Research ArticleCELL BIOLOGY

Promoting the activation of T cells with glycopolymer-modified dendritic cells by enhancing cell interactions

See allHide authors and affiliations

Science Advances  20 Nov 2020:
Vol. 6, no. 47, eabb6595
DOI: 10.1126/sciadv.abb6595


Dendritic cell (DC) modification to enhance antigen presentation is a valuable strategy in cancer immune therapy. Other than focusing on regulating interactions between DC and antigens, we intend to promote cell interactions between DC and T cell by cell surface engineering. T cell activation is greatly improved and generates higher tumor toxicity with the aid of the synthetic glycopolymer modified on the DC surface, although the glycopolymer alone shows no effect. The great promotion of DC–T cell attraction is revealed by cell image tracking in terms of both frequency and duration of contacts. Our findings provide a new method of T cell activation by these engineered “sweet DCs.” This strategy is beneficial for developing more efficient DC-based vaccines.


Dendritic cell (DC)–based vaccination, which induces tumor-specific T cell and directs the immune response toward the antigens of interest (1, 2), has proved to be one of the most promising therapeutic agents for use in cancer therapy (3, 4). Although DC vaccines have been shown to be somewhat effective in the treatment of melanoma, prostate cancer, colon cancer, and breast cancer (58), the clinical benefit is still disappointingly small (1). An important reason may be that deletion of cell surface functional “arms” during DC antigen presentation weakens the interaction between DCs and T cells, thus leading to the failure of T cell induction (9).

Modification of DCs to enhance their antigen presentation is the most common strategy for T cell activation. Protocols developed for loading antigen to DCs in vitro (10) or for the activation of DC surface phagocytosis receptors by modifying antigen with biomaterials (11, 12) have been shown to induce notable DC maturation and a strong T cell response. However, the focus has been mainly on regulation of the interaction between DCs and antigens, while the interaction between DCs and T cells has rarely been studied, although some strategies intend to increase T cell activation by enhancing the intensity of costimulatory molecules (the second signal) (13) or cofactors (the third signal) (14) on the DC surface. However, exposure of the major histocompatibility complex (MHC) to the T cell receptor (the first signal) is the initial step (15, 16). The intensity of their binding directly determines the initiation of antigen presentation and further T cell activation (17). However, very few studies focused on this aspect.

“Engineering” or modifying cell surfaces with synthetic ligands or receptors provides a method for regulating the interactions between different cell types (18, 19). A number of approaches have been devised for this purpose, including DNA hybridization (20), use of lipid aptamers (21), lipid peptides (22), biotin-streptavidin interactions (23), and carbohydrate-lectin binding (24). In particular, carbohydrate-lectin binding, due to its specificity and the stability of the bond, plays an important role in promoting cell-to-cell recognition and adhesion (25). A recent study demonstrated that mannose-modified tumor antigens substantially enhanced the ability of DCs to recognize and bind to the antigens (26). Furthermore, work in our laboratory showed that glycopolymer-modified HeLa cells induced a strong phagocytic response in macrophages (27). In the DC–T cell system, the mannose receptors on the T cell surface have been demonstrated to be essential for T cell functions (28, 29). We considered it highly probable, therefore, that glycopolymer [synthesized poly-mannose (pMAM) or poly-glucose (pMAG), which has an affinity for mannose receptors]–engineered DCs will attach specifically to T cells through carbohydrate-lectin binding. This improved contact between the two types of cells is expected to facilitate binding, increase the stability of the DC–T cell complex, and promote T cell activation.

In the present work, we developed a sweet DC (pMAM- and pMAG-engineered DC) to promote the T cell activation by enhancing interactions between DC and T cell. Then, we further analyzed the mechanism using cell image tracking. This work provides evidence that adding proper synthetic glycopolymers onto the cell surface is a valuable and effective path to the design and optimization of cell vaccines.


Modification of DC with glycopolymers

The schematic process of engineering DC is shown in Fig. 1. Considering the requirement for longer-term stability, membrane-bound halo-tag protein (HTP) was used as an anchor to attach glycopolymers to the DC surface. DCs were first matured by loading antigens from tumor cell lysate (fig. S3), and then DCs were transfected with HTP plasmid (transfection being difficult using common transfection agents such as Lipofectamine). A photo perforation transfection system reported previously (30) was used. Transfection was carried out by loading HTP plasmid and DCs on the photothermal substrate made by multiwalled carbon nanotubes (MCNTs) and followed by laser irradiation (2.5 W/cm2) for 30 s. As shown in Fig. 2C, the transfection efficiency was >95%, while in the lipo2000 group, no transfection was observed (fig. S2).

Fig. 1 Schematic process of DC modification by glycopolymers.

(A) Strategy for DC surface modification with glycopolymers. DCs were first matured and then transfected by HTP using a photoporation delivery system. Then, the synthetic glycopolymers were attached to the cell surface through the reaction between HTP anchors and the polymer end groups. (B) DC stably modified with glycopolymers on the cell surface promotes the efficiency of T cell activation. (C) Molecular structures of synthetic glycopolymers and graphic notes.

Fig. 2 DC cell surface modification with glycopolymers.

