Insulin-like growth factor-binding protein 7 (IGFBP7) functions in several cellular processes including proliferation, senescence and apoptosis. This study analyzed IGFBP7 function in hepatocellular carcinoma (HCC) cells by gene manipulation and investigated the prognostic significance of IGFBP7 expression in clinical HCC samples. In this study, we investigated changes in malignant potential such as cell growth and invasiveness in an HCC cell line, PLC/PRF/5, after transfection with shRNA against IGFBP7. The extent of apoptosis and cell cycle progression were examined after the transfection. The correlation between immunohistochemically determined IGFBP7 expression and long-term postoperative prognosis after curative resection was also investigated in clinical HCC specimens obtained from 104 patients. PLC/PRF/5 cells transfected with shRNA against IGFBP7 showed significantly more rapid growth and stronger invasiveness than control cells. Annexin V assays showed that the IGFBP7-depleted cells were significantly more resistant to apoptosis than the control cells, and showed decreased expression of cleaved caspase-3 and PARP. Cell cycle progression was more rapid in the IGFBP7-suppressed cells. In clinical HCC specimens, IGFBP7 expression was judged as positive in 67 patients (64.4%) and negative in the remaining 37 patients (35.6%). The IGFBP7 downregulation correlated significantly with poor postoperative prognosis, and IGFBP7 status was identified as an independent significant prognostic factor. Our results indicated that IGFBP7 expression correlated significantly with the malignant potential in HCC cells, suggesting that the expression could be a useful prognostic marker for HCC.
Hepatocellular carcinoma (HCC) is a common malignancy worldwide, but especially in Japan and other East Asian countries.1, 2 Although surgery plays a major role in the treatment of HCC, less than 30% of patients with HCC are surgical candidates owing to limiting factors such as severe impairment of reserve hepatic function, bilobar tumor distribution and extrahepatic metastasis. Additionally, no effective chemotherapy regimens have been established for treating HCC.3 Thus, no effective therapy can be offered in many cases of HCC. Such dismal prognosis is not always predicted by conventional prognostic indicators such as vascular invasion, tumor multiplicity and tumor size.4–6 New indicators are thus clearly needed.
Insulin-like growth factor binding protein 7 (IGFBP7), which is also known as IGFBP-rP1 and MAC25, has been implicated in several cellular processes such as proliferation, senescence and apoptosis. IGFBP7 also shows tumor suppressive activity through the induction of apoptosis and it is downregulated in some cancers.7–13 In addition, several studies found a significant association between IGFBP7 and not only apoptosis, but also prognosis, in some kinds of cancers including colorectal and breast cancer.14, 15 However, the functional significance of IGFBP7 in HCC remains unclear.
This study analyzed the function of IGFBP7 in HCC cells in gene manipulation experiments, and investigated the prognostic significance of IGFBP7 expression in clinical HCC samples by immunohistochemical analysis of resected specimens.
Material and Methods
HCC cell lines and clinical tissue specimens
Four human HCC cell lines, PLC/PRF/5, HuH7, HLE and HepG2 were obtained from the Japan Cancer Research Resources Bank (Tokyo, Japan). These cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 mg/ml streptomycin at 37°C in a humidified incubator with 5% CO2 in air.
Surgical specimens were obtained from 104 patients with HCC who underwent curative hepatic resection in the Osaka University Hospital from 2000 to 2007 after informed consent in accordance with the institutional ethical guidelines of Osaka University. Curative resection was defined as complete removal of all macroscopically evident tumors. Patients who underwent transarterial chemoembolization preoperatively were excluded from this study. After hepatic resection, the patients were followed up at regular intervals of 3–4 months with physical examination, assaying of tumor markers including alpha-fetoprotein (AFP) and protein induced by vitamin K absence or antagonists-II (PIVKA-II), liver biochemistry testing, abdominal ultrasonography and abdominal computed tomography. The median duration of clinical follow-up after the initial hepatectomy was 3.5 ± 2.3 years.
