1. INTRODUCTION
Bladder cancer (BC) is a common malignant tumor of the urinary system. Approximately 573,000 new cases of BC and 213,000 related deaths occur worldwide every year [1]. Occupational exposure to industrial chemicals (such as amines) and tobacco smoking contribute to BC risk [2]. BC is generally classified into non-muscle-invasive bladder cancer (NMIBC), which accounts for approximately 75% of newly diagnosed cases and is characterized by isolation to the lamina propria and urothelium, and muscle-invasive bladder cancer (MIBC), which accounts for the remaining 25% of cases, and invades muscles and additional tissues [3]. Although NMIBC has better prognosis than MIBC, its long-term management is costly, and approximately 20% of cases recur and progress to MIBC [4]. In contrast, MIBC has a high risk of recurrence and frequently metastasizes to the lungs, bones, and liver [5, 6]. Although surgery, chemotherapy, radiotherapy, and immunotherapy are among the therapeutic methods available for invasive BC, metastatic disease has a high mortality rate. Consequently, antimetastatic drugs are urgently needed.
Research and development of new drugs largely rely on natural products. Traditional Chinese medicines (TCMs) have a long and widespread history of use in China and regions of Asia. In recent years, TCMs and their active ingredients have been reported to possess antitumor, anti-inflammatory, and analgesic properties [7]. We screened various TCMs classified as heat-clearing and detoxifying herbs, and found that the ethanol extract of Paris polyphylla (chonglou) has outstanding antitumor activity, including inhibition of cell viability, induction of cell cycle arrest, and suppression of tumor growth [8, 9]. P. polyphylla is traditionally used to treat headaches, fevers, burns, wounds, and snake poisoning, and its ingredients are used in more than 70 types of proprietary medicines, such as Jidesheng sheyao tablets, Chonglou jieduding tablets, and Gongxuening capsules [10]. In the Chinese Pharmacopoeia (2020 version), P. polyphylla is identified by the content of polyphyllins (PPs, including PPI, PPII, and PPVII). We further found that PPII, which has excellent antitumor activity, inhibits epithelial-mesenchymal transition (EMT) and the expression of matrix metalloproteinases (MMPs)—proteins that play essential roles in tumor metastasis [8]. However, the detailed mechanism through which PPII mediates antimetastatic functions in BC cells and in vivo conditions remains to be clarified.
Metastasis is a common cause of cancer-related death [11]. Motility and elongation changes triggered by cytoskeletal rearrangement are key factors in tumor cell migration, invasion, and metastasis [11, 12]. The RHO GTPase family, a prominent branch of the Ras superfamily, participates in multiple tumor processes, from initiation to development, including EMT and metastasis [13, 14]. RHOA, CDC42, and RAC1 are well-characterized RHO GTPases that regulate the assembly of actin and the organization of the cytoskeleton through LIMK and CFL1 phosphorylation in various tumor cell types [13, 15]. Our bioinformatics results indicated that PPII affects cytoskeletal rearrangement in T24 cells; however, the underlying mechanisms through which RHO GTPases-associated signaling contribute to inhibiting PPII-induced BC cell migration, invasion, and metastasis remain poorly understood. Herein, we used a variety of in vitro and in vivo methods to elucidate the roles of ROCK1/LIMK/CFL1 pathway-mediated cytoskeletal rearrangement in PPII-mediated suppression of BC migration, invasion, and metastasis; in addition, we evaluated the toxicity of PPII in mice.
2. MATERIALS AND METHODS
2.1 Cell culture
The T24 and 5637 BC cell lines and the SV-HUC-1 (normal bladder) cell line were kindly provided by Professor Shengtian Zhao (Binzhou Medical University, Yantai, Shandong) and grown in RPMI-1640 medium (CM10041; Macgene Technology) (for T24 and 5637 cells) or F12K medium (CM10025; Macgene Technology) (for SV-HUC-1 cells) with 10% fetal bovine serum (10270-106, Gibco). The cell lines were incubated at 37°C in a humidified atmosphere containing 5% CO2.
2.2 Cell viability assays
We assessed the sensitivity of BC cells to PPII (A0387, Chengdu Must Bio-Technology) and determined drug concentrations via CCK-8 assays (E-CK-A362, Elabscience). T24 cells (3 × 103 cells/well) and 5637 cells (6 × 103 cells/well) were seeded in 96-well plates and incubated overnight. The cells were subsequently exposed to cisplatin (0–48 μM) or PPII (0–3.2 μM for T24 cells and 0–8 μM for 5637 and SV-HUC-1 cells) at varying concentrations for 48 h. After treatment, the cells were incubated with CCK-8 reagent at 37°C for 2 h, and the optical density at 450 nm was measured via a spectrophotometer (type 1510, Thermo Fisher Scientific).
