Hispidulin suppresses cell growth and metastasis by targeting PIM1 through JAK2/STAT3 signaling in colorectal cancer.

Colorectal cancer (CRC) accounts for over 600 000 deaths annually worldwide. The current study aims to evaluate the value of proto-oncogene PIM1 as a therapeutic target in CRC and investigate the anticancer activity of hispidulin, a naturally occurring phenolic flavonoid compound, against CRC. Immunohistochemistry analysis showed that PIM1 was upregulated in CRC tissue. The role of PIM1 as an oncogene was evidenced by the fact that PIM1 knockdown inhibits cell growth, induces apoptosis, and suppresses invasion. Our results showed that hispidulin exerts antitumor activity in CRC through inhibiting the expression of PIM1. Moreover, our findings revealed that hispidulin downregulated the expression of PIM1 by inhibiting JAK2/STAT3 signaling by generating reactive oxygen species. Furthermore, our in vivo studies showed that hispidulin can significantly inhibit tumor growth and metastasis in CRC. Collectively, our results provide an experimental basis for trialing hispidulin in CRC treatment. PIM1 can be considered a potential therapeutic target in CRC.

colorectal cancer

2′,7,‐dichlorofluorescein‐diacetate

immunohistochemistry

interleukin

N‐acetylcysteine

nuclear factor‐κB

poly(ADP‐ribose) polymerase

provirus integration site for Moloney murine leukemia virus

reactive oxygen species

signal transducer and activator of transcription

INTRODUCTION

Colorectal cancer ranks as the second most common malignant tumor in women and the third most common malignancy in men,1 and accounts for over 600 000 deaths annually worldwide.2 In China, the incidence of CRC has been increasing with 400 000 new cases diagnosed each year.3 Currently, surgery and chemotherapy are the mainstream treatments for CRC. Surgery cures patients diagnosed at early stages. However, chemotherapy is the only option for approximately one‐quarter of patients diagnosed with CRC at the advanced stage. Despite the use of aggressive therapeutic regimens, the prognosis of patients with CRC is far from ideal.4 Therefore, developing novel therapeutic agents for CRC patients is imperative. The proto‐oncogene PIM1 encodes serine/threonine protein kinases that modulate cell survival, differentiation, proliferation, apoptosis, and tumorigenesis.5 The Pim family comprises PIM1, PIM2, and PIM3. The activity of all 3 kinases directly correlates with their expression levels due to the lack of a regulatory domain.6 Among all members of the Pim family, the diagnostic and prognostic value of PIM1 in human malignancies has been extensively investigated.7 Several human malignancies from hematopoietic and epithelial origins have established the oncogene role of PIM1.8 However, a high level of PIM1 is associated with a favorable prognosis in prostate cancer,9 pancreatic ductal carcinoma,10 and non‐small‐cell lung cancer.11 With regards to CRC, a total PIM1 score that reflects the expression of PIM1 in tumors, tumor stroma, and tumor‐adjacent mucosa has been found to correlate with disease stages, and can be considered a useful prognostic marker of CRC.12 However, the value of PIM1 as a therapeutic target remains to be evaluated.

Flavonoids have been reported to provide health benefits for humans. Hispidulin (4′,5,7‐trihydroxy‐6‐methoxyflavone), is a kind of phenolic flavonoid compound isolated mainly from Salvia involucrata, a medicinal plant traditionally used in oriental medicine.13 In recent years, accumulating evidence has revealed the pleiotropic effects of hispidulin, including antioxidant, anti‐inflammatory, neuroprotective, antithrombotic, anti‐osteoporotic, and anti‐epileptic activities.14, 15, 16, 17, 18, 19, 20 More importantly, many in vivo and in vitro studies have reported that hispidulin shows its antitumor effects against a variety of solid tumors and hematological malignancies, including acute myeloid leukemia, hepatocellular carcinoma, colon carcinoma, gallbladder cancer, and renal cell carcinoma.21, 22, 23, 24, 25 Our previous studies have identified that hispidulin inhibited the growth and metastasis of cells by regulating the activity of sphingosine kinase 1 in renal cell carcinoma.22 However, its role in CRC and its underlying mechanisms remain elusive.

