The leukapheresis procedure was carried out after obtaining a letter of approval from Universiti Sains Malaysia (USM) human ethical research committee (JEPeM code: USM/JEPeM/17120677). The animal study was endorsed by the USM Institutional Animal Care and Use Committee (IACUC) (USM/Animal Ethics Approval / 2015 / (96)/ (628)). All animals were kept and preserved based on the guidelines stated by IACUC.

Leukapheresis was carried out through peripheral vein entree using the Spectra Optia system (Terumo BCT, Lakewood, Colorado, USA) based on the CMNC program (software version 11.3) as previously described14. The anticoagulant utilized in the collection bag was acid-citrate-dextrose formula A (ACD-A) (Terumo BCT, Lakewood, Colorado, USA) without heparin. The flow rate was positioned at 0.8 ml/min/LTBV with a ratio of 1:12 between anticoagulation and inlet. The inlet flow rate during the procedure ranged from 40 to 70 ml/min while the anticoagulant flow rate ranged from 0.7 to 1.1 ml/min/LTBV. The flow rate for the collection was fixed at one ml/min. Approximately 120 ml of mononuclear cells (MNCs) was obtained from the leukapheresis procedure.

Human CD8⁺ T cells were separated from freshly isolated MNCs by negative selection using the human CD8⁺ T cell isolation kit as previously described15 and these were cultured in appropriate medium in the good manufacturing practice (GMP)-certified laboratory owned by Nichi-Asia Life Science Sdn Bhd, Malaysia. Briefly, 2 x 10⁸ cells were mixed into 400 µL of PBS, then 100 μL of non-CD8⁺ T cell biotin-antibody cocktail was added to the mixture and incubated at 4 °C for 5 minutes. Next, 300 μL of phosphate-buffer saline (PBS) and 200 μL of anti-biotin microbeads were added in the cell mixture and incubated at 4 °C for 10 minutes. The CD8⁺ T cells were separated by filling cells into the LS separation column attached to the magnetic cell separator. The flow-through consisting of unlabeled cells, which was enriched with CD8⁺ T cells, was collected, while labelled cells (non-CD8⁺ T cells) were trapped in the column due to the magnetic beads bound to their surface antigens. After separation, the suspension was removed and centrifuged at 300 x g for 10 minutes. The cell pellet was mixed with one ml complete RPMI medium and the cells were counted.

Human moDCs were produced by culturing freshly isolated MNCs in culture plates for 2 hours in the GMP-certified laboratory owned by Nichi-Asia Life Science Sdn Bhd, Malaysia. Non-adherent cells were discarded while adherent cells (monocytes) were cultured in complete RPMI 1640 which was composed of 10% FBS, 1 mM glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin, together with human granulocyte macrophage colony-stimulating factor (GM-CSF) at 50 ng/ml (Peptrotech, Rocky Hill, NJ, USA) and 10 ng/ml recombinant human IL-4 (Peprotech, Rocky Hill, NJ, USA) for 7 days in the 5% CO₂ incubator at 37 °C. The medium was changed every two days with fresh complete RPMI medium together with GM-CSF and IL-4. Immature moDCs were successfully produced after 7 days.

Isolated CD8⁺ T cells and generated moDCs were characterized by staining CD8⁺ T cells for CD3 and CD8, and moDCs for CD11c and CD83 surface expression. The phenotyping and analyses were carried out using BD FACSCanto II (Becton Dickinson, Franklin Lakes, New Jersey, USA) with FlowJo 7.5.3 software.