2.3 Wound healing assays
For wound healing assays, T24 (6 × 105 cells/well) and 5637 (8 × 105 cells/well) cells were seeded in six-well plates to form complete monolayers. Three sets of linear scratch wounds were generated on the cell monolayers with a sterile 200-μl pipette tip. The cells were cultured in RPMI-1640 medium supplemented with 1% FBS. Intercellular distances were measured at 0, 24, and 48 hours. The migration rate was determined as follows: migration distance/primary intercellular distance × 100%.
2.4 Transwell invasion and migration assays
Transwell assays were conducted with an 8.0 μm pore size upper chamber (REF353097, BD Biosciences) with or without precoated Matrigel (354234, BD Biosciences), to assess invasion and migration, respectively. T24 (1 × 105 cells per well) or 5637 (2 × 105 cells per well) cells subjected to various treatments were added to the chambers and incubated 48 h. After incubation, the cells were fixed with 4% paraformaldehyde (PFA) solution (P0099, Beyotime) for 10 minutes. Subsequently, the cells were stained for 30 minutes with 0.1% crystal violet solution (G1061, Solarbio) and washed three times with PBS. Finally, a cotton swab was used to carefully remove the cells and Matrigel from the upper chamber. The invasive cells present on the lower surface were then quantified via an inverted microscope (Vert. A1, Zeiss). For each group, three fields were observed and captured at a magnification of ×100.
2.5 Western blotting and antibodies
Total protein was extracted from T24 and 5637 cells treated with PPII alone or cotransfected with plasmids (pcDNA3.1 or pcDNA3.1-RHOA L63) with jetPRIME transfection reagent (101000027, Polyplus Transfection). Western blot analysis was conducted as previously described [8]. Primary antibodies to the following were used: p-LIMK1/2 (1:1000, #3841S), LIMK1 (1:1000, #3842S), and p-CFL1 (1:1000, #3313S) from Cell Signaling Technology; and RHOA (1:2000, 10749-1-AP), ROCK1 (1:2000, 21850-1-AP), CFL1 (1:5000, 66057-1-Ig), and GAPDH (1:10000, 60004-1-Ig) from Proteintech. Protein band intensities were quantified in ImageJ software (version 1.53q; National Institutes of Health).
2.6 RNA sequencing and data analysis
Total RNA was extracted with an MJzol Animal RNA Extraction Kit according to the manufacturer’s protocol. RNA quality was assessed with a NanoPhotometer® spectrophotometer and Agilent 2100 Bioanalyzer. The RNA-seq libraries were prepared with an Illumina TruSeq RNA Sample Prep Kit and sequenced on the NovaSeq 6000 system. Differential expression analysis was conducted with the DESeq2 package [16], to identify significantly differentially expressed genes (DEGs) on the basis of P < 0.05 and a |log2FoldChange| > 1.25. The TPM values of the DEGs were used for c-means clustering with the Mfuzz package in R to identify dynamic expression patterns [17]. Functional enrichment analysis for Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways was performed with the clusterProfiler package [18], with a significance threshold of P < 0.05.
qPCR was used to detect the mRNA levels of the genes whose expression was most altered in the groups treated with 0.4 μM PPII versus 0 μM PPII. RNA extraction (AC0205, SparkJade) and cDNA template preparation (AG11728, Accurate Biology) were performed according to the manufacturer’s instructions. qPCR was performed with a reagent kit (AG11701, Accurate Biology) and the primers described in Table 1 .
Primer sequences.
| Gene | Primer (5′ to 3′) | Length (bp) |
|---|---|---|
| HNRNPC-F | GCCAAAAGTGAACCGAGGAAAAGC | 24 |
| HNRNPC-R | CCGAGCAATAGGAGGAGGAGGAG | 23 |
| RPL17-F | AGAACACTCGTGAAACTGCTCAGG | 24 |
| RPL17-R | ACCTGCCAACTCCACCATTGTAAC | 24 |
| NAA40-F | TGTAAGCGAGTGTCTGGACTGGAG | 24 |
| NAA40-R | GGCTCGGTCATCTGTCATTTCCTC | 24 |
| ZNF619-F | TGGTTCAGCATCAGCGAGTTCAC | 23 |
| ZNF619-R | GTTGGGGAGGCAGAGGAGAGAG | 22 |
2.7 Phalloidin staining assays
T24 and 5637 cells were treated with a concentration gradient of PPII (0, 0.2, or 0.4 μM) for 48 hours. The cells were then reseeded onto microscope slides in 24-well plates and incubated overnight. The cells were subsequently fixed with 4% PFA (P1110, Solarbio) for 10 minutes, permeabilized with 0.5% Triton X-100 (T8200, Solarbio) for 5 minutes, and incubated with 150 nM TRITC-phalloidin (CA1610, Solarbio) for 1 hour. After nuclei were stained with DAPI (C0065, Solarbio), the cells were washed three times with PBS. Representative images were captured with an Invitrogen EVOS M7000 Imaging System, and ImageJ software was used to assess the average F-actin intensity.