In the present study, the suppressive role of hispidulin on CRC growth and metastasis was investigated in vitro and in vivo. Moreover, our study was aimed at examining what role PIM1 plays in the antitumor effect of hispidulin against CRC, and exploring the pathways responsible for the regulating effect of hispidulin on PIM1.

MATERIALS AND METHODS

Clinical samples

This study was given official approval by the ethics committee of Qingdao University (Qingdao, China) and written consent forms were approved by all patients. A total of 63 patients with CRC who underwent surgical procedures between June 2014 and February 2016 were recruited in this study. Colorectal cancer tissue and adjacent non‐cancer tissue were sampled and stored at −80°C. Diagnosis and staging were carried out by 2 independent senior oncologists blinded to the data. The level of PIM1 in tissue samples was determined by an IHC assay. The primary antibodies used for the IHC assay were as follows: polyclonal antibodies against PIM1 and Ki‐67 (Abcam, Shanghai, China) and polyclonal antibodies against cleaved caspase‐3 (Cell Signaling Technology, Danvers, MA, USA). The expression levels of marker molecules were quantified as previously described.26 To explore the correlation between PIM1 expression and clinicopathological characteristics, the tissue samples were allocated into 2 groups: (i) <25% positive cells defined as (−); and (ii) ≥75% positive cells defined as (+).

Cell culture

Human CRC cell lines HT29 and SW480 were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 10% FBS (Invitrogen) and 1% penicillin/streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) were applied to culture cells at 37°C in a humidified incubator containing 5% CO2.

Cell counting kit‐8 assay

The cell viability was detected and evaluated by CCK‐8 (Beyotime, Shanghai, China) following THE manufacturer's instructions. A spectrophotometer (Thermo Fisher Scientific, Shanghai, China) was applied to measure the optical density of viable cells.

Flow cytometry analysis

Colorectal cancer cells were harvested following treatment, and an apoptosis kit (BD Pharmingen, Franklin Lakes, NJ, USA) was used to stain cells with annexin V–phycoerythrin and propidium iodide. Cell apoptosis was determined using a flow cytometer (Thermo Fisher Scientific, Shanghai).

Transwell invasion assay

Transwell membranes were pretreated with Matrigel (8‐μm pore size; BD Biosciences, San Jose, CA, USA). After starved for 24 hours, cells cultured in serum‐free medium were harvested and resuspended in DMEM supplemented with 1% FBS at a density of 2 × 105 cells/mL. Then 500 μL cells was added to the upper chamber of the Transwell, and MEM containing 10% FBS was used to fill the lower chamber as a chemoattractant. After incubation for 24 hours, the cells and the Matrigel on the surface of the upper chamber were removed with cotton swabs. Staining solution containing 4% formaldehyde and hematoxylin were applied to fix and stain the migrated cells adhered to the surface of the lower membrane. The cells adhered to the lower chamber were counted and photographed at 200× magnification in 5 randomly chosen fields.

Quantitative real‐time PCR

TRIzol reagent (Thermo Fisher Scientific, Waltham) was used to extract the total RNA from cultured cells. Briefly, a High Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA) was applied to synthesize cDNA. The primer sequences for PIM1 mRNA expression were: forward, 5′‐CGGCAAGTTGTCGGAGACG‐3′; and reverse, 5′‐CCTGGAGGTTGGGATGCTCT‐3′. The Super M‐MLV reverse transcriptase (BioTeke, Beijing, China) was used to produce first‐strand cDNA. The SYBR Green Master Mix (Beijing Solarbio Science & Technology, Beijing, China) was used to execute PCR reactions; GAPDH was applied as an internal control to normalize PCR results. The analysis of PCR results was carried out using the comparative ΔCt method (ABPrism software; Applied Biosystems).

