The high-efficiency prokaryotic expression vector pET32a plasmid and the BL21 (DE3) competent Escherichia coli cells were purchased from Invitrogen (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The pMD19-T vector was purchased from Takara Bio, Inc. (Otsu, Japan). The pMD19/TRAIL plasmid and wild-type TRAIL protein were obtained from Chengdu Huachuang Biotechnology Co., Ltd. (Chengdu, China). The TRAIL polyclonal antibody was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). The HCT-15, COLO 205, SW620, HT-29 and HCT 116 colon cancer cell lines were purchased from Wuhan Institute of Virology, Chinese Academy of Sciences (Wuhan, China). Female nude mice (6 weeks old) were purchased from the Animal Experimental Center of the Chinese Academy of Sciences (Shanghai, China).

The 114–121 amino acid coding sequence VRERGPQR of the wild-type TRAIL protein was selected and altered to become RRRRRRRR. The upstream primer (Mu3-TR-NdeI) was 5′-ggt cat atg cgt cgt cgt cgt cgt cgt cgt cgt gtg gct gct cac atc a-3′ and the downstream primer (TR-EcoR) is 5′-gtt gaa ttc tta tta acc aac aag gaa agc acc gaa gaa ag-3′.

The genomic DNA of the pMD19/TRAIL plasmid was used as template DNA to perform polymerase chain reaction (PCR) analysis for the amplification of the TRAIL-Mu3 gene. Amplified products were purified and separated via electrophoresis (3% agarose gel) to confirm whether the amplified products had the desired size (11).

The recovered TRAIL-Mu3 gene was ligated to the Takara pMD19-T vector and digested using EcoRI and HindIII enzymes. The gene was identified by electrophoresis. The plasmids of the corrected transformed bacteria containing the inserted gene segment were used for sequencing. The correctly sequenced bacteria were stored for further study (12).

The TRAIL-Mu3 DNA was ligated to the pET32a plasmid and digested by XbaI and EcoRI enzymes. Subsequently, the gene was identified by electrophoresis. The plasmids of the corrected transformed bacteria containing the inserted gene segment were used for sequencing. The correctly sequenced bacteria were stored for further study (12).

The plasmid pET32a/TRAIL-Mu3 was transformed into BL21 (DE3) competent E. coli cells, which were cultured in Luria-Bertani solid medium containing Ampicillin at 4°C overnight. A single colony was isolated and cultured. The cultured liquid was centrifuged (13,800 × g, 4°C, 10 min) and the precipitate was resuspended to generate the prior-to-induction electrophoresis sample. The remaining liquid was cultured and induced with isopropyl β-D-thiogalactopyranoside to generate the post-induction electrophoresis sample.

The culture liquid was centrifuged (13,800 × g, 4°C, 10 min) and the supernatant was discarded. The precipitate was resuspended, broken down using ultrasound and further centrifuged (13,800 × g, 4°C, 10 min). The precipitate was resuspended. The 20-µl supernatant and resuspended precipitate was used to make separate electrophoresis samples.

All the above electrophoresis samples were disposed in boiling water bath for 10 min. Subsequently, the samples were centrifuged (13,800 × g, 4°C, 10 min) and analyzed by electrophoresis (15% gel, 200 V, 35 min).

The TRAIL-Mu3 expression bacteria were broken down by ultrasound and centrifuged. The supernatant was removed and filtered with a 0.45-µm membrane. Subsequently, the filtered supernatant was purified via cation exchange purification, hydroxyapatite purification and anion exchange purification methods, successively (13).

The TRAIL-Mu3 protein exhibited only 5 amino acids which were altered compared with the N terminal of the wild-type TRAIL protein, and the epitopes of the wild-type TRAIL protein were conserved. Therefore, a TRAIL polyclonal antibody was used for the detection and identification of TRAIL-Mu3 protein by western blotting (14,15). The primary antibody used rabbit anti-human TRAIL polyclonal antibody (cat no. bs-1214R) (1:500) at 4°C overnight. The secondary antibody used was a goat anti-rabbit IgG-HRP (cat no. D2313; 1:5,000) at room temperature for 2 h. Finally, the results were detected using an enhanced chemiluminescence detection reagent (Beyotime Institute of Biotechnology, Haimen, China).

The subcellular localization of TRAIL and TRAIL-Mu3 was detected in SW620 colon cancer cells by immunofluorescence (16).

The antitumor effects of TRAIL-Mu3 and TRAIL were detected in 32 different tumor cell lines (including many TRAIL resistant tumor cell lines) through a Cell Counting Kit-8 (CCK-8). The cell growth inhibition rates of TRAIL-Mu3 and TRAIL in colorectal cancer cell lines (including HCT-15, COLO 205, SW620, HT-29 and HCT 116) measured by CCK-8 assay were discussed in this manuscript (17). The concentrations of reagents that induced a 50% reduction in cell viability [half-maximal inhibitory concentration, (IC₅₀)] were determined from the curves of reagent concentration compared with the cell growth inhibition rate at 48 h of incubation for the cell line analyzed. The sensitivity of cells to a drug is evaluated by calculating the IC₅₀ value; IC₅₀ <10 µg/ml indicates that cells are sensitive to a drug, while IC₅₀ ≥10 µg/ml suggests that cells are relatively resistant to a drug (18–19).