(A) Schematic of HTP transfection with a photo perforation transfection system (PTS) was used. (B) Representative images showing HTP transfection. Nuclei were stained by DAPI (blue), and the HA-tagged HTP was stained by FITC-avidin (green). Scale bar, 100 um. (C) Quantification of transfection efficiency compared with Lip2000. ***P < 0.001 compared with Lip2000. Data are means ± SEM (n = 3). (D) Schematic of DC modification with glycopolymers. (E) Representative images showing green fluorescence on the DC cell surface. Nuclei were stained by DAPI (blue) and biotin-labeled poly-(MAG) (pMB) by FITC-avidin (green). (F) Representative images showing the modified DCs incubated in complete medium for specified times (1, 3, and 7 days). (G) Viability of engineered DC over the 7-day period. Data are means ± SEM (n = 3). N.D., not determined.

Chloroalkane-terminated glycopolymers for attaching to HTP were synthesized via reversible addition-fragmentation chain transfer (RAFT) polymerization of 2-methacrylamido glucopyranose (MAG) or 2-methacrylamido mannose (MAM) as described previously (27). DCs successfully modified with fluorescein isothiocyanate (FITC)–tagged glycopolymers showed green fluorescence on the cell surface (Fig. 2E). Green fluorescence was observable for up to 7 days and then decreased gradually (Fig. 2F), indicating that, although some of the glycopolymers were phagocytosed by DCs, HTP was able to retain the polymers on the cell surface for a relatively long period, thus facilitating longer-term investigation.

T cell activation by glycopolymer-engineered DCs

To determine whether engineered DCs promote T cell activation, T cell proliferation and cytokine release by T cell after T cells and DCs coculture were investigated. Tumor cell lysate was used as an antigen to induce DC maturation (fig. S3); afterward, DCs were modified with glycopolymers as described above. T cells from mouse spleen were cocultured with DCs in a ratio of 10:1. After 24 hours of culturing, T cells were collected for measurement of proliferation, while the supernatant was collected for measuring the activators expressed by the T cells (Fig. 3A). Cell proliferation was verified by 5(6)-carboxyfluorescein diacetate succinimidyl ester (CFSE) staining before coculture and measured by flow cytometry. The fluorescence intensity peak shifted to the left in the modified DC groups, indicating that the T cells experienced a higher division frequency (Fig. 3, B and C). The cytokines interferon-γ (IFN-γ) and tumor necrosis factor–α (TNF-α) expressed by T cells were measured in the supernatant by enzyme-linked immunosorbent assay (ELISA). T cells released more IFN-γ and TNF-α when activated by modified DCs than by native DCs. However, DCs modified with different glycopolymers showed slight differences in inducing the expression of these factors, specifically, DC-pMAG group secreted more IFN-γ, nearly 4-fold compared with DC-pMAM group by 2-fold, while DC-pMAM group secreted more TNF-α, about 3.5-fold compared with DC-pMAG group by 2-fold (Fig. 3, D and E). IFN-γ is the most representative cytokine secreted by helper T cells (TH1) (31, 32), which facilitate the activation of T cells to cytotoxic T lymphocytes (CTLs). TNF-α is a cytotoxic factor secreted by CTL that binds to TNF receptor on the surface of tumor cells and induces tumor apoptosis. Therefore, we considered that the MAG-DC tend to induce TH1 cells.

Fig. 3 Promotion of T cell activation by glycopolymer-engineered DCs.

(A) Schematic of experiment on T cell proliferation and the release of cytokines. (B) Mean fluorescence intensity distribution of CFSE-stained T cells. (C) Mean fluorescence intensity of T cells induced by engineered and native DC. Data are means ± SEM (n = 3). “CON” represents T cells without induction by DCs, DC-T represents T cells induced by native DCs, “DC-pMAG-T” represents T cells induced by pMAG-modified DCs, and “DC-pMAM-T” represents T cells induced by pMAM-modified DCs. ***P < 0.001. (D and E) Cytokine release from T cells induced by different DCs. Data are means ± SEM (n = 3). **P < 0.01 and ***P < 0.001.

T cells induced by glycopolymer-modified DCs: Specific cytotoxicity to cancer cells

To investigate further whether the T cells induced by modified DCs produce a stronger tumor immune effect than by native DCs, two specific T cell types that target melanoma (B16) and colon cancer (CT26), referred to as TB16 and TCT26, respectively, were investigated. After coculturing the T cells with the corresponding tumor cells (2:1 ratio) for 24 hours, the tumor cell growth and viability were determined (Fig. 4A).

Fig. 4 Representation that T cells activated by glycopolymer-modified DCs have increased cancer cytotoxicity and the T cell specificity is not affected.

(A) Schematic showing T cell induction process. B16 antigens were used to stimulate DC to present antigens to T cells, making the T cells specific to B16 (TB16), similarly for CT26 antigen making T cells specific to CT26 (TCT26). (B) Representative cell images after coculturing B16 with T cells: DC-T represents native DC-induced T cells, “pMAG-DC-T” represents T cells induced by pMAG-modified DC, and “pMAM-DC-T” represents T cells induced by pMAM-modified DC. (C) Representative data on LDH release from B16 after treatment with T cells induced by different kinds of DC. Data are means ± SEM (n = 3). ***P < 0.001. (D) Representative cell images after coculturing CT26 with T cells. (E) Representative data on LDH release from B16 after treatment with T cells induced by different kinds of DCs. Data are means ± SEM (n = 3). ***P < 0.001. (F) Showing that glycopolymer-modified DCs had no impact on T cell specificity. Representative cell images show TB16 and TCT26 cells cross-linked to B16 and CT26 cells, respectively. (G) Representative data on LDH release from B16 and CT26 after cross-treatment with specific T cells. Data are means ± SEM (n = 3). ***P < 0.001. OD, optical density.

