Myeloid cells play a key role in tumor progression and metastasis by providing nourishment and immune protection, as well as facilitating cancer invasion and seeding to distal sites. Although advances have been made in understanding the biology of these tumor-educated myeloid cells (TEMCs), their intrinsic plasticity challenges our further understanding of their biology. Indeed, in vitro experiments only mimic the in vivo setting, and current gene-knockout technologies do not allow the simultaneous, temporally controlled, and cell-specific silencing of multiple genes or pathways. In this article, we describe the 4PD nanoplatform, which allows the in vivo preferential transfection and in vivo tracking of TEMCs with the desired RNAs. This platform is based on the conjugation of CD124/IL-4Rα–targeting peptide with G5 PAMAM dendrimers as the loading surface and can convey therapeutic or experimental RNAs of interest. When injected i.v. in mice bearing CT26 colon carcinoma or B16 melanoma, the 4PD nanoparticles predominantly accumulate at the tumor site, transfecting intratumoral myeloid cells. The use of 4PD to deliver a combination of STAT3- and C/EBPβ-specific short hairpin RNA or miR-142-3p confirmed the importance of these genes and microRNAs in TEMC biology and indicates that silencing of both genes is necessary to increase the efficacy of immune interventions. Thus, the 4PD nanoparticle can rapidly and cost effectively modulate and assess the in vivo function of microRNAs and mRNAs in TEMCs.

This article is featured in In This Issue, p.3763

The aberrant differentiation and accumulation of myeloid cells is one of the major hallmarks of cancer (1). Tumor-educated myeloid cells (TEMCs) include inflammatory monocytes, M2-type macrophages, tumor-associated macrophages (TAMs), granulocytic and monocytic myeloid-derived suppressor cells (MDSCs), tolerogenic dendritic cells, and other myeloid cells with a protumoral function. TEMCs are the most abundant innate immune cells present in mouse and human cancers (2, 3), and their presence correlates with increased vascular density, greater metastatic spread, and worse clinical outcomes (4). Although a consensus has not been reached with regard to the definition of each subset, it is clear that TEMCs promote tumor progression by inhibiting tumor immunity (57), as well as by enhancing tumor angiogenesis (8, 9) and metastasis (1, 1012). Only recently has the identification of important microRNAs and key intracellular signaling factors started to define the fine molecular pathways that modulate TEMC differentiation and protumoral activity (1316). However, the intrinsic plasticity of myeloid cells and the difficulties encountered in replicating the tumor micro- and macroenvironment in vitro drastically limit progress in elucidating TEMC biology. For example, in vitro experiments testing MDSC suppressive activity often do not replicate the in vivo phenotype (1719), in part due to many technical limitations, such as the uncharacterized composition of FCS, the plastic variability, oxygen distribution, absence of unknown stimuli, and likely many others (17). Thus, the role of a gene of interest needs to be confirmed in vivo, typically thorough the use of tissue-specific knockout mice, transgenic mice, or bone marrow chimeras. The generation of these models is time consuming and expensive and makes the exploration of possible synergy between pathways, the systematic functional screening of microRNAs, and the evaluation of a particular gene in a precise phase of tumor development especially challenging. Indeed, current technologies (including CRISPR-CAS9) allow the simultaneous knockdown of up to three genes with only 20% of efficiency (20).

RNA interference can be a valuable alternative to the use of bone marrow chimera and knockout mice. MicroRNA, short hairpin RNA (shRNA), and small interfering RNA are commonly used in vitro to evaluate the importance of a gene or pathway. Unfortunately, the delivery of RNA interference in vivo is restricted by the toxicity of common transfection reagents, the lack of optimal targeting moieties, the limited endosomal escape of endocytosed RNA (21, 22), and concerns related to the use of viral vectors (i.e., genome integration, engagement of TLRs). Various nonviral delivery vectors (22, 23) have been designed to target different subsets of myeloid cells in vivo through the use of physical properties (24) (i.e., nanoparticle size, enhanced permeability and retention [EPR] effect), the delivery method (i.e., dermal electroporation), or the functionalization with specific targeting agents [i.e., mannose (25), CpG (26), or MHC class II peptides (27, 28)]. However, most of the developed nanoparticles target only terminally differentiated subsets, whereas immature cells are predominant in tumor bearing hosts (1).

Among different nanoparticles, PAMAM dendrimers are of particular interest for nucleic acid delivery. These biodegradable nanoparticles are synthesized with precise dimensions and are characterized by a roughly spherical shape and a positive charge at physiological pH. Their structure allows for the easy attachment of functional groups (i.e., peptides) to their surface (29). Importantly, negatively charged nucleic acids can be easily loaded on the dendrimer surface where, by electrostatic interaction, they assume a complexed structure that protects them from extracellular nucleases (30). The resulting dendrimer–nucleic acid complexes have a neutral or modestly positive charge, allowing for efficient cell transfection (30): upon internalization in the endosome, possibly because of the lower pH and competition with endosomal anions (31), the nucleic acids are released by the dendrimer into the cytoplasm through a mechanism known as protein sponge endosomal escape (31, 32).

Based on our previous experience with peptide-functionalized G5 PAMAM dendrimer (27), we describe a new nanoparticle called 4PD that allows the loading of nucleic acids on its surface in <5 min, as well as the efficient in vivo transfection of myeloid cells in tumor-bearing hosts through the specific binding to CD124/IL-4Rα.

The (acetyl)-LQRLFRAFR[Abu]LD[Ahx]C(amide) IL-4Rα–targeting peptide was conjugated to G5 PAMAM dendrimer by maleidoamide chemistry. Briefly, G5 PAMAM dendrimers were reacted with eight excess of m-maleimidobenzoyl-N-hydoxysuccinimide ester to generate NHS–maleimide dendrimers. These activated dendrimers were reacted with four excess of the targeting peptide to generate 4PD. It is important to note that, in this case, the maleidoamide ester–cysteine ester also serves as an additional spacer between the dendrimer surface and the targeting peptide.

shRNA and 2′F-shRNA were synthetized in vitro via PCR and T7 RNA polymerase. Briefly, dsDNA templates were generated by fill-in PCR reaction (90°C for 10 min, 60°C for 5 min, and 72°C for 20 min) using recombinant Taq DNA polymerase (Invitrogen), dNTP (10 mM), and the following oligonucleotides (50 μM): STAT3F 5′-TAATACGGACTCACTATAAGGAGGGTGTCAGATCACATGGGCTTTCAAGAGAAGC-3′; STAT3R 5′-AAGAGGGTGTCAGATCACATGGGCTTCTCTTGAAAGCC-3′; ScrambleF 5′-TAATACGACTCACTATAAGGGCAGTAGCATGGTCCGTTGAGATTCAAGAGATCT-3′; ScrambleR 5′-AAGGCAGTAGCATGGTCCGTTGAGATCTCTTGAATC-3′; C/EBPβF 5′-TAATACGACTCACTATAGGGCGGCGACTTCCTCTCCGACCTCTTCTTCAAGAGAGAAGAGG-3′; and C/EBPβR 5′-AAAAACGACTTCCTCTCCGACCTCTTCTCTCTTGAAGAAG-3′. DNA templates were purified using a QIAquick PCR Purification Kit (QIAGEN) and transcribed using a DuraScribe T7 RNA synthesis kit (Epicentre) in a 7-h reaction with 2′-fluorine–modified or unmodified pyrimidines. The reaction products were treated with DNase, purified via NH4AC-based precipitation, and resuspended in ultra-pure water. miR-142-3p (Duplex) 5′-UGUAGUGUUUCCUACUUUAUGGA-3′ and 5′-CAUAAAGUAGAAAGCACUACU-3′ were purchased from Boston Open Labs.

shRNA and the nanoparticles were admixed by adding shRNA or miR-142-3p solution in a drop-wise manner to the nanoparticle mixture while vortexing. The amine/phosphate ratio used was 10:1, unless otherwise indicated. The complex was incubated at room temperature for 5 min, followed by the addition of cardiolipin (0.2 μg of cardiolipin [Sigma-Aldrich] for 1 μg of shRNA) and 1/10 volume of NaCl (9%). In some experiments, BrUTP (Sigma-Aldrich) was admixed with the shRNA and the 4PD at a 10:1 amine/phosphate ratio.

A total of 2 × 105 MSC2 or bone marrow cells was transfected with 30 pmol of shRNA complexed to the nanoparticle by a 1-h incubation at 37°C in Opti-MEM (Invitrogen). Cells were washed twice with complete media and plated into six-well plates for subsequent analysis.

Total RNA was extracted by TRIzol Reagent (Invitrogen), according to the manufacturer’s instructions. cDNA from purified total RNA was produced with SuperScript III Reverse Transcriptase (Invitrogen) and was used in PCR (Invitrogen) and real-time PCR. PCR was performed at 95°C for 1 min, 60°C for 30 s, and 72°C for 30 s, for a total of 30 cycles, on a GS482 Thermal Cycler (G-Storm). Real-time PCR was performed using the TaqMan Gene Expression Assay (Applied Biosystems).

STAT3 was detected using a STAT3 ELISA kit (eBioscience), following the manufacturer’s instructions. The absorbance values were normalized on total protein concentration, as determined by a Coomassie Plus (Bradford) Assay Kit (Pierce).

BALB/c CT26 colon carcinoma (33), BALB/c 4T1 mammary carcinoma (34), C57BL/6 B16 melanoma (35), C57BL/6 MCA203 fibrosarcoma (36), and the immortalized BALB/c MDSC cell line MSC2 (37) were described previously. They were cultured in complete medium (RPMI 1640 medium [Invitrogen] supplemented with 2 mmol/l l-glutamine, 10 mmol/l HEPES, 20 μmol/l 2-ME, 150 μ/ml streptomycin, 200 μ/ml penicillin, and 10% heat-inactivated FBS [Invitrogen]). Mixed lymphocyte tumor cultures (MLTCs) were performed in U-bottom 96-well plates by incubating 2.5 × 103 gamma-irradiated (3000 rad) CT26 tumor cells with 106 Thy1.1+/+ splenocytes and 105 CFSE+ T cells isolated with the mouse Pan T Cell Isolation Kit (Miltenyi Biotec) from Thy1.2 tumor–bearing BALB/c mice.

Suppressive assays.

A total of 105 magnetically purified CD11b+ cells was cultured with 106 CFSE-labeled splenocytes from naive clone-4 mice and stimulated with hemagglutinin (HA)512–520-specific or irrelevant peptide (1 μM) for 3 d. In other experiments, magnetically purified CD11b+ cells were cultured with 7.5 × 105 CFSE-labeled BALB/c splenocytes stimulated with 7.5 × 105 gamma-irradiated splenocytes from C57BL/6 mice. Cells were counterstained with LIVE/DEAD Aqua vital dye and Abs against CD3 and CD8 molecules, and CD8+ T cell proliferation was evaluated by flow cytometry.

Ficoll-purified, deidentified PBMCs from patients with head and neck squamous cell carcinoma (HNSCC) of the oral cavity, who underwent definitive surgical resection in our previously described clinical trial (38), were frozen and used in this study, in accordance with the original Institutional Review Board protocol and informed consent. Briefly, PBMCs were thawed, and 5 × 105 cells were incubated with different amounts (range 15 nM–100 μM) of 4PD loaded with Alexa Fluor (AF)555-conjugated shRNA (BLOCK-iT Alexa Fluor Red Fluorescent Control; Invitrogen) in 500 μl of complete media for 20 min at 4°C. Cells were washed with PBS, counterstained with vital dye and with Abs against IL-4Rα, CD11b, CD14, and HLA-DR, and analyzed by FACS.