All animal procedures were approved by the Animal Care and Scientific Committee of Sichuan University (Chengdu, China). A total of 48 HT-29-bearing nude mice (SPF, aged 6–8 weeks, 18–22 g) were divided into 6 groups (8 mice/group): Group 1, treated with vehicle (saline); group 2, treated with TRAIL-Mu3 at a concentration of 5 mg/kg; group 3, treated with TRAIL-Mu3 at a concentration of 15 mg/kg; group 4, treated with TRAIL-Mu3 at a concentration of 45 mg/kg; and group 5, treated with TRAIL at a concentration of 45 mg/kg. The mice were bred at 23±2°C, with a humidity of 40–70% and a 12/12 light/dark cycle with free access to food and water. The vehicle, TRAIL and TRAIL-Mu3 were injected through the tail vein five times in 5 days. The length, width, and weight of the tumor was measured using a slide caliper every 3 or 4 days. Tumor volume (TV) was estimated using the formula: TV (mm³) = (width² × length)/2. The growth inhibition rate of the tumor was calculated using the formula: Inhibition rate of tumor (%) = (1 - average weight in treated group/average weight in control) ×100 (11,12).

During the experimental period, side effects, including weight loss, mental state, appetite, behavior change and reactions, were observed.

The experiments were repeated at least three times. All data are expressed as the mean ± standard error of the mean unless otherwise stated. Comparisons were analyzed using one-way analysis of variance and Least Significant Difference post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The amplified products of the TRAIL-Mu3 gene were detected by 3% agarose gel electrophoresis, and the results demonstrated that a specific segment of ~500 bp (Fig. 1A) was obtained.

Following ligation of the pMD19-T vector and TRAIL-Mu3 genes, they were transformed into bacteria and cultured. A total of seven single colonies were isolated and the plasmids were extracted. Subsequently, the plasmids were digested and detected by agarose gel electrophoresis. The results demonstrated that there were two segments of ~500 bp and ~2.7 kb in lines 1, 2, 4 and 5 (Fig. 1B), which were equal to the size of the TRAIL-Mu3 gene and pMD19-T vector plasmid, respectively. Therefore, these four colonies were transformed with pMD19/TRAIL-Mu3, and were further assayed by sequencing.

The pET32a plasmid and TRAIL-Mu3 DNA were digested by NdeI and EcoRI enzymes and detected by electrophoresis. The results demonstrated that there were two segments of ~500 bp and ~5.4 kb (Fig. 1C), which were equal to the size of the TRAIL-Mu3 gene and pET32a plasmid, respectively. Subsequently, the plasmids were ligated, transformed into bacteria and cultured. A total of seven single colonies were isolated and the plasmids were extracted. The plasmids were digested and detected by agarose gel electrophoresis. The results demonstrated that there were two segments of ~500 bp and ~5.4 kb in lines 2, 3, 4 and 5 (Fig. 1D), which were equal to the size of the TRAIL-Mu3 gene and pET32a plasmid, respectively. Therefore, these four colonies were transformed with pET32a/TRAIL-Mu3, and they were further analyzed by sequencing.

As presented in Fig. 1E, there was strong expression in the prior-to-induction electrophoresis sample, post-electrophoresis sample, supernatant and precipitate, and that this was most apparent in the supernatant.

The results of this analysis are presented in Fig. 2. Following cation exchange purification (Fig. 2A), 15 ml eluent with a concentration of 2.273 mg/ml was obtained. The purity of TRAIL-Mu3 was thus demonstrated to be high. Via hydroxyapatite purification (Fig. 2B), 12 ml hydroxyapatite eluent with a concentration of 2.080 mg/ml was obtained. The purpose of this step was to attempt to remove contaminating proteins and pyrogen. Following anion exchange purification (Fig. 2C) 20 ml flow through fluid with a concentration of 0.846 mg/ml was obtained. In this step, the pyrogen was further removed. Following repeated purification, sufficient protein was obtained to evaluate its bioactivity in vitro and in vivo.

The results of the western blotting (Fig. 3) demonstrated that there was a positive band of ~19.6 kDa for TRAIL-Mu3 and the TRAIL protein. However, the same band was not present for the supernatant of lysed BL21 (DE3) E. coli.

The results demonstrated that there was little TRAIL aggregated on the cell membrane of SW620 cells (Fig. 4A); however, extensive TRAIL-Mu3 was observed around the cell membrane of the cells (Fig. 4B).

The IC₅₀values of TRAIL and TRAIL-Mu3 in various colorectal cancer cell lines are presented in Table I. The results demonstrated that TRAIL-Mu3 was able to enhance the antitumor effects of TRAIL in all five colorectal cancer cell lines. In addition, TRAIL-Mu3 was able to reverse the resistance of TRAIL-resistant HT-29 colorectal cancer cells.

The treatment commenced on day 10 following the injection of tumor cells. The tumor growth curves and inhibitive rates are presented in Fig. 5. The results demonstrated that the tumor growth inhibition rate increased with the increase in TRAIL-Mu3 concentration. In addition, the tumor growth inhibition rate of TRAIL-Mu3 was significantly increased compared with TRAIL at the same concentration of 45 mg/kg.

All the mice in the TRAIL and TRAIL-Mu3 groups began to manifest slight syndromes such a stunt responses, bad appetite, and little activity on the 8th day after the treatment. However, there were no significant differences between TRAIL and TRAIL-Mu3 groups in mental status, weight, appetite and so on.