Local CpG-Stat3 siRNA treatment improves antitumor effects of immune checkpoint inhibitors

Abstract

Immune checkpoint blockade (ICB) therapy has significantly benefited patients with several types of solid tumors and some lymphomas. However, many of the treated patients do not have a durable clinical response. It has been demonstrated that rescuing exhausted CD8⁺ T cells is required for ICB-mediated antitumor effects. We recently developed an immunostimulatory strategy based on silencing STAT3 while stimulating immune responses by CpG, a ligand for Toll-like receptor 9 (TLR9). The CpG-small interfering RNA (siRNA) conjugates efficiently enter immune cells, silencing STAT3 and activating innate immunity to enhance T cell-mediated antitumor immune responses. In the present study, we demonstrate that blocking STAT3 through locally delivered CpG-Stat3 siRNA enhances the efficacies of the systemic PD-1 and CTLA4 blockade against mouse A20 B cell lymphoma. In addition, locally delivered CpG-Stat3 siRNA combined with systemic administration of PD-1 antibody significantly augmented both local and systemic antitumor effects against mouse B16 melanoma tumors, with enhanced tumor-associated T cell activation. Furthermore, locally delivered CpG-Stat3 siRNA enhanced CD8⁺ T cell tumor infiltration and antitumor activity in a xenograft tumor model. Overall, our studies in both B cell lymphoma and melanoma mouse models demonstrate the potential of combinatory immunotherapy with CpG-Stat3 siRNA and checkpoint inhibitors as a therapeutic strategy for B cell lymphoma and melanoma.

Graphical abstract

Yu, Zhang, and colleagues demonstrate here that locally delivering CpG-Stat3siRNA enhances the efficacies of systemic PD-1 and/or CTLA4 blockade in two mouse tumor models and a xenograft tumor model. Local injection of CpG-Stat3siRNA also increases PD-1 and/or CTLA4 blockade-induced antitumor CD8⁺ T cell activity.

Introduction

Therapeutic immune checkpoint blockade (ICB) has profoundly and positively impacted B cell lymphoma and melanoma treatment. However, many treated patients do not respond, and those who have a good initial response may not have durable clinical benefits.^1,2,3 The limited responses to ICB in patients with cancer are largely attributed to a lack of interferon (IFN) signaling and activation of CD8⁺ T cells even after the immune checkpoint is removed.^4,5,6,7 Extensive studies from our group and others have demonstrated that IFNγ, which is necessary for PD-1-directed therapy responses,^8,9 is inhibited in tumor-associated immune cells by STAT3.¹⁰ STAT3 is also known to promote tumor cell proliferation, survival, and invasion in diverse cancers.^{11,12,13,14,15,16} The immunosuppressive role of STAT3 in tumor cells, T cells, B cells, myeloid cells, and dendritic cells (DCs) has been documented extensively.^{10,11,12,13,14,15,16,17,18,19,20,21,22,23} Additionally, STAT3 signaling in CD4⁺ T cells promotes regulatory T cell (Treg) accumulation in tumors while inhibiting CD8⁺ effector T (T_EFF)tumor infiltration and antitumor immunity.^12,18

STAT3 as a target for inhibiting B cell lymphoma cell survival and activating antitumor immune response has been documented.^24,25,26,27 We previously developed an immunostimulatory strategy by linking Stat3 siRNA with CpG, the ligand for Toll-like receptor 9 (TLR9).^{20,25,28,29,30} CpG not only facilitates the delivery of the small interfering RNA (siRNA) but also, in the absence of STAT3 activity, activates potent antitumor immune responses.³⁰ We further showed that both CpG-Stat3 siRNA and CTLA4 antibody inhibit STAT3 in B malignant cells, leading to tumor cell apoptosis and/or proliferation inhibition.^25,31 Additionally, CpG-Stat3 siRNA activates antitumor T cells by blocking STAT3 in macrophages, B cells, and DCs, while CTLA4 or PD-1 antibody also enables interaction between DCs and T cells to activate antitumor T cell immunity.^20,32,33 Thus, combinatory treatments with CpG-Stat3 siRNA and CTLA4 or PD-1 antibody may significantly boost the antitumor efficacies of CTLA4 or PD-1 antibody in treating B cell lymphoma and melanoma.