The following Abs were used for flow cytometry analysis: allophycocyanin- or Brilliant Violet (BV)711–conjugated rat anti-mouse CD11b (clone M1/70; BD), PerCp-Cy5.5–conjugated rat anti-mouse Ly6G and Ly6C (clone RB6-8C5; BioLegend), allophycocyanin-Cy7–conjugated rat anti-mouse Ly6G (clone 1-A8), V450-conjugated rat anti-mouse Ly6C (clone AL-21), PE-conjugated rat anti-mouse CD124 (clone mIL4R-M1), PE-Cy7–conjugated rat anti-mouse F4/80 (clone BM8; all from BD), FITC-conjugated rat anti-mouse F4/80 (clone A3-1; AbD Serotec), BV650-conjugated rat anti-mouse CD206 (clone C068C2; BioLegend), PE-Cy7–conjugated hamster anti-mouse CD11c (clone HL3; BD), eFluor 450–conjugated rat anti-mouse CD49b (clone DX5; eBioscience), PE-conjugated mouse anti-mouse I-A[d] (clone AMS-32.1), allophycocyanin-conjugated hamster anti-mouse CD80 (clone 16-10A1), FITC-conjugated rat anti-mouse CD86 (clone GL1), allophycocyanin-Cy7–conjugated rat anti-mouse CD4 (clone GK1.5), PE-conjugated rat anti-mouse CD25 (clone 3C7), PerCP-conjugated hamster anti-mouse CD3 (clone 145-2C11; all from BD), PerCP–eFluor 710–conjugated rat anti-mouse CD3 (clone 17A2; eBioscience), allophycocyanin-Cy7–conjugated rat anti-mouse CD19 (clone ID3; BD), allophycocyanin rat anti-mouse Foxp3 (clone FJK-16s; eBioscience), Pacific Blue– or BV711-conjugated rat anti-mouse CD8 (clone 53-6.7; BD), allophycocyanin-conjugated mouse anti BrdU (clone Bu20a; eBioscience), FITC-conjugated mouse anti-human CD33 (clone HIM3-4), allophycocyanin-H7–conjugated mouse anti-human CD14 (clone MϕP9), Pacific Blue–conjugated mouse anti-human CD11b (clone ICRF44; all from BD), allophycocyanin-conjugated mouse anti-human IL-4Rα (clone 25463; R&D Systems), and BV711-conjugated mouse anti-human HLA-DR (clone L243; BioLegend). p-STAT6 AF647 Ab (clone J71-773.58.11; BD) was used with Phosflow Perm Buffer IV, as per the manufacturer’s instructions. Live/dead fixable dead cell stain (Invitrogen) or DAPI (Sigma-Aldrich) were used to exclude dead cells.

Samples were analyzed on an LSRFortessa-HTR flow cytometer (Becton-Dickinson) or on a CytoFLEX (Beckman Coulter), and data were analyzed using FCS express software (De Novo Software). Sorting was performed on a MoFlo Astrios cell sorter (Beckman Coulter).

All mouse experiments were performed in accordance with protocols approved by the University of Miami Institutional Animal Care and Use Committee. Eight- to twelve-week-old female BALB/c and C57BL/6 mice were purchased from Harlan. Clone-4 mice transgenic for the H-2 Kd–restricted TCR recognizing the influenza virus HA peptide (HA512–520) were purchased from the Jackson Laboratory. For in vivo experiments, mice were ear tagged, injected s.c. into the right flank with 0.5 × 106 tumor cells (CT26 and B16) or orthotopically in the third mammary fat pad with 2 × 105 tumor cells (4T1), and randomized before beginning treatment. Tumor size was measured three times a week with a caliper and is reported as the product of the largest diameter and the perpendicular diameter. Mice were euthanized when the tumor reached 100 mm2, unless otherwise specified. shRNA treatment was performed three times a week by injecting 400 μl of shRNA (50 pm/g in 0.9% NaCl) loaded onto 4PD i.v., unless otherwise specified. Treatment began when the tumor reached 25 mm2, unless otherwise indicated. All of the experiments were performed at least twice with five mice per group, unless indicated otherwise in the figures.

Dermal electroporation.

Isoflurane-anesthetized mice were injected intradermally with a total of 50 μg of pcDNA3-gp 70 in the left and right thigh regions. Immediately following the injection, the area was electroporated using a DermaVax (Cellectis) equipped with IDE-4-4-2 electrode needle rows (4-mm gap). A Pulse Agile electroporation protocol, consisting of 10 rectangular wave pulses (one pulse, 450 V, 50 μs duration, 0.2-ms pulse interval plus one pulse, 450 V, 50 μs duration, 50-ms pulse interval, plus eight pulses, 110 V, 10 ms duration, and 20-ms pulse interval), was used. The mice were kept normothermic by using a heating pad until fully recovered from anesthesia.

In vivo imaging system.

CT26 tumor–bearing mice were injected i.v. with AF750-conjugated control shRNA (MWG, 1 μg/g) complexed on nanoparticles. After 2-6-12 and 24 h, the mice were anesthetized with isoflurane and imaged with an ultra-sensitive CCD camera using an In Vivo Imaging System (IVIS; PerkinElmer). Adoptive cell therapy (ACT) was performed in C57BL/6 mice after s.c. challenge with 1 × 106 MCA203 cells, as previously described (39, 40). Nine days later, when the tumor size was ∼20 mm2, mice were treated i.v. with 5 × 106 mouse telomerase reverse transcriptase (mTERT)-specific CTLs or β-galactosidase (β-gal)–specific CTLs; control mice were not treated with CTLs. At the time of CTL transfer, mice were injected i.m. with 5 × 108 PFU the recombinant adenoviral vector coding for the relevant Ag recognized by transferred T cells and then treated with recombinant human IL-2 (30,000 IU), given i.p. twice a day, for three consecutive days. 4PD loaded with miR-142-3p (700 pmol/kg) or control RNA was given the day before and 2 d after ACT. Tumors were measured in a blind manner using a digital caliper.

A Malvern Zetasizer Nano ZS instrument with a back scattering detector (173°, 633-nm laser wavelength) was used to measure the hydrodynamic size (diameter) at 25°C in a low-volume quartz cuvette (path length 10 mm). Twelve measurements per sample were made, and the hydrodynamic size was calculated, using the manufacturer-inbuilt software, as the intensity-weighted average.

One-way ANOVA was performed between groups after normality evaluation by the Kolmogorov–Smirnov test. A pairwise post hoc analysis was performed using the Holm–Sidak test or the Dunn test. ANOVA on ranks was used if the sample was not normally distributed. The Student t test was used to compare two groups. The Kaplan–Meier log-rank test, followed by Holm–Sidak post hoc analysis, was used to evaluate differences in survival between groups.

To target TEMCs in vivo, we opted to decorate G5 PAMAM dendrimers with a peptide targeting IL-4Rα (Fig. 1A). IL-4Rα is upregulated early during myeloid cell differentiation (Supplemental Fig. 1A) in the presence of tumor-conditioned media. This chain is part of type 1 and type 2 IL-4Rs and is characterized by rapid internalization upon binding of the relevant ligands (41). Furthermore, this receptor is a functional marker of MDSCs, TAMs, and tumor-associated monocytes (4245). To design a synthetic ligand that can bind IL-4Rα and targets nanoparticles to TEMCs, we took advantage of previous studies evaluating the interaction between mouse and human IL-4 and their cognate receptors (Fig. 1B) (4648). These studies indicate that regions 79–86 of mouse IL-4 and 77–89 of human IL-4 are important (Fig. 1B); in particular, the arginines in positions 81, 85, and 88 of human IL-4 (47) and the arginines in positions 80, 83, and 86 of mouse IL-4 (46) are involved in cluster 2 and 3 domain formation of the receptor–ligand complex. Surrounding amino acids further facilitate the binding and the correct spacing between these key arginine residues, suggesting that aa 79–86 are the core of the mouse IL-4 binding site (46). Crystallography data show that, upon binding with the cognate receptor, this domain is buried inside a deep pocket of the IL-4Rα receptor, suggesting that a linker is needed to connect the binding domain to the nanocarrier (46). The mouse IL-4 sequence 78–89 (78-LQRLFRAFRCLD-89) was modified in C87 into the synthetic amino acid Abu, to mimic human IL-4 and to facilitate conjugation with the PAMAM dendrimer. Additionally, the hydrophobic spacer aminohexanoic acid (Ahx), followed by a cysteine (C), was added to the N terminus for the subsequent attachment of the peptide to the dendrimer. The amino and C termini were amidated and acetylated, respectively, to neutralize the negative and positive charges. This peptide was unable to inhibit IL-4–mediated STAT6 phosphorylation in the immortalized MDSC line MSC2 (37) (Fig. 1C) or to trigger CD124 signaling (data not shown).

FIGURE 1.

Physical characteristics of the 4PD platform. (A) Schematic diagram of 4PD. An average of two IL-4–derived, IL-4Rα–targeting peptides per G5 PAMAM dendrimer were conjugated via maleidoamide chemistry to generate the 4PD nanoplatform. Therapeutic RNA can be loaded via electrostatic interaction. (B) Alignment of human and mouse IL-4 with the targeting peptide. The arginines that interact with CD124 are in bold. (C) MSC2 cells (1 × 106) were washed, rested in Opti-MEM (1 h at 37°C), and stimulated with IL-4 (741 pM) for 2 h at 37°C in the presence of the indicated relative amount of IL-4Rα–targeting peptide. p-STAT6 was evaluated by flow cytometry. (D) HPLC-purified 4PD was evaluated by MALDI-TOF spectrometry. (E) The size of the shRNA/4PD complexes generated in water or saline was determined by DLS. (F) Different amounts of AF555-shRNA/4PD complexes or the IL-4Rα–targeting FITC–peptide, as control, were incubated for 10 min at 4°C with a fixed amount of Epoxy beads decorated with rCD124. After two washes in PBS, binding and affinity were evaluated by FACS. Data were derived from three independent experiments.

FIGURE 1.

Physical characteristics of the 4PD platform. (A) Schematic diagram of 4PD. An average of two IL-4–derived, IL-4Rα–targeting peptides per G5 PAMAM dendrimer were conjugated via maleidoamide chemistry to generate the 4PD nanoplatform. Therapeutic RNA can be loaded via electrostatic interaction. (B) Alignment of human and mouse IL-4 with the targeting peptide. The arginines that interact with CD124 are in bold. (C) MSC2 cells (1 × 106) were washed, rested in Opti-MEM (1 h at 37°C), and stimulated with IL-4 (741 pM) for 2 h at 37°C in the presence of the indicated relative amount of IL-4Rα–targeting peptide. p-STAT6 was evaluated by flow cytometry. (D) HPLC-purified 4PD was evaluated by MALDI-TOF spectrometry. (E) The size of the shRNA/4PD complexes generated in water or saline was determined by DLS. (F) Different amounts of AF555-shRNA/4PD complexes or the IL-4Rα–targeting FITC–peptide, as control, were incubated for 10 min at 4°C with a fixed amount of Epoxy beads decorated with rCD124. After two washes in PBS, binding and affinity were evaluated by FACS. Data were derived from three independent experiments.