Western blot analysis

Following standard protocols, the PVDF membranes of proteins were incubated with specific antibodies. β‐Actin and specific primary polyclonal antibodies against PIM1, cleaved caspase‐3, and cleaved PARP were obtained from Abcam. The mAb against JAK2 was purchased from Bosterbio (Beijing, China). The polyclonal antibodies against phospho‐JAK2 (Tyr221), STAT3, and phospho‐STAT3 (Ser727) were obtained from Cell Signaling Technology. Goat anti‐rabbit IgG‐HRP (Beyotime) was applied as the secondary antibody in this study. β‐Actin was used to normalize target gene expression as an internal control. Protein levels were measured by chemiluminescent substrate (KPL, Guildford, UK) and visualized by BandScan software (Glyko, Novato, CA, USA).

Reactive oxygen species assay

The generation of ROS in CRC cells was detected by fluorescent probe DCFH‐DA. Colorectal cancer cells were harvested following treatment and resuspended in DCFH‐DA (10 μmol/L) at 37°C for 30 minutes in darkness. A laser scanning confocal microscope (Olympus, Tokyo, Japan) was used to observe the fluorescence signals of cells.

PIM1 knockdown

Two siRNAs were synthesized by GenePharma (Shanghai, China) as previously described.27 The cording strands of the siRNAs were CCAUGGAAGUGGUCCUGCUGAAGAA (Pim1‐1) and GUAUGAUAUGGUGUGUGGAGAUAUU (Pim1‐2). Non‐silencing scramble siRNA was purchased from GenePharma. The siRNA sequence that showed better knockdown efficiency was used in the experiment. The PIM1 targeting the siRNA and scramble sequence as a control were respectively inserted into the plasmid vector pEGFP‐N1 (Wuhan Miaoling Bioscience and Technology, Wuhan, China). Western blot analysis measured the expression of PIM1 48 hours after transfection.

PIM1 overexpression

Empty vectors were carried into CRC cells by Lipofectamine 3000 reagent (Invitrogen, Grand Island, NY, USA) in accordance with the manufacturer's instructions. As previously described, plasmid vector pEGFP‐N1 (Wuhan Miaoling Bioscience and Technology) was used to construct the PIM1 overexpressing vector.28 The expression of PIM1 48 hours after transfection was measured by Western blot analysis.

In vivo tumor growth model

The Institutional Animal Care and Use Committee at Qingdao University approved all animal experiments in this study. Eight‐week‐old male athymic BALB/c nu/nu mice were given sterile food and water in pathogen‐free conditions. The mice were injected with HT29 cells (107 cells) in their left flanks. After 21 days of injection, the mice were randomly divided into 3 groups (8 mice/group) and injected i.p. as follows: (i) vehicle (0.9% sodium chloride plus 1% DMSO); (ii) hispidulin (20 mg/kg/d, dissolved in vehicle) (Beijing Solarbio Science & Technology) HPLC ≥98%; and (iii) hispidulin (40 mg/kg/d, dissolved in vehicle). The body weight and tumor volume of mice were measured twice every week until day 30. Cryostat sections (4 μm/section) of HT29 xenograft tumors were stained using IHC staining and measured using the TUNEL assay according to the protocols described in our previous work.29

Immunohistochemical analysis

An IHC analysis was carried out on 4‐mm, formalin‐fixed, paraffin‐embedded tumor tissue sections according to the following procedures. Briefly, consecutive sections were deparaffinized in xylene, rehydrated in a graded ethanol series, and submerged in EDTA antigenic retrieval buffer for 15 minutes in a microwave oven. The sections were treated with 3% hydrogen peroxide in absolute methanol for 20 minutes to block endogenous peroxidase activity. Then 5% BSA was applied for 15 minutes to prevent non‐specific binding. The sections were incubated overnight at 4°C with primary antibodies (human PIM1 [1:100], Ki‐67 [3 μg/mL], cleaved caspase‐3 [1:200] from Santa Cruz Biotechnology, Santa Cruz, CA, USA). After incubation with the secondary antibody, the visualization signal was developed with 3,3′‐diaminobenzidine tetrachloride.