Drugs and reagents
A polyclonal goat anti-human IGFBP7 antibody and polyclonal rabbit anti-human IGFBP7 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was used for immunohistochemistry and western blot analysis, respectively. Antibodies to caspase-3, cleaved caspase-3, PARP, cleaved PARP, ERK, phoshorylated ERK (pERK), cyclin D1 and p27 were purchased from Cell Signaling Technology (Beverly, MA), antibodies to cyclin E and p21 were purchased from Santa Cruz Biotechnology, and an antibody to actin was purchased from Sigma-Aldrich Co. (Louis, MO).
Plasmids and transfection
Plasmid coding for short hairpin RNA (shRNA) against IGFBP7 and IGFBP7 expression plasmid were purchased from OriGene Technologies (Rockville, MD) and used to transfect HCC cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the instructions provided by the manufacturer. After transfection of the shRNA plasmid and the IGFBP7 expression plasmid into the HCC cells for 24 hr, stable transfectants were selected and maintained in 1.0 μg/ml of puromycin (Sigma-Aldrich, St. Louis, MO) and 600 μg/ml of G418 (Gibco-BRL, Grand Island, NY), respectively. The control vector plasmid expressing non-effective shRNA was similarly introduced into cells to establish negative control cells for the shRNA plasmid experiments. Empty vector plasmid was also similarly used to establish negative control cells for the IGFBP7 expression plasmid for the IGFBP7 expression plasmid experiments.
Cell proliferation assay
Cells were uniformly seeded (4 × 104/well for PLC/PRF/5 and 2 × 104/well for HuH7) in triplicates into 24-well dishes (Day 0). Cells were counted using a CellTac kit (Nihon Koden, Tokyo, Japan) on Days 1–5.
Total RNA isolated from cells was prepared using TRIzol reagent (Invitrogen), and reverse transcription was performed with SuperScript II (Invitrogen) based on the protocols supplied by the manufacturer. Real-time quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was performed using the Light Cycler (Roche Diagnostics, Mannheim, Germany), and the amount of target gene expression was calculated. The expression of the target gene was normalized relative to the expression of porphobilinogen deaminase (PBGD), which was used as an internal control. The designed PCR primers were as follows; IGFBP7 forward primer; 5′-CTGGGTGCTGGTATCTCCTC-3′; IGFBP7 reverse primer; 5′-TATAGCTCGGCACCTTCACC-3′; SMARCB1 forward primer; 5′-TCTGGATTTGAACCCGCTGA-3′; SMARCB1 reverse primer; 5′-TGCTGTATGCGATGGTGGTG-3′; BNIP3L forward primer; 5′-CGGACTCGGCTTGTTGTGTT-3′; BNIP3L reverse primer; 5′-ATGGGTAGCTCCACCCA GGA-3′; PBGD forward primer; 5′-TGTCTGGTAACGGC AATGCGGCTGCAAC-3′; PBGD reverse primer; 5′-TCAA TGTTGCCACCACACTGTCCGTCT-3′
Western blot analysis
Cells grown to semiconfluence were washed and collected with a rubber scraper. After centrifugation, the cell pellets were resuspended, and the extracts were centrifuged and the supernatant fraction was collected. Western blot analysis was carried out as described previously.16, 17 The expression of the target protein was evaluated by comparison to the expression of actin.
Annexin V assay
The binding of annexin V was used here as a sensitive assessment of apoptosis, as described previously.17, 18 Cells were stained by Annexin V-APC and propidium iodide (PI) (BD Biosciences, Franklin Lakes, NJ), and analyzed on a FACS Aria (BD Biosciences).
The invasion assay was performed using transwell culture chambers (BD Biosciences) according to the instructions provided by the manufacturer. The upper chamber was loaded with cell suspension and the lower chamber was loaded with 10% FBS. After incubation (48 hr for PLC/PRF/5 and 24 hr for HuH7), cells that had invaded the undersurface of the membrane were counted under a microscope. Four microscopic fields were randomly selected for cell counting.
Cell cycle analysis
Cell cycle analysis was performed based on flow cytometric analysis, as described previously.19 Briefly, cells were washed and fixed. PI and RNase (Sigma-Aldrich) were then added, and data were acquired on the FACS Calibur (BD Biosciences). The cell cycle analysis was carried out using ModFIT software (BD Biosciences).