Close modal

The resulting (acetyl)-LQRLFRAFR[Abu]LD[Ahx]C(amide) peptide was conjugated through the aminus terminus to G5 PAMAM dendrimer by maleidoamide chemistry, and products were purified by HPLC, as previously described (27). Mass spectrometry analysis (Fig. 1D) of three independently synthetized lots reveals a good purity product with a molecular mass of 32,706 Da, indicating that an average of two peptides (molecular mass 1831.2 Da) are conjugated to each PAMAM dendrimer (molecular mass 28,876 Da). Because 4PD are meant to be used preferentially with shRNA, STAT3-specific shRNA was complexed with 4PD by admixing the two solutions, as previously described (27). The size of the shRNA/4PD complexes generated in water and in saline was evaluated by dynamic light scattering (DLS) and was found to be <200 nm in diameter (Fig. 1E), an ideal size for the EPR effect (49).

To verify whether 4PD can bind to IL-4Rα, 4PD was loaded with AF555-labeled shRNA (AF555-shRNA) and incubated with Epoxy beads conjugated with rIL-4Rα or with an irrelevant protein as negative control. FITC-labeled unconjugated targeted peptide was used as positive control. Binding was evaluated by flow cytometry. As expected, the 4PD/AF555-shRNA complexes showed good affinity for IL-4Rα, with only background binding to the negative control (Fig. 1F). Of note, Kd of the 4PD/RNA complexes was significantly lower than for free peptide, indicating that the multivalency of the 4PD/RNA complexes significantly increases the affinity of this platform for the cognate receptor.

Once 4PD was proven to recognize the rIL-4Rα receptor, we tested whether this complex could also recognize the receptor naturally expressed on MDSCs. To this end, splenic CD11b+ cells magnetically isolated from CT26 tumor–bearing mice were incubated with AF555–shRNA loaded on 4PD or control dendrimer functionalized with a mixture of randomly synthetized irrelevant peptides; fluorescence was then evaluated by fluorescence microscopy (Fig. 2A) and FACS (Fig. 2B). Although negligible fluorescence was detected in the cells treated with the control dendrimer or with shRNA alone, almost all MDSCs were AF555+. Because the targeting peptide was designed to cross-react with human IL-4Rα, PBMCs from patients with HNSCC were incubated with different concentrations of 4PD loaded with AF555–shRNA. Afterward, cells were counterstained with Abs specific for CD33, IL-4Rα, CD14, CD11b, and HLA-DR and evaluated by FACS. Although the control dendrimer did not show any significant binding, 4PD recognized the majority of IL4Rα+ cells, in particular the CD33+IL4Rα+ MDSCs (Fig. 2C–E, Supplemental Fig. 2), which we recently showed to be the main suppressive population in patients with HNSCC (38). Taken together, these data indicate that 4PD can recognize rIL-4Rα, as well as human and mouse IL4Rα+ cells.

FIGURE 2.

4PD recognizes mouse and human MDSCs. Splenic CD11b+ MDSCs isolated from CT26 tumor–bearing mice were transfected with AF555-shRNA alone and loaded onto control dendrimer or onto 4PD. Forty-five minutes later, cells were analyzed by immunofluorescence microscopy (A) and by FACS (B) after gating on live (DAPI) cells. One experiment representative of the other two is depicted in (A). In (B), data were derived from three independent experiments. (C) A fixed amount of PBMCs from patients (n = 3) with HNSCC were incubated (20 min at 4°C in complete media) with different amounts of AF555-shRNA/4PD complexes. Cells were washed with PBS and counterstained with vital dye and Abs specific for CD33, CD14, IL-4Rα, HLA-DR, and CD11b. Transfection was evaluated by FACS by gating on live cells (C) or on the indicated populations (D). (E) Kd was calculated after gating on CD33+IL4Rα+ cells.

FIGURE 2.

4PD recognizes mouse and human MDSCs. Splenic CD11b+ MDSCs isolated from CT26 tumor–bearing mice were transfected with AF555-shRNA alone and loaded onto control dendrimer or onto 4PD. Forty-five minutes later, cells were analyzed by immunofluorescence microscopy (A) and by FACS (B) after gating on live (DAPI) cells. One experiment representative of the other two is depicted in (A). In (B), data were derived from three independent experiments. (C) A fixed amount of PBMCs from patients (n = 3) with HNSCC were incubated (20 min at 4°C in complete media) with different amounts of AF555-shRNA/4PD complexes. Cells were washed with PBS and counterstained with vital dye and Abs specific for CD33, CD14, IL-4Rα, HLA-DR, and CD11b. Transfection was evaluated by FACS by gating on live cells (C) or on the indicated populations (D). (E) Kd was calculated after gating on CD33+IL4Rα+ cells.

Close modal

Although PAMAM dendrimers are generally considered nonimmunogenic (50), some early reports indicated that these polymers might activate the complement system and induce a significant immune reaction (51). Considering that the use of functionalized dendrimer for gene silencing requires multiple administrations, it was important to exclude the possibility that neutralizing Abs could be induced. Briefly, BALB/c mice were injected i.v. every other day for a total of seven injections with 4PD loaded with a scrambled shRNA. Two weeks after the last injection, sera from 4PD/shRNA-treated or naive mice were collected and incubated with Epoxy beads conjugated with 4PD, 4PD/scrambled-shRNA complex, 4PD loaded with STAT3 shRNA, or PADRE–peptide functionalized dendrimer (PPD) (27). After five washes with PBS, bound Abs were evaluated by flow cytometry (Fig. 3A). Although only baseline fluorescence was found on the beads treated with sera from naive mice, a positivity was found on all of the targets incubated with the sera from 4PD-treated mice. Because signal was detected on 4PD- and PPD-loaded Epoxy beads, regardless of the presence of shRNA, these data strongly suggest that neutralizing Abs were generated against the PAMAM dendrimer core. Considering that the surface charges on PAMAM dendrimers may mediate polymer immunogenicity by nonspecifically activating membrane proteins (51, 52), we evaluated whether the use of amphipathic molecules may reduce the generation of anti-PAMAM dendrimer Abs by masking dendrimer’s residual charges. Briefly, the amphipathic molecule cardiolipin was added to the 4PD/shRNA formulation, and the above-described experiments were repeated. Although 4PD/shRNA complex administration confirmed its immunogenicity, the addition of this amphipathic molecule to the complexes completely abolished the generation of 4PD-reactive Abs (Fig. 3B). DLS analysis of 4PD/shRNA/cardiolipin complexes indicated the effective binding of cardiolipin to the nanoparticle, with only a slight increase in the complex’s size (average diameter 170 ± 20 nm). Binding experiments using rIL-4Rα indicated that the affinity for the cognate receptor was not affected by cardiolipin inclusion (data not shown). Because mitochondrial lipids (i.e., β-glucan) may have an effect on myeloid cell polarization (53, 54), we decided to evaluate whether 4PD or 4PD conjugated with cardiolipin could influence MDSC differentiation and suppressive activity using the same concentration as in Fig. 2. Bone marrow cells from naive mice were differentiated into MDSCs with CT26 tumor–conditioned media (55) in the presence of 4PD/control-shRNA or 4PD/control-RNA/cardiolipin or with no nanoparticles. Four days later, leukocyte composition and recovery were analyzed by flow cytometry (Fig. 3C), and magnetically purified CD11b+ cells were tested in a suppressive assay (Fig. 3D). No significant differences were found in the number of recovered leukocyte subsets or changes in the suppressive activity of isolated CD11b+ cells (Fig. 3C, 3D). In summary, these data indicate that the use of amphipathic molecules strongly reduces dendrimer immunogenicity without altering 4PD specificity. Thus, this formulation was used in all of the subsequent in vitro and in vivo experiments.

FIGURE 3.

Cardiolipin reduces dendrimer immunogenicity. (A) BALB/c mice were treated i.v. seven times (three times a week) with PBS or 4PD (14 mg/kg) loaded with scrambled shRNA. Fifteen days after the last inoculation, serum was tested for the presence of reactive Ab against Epoxy beads covalently linked with 4PD, 4PD/scrambled-shRNA, 4PD/STAT3-shRNA, or PPD/scrambled-shRNA. Epoxy beads were washed with PBS and incubated for 10 min with FITC-conjugated, rabbit anti-mouse Ab. Fluorescence was evaluated via FACS. (B) The experiment described in (A) was repeated using 4PD/shRNA or 4PD/shRNA/cardiolipin complexes, and the presence of reactive Abs was evaluated using 4PD linked to Epoxy beads. (C and D) A total of 2 × 105 freshly isolated bone marrow cells was transfected (30 min at 37°C) in Opti-MEM with 0.3 μg of control shRNA/4PD or control shRNA/4PD/cardiolipin and cultured for 4 d with 30% CT26 tumor-conditioned medium. Data were derived from three independent experiments, each performed with sera pooled from three treated mice. Recovery of the different leukocyte subsets was evaluate by flow cytometry (C), and their suppressive function was tested in an MLR (D). Data in (C) and (D) were derived from two independent experiments. pa, ANOVA p value; pt, t test p value.

FIGURE 3.

Cardiolipin reduces dendrimer immunogenicity. (A) BALB/c mice were treated i.v. seven times (three times a week) with PBS or 4PD (14 mg/kg) loaded with scrambled shRNA. Fifteen days after the last inoculation, serum was tested for the presence of reactive Ab against Epoxy beads covalently linked with 4PD, 4PD/scrambled-shRNA, 4PD/STAT3-shRNA, or PPD/scrambled-shRNA. Epoxy beads were washed with PBS and incubated for 10 min with FITC-conjugated, rabbit anti-mouse Ab. Fluorescence was evaluated via FACS. (B) The experiment described in (A) was repeated using 4PD/shRNA or 4PD/shRNA/cardiolipin complexes, and the presence of reactive Abs was evaluated using 4PD linked to Epoxy beads. (C and D) A total of 2 × 105 freshly isolated bone marrow cells was transfected (30 min at 37°C) in Opti-MEM with 0.3 μg of control shRNA/4PD or control shRNA/4PD/cardiolipin and cultured for 4 d with 30% CT26 tumor-conditioned medium. Data were derived from three independent experiments, each performed with sera pooled from three treated mice. Recovery of the different leukocyte subsets was evaluate by flow cytometry (C), and their suppressive function was tested in an MLR (D). Data in (C) and (D) were derived from two independent experiments. pa, ANOVA p value; pt, t test p value.