It was found that the volume of B16 tumor cells was significantly reduced by both MAG-DC– and MAM-DC–induced T cells compared with those T cells induced by unmodified but matured DC group (“DC-T” group) (Fig. 4B). These phenomena were accompanied by increased release of lactate dehydrogenase (LDH), reflecting tumor cell membrane damage. LDH release was 1.77-fold greater in the MAG-DC-T group and 1.82-fold greater in the MAM-DC-T group than in the DC-T group (Fig. 4C). Similar effects were seen for CT26 cells. The network-like structure of the CT26 culture disappeared, cell adhesion decreased, and the cells were damaged (Fig. 4D); these effects were accompanied by 1.72-fold and 1.57-fold greater LDH release, respectively, in the MAG-DC-T and MAM-DC-T groups compared with the DC-T group (Fig. 4E). It may be concluded, therefore, that the glycopolymer-modified DC induces greater T cell tumor toxicity than unmodified DC.

Since DC-induced T cells are specific, we further investigated whether the modified DCs have an impact on this specificity. Both B16 and CT26 cells were treated with TB16 and TCT26 cells, respectively (Fig. 4A). Furthermore, TB16 cells tend to kill B16 cells more specifically, and LDH release was about 2.6-fold greater for B16 than for CT26, while TCT26 induced 2.1-fold greater release of LDH from CT26 than from B16 cells (Fig. 4, F and G). These results indicate that glycopolymer-modified DCs do not alter the specificity of T cell binding to tumors.

Glycopolymer-engineered DCs promoted the interaction with T cells

We speculate that the improvement in T cell activation by glycopolymer-modified DC compared with unmodified DC is due to stronger cell-cell interactions. To investigate this hypothesis, we visualized DC–T cell interactions using cell image tracking. The process was video-recorded every 3 min for a period of 120 min. Intercellular interactions are deemed to have occurred when the contact time of two cells are longer than 10 min. On this basis, the data on contact time of the two cell types and residence time of T cells on the DC surface were analyzed. It was observed that T cells were easy to separate from unmodified DCs, even when the DCs were mature. However, the gathering of T cells around the modified DCs increased over time, with numbers on the DC surface remaining relatively stable (Fig. 5A). Quantitative analysis showed that the contact events between T cell and the two modified DC cell types were significantly favored. Compared with the unmodified DCs, the frequency of T cell contact to DC-pMAG and DC-pMAM improved about 2.5-fold and 3-fold, respectively. In the event of contact, the duration of more than 10 min occurs 8 ± 3 times out of the total numbers in the unmodified DC group, while 35 ± 7 times in DC-pMAG group and 43 ± 9 times in DC-pMAM group. These results indicated that T cells were more prone to adhere to the modified DCs. The duration of T cell adhesion to the modified DCs demonstrated further that glycopolymers on the DC surface promoted the binding of the two cell types (Fig. 5, B and C), due to the specific affinity between synthetic glycopolymer and mannose receptors.

Fig. 5 Glycopolymers modified on the DC cell surface were necessary for enhancing the interaction with T cells.

(A) Image tracking of DC and T cell migration over time. (B) Frequency of contact between DC and T cell. Scale bar, 10 μm. (C) Duration of contact between DC and T cell. Data analysis using Tukey’s multiple comparison test, with T cell number n = 70. Data are means ± SEM (n = 3).*P < 0.05, **P < 0.01, and ***P < 0.001. (D) QCM measurements of interactions between pMAG with proteins PD-1 and CD40L on the T cell surface. (E) Quantitative analysis of protein adsorption difference. Data shown are means ± SEM (n = 3). ***P < 0.001 (t test). (F to H) Free glycopolymers that were added to the medium at concentrations of 0.1 mg/ml did not help to improve the interactions between DCs and T cells. (F) Image tracking of DCs and T cells migration over time. Scale bar, 10 μm. (G) The frequency of contact between DCs and T cells. (H) Duration of contact between DCs and T cells. Data analysis using Tukey’s multiple comparison test, with T cell number n = 70. Data are means ± SEM (n = 3). ***P < 0.001. (I to K) d-(+)-mannose (1 mg/ml) blocked the interactions between glycopolymer-modified DCs and T cells. (I) Image tracking of DC and T cell migration over time. Scale bar, 10 μm. (J) The frequency of contact between DCs and T cells. (K) Duration of contact between DCs and T cells. Data analysis using Tukey’s multiple comparison test, with T cell number n = 70. Data are means ± SEM (n = 3). **P < 0.01, and ***P < 0.001.

In addition to the specific mannose receptors, other proteins may have interactions with the glycopolymers. To investigate this possibility, the interactions between pMAG and PD-1/CD40L (costimulatory factors on the T cell surface) were examined using quartz crystal microbalance (QCM). It was shown that pMAG-modified chips adsorbed a greater quantity of PD-1 than the blank golden chips while adsorption of CD40L was less than the blank (Fig. 5, D and E), demonstrating that pMAG has a stronger interaction with PD-1 than with CD40L.