Resected tissue samples were fixed in 10% buffered formalin and finally embedded in paraffin. Immunohistochemical staining for IGFBP7 in the same samples was performed as described previously.17, 20 Briefly, after deparaffinization and blocking, the sections were incubated overnight at 4°C with polyclonal goat anti-human IGFBP7 antibody, and then counterstained with Mayer's hematoxylin. IGFBP7 expression was defined as the presence of specific staining in the cytoplasm of cancer cells. IGFBP7 expression was evaluated as positive or negative, as previously prescribed.17
Data are expressed as mean ± SD. Differences between groups were assessed using the χ2-test, and continuous variables were compared using Student's t-test. Survival rates were calculated according to the Kaplan-Meier method and compared using the log-rank test. Statistical analysis was performed using StatView (version 5.0; SAS Institute, Cary, NC). A p value < 0.05 was considered statistically significant.
95% CI: 95% confidence interval; AFP: alpha-fetoprotein; Anti-HCV Ab: anti-hepatic C virus antibody; DFS: disease-free survival; HBs-Ag: hepatitis B surface antigen; HCC: hepatocellular carcinoma; IGFBP7: insulin-like growth factor binding protein 7; OR: odds ratio; OS: overall survival; PBGD: porphobilinogen deaminase; pERK: phoshorylated ERK; PI: propidium iodide; PIVKA-II: protein induced by vitamin K absence or antagonists-II; qRT-PCR: quantitative reverse transcription-polymerase chain reaction; shRNA: short hairpin RNA
In vitro studies
IGFBP7 downregulation promotes proliferation and invasive activity
First, IGFBP7 expression was examined by qRT-PCR in 4 HCC cell lines, PLC/PRF/5, HuH7, HLE and HepG2. The IGFBP7 expression levels in PLC/PRF/5 and HuH7 were the highest and lowest of the 4 cell lines, respectively. A plasmid coding for shRNA against IGFBP7 was then transfected into PLC/PRF/5, whose IGFBP7 expression level was the highest in the 4 cell lines. The IGFBP7 expression was suppressed by the transfection, as confirmed by qRT-PCR and western blot analysis (Fig. 1a). The proliferation assay showed significantly more rapid growth in the IGFBP7-suppressed cells compared to control cells (Fig. 1b). In addition, the invasive ability of the IGFBP7-suppressed cells was significantly greater than that of the control cells (Fig. 1c). Based on previous studies that IGFBP7 downregulation promotes cell proliferation through ERK signaling, we analyzed the levels of total ERK and pERK in our cells.10, 21 pERK expression was significantly increased in the IGFBP7-supperessed cells, while total ERK expression was not changed, which coincided with previous reports (Fig. 1d). On the other hand, as we previously reported, there were no significant differences in the expression of total Akt or phoshorylated Akt between the IGFBP7-supperessed cells and the control cells.17
Downregulation of IGFBP7 attenuates apoptosis
The extent of apoptosis of these cells was examined by the Annexin V assay. The percentages of early apoptotic cells and late apoptotic cells defined by Annexin V-positive/PI-negative cells and Annexin V-positive/PI-positive cells respectively were significantly lower in the IGFBP7-suppressed cells than those in the control cells (Fig. 2a).22 This significant difference of the extent of apoptosis was also confirmed under the condition where apoptosis is induced by some agents, which was reported previously.17 Next, the expression of proteins related to apoptosis was examined. The result showed that cleaved caspase-3 and cleaved PARP are significantly decreased in the IGFBP7-suppressed cells, while total caspase-3 and PARP expressions were not changed (Fig. 2b). In addition, since apoptosis induced by IGFBP7 was reported to occur via SMARCB1 and BNIP3L upregulation, we also evaluated the expression levels of SMARCB1 and BNIP3L by qRT-PCR.10 The results showed that SMARCB1 and BNIP3L expressions were significantly decreased in the IGFBP7-supressed cells compared with the control cells (Fig. 2c).
