Fresh frozen tissue samples from 12 patients (8 males and 4 females aged from 49–83 years old) with CRC were provided by the Department of Colorectal Surgery, Tianjin Union Medical Center (Tianjin, China). All the tissue samples were collected from patients pathologically diagnosed with CRC between Janunary and March 2018, and the characteristics of each patient are summarized in Table I. This study was approved by the Tianjin Nankai Hospital ethics commission (Tianjin, China; approval no. NKYY_YX_IRB_044_01).

The human CRC cell lines, HT29, HCT116, Colo205 and SW620 were obtained from the Tianjin Institute of Integrative Medicine for Acute Abdominal Diseases, Tianjin Nankai Hospital and the 293T cell line was provided by Professor Xiong Dongsheng from the Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College (Tianjin, China). Cells were maintained in Dulbecco's modified Eagle medium (DMEM; cat. no. 1791922; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; cat. no. 25030-081; Gibco; Thermo Fisher Scientific, Inc.) at 37°C. Cells cultured under hypoxic conditions were placed in a hypoxic chamber as previously described (14,15). The air inside the chamber was flushed with a mixture of N₂ (95%) and CO₂ (5%) and when the oxygen concentration decreased to 1%, the chamber was sealed and kept at 37°C.

Peripheral blood lymphocytes (PBLs) were isolated from the blood of healthy donors from the Tianjin Blood Center (Tianjin, China; dataset no. ISCP17021885; tjbc.org.cn) by Ficoll^® solution (cat. no. LTS10771; TBD Science, Tianjin, China). The cells were maintained in Roswell Park Memorial Institute-1640 medium (cat. no. 1721503; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS and 100 U/ml recombinant human IL-2 (cat. no. 202-IL-010; R&D Systems, Inc., Minneapolis, MN, USA) at 37°C every 48 h.

Total RNA was extracted from cells or tissue samples using TRIzol^® reagent (cat. no. 15596026; Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Total RNA (4 µg) was reverse transcribed into cDNA using the RevertAid First Strand cDNA Synthesis kit (cat. no. K1622; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. qPCR was subsequently performed using the SYBR^® Premix Ex Taq™ kit (cat. no. RR420L; Takara Biotechnology Co., Ltd., Dalian, China). The primer pairs used are presented in Table II, and β-actin was used as a reference gene for normalization. The following thermocycling conditions were used for the qPCR: Initial denaturation at 95°C for 30 sec; 40 cycles of 95°C for 5 sec and 60°C for 34 sec. To analyze the correlation of gene expression in tumor tissues of patients with CRC, the Cq values from patient 1 were used as a baseline to correlate the Cq values of other patients, and the Pearson product-moment correlation coefficient (Pearson's R) was measured. Relative gene expression in CRC cell lines was quantified using the 2^−∆∆Cq method (16). These experiments were repeated in triplicate.

Cells were seeded onto 96-well plates at a density of 1×10⁴ cells/well and treated with 0, 12.5, 25, 50, 100 or 200 µM rhein (cat. no. SG8100; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) for 48 h. Then the purple formazan were dissolved in DMSO (cat. no. D8371-50; Beijing Solarbio Science & Technology Co., Ltd.) and measured at a wavelength of 492 nm. In CRC cell lines, cell viability was measured by MTT reduction assay (cat. no. M2128; Sigma-Aldrich, Merck KGaA, Darmstadt, Germany). In PBLs, cell viability was measured by cell counting kit-8 (cat. no. CK04; Dojindo Molecular Technologies, Inc., Kumamoto, Japan) as manufacturer's protocol. Untreated cells were used as a negative control and cell viability was set as 100%.

The mCD3scfv sequence was provided by Professor Xiong Dongsheng and amplified by PCR using Pyrobest^™ DNA Polymerase (cat. no. R005A; Takara Biotechnology Co., Ltd.) and the following primers: 5′-CGTAGAATTCGCCACCATGGAGACAGACACACTCCTG-3′ and 5′-CGTAGGATCCCTAACGTGGCTTCTTCTCGT0-3′. The thermocycling condition was as follows: Initial denaturation at 95°C for 5 min; 35 cycles of 95°C for 10 sec, 60°C for 60 sec and 72°C for 60 sec. The PCR product was purified and cloned into the lentiviral expression vector pCDH1-CMV-MSC-EF1 α-Puro (cat. no. CD510B-1; System Biosciences, LLC., Palo Alto, CA, USA) using EcoRI and BamHI restriction sites. The lentiviral expression construct was co-transfected with backbone plasmids [5 µg MD2G, 3 µg PAX2 (Invitrogen; Thermo Fisher Scientific, Inc.) and 7 µg pCDH1] in 293T cells using X-tremeGENE^™ HP DNA transfection reagent (cat. no. 06366236001; Roche Molecular Diagnostics, Pleasanton, CA, USA) to produce the lentivirus (17). HT29 cells were transduced with the lentivirus for 48 h and selected with 30 µg/ml puromycin (cat. no. P8230; Beijing Solarbio Science & Technology Co., Ltd.) for 2 weeks. The HT29 cell line stably expressing mCD3scfv, was termed HT29-CD3scfv.