DC surface modification was necessary for promotion of the interaction with T cells

To further investigate the effects of the glycopolymers on DC–T cell interactions, free pMAG or pMAM was added directly to the medium at the same concentration used to modify the DCs. This comparison allowed determination of whether the free glycopolymers in the medium were able to promote the same DC–T cell interactions seen with the glycopolymer-modified DCs. Image tracking over a 2-hour period, however, did not show any obvious interactions between the two cell types (Fig. 5F). Both the frequency of contact and duration of contact were similar to those for experiments without glycopolymers in the medium and much less than for the engineered DCs (Fig. 5, G and H). These results indicate that the glycopolymers must be attached to the DC surface to promote DC–T cell interactions. Glycopolymers in the medium may occupy mannose receptors on the T cell surface, thus further inhibiting close contact of DCs with T cells. To test this hypothesis, d-(+)-mannose at a concentration of 1 mg/ml was used to block the mannose receptors. A significant decrease in modified DC–T cell interactions was observed when d-(+)-mannose was present (Fig. 5I): Both the frequency and duration of contact decreased strongly (Fig. 5, J and K). These results provide additional evidence that the glycopolymers attached to the DC surface rather than in the medium promote interactions with T cells through mannose receptors.

Fig. 6 Glycopolymer-modified BMDCs enhanced T cell viability that was inhibited by blocking sugar receptors on the T cell surface.

(A) Representative images of BMDCs modified with pMAG. (B) TNF-α expression of T cells induced by pMAG-modified BMDCs. ***P < 0.001 compared with mDC. (C) IFN-γ expression of T cells induced by pMAG-modified BMDCs. (D) Representative image of B16 cancer cells treated by unmodified DC-induced T cells. (E) Representative images of B16 cancer cells treated by glycopolymer-engineered T cells. pMAG-DC-T represents T cells induced by MAG-modified DC, and pMAM-DC-T represents T cells induced by MAM-modified DC. (F) Representative images of B16 cancer cells treated by sugar receptor–blocking T cells. “IG-T” represents T cells incubated with glucose and then induced by pMAG-modified matured DCs. “IM-T” represents T cells incubated with mannose and then induced by pMAG-modified matured DCs. Scale bar, 10 μm. (G) B16 cancer cell viabilities treated by different BMDC-induced T cells. Data analysis using Tukey’s multiple comparison test. Data are means ± SEM (n = 3). *P < 0.05 compared with pMAG-DC-T, #P < 0.05 compared with pMAM-DC-T.

Glycopolymer-modified primary bone marrow–derived DCs promote T cell activation

To expand the application of glycopolymer-modified DCs, we conducted a series of experiments to verify the feasibility on primary bone marrow–derived DCs (BMDCs). Using cholesterol as anchors, we inserted the glycopolymers onto the DC surface by simply incubating glycopolymers (0.1 mg/ml) and DCs for 30 min. As shown in Fig. 6, these glycopolymer-modified DCs induced higher expression of TNF-α and IFN-γ of T cells (Fig. 6, B and C), and these T cells exhibited stronger antitumor activities. However, the cytotoxicity of T cell to cancer cells weakened when T cells were first incubated with glucose and mannose, which blocked the mannose/glucose receptors on the T cell surface (Fig. 6, D to F). These results indicate that coating the BMDC surface with glycopolymers can also enhance the interactions between DC and T cells, further inducing higher T cell activation.


In this work, we successfully engineered DC by glycopolymers in a safe and stable manner. The engineered DCs showed higher efficiency of T cell activation than native DCs. We believe that this is due to the enhancement of interactions between the two types of cells. To confirm this, we used cell image tracking to record the movement process when DCs and T cells were cocultured together. We visualize that the glycopolymer-engineered DCs and T cells are greatly enhanced with increased help to improve the binding possibility of intermolecular signals, which has been proved in our experiment with increased specific cancer cytotoxicity.

Moreover, we confirmed that the enhancement of engineered DCs and T cells may rely on the mannose receptors on the T cell surface (28, 29). We verified that the glycopolymers modified on the DC surface, rather than in medium, enhance the DC–T cell interaction. Furthermore, we found that the enhancement of cell interactions was nullified by the addition of the small-molecule d-(+)-mannose. We speculate that these small molecules competitively occupy the mannose binding sites on the surface of T cells, thus leading to the failure of contact between sweet DCs and T cells. This was also confirmed in the cytotoxicity study of T cells induced by glycopolymer-modified BMDCs. The incubation of T cells with glucose and mannose, which occupy the sugar binding sites on the surface of T cells, leads to a decrease of T cell activation efficiency (Fig. 6).

