Mechanism of Jianpi Shengxue Decoction in Alleviating Cyclophosphamide-Induced Myelosuppression in Mice

Chemotherapy is a cornerstone in the treatment of various malignancies, primarily by inhibiting cancer cell proliferation and survival^1,2, thereby extending patient survival. However, due to the lack of selectivity in targeting normal tissues, chemotherapy drugs often cause severe damage to healthy cells and tissues, leading to a range of adverse effects³. One of the most common and serious side effects of chemotherapy is myelosuppression, characterized by a reduction in the production of blood cells, including white blood cells (WBCs), red blood cells (RBCs), and platelets (PLTs)^2,4. Myelosuppression severely impairs the immune defense system, increasing the risk of infections, bleeding, anemia, and other complications. It also significantly diminishes patients' quality of life and can even be life-threatening⁵. Thus, alleviating and preventing chemotherapy-induced myelosuppression is critical to improving patients' tolerance to chemotherapy and overall survival.

Current clinical treatments for myelosuppression primarily involve hematopoietic growth factors such as granulocyte colony-stimulating factor [G-CSF], erythropoietin [EPO]) and transfusion therapy^6,7. However, these approaches are often associated with high costs, potential side effects, and limited efficacy. As a result, exploring safer and more effective methods for preventing and alleviating myelosuppression remains a major research focus. In recent years, traditional Chinese medicine (TCM) has garnered significant attention for its role in cancer treatment and managing chemotherapy-induced side effects^8,9. Many TCM formulations are believed to have tonifying and immune-regulating effects, such as Shengxuebao decoction¹⁰, Dihuang Shengbai tablets¹¹, and ginsenoside compounds¹², which not only protect normal cells and reduce toxic side effects but also enhance anti-tumor efficacy.

Jianpi Shengxue Decoction (JPSXD) is a classical TCM formula that exemplifies the therapeutic principles of tonifying Qi and nourishing blood¹³. This formula is a sophisticated modification of the Sijunzi Decoction combined with Er Zhi Wan, integrating the synergistic effects of multiple herbs to address Qi and blood deficiency¹⁴. The formula comprises of ginseng (Ren Shen), astragalus (Huang Qi), epimedium (Yin Yang Huo), atractylodes (Bai Zhu), yam (Shan Yao), salvia (Dan Shen), prepared rehmannia (Shu Di Huang), dodder seed (Tu Si Zi), wolfberry (Gou Qi Zi), ligustrum (Nv Zhen Zi), angelica (Dang Gui), fried barley (Chao Mai Ya), fried chicken gizzard (Chao Ji Nei Jin), dried tangerine peel (Chen Pi), coix seed (Yi Yi Ren), and several other herbs. Each herb plays a specific role in the formula, adhering to the TCM principle of sovereign, minister, assistant, and guide herbs, which ensures a balanced and targeted therapeutic effect.

The Sijunzi Decoction, a foundational formula in TCM, is renowned for its ability to tonify Qi and strengthen the spleen. It has been widely used to treat conditions associated with Qi and blood deficiency, such as fatigue, poor appetite, and weakened immunity. Modern pharmacological studies have demonstrated that Sijunzi Decoction enhances immune function, promotes hematopoiesis, and improves the body's self-repair capabilities^15,16. Building on this foundation, JPSXD incorporates additional herbs to further enhance its efficacy in nourishing blood and address more complex deficiencies. For instance, prepared rehmannia and angelica are key blood-nourishing herbs¹⁷, while astragalus and ginseng work synergistically to boost Qi and improve overall vitality¹⁸.

Recent studies have highlighted the potential anti-tumor effects of JPSXD, suggesting that its multi-component, multi-target approach may inhibit tumor growth and enhance the efficacy of conventional cancer treatments^19,20. The formula's ability to modulate immune responses and improve hematopoietic function makes it particularly valuable in supporting patients undergoing chemotherapy or radiation therapy, who often experience Qi and blood deficiency as a side effect of treatment²¹. Furthermore, the inclusion of herbs like salvia and epimedium has been shown to improve microcirculation and enhance cellular repair mechanisms, contributing to the formula's overall therapeutic benefits²².

Modern pharmacological studies have demonstrated that each individual herb in JPSXD possesses pharmacological activities related to hematopoietic function, anti-myelosuppression, immunomodulation, and antioxidation, as well as microcirculatory improvement, providing indirect research support for the use of JPSXD in alleviating chemotherapy-induced myelosuppression. Table 1 summarizes the primary pharmacological effects of each herb along with the corresponding references, with a focus on studies investigating hematopoiesis, anti-myelosuppressive activity, immunomodulation, and antioxidative or microcirculation-enhancing properties. The complex synergy of components in JPSXD involves both cooperative and antagonistic interactions among the herbs, targeting multiple pathways to achieve a holistic therapeutic effect. For example, while ginseng and astragalus work together to tonify Qi, salvia and prepared rehmannia collaborate to nourish blood and improve circulation. At the same time, herbs like dried tangerine peel and fried barley help regulate digestion and prevent stagnation, ensuring that the tonifying effects of the formula are effectively absorbed and utilized by body²³. This multi-faceted approach aligns with the TCM philosophy of treating the root cause of disease while addressing its symptoms.

In conclusion, JPSXD is a well-balanced and scientifically supported TCM formula that embodies the principles of Qi and blood detoxification. Its potential applications in enhancing immunity, promoting hematopoiesis, and supporting anti-tumor therapy make it a valuable addition to both traditional and modern medical practices. However, high-quality clinical evidence directly evaluating the efficacy of JPSXD in alleviating chemotherapy-induced myelosuppression remains lacking. Further research is needed to elucidate its underlying mechanisms and to optimize its clinical application strategies. Therefore, this study aims to provide proof of concept for the potential therapeutic effects of JPSXD from a preclinical mechanistic perspective.

Network pharmacology, with its systemic and integrative nature, has been widely used to identify active ingredients of Chinese medicines and elucidate their mechanisms of action^24,25. In this study, network pharmacology was employed to analyze the potential signaling pathways and key targets affected by JPSXD in the context of myelosuppression. Additionally, a cyclophosphamide (CTX)-induced myelosuppression mouse model was established to assess relevant indicators and evaluate the effects of JPSXD on myelosuppression. Preliminary validation of the key targets and pathways identified through network pharmacology was also conducted. This is the first study to systematically integrate network pharmacology with animal experiments to elucidate the molecular mechanism by which JPSXD alleviates chemotherapy-induced myelosuppression through the multi-target regulation of signaling pathways, such as the PI3K-AKT pathway, thereby filling a gap in the mechanistic understanding of the hematopoietic protective effects of this compound formula.

All animal experiments were conducted in strict accordance with the Regulations on the Administration of Laboratory Animals and the ethical guidelines of the Experimental Animal Center of Hunan Cancer Hospital. The study protocol was approved by the Animal Ethics Committee of Hunan Cancer Hospital (Approval No.: KNZY-202416). Throughout the study, researchers made every effort to minimize the number of animals used and to alleviate their suffering.