2.8 Immunohistochemistry
Lung samples were fixed overnight with PFA solution (G1101, Servicebio). The samples were subsequently dehydrated, embedded in paraffin, and sectioned. Antigen retrieval was performed by boiling the sections in sodium citrate buffer (EE0005, SparkJade) for 20 minutes. To suppress endogenous peroxidase activity, the sections were immersed in blocking solution (EE0007, SparkJade) at room temperature for 10 minutes, then blocked with 10% serum at 37°C for 30 minutes to minimize nonspecific staining. Subsequently, the sections were incubated overnight at 4°C with primary antibodies. After being rinsed, the sections were treated with HRP-conjugated secondary antibody (1:200, EF0002, SparkJade) and incubated for 40 minutes at 37°C. Immunoreaction sites were visualized with DAB, and this was followed by counterstaining with hematoxylin and mounting on glass slides. The mean integrated optical density (IOD) of the staining signal (indicating positive expression) was quantified for five samples per group in Image-Pro Plus software (version 6.0, Media Cybernetics). Primary antibodies to the following were used for immunohistochemistry (IHC): Ki-67 at 1:500, GB111141, Servicebio; ROCK1 at 1:800, 21850-1-AP, Proteintech; p-LIMK1 at 1:200, AF3345, Affinity; and p-CFL1 at 1:100, AP0178, ABclonal Technology.
2.9 Hematoxylin and eosin staining
The tissue sections were deparaffinized and rehydrated. The sections were stained with hematoxylin solution for 5 minutes, washed, treated with differentiation medium for 0.5 minutes, rinsed with distilled water, and stained with eosin solution for 2 minutes. The sections were subsequently dehydrated with a graded alcohol series, soaked in xylene, and sealed on slides. Imaging was performed with a scanning imaging system (WS-10, Wisleap).
2.10 Detection of serum biochemical parameters
Serum biochemical parameters, including alkaline phosphatase (ALP), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels, were measured with specific assay kits (BC2145 for ALP, BC1555 for ALT, and BC1565 for AST, Beijing Solarbio Science & Technology) according to the manufacturer’s instructions.
2.11 Animal studies
Male Nu/Nu nude mice (5 weeks old) were obtained from Beijing Vital River Laboratory Animal Technology (license number: SCXK (Beijing) 2021-0006). The mice were housed under pathogen-free conditions with a 12-hour light/dark cycle, and given ad libitum access to food and water. All animal experiments were approved by the Ethics Committee of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine (approval No. AWE-2022-016).
For subchronic toxicity testing of PPII in vivo, the mice received intraperitoneal (IP) injections of 0, 3, or 6 mg/kg PPII every 2 days. After 63 days of treatment, the mice were euthanized, and blood samples were collected. Heart, liver, spleen, lung, kidney, and stomach tissues were harvested and fixed in tissue fixative (G1101, Servicebio). Hematoxylin and eosin (HE) staining was performed to examine morphological changes in the tissues, and the serum concentrations of ALP, ALT, and AST were determined as described above.
For analysis of lung metastasis, T24 cells (2.0 × 106) resuspended in 100 μl PBS were injected into the tail vein of each Nu/Nu nude mouse. The mice received IP injections of PPII at doses of 0, 1.5, or 3.0 mg/kg every 2 days for 63 days. After treatment was completed, the mouse lungs were dissected and embedded in paraffin for HE and IHC analysis. The weights of the lungs and livers were recorded for each group.
2.12 Statistical analysis
Statistical analysis was performed via analysis of variance. The data are presented as mean ± standard deviation. The levels of statistical significance are indicated as follows: *P < 0.05, **P < 0.01, and ***P < 0.001. Nonsignificant differences (P > 0.05) are denoted with NS.
3. RESULTS
3.1 PPII inhibits BC cell migration and invasion
The sensitivity of BC cells to PPII was evaluated by treatment of T24, 5637, and SV-HUC-1 cells with a concentration gradient of cisplatin (0–48 μM for cells; Figure 1A–C ) as a control or increasing concentrations of PPII (0–3.2 μM for T24 cells and 0–8 μM for 5637 and SV-HUC-1 cells; Figure 1D–F ) for 48 h. The IC50 values of cisplatin and PPII were then calculated for the three cell lines. PPII showed significant inhibitory effects on the viability of both T24 and 5637 cells, with IC50 values of 1.00 ± 0.04 μM and 1.87 ± 0.17 μM, respectively, thus indicating that PPII had greater inhibitory effects than cisplatin (17.95 ± 1.08 μM for T24 and 20.57 ± 0.27 μM for 5637). Moreover, SV-HUC-1 cells were less sensitive to cisplatin (29.53 ± 1.71 μM) and PPII (3.89 ± 0.42 μM) than T24 and 5637 cells.