Downregulation of IGFBP7 promotes cell cycle progression
The influence of IGFBP7 on cell cycle was examined by flow cytometric analysis. Prior to the examination, cells were synchronized in the G0/G1 phase by serum starvation for 72 hr, and then put back in the regular medium with 10% fetal bovine serum. Dynamic changes in percentage between G0/G1 phase and S phase are shown in Figure 3a. The proportion of G0/G1 phase and S phase on the end of the starvation (0 hr) was almost comparable between the IGFBP7-suppressed cells and the control cells. As shown in Figure 4, the time with minimum percentage of G0/G1 phase and maximum percentage of S phase was 24 hr in the control cells, while the time was 12 hr in the IGFBP7-suppressed cells, which suggests that the cell cycle progression was more rapid in the IGFBP7-suppressed cells than that in the control cells. Furthermore, we found that cyclin D1 and cyclin E were increased and p27 was decreased in the IGFBP7-suppressed cells than those in the control cells, and that there was no significant difference in p21 expression between the 2 cells, which was agreement with more rapid cell cycle progression in the IGFBP7-suppressed cells (Fig. 3b).
Transfection of IGFBP7 attenuates proliferation and invasive activity
Next, the IGFBP7 expression plasmid was transfected into HuH7, whose IGFBP7 expression level was the lowest in the 4 HCC cell lines. The IGFBP7 expression was increased by the transfection, as confirmed by qRT-PCR and western blot analysis (Fig. 4a). The proliferation assay showed significantly less rapid growth in the IGFBP7-overexpressing cells compared to the control cells (Fig. 4b). In addition, the invasive ability of the IGFBP7-overexpressing cells was significantly weaker than that of the control cells (Fig. 4c), which were consistent to the results of the above shRNA plasmid experiments.
In vivo studies
IGFBP7 expression correlates with tumor-related factors in clinical HCC samples.
Next, IGFBP7 expression in the tumoral leison was evaluated in clinical sample by immunohistochemical staining. The immunohistochemical analysis showed that among the 104 patients examined, 67 patients (64.4%) showed positive staining for IGFBP7 and the remaining 37 patients (35.6%) were negative for IGFBP7. The immunohistochemical findings of representative cases are shown in Figure 5a. The clinicopathological factors related to IGFBP7 expression status of the 104 patients are summarized in Table 1. The data indicated that IGFBP7 expression was significantly associated with maximum tumor size and vascular invasion (p < 0.0001, p = 0.0095, respectively).
Table 1. Clinicopathological characteristics of patients with HCC according to IGFBP7 status
On the other hand, the IGFBP7 expression in non-tumoral lesion was homogenously observed in the cytoplasm of cells in all the 104 patients. The immunohistochemically determined IGFBP7 expression level was similar between 53 cirrhotic patients and the remaining 51 non-cirrhotic patients.
IGFBP7 downregulation is an independent significant predictor for postoperative outcome in HCC patients
The disease-free survival (DFS) in patients without IGFBP7 expression (1-/3-/5-year: 42.9%/15.3%/15.3%) was significantly poorer than that in patients showing IGFBP7 expression (1-/3-/5-year: 70.9%/51.8%/43.6%) (p = 0.0002; Fig. 5b). Univariate analyses showed that number of tumors (p = 0.0025), maximum tumor size (p = 0.0001), presence/absence of vascular invasion (p < 0.0001), and Edmondson-Steiner grade (p = 0.0392) all significantly correlated with DFS, in addition to IGFBP7 status (Table 2). Multivariate analysis for DFS using the above 5 factors identified presence/absence of vascular invasion and IGFBP7 status as independent significant factors (Table 2).
Table 2. Statistical analysis of disease-free survival and overall survival of patients with HCC
The overall survival (OS) rate in patients without IGFBP7 expression (1-/3-/5-year: 83.4%/70.2%/59.2%) was also significantly lower than that in patients with positive IGFBP7 expression (1-/3-/5-year: 98.5%/88.4%/82.0%) (p = 0.0063; Fig. 5c). By univariate analysis, number of tumors (p = 0.0007), maximum tumor size (p = 0.0290), presence/absence of vascular invasion (p < 0.0001), and Edmondson-Steiner grade (p = 0.0180) were also significantly correlated with OS (Table 2). Multivariate analysis using the above 5 factors identified presence/absence of vascular invasion, Edmondson-Steiner grade, and IGFBP7 status as independent significant factors in OS (Table 2). Thus, IGFBP7 expression was an overall independent significant factor for postoperative prognosis in HCC patients.