We further look into whether glycopolymers interact with other molecules on the T cell surface that participate in T cell activation. We investigated the interactions between glycopolymers with PD-1 and CD40L by QCM. Results showed that pMAG has a higher tendency to interact with PD-1 but not with CD40L. The combination of PD-1 to PD-L1 on the DC surface initiates the programmed death of T cells, thus inhibiting T cell proliferation (33). Attachment of CD40 to CD40L on the surface of DC increases their antigen presentation and costimulatory capacity (34) so that CD40-CD40L interactions are vital in CTL priming (35). Therefore, we consider that the strong interactions between pMAG and PD-1 not only improve the interaction between DCs and T cells but also block PD-1 signaling, which inhibits T cell proliferation. On the other hand, the “rejection” of CD40L frees up space for CD40 signaling priming. These results give the evidence that sweet DCs not only enhance the binding intensity between DCs and T cells but also give a helping hand in terms of signaling pathways within the cells.

We not only engineered DC2.4 cells with glycopolymers but also successfully applied this approach on primary DC cells (Fig. 6). This work provides a new strategy to prepare an enhanced DC vaccine by modifying the DC surface with glycopolymers. We have preliminarily discussed the effect of glycopolymer modification on primary BMDCs in vitro (figs. S2, S7, and S9). Further experiments are still needed to study the function of glycopolymer-modified DCs in vivo. In the future, we will conduct a series of in vivo tests to verify the feasibility of glycopolymer-modified DCs as vaccines and their application in immunotherapy.


Materials and reagents

Chemical reagents. 4-Cyanopentanoic acid dithiobenzoate (CPADB), MCNTs, and sylgard(r) 184 were from Sigma-Aldrich Chemical Co. (Shanghai, China). 2-(2-Boc-aminoethoxy) ethanol was from Adamas Co. Ltd. (Shanghai, China). N-hydroxysuccinimide (NHS), 1-chloro-6-iodohexane (stabilized with copper chips), sodium hydride (60%, dispersion in paraffin liquid), dicyclohexylcarbodiimide (DCC), and anhydrous tetrahydrofuran (THF; stabilized with Butylated hydroxytoluene, BHT) were from TCI Co. Ltd. (Shanghai, China). 2,2′-Azoisobutyronitrile (AIBN) from Sinopharm Chemical Reagent Co. (Shanghai, China) was recrystallized from ethanol and dried under vacuum before use. Anhydrous N,N-dimethylformamide (DMF) was from Aladdin Co. Ltd. (Shanghai, China). MAG and MAM were synthesized as reported previously (36, 37). d-(+)-mannose was from TCI (Shanghai, China). Other organic solvents were from Sinopharm Chemical Reagent Co. (Shanghai, China) and distilled before use. Deionized water (DIW), purified to a minimum resistivity of 18 megohms∙cm using a Millipore water purification system, was used in all experiments.

Biological reagents and methods. RPMI 1640 medium, penicillin-streptomycin solution, and fetal bovine serum (FBS) were from Gibco (Grand Island, NY, USA). Plasmoc in transmembrane was from InvivoGen Co. Ltd. (San Diego). Bovine serum albumin (BSA), paraformaldehyde, and Triton X-100 were from Sigma-Aldrich Chemical Co. (Shanghai, China). DAPI (4′,6-diamidino-2-phenylindole) was from Invitrogen (Waltham, MA). FITC-conjugated avidin (FITC-avidin) was from Sigma-Aldrich Chemical Co. (Shanghai, China). Primary antibody hemagglutinin (HA) and FITC-conjugated secondary antibody were from Wuhan Boster Biological Technology Ltd. (Wuhan, China). Dopamine was from Sigma-Aldrich Chemical Co. (Shanghai, China). Recombinant mouse granulocyte-macrophage colony-stimulating factor (GM-CSF) was from PeproTech EC Ltd. (UK). Lipopolysaccharide (LPS) was from Sigma-Aldrich Chemical Co. (Shanghai, China). DC2.4 cells were from Millipore. Fluorescence images were captured using an SP5II confocal microscope (Leica Microsystems). The images were overlapped using Image-Pro Plus 6.0 software. Flow cytometry was performed on a BD FACSVerse system, and the data were analyzed using FlowJo with 10,000 events during collection. CFSE was from Invitrogen (CA, USA), 1,1’-Dioctadecyl-3,3,3‘,3’-tetramethylindodicarbocyanine perchlorate, DID staining buffer was from Sangon (Shanghai, China), mouse IFN-γ and TNF-α ELISA kit were from DAKEWE (Suzhou, China), and LDH cytotoxicity colorimetric assay kit was from BioVision (CA, USA). CD80 and CD86 staining kits were from eBioscience (CA, USA). CD40L and PD-1 protein were from Proteintech (Chicago, USA). Anti–PD-L1 antibody was from Abcam (Cambridge Science Park in Cambridge, UK). CD80, CD86, CD40, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) rabbit monoclonal antibody were from Cell Signaling Technology (Boston, USA). PD-L1 and MHC I rabbit polyclonal antibody were from Proteintech (Chicago, USA). FITC anti-mouse CD11c, allophycocyanin (APC) anti-mouse CD86, phycoerythrin (PE) anti-mouse CD80, Alexa Fluor 488 anti-mouse CD11b, PE anti-mouse CD207, and APC anti-mouse CD103 were from BioLegend (San Diego, CA).

