CXCR4 antagonist AMD3100 enhances therapeutic efficacy of transcatheter arterial chemoembolization in rats with hepatocellular carcinoma

Abstract

This study aims to discover the therapeutic effect of chemokine (CXC motif) receptor 4 (CXCR4) antagonist AMD3100 combined with transcatheter arterial chemoembolization (TACE) in a rat model with hepatocellular carcinoma (HCC). An orthotopic model of HCC was established and treated with TACE (doxorubicin‐lipiodol emulsion) with or without AMD3100. The tumor volume was measured by magnetic resonance imaging (MRI). Histopathological changes were detected by hematoxylin–eosin (HE) staining. HCC cell apoptosis was assessed by terminal deoxyribonucleotidyl transferase (TdT)‐mediated biotin‐16‐dUTP nick‐end labeling (TUNEL) staining. Immunohistochemistry was used to detect the expression of CD34, hypoxia‐inducible factor 1α (HIF‐1α), vascular endothelial growth factor (VEGF), and Ki67. Gene and protein expressions were quantified by quantitative reverse‐transcription polymerase chain reaction (qRT‐PCR) and western blotting, respectively. Both TACE and AMD3100 reduced the tumor volume in orthotopic rat model of HCC with the decreased CXCR4 expression in tumor tissues, and the combination had better effect. However, TACE increased the microvessel density (MVD) in HCC tissues of rats, while AMD3100 treatment reduced MVD in HCC tissues. AMD3100 reduced the TACE induced MVD in HCC tissues with the reduction of HIF‐1α and VEGF expression. Either AMD3100 or TACE could promote HCC cell apoptosis accompanying by decreased cell proliferation, and their combined use had better therapeutic effects. CXCR4 antagonist AMD3100 enhance therapeutic efficacy of TACE in rats with HCC via promoting the HCC cell apoptosis, reducing cell proliferation, and inhibiting MVD, thus reducing tumor volume.

INTRODUCTION

Liver cancer is the fifth most common malignancies and the second leading cause of cancer-related deaths worldwide, while hepatocellular carcinoma (HCC) accounts for approximately 90% of all liver cancer cases.¹ Although preventing the recurrence of tumors is very important to ensure the therapeutic effect in the cancer treatment, there are still some liver cancer patients who would lose the opportunity of radical resection due to their poor liver function or special tumor location.² So far, the interventional therapy is recognized as the first-line treatment for unresectable liver cancer, playing significant effects on inhibiting tumor growth and improving the quality of patients' life.³ Transcatheter arterial chemoembolization (TACE), on the basis of transcatheter arterial embolization (TAE), was initially developed in the 1980s in Japan, which would deliver chemotherapy drugs (mainly doxorubicin or cisplatin) to cancer tissues, thereby blocking the blood supply of liver cancer tissues and kill the cancer cells at the same time.^4,5 Based on a randomized controlled trial and meta-analysis, TACE has become the best treatment option for intermediate-stage HCC patients.⁶ However, the long-term efficacy of TACE is still not satisfactory mainly due to the recurrence and metastasis of HCC.⁷ It is worthy to mention that angiogenesis is essential for tumor growth and metastatic dissemination, whereas the antiangiogenics can directly inhibit the growth of tumor blood vessels and the dissemination of metastatic tumors, demonstrating the efficacy in HCC patients.⁸ Currently, it is generally acknowledged that the combination of TACE and antiangiogenics can be an effective therapy to improve the outcome of HCC patients.⁹

Chemokines (8–14 kDa) are a group of small-molecule proteins that can activate chemokine receptors, regulate leukocyte migration, and participate in inflammation responses.¹⁰ In addition to the effects on inflammatory diseases, chemokines are also involved in the malignant tumor progression, including tumor growth and metastasis caused by tumor cell migration, invasion, and angiogenesis.¹¹ Accumulating evidences have shown the immunohistochemical staining of chemokine (CXC motif) receptor 4 (CXCR4) in HCC tissues, but not in normal hepatic tissues.^12,13 High CXCR4 expression in the liver has been considered as a biomarker of poor prognosis in HCC patients.¹⁴ By contrast, AMD3100 is a CXCR4 antagonist¹⁵ that has been found to be able to inhibit metastasis and development of multiple solid tumors, like breast cancer and glioblastoma, as well as pulmonary metastasis of melanoma.¹⁶ Additionally, AMD3100 was shown to inhibit the tube formation of endothelial cells in vitro¹⁷ and suppress tumor angiogenesis in vivo.¹⁸ Thus, we speculated that AMD3100 combined with TACE can treat transplanted HCC tumor and inhibit tumor angiogenesis. In this study, we established the orthotopic rat model of HCC and treated HCC rats with TACE and AMD3100 alone or in combination prior to a series of tests.

