Synergistic potentiation of the anti-metastatic effect of a Ginseng-Salvia miltiorrhiza herbal pair and its biological ingredients via the suppression of CD62E-dependent neutrophil infiltration and NETformation

Abstract

Introduction: The combination of the roots of ginseng and Salvia miltiorrhiza is an effective approach for treating metastatic cancer in patients with Qi stagnation and blood stasis patterns. However, the molecular mechanism underlying the combined use of ginseng and Salvia miltiorrhiza is unknown.

Objectives: This study unveils the pharmacological foundation of ginseng and Salvia miltiorrhiza by examining the involvement of neutrophils in the critical process of tumor hematogenous metastasis. Additionally, by employing a reverse pharmacology research model (effect-target-constituent), potential potent components were screened, and the dominant component formulations were determined.

Methods: An experimental lung metastatic model was constructed to compare the antitumor effects of ginseng and Salvia miltiorrhiza. RNA sequencing was employed to identify pivotal biological events and key targets, while the detection of CD62E expression and neutrophil extracellular traps (NETs) release was used to screen for effective substances in ginseng and Salvia miltiorrhiza. In addition, a comprehensive array of in vitro and in vivo experiments was conducted to explore the underlying mechanisms and therapeutic significance.

Results: Compared with single-herb use, the use of ginseng or Salvia miltiorrhiza significantly reduced tumor metastasis, which was accompanied by reduced neutrophil infiltration into the lungs. Cryptotanshinone (CPT), an active constituent of Salvia miltiorrhiza, can inhibit neutrophil adhesion and recruitment to lung tissue by downregulating the expression of E-selectin (CD62E) in endothelial cells. Moreover, the ginseng-derived ginsenoside Rg1 mitigated the formation of NETs in lung tissues and reversed the protumor effects of NETs. We further explored the efficacy of combination therapy with Rg1 and CPT, which also reduced tumor metastasis in vivo.

Conclusion: Ginseng and Salvia miltiorrhiza exhibited a mutual potentiation of the anti-metastatic effect by suppressing both early and late stages of neutrophil-initiated metastasis cascade. Rg1 and CPT represent the synergistic ingredients from ginseng and Salvia miltiorrhiza, respectively.

Keywords: Cryptotanshinone; E-selectin; Ginseng-Salvia miltiorrhiza herbal pair; Ginsenoside Rg1; Neutrophil extracellular traps; Neutrophil response.

Graphical abstract

Fig. 1

Ginseng and Salvia miltiorrhiza modulated the neutrophil-associated TIME in the lungs. (A) Experimental setup. The mice were constructed as experimental metastasis models after a period of acclimatization. The mice were prepared as experimental metastasis models following an acclimatization period. Ginseng and Salvia miltiorrhiza extracts were administered daily to the relevant groups according to the schematic diagram, and lung tissues were collected on day 28. (B) Representative bioluminescence images of metastasis in the intravenously implanted LLC mice. (C) Representative H&E images of the tumor area in the lungs of the mice (scale bar: 50 μm). (D) Quantitative analysis of the lung tumor area (n = 3). (E) Quantitative analysis of the number of lung metastatic nodules in tumor-bearing mice(n = 15). (F) Metastasis rates of LLC tumor-bearing mice (n = 15). (G) GOBP analysis of intersecting genes in the control, model and G + S groups. (H) GSEA of genes related to neutrophil migration. (I) Representative images of immunohistochemical staining and quantitative analysis of Ly6G in the lungs (n = 3; scale bar: 50 μm). (J) Representative images of immunofluorescence staining and quantitative analysis of Ly6G in the lungs (n = 3; scale bar: 100 μm). All the data are presented as mean ± SEM. #### p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs the model group.

Fig. 2

Ginseng and Salvia miltiorrhiza reduced CD62E expression, and effective substances from Salvia miltiorrhiza were screened. (A) GSEA of genes related to leukocyte adhesion to vascular endothelial cells. (B) Gene expression analysis of adhesion molecules. (C) Representative immunohistochemical images and quantification of CD62E expression in the lungs (n = 3, scale bar 100 μm). (D) HUVECs were stimulated with TNF-α for 4 h with or without pretreatment with various compounds (10 μM) for 24 h. The mRNA expression levels of CD62E were determined by qRT–PCR. The chemical structure of CPT is shown on the right. (E–F) Representative western blot images of CD62E levels in HUVECs and quantitative data. GAPDH was used as a loading control. (G) Representative immunofluorescence images of CD62E levels in HUVECs (scale bar, 50 μm). All the data are presented as mean ± SEM. #p < 0.05, ##p < 0.01, ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. the model group.

Fig. 3

CPT inhibited CD62E–mediated neutrophil adhesion and recruitment to metastases. (A) Representative images and quantification of HL–60 cell adhesion to HUVECs (scale bar: 200 μm). (B) Schematic diagram of the parallel plate flow chamber combined with the microscope. (C) Number of neutrophils firmly adhered to HUVECs. (D) Representative immunofluorescence images and quantification of CD62E in the lungs (n = 3; scale bar: 50 μm). (E) Representative western blot images of CD62E levels in the lungs and quantitative data (n = 3). GAPDH was used as a loading control. (F) The mRNA expression levels of TNF–α, ICAM1, and VCAM1 in lung tissues were determined by qRT–PCR (n = 3). (G) The correlation between CD62E expression and neutrophil infiltration in LUAD and LUSC was analyzed via TIMER. (H) Representative images and quantification of Ly6G⁺ neutrophils in lungs detected by flow cytometry (n = 4). (I) Representative immunofluorescence images and quantification of Ly6G⁺ neutrophils in the lungs (n = 3; scale bar: 50 μm). All the data are presented as mean ± SEM. ##p < 0.01, ###p < 0.001, ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ****p < 0.0001 vs. the model group.

