Growth conditions for human breast cancer MDA-MB-231 and BT-474 cell lines, and murine NIH3T3 fibroblasts (American Type Culture Collection, ATCC, Manassas, VA) were previously reported . The murine TNBC 4T1 cells (ATCC) were grown in Roswell Park Memorial Institute-1640 medium (RPMI-1640, Sigma-Aldrich, Milan, Italy) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Sigma-Aldrich), in 95% air/5% CO2 atmosphere at 37°C.
Human peripheral blood mononuclear cells (hPBMCs) were isolated and grown as previously described [38, 39]. Mouse lymphocytes were isolated from mouse spleen and grown in R10 medium consisting of RPMI-1640 medium, supplemented with 10% heat-inactivated FBS, 50 U/ml penicillin, 50 µg/ml streptomycin, 2 nM L-glutamine, 10 mM HEPES and 50mM β-mercaptoethanol.
Aptamers and monoclonal antibodies
The sequences of the 2'Fluoro-pyrimidines (2'F-Py) RNA PDGFRβ Gint4.T and scrambled (Scr) aptamer, used as negative control, were previously reported . Unlabeled and FAM-labeled aptamers were synthesized by TriLink Biotechnologies (San Diego, CA, USA). 5'-biotinylated Gint4.T and Scr were synthesized by LGC Biosearch Technologies (Risskov Denmark). The handling protocols for aptamers, prior to each treatment, were previously described .
Anti-human PD-L1 10_12 mAb, anti-mouse PD-L1 mAb (clone 10F.9G2, BioXcell) and unrelated IgG, used as negative control, were previously reported .
Binding of Gint4.T aptamer to PDGFRβ-positive murine cells
Binding affinity (Kd value) calculation
Binding of Gint4.T to 4T1 cells was assessed by streptavidin-biotin-based assay, as previously described . Briefly, 4T1 cells (2.0×104 cells/well in clear round bottom 96-well plate) were incubated for 10 min at room temperature (RT) with increasing concentrations of 5'-biotinylated Gint4.T or Scr aptamers (10 nM, 20 nM, 50 nM, 100 nM, 200 nM and 500 nM), diluted in the binding buffer (BB) consisting of BlockAid™ blocking solution (Invitrogen, Carlsbad, CA, USA) with 1 mg/ml yeast tRNA and 1 mg/ml ultrapure™ salmon sperm DNA (Invitrogen), as nonspecific competitors. The binding affinity (Kd value) was calculated as previously reported , by using Scr to determine the nonspecific binding.
To visualize Gint4.T on the surface of PDGFRβ-positive murine cells, 4T1 and NIH3T3 (1.0×105 cells/well in 24-well), previously seeded on a coverslip for 24 h, were incubated with FAM-labeled Gint4.T or FAM-labeled Scr (500 nM-final concentration in BB) for 10 min at RT. PDGFRβ-negative BT-474 cells were treated in the same condition and used as negative control. After three washes in Dulbecco’s phosphate-buffered saline (DPBS), cells were fixed with 4% paraformaldehyde in DPBS for 20 min, washed three times in DPBS and incubated with 1.5 μM 4’,6-Diamidino- 2-phenylindole (DAPI, D9542, Sigma-Aldrich). Finally, coverslips were mounted with glycerol/DPBS. The fluorescence images were taken under a Zeiss LSM 700 META confocal microscopy equipped with a Plan-Apochromat 63x/1.4 Oil DIC objective.
Inhibition of murine cells migration by Gint4.T aptamer
4T1 and NIH3T3 cells were serum starved overnight in the presence of 500 nM Gint4.T or Scr. Then, cells (4T1, 5×104/well; NIH3T3, 2×105/well) were seeded into the upper chamber of a 24-well transwell (Transwell filters 8 μm pore size; Corning Incorporate, Corning, NY) in the presence of 500 nM Gint4.T or Scr and exposed to medium containing PDGF-BB (50 ng/ml, R&D Systems, Minneapolis, MN), as inducer of migration. After incubation for 24 h at 37°C in a humidified incubator in 5% CO2, the migrated cells were visualized by staining with 0.1% crystal violet in 25% methanol and photographed. Stained cells were lysed in 1% sodium dodecyl sulfate and absorbance at 595 nm was measured on a microplate reader.
PDGF-BB stimulation of murine cells
NIH3T3 cells (1.5×105 cells/well in 6-well) were mock-treated or serum-starved for 18 h and then left untreated or stimulated for 10 min with 20 ng/mL PDGF-BB. Cell lysates preparation and immunoblotting analyses with anti-phospho-PDGFRβ (Tyr771, indicated as p-PDGFRβ) (Cell Signaling Technology Inc.) primary antibodies were performed as previously reported  by using anti-vinculin (N-19) (Santa Cruz Biotechnology, Santa Cruz, CA) as loading control. Blots shown are representative of at least three independent experiments.
Cell growth inhibition by Gint4.T aptamer/anti-PDL1 mAb combined treatment
MDA-MB-231 (5.0×103 cells/well) and 4T1 (3.0×103 cells/well) were plated in 96-well and, after 16 h at 37 °C, were either untreated or treated for 96 h with 200 nM Gint4.T and 100 nM human anti-PD-L1 10_12 (MDA-MB-231) or murine anti-mPD-L1 (4T1), used alone or in combination. Unrelated IgG (100 nM) and Scr (200 nM) were used as negative controls. The treatment with the aptamers was renewed after 72 h. Cell counts were measured by the trypan blue exclusion test.