Effects of PPII on T24 and 5637 cell viability.
Cytotoxicity of cisplatin as a positive control against T24 (A), 5637 (B), and SV-HUC-1 (C) cells after 48 h treatment. Cytotoxicity of PPII in T24 (D), 5637 (E), and SV-HUC-1 (F) cells after 48 h treatment. Screening for PPII concentrations not affecting cell viability in T24 (G) and 5637 (H) cells. NS, no significant difference; ***, P < 0.001; PPII, polyphyllin II.
Considering the potential interference of drug-induced toxicity on the invasion and migration of tumor cells, we treated BC cells with low concentrations of PPII (0–0.8 μM) to identify noncytotoxic ranges. CCK-8 assays suggested that cell viability remained unaffected under 0–0.4 μM PPII treatment, whereas significant inhibition of viability was observed under 0.8 μM PPII treatment (P < 0.001, Figure 1G, H ) in both T24 and 5637 cells.
The effect of PPII on BC migration was investigated with wound healing assays in T24 and 5637 cells, which revealed that 0.4 μM PPII significantly inhibited wound closure in both cell lines (P < 0.01, Figure 2A–D ). To assess the inhibitory effect of PPII on invasion, we subjected T24 and 5637 cells to Transwell invasion assays after treatment with PPII (0, 0.2, or 0.4 μM) for 48 h. PPII, even at low concentrations, significantly restrained the invasion of both T24 and 5637 cells (P < 0.05, Figure 2E–H ). Therefore, PPII suppressed BC cells’ migratory and invasive ability.
PPII inhibits migration and invasion of BC cells.
Wound healing assays of T24 (A) and 5637 (C) cells treated with PPII (0, 0.2, or 0.4 μM) for 48 h, and statistical analysis of the migration rates are presented in B and D, respectively. Transwell invasion assays of T24 (E) and 5637 (G) cells treated with PPII (0, 0.2, or 0.4 μM) for 48 h. Cells that crossed the membrane were counted and are presented in F and H. Scale bar: 50 μm; NS, no significant difference; *, P < 0.05, **, P < 0.01, ***, P < 0.001.
3.2 RNA sequencing analysis of T24 cells treated with PPII
For bioinformatics analysis of the molecular mechanism underlying the PPII-induced inhibition of BC cell migration and invasion, we treated T24 cells with PPII (0, 0.2, or 0.4 μM) for 48 h, and subsequently performed RNA sequencing. We identified 120 upregulated genes and 117 downregulated genes in the 0.2 μM PPII-treated group with respect to the control group (0 μM PPII, Figure 3A ), and 230 upregulated genes and 367 downregulated genes in the 0.4 μM PPII-treated group with respect to the control group ( Figure 3B ) with a fold change value > 1.25 of DEGs. The qPCR results confirmed that the mRNA levels of HNRNPC and RPL17 were increased, whereas the levels of NAA40 and ZNF619 were decreased in 0.4 μM PPII-treated group compared to control group ( Figure 3C ), thus confirming the reliability of the sequencing results.
RNA sequencing analysis of T24 cells treated with PPII.
Volcano plot showing the distribution of upregulated and downregulated genes in the 0.2 μM PPII-treated group versus the 0 μM PPII-treated group (A) and the 0.4 μM PPII-treated group versus the 0 μM PPII-treated group (B). qPCR detection of the mRNA levels of the top DEGs in the 0.4 μM PPII-treated group versus the 0 μM PPII-treated group (C). Gene expression heatmap showing a targeted cluster of genes whose expression levels decreased with increasing drug concentration (D). GO (E) and KEGG (F) analyses of the targeted cluster genes.
Among all sequenced genes, one cluster (denoted the target cluster) of genes ( Figure 3D ) notably showed changes in gene expression consistent with the changes in drug concentration. That is, the expression levels of genes in this cluster, as compared with those in the control group, changed slightly after treatment with low concentrations of PPII (0.2 μM) but clearly decreased after treatment with high concentrations of PPII (0.4 μM).