Pulmonary metastasis animal model

Mice were injected with HT29 cells (1 × 106) in 0.1 mL saline into the tail vein, and killed after a 12‐week inoculation period. Hematoxylin–eosin was used to stain consecutive sections of the whole lung. All metastatic lesions in the lungs of the mice were assessed and counted to evaluate the role of hispidulin on the pulmonary metastasis of CRC.

Statistical analysis

The association of PIM1 expression with clinicopathological characteristics was analyzed using Spearman's correlation analysis. Comparisons were undertaken using one‐way ANOVA followed by Dunnett's t‐test. The statistical analysis of all data was carried out using SPSS software (SPSS, Chicago, IL, USA). P < .05 was considered statistically significant.

RESULTS

PIM1 aberrantly upregulated in CRC tissues and associated with clinicopathological features of patients

Expression of PIM1 in CRC and matched non‐cancerous tissues was determined by IHC staining. As shown in Figure 1A, our findings indicated that the level of PIM1 was remarkably higher in CRC tissues relative to matched non‐cancerous tissues. The association between PIM1 levels and the TNM stage of patients was also examined, which showed that PIM1 expression increased along with advanced stages of CRC (Figure 1B). Moreover, patients with lymph node metastasis showed higher PIM1 levels compared with those without lymph node metastasis (Figure 1C). Our results showed that a high expression of PIM1 was positively associated with Ki‐67 expression and inversely correlated with the expression of cleaved caspase‐3, which indicated that PIM1 promoted tumor growth and reduced apoptosis in human CRC (Figure 2). The clinicopathological characteristics of all patients were analyzed to further investigate the role of PIM1 in CRC. The results indicated that PIM1 expression in tumor tissues has significant correlations with the degree of local invasion and lymph node metastasis (Table 1). In contrast, the protein expression of PIM1 was not associated with sex, age, or tumor focus size (Table 1). Collectively, our results suggested the role of PIM1 as oncogenic factor in CRC.

Figure 1

expression in colorectal cancer (CRC) and matched normal tissues. A, Aberrantly higher expression in CRC tissue than matched normal tissue. B, Association between expression and disease stage. C, Correlation between expression and lymph node metastasis. **P < .01

Figure 2

Correlation of expression with Ki‐67 and cleaved caspase‐3 expression in colorectal cancer tissue. A, Positive association between and Ki‐67 expression. B, Inverse association between and cleaved caspase‐3 expression. **P < .01

Table 1

Association between PIM1 expression and clinical pathological characteristics

Parameters	PIM1 expression		P value
	High (26)	Low (37)
Age
≤60	15	16	.3114
>60	11	21
Gender
Male	15	17	.4455
Female	11	20
Tumor size
≤5 cm	17	18	.2095
>5 cm	9	19
Local invasion
T1‐T2	8	25	.0052**
T3‐T4	18	12
Lymph node metastasis
Absence	9	25	.0119*
Presence	17	12

*P < .05, **P < .01.

Knockdown of PIM1 inhibits cell growth, induces apoptosis, and suppresses invasion

To further investigate the role of PIM1 in the cellular activities of CRC, CRC cell lines were transfected with siRNA targeting PIM1. As shown in Figure 3A, the protein expression of PIM1 was reduced to approximately 30% compared to cells transfected with control siRNA or parental cell lines. The results of the CCK‐8 assay indicated that the knockdown of PIM1 exerted marked inhibitory effects on cell growth (Figure 3B). The contribution of apoptosis to the antigrowth effect of PIM1 siRNA was confirmed by flow cytometry and Western blot analysis (Figure 3C,D). Moreover, knockdown of PIM1 also suppressed cell invasion (Figure 3E,F). In general, these results indicated that PIM1 presented as a promising therapeutic target in CRC.