Animal preparation
A total of 60 10-week-old male C57/BL6 mice were obtained from Hunan Slack Jingda Laboratory Animal Co., Ltd. (License No.: SCXK [Hu] 2021-0002). All mice were housed in the Experimental Animal Center of Hunan Cancer Hospital under controlled environmental conditions: temperature: 23 ± 2 °C, humidity: 50% ± 10%, and a 12 h light/dark cycle (8:30-20:30). Mice had ad libitum access to food and water. Prior to experimentation, they were acclimated to the environment through daily handling for 5 min over a 3 day period. Following acclimatization, all 60 mice were stratified by body weight and randomly assigned into six groups (n = 10 per group) using a computer-generated random number table. The groups included: blank control, model, normal saline (NS) control, high-dose JPSXD (40 mg/kg), medium-dose JPSXD (30 mg/kg), and low-dose JPSXD (20 mg/kg). The randomization process ensured no significant differences in mean body weight across groups (p > 0.05). The randomization sequence was generated and executed by researchers not involved in subsequent experimental procedures to minimize bias.

Material preparation
Safety and waste disposal: This study involved CTX, a highly toxic chemotherapeutic agent. All procedures were conducted in strict accordance with the Regulations on the Safety Management of Hazardous Chemicals and Biosafety Level 2 laboratory standards. Specific precautions included the use of disposable protective clothing, nitrile gloves, N95 respirators, and protective goggles by all personnel. CTX preparation and administration were performed inside a Class II biosafety cabinet. Preparation of the CTX injection solution was carried out in a fume hood using dedicated instruments to prevent the formation of aerosols. Sharps and items contaminated with CTX, such as syringes and gloves, were discarded in puncture-resistant sharps containers. Other contaminated materials (e.g., bedding, animal tissues) were inactivated by immersion in 10% sodium hypochlorite for 24 h and subsequently incinerated by a certified medical waste disposal provider. In the event of CTX spillage, the area was immediately covered with absorbent cotton and neutralized with a 1% sodium thiosulfate solution. The incident was then reported to the safety supervisor.

Drugs and reagents: CTX injection was obtained from Baxter Oncology GmbH, Germany (Batch No.: HJ20160467). JPSXD was prepared by the Department of Traditional Chinese Medicine at Hunan Cancer Hospital. It consisted of 15 traditional herbs -- Panax ginseng, Astragalus membranaceus, Epimedium brevicornum, Atractylodes macrocephala, Dioscorea opposita, Salvia miltiorrhiza, Rehmannia glutinosa, Cuscuta chinensis, Lycium barbarum, Ligustrum lucidum, Angelica sinensis, fried malt, chicken gizzard lining, dried tangerine peel, and Coix seed -- combined according to classical compatibility ratios²⁰. The decoction was concentrated to a final concentration of 3 kg/L (i.e., 3 g/mL). Before gavage, the concentrate was diluted with NS to achieve the required concentrations for each dose group: 4 mg/mL for the high-dose group (40 mg/kg), 3 mg/mL for the medium-dose group (30 mg/kg), and 2 mg/mL for the low-dose group (20 mg/kg)²⁶. Gavage volume was calculated as 0.1 mL per 10 g of body weight, resulting in a final administration volume of 0.2-0.25 mL per mouse (based on a weight range of 20-25 g)²⁷.

ELISA kits: Commercially available ELISA kits for mouse Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF), EPO, and Thrombopoietin (TPO) were used. Serum samples were diluted 1:5 in PBS buffer (pH 7.4) prior to testing. Standards were reconstituted in 100 µL per well.

Secondary antibody: HRP-conjugated goat anti-rabbit IgG was diluted 1:5000 (final concentration: 0.04 µg/mL) in TBST buffer containing 5% non-fat dry milk and 0.05% Tween-20. The solution was freshly prepared and incubated at room temperature for 1 h.

Primary antibodies for Western blot: GAPDH (rabbit polyclonal), AKT (rabbit polyclonal), JAK2 (rabbit monoclonal), and EGFR (mouse monoclonal) were used. All antibodies were diluted in TBST buffer containing 5% BSA: GAPDH (1:1000), AKT (1:1000), JAK2 (1:800), and EGFR (1:800). The incubation volume was 100 µL per membrane.

Tissue Fixative: A 4% paraformaldehyde (PFA) solution was prepared by dissolving PFA powder in PBS (pH 7.4). Tissues were fixed for 24 h.

Hematoxylin and eosin (H&E) staining solution: Hematoxylin solution (0.1% Harris hematoxylin) and eosin Y solution (0.5%) were prepared according to standard pathology protocols²⁸. Staining durations were 5 min for hematoxylin and 3 min for eosin.

Immunofluorescence blocking buffer: PBS buffer containing 10% goat serum and 1% BSA was used for blocking, with a 30 min incubation.

Antibody dilution buffer for immunofluorescence: Primary antibodies (CD34+, TPO, EPO, GM-CSF) for immunofluorescence were diluted in PBS buffer containing 1% BSA, following the manufacturer's recommended dilution ratio (1:200), yielding a final concentration of 2.5 µg/mL. The incubation volume was 50 µL per tissue section.

Induction of an animal model
60 mice were randomly divided into six groups (n = 10 per group): blank control, model, high-dose, medium-dose, low-dose, and NS control. Except for the blank control group, all mice received intraperitoneal injections of CTX (20 mg/kg) 1x daily for 3 consecutive days to establish a myelosuppression model. All model mice exhibited WBC < 1.8 × 10⁹/L, PLT < 85 × 10⁹/L, and >55% reduction in bone marrow nucleated cells at 24 h post-final injection. This protocol aligns with established CTX-induced myelosuppression models^29,30. Following injection, mice were monitored daily for general condition, including activity level, coat appearance, and changes in body weight. From the onset of modeling until the end of the experiment, body weight was recorded at the same time each day using an electronic balance. The body weight change rate was calculated as follows: (final weight / initial weight) x 100%, to evaluate the general physiological status and the potential protective effect of JPSXD. Post-injection, mice were housed in individually ventilated cages (IVCs), each clearly labeled as CTX-contaminated. Bedding was changed every 2 days, and all used bedding was disposed of in accordance with hazardous waste protocols described above.

Intragastric administration
Mice in the high-, medium-, and low-dose groups received JPSXD via gavage at doses of 40 mg/kg, 30 mg/kg, and 20 mg/kg, respectively. The saline control group received an equal volume of NS. Gavage solutions were prepared at concentrations of 4 mg/mL, 3 mg/mL, and 2 mg/mL, respectively, and administered at a volume of 0.1 mL per 10 g of body weight. All mice were gavaged 2x daily for 7 consecutive days. The saline control group received 0.2-0.25 mL per mouse of NS. No treatment was given to the blank and model groups. Absence of coughing or struggling during gavage indicated correct administration technique.