GO and KEGG analyses of the target cluster suggested enrichment in a series of terms associated with cell migration, invasion, and metastasis, such as positive regulation of cell adhesion, extracellular structure organization, basement membrane organization, and ECM-receptor interaction ( Figure 3E, F ). GO analysis further revealed that genes associated with cytoskeletal regulation, such as the regulation of actin filament-based process, regulation of actin cytoskeleton organization, and actin filament organization, were enriched in the target cluster ( Figure 3E ), whereas regulation of actin cytoskeleton was enriched in the KEGG analysis ( Figure 3F ). However, no specific cytoskeletal regulatory pathway was identified. Sorting of genes associated with cytoskeletal regulation in the GO ( Suppl. Table S1 ) and KEGG ( Suppl. Table S2 ) enrichment analyses revealed involvement of RHO GTPase-associated effectors, including RHO, RAC, and LIMK. Our findings indicated that the ability of PPII to inhibit BC cell motility was closely associated with cytoskeletal rearrangement, and that RHO GTPase-associated pathways might play important roles in PPII-mediated cytoskeletal regulation in BC cells.
In summary, the treatment of T24 cells with PPII resulted in enrichment in gene expression associated with cell motility and cytoskeletal regulation. Consequently, we focused on the changes in the cytoskeleton induced by PPII.
3.3 PPII represses cytoskeletal formation and CFL1 phosphorylation in BC cells
Given that cytoskeletal rearrangement plays a crucial role in cell migration, invasion, and metastasis, we conducted F-actin staining assays with phalloidin to assess whether PPII treatment might induce cytoskeletal changes in BC cells. Representative images for each group revealed noticeably diminished cell protrusion and staining intensity under treatment with high concentrations of PPII (0.4 μM, Figure 4A ). Statistical analysis of the average F-actin intensity suggested a significant decrease in staining intensity, particularly in the 0.4 μM groups of T24 and 5637 cells compared to 0 μM groups (P < 0.05, Figure 4B, C ).
PPII suppresses cytoskeletal formation in BC cells.
T24 and 5637 cells were treated with PPII for 48 h, after which actin filaments were stained with TRITC-phalloidin (red), and nuclei were stained with DAPI (blue). Scale bar: 75 μm (A). Average calculated F-actin intensity of T24 (B) and 5637 (C) cells subjected to TRITC-phalloidin staining. T24 cells were treated with a PPII concentration gradient, as indicated, for 48 h, and the expression levels of cytoskeletal regulation-associated ROCK1, p-LIMK1/2, and p-CFL1 were detected by western blotting. Relative expression levels are displayed under the corresponding lanes (D). Bar chart showing relative expression levels of ROCK1 to GAPDH; p-LIMK1/2 to LIMK1; and p-CFL1 to CFL1 (E). The 5637 cells were treated similarly to T24 cells, and western blotting results and relative protein expression levels are presented in D and F, respectively. NS, no significant difference; *, P < 0.05, **, P < 0.01, ***, P < 0.001.
The RHO GTPase pathway is widely recognized to play an essential role in cytoskeletal regulation. Our RNA sequencing results revealed the involvement of RHO GTPase-associated pathways in cytoskeletal changes after PPII treatment. CFL, which can sever actin filaments, is a common effector of multiple RHO GTPase pathways that regulates cytoskeletal polymerization and depolymerization; therefore, we detected the phosphorylation levels of CFL1 and its kinase LIMK1/2 via western blotting. Comparing to the control group (0 μM PPII), PPII decreased the levels of p-LIMK1/2 and p-CFL1 ( Figure 4D ). Further grayscale analysis revealed that p-LIMK1/2/LIMK1 and p-CFL1/CFL1 levels significantly decreased in the groups after treatment with 0.4 μM PPII (P < 0.05, Figure 4E, F ). The relative expression level of ROCK1, an upstream regulator of LIMK1/2, also decreased after 48 h treatment with 0.2 μM and 0.4 μM PPII in T24 and 5637 cells (P < 0.05, Figure 4D–F ). Thus, PPII decreased the expression levels of ROCK1 and suppressed the phosphorylation of effectors, including LIMK1/2 and CFL1.
3.4 PPII inhibits RHOA-mediated migration, invasion, and CFL1 phosphorylation
Given that ROCK1 acts as an effector of RHOA instead of CDC42 and RAC1, and that RHOA is abnormally activated in BC, we further assessed the changes in cell migration, invasion, and CFL1 phosphorylation under RHOA activation in BC cells.
The effects of PPII on the invasion and migration of BC cells transfected with constitutively active RHOA (RHOA L63) were detected through Transwell assays. Comparing to the control group, migration assays without Matrigel revealed that RHOA L63 overexpression increased the numbers of T24 and 5637 cells that crossed the Transwell membrane, whereas PPII treatment inhibited the increase induced by the activation of RHOA ( Figure 5A, B ). Statistical analysis of cell count data confirmed that RHOA L63 overexpression significantly increased the numbers of BC cells that crossed the Transwell membrane; this number was further suppressed by the addition of 0.4 μM PPII ( Figure 5C, D ). Invasion assays with Matrigel were also performed with T24 ( Figure 5E, G ) and 5637 ( Figure 5F, H ) cells and yielded results consistent with those of the migration analysis. Briefly, RHOA L63 transfection increased the invasive ability of BC cells, whereas this increase was inhibited by the addition of 0.4 μM PPII.