MDA-MB-231 originating from human breast adenocarcinoma cell lines was obtained from the American Type Culture Collection (Manassas, Virginia, USA) and maintained in the Dulbecco's Modified Eagle's medium (DMEM) (Thermo Fisher Scientific, Waltham, Massachusetts, USA) containing 5% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, Massachusetts, USA), 100 U/ml penicillin, 100 µg/ml streptomycin (Thermo Fisher Scientific, Waltham, Massachusetts, USA), 600 µg/ml L-glutamine (Thermo Fisher Scientific, Waltham, Massachusetts, USA) and 6/500 ml 100X non-essential amino acids (Thermo Fisher Scientific, Waltham, Massachusetts, USA), and kept in the 5% CO₂ incubator at 37 °C in the GMP-certified laboratory (Nichi-Asia life science Sdn Bhd, Malaysia). MDA-MB-231 cells were accumulated at the exponential phase of growth before injection in the mammary fat pads of mice.

Immature DCs were collected on day 7, and 10 x 10⁶ cells were plated into a T25 flask. Mature DCs were induced with whole tumor lysate from MDA-MB-231 cells. Briefly, tumor cells were collected at 80% confluence by depleting using 2 ml 0.25% trypsin-EDTA, incubated in the 5% CO₂ incubator at 37 °C for 5 min, and washed in 5 ml of PBS. The trypsinased cells were centrifuged at 300 x g for 5 minutes. The cell pellets were lysed using the RIPA Lysis Buffer System. The BCA Protein Assay Kit was used to determine the total protein concentration. The tumor lysates were stored in aliquots at -80 °C until further use. Tumor lysates were mixed with DCs in the culture flask at a ratio of one DC to five tumor cell equivalents for 24 hours.

Mature moDCs and T cells were co-cultured in a T25 flask at a ratio of 1∶20 (DCs:T), as this ratio was indicated to produce the most desired T cell proliferation. This high ratio was expected due to the relative infrequency of DCs bearing in vivo-captured antigen. Co-cultures were preserved for 48 hours for further experiments.

Eight weeks old NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) mice were acquired from Jackson Laboratory (Bar Harbor, Maine, USA). After one-week of acclimation to the animal research facility, Advanced Medical and Dental Institute, USM, NSG mice were subjected to tumor inoculation. MDA-MB-231 cells were suspended into 200 μl of PBS at a concentration of 5 x 10⁶ cells. Before tumor inoculation, mice were anesthetized with ketamine-xylazine-acepromazine (80 – 120 mg/kg) by intra-peritoneal injection. A small incision was made over the right or left auxiliary mammary fat pad and the tumor cells were injected using an insulin syringe with a 27 to 30-gauge needle. The incision was closed with a simple suture. Mice were monitored daily for well-being (physical behavior, food and drink update, and animal body weight) and tumor development (tumor volume and size). Tumor size was measured twice a week using caliper measurement and the volume were calculated as D x d² x 0.52, where D is the long diameter and d is the perpendicular short diameter. After tumor size reached approximately 0.2 cm, NSG mice were injected via tail vein with 20 x 10⁶ CD8⁺ T cells and tumor-activated moDCs pulsed CD8⁺ T cells twice a week for 3 weeks. Treated NSG mice were terminated four weeks after immune cell treatment. Blood samples were collected from the submandibular vein of NSG mice at the experiment initiation and before termination. After treated NSG mice were terminated, primary mammary fat pad tumor tissue and lung were harvested. Harvested tumor organs were been subjected to pathophysiological examination.

Formalin-fixed mammary fat pad and liver were fixed with paraffin, cut at 5 μm, and stained with haematoxylin and eosin using the established procedure. Briefly, the sectioned tissue on the glass slides were subjected to deparaffinization using xylene for 2 minutes, rehydration using various percentages of graded ethanol (100%, 90%, 80%, and 70%) each for 2 minutes, left under tap water flow for 2 minutes, stained with Harris’ haematoxylin (Sigma Aldrich, St. Louis, Missouri, USA) for 11 minutes, 0.5% acid alcohol for 2 seconds, tap water for 2 minutes, 0.2% ammonia water (Merck, Kenilworth, New Jersey, USA) for 20 seconds, and tap water for 2 minutes. Then, the sample was counter-stained with eosin (Sigma Aldrich, St. Louis, Missouri, USA) for 3 minutes. The dehydration process was started by using graded ethanol (70%, 80%, 90%, and 100%) for 10 seconds followed by clearing using xylene for 2 minutes. The slides were then mounted using DPX mountant (Sigma Aldrich, St. Louis, Missouri, USA) and the staining was visualized using an Olympus CX31 light microscope (Olympus, Shinjuku City, Tokyo, Japan).