MATERIALS AND METHODS Ethical statement

Animal experiments were approved by Ethics Committee of Laboratory Animals of our hospital and conducted in accordance with the Guide for the Care and Use of Laboratory Animals: Eighth Edition.¹⁹

Cell culture

N1-S1 HCC cells (CRL-1604, American type culture collection, ATCC,Rockville, MD, USA) were established from a hepatoma by feeding a male rat with 4-dimethylaminoazobenzene, and then cultured in Iscove's Modified Dulbecco's Medium (30-2005, ATCC, Rockville, MD, USA) with 10% fetal-buffered saline (FBS) (culture conditions: 95% air, 5% CO₂, 37°C).

Establishment of orthotopic rat models

N1-S1 cells (1 × 10⁶ cells) were suspended in Dulbecco's modified Eagle's medium (DMEM) and implanted into the left hepatic lobe of 48 male Sprague–Dawley (SD) rats (12 weeks old) by mini-laparotomy (day 1).²⁰ On day 7 when tumor grew to 5 mm in diameter, rats were randomized into four groups (12 in each): Sham group, TACE group, AMD3100 group, and TACE + AMD3100 group.

TACE treatment

On day 7, rats were anesthetized, prepared, and disinfected for laparotomy. The skin was open with an ophthalmic eyelid spreader and the liver lobe was pulled gently with a sterile cotton swab to fully expose the visual field of hepatic porta of rats. Ophthalmic curved forceps were used to separate common hepatic artery and gastroduodenal artery. The suture 9-0 was used to ligate the distal end of gastroduodenal artery. The proximal end of gastroduodenal artery and the common hepatic artery were pulled up in turn to guarantee that the free gastroduodenal artery segment was filling with blood and to prevent the perfusion fluid from flowing back into adjacent tissues. The self-made microcatheter was retrogradely inserted into the proper hepatic artery along the proximal end of gastroduodenal artery. Rats in TACE group and TACE + AMD3100 group were injected with 0.1 ml doxorubicin-lipiodol emulsion (DLE) (equivalent to 1 mg/kg doxorubicin) via the catheter. Of note, the recommended dose of doxorubicin and lipiodol was ≤2 mg/kg for human adults.²¹ Rats in Sham group and AMD3100 group received laparotomy but not treated with TACE. The principle of aseptic operation was strictly observed throughout the operation. Lateral position was taken after operation, fluid was injected subcutaneously with Ringer's lactate solution, and penicillin was used for anti-infection treatment.

AMD3100 treatment

The CXCR4 antagonist AMD3100 was provided by Selleck Chemicals (S9725, Houston, TX, USA). On day 10, rats in AMD3100 group and TACE + AMD3100 group were injected with 5 mg/kg/day AMD3100 by intraperitoneal injection for seven consecutive days (from day 10 to day 16).^22,23 Rats in the other two groups were injected with the equal volume of normal saline. At 24 h after the last administration (on day 17), half of the rats in each group were killed for subsequent examinations. The flowchart of the experiment is shown in Figure 1.

Tumor volume measurement

Magnetic resonance imaging (MRI) was used to measure the width (a) and length (b) of tumors on the day 7, 10, and 17. The tumor volume (V) = (a² × b)/2. For comparison, the tumor volume was normalized by the initial volume: V/V_day7. To perform MRI, the 7-T MRI unit (BioSpec) was used with the following parameters: 2000 ms (repetition time); 30 ms (echo time); 1 mm section thickness; visual field, 71 × 85 mm; 216 × 256 matrix; and respiratory triggering with an MRI-compatible small animal gating system (model 1025; SA Instruments, Stony Brook, NY, USA).

Detection of liver function enzymes

Serum was collected for analyzing the liver function enzymes using a Cobas Integra 400 plus analyzer (Roche, Switzerland), including alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (AP).