Total protein was extracted from cells using radioimmunoprecipitation lysis buffer (cat. no. R0010) and the protein concentration was determined by BCA Protein Assay kit (cat. no. PC0020; both Beijing Solarbio Science & Technology Co., Ltd.). The proteins were separated using SDS-PAGE (10% polyacrylamide gel; 40 µg total protein per lane) and transferred onto polyvinylidene difluoride membranes (cat. no. IPVH00010; EMD Millipore, Billerica, MA, USA). Membranes were blocked by 5% skim milk at room temperature for 4 h and then incubated with primary antibodies (1:500) against HIF-1α (cat. no. 100-449), β-tubulin (cat. no. 66240-1; both 1:500; Wuham Sanying Biotechnology, Inc., Wuhan, China), and HA-tag (cat. no. ab1818; 1:500; Abcam, Cambridge, MA, USA) overnight at 4°C. Following primary incubation, membranes were incubated with horseradish peroxidase-conjugated goat anti-mouse (cat. no. SE12) or anti-rabbit (cat. no. SE13) secondary antibodies (both 1;2,000; Beijing Solarbio Science & Technology Co., Ltd.) for 1 h at room temperature. Protein bands were visualized using the BioVision ECL Western Blotting Substrate kit (cat. no. K820-500; BioVision, Inc., Milpitas, CA, USA).

HT29-CD3scfv or HT29-control cells were fixed with 4% paraformaldehyde for 10 min and blocked with 1% bovine serum albumin (cat. no. A8010; Beijing Solarbio Science & Technology Co., Ltd.) for 1 h. Cells were incubated with mouse anti-HA-tag primary antibody (cat. no. ab18181) for 1 h, followed by incubation with allophycocyanin (APC)-conjugated goat anti-mouse IgG (cat. no. ab130782; both 1:1,000; Abcam) secondary antibody for 1 h. Cells were stained with DAPI (Sigma-Aldrich, Merck KGaA) for 5 min. All the steps were performed at room temperature. Images were captured using a two-photon laser scanning confocal microscope (Olympus Corporation, Tokyo, Japan; FV1200 MPE).

HT29-CD3scfv or HT29-control cells were incubated with mouse anti-HA tag primary antibody (cat. no. ab18181; 1:1,000; Abcam) for 30 min at room temperature. Following primary incubation, cells were washed twice using PBS and incubated with APC-conjugated goat anti-mouse IgG (cat. no. ab130782; 1:1,000; Abcam) secondary antibody for 30 min at room temperature. Following one wash by PBS, cells were analyzed using a flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA; FACS LSRII). The data were analyzed by FlowJo software (version 7.6; Tree Star, Inc., Ashland, OR, USA).

The specific lysis of target cells was measured by lactate dehydrogenase release assay using the CytoTox96^® Non-Radioactive Cytotoxicity Assay kit (Promega Corporation, Madison, WI, USA) according to the manufacturer's protocol. HT29 and HT29-CD3scfv cells were seeded onto 96 well plates (1×10⁴ cells/well) as target cells. PBLs were added at different effector-to-target (E:T) cell ratios 5:1, 10:1 and 20:1 to 96-well culture plates and co-cultured with target cells for 16 h.

For assessment under hypoxic conditions, HT29-CD3scfv cells were pretreated with or without 50 µM rhein for 8 h. Untreated cells in normoxia were used as the negative control. PBLs were subsequently added at different E:T cell ratios and co-cultured with target cells for 16 h, and specific cytotoxicity was detected as described above.

