated with Wilcoxon matched-pairs signed-rank tests and Kruskal-Wallis tests. A p-value of less than 0.05 (p < 0.05) was considered signi cant in all statistical analyses.
Target genes of GA GA-associated targets were retrieved from the Online Mendelian Inheritance in Man (OMIM) database (https://www.omim.org/) and GeneCard Database (https://www.genecards.org/), using "gynecological anemia "as the keyword. Data were extracted and summarized without duplication.

Network construction
Potential therapeutic targets are the intersections between the compound targets of BYP and GA-associated targets, which were generated with the VennDiagram R package. After importing the main ingredients, bioactive compounds, and potential therapeutic targets, the compound-target-disease network was built and visualized by Cytoscape software (version 3.6.1, http://cytoscape.org/).

Functional analysis
Based on the obtained potential therapeutic targets, enrichment analysis was performed using the online tool Metascape (https://metascape.org) 15 , including Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis.
A protein-protein interaction (PPI) network was constructed via STRING (Version 11.0, https://string-db.org/) with a minimum required interaction score of 0.400. We subsequently used the Molecular Complex Detection tool (MCODE) plugin to identify signi cant clusters in this PPI network.

Baseline characteristics
Out of 170 enrolled patients with GA,138 patients (81.2%) completed the four-week treatment and were included in the data analysis: BYP group (n=58), oral iron group (n=37), and the combination group (n=43). The baseline characteristics of all three groups showed no statistical difference (Table 1). Table 2 shows the values of laboratory parameters before and after treatment. After four weeks of treatment, patients in three groups exhibited a signi cant increase in Hb, RBC, and HCT (p<0.01), but no signi cant differences in RET. The changes of Hb (g/L) are listed in the following: BYP group (16.44±14.12), oral iron group (30.84±15.46), and the combination group (37.82±16.9). Pairwise comparison show there are statistically signi cant differences between BYP group and oral iron group (p<0.01) or BYP group and the combination group (p<0.01), while differences between the two iron-containing groups were not remarkable (p=0.062) (Figure 1).

Laboratory parameters
According to the iron indexes, it is expected that there were signi cant increases in SI and SF accompanied by a substantial decrease in TIBC in the oral iron group (p<0.01), but no differences were detected in BYP group (p>0.05). However, unexpectedly, the combination group, whose Hb was improved most signi cantly after treatment, showed noticeable increases in SI and SF but no apparent change in TIBC.

Compound-target-disease network
Thirty-two molecules, the bioactive compounds, with oral bioavailability ≥ 30% and drug-likeness ≥0.18 were identi ed through network pharmacology analysis (Table 3) Figure 2(a)). The compound-target-disease network was constructed with 818 nodes and 1436 connecting edges and shown in Figure 2(b). After analyzing the network with the NetworkAnalyzer plugin in Cytoscape, quercetin (MOL000098, mol17), which is shared by Huangqi and Yimucao, was identi ed as the most important compound with the overwhelmingly highest degree of 107. Evidence from previous studies has shown that quercetin can improve in ammation and chemotherapy-related anemia through reducing oxidative stress, ameliorating the iron status and protecting red blood cells [16][17][18] . Figure 3 shows the results of functional enrichment analysis: KEGG pathway, Gene Ontology Biological Process (GOBP), Gene Ontology Cellular Component (GOCC), and Gene Ontology Molecular Function (GOMF), which were mostly associated with cancer, in ammation, steroid hormone regulation, oxidative stress, metabolic syndrome, nervous system diseases, and infectious diseases. Many of these terms are too general for annotation, so we took a deeper insight into their child function clusters and further summarized the biological processes into several underlying functional modules: regulation of in ammation, regulation of steroid hormone, angiogenesis and hemostasis, response to decreased oxygen levels, effects on myeloma cell and response to metal ion (Supplementary Table1). The biological processes with the highest enrichment score in each module were selected to build a compound-target-underlying biological process network (Figure 4), with 139 nodes, 483 edges.

Functional analysis
Protein-protein interaction (PPI) network of intersecting targets was constructed with STRING, including 145 nodes and 3196 edges ( Figure 5 (a)).