PPII inhibits cell migration and invasion, and CFL1 phosphorylation mediated by RHOA.
Transwell migration assays of T24 (A) and 5637 (B) cells transfected with constitutively active RHOA mutant (RHOA L63) and treated with PPII for 48 h. Cells that crossed the membrane were counted and are presented in C and D, respectively. Transwell invasion assays of T24 (E) and 5637 (F) cells transfected with RHOA L63 and treated with PPII for 48 h. Cells that crossed the membrane were counted and are presented in G and H. Cells were treated as described above, and the expression levels of RHOA, ROCK1, and p-CFL1 were detected by western blotting. Relative expression levels are displayed under the corresponding lanes (I). Relative expression levels of ROCK1 to GAPDH, and p-CFL1 to CFL1 are shown in J and K, respectively. Scale bar: 50 μm; Statistical results are presented as means ± standard deviations of three independent experiments. *, P < 0.05, **, P < 0.01, ***, P < 0.001.
BC cells were treated with 0.4 μM PPII, and RHOA L63 was simultaneously overexpressed. Western blotting revealed that RHOA L63 overexpression increased the expression levels of ROCK1 and p-CFL1 compared to the control group, whereas treatment with 0.4 μM PPII inhibited this increase ( Figure 5I ). Grayscale analysis also showed that the levels of ROCK1/GAPDH and p-CFL1/CFL1 increased with overexpression of RHOA L63 but decreased after PPII treatment (P < 0.05, Figure 5J, K ).
In summary, the invasion and migration of T24 and 5637 cells were inhibited by PPII under RHOA activation. Mechanistically, PPII not only suppressed the expression of ROCK1 and p-CFL1 but also attenuated their expression in BC cells overexpressing RHOA L63.
3.5 Subchronic toxicity of PPII in mice
PPII treatment at 15 mg/kg and 25 mg/kg has been described in an in vivo study [19]; nevertheless, PPI and PPVI were used at a concentration of 5 mg/kg [20, 21]. Given the major differences in the concentrations used in animal experiments with different PPs and the lack of subchronic toxicity evaluation of PPII, we administered PPII (0, 3, or 6 mg/kg) to Nu/Nu nude mice via the IP route every 2 days.
After 63 days of PPII treatment, abdominal distension was observed in the 6 mg/kg treatment group ( Suppl. Figure S1A–C ; weight curve in Figure 6A ). Mice were euthanized to obtain heart, liver, spleen, lung, kidney, and stomach tissues, which were weighed separately. Increases in the dimensions of the liver ( Figure 6B ) and spleen ( Suppl. Figure S1D ) were observed in 6 mg/kg PPII-treated group compared to control group. Furthermore, the weights of heart, liver, spleen, and lung tissues in the 6 mg/kg treatment group were significantly greater than those in the control group (0 mg/kg; P < 0.05, Figure 6C and Suppl. Figure S1E ). Moreover, no significant differences were observed in the weights of kidney and stomach tissues between the PPII-treated groups and the control group ( Figure 6C and Suppl. Figure S1E ).
PPII induced weight increases and morphological changes in the livers and kidneys of Nu/Nu mice.
Mice were euthanized after 63 days of PPII administration (0, 3, and 6 mg/kg; five mice per group) every 2 days. Weight curves of the treated groups are shown (A). Livers from the groups treated with a concentration gradient of PPII are presented (B). Bar charts showing the differences in the weights of the liver, lungs, and kidneys between groups (C). HE staining indicating morphological changes in the liver, lungs, and kidneys of groups treated with PPII. Scale bar: 100 μm (D). ALP, ALT, AST, and the AST/ALT ratio are displayed in bar charts (E). NS, no significant difference; **, P < 0.01, ***, P < 0.001.
Subsequently, we conducted HE staining analysis on the above tissues. Compared with the control group, the group treated with 6 mg/kg PPII showed mild granular degeneration and visible focal necrosis in liver cells. In addition, aggregation of foam cells in the lungs, mild hypertrophy of myocardial cells, and extramedullary hematopoiesis in the spleen were found in the 6 mg/kg PPII treatment group, whereas the morphology of the kidney and stomach tissues did not change at the macroscopic level between treatments ( Figure 6D and Suppl. Figure S1F ). Furthermore, we detected liver serological indicators. Although no significant difference was observed in ALP, ALT, or AST between the PPII-treated groups and the control group, the AST/ALT ratio was significantly elevated in the 6 mg/kg treatment group (P < 0.01, Figure 6E ), thus indicating negative effects on liver function.