Local treatment can lead to direct antitumor effects at the injected tumor sites and induction of systemic antitumor immune responses.^34,35,36,37 Both lymph-node-resident B cell lymphoma and cutaneous and subcutaneous melanoma tumors provide the opportunity for local treatment for easier delivery and generally lower toxicity. Previously, we showed that local treatments with CpG-Stat3 siRNA inhibit both B cell lymphoma²⁵ and melanoma tumor growth²⁰ while resulting in effective tumor growth inhibition, including complete tumor eradication, when combined with localized radiation therapy.^25,38 However, whether CpG-Stat3 siRNA can boost the efficacies of ICB in these two tumor types remains unknown. The CpG oligonucleotide TLR9 ODN 1826 agonist has been shown to effectively augment the therapeutic potential of checkpoint blockade through locally delivered ODN 1826 and systemic CTLA4 or PD-1 blockade antibody in the B16 melanoma mouse model.³⁷ However, high CpG dosing (30 μg/intra-tumorally) was required to enhance the ICB response in this study.³⁷ High-dose CpG injections induced sharp and sustained increased local and serum levels of interleukin (IL)-12, IL-6, and tumor necrosis factor alpha (TNF-α), followed by high levels of the acute-phase proteins serum amyloid A³⁹ and serum amyloid P (SAP), which could cause strong side effects.^40,41 Therefore, we hypothesized that CpG-Stat3 siRNA, at significantly lower and safer effective CpG concentrations, may enhance the antitumor efficacy of ICB by blocking STAT3 in the tumor-associated myeloid cells, thereby stimulating antigen presentation and activating CD4⁺ and CD8⁺ T cell-mediated antitumor immune responses. Our study shows that at relatively low CpG-Stat3 siRNA concentrations, local injection can augment the therapeutic potential of PD-1 antibody not only in CpG-Stat3 siRNA-treated tumors but also in distal non-treated tumors. In the xenograft tumor model, CpG-Stat3 siRNA local injection can enhance the antitumor effects induced by systemic anti-PD-1 antibody treatment.

Results

Local CpG-STAT3 siRNA treatment enhances anti-lymphoma effects of CTLA4 and PD-1 blockade

Although some advances have been made in treating patients with cancer with ICB, the efficacy is limited, which is contributed to by the lack of CD8⁺ T cell activation.^2,6 We have shown previously that CpG-Stat3 siRNA treatment in vivo silences Stat3 in DCs, B cells, and macrophages, leading to T cell activation and potent antitumor immunity.²⁰ We further showed that both CpG-Stat3 siRNA and CTLA4 antibody inhibit STAT3 in B malignant cells, leading to tumor cell apoptosis and/or growth inhibition.^25,31 Here, we assessed whether silencing Stat3 with CpG-Stat3 siRNA can significantly enhance the efficacies of CTLA4 antibody therapy in B cell lymphoma by augmenting the immunostimulatory effects. For our experiments, we selected the commonly used mouse A20 B cell lymphoma model, which exhibits TLR9 expression and constitutive Stat3 activation.²⁵ We have previously shown that CpG-Stat3 siRNA efficiently targeted A20 tumor cells to silence Stat3.^20,25 To evaluate gene silencing effect of CpG-Stat3 siRNA, we used quantitative real-time PCR (real-time qPCR) analysis of Stat3 mRNA of CpG-Stat3 siRNA-treated A20 tumor cells. The effect of Stat3 gene silencing was observed in cultured A20 cells at 48 h since the start of CpG-Stat3 siRNA treatment (Figure 1A). Next, we tested the feasibility of using this strategy for targeting Stat3 in vivo. BALB/c mice with established, subcutaneously engrafted (s.c.-engrafted) A20 lymphoma tumors were treated with PBS or CpG-Stat3 siRNA at 12.5 or 25 μg/mouse by daily intra-tumoral (IT) injections. One day after the third injection of PBS or CpG-Stat3 siRNA, mice were euthanized, and the tumor tissues were harvested to assess Stat3 mRNA expression by real-time qPCR. The CpG-Stat3 siRNA treatment reduced the Stat3 mRNA expression in a dose-dependent manner (Figure 1B).