Synthesis of chloroalkane-conjugated chain transfer agent.
tert-Butyl (2-(2-((6-chlorohexyl)oxy)ethoxy)ethyl)carbamate (B)

To a solution of 2-(2-boc-aminoethoxy) ethanol (4.30 g, 20.95 mmol) in anhydrous THF (40 ml) and DMF (20 ml) at 0°C, NaH (60% dispersion in paraffin liquid, 1.12 g, 28.00 mmol) was added portionwise. After stirring at 0°C for 30 min, 1-chloro-6-iodohexane (4.80 ml, 31.60 mmol) was added to the mixture at 0°C. The reaction mixture was stirred at 0°C for 20 min, at room temperature for 16 hours, and then quenched with saturated aqueous NH4Cl solution. The mixture was extracted twice with ethyl acetate, and the combined extracts were washed with brine, dried over Na2SO4, filtered, and concentrated. The residue was chromatographed on silica gel using 3:1 hexane/ethyl acetate to afford tert-butyl (2-(2-((6-chlorohexyl)oxy)ethoxy)ethyl)carbamate (B). 1H nuclear magnetic resonance (NMR; 300 MHz, CDCl3): δ 3.61-3.50 (m, 8H), 3.47-3.43 (t, J = 6 Hz, 2H), 3.32-3.28 (t, J = 6 Hz, 2H), 1.81-1.72 (m, 2H), 1.65-1.55 (m, 2H), 1.49-1.33 (m, 13H).

2-(2-(Chloromethoxy)ethoxy)ethan-1-amine (C)

To a solution of B (1.35 g, 4.17 mmol) in 30 ml of CH2Cl2 at 0°C was added 5 ml of trifluoroacetic acid (TFA). After stirring for 2.5 hours at 0°C, TFA and solvent were removed, and the residue was diluted with 30 ml of CH3OH. The solution was cooled to 5°C, and K2CO3 (1.65 g, 11.93 mmol) was added to the mixture. The mixture was stirred at 5°C for 10 min, filtered, and evaporated. The residue was diluted with H2O (20 ml), and the mixture was extracted four times with ethyl acetate. The combined extracts were dried over Na2SO4, filtered, and concentrated. The crude product was purified by silica gel chromatography to afford 2-(2-(chloromethoxy)ethoxy)ethan-1-amine (C). 1H NMR (300 MHz, CDCl3): δ 3.63-3.50 (m, 8H), 3.48-3.44 (t, J = 6 Hz, 2H), 2.92-2.88 (t, J = 6 Hz, 2H), 2.66 (s, 2H), 1.81-1.72 (m, 2H), 1.64-1.55 (m, 2H), 1.50-1.30 (m, 4H).

5-((2-(2-((6-Chlorohexyl)oxy)ethoxy)ethyl)amino)-2-cyano-5-oxopentan-2-yl benzodithioate (A)

CPADB (0.50 g, 1.91 mmol), NHS (0.33 g, 2.87 mmol), and DCC (0.74 g, 3.82 mmol) were dissolved in 10 ml of DMF. The reaction was followed by thin-layer chromatography. 2-(2-(chloromethoxy)ethoxy)ethan-1-amine (C) (387 mg, 1.73 mmol, dissolved in 10 ml of DMF) was added to the reaction mixture when the spot from CPADB had disappeared. The mixture was stirred overnight at 30°C. The mixture was washed with water and then concentrated. Purification by silica gel chromatography was done using 1:100 CH3OH/CH2Cl2.

Preparation of glycopolymers containing chloroalkane end groups. Glycopolymers containing chloroalkane end groups were synthesized via RAFT polymerization of MAG or MAM using the synthesized chain transfer agent A. Briefly, MAG or MAM (0.1240 g, 0.5 mmol), A (0.0024 g, 5.0 μmol), and AIBN (0.0004 g, 2.5 μmol) were dissolved in DMF. The solution was then deoxygenated by bubbling with nitrogen for 30 min. The polymerization was carried out at 70°C for 12 hours under a nitrogen atmosphere. The reaction mixtures were dialyzed for 2 days against DIW to remove unreacted monomer. The glycopolymer solutions were lyophilized to give fluffy solid products. The biotin-labeled glycopolymer pMB was generated via RAFT polymerization of MAG and biotin monomer (a custom-synthesized compound from Lianke Biotechnology Service Department, Suzhou, China). Briefly, MAG (0.1112 g, 0.45 mmol), biotin monomer (0.0257 g, 0.05 mmol), A (0.0024 g, 5.0 μmol), and AIBN (0.0004 g, 2.5 μmol) were dissolved in DMF. The polymerization was carried out by the same methods as described above.

Characterization. All synthetic products were analyzed by 1H NMR (Bruker, 300 MHz, tetramethylsilane as internal standard at 293 K) to verify their chemical structure and composition. Spectra were recorded in CDCl3 or D2O solution [1H NMR: CDCl3, 7.26 parts per million (ppm); D2O, 4.79 ppm; for 13C NMR: CDCl3, 77.0 ppm]. Mass spectra were acquired using a Bruker microTOF-Q III instrument. Fourier transform infrared spectra were recorded using a Nicolet 6700 spectrometer with 32 scans per sample. The number-average molecular weights (Mn) and polydispersity indices of the polymers were determined by size exclusion chromatography using a Waters 1515 gel permeation chromatograph with polyethylene glycol standards.