Measurement of IFN-γ, IL-4, IL-10, and IL-17 in mouse serum from blood that was collected from the submandibular veins at the initiation, and before the end, of the experiments. Bloods were centrifuged at 2000 x g for 10 minutes to isolate serum from other blood components. IFN-γ, IL-4, IL-10, and IL-17 secretion was evaluated by using mouse magnetic luminex assay (R&D Systems, McKinley Place, Minneapolis, Minnesota, USA) and analyzed on the Luminex® 100 (Luminex Corp., Austin, Texas, USA).

The collected mammary fat pad and liver were cut removed and kept in RNAlater® solution (Ambion™). Approximately 30 mg of the mammary fat pad and the liver were transferred in a lysing matrix D tube (MP Biomedicals, Irvine, CA, USA). TRIzol was added into the tube and FASTPrep®-24 (MP Biomedicals, Irvine, CA, USA) was used to homogenize the tissues. The homogenates were transferred into a new tube and 200 μl of chloroform was added. The cap of the sample tubes was closed tightly, and tubes were vortexed firmly for 15 seconds. Then, the sample tubes were incubated for 2 to 3 minutes at room temperature. Sample tubes were centrifuged at 12,000 x g for 15 minutes at 4 °C to produce four layers: a colorless upper aqueous phase, interphase, phenol-chloroform phase, and lower red. The upper aqueous phase composed of RNA was relocated carefully into the new assigned tube. Isopropyl alcohol (0.5 ml) was added into the tube containing the aqueous phase and the tube was placed at room temperature for 10 minutes. Following incubation, that tube was centrifuged at 12,000 x g for 10 minutes at 4 °C. The supernatants were discarded, and the RNA pellets were mixed with 1 ml 75% ethanol followed by spinning at 7,500 x g for 5 minutes at 4 °C. The RNA pellets were air-dried for 5–10 minutes and were eluted into 30 μl of RNAase free water. Subsequently, the tetro cDNA synthesis kit (Bioline, London, UK) was utilized to reverse-transcribe 2 μg of total RNA. Briefly, the reaction mixture tube containing total RNA, ribosafe RNA inhibitor, oligo (dT) 18, 5x RT buffer 10 mM dNTP, tetro reverse transcriptase, and DEPC-treated water, totaling 20 μl, was prepared. The sample tubes were vortexed firmly and incubated at 45 °C for 30 minutes. The reactions were terminated by incubating at 85 °C for 5 minutes and placed on ice.

A SensiFAST SYBR Hi-Rox Kit (Bioline, London, UK) was used to quantify relative mRNA expression levels on a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The reaction mixtures containing 10 μl of 2x SensiFAST SYBR® Hi-ROX Mix, 0.8 μl of 10 μM primer pair mixture, 7.4 μL H₂O and 1 μl of cDNA were prepared and subsequent qRT-PCR was carried out as follows: a denaturing step at 95 °C for 2 min, then 95 °C for 10 s (40 cycles) and finally 60 °C for 30 s. The sequences of the primers for target genes were T-bet: 5’-CCTGGACCCAACTGTCAACT-3’ (F) 5’-AACTGTGTTCCCGAGGTGTC-3’ (R), RORγt: 5’TGCAAGACTCATCGACAAGG-3’(F) 5’-AGGGGATTCAACATCAGTGC-3’ (R), Foxp3: 5’-CAACCTAGCCCCAAGATGAA-3’ (F) 5’-CCAGATGTTGTGGGTGAGTG-3’ (R), and GATA-3: 5’-CCGAAACCGGAAGATGTCTA-3’ (F) 5’-AGGGCTCTGCCTCTCTAACC-3’ (R). ΔΔCt method was used to calculate the relative mRNA expression.