Figure 1.

CpG-Stat3 siRNA gene silencing effect in A20 B cell lymphoma cells and therapeutic effect of CpG-Stat3 siRNA with or without CTLA4 or PD-1 blockade

(A) A20 cells were cultured in the presence of CpG-Stat3 siRNA or CpG-Luc-siRNA for 48 h. CpG-Stat3 siRNA silences Stat3 gene expression at the mRNA level, as shown by qPCR. Data are shown as means ± SEM, n = 3. (B) BALB/c mice with subcutaneous (s.c) A20 lymphoma tumors were treated by intra-tumoral injections of CpG-Stat3 siRNA using the indicated amounts of PBS. CpG-Stat3 siRNA gene silencing effect is shown by qPCR. Data are shown as means ± SEM, n = 3 mice. (C and D) BALB/c mice with s.c. A20 lymphoma tumors were treated by intra-tumoral injections of CpG-Stat3 siRNA, i.p injection of anti-CTLA4 or anti-PD-1 antibodies, or combination treatment with CpG-Stat3 siRNA and anti-CTLA4 or anti-PD-1 antibodies every other day, starting 9–10 days after tumor implantation (5 × 10⁵ A20 cells/tumor). Data are shown as means ± SEM, n = 5. Student’s t test or one-way ANOVA was used for statistical analysis (∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001). Tumor size was monitored every other day. (E) In vivo study design for (C) and (D).

We next tested whether combining CpG-Stat3 siRNA with CTLA4 or PD-1 inhibition would lead to superior antitumor effects relative to the single agents in the mouse B cell lymphoma. As shown in Figures 1C and 1D, silencing Stat3 through locally delivered CpG-Stat3 siRNA significantly enhanced the antitumor efficacies of systemic administration of PD-1- and CTLA4-specific antibodies in mice bearing the A20 B cell lymphoma arresting tumor growth. However, locally delivered CpG-scramble siRNA did not significantly increase the antitumor effects of PD-1 or CTLA4 antibody (Figure S1).

Combining CpG-Stat3 siRNA with CTLA4 or PD-1 blockade induces significantly higher activities of tumor-infiltrating T cells

Tumor infiltration of activated cytotoxic CD8⁺ T cells is critical for the successful outcome of ICB therapy.^2,6,42 Therefore, we determined whether T cell activation is a critical contributor for the antitumor effect of combinatory immunotherapy with CpG-Stat3 siRNA and checkpoint blockades. Combinatory treatment with CpG-Stat3 siRNA and CTLA4 antibody in A20 lymphoma tumor-bearing mice significantly increased the percentages of IFNγ and/or granzyme B (GZMB) producing CD4⁺ T cells and CD8⁺ T cells in tumors compared to either CpG-Stat3 siRNA or CTLA4 antibody alone (Figures 2A–2C). Similarly, combinatory treatments with CpG-Stat3 siRNA and PD-1 antibody in A20 lymphoma tumor-bearing mice led to significantly higher percentages of activated CD8⁺ T cells positive for IFNγ, CD107α, which measures T cell cytotoxicity, and GZMB at the tumor sites (Figures 2D–2F). Furthermore, combinatory treatment led to the inhibition of FoxP3⁺ Tregs (Figure 2G). These results support that local CpG-Stat3 siRNA treatment enhances antitumor effects of CTLA4 or PD-1 blockade by increasing the antitumor effector functions of tumor-infiltrating T cells.