The results are reported as the mean ± standard deviation of at least three independent experiments. All experimental data were analyzed using one-way analysis of variance (ANOVA) with Dunnett's post-test. The strength of a linear association between two variables was calculated with the Pearson product-moment correlation coefficient (Pearson's R coefficient). Pearson's R coefficient and ANOVA were performed using GraphPad Prism 5 software (GraphPad Software Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

To determine the effect of HIF-1α in CRC, the mRNA expression levels of HIF-1α and six immunosuppressive molecules were determined by RT-qPCR from 12 CRC tumor tissue samples. The correlation between the mRNA expression level of HIF-1α and the six immunosuppressive molecules was analyzed by Pearson's R coefficient. The mRNA expression levels of PD-L1, VEGF, COX-2, and galectin-1 were positively correlated with that of HIF-1α in CRC tissues (Fig. 1). However, there was no significant correlation between HIF-1α and IL-10 or TGF-β1 in the CRC tissue samples analyzed.

To determine the effect of rhein in CRC, cell viability was examined using the MTT and CCK-8 assays. The MTT assay was performed in CRC cell lines (HT29, HCT116, Colo205, SW620) following treatment with rhein (12.5, 25, 50, 100 and 200 µM) for 48 h. Rhein (100 and 200 µM) significantly decreased cell viability in a what appeared to be a dose-dependent manner in all four CRC cell lines (P<0.05; Fig. 2A). The CCK-8 assay was performed in PBLs following treatment with rhein (25 or 50 µM) for 48 h. Rhein had no affect on PBL cell viability (Fig. 2B).

To determine the inhibitory effect of rhein in hypoxia, the expression levels of HIF-1α and six immunosuppressive molecules were examined using western blotting in HT29 cells cultured in 4 conditions: Normoxia, hypoxia, hypoxia + 25 µM rhein, and hypoxia + 50 µM rhein for 16 h. The protein expression level of HIF-1α was increased in hypoxia, compared with normoxia. Additionally, rhein markedly decreased the protein expression level of HIF-1α under hypoxic conditions in a dose-dependent manner (Fig. 3A). The mRNA levels of six immunosuppressive molecules were detected by RT-qPCR. Similar to HIF-1α protein expression, the mRNA expression levels of PD-L1, VEGF, COX-2, galectin-1, IL-10 and TGF-β1 were upregulated in hypoxia, compared with normoxia. Rhein significantly decreased the mRNA expression levels under hypoxic conditions in a dose-dependent manner (Fig. 3B). These results demonstrated that the mechanism underlying the inhibitory effect of rhein may be HIF-1α-dependent. Although the mRNA expression levels of IL-10 and TGF-β1 were also downregulated following treatment with rhein under hypoxic conditions (Fig. 3B), there was no correlation between HIF-1α and IL-10 or TGF-β1 observed (Fig. 1). These results identified a potential HIF-1α-independent mechanism underlying the inhibitory effect of rhein.

To examine the cytotoxicity of PBLs in CRC, a CRC cell model specifically recognized by PBLs was established using a lentivirus containing the membrane-bound anti-CD3scFv expression cassette (Fig. 4A). HT29 cells were transduced with the lentiviral vector or blank control for 48 h and then selected with puromycin for an additional 2 weeks. To identify HT29 cells stably expressing mCD3scfv, western blot analysis, confocal microscopy and flow cytometry were used to detect the N-terminal HA-tag of the fusion protein. The results revealed that mCD3scfv was expressed in HT29-CD3scfv cells and mCD3scfv localized on the cell membrane (Fig. 4B and C). Furthermore, HT29 cells expressing the fusion protein accounted for 96.8% of the total HT29-CD3scfv cells examined (Fig. 4D). Finally, HT29-CD3scfv and HT29-control target cells were co-cultured with PBLs at different E:T cell ratios, respectively for 16 h. CytoTox96^® non-radioactive cytotoxicity assays revealed that PBLs were cytolytic toward HT29-CD3scfv cells. Their cytotoxic effect increased as the E:T ratio using PBLs as effector cells and HT29-CD3scfv cells as target cells increased. PBLs were slightly cytolytic toward HT29-control cells. Furthermore, the cytotoxity of HT29-CD3scfv singicantly increase compared with HT29-control cells at the corresponding E:T ratio (P<0.05; Fig. 4E).

To further examine the inhibitory effect of rhein in hypoxia, the cytotoxicity of PBLs toward CRC cells was examined following treatment with rhein under hypoxic conditions. HT29-CD3scfv cells were cultured under conditions of normoxia, hypoxia and hypoxia + 50 µM rhein, followed by co-culture with PBLs at different effector-to-target cell ratios (5:1, 10:1 and 20:1). CytoTox96^® non-radioactive cytotoxicity assays revealed that PBLs are cytolytic toward HT29-CD3scfv cells, and their cytotoxic effect was inhibited by hypoxia. Treatment with rhein enhanced the cytotoxicity of PBLs in hypoxia, compared with normoxia (Fig. 5).