Close modal

To determine 4PD-mediated shRNA pharmacokinetics and biodistribution in vivo, AF750-conjugated shRNA was loaded onto 4PD and injected i.v. into CT26 tumor–bearing mice. Fluorescence emission in the liver, bladder, tumor region, and contralateral cutaneous site was quantified by an IVIS at 2, 6, 12, and 24 h after injection (Fig. 4A). As expected with the G5 PAMAM dendrimer (56), this analysis suggests rapid accumulation throughout the body, rapid kidney clearance, and liver detention. Interestingly, RNA signal in the tumor seems to be retained longer, suggesting uptake from the tumor microenvironment. To better evaluate 4PD-mediated shRNA biodistribution, AF555-shRNA was loaded onto 4PD and injected i.v. into BALB/c mice bearing the CT26 colon carcinoma (Fig. 4B) or C57BL/6 mice bearing the B16 melanoma (Supplemental Fig. 3A). As control, mice were injected with AF555–shRNA loaded onto the control dendrimer. Biodistribution and 4PD targeting selectivity were evaluated by FACS 2 h later, using the gating strategy depicted in Supplemental Fig. 2. This analysis revealed that 4PD preferentially binds to granulocytic and monocytic MDSCs, as well as to macrophages (Fig. 4B, Supplemental Fig. 3A), in the tumor, and, to a lesser extent, in the spleen of both strains of mice. On the contrary, no appreciable differences from control were detected in the liver of CT26 cancer–bearing BALB/c mice, whereas, in the liver of B16 melanoma–bearing C57BL/6 mice, a selectivity toward MDSCs and macrophages was still detectable. Interestingly, liver monocytic MDSCs (mMDSCs) from BALB/c mice seem to be characterized by greater uptake of the control dendrimer compared with the livers from C57BL/6 mice. This can be explained by strain-specific differences in type 1 and type 2 polarization that seem to influence nanoparticle clearance (57).

FIGURE 4.

In vivo 4PD recognizes MDSCs and macrophages. (A) CT26 tumor–bearing (25 mm2) BALB/c mice were injected with AF750-conjugated shRNA/4PD, and the fluorescence from tumor, contralateral cutaneous, liver, and bladder regions was analyzed at different time points using an IVIS (n = 3 per group). (B) BALB/c mice (n = 5) bearing the CT26 colon carcinoma (0.7 cm in diameter) were injected with AF555-shRNA loaded onto 4PD (black bars) or, as control, onto G5 PAMAM dendrimer functionalized with a random peptide (CTRL DR; gray bars). Mice were sacrificed 2 h later. Single-cell suspensions from spleen, tumor, and liver were labeled with Abs specific for Gr1, CD11b, F4/80, CD11c, CD3, CD4, and CD8 to identify gMDSCs, mMDSCs, macrophages, dendritic cells, and helper and cytotoxic T cells using the gating strategy depicted in Supplemental Fig. 3. Fluorescence in the AF555 channel was evaluated by flow cytometry after gating in each population. (C) CT26 tumor–bearing BALB/c mice (n = 5) were injected with the indicated amounts of the AF555-shRNA/4PD complex when the tumor reached 25 mm2. Two hours later, spleen and tumor were labeled with Abs against CD11b, Ly6G, Ly6C, F4/80, CD19, CD3, and CD49b and analyzed by flow cytometry.

FIGURE 4.

In vivo 4PD recognizes MDSCs and macrophages. (A) CT26 tumor–bearing (25 mm2) BALB/c mice were injected with AF750-conjugated shRNA/4PD, and the fluorescence from tumor, contralateral cutaneous, liver, and bladder regions was analyzed at different time points using an IVIS (n = 3 per group). (B) BALB/c mice (n = 5) bearing the CT26 colon carcinoma (0.7 cm in diameter) were injected with AF555-shRNA loaded onto 4PD (black bars) or, as control, onto G5 PAMAM dendrimer functionalized with a random peptide (CTRL DR; gray bars). Mice were sacrificed 2 h later. Single-cell suspensions from spleen, tumor, and liver were labeled with Abs specific for Gr1, CD11b, F4/80, CD11c, CD3, CD4, and CD8 to identify gMDSCs, mMDSCs, macrophages, dendritic cells, and helper and cytotoxic T cells using the gating strategy depicted in Supplemental Fig. 3. Fluorescence in the AF555 channel was evaluated by flow cytometry after gating in each population. (C) CT26 tumor–bearing BALB/c mice (n = 5) were injected with the indicated amounts of the AF555-shRNA/4PD complex when the tumor reached 25 mm2. Two hours later, spleen and tumor were labeled with Abs against CD11b, Ly6G, Ly6C, F4/80, CD19, CD3, and CD49b and analyzed by flow cytometry.

Close modal

Similar experiments performed on IL4Rα+/+ and IL4Rα−/− BALB/c mice bearing the 4T1 mammary carcinoma revealed that, in this metastatic model, 4PD accumulation in the tumor, lung, and spleen is mediated by IL-4Rα gene expression, as well as confirmed the selectivity of this platform for macrophages and MDSCs (Supplemental Fig. 3B). To evaluate whether increasing the dose could increase the number of transfected TEMCs without affecting the specificity, different doses of AF555-shRNA loaded onto 4PD were injected i.v. into CT26 tumor–bearing mice, and specificity was evaluated by flow cytometry 2 h later in the spleens and tumors (Fig. 4C). No significant differences were found between groups injected with the different doses, suggesting an early saturation of the receptors. However, although the 16.6 pmol/g dose (∼7 μg per mouse) showed greater variation, the 150 pmol/g dose (∼60 μg per mouse) caused an increase in splenocyte binding. Taken together, these experiments indicate that 4PD can recognize macrophages and MDSCs in different strains and tumor models and may serve as a useful platform for TEMC-targeted shRNA delivery.

We next evaluated whether 4PD could be used to deliver functionally active shRNA. To this aim, STAT3-specific shRNA (58) was produced using unmodified or 2′F-pyrimidines; the fluorinated nucleotides confer RNase resistance, limit the interaction with TLRs, and were suggested to prolong gene silencing (59). STAT3-specific shRNA was loaded onto 4PD and used to transfect the IL4Rα+ immortalized MDSC line MSC2 (37). As expected, 4PD-mediated delivery of unmodified and fluorinated shRNA effectively reduced STAT3 expression (Fig. 5A); however, fluorinated shRNA outperformed the unmodified shRNA (Fig. 5A) and was effective in reducing STAT3 protein for up to 4 d following transfection (Fig. 5B). Similar results were obtained with splenic CD11b+ cells isolated from 4T1 tumor–bearing mice (data not shown). Because of the superiority of fluorinated shRNA in mediating gene silencing, this nucleic acid composition was used in subsequent experiments. To evaluate whether 4PD could mediated the delivery of functional shRNA to MDSCs in vivo and affect myeloid cell function, BALB/c mice bearing the CT26 colon carcinoma (0.5 cm in diameter) were injected i.v. with 4PD loaded with STAT3-specific or scrambled shRNA at 9, 12, 15, and 18 d after challenge. As an additional control, mice were injected with irrelevant dendrimers loaded with STAT3 shRNA. The day after the last inoculation, mice were sacrificed, and CD11b+ cells were magnetically isolated from the spleen and the tumor. In vivo, 4PD treatment with the relevant shRNA reduced STAT3 expression, as measured by quantitative RT-PCR (qRT-PCR), by ∼10 and 50% in splenic- and tumor-associated MDSCs, respectively (data not shown). No changes were observed in the groups treated with scrambled shRNA or with STAT3-specific shRNA loaded on irrelevant dendrimer (data not shown). Because STAT3 was shown to play a key role in MDSC biology, the suppressive activity of myeloid cells isolated from the tumor of treated mice was tested. Briefly, magnetically purified CD11b+ cells were cultured with CFSE-labeled HA-specific CD8+ T cells stimulated with the relevant peptide. CD8+ T cell proliferation was evaluated 3 d later by FACS. Tumor-associated MDSCs from mice treated with 4PD loaded with control shRNA significantly inhibited T cell proliferation (Fig. 5C). Similar results were obtained when tumor-infiltrating CD11b+ cells from mice treated with STAT3-specific shRNA loaded onto control dendrimers were used. Conversely, a strong reduction in MDSC suppressive activity was observed when CD11b+ cells from mice treated with STAT3-specific shRNA loaded on 4PD were used in the suppression assay (Fig. 5C). The suppressive activity of splenic MDSCs, which was present, but reduced, compared with the tumor counterpart, was not modulated by any treatment (data not shown). These data suggest that 4PD-mediated STAT3-specific shRNA delivery can significantly inhibit MDSC-dependent immunosuppression at the tumor site.

FIGURE 5.

4PD delivers functional shRNA in vitro and in vivo. (A) The immortalized cell line MSC2 was transfected with 4PD loaded with STAT3-specific shRNA (normal or fluorinated) or with irrelevant shRNA. After 48 h, STAT3 expression was evaluated by RT-PCR. Each circle represents one independent experiment. (B) Mitomycin-treated MSC2 cells were transfected with 4PD loaded with STAT3-specific shRNA, and STAT3 protein expression was evaluated by ELISA 3 and 4 d later. (C) BALB/c mice were challenged with the CT26 cell line and treated 13 d later with 4PD loaded with STAT3 or scrambled shRNA, every other day, for 9 d. Additional control mice were injected with control dendrimer loaded with STAT3 shRNA. Twenty-four hours after the last inoculation, mice were sacrificed, and tumor-infiltrating CD11b+ cells were magnetically isolated. CD11b+ cells’ suppressive activity was tested against CFSE-labeled HA-specific CD8+ T cells stimulated with the relevant peptide. CD8+ T cell proliferation was evaluated 3 d later by FACS. Data were derived from three independent experiments in which MDSCs from each group (five mice per group) were pooled. (D) CT26 tumor–bearing (25 mm2) mice (n = 3 per experiment) were treated three times a week i.v. with 4PD loaded with C/EBPβ-specific shRNA, STAT3-specific shRNA, or with both shRNAs. As negative control, mice were treated with 4PD loaded with irrelevant shRNA. Nine days later, mice were sacrificed, CD11b+ cells were isolated, and the expression of STAT3 and C/EBPβ was evaluated by qRT-PCR. Data are derived from three experiments in which CD11b+ cells from each group were pooled. (E) CT26 tumor–bearing (0.5 cm in diameter) mice (n = 5 per experiment) were treated three times a week i.v. with 4PD loaded with both C/EBPβ- and STAT3-specific shRNA or with irrelevant shRNA. In the last two injections, AF555 was added to the nanoparticles. Two hours after the last inoculation, mice were sacrificed, and the tumors were pooled and disaggregated into a single-cell suspension. The tumor cells were labeled with anti-CD11b Ab and counterstained with the vital dye. AF555 fluorescence was used to discriminate and sort the transfected and nontransfected CD11b+ cells. Real-time PCR was subsequently performed to determine the expression of C/EBPβ and STAT3 in the sorted populations. Data were derived from two experiments in which tumor cells from each group were pooled.

FIGURE 5.