Blood collection and serum preparation
On the 3^rd day after CTX injection, 0.2 mL of blood was collected from the tail vein to evaluate the success of model establishment by measuring peripheral blood cell counts using a fully automated hematology analyzer. At 24 h after the final gavage of JPSXD, blood was collected from the orbital sinus of randomly selected mice (n = 10 per group). A portion of the blood was used for complete blood count analysis, while the remainder was processed for serum preparation. Samples were centrifuged at 3,000 x g for 15 min at 4 °C. The supernatant (a clear, pale-yellow upper layer) was collected as serum and aliquoted for storage at -80 °C for subsequent ELISA assays and liver/kidney function tests. The serum layer was observed to be clearly separated from the blood cell layer, with no signs of hemolysis or turbidity. Hematological parameters measured included WBC, RBC, PLT, and hemoglobin (Hb) levels.

Liver and kidney function assessment
A fully automated biochemical analyzer was used to measure serum liver function markers -- alanine aminotransferase (ALT) and aspartate aminotransferase (AST) -- and kidney function markers-blood urea nitrogen (BUN) and creatinine (Cre). All procedures followed the instrument's standard operating protocol. Specifically, 100 µL of serum was mixed with the corresponding assay kits (ALT, AST, BUN, and Cre) and incubated at 37 °C for 5 min. Absorbance was measured at 340 nm. Each sample was tested in duplicate, and the average value was used for analysis.

Organ index calculation
At 24 h after the final gavage administration of JPSXD, mice were euthanized by gradual-fill carbon dioxide (CO₂) inhalation (flow rate: 20% chamber volume/min) followed by cervical dislocation as a secondary method³¹. Immediately after euthanasia, the liver, kidneys, spleen, and thymus were aseptically harvested through a ventral midline incision. Organs were gently dissected free from surrounding connective tissues and fat, then rinsed in ice-cold normal saline (NS) to remove residual blood. Organs were blotted dry with pre-weighed filter paper under standardized pressure (3 gentle presses) and weighed immediately on a calibrated electronic balance (accuracy: ±0.1 mg) at room temperature (22 ± 1°C).

All procedures were performed on a pre-chilled stainless-steel tray to minimize autolysis. The organ index was calculated as organ weight (mg) divided by body weight (g) to evaluate the effects of JPSXD on major organs. Portions of each tissue were reserved for Western blot analysis, while the remaining samples were fixed in 4% PFA for subsequent histological evaluation.

ELISA
Serum concentrations of TPO, EPO, and GM-CSF were measured using ELISA kits according to the manufacturers' instructions. All centrifugation steps, including sample washing and reagent reconstitution, were performed at room temperature at 1,000 x g for 5 min. Washing steps were conducted using PBST buffer (PBS containing 0.05% Tween-20), with 300 µL of wash solution added per well and repeated 3x. After washing, no residual liquid remained at the bottom of the wells. Following color development, a distinct blue precipitate was visible. Absorbance was measured at 450 nm using a microplate reader after the reaction was terminated.

H&E staining
Following euthanasia, the thymus and femur were harvested. The thymus was weighed to calculate the thymus index (thymus weight [mg]/body weight [g]). After weighing, part of the thymus was used for Western blot analysis, while the remainder, along with the femur, was fixed in 4% PFA. Femur samples underwent graded ethanol dehydration (70%, 80%, 95%, and 100%, each for 1 h). Proper dehydration was indicated by a uniformly white appearance, without softening or deformation. Paraffin-embedded tissue blocks were sectioned at 4 µm thickness using a rotary microtome fitted with high-precision tungsten-carbide blades (Type H31). Sections were floated on a 40 °C water bath containing 0.5% gelatin solution to eliminate wrinkles, then mounted on microscope slides. All sections exhibited intact morphology, smooth edges, and uniform translucency under light microscopy, confirming optimal sectioning quality. H&E staining was performed, followed by mounting with an aqueous mounting medium²⁸. After staining, nuclei appeared blue and cytoplasm pink, with well-defined cellular architecture under microscopic observation using an inverted microscope. Histomorphological analysis was conducted using software. For each H&E-stained section, three random fields were selected at 200x magnification. The software automatically quantified the number of nucleated cells per unit area after manual delineation of the region of interest, and results were expressed as mean ± standard deviation (x̄ ± s).

Western blot
Thymus tissues (50 ± 2 mg/sample) were homogenized in ice-cold RIPA buffer containing 1x protease inhibitor cocktail at a precise ratio of 100 mg tissue per 1 mL buffer, using a motorized pellet pestle homogenizer fitted with RNase-free pestles. Homogenization was performed on ice with 15 pulses (1 s pulse/2 s pause), followed by 30 min incubation on ice with vortexing every 10 min. The homogenate was centrifuged at 12,000 x g for 20 min at 4 °C, and the resulting supernatant was collected for protein quantification. The supernatant appeared clear and transparent, free of precipitates or lipid layers. Protein concentration was measured using a BCA protein assay kit, with a standard curve range of 0-2000 µg/mL. Western blot was performed to assess the expression levels of AKT1, JAK2, and EGFR in thymus tissue, using GAPDH as the internal control. After electrophoresis and membrane transfer, the protein marker (10-200 kDa) displayed clear bands, and the target bands aligned with the expected molecular weights (AKT1: 60 kDa, JAK2: 130 kDa, EGFR: 170 kDa, and GAPDH: 37 kDa).

Densitometric analysis of Western blot bands was conducted using software. Prior to analysis, bands were visually inspected to confirm the absence of smearing or diffusion. The analysis procedure included: Manual selection of target band regions following image import, with automated edge detection by the software, followed by local background subtraction using regions adjacent to the bands to correct gray values. Then, the calculation of the gray value ratio between the target protein and internal control for relative quantification was done. The normalization formula used was:

Relative expression of target protein = (gray value of target band −local background) / (gray value of internal control −local background)

Immunohistochemistry
Immunohistochemistry staining was performed on femur sections. Slices were permeabilized with 0.1% Triton X-100 for 10 min, followed by blocking for 30 min. Primary antibodies were incubated overnight at 4 °C, and secondary antibodies were incubated for 1 h at room temperature. After staining, distinct positive signals with low background and no non-specific staining were observed under a light microscope. Slides were mounted using an anti-fade medium containing DAPI, which stained nuclei blue. The expression of CD34+, TPO, EPO, and GM-CSF was then evaluated. Immunohistochemistry images were analyzed using digital pathology software (Version 2.1.0). Prior to analysis, images were verified to be free of overexposure or signal degradation. The analysis procedure³²included:

Model Selection: A pre-trained bone marrow segmentation model (U-Net-based) was used to automatically identify tissue regions³³.
Threshold Setting: Positive signal thresholds were defined based on the negative control group. Regions with chromogenic intensity exceeding the background mean by three standard deviations were classified as positive.
ROI Definition: The software automatically divided each field into 500 × 500 µm grid regions and calculated the number of positive cells per unit area (positive cell count/field area).
Data Export: A statistical report of positive cell density was automatically generated.