The subchronic toxicity assay of PPII in Nu/Nu nude mice suggested that IP administration of 6 mg/kg PPII led to alterations in the weight and morphology of multiple tissues, including those of the heart, liver, spleen, and lungs, along with a significant increase in the AST/ALT ratio. In contrast, these effects were not observed at the 3 mg/kg dose. On the basis of these findings, we conducted subsequent animal experiments with drug concentrations of 0, 1.5, or 3.0 mg/kg.
3.6 PPII inhibits BC cell metastasis in vivo
On the basis of a previous report verifying successful lung colonization of T24 cells after injection into the caudal vein in mice [22], we further examined the effects of PPII on lung colonization and the formation of metastatic nodules by T24 cells, according to previously described methods. Nu/Nu nude mice were subjected to IP administration of PPII for 63 days after tail vein injection of T24 cells (2.0 × 106 cells per mouse, Suppl. Figure S2A ), and the weights of mice in the three groups (0, 1.5 and 3.0 mg/kg PPII) were plotted as weight curves ( Suppl. Figure S2B ). The lung tissues obtained from euthanized mice and the surface nodules on the lung tissue of each group are shown in Figure 7A . In agreement with the subchronic toxicity assessment results, PPII at concentrations of 1.5 or 3.0 mg/kg did not lead result in weights of the lungs ( Figure 7B ) or livers ( Suppl. Figure S2C ) differing from those in the control group.
PPII inhibits lung metastasis of BC cells.
Mice were injected with T24 cells via the tail vein and euthanized after 63 days of PPII administration (0, 1.5, or 3.0 mg/kg; five mice per group) every 2 days. Representative pulmonary metastatic nodules are shown. Scale bar: 5 mm (A). Differences in the weights of lung tissue (B) and the numbers of nodules (C) among the three treatment groups are shown in bar charts. HE staining revealing representative pulmonary metastatic nodules in the indicated groups (D). IHC showing the expression levels of ROCK1 (E) and p-CFL1 (F) in the indicated PPII treatment groups (scale bar: 100 μm). Differences in the mean IOD between groups were calculated. Arrows represent representative metastatic nodules. IOD, integrated optical density; NS, no significant difference; *, P < 0.05, **, P < 0.01, ***, P < 0.001.
Statistical analysis of the numbers of tumor nodules among treatment groups revealed a decreasing trend in the group treated with 1.5 mg/kg PPII (IP) compared to the control group; however, no significant difference was observed. In contrast, the 3.0 mg/kg PPII group exhibited a significant decrease compared to 0 mg/kg PPII-treated group ( Figure 7C ). Representative tumor nodules detected by HE staining in each group are shown in Figure 7D . Furthermore, PPII (3.0 mg/kg) treatment significantly inhibited F-actin intensity at lung metastasis sites ( Suppl. Figure S2D and E ). The IHC results suggested significant downregulation of cell proliferation-associated Ki-67 expression levels in the 3.0 mg/kg PPII treatment group (P < 0.001, Suppl. Figure S2F, G ). Moreover, the protein levels of ROCK1 (P < 0.001, Figure 7E ), p-LIMK1 (P < 0.05, Suppl. Figure S2H, I ), and p-CFL1 (P < 0.05, Figure 7F ), which are associated with cytoskeletal regulation, were also significantly lower in the treatment group than the control group.
In conclusion, PPII significantly inhibited the expression of the cytoskeletal regulatory proteins ROCK1, p-LIMK1, and p-CFL1 in vivo, and consequently impeded the formation of metastatic nodules after tail vein injection of T24 cells.
4. DISCUSSION
Our results indicated that PPII inhibits the invasion and migration of T24 and 5637 cells. Bioinformatics analysis revealed that PPII affects cytoskeletal rearrangement in BC cells, as indicated by diminished TRITC-phalloidin staining. Therefore, the inhibitory effect of PPII on BC cell invasion and migration might be achieved through induction of cytoskeletal rearrangement. Moreover, PPII treatment not only decreased the levels of ROCK1 and p-LIMK1, thereby inhibiting CFL1 phosphorylation, but also decreased the level of p-CFL1 when constitutively active RHOA L63 was overexpressed, thereby indicating that PPII suppressed RHOA-mediated CFL1 phosphorylation in BC cells. Moreover, the enhanced invasion and migration of BC cells induced by RHOA L63 was inhibited by PPII. Furthermore, subchronic toxicity detection demonstrated that IP administration of 6 mg/kg PPII in Nu/Nu nude mice led to morphological changes in multiple organs, including the liver and spleen, as well as dysregulation of serological indicators, such as the AST/ALT ratio. However, the 3 mg/kg concentration had limited adverse effects and was used in subsequent experiments. PPII was found to inhibit the formation of metastatic pulmonary nodules induced by tail vein injection of T24 cells, as well as the expression of ROCK1, p-LIMK1, and p-CFL1 associated with cytoskeletal regulation in vivo.