Figure 2.

Local CpG-Stat3 siRNA treatment significantly enhances the antitumor effector functions of tumor-infiltrating T cells in mice treated with CTLA4 or PD-1 blockade

Single-cell suspensions prepared from A20 tumors from mice receiving indicated treatments were analyzed by flow cytometry for IFNγ (A, B, and D), CD107α (E), granzyme B (C and F), and FoxP3 (G) expression in T cells. Flow cytometry data showing IFNγ⁺ or GzmB⁺ cell frequencies in tumor-infiltrating CD4⁺ or CD8⁺ T cells. Data are shown as means ± SEM (n = 3, n is for number of samples, each of which was pooled from 2–3 mice), One-way ANOVA was used for statistical analysis (∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001).

Local CpG-STAT3 siRNA combined with PD-1 antibody results in systemic antitumor effects against melanoma in mice

We previously showed that local treatments with CpG-Stat3 siRNA inhibit melanoma tumor growth by silencing Stat3 and activating tumor-infiltrating immune cells.²⁰ However, the ideal cancer therapy should not only inhibit local tumor regression but also induce a systemic antitumor immunity that could effectively eradicate distant metastases. Whether CpG-Stat3 siRNA can enhance both local and systemic antitumor responses of ICB in melanoma remains unknown. Therefore, we evaluated whether combinatory treatment with locally delivered CpG-Stat3 siRNA and systemic PD-1 antibody treatment could lead to both local and systemic antitumor effects. We used a bilateral tumor model in which the C57BL/6 mice were challenged at both flanks by mouse B16 melanoma tumor cells through s.c. injection. Only one tumor was treated with locally delivered CpG-Stat3 siRNA to assess local and systemic effects of single and combinatory treatments. Our results show that local CpG-Stat3 siRNA treatment suppressed the growth of both treated and non-treated distal tumors (Figure 3), suggesting the generation of systemic antitumor immune responses. In addition, combination treatment with CpG-Stat3 siRNA and systemic PD-1 antibody significantly enhanced the systemic antitumor responses compared to either CpG-Stat3 siRNA or PD-1 antibody treatment alone (Figure 3). These findings provide evidence that CpG-Stat3 siRNA local treatment can enhance the PD-1 antibody-mediated systemic antitumor response.

Figure 3.

Combination treatment with intra-tumoral CpG-Stat3 siRNA and systemic PD-1 antibody reduces growth of both treated and distal tumors

C57BL/6 mice were injected with 2 × 10⁵ B16 melanoma cells on both right and left flanks. The left flank tumors were treated by intra-tumoral injections of CpG-Stat3 siRNA, i.p injection of anti-PD-1 antibody, or CpG-Stat3 siRNA and anti-PD-1 every other day, starting day 7 post-tumor challenge (2 × 10⁵ B16-F10 cells/tumor), n = 7–8. Both treated tumor (A) and distal tumor (B) size was monitored every other day. Data are shown as means ± SEM, Student’s t test and one-way ANOVA were used for statistical analysis (∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001). (C) In vivo study design for (A) and (B).

Combination of the local CpG-Stat3 siRNA treatment with PD-1 blockade promotes tumor T cell recruitment and activity

To assess the effect of CpG-Stat3 siRNA/anti-PD1 combination treatment on T cell tumor infiltration and activation, we performed experiments using bilateral B16 tumors as described above. On day 15 after tumor implantation, we harvested the tumor tissues and prepared single-cell suspension, followed by antibody staining and flow cytometric analysis. Compared to the PBS control, local CpG-Stat3 siRNA treatment enhanced the tumor infiltration of CD8⁺ T cells (Figure 4A). IT ratios of CD8⁺ T cells vs. Tregs in treated tumors were significantly increased (Figure 4B). Consistent with the antitumor effects of each treatment in Figure 3, either local CpG-Stat3 siRNA or systemic PD-1 antibody treatment improved CD8⁺ T cell activity with increased IFNγ and GZMB producing CD8⁺ T cell frequency, but the most significant enhancement of the activated CD8⁺ T cells in bilateral tumors was observed in mice given the combinatory treatment (Figure 4C). These results indicated that the combination treatment with local CpG-Stat3 siRNA and systemic PD-1 antibody increases the tumor infiltration of activated T cells.