Statistical analysis
Statistical analysis was performed using SPSS software (Version 22.0). Data were expressed as x̄ ± s. Prior to intergroup comparisons, normality was assessed using the Shapiro-Wilk test, and homogeneity of variances was evaluated with Levene's test. If both normal distribution (p > 0.05) and variance homogeneity (p > 0.05) were confirmed, one-way ANOVA was used for group comparisons. Post hoc multiple comparisons were conducted using the Least Significant Difference (LSD) method; to control for Type I error, all post hoc P-values were adjusted using the Bonferroni correction, with adjusted p < 0.05 considered statistically significant.

If data failed to meet the assumptions of normality or homogeneity of variance (p ≤0.05 in either test), the Kruskal-Wallis test was applied, followed by Dunn's test for post hoc analysis with Bonferroni correction. All statistical graphs and visualizations were generated using GraphPad Prism (Version 9.0). The significance level was set at α = 0.05. All reported p-values were adjusted values. Outliers were identified using Grubbs' test (α = 0.05) and excluded only if clearly attributable to experimental error. All remaining data were included in the final analysis. Results are presented as x̄ ± s.

Network pharmacology methodology
This study employed a network pharmacology approach to explore the potential mechanisms by which JPSXD alleviates chemotherapy-induced myelosuppression. First, the chemical constituents of JPSXD and their corresponding targets were retrieved from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP, https://old.tcmsp-e.com/tcmsp.php). The screening criteria were set as follows: oral bioavailability (OB) ≥30% and drug-likeness (DL) ≥0.18. For compounds not included in the TCMSP database, target prediction was conducted using the SwissTargetPrediction platform (https://www.swisstargetprediction.ch/).

To ensure the selected compounds covered the major bioactive constituents of JPSXD, reference was made to the Pharmacopoeia of the People's Republic of China³⁴. Specifically, we identified marker compounds such as ginsenosides, astragaloside IV, icariin, and salvianolic acids by reviewing the Identification and Assay sections for each herb. These compounds, commonly used as quality control markers³⁵, have defined chemical structures and quantitative requirements. For accuracy, we cross-referenced the General Notices and Appendices of the Pharmacopoeia to confirm naming conventions and analytical methods (e.g., high-performance liquid chromatography, HPLC). Analysis confirmed that the screening criteria (OB ≥30%, DL ≥0.18) effectively captured the core compounds of the formula³⁶, including key pharmacopeial constituents such as ginsenosides Rg1 and Rb1 (from Panax ginseng), astragaloside IV (from Astragalus membranaceus), and icariin (from Epimedium brevicornum), indicating that the selected compounds are representative and encompass the primary active ingredients responsible for JPSXD's qi- and blood-tonifying effects.

Subsequently, gene targets associated with myelosuppression were retrieved from the GeneCards database (https://www.genecards.org/), and the results were downloaded into a spreadsheet. Genes with a relevance score ≥44 were selected for further analysis. The intersection of drug targets and disease-related genes was visualized using a Venn diagram, and the overlapping genes were imported into the STRING database (https://string-db.org/). The organism was set to human, and interaction sources were limited to Experiments and Databases with a minimum required interaction score > 0.7. The resulting protein-protein interaction (PPI) network was then analyzed for topological characteristics to identify core targets and visualized using Cytoscape 3.10.2.

Based on the identified core targets, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were conducted using the DAVID database (https://david.ncifcrf.gov/). The parameters were set as: Count ≥5, EASE score ≥0.05, and p-value < 0.01. Enrichment results were visualized through the Microbiological Letter platform (http://www.bioinfo-cloud.org/). Pathway analysis further clarified the key signaling routes and regulatory markers involved, thereby revealing the potential mechanisms through which JPSXD mitigates chemotherapy-induced myelosuppression.

Network pharmacology analysis
Identification of active compounds and targets
We first identified the primary marker compounds of each herb in JPSXD based on the literature to ensure comprehensive selection34. Specifically, by consulting the pharmacopeial entries for relevant herbs (e.g., Panax ginseng, Astragalus membranaceus, Angelica sinensis), we compiled a list of the official marker constituents for all 15 herbal components in the formula. These include: ginsenosides Rg1, Re, and Rb1 (from Panax ginseng); astragaloside IV (from Astragalus); icariin (from Epimedium); salvianolic acid B (from Salvia miltiorrhiza); catalpol (from Rehmannia glutinosa); ligustilide (from Angelica sinensis); atractylenolide (from Atractylodes macrocephala); diosgenin (from Dioscorea opposita); cuscatin (from Cuscuta chinensis); Lycium barbarum polysaccharide (from Lycium barbarum); nuezhenide (from Ligustrum lucidum); maltose (from roasted malt); Gallus gallus polysaccharide (from roasted chicken gizzard lining); hesperidin (from Citrus reticulata); and coixenolide (from Coix lacryma-jobi). All of these compounds are clearly defined in the Pharmacopoeia and serve as quality control indicators.

Subsequently, in the network pharmacology screening, we used the TCMSP database and the SwissTargetPrediction platform, applying criteria of OB ≥30% and DL ≥0.18. By comparing the pharmacopoeial marker compounds with the database screening results, we confirmed that all listed compounds met the OB and DL thresholds. For example, ginsenoside Rg1 (OB = 35.2%, DL = 0.22) and astragaloside IV (OB = 45.1%, DL = 0.25) were successfully captured. This alignment ensured that the screening comprehensively covered the major constituents of JPSXD and accurately represented the overall pharmacological profile of the formula. The TCMSP and Swiss Target Prediction databases were used to identify target proteins associated with the main components of each drug. After merging and removing duplicates, 172 key compounds with 454 targets were obtained. These data were visualized using Cytoscape 3.10.2 (Figure 1).

Network topology parameters were analyzed using the Network Analysis tool, and a network diagram was generated based on the degree values. Among the top 10 components with the highest Degree values were kaempferol, gomisin B, licorice flavanone (Licoagrocarpin), licorice isoflavone (Licoagroisoflavone), licorice flavin B (Glyasperin B), deoxyharringtonine, glabridin, licoisoflavanone, kanzonols W, and glycyrin. The top 20 effector genes ranked by Degree value included CYP19A1 (cytochrome P450 19A1), NOS2 (nitric oxide synthase 2), CDK2 (cyclin-dependent kinase 2), ACHE (acetylcholinesterase), CDK1 (cyclin-dependent kinase 1), CHEK1 (checkpoint kinase 1), PPARD (peroxisome proliferator-activated receptor δ), MAPK14 (mitogen-activated protein kinase 14), SLC6A2 (sodium/chlorine cotransporter protein family member 2), AURKB (aurora kinase B), EGFR (epidermal growth factor receptor), TNF (tumor necrosis factor), RET (RET proto-oncogene), KDR (kinase insert domain receptor), FLT3 (fibroblast growth factor receptor 3), TERT (telomerase reverse transcriptase), TOP1 (topoisomerase 1), MMP2 (matrix metalloproteinase 2), F2 (coagulation factor II), and CYP2C19 (cytochrome P450 2C19).