Effects of combinatorial treatments on co-cultures of tumor cells and lymphocytes
MDA-MB-231 (1.5×104 cells/well) or 4T1 (1.5×104 cells/well) cells, previously seeded in 96-well flat-bottom plates for 16 h at 37 °C, were co-cultured with human or murine lymphocytes, respectively, at effector:target cells ratio 10:1, in the absence or presence of 200 nM Gint4.T or 100 nM anti-PD-L1 10_12 (human setting) or anti-mPD-L1 (mouse setting), used alone or in combination. Unrelated IgG (100 nM) and Scr (200 nM) were used as negative controls. After 24 h incubation at 37 °C in a humidified incubator in 5% CO2, lymphocytes were removed and adherent cells were washed and counted by the trypan blue exclusion test.
For determination of tumor cell lysis, the release of lactate dehydrogenase (LDH) in the cellular co-culture supernatants was measured by a LDH detection kit (Thermofisher Scientific, Meridian Rd., Rockoford, IL, USA), as previously described [38, 39].
The concentration of interleukin-2 (IL-2) or interferon gamma (IFN-γ) cytokines secreted in the cellular co-cultures supernatant was measured by ELISA assays (DuoSet ELISA, R&D Systems, Minneapolis, MN, USA), as previously described [38, 39].
Binding of Gint4.T aptamer to human and mouse lymphocytes
Binding of 5'-biotinylated Gint4.T to human or mouse activated lymphocytes was assessed as previously described  by using increasing concentrations (50 nM, 100 nM and 200 nM) of 5'-biotinylated Gint4.T or Scr aptamers.
In vivo experiments
4T1 cells (3×104) were re-suspended in 0.1 ml of 1:1 mix of physiological saline and Matrigel (BD Biosciences, Franklin Lakes, NJ) and orthotopically injected into the mammary fat pads of five-week-old Female Balb/c mice, which weighed about 20-22 g (Charles River, Milan, Italy). Once tumors became approximately 150 mm3 [volume = 0.5 × long diameter× (short diameter)2], mice were randomized into four groups (five mice for each group): Ctrl (1400 pmol Scr/intravenous injection, at day 0, 2, 4, 7 and 9); Gint4.T aptamer (1400 pmol Gint4.T/intravenous injection, at day 0, 2, 4, 7 and 9); anti-mPD-L1 (200 µg/intraperitoneal injection, at day 0, 4 and 9); Gint4.T plus anti-mPD-L1 (1400 pmol Gint4.T/intravenous injection at day 0, 2, 4, 7 and 9 plus 200 µg anti-mPD-L1/intraperitoneal injection, at day 0, 4 and 9). The long and short diameters of the tumors were measured using slide calipers up to 12 days after first treatment and the body weight was also measured.
Ex vivo analyses
After sacrificing mice, tumors from each animal were excised and cut into two pieces for sample processing: one piece was stored in 10% neutral buffered formalin for immunohistochemistry analyses, and the other piece was frozen, using liquid nitrogen, for RNA extraction and reverse transcription quantitative polymerase chain reaction (RT-qPCR) and protein lysis for immunoblotting analyses. Lung from each animal were harvested and stored in 10% neutral buffered formalin for immunohistochemistry analyses.
RNA extraction and RT-qPCR were performed as previously described  on tumors from four animals per group. Primers used were: IL-2, Fwd 5′-TTGTCGTCCTTGTCAACAGC-3′, Rev 5′- CTGGGGAGTTTCAGGTTCCT-3′; IFN-γ, Fwd, 5'-AGCGGCTGACTGAACTCAGATTGTAG-3', Rev 5'-GTCACAGTTTTCAGCTGTATAGGG-3'. The following primers were used for internal control: Glucose-6-phosphate dehydrogenase, Fwd 5′-ttatcatcatgggtgcatcg-3′, Rev 5′-GCATAGCCCACAATGAAGGT-3′.
Immunoblotting analyses on tumor lysates
Tumor lysates preparation and immunoblotting analyses were performed as previously reported . Filters were probed with the indicated primary antibodies: anti-PDGFRβ, anti-phospho-44/42 MAPK (extracellular signal-regulated kinase 1/2, ERK1/2, indicated as p-ERK1/2), anti-phospho-Akt (Ser473, indicated as p-Akt) (Cell Signaling Technology Inc.), anti-PD-L1/CD274 (Proteintech Group, Inc.), and anti-vinculin (N-19) (Santa Cruz Biotechnology, Santa Cruz, CA). Blots shown are representative of at least three independent experiments.
Formalin-fixed tumors and lungs were paraffin-embedded and sectioned (4 μm) and three samples per group were stained with hematoxylin and eosin (H&E) or immunostained as reported . The primary antibodies used were: anti-CD8 alpha antibody (clone D4W2Z, #98941S, dilution 1:500, 1 h 4°C incubation, Cell Signaling Technology Inc.); anti-FoxP3 (clone D608R, #12653S, dilution 1:150, 1 h 4°C incubation, Cell Signaling); anti-Granzyme B (GRZB ab4059; dilution 1:150 diluted, 4°C RT, Abcam, Cambridge, MA); anti-Ki-67 (dilution 1:75, overnight 4°C incubation, Invitrogen); anti-PDGFRβ (diluition 1:50, 1 h 4°C incubation, Cell signaling Technology Inc.); anti-PD-L1 (diluition 1:50, 1 h 4°C incubation, Proteintech Group).
Results were interpreted using Olympus BX43 light microscope (Olympus, Center Valley, PA). Each slide was reviewed blinded and, to ensure accuracy, the number of metastases and of Ki-67, CD8, GRZB and Foxp3-positive cells was determined by two independent counts.
Statistical values were defined using GraphPad Prism version 8.00 by unpaired t-test (two variables) or one-way ANOVA followed by Tukey's multiple comparison test (more than two variables). P value < 0.05 was considered significant for all analyses.