Fig. 4

Neutrophil RNA sequencing analysis and correlation analysis of NETs and lung cancer. (A) Schematic diagram of neutrophil isolation and RNA sequencing from mouse lung tissues. (B-C) Heatmap analysis of neutrophil-related genes and KEGG analysis of altered genes. (D) Results of partial GOBP analysis of intersecting genes in the control, model and G + S groups. (E) The expression of HIST3H3 in different cancers and adjacent normal tissues was analyzed via TIMER. (F–H) The correlation between copy number alterations in HIST3H3, ELANE, and MPO and neutrophil infiltration was analyzed via TIMER. (I) The correlations between cumulative survival and HIST3H3, ELANE, MPO and neutrophil infiltration in lung cancer patients were analyzed via TIMER. A two-sided Wilcoxon rank-sum test was used. *p < 0.05, **p < 0.01, ***p < 0.001 vs. the control group.

Fig. 5

Ginseng and Salvia miltiorrhiza reduced the generation of NETs in lung tissues, and effective substances from ginseng were screened. (A) Representative immunohistochemical images of MPO expression in the lungs (scale bar: 100 μm). (B) Representative immunofluorescence images of H3cit and NE in the lungs (scale bar: 100 μm). (C–E) Quantitative analysis of MPO, H3cit and NE expression in the lungs (n = 3). (F) Neutrophil MPO release was detected by ELISA. The chemical structure of Rg1 is shown on the right. (G) Representative images and quantification of extracellular DNA release after Sytox™ Green staining (scale bar: 50 μm). All the data are presented as mean ± SEM. ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. the model group.

Fig. 6

Rg1 reduced NET formation by inhibiting ROS production and ERK1/2 and MAPK phosphorylation. (A) Representative immunofluorescence images of the NET components H3cit and NE in neutrophils (n = 3; scale bar: 50 μm). (B–C) Representative western blot images of H3cit expression in neutrophils and quantitative data. GAPDH was used as a loading control (n = 3). (D-E) ROS production in dHL-60 cells (D) and neutrophils (E) was assessed by flow cytometry. The median fluorescence intensity of ROS was used for quantification. (F-G) Representative western blot images and quantification of p-ERK1/2, ERK1/2, p-MAPK and MAPK expression in dHL-60 cells (F) and neutrophils (G). GAPDH was used as a loading control. All the data are presented as mean ± SEM. #p < 0.05, ###p < 0.001, ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ****p < 0.0001 vs. the model group.

Fig. 7

Rg1 reduced NETs in lung tissues. (A) Representative immunofluorescence images and quantification of H3cit and NE in the lungs (n = 3; scale bar: 50 μm). (B) Representative western blot images of H3cit and NE levels in the lungs and quantitative data (n = 3). GAPDH was used as a loading control. (C-E) Serum levels of CXCL1, CXCL2, and G-CSF were detected via ELISA (n = 10). (F-H) The mRNA expression levels of CXCL1, CXCL2 and G-CSF in lung tissues were determined by qRT–PCR (n = 3). All the data are presented as mean ± SEM. #p < 0.05, ##p < 0.01, ###p < 0.001, ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. the model group.

Fig. 8

Rg1 reversed NET-induced invasion, EMT, and adhesion of LLC cells and increased vascular permeability. (A) LLC cell proliferation was detected via a CCK8 assay. (B) Schematic diagram of NET isolation and coculture with LLC cells in the transwell system. (C) Images of the transwell invasion assay (scale bar, 200 μm). The number of infiltrated cells and the OD values were used for quantification. (D-E) Representative immunofluorescence images and quantification of MMP9 in the lungs (n = 3, scale bar 50 μm). (F) Representative immunofluorescence images of E-cadherin and N-cadherin in LLC cells treated with or without Rg1 or NETs. (G-I) Representative immunofluorescence images and quantification of E-cadherin and N-cadherin in the lungs (n = 3, scale bar 50 μm). (J) Schematic diagram of the adhesion assay of LLC cells by NETs. (K) Representative fluorescence images of the adhesion assay for DiI-labeled LLC cells trapped within PMA- or Rg1-treated neutrophils. (L) RNA-seq analysis of neutrophils. GSEA of genes related to the regulation of vascular permeability. (M) Schematic diagram of the endothelial permeability assay. (N) The effects of Rg1 or NETs on vascular permeability were tested via an endothelial permeability assay. (O-P) Representative immunofluorescence images and quantification of CD31 and FITC-dextran in the lungs (n = 3; scale bar, 50 μm). All the data are presented as mean ± SEM. #p < 0.05, ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ****p < 0.0001 vs. the model group.

Fig. 9

Rg1 combined with CPT inhibited lung cancer metastasis in vivo. (A) Experimental design of Rg1- and CPT-treated mice. (B) Monitoring of mouse body weight (n = 10). (C) Representative micro-CT images of mouse lungs. (D) Quantitative analysis of the number of nodules in the lungs (n = 10). (E) Representative images of the lungs. (F) Representative H&E staining images of the lungs and quantification of the tumor area (n = 3; scale bars: 800 μm and 80 μm). (G) Representative H&E staining images of livers and spleens (scale bar: 100 μm). All the data are presented as mean ± SEM. ####p < 0.0001 vs. the control group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. the model group.