Identification of disease-related genes
The GeneCards database was used to identify disease-associated target genes, yielding 286 genes. After de-duplication and data processing, 719 target genes of Jianpi Shengxue Tang's components were integrated with the disease-related genes. The PPI network was constructed and visualized using Cytoscape 3.10.2 (Figure 2).

Analysis of the PPI network revealed that the key genes associated with the myelosuppressive effects of Jianpi Shengxue Tang included TP53 (tumor protein P53), AKT1 (protein kinase B1), EGFR (epidermal growth factor receptor), JAK2 (Janus kinase 2), IL6 (interleukin 6), TNF (tumor necrosis factor), BCL2 (B-cell lymphoma 2), IL1B (interleukin 1β), MMP9 (matrix metalloproteinase 9), CDK9 (cyclin-dependent kinase 9), CDK2 (cyclin-dependent kinase 2), KDR (kinase insert domain receptor), and IL2 (interleukin 2), among others.

GO and KEGG pathway enrichment analysis
The intersecting genes were uploaded to the DAVID database for enrichment analysis, covering biological processes (BP), cellular components (CC), and molecular functions (MF). A total of 3,197 BP-related, 326 CC-related, and 197 MF-related genes were identified. The top 10 pathways were visualized using the Microbiology Platform (Figure 3).

KEGG pathway enrichment analysis in the DAVID database identified 137 pathways, with the top 20 pathways visualized in WBC (Figure 4). The results suggest that Jianpi Shengxue Tang may mitigate chemotherapy-induced myelosuppression through multiple signaling pathways, including the PI3K-AKT signaling pathway, MAPK signaling pathway, and AGE-RAGE signaling pathway in diabetic complications. Other enriched pathways included human T-cell leukemia virus 1 infection, hepatitis B, epidermal growth factor receptor tyrosine kinase inhibitor resistance, fluid shear stress and atherosclerosis, endocrine drug resistance, central carbon metabolism in cancer, and antifolate resistance.

Animal experiments
Peripheral blood leukocyte count in mice
At 3 days following intraperitoneal administration of CTX, a significant reduction in peripheral blood leukocyte levels was observed in both the model and experimental groups compared to the blank control group (p < 0.01, Figure 5A). By day 7 post-treatment, the low-dose group exhibited a marked increase in peripheral blood leukocyte count, with significant differences compared to the model group (p < 0.01; Figure 5B).

Serum TPO, EPO, and GM-CSF Levels
Serum levels of TPO, EPO, and GM-CSF were quantified using ELISA after 7 days of oral administration. In the model group, GM-CSF levels were significantly elevated compared to the high-dose, medium-dose, control, and saline groups (p < 0.05). The low-dose group exhibited a noticeable increase in GM-CSF levels, although this did not reach statistical significance (p > 0.05; Figure 5C). GM-CSF is a critical factor that promotes the proliferation and differentiation of immune cells, particularly affecting leukocyte and macrophage function37. The increased GM-CSF levels in the model group suggest potential immune system activation, stimulating immune cell proliferation or secretion. The observed differences between the high-dose and the control groups further indicate that dose-dependent effects on GM-CSF levels are plausible. This suggests that JPSXD may modulate the immune system to influence leukocyte levels.

For EPO, the model group exhibited significantly higher levels compared to the high-dose group (p < 0.05), while no significant differences were observed between the model group and other groups (p > 0.05; Figure 5D). EPO, a key regulator of erythropoiesis38, was elevated in the model group, indicating a potential impact on erythropoiesis, particularly in the high-dose group. This elevation may be associated with hypoxia, anemia, or other physiological changes. However, the lack of significant differences between other groups and the model group suggests that, aside from the high-dose group, other treatments had minimal effects on EPO levels.

TPO levels did not show significant differences between the model group and other groups (p > 0.05; Figure 5E), suggesting that the treatment had limited effects on platelet production or that the changes in TPO were not sufficiently sensitive to be detected statistically significant difference compared to the model group.

Thymus index measurement
As presented in Supplementary Figure 1A, the thymus index was significantly elevated in the high-dose, medium-dose, and low-dose groups compared to the model group, with the high-dose group exhibiting the most pronounced increase (p < 0.01). This finding suggests a dose-dependent enhancement of immune function, particularly in the high-dose group, where thymus enlargement is indicative of immune system activation. The observed increase in thymus index may be attributed to the higher dosage stimulating the immune system, thereby promoting thymocyte proliferation and maturation. The significant differences in the high-dose group imply that immune activation is more robust at higher doses, potentially reflecting the pharmacodynamic properties of the active compounds within the treatment.

Body weight changes, liver and kidney function, and organ index
Body weight monitoring revealed a significant decrease in the model group compared to the blank group (p < 0.05), indicating that CTX induced marked toxicity. In contrast, mice in all JPSXD treatment groups -- particularly the high-dose group -- exhibited less weight loss and showed significant improvement compared to the model group (p< 0.05), suggesting that JPSXD may alleviate chemotherapy-induced cachexia (Supplementary Figure 1B).

Liver and kidney function tests showed significantly elevated serum levels of ALT, AST, BUN, and creatinine in the model group (p < 0.05), indicating potential hepatic and renal impairment caused by CTX. Following treatment with JPSXD, especially in the high-dose group, these parameters decreased significantly (p < 0.05), approaching levels observed in the blank group. This suggests that JPSXD exerted protective effects on liver and kidney function (Supplementary Figure 1C).

Organ index analysis demonstrated a significant reduction in both thymus and spleen indices in the model group (p < 0.05), while liver and kidney indices remained unchanged. In the JPSXD treatment groups -- particularly the high-dose group -- thymus and spleen indices increased significantly (p < 0.05), further confirming the protective effect of JPSXD on immune organs. Additionally, no abnormalities were observed in the liver and kidney indices, ruling out potential hepatorenal toxicity of JPSXD (Supplementary Figure 1D).