Figure 3

Oncogenic role of in colorectal cancer. A, Knockdown of in HT29 and SW480 cells using siRNA. B–F, Effect of knockdown on cell growth (B), apoptosis (C), activation of caspase‐3 and poly(ADP‐ribose) polymerase (PARP) (D), cell invasion (E), and expression of MMP‐2 and MMP‐9 (F). *P < .05; **P < .01

Hispidulin inhibits cell growth and decreases cell motility in CRC cells

To explore the therapeutic potential of hispidulin in CRC cells, the role of hispidulin on CRC cell viability was assessed. As Figure 4A shows, hispidulin inhibited the growth of the CRC cell lines HT29 and SW480 in a concentration‐ and time‐dependent manner. To investigate the underlying mechanisms involved in the growth inhibition of hispidulin, cell apoptosis was analyzed using flow cytometry following treatment. Following treatment with hispidulin, both HT29 and SW480 cells presented with significant increases in apoptotic cell death in a dose‐dependent manner (Figure 4B). As 2 crucial markers of cell apoptosis, the cleavage of caspase‐3 and PARP were dose‐dependently increased by treatment with hispidulin, which further indicated that hispidulin induced the apoptosis of HT29 and SW480 cells (Figure 4C). Furthermore, hispidulin inhibited the invasion of the CRC cell lines HT29 and SW480 in a concentration‐dependent manner (Figure 4D). The protein expression of MMP‐2 and MMP‐9 was also decreased in a concentration‐dependent manner (Figure 4E).

Figure 4

Anticancer effect of hispidulin against colorectal cancer (CRC). A, HT‐29 and SW480 cells were treated with hispidulin (10 or 20 μmol/L) for 24 and 48 hours, the antigrowth effect of hispidulin on CRC cells was measured by CCK‐8 assay. B, Cell apoptosis was quantified by flow cytometry after CRC cells were treated with different concentrations of hispidulin (10 or 20 μmol/L) for 24 hours. C, Effect of hispidulin on activation of caspase‐3 and poly(ADP‐ribose) polymerase (PARP). D, After cells were incubated with hispidulin for 24 hours, the effect of hispidulin (10 or 20 μmol/L) on cell invasion ability was analyzed using Transwell invasion assay. E, HT‐29 and SW480 cells were treated with hispidulin (10 or 20 μmol/L) for 24 h. Western blot analysis was used to detect the expression of MMP‐2 and MMP‐9 using indicated the antibodies. **P < .01

PIM1 mediates the antitumor activity of hispidulin in CRC cells

The function of JAK2/STAT3 as an upstream signaling regulator of PIM1 expression has also been reported.30, 31 In addition, our previous study showed that hispidulin functioned as an inhibitor of STAT3 signaling.21, 32 Therefore, we hypothesized that hispidulin exerted an antitumor effect by regulating PIM1 through inhibiting JAK2/STAT3. As shown in Figure 5A,B, hispidulin treatment could dose‐dependently repress PIM1 expression at mRNA and protein levels, indicating that hispidulin modulated PIM1 at a transcriptional level. To further investigate the role of PIM1 on the antitumor effect of hispidulin, overexpressing plasmid or siRNA was used to manipulate the level of PIM1. PIM1 expression following transfection with the overexpression vector was increased by approximately 3.2‐fold (Figure S1). The effects of hispidulin were then compared in relation to cell growth, apoptosis, and invasion in CRC cells with PIM1 overexpression or knockdown. As shown in Figure 5C, the growth inhibition of hispidulin on cells was markedly compromised by the ectopic expression of PIM1, but enhanced by PIM1 siRNA. Correspondingly, hispidulin‐induced apoptosis and the upregulation of cleaved caspase‐3 and cleaved PARP were significantly attenuated in cells transfected with the PIM1 vector, but augmented in cells transfected with PIM1 siRNA (Figure 5D,E). Furthermore, PIM1 overexpression abrogated the suppressive effect of hispidulin on cell invasion. In contrast, PIM1 knockdown further reduced the invasion of CRC cells (Figure 5F). Hispidulin‐induced downregulation of MMP‐2 and MMP‐9 was significantly attenuated in cells transfected with the PIM1 vector, yet augmented in cells transfected with PIM1 siRNA (Figure 5G). Collectively, our results verified the first part of our postulation that hispidulin exerted an antitumor effect through modulating PIM1.