Discussion
Consistent with previous studies, after four-week treatment, patients taking ferrous sulfate or ferrous sulfate &BYP showed an elevation of Hb in accompany with signi cant changes in iron metabolism markers, typically with elevated SI, Fer, and decreased TIBC (oral iron group). One unexpected nding was that we did not nd a signi cant decline of TIBC in the combination group, even though it did show a most signi cant increase in Hb. BYP group also exhibited hemoglobin improvement without obvious changes in iron metabolism markers. This result may be explained by the fact that patients with GA are commonly found to have several closely related features: abnormal uterine bleeding (AUB), iron de ciency anemia (IDA), chronic in ammatory state 19 . On the one hand, an in ammatory state can upregulate the level of hepcidin, a liver-speci c peptide, increasing the degradation of ferroportin (FPN) at enterocytes, hepatocytes, and macrophages, leading to iron overload in nonhematopoietic cells but the iron de ciency in erythroid precursors 20 . In addition, chronic in ammation can cause unbalanced blood production of hematopoietic stem cells (HSCs), impaired HSC self-renewal as well as impaired erythroid production [21][22][23] . The previous study has found a negative correlation between Hb concentration and in ammatory markers, especially interleukin 6 (IL-6) 9 . As mentioned before, Buxue Yimu Pills can reinforce qi and nourish the blood, whose main components have previously been observed to have different e cacies of promoting hematopoiesis and anti-in ammatory effects [12][13][14] . After treatment, with decreased in ammation and increased hematopoiesis, iron in nonhematopoietic cells can be transported to erythroid precursors more e ciently; subsequently, levels of circulating iron decrease. As a consequence, we did observe a slight upward trend of TIBC in the BYP group and an offsetting decline of TIBC in the combination group, while Hb in both groups elevated.
We also used network pharmacology technology to investigate the mechanisms of Buxue Yimu Pills further. From the compound-target-disease network, 27 bioactive compounds in Buxue Yimu Pills were identi ed, among which quercetin, a common component of Huangqi and Yimucao, may play a critical role. One hundred forty-ve potential therapeutic targets of these compounds were con rmed. Biological process and pathway enrichment analysis demonstrated that multiple mechanisms including regulation of in ammation, regulation of steroid hormone, angiogenesis and hemostasis, response to decreased oxygen levels, effects on myeloma cell as well as response to metal ions, may contribute to improving GA. Previous studies have demonstrated that reduction of in ammation and oxidative stress can alleviate the absorption and utilization of iron, increase the production of erythropoietin (EPO) and improve hematopoietic function 24,25 . Metal ion transmembrane transport contributed to iron uptake and maintenance of iron homeostasis 26,27 . In addition, ovulation disorders and coagulation dysfunction have been the main nonstructural etiologies of AUB 28,29 . Therefore, the regulation of steroid hormone and hemostasis can have signi cant implications for this condiction.
Since Buxue Yimu Pills may improve anemia through multiple targets, it could have potential applications in diseases needing to deal with oxidative stress damage, such as thalassemia and in ammatory anemia. But on the other hand, one crucial point that needs to be raised is that reduction of oxidative stress might reduce the therapeutic e cacy of anti-neoplastic drugs acting through mechanisms of reactive oxygen species (ROS)-induced injury 30 , which was also indicated in KEGG enrichment (hsa01524, platinum drug resistance). Therefore, Buxue Yimu Pills may not be recommended to patients undergoing cancer treatment.

Conclusion
A combination of clinical study and network pharmacology approaches not only demonstrates that Buxue Yimu Pills contributes to the effective improvement of GA through multiple mechanisms but also provides us more comprehensive insights into integrated multidimensional relationships between drugs and diseases. Finally, some limitations need to be acknowledged. The sample size and the follow-up duration may have been insu cient to reveal apparent differences between cohorts, and in ammatory indicators were not monitored. Accordingly, based on the results of our study, continued efforts are needed to further veri cation through larger-scale, more prolonged follow-up cohort research with targeted designs, as well as molecular mechanism studies.

Consent to publish
Not applicable.

Availability of data and materials
All relevant data and materials are within the paper and its additional les.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was nancially supported by ZhuZhou QianJin Pharmaceutical Co., Ltd.
Authors' Contributions WYF implemented the research plan and was responsible for data curation, data mining formal analysis, conceptualization and writing of the manuscript. DY, MRL, LD, LB, WX, DM participated in the enrollment of the patients and data collection. SAJ provided a critical contribution to the organization and cooperation of this multi-center task.  Figure 1 Changes in hemoglobin after 4-week treatment in three groups.  Compound-target-underlying biological process network with 139 nodes, 483 edges. Red oblong nodes represent target genes; green square nodes represent bioactive compounds; Blue polygon nodes represent main underlying biological processes. Edges between biological processes and targets are shown in red; edges between bioactive compounds and targets are shown in grey.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.