We previously reported that P. polyphylla ethanol extract upregulates the expression of CDKN1A; inhibits the expression of CCNB1 and CDK1; induces G2/M phase arrest in J82 cells; and inhibits tumor growth in vivo [9]. In addition, we previously demonstrated that PPII, a main component of P. polyphylla, regulates EMT-associated factors, and decreases levels of MMP2 and MMP9 [8]. In the present study, we further revealed the molecular mechanism through which PPII inhibits the invasion and migration of T24 and 5637 cells through the ROCK1/LIMK/CFL1 pathway and cytoskeletal rearrangement; evaluated the subchronic toxicity of PPII; and finally determined the role of PPII in inhibiting metastasis in vivo ( Figure 8 ).
PPII inhibits the metastasis signaling cascade through cytoskeletal rearrangement in BC cells (created in Figdraw; https://www.figdraw.com/#/).
PPs from P. polyphylla have been widely used in antitumor experiments in vivo. For example, 1–3 mg/kg PPI has been used in animal models of lung cancer and ovarian cancer [23–25], and glioma models established in BALB/c nude mice have been treated with 5 mg/kg PPVI [21]. However, PPII has also been administered at concentrations of 15 mg/kg and 25 mg/kg via IP injection into BALB/c nude mice with ovarian cancer xenograft tumors [19]. In addition, most of the animal studies on PPs described above used the IP administration route. The concentration of PPII used for lung metastasis of BC in Nu/Nu nude mice and its toxicity to the body had not been reported; therefore, we used IP administration to detect the effects of a concentration gradient of PPII (0, 3, and 6 mg/kg) on multiple organs in nude mice. Our results suggested increased liver, spleen, lung, and heart tissue weights; changes in tissue morphology revealed by HE staining; and an elevated AST/ALT ratio in the 6 mg/kg PPII treatment group, whereas 3 mg/kg PPII had relatively mild effects and was used in subsequent antimetastatic analyses. In subchronic toxicity evaluation, the 6 mg/kg treatment group also had ascites (four of five mice); therefore, the ratio of tissue to body weight was not used in the statistical analysis to minimize bias.
The inactive form of CFL is phosphorylated (Ser3), whereas its active form is nonphosphorylated; its cleavage activity increases the efficiency of G-actin dissociation from F-actin, thus leading to F-actin depolymerization [26]. However, our conclusions did not consider the reuse of F-actin, although the depolymerized F-actin end can be reused for actin polymerization [26]. Therefore, CFL plays a dual regulatory role in both actin polymerization and depolymerization. Evidence suggests that CFL is involved in tumor cell invasion and migration [27, 28]; that diminished CFL phosphorylation inhibits invasion and migration [29]; and the decrease of p-CFL level also reported to promote cell motility [30]. Our results indicated that PPII inhibits the invasion, migration, and metastasis of BC cells by inhibiting CFL1 phosphorylation mediated by the ROCK1/LIMK axis, thus inhibiting tumor cell motility.
RHO GTPases, an important branch of the Ras superfamily, regulate cytoskeletal rearrangement and are abnormally expressed in tumors. Moreover, these GTPases are considered potential biomarkers or therapeutic targets for cancer [13]. As important RHO GTPases, RHOA [31], RAC1 [32], and CDC42 [33] have been extensively studied in BC, and are closely associated with tumor progression, particularly migration, invasion, and metastasis [34–36]. In the present study, PPII treatment was found to inhibit the expression of ROCK1, and the phosphorylation of its downstream effector factors LIMK1/2 and CFL1. Subsequently, the upstream regulator of ROCK1, the constitutively active mutant form of RHOA (RHOA L63), was overexpressed. Western blot and Transwell assays revealed that PPII inhibited RHOA-mediated CFL1 phosphorylation, as well as BC cell invasion and migration. Nevertheless, we cannot rule out the possibility RAC1 and CDC42 might play roles in the above process. Whether RAC1 and CDC42 participate in the regulatory effects of PPII on cytoskeletal rearrangement in BC cells must be verified in further studies.
5. CONCLUSION
In conclusion, this study revealed that PPII inhibits the phosphorylation of CFL1, and the invasion and migration of T24 and 5637 cells, via the ROCK1/LIMK axis in BC cells. Subchronic toxicity evaluation revealed that a drug concentration of 3.0 mg/kg had limited adverse effects on tissue morphology and the AST/ALT ratio in Nu/Nu mice. This concentration was confirmed to have inhibitory effects on BC lung metastasis after tail vein injection. Because of its potential benefits, PPII may be a valuable candidate for the development of future treatments aimed at managing BC metastasis.