In this study, we first analyzed IGFBP7 function in vitro experiments. The results demonstrated that IGFBP7 downregulation was significantly associated with rapid growth and proliferation of HCC cells. In addition, the cells showed decreased apoptotic cell numbers and expression of apoptosis-related proteins, enhancement of ERK signaling, and rapid cell cycle progression. Considering the implicated tumor suppression activity of IGFBP7, the results of this study are consistent with previous similar reports.7, 9–11, 14, 15, 21 We also previously reported a significant association of IGFBP7 downregulation with resistance to some chemotherapeutic drugs in HCC cells.17 Taken together, it seems apparent that IGFBP7 downregulation is significantly associated with the malignant potential of cancer cells including proliferation, invasiveness, and resistance to chemotherapeutic drugs. To our knowledge, this is the first study to examine the functional role of IGFBP7 in HCC. On the other hand, the cause-and-effect relationship between the IGFBP7 downregulation and the malignant potential is still unsolved, which is expected to be elucidated by further studies in future.
This study also assessed the prognostic significance of IGFBP7 expression in resected human HCC samples. From these findings, IGFBP7 downregulation was significantly associated with tumor progression and postoperative poor prognosis in our patients group, and IGFBP7 status was identified as an independent significant prognostic factor, in addition to other well-known factors. This finding was consistent with the results of the in vitro experiments and serves to suggest that assessing the IGFBP7 expression status of patients with HCC could improve the prediction of prognosis.
We have reported some studies of significant prognostic predictors after surgery for HCC.20, 23–38 In one of the studies, based on cDNA microarray analysis, we identified a set of multiple genes whose expressions were significantly different between patients with good prognosis and those with poor prognosis, and revealed that the gene set was one of the independent prognostic factors.38 IGFBP7 was not included in the gene set because the difference of the expression level was not large between the 2 groups, but the IGFBP7 expression level examined by the microarray analysis was also significantly correlated to the prognosis. This result seems to be consistent with the result of this study. In addition, considering the significant inverse correlation of IGFBP7 expression to the extent of apoptosis and cell cycle progression confirmed by the in vitro experiments, our previous reports of apoptosis- and cell cycle-related molecules as prognostic factors are in agreement with the prognostic impact of IGFBP7.25, 26, 34, 35 Furthermore, we have reported that angiogenesis-related molecules such as angiopoietin-2 and hypoxia-induced factor-1α are significant factors for postoperative prognosis.33 Considering that IGFBP7 was reported to block angiogenesis in human vascular endothelial cells, it may be possible that there is a significant correlation between IGFBP7 expression and the angiogenesis-related molecules, though we did not examine it in this study.39
IGFBP7 is a secreted protein, and recombinant IGFBP7 is commonly purified.40, 41 Indeed, previous studies of IGFBP7 function have expressed the protein exogenously using IGFBP7 viral vectors or by administering recombinant protein.39 In addition, IGFBP7 has also been studied as a possible therapeutic agent for treatment of malignancies that are dependent on BRAF-MEK-ERK signaling.10, 42 Thus, recombinant IGFBP7 could be potentially suitable therapeutically to improve the poor prognosis of HCC patients lacking IGFBP7 expression. In addition, IGFBP7 expression is also subject to epigenetic modification, and aberrant methylation of CpG islands in the IGFBP7 promoter region was confirmed in several kinds of cancers.10, 43 In this regard, Wajapeyee et al.10 showed that treatment of melanoma cell lines with DNA methyltransferase inhibitor, 5-aza-2′-deoxycytidine, restored IGFBP7 expression. Such a finding suggests a therapeutic application whereby IGFBP7 expression and thus function are restored using a DNA methyltransferase inhibitor. Exploring these therapeutic interventions against HCC was unfortunately beyond the scope of this study, and further studies are definitely needed in this regard.
In summary, we found that IGFBP7 downregulation was significantly associated with both tumor progression and clinical outcome in HCC. This result suggested that analysis of IGFBP7 expression in patients might help to predict prognosis, and that IGFBP7 could be a novel therapeutic target in HCC patients with poor prognosis.