LX-2 human liver stellate cell line was provided by Dr Scott L. Friedman (Mount Sinai School of Medicine, NY, USA). The cell line is characterized as an immortal, non-malignant cell line that retains key features of the hepatic stellate lineage. It was included in the current study as a negative control/normal cell. The HepG2 human liver cancer cell line and Ad-293 cell line were purchased from the National Cell Bank, Pastor Institute of Tehran (Tehran, Iran). All cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 10 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin (Thermo Fisher Scientific, Inc., Waltham, MA, USA) under 5% CO₂ atmosphere and 37°C condition.

The original MDA-7 plasmid (pGEX-5X1/GST-IL-24) was a gift from Dr Stephanie Kreis (Laboratoire de Biologie et Physiologie Integree University of Luxembourg, Esch-sur-Alzette, Luxembourg) (19). The plasmids, expressing IL-24 and SP.RGD.IL-24, were constructed by stepwise cloning. Using the P1 and P2 primers, which harbored the BamHI and XhoI recognition sites (Table I), the normal MDA-7/IL-24 was amplified by Taq DNA polyemerase (Thermo Fisher Scientific, Inc.) using the following cycling conditions: Denaturation at 95°C for 5 min; followed by 35 cycles at 95, 58, and 72°C for 15 sec, 40 sec, and 1 min, respectively. The amplified segment was directly cloned into the pcDNA3.1 (Addgene, Inc., Cambridge, MA, USA) expression vector. The resulting vector was designated as pc/IL-24.

The modified IL-24 coding sequence with proceeding RGD4C sequence was constructed by replacing the intrinsic IL-24 signal peptide sequence with a fusion of the artificial signal peptide RGD4C sequence. The secrecon is a bioinformatically-designed signal sequence with a high secretion potency, which is composed of 21 amino acids (20). The cleavage site of the secrecon in the fusion sequence was predicted just before the RGD sequence using SignalP v.4.0 software (http://www.cbs.dtu.dk/services/SignalP/) (21).

The fusion segment of the artificial signal peptide RGD4C was made by extending the forward (S1) and the reverse (S2) primers. This sequence was termed SP-RGD4C. The S1 and S2 primers had an overlapping complementary 20-base sequence in their 3′ ends (double underline), which allows the primers to extend each other. Additionally, the IL-24 coding region was amplified without its intrinsic signal peptide sequence by the P2 and P3 primers. As the forward primer (P3) was designed immediately following the signal sequence, the amplified segment did not contain the signal sequence. The S2 primer had 25 overlapped bases in the 5′end (underlined) with the IL-24 sequence immediately following the signal sequence. Therefore, the IL-24 without intrinsic signal sequence was fused to the SP-RGD4C using the synthesis by overlap extension polymerase chain reaction (PCR) method. The total sequence was termed SP.RGD.IL-24. Due to the features of the S1 and P2 primers with the BamHI and the XhoI recognition sites, respectively, SP.RGD.IL-24 final amplicons were adjustable for the directed cloning into the pcDNA3.1 expression vector. The vector, containing the SP.RGD.IL-24 sequence, was termed pc/SP.RGD.IL-24. The gene insertion and the integrity of the constructs were assessed through PCR, restriction analysis and sequencing.

Cells were plated 1 day before transfection. A total of 4×10⁵ cells were seeded per well in 6-well plates and transfected with 1 µg plasmid vectors using Lipofectamine-LTX™ (Thermo Fisher Scientific, Inc.) transfection reagent according to the manufacturer's instructions. The cell lines were transfected separately with empty pCDNA3.1 (pc), pc/IL-24 or pc/SP.RGD.IL-24. To optimize plasmid transfection method and estimate transfection rate, a control GFP-expressing plasmid (pAdenovator-CMV5-IRES-EGFP vector; Qbiogene; MP Biomedicals, LLC, Santa Ana, CA, USA) was also included. The percentage of GFP expression was observed using fluorescent microscopy and used to estimate the transfection rate.