Figure 5

Mediating role of in the antitumor effect of hispidulin in colorectal cancer (CRC) cells. A, Modulation of mRNA expression by hispidulin in CRC cells. B, Modulation of PIM1 protein by hispidulin in CRC cells. C–G, Effect of overexpression or knockdown on hispidulin‐induced growth inhibition (C), apoptosis (D), activation of caspase‐3 and poly(ADP‐ribose) polymerase (PARP) (E), suppression of cell invasion (F), and repression of MMP‐2 and MMP‐9 expression (G). **P < .01

Hispidulin modulates PIM1 through ROS/JAK2/STAT3 signaling

Next, we undertook experiments to verify the second part of our postulation that hispidulin modulated PIM1 through the inhibition of JAK2/STAT3. The inhibition of JAK2/STAT3 by hispidulin was confirmed in both cell lines by Western blot analysis (Figure 6A). Moreover, the activation of JAK2/STAT3 by IL‐6 significantly attenuated the repression of PIM1 resulting from hispidulin treatment (Figure 6B). Flavonoid compounds have been found to inhibit JAK2/STAT3 signaling by generating ROS in cancerous cells.33 Therefore, the extent to which hispidulin inhibited JAK2/STAT3 signaling in a ROS‐dependent manner was explored. In line with the findings in breast cancer cells,34 the intracellular level of ROS was increased by hispidulin in a dose‐dependent manner (Figure 6C). N‐acetylcysteine, a well‐established ROS scavenger, was used to examine the role of ROS in modulating PIM1. As shown in Figure 6D, pretreatment with NAC restored the activation of JAK2/STAT3 signaling and the expression of PIM1, suggesting that ROS generation was involved in the inhibition of JAK2/STAT3 and thus the repression of PIM1 by hispidulin. The role of ROS/JAK2/STAT3 signaling in the antitumor effect of hispidulin against CRC cells was further investigated by testing to what extent the activation of JAK2/STAT3 by IL‐6 or ROS scavenger NAC could dampen the effects of hispidulin on cell growth, apoptosis, and invasion. As shown in Figure 7A–C, both IL‐6 and NAC could successfully rescue CRC cells from hispidulin‐induced cell death, apoptosis, and the upregulation of cleaved caspase‐3 and cleaved PARP. In addition, pretreatment with IL‐6 or NAC reversed the inhibition of cell invasion and expression of MMP‐2 and MMP‐9 by hispidulin (Figure 7D,E). Therefore, our results revealed that hispidulin exerted an anti‐tumor effect by targeting PIM1 through ROS and JAK2/STAT3 signaling.

Figure 6

Effect of hispidulin on reactive oxygen species (ROS)/JAK2/signal transducer and activator of transcription 3 (STAT3) signaling in colorectal cancer (CRC) cells. A, Inhibition of JAK2/STAT3 by hispidulin in CRC cells. B, Effect of activation of JAK2/STAT3 by interleukin‐6 (IL‐6) on the modulation of by hispidulin (His). C, Generation of ROS by hispidulin in CRC cells. D, Role of ROS generation by hispidulin in JAK2/STAT3 signaling and modulation. **P < .01. NAC, N‐acetylcysteine

Figure 7

Role of reactive oxygen species (ROS)/JAK2/signal transducer and activator of transcription 3 (STAT3) signaling in the anticancer effect of hispidulin against colorectal cancer. A–E, Effect of JAK2/STAT3 activator interleukin‐6 (IL‐6) and ROS inhibitor N‐acetylcysteine (NAC) on hispidulin‐induced growth inhibition (A), apoptosis (B), activation of caspase‐3 and poly(ADP‐ribose) polymerase (PARP) (C), suppression of cell invasion (D), and repression of MMP‐2 and MMP‐9 expression (E). **P < .01