Figure 4.

Intra-tumoral CpG-Stat3 siRNA and systemic PD-1 antibody combined treatment significantly increases CD8⁺ T cell tumor infiltration and activity in treated and distal tumors

Single-cell suspensions from the B16 tumors from mice with indicated treatments were analyzed by flow cytometry for CD8⁺ (A) and CD8⁺/CD4⁺FoxP3⁺ immune cells (B) as well as IFNγ⁺ or GzmB⁺ CD8⁺ (C) cells from the indicated treatments. Data were shown as means ± SEM, n = 3 (n is for number of samples, each of which was pooled from 2–3 mice). One-way ANOVA was used for statistical analysis (∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001).

Local CpG-Stat3 siRNA treatment systematically increases IFNγ and GZMB production

In addition to immune cells, cytokines also play a critical role in antitumor immune therapy. Both IFNγ and GZMB are well known for their direct cytotoxic effects on tumor cells and being able to stimulate production of immune-stimulating cytokines.^43,44,45 Thus, we co-cultured B16 tumor cells with splenic cells from B16 tumor-bearing mice receiving various treatments. 3 or 4 days after co-culturing, the levels of IFNγ and GZMB in the cultured cell supernatants were measured by ELISA. Both IFNγ and GZMB were noticeably increased, particularly in the combinatory treatment group at day 4 after co-culturing, although there was also a significant induction of IFNγ and GZMB in the local CpG-Stat3 siRNA treatment group (Figure S2). These findings suggest that local CpG-Stat3 siRNA treatment induces and enhances systemic IFNγ and GZMB production triggered by PD-1 antibody treatment.

Local CpG-Stat3 siRNA treatment enhances PD-1 blockade-induced antitumor effects in a melanoma xenograft tumor model

NSG mice were implanted with a mixture of human peripheral blood mononuclear cells (hPBMCs) and human melanoma A2058 cells, followed by local CpG-Stat3 siRNA injection or systemic PD-1 antibody treatment, alone or in combination. Compared to either local CpG-Stat3 siRNA or systemic PD-1 antibody treatment alone, the combinatory treatment resulted in significantly more tumor inhibition (Figure 5A). Moreover, decreased tumor cell proliferation was observed in the mice given the combinatory treatment, as indicated by reduced Ki67⁺ cells in HMB-45⁺ tumor areas (HMB45 is a human melanoma marker) (Figure S3). This is accompanied by significantly increased infiltration of GZMB⁺ CD8⁺ T cells in the tumors in mice treated with the local CpG-Stat3 siRNA and systemic PD-1 antibody combination (Figure 5C). In addition, CpG-Stat3 siRNA local treatment induced p-STAT3 downregulation in CD11b⁺ myeloid cells (Figure S4). These findings further suggest that CpG-Stat3 siRNA local treatment can enhance the PD-1 antibody-mediated systemic antitumor response in a human melanoma xenograft model.

Figure 5.

Local CpG-Stat3 siRNA treatment enhances PD-1 blockade-induced antitumor effects in a melanoma xenograft tumor model

(A) Freshly isolated hPBMCs were mixed with A2058 tumor cells at 1:4 (hPBMCs:A2058 cells) and implanted subcutaneously in the left flank of NSG mice. The mice were treated by intra-tumoral injections of CpG-Stat3 siRNA or i.p injections of anti-PD-1 antibody, alone or in combination, every other day, starting day 15 post-tumor challenge (5 × 10⁶ A2058 tumor cells and 1.25 × 10⁶ hPBMCs/tumor), n = 8. Tumor size was monitored. Data are shown with means ± SEM, Student’s t test was used for statistical analysis (∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001). (B) In vivo study design for (A). (C) Representative immunofluorescence images of CD8 (green), granzyme B (red), and Hoechst (blue) on the tumor tissue sections as indicated. Scale bars represent 50 μm.