Preparation of photoporated delivery platform. Polydimethylsiloxane precursor and sylgard 184 silicone rubber curing agent were mixed with a mass ratio of 10:1, and then the mixture was dispersed with ultrasound. After dispersion, the MCNTs were added with a mass ratio of 1%. Then, the mixture was poured into the culture dish, stirred well, and incubated at 65°C until solidification. The matrix was cut into pieces with a diameter of 1 cm. Before use, the samples were washed with 95% ethanol and water three times, respectively, and then dried with nitrogen.

QCM measurements. QCM chips were immersed separately in a suspension of thiolated pMAG at room temperature overnight. The chips were then washed three times with sterile water and placed in QCM chambers. Phosphate-buffered saline (PBS) was circulated in the system at a flow rate of 50 μl/min until the baseline was stable. Proteins in PBS solution were then injected at a speed of 10 μl/min for 105 min. PBS was circulated again to reach a new stable baseline, and the change in frequency due to protein adsorption was determined.

Cell culture. DC line DC2.4, melanoma cell line B16, mouse colon cancer cell line CT26, and T cells were used for most of the experiments. T cells were obtained from mouse spleen. Cell extraction was done using the MojoSort kit following the instructions of the supplier. All cells were cultured in RPMI 1640 medium with 10% FBS, penicillin G (100 U/ml), and streptomycin (100 g/ml). Cells were maintained in a humidified incubator at 37°C with 5% CO2/95% air and subcultured every 48 hours.

Primary DC culture. BMDCs were prepared as follows. Briefly, bone marrow cells were collected by flushing tibias and femurs of BALB/c mice (7 to 12 weeks old) using a 26G needle syringe with PBS, and then the cell suspension was filtered through nylon mesh to remove small tissue pieces and debris. Bone marrow cells (2 × 105/ml) were cultured in (RPMI 1640 with 10% fetal calf serum and 50 mM 2-mercaptoethanol) supplemented with recombinant mouse GM-CSF (5 ng/ml) with medium change every 4 days. After 7 to 9 days of culture, nonadherent cells were harvested, and final maturation of DCs was induced by incubation with LPS at 1 mg/ml for the final 17 to 24 hours of culture.

Cell transfection and harvesting. Macromolecular delivery experiments were performed as reported previously (38). Briefly, samples were placed in a 48-well culture plate. Each well was seeded with 5 × 104 DC2.4 cells and incubated for 12 hours to allow the cells to spread fully. The cell culture medium was then replaced by 250 μl of RPMI 1640 containing 250 μg of tetramethyl rhodamine isothiocyanate (TRITC) dextran, 250 μg of Rhodamin B isothiocyanate-Bovine Serum Albumin, (RBITC-BSA) or 1.5 μg of pHTP. The wells were then irradiated with a near-infrared laser (808 nm wavelength) at 2 W/cm2 for 30 s. DC2.4 cells delivered with TRITC-dextran and RBITC-BSA were stained with DAPI (blue) 30 min after laser treatment, and cells delivered with pGFP were stained with DAPI 48 hours after laser treatment and observed by fluorescence microscopy using a 20× objective. Images of 15 randomly chosen fields were captured, and three replicates were examined. The delivery efficiency was calculated as the ratio of red or green fluorescent cell density to blue fluorescent cell density. The relative viability of DC2.4 cells 48 hours after delivery was measured by cholecystokinin-8 assay. Absorbance at 450 nm was recorded on a microplate reader. Three replicates were examined.

Construction of DC2.4 cells stably expressing HTP. The HA-tagged HTP gene inserted between the sequence encoding the immunoglobulin κ-chain leader and the platelet-derived growth factor receptor transmembrane domain 3 was a customized plasmid from Genewiz Co. Ltd. (Suzhou, China). DC2.4 cells were plated and transfected with HTP plasmid using an intracellular delivery platform based on photoporation as described above. After expression of plasmid DNA in DC2.4 for 2 days, an HA immunofluorescence staining assay of the cells was carried out to demonstrate the display of HTP on the cell membranes. In detail, the DC2.4 cells were plated and cultured on the chamber slides in complete RPMI 1640 medium overnight. They were then washed with PBS, fixed in 4% paraformaldehyde for 10 min, and rinsed three times in PBS. The cells were permeabilized with 0.1% Triton X-100 for 5 min. After washing twice with PBS, the cells were blocked using 3% BSA/PBS blocking solution for 30 min. They were then incubated with primary anti-HA antibody overnight at 4°C. After washing three times with PBS, the cells were incubated with FITC-conjugated secondary antibody at room temperature for 1 hour and washed twice with PBS. Representative images were captured using a laser scanning confocal microscope.

Engineering DC2.4 cell surfaces with well-defined synthetic glycopolymers. DC2.4 cells stably expressing HTP were washed twice with 200 μl of PBS and incubated with 150 μl of pMB solution (0.1 mg/ml) in serum-free RPMI 1640 at 37°C for 1 hour. PBS (200 μl per well) was added to wash away excess polymers, and the cells were incubated with FITC-avidin solution for 1 hour at 4°C. Images were captured using a laser scanning confocal microscope.

Stimulation of DC maturation. Antigens were from tumor cell lysate, which was obtained by alternately rapid-freezing in liquid nitrogen and thawing at 37°C for 5 cycles. After that, the cell lysate was added to the DCs to stimulate DC maturation overnight. Flow cytometry was used to determine the CD80/CD86 expression among CD11c subgroups.