4PD delivers functional shRNA in vitro and in vivo. (A) The immortalized cell line MSC2 was transfected with 4PD loaded with STAT3-specific shRNA (normal or fluorinated) or with irrelevant shRNA. After 48 h, STAT3 expression was evaluated by RT-PCR. Each circle represents one independent experiment. (B) Mitomycin-treated MSC2 cells were transfected with 4PD loaded with STAT3-specific shRNA, and STAT3 protein expression was evaluated by ELISA 3 and 4 d later. (C) BALB/c mice were challenged with the CT26 cell line and treated 13 d later with 4PD loaded with STAT3 or scrambled shRNA, every other day, for 9 d. Additional control mice were injected with control dendrimer loaded with STAT3 shRNA. Twenty-four hours after the last inoculation, mice were sacrificed, and tumor-infiltrating CD11b+ cells were magnetically isolated. CD11b+ cells’ suppressive activity was tested against CFSE-labeled HA-specific CD8+ T cells stimulated with the relevant peptide. CD8+ T cell proliferation was evaluated 3 d later by FACS. Data were derived from three independent experiments in which MDSCs from each group (five mice per group) were pooled. (D) CT26 tumor–bearing (25 mm2) mice (n = 3 per experiment) were treated three times a week i.v. with 4PD loaded with C/EBPβ-specific shRNA, STAT3-specific shRNA, or with both shRNAs. As negative control, mice were treated with 4PD loaded with irrelevant shRNA. Nine days later, mice were sacrificed, CD11b+ cells were isolated, and the expression of STAT3 and C/EBPβ was evaluated by qRT-PCR. Data are derived from three experiments in which CD11b+ cells from each group were pooled. (E) CT26 tumor–bearing (0.5 cm in diameter) mice (n = 5 per experiment) were treated three times a week i.v. with 4PD loaded with both C/EBPβ- and STAT3-specific shRNA or with irrelevant shRNA. In the last two injections, AF555 was added to the nanoparticles. Two hours after the last inoculation, mice were sacrificed, and the tumors were pooled and disaggregated into a single-cell suspension. The tumor cells were labeled with anti-CD11b Ab and counterstained with the vital dye. AF555 fluorescence was used to discriminate and sort the transfected and nontransfected CD11b+ cells. Real-time PCR was subsequently performed to determine the expression of C/EBPβ and STAT3 in the sorted populations. Data were derived from two experiments in which tumor cells from each group were pooled.

Close modal

To evaluate whether multiple genes could be simultaneously silenced in vivo, BALB/c mice were challenged with the CT26 colon carcinoma. When the tumor reached 0.5 cm in diameter, mice were treated three times a week with 4PD loaded with shRNAs specific for STAT3 and/or C/EBPβ (60), two genes that have been shown to be important for MDSC function (58, 6167). Nine days later, mice were sacrificed, and CD11b+ cells were magnetically isolated from the spleen and the tumor. Although a modest and nonsignificant inhibition of gene expression was found in the spleen (Fig. 5D), STAT3 and C/EBPβ were downregulated ∼50% in the tumor, as assessed by qRT-PCR, whereas no effect was seen when scrambled shRNA was used (Fig. 5D). To determine whether the 50% decrease in the expression of STAT3 and C/EBPβ was the result of the presence of the shRNAs only in a fraction of cells, the experiment was repeated by treating CT26 tumor–bearing mice with 4PD/STAT3–C/EBPβ (three times a week for four times) and with 4PD complexed with the relevant shRNAs and AF555-shRNA (to detect transfected cells) twice before analysis. Twenty-four hours after the last administration, mice were sacrificed, and transfected (CD11b+AF555+) or nontransfected (CD11b+AF555) cells were sorted by FACS and analyzed by quantitative PCR (Fig. 5E). As expected, only transfected cells showed a significant reduction ∼ 70 and 60% for STAT3 and C/EBPβ transcripts, respectively. Interestingly, silencing of STAT3 or C/EBPβ induces a transcriptional regulation in the other gene, confirming their reciprocal regulation (Fig. 5D) (68). Taken together, these data indicate that 4PD can recognize myeloid cells in the tumor and can also effectively mediate the delivery of functional shRNAs in vitro and in vivo.

As mentioned above, STAT3 and C/EBPβ have been proposed as relevant genes in MDSC biology, differentiation, and suppressive activity (6067, 69, 70). Having shown that 4PD can mediate silencing of these genes in tumor-infiltrating myeloid cells, we aimed to follow the fate of transfected cells. To this end, we took advantage of the possibility to load 4PD/shRNA complexes with BrUTP. Because BrUTP cannot penetrate the cell membrane unless loaded onto 4PD (data not shown), this allows the labeling of in vivo–transfected cells. Upon 4PD internalization, BrUTP is released and incorporated into nascent rRNA and mRNA and can be detected by FACS up to 2 wk after treatment (71). When tumors reached ∼0.5 cm in diameter, CT26 tumor–bearing mice were injected with 4PD loaded with BrUTP and STAT3-specific shRNA, C/EBPβ shRNA, both shRNAs, or scrambled shRNAs as control. The composition of the BrdU+ cells infiltrating the tumor was evaluated longitudinally by flow cytometry (Fig. 6A). Initially, 4PD transfected mostly tumor-infiltrating CD11b+Ly6C+ mMDSCs, supporting the data obtained with AF555-shRNA (Fig. 4). At 24 h, a strong decrease in mMDSCs and a concomitant increase in CD11b+Ly6G+ granulocytic MDSCs (gMDSCs) is observed in the control group (4PD scrambled-shRNA). In the control group, the entire transfected population decreases in the 5 d after injection. The BrU+gMDSC+ population decreases at 72 h, whereas the percentage of BrU+ TAMs with a strong M2 polarization increases. Compared with control, although STAT3 silencing alone does not modify the number of BrU+CD11b+ cells in the tumor, C/EBPβ silencing seems to delay the decrease in the transfected myeloid cells. This can be explained by the suggested proapoptotic function of C/EBPβ on myeloid cells exposed to IFN-γ (72).

FIGURE 6.

4PD-mediated in vivo silencing of STAT3 and C/EBPβ restores the efficacy of antitumor vaccines. (A) CT26 tumor–bearing (25 mm2) mice (n = 3) were injected once with 4PD loaded with BrUTP and STAT3 and/or C/EBPβ shRNA. At 2, 24, 72, and 120 h postinjection, single-cell suspensions from the tumors were labeled with Abs against CD11b, Ly6G, Ly6C, F4/80, CD206 (to identify gMDSCs, mMDSCs, TAMs, and M1/M2 TAMs), and anti BrU Ab to identify the in vivo–transfected cells (n = 3 mice per group per time point). Data were derived from two independent experiments. The table shows the ANOVA p values comparing the effect of treatment on each population at each time point. (B) CT26 tumor–bearing mice (n = 5 per group) were treated i.v. with 4PD loaded with STAT3- and/or C/EBPβ-specific shRNAs. Twenty-four hours after the last injection, T cells were magnetically sorted, CFSE labeled, and tested in MLTCs against CT26. (C, D and E) Starting 9 d after challenge, CT26 tumor–bearing mice were treated i.v. three times a week with PBS or with 4PD loaded with STAT3-specific shRNA, C/EBPβ-specific shRNA, or with both shRNAs (20 μg per mouse). At 10 and 17 d after challenge, mice were vaccinated via electroporation with pcDNA3 (D) or with gp70 encoding pcDNA3 (E). Tumor growth was monitored. *p < 0.05 versus control.

FIGURE 6.

4PD-mediated in vivo silencing of STAT3 and C/EBPβ restores the efficacy of antitumor vaccines. (A) CT26 tumor–bearing (25 mm2) mice (n = 3) were injected once with 4PD loaded with BrUTP and STAT3 and/or C/EBPβ shRNA. At 2, 24, 72, and 120 h postinjection, single-cell suspensions from the tumors were labeled with Abs against CD11b, Ly6G, Ly6C, F4/80, CD206 (to identify gMDSCs, mMDSCs, TAMs, and M1/M2 TAMs), and anti BrU Ab to identify the in vivo–transfected cells (n = 3 mice per group per time point). Data were derived from two independent experiments. The table shows the ANOVA p values comparing the effect of treatment on each population at each time point. (B) CT26 tumor–bearing mice (n = 5 per group) were treated i.v. with 4PD loaded with STAT3- and/or C/EBPβ-specific shRNAs. Twenty-four hours after the last injection, T cells were magnetically sorted, CFSE labeled, and tested in MLTCs against CT26. (C, D and E) Starting 9 d after challenge, CT26 tumor–bearing mice were treated i.v. three times a week with PBS or with 4PD loaded with STAT3-specific shRNA, C/EBPβ-specific shRNA, or with both shRNAs (20 μg per mouse). At 10 and 17 d after challenge, mice were vaccinated via electroporation with pcDNA3 (D) or with gp70 encoding pcDNA3 (E). Tumor growth was monitored. *p < 0.05 versus control.

Close modal

STAT3 and C/EBPβ silencing significantly increased the percentage of BrU+ TAMs with an M1 polarization; this effect also was detectable in the mice treated only with C/EBPβ. These data seem to confirm previous findings suggesting that mMDSCs can differentiate into gMDSCs and TAMs (69, 70). Interestingly, at 120 h after treatment, the number of TAMs with a reduced CD206+ M2 polarization is increased in the STAT3 and C/EBPβ shRNA group, suggesting that this combined treatment may modify the tumor microenvironment. It is important to emphasize that this assay evaluates the entire population of transfected cells but not the differentiation of the single cell. Thus, it cannot differentiate whether gMDSCs and TAMs differentiate directly from mMDSC or whether a transdifferentiation of gMDSCs in TAMs occurs.

To determine the effect of 4PD-mediated silencing of STAT3 and/or C/EBPβ on tumor-specific T cells, MLTCs from treated mice were performed. Briefly, CFSE-labeled splenic T cells from CT26 tumor–bearing Thy1.2 mice treated with 4PD loaded with scrambled, STAT3, C/EBPβ, or STAT3 and C/EBPβ shRNAs were magnetically purified and stimulated with gamma-irradiated CT26 cells in the presence of Thy1.1+/+ splenocytes as feeder. Five days later, proliferating (CFSElow) CD8+Thy1.2+ T cells were analyzed by CytoFLEX flow cytometry, which allows the exact enumeration of cell subsets in the proliferation wells (Fig. 6B). This analysis shows that greater numbers of T cells reactive to the tumor can be recovered from the mice treated with both inhibitors.

To determine whether the observed changes in the tumor micro- and macroenvironment modify tumor progression, BALB/c mice were treated three times a week, starting 9 d after CT26 injection, with 4PD loaded with control, STAT3, or C/EBPβ shRNA or STAT3 and C/EBPβ shRNA. In each group, mice were vaccinated via dermal electroporation with control pcDNA3 (Fig. 6A) or pcDNA3 encoding the poorly immunogenic tumor-specific gp70-derived Ag, AH1 (Fig. 6B) (73). Only the simultaneous administration of STAT3 and C/EBPβ shRNA, combined with gp70 vaccination, significantly delayed tumor growth and improved mice survival, with ∼30% of the animals completely rejecting the tumor; no therapeutic effect was observed in any other experimental group. Interestingly, a nonhealing wound developed in all of the surviving animals that, although forcing the euthanization of the animals to remain in compliance with Institutional Animal Care and Use Committee policy, may suggest an important role for MDSCs and these genes in the wound-healing process. Taken together, these data indicate that 4PD can be used to exploit therapeutic synergies between the simultaneous silencing of different genes in association with immunotherapeutic approaches.