Discussion

Despite the promising outcomes of anti-CTLA4 and anti-PD-1 antibodies in treating cancer, including solid tumors and hematopoietic malignancies,^1,3,6,46 many of the treated patients do not respond well to the ICB, and those who initially do may not have durable clinical benefits.^1,2,3 Thus, there is an urgent need to develop immunotherapeutic strategies to improve the efficacy of ICB. The limited responses to ICB in patients with cancer, including those with B cell lymphoma and melanoma, are largely attributed to the lack of IFNγ production required for the activation of CD8⁺ T cells.^4,5,6,7 Extensive studies from our group have demonstrated that STAT3, which is persistently activated in multiple types of immune cells in the tumor microenvironment, contributes to the suppression of IFNγ production in the tumor microenvironment.^10,20,23

In this study, we evaluated the antitumor effects of local CpG-Stat3 siRNA treatment at tumor sites to boost the antitumor effects of CTLA4 and/or PD-1 antibody systemic treatments in the A20 mouse lymphoma model and the B16 melanoma model. Using primary tumors as the injection site has been well documented.^36,47,48 This approach has several merits, including easier delivery, reduced drug dosing and associated toxicity, and having the potential to stimulate systemic antitumor immune responses. Our results demonstrate that combining CpG-Stat3 siRNA with CTLA4 or PD-1 antibody suppressed A20 lymphoma tumor growth more effectively than either CpG-Stat3 siRNA or CTLA4 or PD-1 antibody alone. In the case of the B16 melanoma tumor model, we showed that CpG-Stat3 siRNA local treatment significantly potentiates the antitumor effects of PD-1 antibody, not only at the tumors that have received the treatment but also at non-treated distal tumors. The antitumor effects on the distal non-treated tumors provide evidence supporting the activation of systemic antitumor immune responses. The detection of increased tumor CD8⁺ T cell infiltrates that are activated and able to produce IFNγ and GZMB and reduced tumor-associated Tregs when CpG-Stat3 siRNA is added to ICB therapy show that CpG-Stat3 siRNA can serve as an effective adjuvant for ICB. The finding that CpG-Stat3 siRNA can elevate IFNγ and GZMB production by splenic cells in mice receiving anti-PD-1 treatment further supports that CpG-Stat3 siRNA can boost the antitumor efficacy of anti-PD-1 treatment. We have noted that the antitumor effectiveness is stronger at the treated tumor than the distal tumor. Similar findings were reported,^49,50 likely due to concentrated immunostimulatory factors produced at the tumor site receiving CpG-Stat3siRNA, attracting more antigen-presenting cells and other immune cells, including both innate immune cells and T cells.^20,49,50