Hispidulin suppresses CRC tumor growth and pulmonary metastasis

Based on the results of our in vitro studies, a mouse model was used to evaluate the antitumor in vivo effect of hispidulin. As Figure 8A shows, hispidulin at a dosage of 20 mg/kg could exert significant suppression on tumor growth (P < .01 vs control). Tumor weight was dose‐dependently decreased by hispidulin treatment (Figure 8B). In addition, our results showed that hispidulin at both doses did not cause marked change in the body weight of model animals, suggesting the selective toxicity of hispidulin against tumors. Moreover, H&E examination of brain, lung, heart, spleen, kidney, and liver showed that hispidulin at 40 mg/kg did not cause changes in the morphology of these tissue (Figure S2), further supporting that hispidulin did not damage the healthy tissues. In line with the findings of tumor growth, the TUNEL assay showed that treatment with hispidulin was associated with dose‐dependent increases in apoptosis (Figure 8C). Immunohistochemistry results showed that hispidulin treatment resulted in decreased levels of PIM1, p‐STAT3, and p‐JAK2, thus, further supporting our findings in vitro (Figure 8C). Hispidulin has been confirmed to affect lung metastasis. As shown in Figure 8D, the number of metastatic nodules in the lungs of mice was significantly decreased by hispidulin treatment. Therefore, our in vivo studies indicated that hispidulin can significantly inhibit cell growth and metastasis in CRC.

Figure 8

In vivo efficacy of hispidulin in colorectal cancer tumor growth and metastasis. A, Suppression of tumor growth by hispidulin (His) in a xenograft animal model. B, Reduction in tumor weight by hispidulin treatment. C, Repression in expression of provirus integration site for Moloney murine leukemia virus (PIM1), phosphorylated signal transducer and activator of transcription 3 (p‐STAT3), and p‐JAK2 by hispidulin in tumor tissue. D, Decrease in pulmonary metastatic nodules by hispidulin in a xenograft animal model. **P < .01

DISCUSSION

It has been reported by Verbeek et al that PIM1 inhibited the apoptosis induced by myc in mouse models of lymphoma.35 It also has been proven that PIM1 is involved in the development and progression of several solid tumors, including CRC.12 Of interest, PIM1 inhibitors have been investigated for their efficacy in treating malignancies. For instance, AZD1208 is an effective and selective pan‐Pim kinase inhibitor, and has been found to have a good efficacy in preclinical models of acute myeloid leukemia.36 AZD1208 has also been found to treat MYC‐driven prostate cancers in animal models.37 Moreover, the pivotal role of PIM kinases in hematopoietic tumors has promoted the development of various pan‐PIM inhibitors. For example, this involves phase I trials in patients with relapsed and/or refractory multiple myeloma, and in phase Ib/II trials (CLGH447X2103C, NCT02144038, and EudraCT2013‐004959‐21) for patients with relapsed/refractory acute myeloid leukemia who have a high risk of myelodysplastic syndrome and myelofibrosis.38, 39 Given that PIM1 is constitutively active and not regulated by phosphorylation,6 modulating the expression of PIM1 at a transcriptional level provides a novel insight for cancer therapy. In the present study, our findings indicated that PIM1 functioned as an oncogene in CRC, and hispidulin suppressed tumor growth and metastasis of CRC through modulating PIM1 expression.

An aberrant upregulation of PIM1 has been documented in solid tumors and hematological malignancies.8 The role of PIM1 as a prognostic marker has also been examined. For instance, overexpression of the PIM1 protein was associated with the tumorigenesis and lymphatic metastasis of esophageal squamous cell carcinoma.40 In breast cancer, PIM1 expression was higher in triple‐negative tumors than in receptor‐positive tumors, and was associated with poor prognosis in patients with hormone‐ and human epidermal growth factor receptor 2‐negative tumors.41 However, high levels of PIM1 are associated with a good prognosis in prostate cancer,9 pancreatic ductal carcinoma,10 and non‐small‐cell lung cancer.11 In terms of CRC, previous studies have shown a total PIM1 score that reflected the expression of PIM1 in tumors, tumor stroma, and tumor‐adjacent mucosa, and has been found to correlate with disease stage; thus, it can be considered a useful prognostic marker for CRC.12 Interestingly, Peng et al did not observe any correlation between PIM1 expression in tumor tissues and the clinicopathological features of patients.12 On the contrary, our findings revealed a positive association between PIM1 expression and advanced disease stage and lymph node metastasis. These conflicting results might reflect the difference in scoring systems for PIM1 expression and (or) sample pools.