By regulating gene expression and splicing, microRNAs play important roles in myeloid cell biology, and high-throughput screening methods have been developed to identify those that are differentially regulated in cancer. However, functional screening of these important mediators is hindered by the absence of simple methods to selectively transfect myeloid cells in vivo. We used miR-142-3p to evaluate whether 4PD could be used to functionally screen the role of microRNAs. We demonstrated previously that miR-142-3p regulated C/EBPβ and STAT3 in MDSCs, reduced their immune-suppressive properties, and restored the antitumor efficacy of adoptively transferred tumor-specific T cells (74). Indeed, delivery of miR-142-3p in BALB/c mice bearing the CT26 tumor significantly downregulates STAT3 and C/EBPβ expression in transfected cells, indicating that, in addition to the translational modulation of C/EBPβ (74), this microRNA may regulate C/EBPβ and STAT3 at the transcriptional level (Supplemental Fig. 4). C57BL/6 mice were injected with MCA203 fibrosarcoma at day 0; when the tumor size reached ∼ 20 mm2, mice were treated with CTLs specific for the tumor-associated Ag mTERT or β-gal as control. 4PD loaded with miR-142-3p or control RNA was given i.v. the day before and 3 d after ACT. As shown in Fig. 7, tumor progression was significantly delayed in mice treated with 4PD loaded with miR-142-3p and mTERT-specific T cells, whereas no effects were detected in mice treated with 4PD/miR-142-3p or mTERT ACT alone. These data suggest that 4PD can be used to deliver functional microRNAs to myeloid cells in tumor-bearing mice. Notably, these results confirm the role of miR-142-3p in myeloid cell biology that was previously proven with complex and time-consuming efforts, through the use of chimeric mice reconstituted with lentivirus-transduced hematopoietic precursors (74).

FIGURE 7.

4PD in vivo delivery of miR-142-3p mimic synergizes with passive immunotherapy. C57BL/6 mice (n = 10 per group) were injected with MCA203 fibrosarcoma at day −10. When the tumor reached ∼20 mm2 (day 0), mice were treated with CTLs specific for mTERT or CTLs specific for β-gal, the same day mice were vaccinated with adenoviral vector encoding for the peptide recognized by transferred T cells. 4PD loaded with miR-142-3p (700 pmol/kg) or control RNA was given the day before and 2 d after ACT; IL-2 was administered i.p. twice a day at days 0, 1, and 2.

FIGURE 7.

4PD in vivo delivery of miR-142-3p mimic synergizes with passive immunotherapy. C57BL/6 mice (n = 10 per group) were injected with MCA203 fibrosarcoma at day −10. When the tumor reached ∼20 mm2 (day 0), mice were treated with CTLs specific for mTERT or CTLs specific for β-gal, the same day mice were vaccinated with adenoviral vector encoding for the peptide recognized by transferred T cells. 4PD loaded with miR-142-3p (700 pmol/kg) or control RNA was given the day before and 2 d after ACT; IL-2 was administered i.p. twice a day at days 0, 1, and 2.

Close modal

Myeloid cells play a key role in tumor growth and progression. However, their intrinsic plasticity makes the results from in vitro experimentation questionable and requires the validation of any results in vivo using elaborate and time-consuming experiments and models, such as conditional knockout mice and bone marrow chimeras. By providing effective and selective delivery of the desired nucleic acids, nanoparticles and macromolecules can be an attractive and cost-effective alternative to these methods. Indeed, different nanoparticles have been proposed to target different subsets of myeloid cells in vivo through the use of physical properties or functionalized surfaces (2428). However, their use is often limited by the intrinsic immunogenicity and chemical expertise required for loading the nucleic acid of interest. For example, we have recently described a liposomal form of gemcitabine that specifically targets mMDSC but requires cargo modification and ad hoc synthesis for each drug (75). In this article, we described 4PD, a new nanoplatform based on the conjugation of the G5 PAMAM dendrimer with the IL-4–derived IL-4Rα–targeting peptide (Fig. 1). This platform allows the easy transfection of mouse and human MDSCs with minimal toxicity and increased specificity (Fig. 2, Supplemental Fig. 2). The formulation has been optimized by the addition of amphipathic molecules that greatly reduce 4PD immunogenicity, allowing for multiple administrations of the shRNA/4PD complexes, without the generation of neutralizing Abs (Fig. 3). Furthermore, multiple administrations of large doses of 4PD do not alter general blood toxicology parameters, indicating a good treatment tolerability (data not shown).

In vivo biodistribution studies using the optimized formulations indicate the preferential targeting of MDSCs and TAMs at the tumor site (Fig. 4), where functional therapeutic RNA can silence the gene(s) of interest (Fig. 5). The biodistribution and preferential selectivity for myeloid cells are most likely mediated by different factors. Indeed, although IL-4Rα has been proposed as a functional marker of myeloid cells in tumor-bearing mice and patients with cancer (38, 42, 44), this marker is not exclusively expressed in myeloid cells (76). It can be expressed by some tumors (but it is undetectable on the CT26 tumors used in this study, Supplemental Fig. 1B), and its expression is insufficient to define MDSCs and macrophages (77). Nevertheless, pharmacokinetic biodistribution and analysis of gene silencing (Figs. 4, 5) indicate a preferential shRNA uptake from myeloid cells within the tumor, with minimal transfection of other cell types (i.e., B and T cells). These results can be explained, in part, by shRNA/4PD complex sizes that can mediate the EPR effect by the enhanced receptor-mediated phagocytosis of myeloid cells compared with lymphoid populations (78). Furthermore, cardiolipin, used in this study to mask the residual positive charges in the dendrimers, was shown to bind preferentially and be taken up by MDSCs (79).

Contrary to most nanoparticles, therapeutic RNA can be loaded by electrostatic interactions on the 4PD in <5 min, by simply mixing the nanoparticles with the desired nucleic acid. This characteristic is extremely important because it allows the direct loading of one or multiple validated shRNA(s) onto the platform and the direct testing in vivo of the silencing effect of the desired gene(s). Additionally, BrUTP can be easily loaded, together with the shRNA of interest, onto the 4PD. This greatly facilitates in vivo tracking of the effect of gene silencing on the fate of transfected cells and drastically simplifies the experiments to elucidate, in vivo, the roles of genes in cell differentiation and phenotype. However, it is important to emphasize that the use of BrUTP only allows the tracking of the differentiation of the entire population of transfected cells rather than a single transfected cell, for which in vivo tracking techniques are necessary.

Using the 4PD platform, we successfully silenced up to five genes in myeloid cells infiltrating the 4T1 tumor and rapidly determined which of those were more relevant for MDSC differentiation (data not shown).

We successfully used this nanoparticle to deliver shRNA specific for STAT3 and/or C/EBPβ. We also showed that preferential simultaneous silencing of STAT3 and C/EBPβ in the tumor-infiltrating TEMCs may modify the polarization of these cells, allowing the priming of antitumor response, and synergizing with DNA vaccination against a mouse colon carcinoma. These results confirm the previous data indicating that STAT3 and C/EBPβ are important in MDSC biology (6067), seem to confirm that mMDSCs are precursors of gMDSCs and TAMs (69, 70), and suggest that silencing both genes may have an important additive/synergistic therapeutic effect.

Similarly, we tested this nanoparticle with miR-142-3p, a microRNA that we have previously shown to modulate C/EBPβ isoforms in MDSCs and consequently alter their differentiation into protumoral cells (74). By delivering miR-142-3p mimic via 4PD to MDSCs and macrophages, we confirmed the antitumoral role of the forced expression of this microRNA in the myeloid lineage, and we also demonstrated that this platform can be used for the rapid functional screening of microRNAs.

In summary, we have described a versatile and cost-effective nanoplatform that can be used to preferentially transfect myeloid cells in vivo and rapidly explore the putative roles of genes or microRNAs of interest.

We thank the flow cytometry cores of the Sylvester Comprehensive Cancer Center and the Diabetes Research Institute at the University of Miami for help with flow cytometry analysis and FACS.

This work was supported by Department of Defense Concept Award CA101118, Department of Defense Idea Award BC121157, by a Sylvester Comprehensive Cancer Center-NIH funding award, and by the Glaser Award.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ACT

adoptive cell therapy

AF

Alexa Fluor

AF555-shRNA

AF555-labeled shRNA

BV

Brilliant Violet

DLS

dynamic light scattering

EPR

enhanced permeability and retention

β-gal

β-galactosidase

gMDSC

granulocytic MDSC

HA

hemagglutinin

HNSCC

head and neck squamous cell cancer

IVIS

In Vivo Imaging System

MDSC

myeloid-derived suppressor cell

mMDSC

monocytic MDSC

MLTC

mixed lymphocyte tumor culture

mTERT

mouse telomerase reverse transcriptome

PPD

PADRE–peptide functionalized dendrimer

qRT-PCR

quantitative RT-PCR

shRNA

short hairpin RNA

TAM

tumor-associated macrophage

TEMC

tumor-educated myeloid cell.