Expression of p-Akt, JAK2, and EGFR in the Thymus
Western blot analysis and quantitative assessments (Figure 6A-D) revealed significantly diminished expression levels of EGFR, JAK2, and p-Akt in the model group relative to the control group. Raw data of western blot with full-size blot shown in Supplementary Figure 2. This downregulation indicates suppression of the PI3K-AKT signaling pathway under pathological conditions, which is likely associated with cellular dysfunction and impaired signal transduction. Following treatment with JPSXD, particularly in the low-dose group, the expression levels of EGFR and JAK2 were significantly upregulated, and the phosphorylation level of p-Akt was notably restored. These results suggest that JPSXD may activate the PI3K-AKT signaling pathway by upregulating upstream regulators such as EGFR and JAK2, thereby restoring normal signal transduction. The PI3K-AKT pathway is pivotal in mediating cell survival, proliferation, and apoptosis inhibition39,40. Thus, JPSXD may enhance cell survival, reduce apoptosis under pathological conditions, and promote the recovery of hematopoietic function by modulating this pathway. By targeting the PI3K-AKT pathway, JPSXD demonstrates therapeutic potential in alleviating pathological conditions associated with its dysfunction, highlighting its broad clinical applicability.

Histological examination of mouse femurs
Following a 7-day treatment period, femur samples were collected from mice, fixed in 4% formaldehyde, and processed according to standard histopathological protocols for sectioning, staining, and mounting. Morphological examination of the femur revealed significant infiltration of fat cells in the bone marrow of the model and saline groups41,42, with prominent vacuoles observed (Figure 7). In contrast, no significant vacuoles were detected in the femoral bone marrow of the high-, medium-, and low-dose treatment groups, suggesting that the treatment effectively reversed this pathological change. This phenomenon is likely associated with myelosuppression, which can induce abnormal proliferation of adipocytes within the bone marrow, thereby disrupting normal hematopoietic function.

For nucleated cell counting in femoral bone marrow, the results were analyzed based on the number of nucleated cells per unit area (Figure 8). JPSXD significantly ameliorated CTX-induced myelosuppression. Compared to the model group, the high- and medium-dose groups exhibited a marked increase in nucleated cell count, with the high-dose group achieving the highest count (3,130 cells). This indicates that JPSXD promotes the recovery of bone marrow cells, with therapeutic efficacy potentially correlating with dosage. However, other groups did not fully recover to the levels observed in the blank control group, suggesting that higher doses or extended treatment durations may be necessary to achieve optimal recovery.

Analysis of CD34+, TPO, EPO, and GM-CSF positive areas in mouse femurs
In the immunofluorescence staining analysis of mouse femurs, CTX treatment significantly reduced the number of CD34+ positive cells, which are closely associated with vascularization and endothelial cell function (Figure 9A). Compared to the blank group, CD34+ positive cells were notably decreased in the CTX group (p< 0.01; Figure 9B). However, JPSXD treatment significantly reversed this effect, particularly in the high-dose group, where the rebound of CD34+ positive areas was significantly greater than in both the model and saline groups (p< 0.01; Figure 9B). This suggests that JPSXD may promote neovascularization by enhancing endothelial cell proliferation and migration.

Regarding EPO expression, the density of EPO-positive cells was significantly reduced in the model group (p < 0.01), whereas JPSXD treatment at both high and medium doses markedly restored EPO expression levels (Figure 9C). Quantitative analysis (Figure 9D) revealed that the EPO-positive area in the high-dose group increased by approximately 2.5-fold compared to the model group (p< 0.01), exhibiting a dose-dependent trend. These findings indicate that JPSXD effectively stimulates EPO secretion, thereby facilitating the recovery of erythroid hematopoietic function.

As for TPO expression, the intensity of TPO-positive signals was significantly diminished in the model group (p< 0.01), while the JPSXD-treated groups -- particularly the high-dose group -- showed markedly increased TPO expression (Figure 10A). Quantitative results (Figure 10B) demonstrated that the number of TPO-positive cells in the high-dose group was approximately three times higher than in the model group (p< 0.01), suggesting that JPSXD may promote megakaryocyte differentiation and platelet production by enhancing TPO expression.

Additionally, the expression of TPO and EPO was significantly reduced in the model group, but high- and medium-dose interventions with JPSXD significantly increased the expression of these factors in a dose-dependent manner (p< 0.01; Figure 9D, Figure 10B). This suggests that JPSXD can effectively restore hematopoietic factors and stimulate bone marrow hematopoiesis.

For GM-CSF, although the proportion of positive cells decreased in the model group, the density of positive cells increased, potentially due to an inflammatory response. High- and medium-dose treatments of JPSXD significantly elevated GM-CSF expression (p< 0.01; Figure 10D), with the most pronounced effect observed in the medium-dose group.

Data availability:
The datasets generated during this study are fully available within the article and supplementary files.

Figure 1: Active compound-target network of JPSXD. A visual network of active compounds and their potential targets in JPSXD was constructed using Cytoscape 3.10.2. Green square nodes represent target genes, while the outer circular nodes in different colors represent active compounds derived from different herbs. Edges between nodes indicate interactions between compounds and their corresponding targets. Network data were integrated from the TCMSP database and the Swiss Target Prediction platform, comprising a total of 172 active compounds and 454 potential targets. Please click here to view a larger version of this figure.

Figure 2: PPI network of JPSXD targets related to myelosuppression. The intersection between potential targets of JPSXD and disease-related genes was visualized using Cytoscape 3.10.2. Network data were obtained from the GeneCards database and processed through integration and de-duplication. Each node represents a protein target, and edges indicate PPIs. The color intensity of the nodes corresponds to their degree value, with darker colors indicating higher connectivity. PPI data were sourced from the STRING database with a confidence score threshold set at 0.7. Please click here to view a larger version of this figure.

Figure 3: GO enrichment analysis of target genes affected by JPSXD. The overlapping genes identified from the screening were submitted to the DAVID database (version 6.8) for GO enrichment analysis, covering three categories: biological processes (BP), cellular components (CC), and molecular functions (MF). The top 10 significantly enriched terms (p< 0.05) from each category were selected and visualized using the Microbiology Platform software. Green, orange, and blue bars represent BP, CC, and MF categories, respectively. The x-axis denotes the GO terms, and the y-axis represents the enrichment score. Please click here to view a larger version of this figure.

Figure 4: KEGG pathway enrichment analysis of target genes affected by JPSXD. Overlapping genes were submitted to the DAVID database for KEGG pathway enrichment analysis. The top 20 pathways with the highest significance (p< 0.05) were selected and visualized using WBC software. Bubble color indicates the level of statistical significance, with a gradient from green to red representing decreasing P-values. Bubble size corresponds to the number of genes involved in each pathway. The x-axis shows the enrichment ratio (rich factor), and the y-axis lists the pathway names. Please click here to view a larger version of this figure.

Figure 5: Effects of JPSXD on GM-CSF, EPO, and TPO levels in mice. (A) Comparison of peripheral blood leukocyte counts among groups on Day 3 after cyclophosphamide injection. (B) Comparison of peripheral blood leukocyte counts among groups after 7 consecutive days of intragastric administration. (C) Comparison of serum granulocyte-macrophage colony-stimulating factor (GM-CSF) concentrations among groups. (D) Comparison of serum erythropoietin (EPO) concentrations among groups. (E) Comparison of serum thrombopoietin (TPO) concentrations among groups. Data are presented as mean ± standard deviation (n = 10). Group comparisons were performed using one-way ANOVA followed by the LSD post hoc test. *p < 0.05, **p < 0.01, ****p < 0.0001 indicate statistically significant differences compared with the model group. Please click here to view a larger version of this figure.