Hematoxylin–eosin staining

HCC tissues of rats were embedded in paraffin and sliced as tissue sections, which were de-paraffinized in xylene for 5 min × 2 times, dehydrated with gradient alcohol, and stained in hematoxylin dye for 5–15 min. Excess dye in the slides was washed away by distilled water, followed by differentiation with 0.5% hydrochloric ethanol (prepared with 70% ethanol) for 2–5 s. After rinsing with running water for 15–30 min, the sections were stained with Eosin dye, dehydrated with gradient alcohol, and hyalinized with xylene. The excess xylene around the slice was wiped off. Then, the resin was added and the cover glass was used to seal the tissue slices.

Quantitative reverse-transcription polymerase chain reaction

The total RNA in HCC tissues was extracted with TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) and quantified for the concentration via NanoDrop 1000 (Thermo Scientific, Massachusetts, USA). Then, RNA was reversely transcribed into cDNA with high-capacity cDNA Reverse Transcription Kit (Applied Biosystems,Foster City, CA, USA). The appropriate amount of cDNA was used as template for PCR reaction. The software Primer 5.0 was used to design primer sequences (Table 1). PowerUp SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) was used to do quantitative reverse-transcription polymerase chain reaction (qRT-PCR) in the StepOnePlus™ Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The mRNA level was quantified via 2^−ΔΔCT analysis using GAPDH mRNA as reference gene.

TABLE 1 Primers for quantitative reverse-transcription polymerase chain reaction (qRT-PCR)

Gene	Primers (5′–3′)
CXCR4	Forward: CGCAAGGCCCTCAAGACTAC
	Reverse: CGAGGAAGGCGTAGAGGATG
VEGF	Forward: ATCTTCAAGCCATCCTGTGTGC
	Reverse: CAAGGCCCACAGGGATTTTC
HIF-α	Forward: CCGAATTGATGGGATATGAGCCAG
	Reverse: TTGGCAAGCATCCTGTACTGTCCT
GAPDH	Forward: ATCACCATCTTCCAGGAG
	Reverse: ATCGACTGTGGTCATGAG

Western blotting

Total proteins were extracted from HCC tissues, quantified for concentration with Bradford method, and adjusted to the consistent concentration for all samples. Equal volume of proteins (30–50 μg) was separated via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membrane. The membrane was blocked in defatted milk powder at room temperature, washed with phosphate-buffered saline with Tween 20 (PBST) buffer and hybridized for 1 h with anti-CXCR4 antibody (ab124824, Abcam, Cambridge, MA, USA) at 1/1000 dilution, anti-GAPDH antibody-Loading Control (ab8245, Abcam, Cambridge, MA, USA) at 2.5 μg/ml at room temperature. The membrane was washed with PBST (5 times × 3 min), followed by incubation with Goat Anti-Rabbit IgG (ab97051, Abcam, Cambridge, MA, USA) at 1/20,000 dilution. Enhanced chemiluminescence (ECL) was used for development and visualization. The relative expression of target proteins was calculated as the gray value ratio of target protein to GAPDH.

Immunohistochemical staining and terminal deoxyribonucleotidyl transferase-mediated biotin-16-dUTP nick-end labeling (TUNEL) staining

Tissue sections were deparaffinized in xylene, dehydrated with gradient ethanol, transferred to citric acid buffer, and boiled for 3–5 min in microwave oven for antigen retrieval. Next, tissue sections were incubated in 3% hydrogen peroxide for 15 min in avoidance of light and rinsed with distilled water. Primary antibody was added for overnight reaction at 4°C. Tissue sections were rinsed with PBST, reacted with HRP-labeled secondary antibody for 60 min at room temperature, rinsed again with PBST (5 min × 3 times), and incubated with 50–100 μl diaminobenzidine (DAB) solution. Then, tissue sections were rinsed with tap water, stained in hematoxylin for 5 min, rinsed with tap water, differentiated in 0.1% hydrochloric ethanol for 2–5 s, and rinsed with tap water for 10–15 min for re-bluing. After the dehydration with gradient ethanol, the sections were hyalinized with xylene, and embedded with resin. The analysis was used to quantify positive nuclei (DAB-brown) and negative nuclei (counter-stained blue with hematoxylin) for automated quantification of Ki67.²⁴ By this way, ~10,000 to 200,000 cells can be quantified for each tumor. Apoptotic cells were identified by the terminal deoxyribonucleotidyl transferase (TdT)-mediated biotin-16-dUTP nick-end labeling (TUNEL) assay using an in-situ ApopTag kit (Chemicon International, Temecula, CA, USA).

Statistical methods

All statistical data were analyzed with SPSS 21.0 (SPSS, Inc., Chicago, IL). Measurement data were expressed as mean ± standard deviation (SD). Comparison among multiple groups was analyzed with One-Way ANOVA, while inter-group difference was analyzed by Tukey' HSD test. p <0.05 indicates the statistical significance of differences.