1
Condamine
T.
,
Ramachandran
I.
,
Youn
J. I.
,
Gabrilovich
D. I.
.
2015
.
Regulation of tumor metastasis by myeloid-derived suppressor cells.
Annu. Rev. Med.
66
:
97
110
.
2
Sica
A.
,
Bronte
V.
.
2007
.
Altered macrophage differentiation and immune dysfunction in tumor development.
J. Clin. Invest.
117
:
1155
1166
.
3
Pollard
J. W.
2004
.
Tumour-educated macrophages promote tumour progression and metastasis.
Nat. Rev. Cancer
4
:
71
78
.
4
Zhang
J.
,
Patel
L.
,
Pienta
K. J.
.
2010
.
CC chemokine ligand 2 (CCL2) promotes prostate cancer tumorigenesis and metastasis.
Cytokine Growth Factor Rev.
21
:
41
48
.
5
Schmielau
J.
,
Finn
O. J.
.
2001
.
Activated granulocytes and granulocyte-derived hydrogen peroxide are the underlying mechanism of suppression of T-cell function in advanced cancer patients.
Cancer Res.
61
:
4756
4760
.
6
Nagaraj
S.
,
Gupta
K.
,
Pisarev
V.
,
Kinarsky
L.
,
Sherman
S.
,
Kang
L.
,
Herber
D. L.
,
Schneck
J.
,
Gabrilovich
D. I.
.
2007
.
Altered recognition of antigen is a mechanism of CD8+ T cell tolerance in cancer.
Nat. Med.
13
:
828
835
.
7
Serafini
P.
,
Mgebroff
S.
,
Noonan
K.
,
Borrello
I.
.
2008
.
Myeloid-derived suppressor cells promote cross-tolerance in B-cell lymphoma by expanding regulatory T cells.
Cancer Res.
68
:
5439
5449
.
8
De Palma
M.
,
Naldini
L.
.
2009
.
Tie2-expressing monocytes (TEMs): novel targets and vehicles of anticancer therapy?
Biochim. Biophys. Acta
1796
:
5
10
.
9
Schmid
M. C.
,
Varner
J. A.
.
2010
.
Myeloid cells in the tumor microenvironment: modulation of tumor angiogenesis and tumor inflammation.
J. Oncol.
2010
:
201026
.
10
Qian
B. Z.
,
Pollard
J. W.
.
2010
.
Macrophage diversity enhances tumor progression and metastasis.
Cell
141
:
39
51
.
11
Wyckoff
J. B.
,
Wang
Y.
,
Lin
E. Y.
,
Li
J. F.
,
Goswami
S.
,
Stanley
E. R.
,
Segall
J. E.
,
Pollard
J. W.
,
Condeelis
J.
.
2007
.
Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors.
Cancer Res.
67
:
2649
2656
.
12
Konstantopoulos
K.
,
Thomas
S. N.
.
2009
.
Cancer cells in transit: the vascular interactions of tumor cells.
Annu. Rev. Biomed. Eng.
11
:
177
202
.
13
El Gazzar
M.
2014
.
microRNAs as potential regulators of myeloid-derived suppressor cell expansion.
Innate Immun.
20
:
227
238
.
14
Ugel
S.
,
Delpozzo
F.
,
Desantis
G.
,
Papalini
F.
,
Simonato
F.
,
Sonda
N.
,
Zilio
S.
,
Bronte
V.
.
2009
.
Therapeutic targeting of myeloid-derived suppressor cells.
Curr. Opin. Pharmacol.
9
:
470
481
.
15
Arina
A.
,
Corrales
L.
,
Bronte
V.
.
2016
.
Enhancing T cell therapy by overcoming the immunosuppressive tumor microenvironment.
Semin. Immunol.
28
:
54
63
.
16
Condamine
T.
,
Mastio
J.
,
Gabrilovich
D. I.
.
2015
.
Transcriptional regulation of myeloid-derived suppressor cells.
J. Leukoc. Biol.
98
:
913
922
.
17
Hamilton
M. J.
,
Banáth
J. P.
,
Lam
V.
,
Lepard
N. E.
,
Krystal
G.
,
Bennewith
K. L.
.
2012
.
Serum inhibits the immunosuppressive function of myeloid-derived suppressor cells isolated from 4T1 tumor-bearing mice.
Cancer Immunol. Immunother.
61
:
643
654
.
18
Haverkamp
J. M.
,
Crist
S. A.
,
Elzey
B. D.
,
Cimen
C.
,
Ratliff
T. L.
.
2011
.
In vivo suppressive function of myeloid-derived suppressor cells is limited to the inflammatory site.
Eur. J. Immunol.
41
:
749
759
.
19
Schmidt
K.
,
Zilio
S.
,
Schmollinger
J. C.
,
Bronte
V.
,
Blankenstein
T.
,
Willimsky
G.
.
2013
.
Differently immunogenic cancers in mice induce immature myeloid cells that suppress CTL in vitro but not in vivo following transfer.
Blood
121
:
1740
1748
.
20
Wang
H.
,
Yang
H.
,
Shivalila
C. S.
,
Dawlaty
M. M.
,
Cheng
A. W.
,
Zhang
F.
,
Jaenisch
R.
.
2013
.
One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering.
Cell
153
:
910
918
.
21
Roberts
T. C.
,
Ezzat
K.
,
El Andaloussi
S.
,
Weinberg
M. S.
.
2016
.
Synthetic siRNA delivery: progress and prospects.
Methods Mol. Biol.
1364
:
291
310
.
22
Crotty
S.
,
Pipkin
M. E.
.
2015
.
In vivo RNAi screens: concepts and applications.
Trends Immunol.
36
:
315
322
.
23
Amoozgar
Z.
,
Goldberg
M. S.
.
2015
.
Targeting myeloid cells using nanoparticles to improve cancer immunotherapy.
Adv. Drug Deliv. Rev.
91
:
38
51
24
Kourtis
I. C.
,
Hirosue
S.
,
de Titta
A.
,
Kontos
S.
,
Stegmann
T.
,
Hubbell
J. A.
,
Swartz
M. A.
.
2013
.
Peripherally administered nanoparticles target monocytic myeloid cells, secondary lymphoid organs and tumors in mice.
PLoS One
8
:
e61646
.
25
Zhu
S.
,
Niu
M.
,
O’Mary
H.
,
Cui
Z.
.
2013
.
Targeting of tumor-associated macrophages made possible by PEG-sheddable, mannose-modified nanoparticles.
Mol. Pharm.
10
:
3525
3530
.
26
de Titta
A.
,
Ballester
M.
,
Julier
Z.
,
Nembrini
C.
,
Jeanbart
L.
,
van der Vlies
A. J.
,
Swartz
M. A.
,
Hubbell
J. A.
.
2013
.
Nanoparticle conjugation of CpG enhances adjuvancy for cellular immunity and memory recall at low dose.
Proc. Natl. Acad. Sci. USA
110
:
19902
19907
.
27
Daftarian
P.
,
Kaifer
A. E.
,
Li
W.
,
Blomberg
B. B.
,
Frasca
D.
,
Roth
F.
,
Chowdhury
R.
,
Berg
E. A.
,
Fishman
J. B.
,
Al Sayegh
H. A.
, et al
.
2011
.
Peptide-conjugated PAMAM dendrimer as a universal DNA vaccine platform to target antigen-presenting cells.
Cancer Res.
71
:
7452
7462
.
28
Daftarian
P. M.
,
Stone
G. W.
,
Kovalski
L.
,
Kumar
M.
,
Vosoughi
A.
,
Urbieta
M.
,
Blackwelder
P.
,
Dikici
E.
,
Serafini
P.
,
Duffort
S.
, et al
.
2013
.
A targeted and adjuvanted nanocarrier lowers the effective dose of liposomal amphotericin B and enhances adaptive immunity in murine cutaneous leishmaniasis.
J. Infect. Dis.
208
:
1914
1922
.
29
Tekade
R. K.
,
Kumar
P. V.
,
Jain
N. K.
.
2009
.
Dendrimers in oncology: an expanding horizon.
Chem. Rev.
109
:
49
87
.
30
Pietersz
G. A.
,
Tang
C. K.
,
Apostolopoulos
V.
.
2006
.
Structure and design of polycationic carriers for gene delivery.
Mini Rev. Med. Chem.
6
:
1285
1298
.
31
Waite
C. L.
,
Sparks
S. M.
,
Uhrich
K. E.
,
Roth
C. M.
.
2009
.
Acetylation of PAMAM dendrimers for cellular delivery of siRNA.
BMC Biotechnol.
9
:
38
.
32
Liang, W. and J. K. Lam. 2012. Endosomal escape pathways for non-viral nucleic acid delivery systems. In Molecular Regulation of Endocytosis. B. Ceresa, ed. InTech. Available at: http://www.intechopen.com/books/molecular-regulation-of-endocytosis/endosomal-escape-pathways-for-non-viral-nucleic-acid-delivery-systems
33
Griswold
D. P.
,
Corbett
T. H.
.
1975
.
A colon tumor model for anticancer agent evaluation.
Cancer
36
(
6
Suppl.
):
2441
2444
.
34
Aslakson
C. J.
,
Miller
F. R.
.
1992
.
Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor.
Cancer Res.
52
:
1399
1405
.
35
Fidler
I. J.
1975
.
Biological behavior of malignant melanoma cells correlated to their survival in vivo.
Cancer Res.
35
:
218
224
.
36
Spiess
P. J.
,
Yang
J. C.
,
Rosenberg
S. A.
.
1987
.
In vivo antitumor activity of tumor-infiltrating lymphocytes expanded in recombinant interleukin-2.
J. Natl. Cancer Inst.
79
:
1067
1075
.
37
Apolloni
E.
,
Bronte
V.
,
Mazzoni
A.
,
Serafini
P.
,
Cabrelle
A.
,
Segal
D. M.
,
Young
H. A.
,
Zanovello
P.
.
2000
.
Immortalized myeloid suppressor cells trigger apoptosis in antigen-activated T lymphocytes.
J. Immunol.
165
:
6723
6730
.
38
Weed
D. T.
,
Vella
J. L.
,
Reis
I. M.
,
De la Fuente
A. C.
,
Gomez
C.
,
Sargi
Z.
,
Nazarian
R.
,
Califano
J.
,
Borrello
I.
,
Serafini
P.
.
2015
.
Tadalafil reduces myeloid-derived suppressor cells and regulatory T cells and promotes tumor immunity in patients with head and neck squamous cell carcinoma.
Clin. Cancer Res.
21
:
39
48
.
39
Mennuni
C.
,
Ugel
S.
,
Mori
F.
,
Cipriani
B.
,
Iezzi
M.
,
Pannellini
T.
,
Lazzaro
D.
,
Ciliberto
G.
,
La Monica
N.
,
Zanovello
P.
, et al
.
2008
.
Preventive vaccination with telomerase controls tumor growth in genetically engineered and carcinogen-induced mouse models of cancer.
Cancer Res.
68
:
9865
9874
.
40
Ugel
S.
,
Zoso
A.
,
De Santo
C.
,
Li
Y.
,
Marigo
I.
,
Zanovello
P.
,
Scarselli
E.
,
Cipriani
B.
,
Oelke
M.
,
Schneck
J. P.
,
Bronte
V.
.
2009
.
In vivo administration of artificial antigen-presenting cells activates low-avidity T cells for treatment of cancer.
Cancer Res.
69
:
9376
9384
.
41
Galizzi
J. P.
,
Zuber
C. E.
,
Cabrillat
H.
,
Djossou
O.
,
Banchereau
J.
.
1989
.
Internalization of human interleukin 4 and transient down-regulation of its receptor in the CD23-inducible Jijoye cells.
J. Biol. Chem.
264
:
6984
6989
.
42
Roth
F.
,
De La Fuente
A. C.
,
Vella
J. L.
,
Zoso
A.
,
Inverardi
L.
,
Serafini
P.
.
2012
.
Aptamer-mediated blockade of IL4Rα triggers apoptosis of MDSCs and limits tumor progression.
Cancer Res.
72
:
1373
1383
.
43
Serafini
P.
,
Bronte
V.
.
2008
.
Myeloid-derived suppressor cells in cancer
. In
Tumor-Induced Immune Suppression Mechanisms and Therapeutic Reversal.
Gabrilovich
D. I.
,
Hurwitz
A. A.
, eds.
Springer
,
New York
, p.
157
195
.
44
Gallina
G.
,
Dolcetti
L.
,
Serafini
P.
,
De Santo
C.
,
Marigo
I.
,
Colombo
M. P.
,
Basso
G.
,
Brombacher
F.
,
Borrello
I.
,
Zanovello
P.