GraphPad Prism Version 8.0 software was used to carried out statistical analysis. Values of the data were presented as mean ± standard deviation (SD) from three mice per treatment group and were analyzed using one-way analysis of variance (ANOVA) followed by the Bonferroni post-hoc test. The expressions of IFN-γ, IL-4, IL-10, and IL-17 in the serum of blood were analyzed by two-way ANOVA followed by Bonferroni post hoc test.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

We examined expanded CD8⁺ T cells by performing basic phenotyping for CD3⁺CD8⁺ cells using flow cytometry analysis. Figure 1A shows that the percentage of CD3⁺CD8⁺ T cells was 74.6%. The CD8⁺ T cell growth from two different donors showed consistent shape as indicated by microscopy (Figure 1 B). After 1 and 2 hours of culture, the microscopy showed round-shaped and non-adherent cell characteristics (Figure 2 B). However, the generated moDCs appeared as single cells or loosely adherent aggregates (Figure 2 B). The generated moDCs were comprised of 82.9% CD3^-CD11c⁺ population and CD3^-CD83⁺ population. MDA-MB-231 cell lysate-activated moDCs pulsed CD8⁺ T cells at a ratio of 1 in 20 stimulated CD8⁺ T cell proliferation (Figure 3).

After approximately 2 months, tumor-bearing NSG mice developed primary metastatic breast cancer, as the anatomy of the mammary fat pad and the liver following haematoxylin and eosin staining showed the presence of numerous macroscopic metastatic foci in untreated and treated tumor-bearing NSG mice (Figure 4). The shape and size of the mammary fat pad and the liver did not show any clear differences between untreated and treated tumor-bearing NSG mice, with the size of both organs being 1.6 – 2.0 mm in all cases. No significant reduction of tumor necrosis in the mammary fat pad was seen between tumor-bearing NSG mice treated with CD8⁺ and tumor-activated moDCs pulsed CD8⁺ T cells and untreated tumor-bearing NSG mice (Figure 4 A). However, tumor-bearing NSG mice treated with CD8⁺ T cells and tumor-activated moDCs pulsed CD8⁺ T cells significantly reduced tumor necrosis in the liver compared to untreated tumor-bearing NSG mice (p < 0.05 and p < 0.01, respectively) (Figure 4 B).

We assessed the expression of IFN-γ, IL-4, IL-10, and IL-17 in the serum of blood collected from the submandibular vein of untreated and treated tumor-bearing NSG mice to investigate their potential association with tumor progression. Our results indicated that IFN-γ, IL-4, IL-10, and IL-17 secretion in the blood of untreated and treated tumor-bearing NSG mice did not show any significant differences at experiment initiation and after termination.

We also evaluated the expression of Th1, Th2, Treg, and Th17 related transcription factor (TF) genes in the mammary fat pads and livers of untreated and treated tumor-bearing NSG mice to associate these with tumor necrosis. The mRNA level of Th1-specific TF, T-bet, was increased in the mammary fat pad and the liver of tumor-bearing NSG mice treated with CD8⁺ T cells and tumor-activated moDCs pulsed CD8⁺ T cells (p < 0.001) compared to untreated tumor-bearing NSG mice. Interestingly, tumor-bearing NSG mice treated with CD8⁺ T cells and tumor-activated moDCs pulsed CD8⁺ T cells had reduced expression of Th2-specific TF, GATA-3, and Treg-specific TF, Foxp3, in the mammary fat pad and the liver (p < 0.001) compared to untreated tumor-bearing NSG mice. However, only the livers of tumor-bearing NSG mice treated with CD8⁺ T cells and tumor-activated moDCs pulsed CD8⁺ T cells showed a reduction of Th17-specific TF, RORγt (p < 0.001) compared to untreated tumor-bearing NSG mice.