Figure 6: Effects of JPSXD on the expression levels of EGFR, JAK2, and p-Akt proteins in mouse thymus tissue. (A) Relative expression levels of epidermal growth factor receptor (EGFR). (B) Relative expression levels of Janus kinase 2 (JAK2). (C) Relative expression levels of phosphorylated protein kinase B (p-Akt). (D) Representative Western blot bands of JAK2, EGFR, and p-Akt, with GAPDH as the internal control. Data are presented as mean ± standard deviation (n = 6). Group comparisons were performed using one-way ANOVA followed by the LSD post hoc test. ****p < 0.0001 indicates statistically significant differences compared with the model group. Please click here to view a larger version of this figure.

Figure 7: Representative H&E staining images of femoral bone marrow tissue in mice from different groups. Shown are histological sections of femoral bone marrow from the control, model, NS, and JPSXD high (H), medium (M), and low (L) dose groups. The model and NS groups exhibited prominent adipocyte infiltration (indicated by arrows) and vacuolar degeneration, whereas vacuole formation was markedly reduced in all JPSXD-treated groups. Sample size: n = 10. Scale bar = 50 µm. Please click here to view a larger version of this figure.

Figure 8: Effects of JPSXD on the number of nucleated cells in femoral bone marrow of mice with CTX-induced myelosuppression. Quantitative analysis of H&E-stained sections was performed using image analysis software to determine the number of nucleated cells per unit area in the femoral bone marrow of mice across different groups. Data are presented as x̄ ± s (n = 10). One-way ANOVA was used for intergroup comparisons, followed by the LSD post hoc test. "", "", and "***" denote statistically significant differences compared with the model group at p < 0.01, p < 0.001, p < 0.0001, respectively. Please click here to view a larger version of this figure.

Figure 9: Immunohistochemical analysis of CD34+ and EPO expression in the femur of mice treated with JPSXD. (A-B) Immunostaining results of CD34-positive regions (green) and quantitative analysis of positive cell count per unit area. (C-D) Immunostaining results of erythropoietin (EPO)-positive regions (red) and quantitative analysis of positive cell count per unit area. Data are expressed as mean ± standard deviation (n = 6). Group comparisons were performed using one-way ANOVA followed by the LSD post hoc test. "", "", and "***" indicate statistically significant differences compared with the model group at p < 0.01, p < 0.001, p < 0.0001, respectively. Please click here to view a larger version of this figure.

Figure 10: Immunohistochemical analysis of TPO and GM-CSF expression in the femur of mice treated with the JPSXD. (A-B) Immunofluorescence staining of thrombopoietin (TPO)-positive regions (red) and quantitative analysis of positive cell counts per unit area. (C-D) Immunofluorescence staining of granulocyte-macrophage colony-stimulating factor (GM-CSF)-positive regions (green) and quantitative analysis of positive cell counts per unit area. Data are expressed as mean ± standard deviation (n = 6). Group comparisons were performed using one-way ANOVA followed by the LSD post hoc test. "**" and "****" indicate statistically significant differences compared with the model group at p < 0.01, p < 0.0001, respectively. Please click here to view a larger version of this figure.

Figure 11: Schematic illustration of the proposed mechanism by which JPSXD alleviates CTX-induced myelosuppression. JPSXD enhances hematopoietic function by upregulating hematopoietic cytokines, such as EPO, TPO, and GM-CSF, activating the EGFR/JAK2-PI3K/AKT signaling pathway, and promoting the phosphorylation of AKT. These effects improve the thymus index, restore bone marrow morphology, and increase CD34⁺ HSC populations, thereby alleviating bone marrow suppression. Please click here to view a larger version of this figure.

Supplementary Figure 1: Effects of JPSXD on body weight, thymus index, and hepatorenal function in cyclophosphamide-induced immunosuppressed mice. (A) Thymus index: body weight and thymus weight were measured at the end of the experiment, and the thymus index was calculated as thymus weight/body weight (mg/g), n = 10. (B) Body weight changes: body weight was recorded every 3 days during administration, data expressed as mean ± SD (n = 10). (C) Hepatic and renal function: after euthanasia, serum was collected for measurement of ALT, AST, BUN, and creatinine (Cre) using a biochemical analyzer, n = 10. (D) Organ indices: thymus, spleen, liver, and kidneys were weighed, and corresponding organ indices (organ weight/body weight, mg/g) were calculated, n = 10. "", "", and "***" in the figure denote statistically significant differences compared with the model group at p < 0.01, p < 0.001, p < 0.0001, respectively. Please click here to download this File.

Supplementary Figure 2: Full-size raw Western blots. Please click here to download this File.

Table 1: Pharmacological effects of herbs in the spleen-strengthening and blood-generating decoction (JPSXD). Summary of major pharmacological actions of each herb included in JPSXD, based on published studies. These effects mainly involve hematopoietic promotion, immunomodulation, and antioxidation.

Amid the escalating global incidence of cancer, the medical community is confronted with dual challenges: not only the treatment of cancer itself but also the management of associated side effects, among which myelosuppression is a particularly common and severe adverse reaction⁴³. This study systematically investigated the role and potential mechanisms of JPSXD in alleviating CTX-induced myelosuppression in mice, employing a combination of network pharmacology and animal experiments. The results suggest that JPSXD may improve symptoms of myelosuppression through a multi-target, multi-pathway regulatory mechanism, highlighting the unique advantages of TCM in leveraging multiple components and targets.

Network pharmacology analysis identified key compounds in JPSXD, including kaempferol, Gomisin B, Licoagrocarpin, and Licoagroisoflavone. These compounds have demonstrated a variety of pharmacological activities in previous studies, including anti-inflammatory, antioxidant, and hematopoietic effects^44,45,46, making them promising candidates for further experimental validation in treating myelosuppression. These compounds represent promising candidates for further experimental validation. Given the complexity of myelosuppression⁴⁷, the synergistic effects of these bioactive compounds, when acting on multiple targets, could offer a multifaceted approach to mitigate the condition. A PPI network diagram, combined with GO and KEGG pathway analyses, highlighted key targets and pathways associated with myelosuppression. Notably, p-Akt, JAK2, and EGFR emerged as crucial targets, with the PI3K-AKT signaling pathway identified as a significant regulator.Studies have demonstrated that the PI3K-AKT signaling pathway plays a vital role in maintaining the proliferation and survival of hematopoietic stem cells (HSCs)^48,49. This pathway regulates key anti-apoptotic and pro-apoptotic proteins, such as the upregulation of Bcl-2 (an anti-apoptotic protein) and the downregulation of Bax (a pro-apoptotic protein). By balancing these proteins, the PI3K-AKT pathway can protect hematopoietic cells from oxidative stress and chemotherapy-induced damage, ultimately promoting hematopoietic recovery⁵⁰. EGF binding to EGFR triggers receptor autophosphorylation, recruits PI3K, and activates Akt, while JAK activation further stimulates the PI3K-AKT pathway⁵¹. In this study, significant increases in p-Akt, JAK2, and EGFR levels were detected in the thymus of mice treated with JPSXD, indicating activation of the PI3K-AKT pathway. This suggests that the treatment may correct myelosuppression through this signaling route.