RESULTS AMD3100 combined with TACE inhibits the tumor volume of HCC orthotopic rat models

As shown in Figure 2A,B, there was no significant difference in tumor volume among rat groups before TACE treatment (Day 7) (all p >0.05). On day 10, rat in the TACE group and TACE + AMD3100 group showed smaller in tumor volume than the Sham group and the AMD3100 group (all p <0.05). On day 17, Sham group had the largest tumor volume while TACE + AMD3100 group had the smallest tumor volume, both of which were significantly different from TACE group and TACE + AMD3100 group (all p <0.05), although no obvious difference was observed between TACE group and AMD3100 group (p >0.05).

AMD3100 combined with TACE affects pathological tissues of orthotopic HCC tumor

Hematoxylin–eosin (HE) staining of HCC tissues (Figure 2C) showed that Sham group had the largest number of tumor cells. By contrast, TACE group and AMD3100 group had more necrotic HCC cells, with the cytoplasm of the surviving HCC cells stained red, the nucleus heavily stained and moderate mitotic rate, while TACE + AMD3100 group demonstrated more thorough necrosis of tumor cells with few surviving cells. Besides, all these treatments did not appear to cause significant toxicity in these animals with similar levels of AST, ALT, and AP in serum on day 17 (Figure 2D).

CXCR4 expression in HCC tissues of rats in each group

CXCR4 gene expression in HCC tissues of rats was detected by qRT-PCR (Figure 3A). Compared with Sham group, rats in both AMD3100 group and TACE group had decreased CXCR4 expression in HCC tissues (all p <0.05). Besides, rats in AMD3100 combined with TACE group further decreased the CXCR4 expression compared to those groups treated alone (both p <0.05). By detecting the protein expression of CXCR4 using western blotting, we observed the same trend with that of CXCR4 gene expression (Figure 3B).

AMD3100 combined with TACE regulates microvessel density in tumor tissues from HCC orthotopic rats

The expression of CD34, as a marker of microvessel density (MVD), in tumor tissues was determined by immunohistochemical assay (Figure 4). As a result, TACE led to the obvious increase of MVD in tumor tissues of HCC rats, while AMD3100 treatment reduced MVD in HCC tissues (all p <0.05). Moreover, TACE combined with AMD3100 could further reduce MVD in tumor tissues, which was significantly different from AMD3100 group (p <0.05).

AMD3100 combined with TACE regulates the expression of HIF-1α and VEGF in HCC tissues

The gene expression of HIF-1α and VEGF in HCC tissues were quantified by qRT-PCR (Figure 5A). AMD3100 inhibited the expression of HIF-1α and VEGF in HCC tissues, while TACE led to the increased expression of HIF-1α and VEGF (all p <0.05). Compared with other three groups, AMD3100 combined with TACE significantly mitigated the expression of HIF-1α and VEGF in HCC tissues (all p <0.05). HCC tissues were embedded in paraffin and stained by immunohistochemistry to observe the expression of HIF-1α and VEGF proteins to assess the therapeutic effects (Figure 5B). TACE group was increased dramatically in the expression of HIF-1α and VEGF proteins, whereas AMD3100 group was reduced in the expression of HIF-1α and VEGF proteins. Furthermore, AMD3100 + TACE group showed a few expressions of HIF-1α and VEGF proteins only in residual HCC cells, which was significantly lower than AMD3100 group.

AMD3100 combined with TACE affects the apoptosis and proliferation of HCC tissues

As illustrated by TUNEL staining, either AMD3100 or TACE could promote the HCC cell apoptosis of rats, but their combination had better therapeutic effects. In other words, AMD3100 + TACE group was significantly higher in TUNEL-positive cell number than either AMD3100 group or TACE group (all p <0.05, Figure 6A,B). The HCC cell proliferation was compared among groups by comparing the expression of Ki67 in HCC tissues. Consequently, Sham group was significantly higher in the number of proliferative cells than other three treatment groups (all p <0.05) and AMD3100 + TACE group had the least number of proliferative cells (p <0.05, Figure 6C,D). AMD3100 group and TACE group showed no obvious difference in the apoptosis and proliferation of HCC cells (p >0.05).