The involvement of PIM1 in several cancerous cellular activities has been reported. For example, PIM1 can help cancer cells escape from apoptotic signaling by modulating Bcl‐xL and Bcl‐2.42 Moreover, recent studies have shown that the inhibition of PIM1 expression by miRNAs resulted in a reduction in the invasiveness of tumor cells, which provides indirect evidence for the role of PIM1 in metastatic behavior.43, 44, 45 Consistent with these studies, our findings indicated that knockdown of PIM1 using siRNA led to increased apoptotic cell populations and decreased cell motility. Our findings also indicated that repressing PIM1 expression by hispidulin led to the suppression of cell growth, induction of apoptosis, and inhibition of migration and invasion, all of which provided strong evidence for the role of PIM1 as an oncogenic factor in CRC. In addition to cell apoptosis and invasive behavior, PIM1 has been found to regulate the energy metabolic process of glioma and hepatocellular carcinoma cells.28, 44 Santio et al have also reported that PIM1 promotes angiogenesis and lymphangiogenesis in orthotopic xenografts of pancreatic cancer.46 Gu et al have revealed that suppressing PIM1 reduced hypoxia‐induced autophagy and enhanced the radiosensitivity of prostate cancer cells.47 In addition, PIM1 was observed to induce chemoresistance in cancer cells by interaction with phosphorylation of Etk, fms‐like tyrosine kinase 3, breast cancer‐resistant protein, and P‑glycoprotein.48 Given that hispidulin can modulate the expression of PIM1, hispidulin could affect angiogenesis, autophagy, and glycolysis, all of which could contribute to its antitumor activity.

The regulation of PIM1 occurs through transcriptional regulation, translational regulation, and proteasomal degradation. In human malignancies, multiple transcription factors and pathways have been found to regulate the gene expression of PIM1.49, 50, 51 Following activation by cytokines, growth factors, and hormones, the transcription factors STAT3 and STAT5 have been observed to elevate the gene expression of PIM1 by directly binding to promoter regions of PIM1.52 Hypoxic microenvironments have been identified as another mechanism that promotes PIM1 expression, resulting in hypoxia‐induced tumorigenesis and chemoresistance.51, 53 Various recent studies have indicated that miRNAs could regulate the expression of PIM1 through targeting the mRNA of PIM1.29 Moreover, inhibiting NF‐κB activation in diffuse large B‐cell lymphoma attenuated CD40‐ellicited increases in PIM1 expression, suggesting the regulatory function of NF‐κB signaling on PIM1.54 In our studies, we found that hispidulin modulated the expression of PIM1 by suppressing JAK2/STAT3 signaling through generating ROS. However, a previous study by Clavin indicated that hispidulin exerted anti‐inflammatory activities by modulating NF‐κB.55 Therefore, more research needs to be undertaken to examine to what extent the suppression of NF‐κB activation contributes to the modulation of PIM1 by hispidulin.

In summary, our findings indicated that PIM1 was upregulated in CRC tumor tissue and associated with advanced disease stage and lymph node metastasis. Knockdown of PIM1 by siRNA or the downregulation of PIM1 by a naturally occurring compound, hispidulin, has been shown to exert an antitumor effect in vitro and in vivo in CRC. Furthermore, our findings also revealed that hispidulin repressed the expression of PIM1 by suppressing JAK2/STAT3 signaling by generating ROS.

CONFLICT OF INTEREST

The authors have no conflict of interest.

Click here for additional data file.

Click here for additional data file.