Key hematopoietic factors, including TPO, EPO, and GM-CSF, regulate bone marrow function and play pivotal roles in recovery from marrow suppression⁵². TPO, primarily secreted by the liver and kidneys, stimulates megakaryocyte differentiation and maturation, promoting platelet production while maintaining HSC homeostasis⁵³. EPO, secreted by the kidneys, promotes the proliferation and differentiation of erythroid progenitor cells via the JAK2/STAT5 signaling pathway, which is critical for correcting anemia⁵⁴. GM-CSF, secreted by endothelial and T cells, induces HSC differentiation into granulocytes and monocytes, enhances phagocytic activity, and accelerates WBC recovery, thereby reducing infection risks⁵⁵. In our experiment, elevated levels of TPO, EPO, and GM-CSF were observed in serum and femur sections from mice treated with JPSXD, suggesting that the treatment's improvement of hematopoietic function is closely linked to enhanced secretion of these regulatory factors.Notably, the thymic index in the JPSXD-treated group also showed a significant increase, further supporting the hypothesis that JPSXD enhances immune function. The activation of the PI3K-Akt signaling pathway, along with the increased thymic index, suggests that JPSXD may promote the survival, proliferation, and differentiation of thymocytes, which play a crucial role in restoring and enhancing immune function⁵⁶. These experimental results strongly indicate that JPSXD has potential therapeutic effects on chemotherapy-induced immunosuppression by regulating key signaling molecules, particularly in enhancing thymic function and immune cell activity. This provides a solid scientific basis for its clinical application in improving immune recovery during chemotherapy⁵⁷.

Histological analysis revealed significant structural damage to the bone marrow in the model group, including disorganized cell arrangement, reduced cellular density, increased adipocyte vacuoles, and a decrease in nucleated cells. These changes indicated a suppression of hematopoietic function and disruption of the hematopoietic microenvironment, which in turn affected normal blood cell generation. Specifically, the increase in adipocyte vacuoles is often associated with the exacerbation of bone marrow adiposity, which may further hinder the proliferation and differentiation of HSCs⁵⁸. Studies have shown that bone marrow adiposity is closely linked to impaired hematopoietic function, as adipocytes can secrete factors that inhibit HSC activity and alter the bone marrow niche⁵⁹.

Treatment with JPSXD resulted in significant improvements in these pathological alterations. JPSXD treatment restored the normal bone marrow structure, reduced the formation of adipocyte vacuoles, and notably increased the proportion of nucleated cells. This suggests that JPSXD not only improved the bone marrow microenvironment but also effectively promoted the survival and proliferation of HSCs, thereby aiding in the restoration of normal hematopoietic function. The ability of JPSXD to modulate the bone marrow microenvironment may be attributed to its anti-inflammatory and antioxidant properties, which have been shown to mitigate chemotherapy-induced damage and support hematopoietic recovery⁶⁰.

In conclusion, the beneficial effects of JPSXD on bone marrow structure are not only evident from histological changes but also reflect its comprehensive regulatory role in the bone marrow hematopoietic microenvironment, as it repairs bone marrow damage and improves cell arrangement. In this experiment, we assessed the thymus of mice for p-Akt, JAK2, and EGFR protein levels and found that these proteins were significantly elevated in the JPSXD-treated mice compared to the model group. This suggests that the PI3K-Akt signaling pathway was activated. Activation of these pathways may promote the survival, proliferation, and differentiation of thymocytes, which contribute to the restoration and enhancement of immune function. The experimental data strongly indicate that JPSXD has potential therapeutic effects on chemotherapy-induced immunosuppression by regulating key signaling molecules, particularly in enhancing thymic function and immune cell activity, thus providing a solid scientific basis for its clinical application. Restoring the proportion of nucleated cells, JPSXD holds potential as an effective therapeutic agent for bone marrow suppression-related diseases, especially in the treatment of chemotherapy-induced bone marrow damage and other hematopoietic dysfunctions. In addition, further studies are needed to elucidate the specific molecular mechanisms underlying the protective effects of JPSXD. For example, flow cytometric analysis of bone marrow LSK cells (Lin⁻Sca-1⁺c-Kit⁺) and downstream apoptosis-related proteins (such as Bcl-2, Bax, and Caspase-3) may provide deeper insights into its mechanism of action. Clinical validation is also necessary to confirm its efficacy and safety, thereby laying the groundwork for broader clinical application.

Despite these promising results, this study has limitations. First, this study was a preclinical animal experiment, and its conclusions require further validation through subsequent clinical trials. Second, the experimental design did not include clinically used leukocyte-boosting agents, such as recombinant human G-CSF, as a positive control. This limitation restricts our ability to directly assess the efficacy of JPSXD relative to standard therapies. However, it is important to emphasize that the primary objective of this study was to elucidate, for the first time, the potential signaling pathways -- such as PI3K-AKT -- through which JPSXD mitigates myelosuppression, based on network pharmacology and experimental validation, rather than to perform a comparative efficacy analysis. Currently, high-quality clinical data on the use of JPSXD for chemotherapy-induced myelosuppression are lacking, underscoring the necessity of this study and highlighting that its clinical translational value remains to be confirmed. Third, current gene databases linking drugs and diseases remain incomplete. This study primarily relied on the GeneCards database to identify disease-related targets. Although high-confidence targets were selected using a relevance score threshold, relying on a single database may not comprehensively capture all potential therapeutic targets, thereby limiting the completeness and accuracy of the network pharmacology analysis and pathway identification. Future research should integrate multiple specialized databases -- such as DisGeNET, OMIM, and TTD -- to leverage their complementary strengths for target identification and validation, thus improving the coverage and reliability of the target dataset. In addition, further research is necessary to identify the specific bioactive compounds within JPSXD and elucidate their mechanisms of action. Future studies should incorporate both in vitro assays and clinical trials to substantiate these findings. Nevertheless, the present results provide valuable preliminary experimental evidence supporting the potential of JPSXD as a therapeutic candidate for chemotherapy-induced myelosuppression.

Conclusion
Our findings demonstrate that JPSXD ameliorates chemotherapy-induced myelosuppression via modulation of the PI3K-AKT signaling pathway (Figure 11). These results position JPSXD as a promising adjuvant therapy for CTX-induced myelosuppression, with potential for broader clinical application in mitigating chemotherapy-associated adverse effects.