DISCUSSION

The ideal treatment outcome of TACE is that embolization materials like alcohol-lipiodol can thoroughly embolize tumor blood-supply vessels to induce total necrosis of tumor cells because of “hunger.”²⁵ In the current study, HCC rats treated with TACE decreased significantly in tumor volume with many necrotic HCC cells. AMD3100, also known as plerixafor, is a CXCR4 antagonist approved by U.S. Food and Drug Administration (FDA) for autologous transplantation in patients with Non-Hodgkin's lymphoma or multiple myeloma.²⁶ Meanwhile, AMD3100 demonstrates huge potential for the treatment of a variety of other diseases, such as leukemia and solid tumors.^27,28 CXCR4 antagonist AMD3100 could prevent the polarization to an immunosuppressive microenvironment after sorafenib treatment, thereby inhibiting HCC tumor growth with reduced lung metastasis and improved survival.²⁹ Besides, AMD3100 can block CXCR4/SDF1α to reduce the infiltration of tumor-associated macrophages, and then enhance anti-angiogenesis and slow down tumor progression, thus improving the overall survival of the orthotopic HCC model.³⁰ In this study, AMD3100 significantly inhibited HCC tumor growth, accompanied by obvious increase of apoptotic cells and the inhibition of Ki-67 expression (a cellular proliferation marker). Similarly, AMD3100 was also previously found to be able to inhibit Ki-67 expression in various types of tumors,^31,32 highlighting its anti-tumor effects.

Based on the publication data, CXCR4 was involved in tumor angiogenesis, and its over-expression in cancer cells could enhance the tumor angiogenesis.³³ In another in vitro study, human umbilical vein endothelial cells (HUVECs) were pretreated with AMD3100 and cocultured with HCC cells, and the result showed AMD3100 can selectively inhibit the migration of HUVECs induced by SOX4 over-expression,¹⁷ suggesting the potential anti-angiogenesis effect of AMD3100. Also, an in vivo study found that AMD3100 could induce the apoptosis of a metastatic tumor derived from a cervical metastatic lesion of human adenoid cystic carcinoma (ACC) and inhibit tumor-induced angiogenesis and proliferation of ACCIM.³⁴ Of note, AMD3100 could significantly reduce the MVD in the lung cancer xenograft model mice.³⁵ In this study, AMD3100 treatment exhibited the reduced MVD of HCC tissues, accompanied by the decreased expression of HIF-1α and VEGF, which supported the anti-angiogenesis function of AMD3100 in HCC.

In the clinical practice, most HCC patients cannot reach the ideal state of complete necrosis of tumor cells after TACE treatment, and to complicate the matter, the aggregation and stimulation of lipiodol and chemotherapy drugs usually induce local hypoxia and ischemia of tumor tissues.³⁶ Angiogenesis refers to the process of forming new blood vessels from the budding in existing vascular bed, involving in the growth, invasion, and metastasis of malignant tumors.³⁷ At present, angiogenesis was often evaluated by MVD via anti-CD34 immunohistochemical labeling.³⁷ Using immunohistochemical staining, we observed the increased CD34 expression in xenograft HCC tissues after TACE treatment, indicating the increase of MVD, which was consistent with a previous study.²⁰ As for HCC, it is a strongly angiogenesis-dependent tumor and the high HIF-1α and VEGF expression in cancer cells is a critical factor to maintain tumor growth, promote neovascularization, and induce invasion and metastasis.³⁸ In terms of VEGF, it is a strong stimulator of endothelial cell proliferation and the only factor in HCC tumor tissue to stimulate the division and proliferation of tumor vascular endothelial cells, which could promote angiogenesis and participate in the formation of tumor blood vessels.³⁹ More Importantly, HIF-1α up-regulation induced by ischemia and hypoxia of liver cancer tissues after TACE would also cause the up-regulation of VEGF expression,³⁶ which was in agreement with our findings. After the combined therapy of TACE with AMD3100 in our experiments, the expression of CD34, HIF-1α, and VEGF was decreased, and even lower than those groups treated by either TACE or AMD3100 alone, further indicating that AMD3100 could inhibit the HCC progression by inhibiting the formation of microvessel.

To sum up, TACE treatment can reduce CXCR4 expression, but increase the MVD in tumor tissues of HCC rats. However, CXCR4 antagonist AMD3100 combined with TACE could not only reduce tumor volume of HCC rats, but also promote HCC cell apoptosis, with the inhibited HCC cell proliferation by inhibiting tumor angiogenesis.

CONFLICT OF INTEREST

The authors declare no conflict of interest.