In order to evaluate the gene expression potency of the new constructs, after cultivating the transfected and non-transfected Ad-293 for 72 h, the supernatants were collected and centrifuged at 3,000 × g for 5 min at room temperature. The concentrations of IL-24 or SP.RGD.IL-24 in Ad-293 supernatants were quantified using an IL-24 human ELISA kit (Abcam, Cambridge, MA, USA; cat. no. ab171345), following the manufacturer's instructions. All the experiments were performed in triplicate and the mean value was included in the analysis.

As the final product of the novel SP.RGD.IL-24 sequence was proposed to anchor to integrin receptors, a simple attachment assay was performed to determine the potency. For this purpose, an ELISA-based method was performed. In this regard, at first Ad-293 cells were transfected with each of the pc.1/IL-24 or pc/SP.RGD.IL-24 constructs. After 72 h, the supernatants of these cells were collected and the concentrations of the gene products were estimated using an IL-24 ELISA kit. The gene products were then diluted to the same concentrations. HepG2 cells were plated in a 6-well plate (4×10⁵ cells per well), 2 days prior to the attachment assay. The HepG2 cell is among the types of liver cell to express integrins, thus, it can support the attachment of RGD-modified protein.

Subsequently, the supernatant of cells was removed and the plated cells were washed twice with PBS. Then, different supernatants from transfected Ad-293 cells, containing different gene products, but with the same concentration, were added to each well of HepG2 cells. After 2 h, the supernatants were removed slowly and the concentrations of the protein products were estimated by the IL-24 ELISA kit again. The reduced amount of the protein concentrations were estimated as the cell attachment capability of the gene products. The assays were performed in triplicate.

Total cellular RNA was isolated from cells 36 h after transfection by using the Total RNA Isolation System kit (Promega Corporation, Madison, WI, USA) in accordance with the manufacturer's instructions. The relative RNA integrity was, then, checked by visualizing the ribosomal RNA bands via gel electrophoresis and using a NanoDrop (Thermo Fisher Scientific, Inc., Pittsburgh, PA, USA). Total RNA (~1 µg) was subjected to reverse transcription using the cDNA Synthesis Premix (GeneAll Biotechnology Co. Ltd, Seoul, South Korea), according to the manufacturer's instructions.

qPCR was performed on the ABI 7500 Sequence Detection System (Applied Biosystems; Thermo Fisher Scientific, Inc.) using qPCR SYBR GreenMaster mix (Jena Bioscience GmbH, Jena, Germany). Thermal cycling was performed using the following conditions: 95°C for 5 min as the first denaturation step, followed by 40 cycles at 95, 60, and 72°C for 15 sec, 40 sec, and 1 min, respectively. Each individual run was followed by a melting curve analysis for 65–95°C to ensure the homogeneity of the PCR products. The phosphoglycerate kinase 1 gene was used as a reference gene to normalize the expression levels. Based on the similar efficiency of the qPCRs, the relative quantification of genes was measured using the 2^−ΔΔCq method and represented as fold change in expression (22). The assays were performed in triplicate independently. The primer sequences, employed in the qPCR assays, are listed in Table I.

An apoptosis assay was performed using a propidium iodide (PI)/annexin V-APC staining kit (eBioscience, Inc., San Diego, CA, USA), according to the manufacturer's protocols. In brief, the transfected cells were trypsinized and harvested 48 h after transfection, washed with PBS, and resuspended in 1% binding buffer (200 µl). The cells of different wells were, then, aliquoted equally and stained sequentially with annexin V-conjugated APC and PI for 15 min in the dark. Subsequently, 300 µl binding buffer was added to the mixture and the acquisition was immediately performed on a FACSCalibur flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA). The analyses were completed by using Cell Quest Pro 5.1 software package (BD Biosciences). The sum of early and late apoptosis cells percentages was considered as the total percentage of apoptotic cells for further analysis. The assays were performed in three independent experiments and subsequently, the means of each group were statistically compared.

In all of the experiments, the statistical differences between the means were evaluated by one-way analysis of variance followed by Tukey post test evaluation. P<0.05 was considered to indicate a statistically significant difference. Values are presented as the mean ± standard deviation.