, et al
.
2006
.
Tumors induce a subset of inflammatory monocytes with immunosuppressive activity on CD8+ T cells.
J. Clin. Invest.
116
:
2777
2790
.
45
Bronte
V.
,
Serafini
P.
,
De Santo
C.
,
Marigo
I.
,
Tosello
V.
,
Mazzoni
A.
,
Segal
D. M.
,
Staib
C.
,
Lowel
M.
,
Sutter
G.
, et al
.
2003
.
IL-4-induced arginase 1 suppresses alloreactive T cells in tumor-bearing mice.
J. Immunol.
170
:
270
278
.
46
Yao
G.
,
Chen
W.
,
Luo
H.
,
Jiang
Q.
,
Xia
Z.
,
Zang
L.
,
Zuo
J.
,
Wei
X.
,
Chen
Z.
,
Shen
X.
, et al
.
2006
.
Identification of core functional region of murine IL-4 using peptide phage display and molecular modeling.
Int. Immunol.
18
:
19
29
.
47
Zhang
J. L.
,
Simeonowa
I.
,
Wang
Y.
,
Sebald
W.
.
2002
.
The high-affinity interaction of human IL-4 and the receptor alpha chain is constituted by two independent binding clusters.
J. Mol. Biol.
315
:
399
407
.
48
Hage
T.
,
Sebald
W.
,
Reinemer
P.
.
1999
.
Crystal structure of the interleukin-4/receptor alpha chain complex reveals a mosaic binding interface.
Cell
97
:
271
281
.
49
Kobayashi
H.
,
Watanabe
R.
,
Choyke
P. L.
.
2013
.
Improving conventional enhanced permeability and retention (EPR) effects; what is the appropriate target?
Theranostics
4
:
81
89
.
50
Kesharwani
P.
,
Banerjee
S.
,
Gupta
U.
,
Mohd Amin
M. C.
,
Padhye
S.
,
Sarkar
F. H.
,
Iyer
A. K.
.
2015
.
PAMAM dendrimers as promising nanocarriers for RNAi therapeutics.
Mater. Today
18
:
565
572
.
51
Plank
C.
,
Mechtler
K.
,
Szoka
F. C.
 Jr.
,
Wagner
E.
.
1996
.
Activation of the complement system by synthetic DNA complexes: a potential barrier for intravenous gene delivery.
Hum. Gene Ther.
7
:
1437
1446
.
52
Zolnik
B. S.
,
González-Fernández
A.
,
Sadrieh
N.
,
Dobrovolskaia
M. A.
.
2010
.
Nanoparticles and the immune system.
Endocrinology
151
:
458
465
.
53
Tian
X.
,
Tian
J.
,
Tang
X.
,
Rui
K.
,
Zhang
Y.
,
Ma
J.
,
Wang
Y.
,
Xu
H.
,
Lu
L.
,
Wang
S.
.
2015
.
Particulate β-glucan regulates the immunosuppression of granulocytic myeloid-derived suppressor cells by inhibiting NFIA expression.
OncoImmunology
4
:
e1038687
.
54
Tian
J.
,
Ma
J.
,
Ma
K.
,
Guo
H.
,
Baidoo
S. E.
,
Zhang
Y.
,
Yan
J.
,
Lu
L.
,
Xu
H.
,
Wang
S.
.
2013
.
β-Glucan enhances antitumor immune responses by regulating differentiation and function of monocytic myeloid-derived suppressor cells.
Eur. J. Immunol.
43
:
1220
1230
.
55
Nefedova
Y.
,
Huang
M.
,
Kusmartsev
S.
,
Bhattacharya
R.
,
Cheng
P.
,
Salup
R.
,
Jove
R.
,
Gabrilovich
D.
.
2004
.
Hyperactivation of STAT3 is involved in abnormal differentiation of dendritic cells in cancer.
J. Immunol.
172
:
464
474
.
56
Longmire
M.
,
Choyke
P. L.
,
Kobayashi
H.
.
2008
.
Clearance properties of nano-sized particles and molecules as imaging agents: considerations and caveats.
Nanomedicine (Lond)
3
:
703
717
.
57
Jones
S. W.
,
Roberts
R. A.
,
Robbins
G. R.
,
Perry
J. L.
,
Kai
M. P.
,
Chen
K.
,
Bo
T.
,
Napier
M. E.
,
Ting
J. P.
,
Desimone
J. M.
,
Bear
J. E.
.
2013
.
Nanoparticle clearance is governed by Th1/Th2 immunity and strain background.
J. Clin. Invest.
123
:
3061
3073
.
58
Kujawski
M.
,
Kortylewski
M.
,
Lee
H.
,
Herrmann
A.
,
Kay
H.
,
Yu
H.
.
2008
.
Stat3 mediates myeloid cell-dependent tumor angiogenesis in mice.
J. Clin. Invest.
118
:
3367
3377
.
59
Rao
D. D.
,
Vorhies
J. S.
,
Senzer
N.
,
Nemunaitis
J.
.
2009
.
siRNA vs. shRNA: similarities and differences.
Adv. Drug Deliv. Rev.
61
:
746
759
.
60
Carro
M. S.
,
Lim
W. K.
,
Alvarez
M. J.
,
Bollo
R. J.
,
Zhao
X.
,
Snyder
E. Y.
,
Sulman
E. P.
,
Anne
S. L.
,
Doetsch
F.
,
Colman
H.
, et al
.
2010
.
The transcriptional network for mesenchymal transformation of brain tumours.
Nature
463
:
318
325
.
61
Xin
H.
,
Zhang
C.
,
Herrmann
A.
,
Du
Y.
,
Figlin
R.
,
Yu
H.
.
2009
.
Sunitinib inhibition of Stat3 induces renal cell carcinoma tumor cell apoptosis and reduces immunosuppressive cells.
Cancer Res.
69
:
2506
2513
.
62
Zhang
H.
,
Nguyen-Jackson
H.
,
Panopoulos
A. D.
,
Li
H. S.
,
Murray
P. J.
,
Watowich
S. S.
.
2010
.
STAT3 controls myeloid progenitor growth during emergency granulopoiesis.
Blood
116
:
2462
2471
.
63
Hossain
D. M.
,
Pal
S. K.
,
Moreira
D.
,
Duttagupta
P.
,
Zhang
Q.
,
Won
H.
,
Jones
J.
,
D’Apuzzo
M.
,
Forman
S.
,
Kortylewski
M.
.
2015
.
TLR9-targeted STAT3 silencing abrogates immunosuppressive activity of myeloid-derived suppressor cells from prostate cancer patients.
Clin. Cancer Res.
21
:
3771
3782
.
64
Marigo
I.
,
Bosio
E.
,
Solito
S.
,
Mesa
C.
,
Fernandez
A.
,
Dolcetti
L.
,
Ugel
S.
,
Sonda
N.
,
Bicciato
S.
,
Falisi
E.
, et al
.
2010
.
Tumor-induced tolerance and immune suppression depend on the C/EBPbeta transcription factor.
Immunity
32
:
790
802
.
65
van Dijk
T. B.
,
Baltus
B.
,
Raaijmakers
J. A.
,
Lammers
J. W.
,
Koenderman
L.
,
de Groot
R. P.
.
1999
.
A composite C/EBP binding site is essential for the activity of the promoter of the IL-3/IL-5/granulocyte-macrophage colony-stimulating factor receptor beta c gene.
J. Immunol.
163
:
2674
2680
.
66
Ruffell
D.
,
Mourkioti
F.
,
Gambardella
A.
,
Kirstetter
P.
,
Lopez
R. G.
,
Rosenthal
N.
,
Nerlov
C.
.
2009
.
A CREB-C/EBPbeta cascade induces M2 macrophage-specific gene expression and promotes muscle injury repair.
Proc. Natl. Acad. Sci. USA
106
:
17475
17480
.
67
Schoenen
H.
,
Huber
A.
,
Sonda
N.
,
Zimmermann
S.
,
Jantsch
J.
,
Lepenies
B.
,
Bronte
V.
,
Lang
R.
.
2014
.
Differential control of Mincle-dependent cord factor recognition and macrophage responses by the transcription factors C/EBPβ and HIF1α.
J. Immunol.
193
:
3664
3675
.
68
Shackleford
T. J.
,
Zhang
Q.
,
Tian
L.
,
Vu
T. T.
,
Korapati
A. L.
,
Baumgartner
A. M.
,
Le
X.-F.
,
Liao
W. S.
,
Claret
F. X.
.
2011
.
Stat3 and CCAAT/enhancer binding protein beta (C/EBP-beta) regulate Jab1/CSN5 expression in mammary carcinoma cells.
Breast Cancer Res.
13
:
R65
.
69
Youn
J. I.
,
Gabrilovich
D. I.
.
2013
.
New roles of Rb1 in expansion of MDSCs in cancer.
Cell Cycle
12
:
1329
1330
.
70
Kumar
V.
,
Cheng
P.
,
Condamine
T.
,
Mony
S.
,
Languino
L. R.
,
McCaffrey
J. C.
,
Hockstein
N.
,
Guarino
M.
,
Masters
G.
,
Penman
E.
, et al
.
2016
.
CD45 phosphatase inhibits STAT3 transcription factor activity in myeloid cells and promotes tumor-associated macrophage differentiation.
Immunity
44
:
303
315
.
71
Thiry
M.
,
Lamaye
F.
,
Thelen
N.
,
Chatron-Colliet
A.
,
Lalun
N.
,
Bobichon
H.
,
Ploton
D.
.
2008
.
A protocol for studying the kinetics of RNA within cultured cells: application to ribosomal RNA.
Nat. Protoc.
3
:
1997
2004
.
72
Gade
P.
,
Roy
S. K.
,
Li
H.
,
Nallar
S. C.
,
Kalvakolanu
D. V.
.
2008
.
Critical role for transcription factor C/EBP-beta in regulating the expression of death-associated protein kinase 1.
Mol. Cell. Biol.
28
:
2528
2548
.
73
Bronte
V.
,
Cingarlini
S.
,
Apolloni
E.
,
Serafini
P.
,
Marigo
I.
,
De Santo
C.
,
Macino
B.
,
Marin
O.
,
Zanovello
P.
.
2003
.
Effective genetic vaccination with a widely shared endogenous retroviral tumor antigen requires CD40 stimulation during tumor rejection phase.
J. Immunol.
171
:
6396
6405
.
74
Sonda
N.
,
Simonato
F.
,
Peranzoni
E.
,
Calì
B.
,
Bortoluzzi
S.
,
Bisognin
A.
,
Wang
E.
,
Marincola
F. M.
,
Naldini
L.
,
Gentner
B.
, et al
2013
.
miR-142-3p prevents macrophage differentiation during cancer-induced myelopoiesis.
Immunity
38
:
1236
1249
.
75
Sasso
M. S.
,
Lollo
G.
,
Pitorre
M.
,
Solito
S.
,
Pinton
L.
,
Valpione
S.
,
Bastiat
G.
,
Mandruzzato
S.
,
Bronte
V.
,
Marigo
I.
,
Benoit
J.-P.
.
2016
.
Low dose gemcitabine-loaded lipid nanocapsules target monocytic myeloid-derived suppressor cells and potentiate cancer immunotherapy.
Biomaterials
96
:
47
62
.
76
Suzuki
A.
,
Leland
P.
,
Joshi
B. H.
,
Puri
R. K.
.
2015
.
Targeting of IL-4 and IL-13 receptors for cancer therapy.
Cytokine
75
:
79
88
.
77
Youn
J. I.
,
Nagaraj
S.
,
Collazo
M.
,
Gabrilovich
D. I.
.
2008
.
Subsets of myeloid-derived suppressor cells in tumor-bearing mice.
J. Immunol.
181
:
5791
5802
.
78
Hirota, K. and H. Terada. 2012. Endocytosis of particle formulations by macrophages and its application to clinical treatment. In Molecular Regulation of Endocytosis. B. Ceresa, ed. InTech. Available at: http://www.intechopen.com/books/molecular-regulation-of-endocytosis/endocytosis-of-particle-formulations-by-macrophages-and-its-application-to-clinical-treatment
.
79
Waltenbaugh, A. K., S. Poe, B. B. Chen, P. Ray, and R. K. Mallampalli. 2011. Molecular and cellular regulation of cardiolipin uptake. American Thoracic Society. Available at: http://www.atsjournals.org/doi/abs/10.1164/ajrccm-conference.183.1_MeetingAbstracts.A4545
.

J.L.V., S.Z., P.M.D., V.B., A.K., and P.S. are named inventors in a nonprovisional patent filed by the University of Miami related to the use of the 4PD nanoparticles described in this article and as such have a potential financial interest should the 4PD nanoparticle be commercialized. The other authors have no financial conflicts of interest.

Supplementary data