Coculture of T cells and glycopolymer-engineered DC cells. DC2.4 cells stably expressing HTP were incubated for 24 hours at 37°C (5% CO2) in complete RPMI 1640 medium. The cells were then washed twice with PBS and incubated, respectively, with pMAG (Mn = 5100, Ð = 1.42) or pMAM (Mn = 5000, Ð = 1.35) solution (0.1 mg/ml) in RPMI 1640 or normal medium without FBS at 37°C for 1 hour. The cells were then washed three times with PBS to remove unbound glycopolymers. For BMDCs, cells were first suspended in RPMI 1640 without FBS, and then the CHO-pMAG (Mn = 8600, Ð = 1.76) or CHO-pMAM (Mn = 6600, Ð = 1.13) was added into the medium at the final concentration from 0.1 to 0.001 mg/ml and cultured in a humidified incubator at 37°C with 5% CO2/95% air for 30 min. After that, the DCs were washed by PBS twice to remove the residual glycopolymers. T cells were then mixed with the DCs (DCs:T cells, 1: 2) and incubated for 24 hours.

Cell image tracking and image analysis. Cell image tracking was performed by using the Nikon Eclipse Ti Imaging System. T cells (total of 60 × 104) obtained from mouse spleen were added into 30 × 104 DCs in 1 ml of RPMI 1640 containing 10% FBS in six-well chambers. Images were acquired sequentially every 3 min for 120 min. T cells that appeared to be caught passively in DC clusters were excluded from the analysis. T cells and DCs in contact were enumerated and expressed as contact frequency and duration. T cells that appeared to be caught passively in DC clusters were excluded from the analysis.

T cell proliferation. The proliferation of T cells was determined by CFSE staining. Before coculturing with DC, T cells were incubated with 5 μM CFSE for 15 min at 37°C. The cells were then washed three times with PBS to remove excess CFSE. After coculturing T cells and DC for 24 hours, T cells were separated and quantified by flow cytometry with 488 nm excitation light. Fluorescence intensity was analyzed by Tree Star FlowJo X 10.0.7.

T cell staining. T cells were first resuspended in RPMI 1640 without FBS. Then, DID staining reagent was added into the medium at the final concentration of 5 μM. Then, the cells were cultured in a humidified incubator at 37°C with 5% CO2/95% air for 30 to 40 min. The T cells were washed by PBS twice and prepared for use.

Identification of cytokines released by T cells. Supernatants obtained after coculturing T cells and DC for 24 hours were used for cytokine determination. IFN-γ was determined using the BioRay Mouse IFN-γ ELISA kit following the manufacturer’s instructions. TNF-α was determined by a TNF-α ELISA kit following the manufacturer’s instructions. All data were analyzed using GraphPad 7.0.

LDH release assay. The cytotoxicity of T cells induced by glycopolymer-engineered DCs was compared with that induced by unmodified DC. After activation by DC, T cells were collected and cocultured with the specific tumor cells B16 and CT26 for 24 hours at 37°C(5% CO2) in complete RPMI 1640 medium. Supernatants were used for LDH determination using a Pierce LDH Cytotoxicity Assay Kit (Thermo Fisher Scientific). In brief, 50 μl of reaction mixture was added to each well, and the plate incubated at room temperature for 30 min in the dark. The reaction was stopped by adding 50 μl of the stop buffer included in the kit, and the absorbance of each well at 490 nm was determined.

Western blot. For Western blot analysis, equal amounts of protein (20 μg) were separated on a SDS–polyacrylamide gel electrophoresis gel and transferred to a polyvinylidene difluoride membrane. The membrane was rinsed and blocked with 5% nonfat skim milk in tris-buffered saline with 0.1% Tween 20 for 2 hours at room temperature. Then, the following primary antibodies were incubated at 4°C overnight. Then, the secondary antibodies were incubated at room temperature for 1 hour. Chemiluminescence was imaged on a FUJIFILM LAS-3000 system. The basal levels of proteins were normalized to the level of GAPDH protein.

Statistical analysis. Student’s t test was used for comparisons between two groups, while one-way analysis of variance (ANOVA) or two-way repeated-measures ANOVA with the Tukey’s multiple comparison test was used for comparisons in multiple groups. Data are presented as the means ± SEM, and P values of <0.05 were considered statistically significant.


Supplementary material for this article is available at

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.


Acknowledgments: We are grateful to Z. Liu, Q. Chen, and C. Wang for help and suggestions. We also thank J. L. Brash for the valuable advice and for proofreading the manuscript. Many thanks to R. Chen and J. Wang for assistance in polymer synthesis and suggestions in manuscript revision. Funding: This study was supported by the National Natural Science Foundation of China (nos. 21935008 and 21774084). Author contributions: L.Y., G.C., and H.C. designed the research. L.Y., R.F., L.Z., Q.H., and J.C. performed research. L.Y., R.F., L.Z., Y.G., Y.L., and Z.Z. analyzed data. L.Y., R.F., L.Z., Z.Z., G.C., and H.C. wrote the paper. Competing interests: H.C. and L.Y. are inventors on a pending patent related to this work filed by Soochow University (no. 201910654432.7, filed on 19 July 2019). The authors declare that they have no other competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.
View Abstract

Stay Connected to Science Advances

Navigate This Article