Following cloning of two different sequences into the pcDNA3.1 plasmid, all PCR reactions, restriction analysis and sequencing result confirmed the vector integrity, sequence accuracy, and the correct direction of inserts. The sequencing results confirmed that SP.RGD.IL-24 had an artificial signal sequence, followed by the RGD4C sequence, and the sequence of IL-24 beyond its intrinsic signal sequence.

The IL-24 gene, with an intrinsic signal peptide, produces a secretory protein, which is a common feature of all cytokines. Additionally, the novel SP.RGD.IL-24 sequence, with an artificial signal peptide, was expected to similarly produce a secretory protein. As the most reliable test for the protein secretion assay, the concentration of IL-24 in the supernatants of the transfected cells was quantified by ELISA. The results demonstrated that the concentration of IL-24 in the supernatant of the A-293 cells transfected with pc/SP.RGD.IL-24, was fairly similar to pc/IL-24, albeit the former construct exhibited less production (P>0.05). As expected, the concentration of IL-24 in the supernatant of the transfected cells, with an empty pcDNA3.1, was negligible (data not shown).

To assess whether the RGD motif of SP.RGD.IL-24 is functional and if is accessible to its cognate integrin, αvβ3, an ELISA based method was employed. Reduction of gene products in the presence of HepG2 cells indirectly indicated that adhesion of SP.RGD.IL-24 to integrin αvβ3 is increased compared with IL-24 (Fig. 1).

Three prepared plasmids were transfected into LX-2 cells (as the normal control) and HepG2 liver cancer cells. Ad-293, LX-2 and HepG2 cells exhibited different rates of susceptibility to transfection. The GFP signal counting under microscope (data not shown) indicated that transfection rate for Ad-293, LX-2 and HepG2 cells were estimated to be ~80, 70 and 50% respectively, during experiments. Subsequently, qPCR was employed to compare the expression level of two pro-apoptotic genes (Gadd153 and Bax), following transfection of plasmids expressing IL-24 and the modified SP.RGD.IL-24 in LX-2 and HepG2 cells. Notably, the attained data exhibited a significant difference of the response to plasmids in normal and tumor cells. The expression data analysis demonstrated that the expression of the modified SP.RGD.IL-24, similarly to IL-24, significantly upregulated the Gadd153 (2.5 fold) and Bax (1.9 fold) expression levels in RNA extracted from plasmid-transfected LX-2 and HepG2 cells, compared with the empty plasmid when assayed on HepG2 (P<0.05). However, the gene expression induced by the SP.RGD.IL-24 construct was significantly greater than that induced by IL-24. As a result, Gadd153 and Bax exhibited 74 and 73% overexpression, respectively (P<0.05). In contrast to the HepG2 cells, the expression of IL-24 or SP.RGD.IL-24 in LX-2 cells did not have a significant effect on Gadd153 and Bax mRNA expression levels compared with the empty plasmid (P>0.05; Fig. 2).

The annexin/PI method was used to investigate the apoptosis induction of different constructs by flow cytometry. As a result, in the analysis, the Q2 and Q3 quadrants were determined as the early and late apoptosis-induced cells, respectively (Fig. 3A). The percentage of total apoptosis (early and late stages) for the acquired population following HepG2 transfection were as follows: 19.8±4.6% for the native IL-24 (P=0.00051), 27.5±4.8% for the modified SP.RGD.IL-24, and 3.9±1.4% for the empty plasmid (Fig. 3). Modified SP.RGD.IL-24 had significantly higher apoptosis induction potency when compared with unmodified IL-24 (P<0.05). Additionally, the amount of necrosis events was negligible according to the data from the Q1 section (Fig. 3A). In transfected LX-2 cells, no significant changes in the apoptotic population was detected compared with the control group, and the quantity of cells in Q2 and Q3 was similar (<4%) in all groups (Fig. 3).

In brief, the results demonstrated that the modified SP.RGD.IL-24 was more efficient in the induction of apoptosis in HepG2 cells compared with the IL-24. However, SP.RGD.IL-24, similar to IL-24, had no detectable effect on apoptosis of non-cancerous LX-2 cells.