Pioneering studies by others have demonstrated CpG as an immunostimulatory molecule that activates innate immunity and induces systemic immune response through TLR9 in solid and blood cancer, including B cell lymphoma and melanoma.^{37,48,50,51,52} However, high CpG dosing (30–100 μg/intra-tumorally) is required to enhance the antitumor immune response in these studies.^37,48,50,51 It has been demonstrated that high-dose CpG injections can cause strong side effects associated with innate immune cell activation or toxicities related to chemical modification of the oligonucleotide, such as phosphorothioation.^40,41,52 CpG in CpG-Stat3 siRNA acts as not only an immune-stimulatory molecule but also a carrier for Stat3 siRNA. In the present study, we showed that a local injection of relatively low-dose (0.5–2.5 mg/kg body weight) CpG-Stat3 siRNA, in which CpG represents roughly a quarter of the conjugate, is effective in increasing both CTLA4 and PD-1 antibody-mediated antitumor immune responses, including recruitment and activation of tumor CD8⁺ T cells, reduction of tumor-infiltrating CD4⁺ Tregs, and improved antitumor effects. Stat3 gene silencing by Stat3 siRNA in the CpG-Stat3 siRNA conjugate has been shown to reduce the expression of immunosuppressive factors while increasing expression of immunostimulatory molecules including IFNγ.^20,29 Because of this, it has been shown that CpG-Stat3 siRNA is more effective in activating antitumor immune responses than CpG stimulation alone.²⁰ In addition to CpG, immune checkpoint antibodies are associated with autoimmune toxicities.⁵³ Recently, similar therapies in cancer treatment have been reported.^54,55 One study showed that RNAi-mediated silencing of STAT3 and PD-L1 led to tumor growth inhibition by activating tumor-infiltrating immune cells, while RNAi-mediated silencing of STAT3/PD-L1 in tumor-associated immune cells induces robust antitumor effects in immunotherapy resistant tumors.⁵⁴ Another publication shows that silencing STAT3 via peptidomimetic lipid nanoparticles (LNP)-mediated systemic delivery of RNAi downregulated PD-L1, resulting in the inhibition of melanoma growth.⁵⁵ Although there are some similarities to these reports, our current study shows that local injection of CpG-Stat3 siRNA can boost the effects PD-1 antibody systemic treatment at reduced doses (100 vs. 200 μg per mouse, which is expected to reduce toxicity) to induce systemic antitumor immune responses.

Tumor cells from melanoma and many types of B cell lymphomas require persistently activated STAT3 for growth and/or survival.^25,26,56,57 Our previous studies have shown that TLR9 is required to process CpG-Stat3 siRNA to be an effective siRNA.⁵⁸ B cell lymphoma cells, including mouse A20 tumor cells, express TLR9.^25,26 Follow-up studies are needed to demonstrate the contribution of Stat3 silencing in the tumor cells in enhancing the efficacies of ICB in A20 tumor model. Several tumor cells, especially cancer stem cells, in some solid tumors, including glioma, express TLR9.²⁸ The glioma cancer stem cells can also process CpG-Stat3 siRNA into effective siRNA.²⁸ Nevertheless, the B16 melanoma cells do not display TLR9.²⁰ Therefore, at least in the B16 tumor model, CpG-Stat3 siRNA-induced enhancement of the antitumor effects of PD-1 is largely contributed by immune activation.

Taken together, the results of our study showed that local administration of low-dose CpG-Stat3 siRNA can enhance the therapeutic potential of IBC in B cell lymphoma and melanoma mouse tumor models, supporting further testing and development of CpG-Stat3 siRNA and ICB combinatory treatment for clinical application.

Materials and methods

Cell lines

Mouse lymphoma cell line A20 (American Type Culture Collection [ATCC]; A-20) was cultured in RPMI containing 10% fetal bovine serum (FBS; Omega Scientific) and 1× antibiotic antimycotic (AA; Gibco), supplemented with 0.05 mM 2-mercaptoethanol (2-ME; Gibco). Either mouse melanoma cell line B16 (ATCC) or human melanoma cell line A2058 (ATCC) was cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% FBS and 1× AA.

In vivo mouse experiments

Mouse care and experimental procedures were performed under pathogen-free conditions in accordance with established institutional guidance and approved protocols (IACUC: 08026) from the Institutional Animal Care and Use Committee at the Beckman Research Institute of the City of Hope Medical Center. C57BL/6 and BALB/c mice were obtained from Jackson Laboratory. 8- to 10-week-old female NSG mice from The Jackson Laboratory or animal resource core facility were used to establish the human melanoma xenograft model. For the A20 lymphoma mouse model, we injected 5 × 10⁵ A20 tumor cells s.c. into BALB/c mice. When the tumors reached an average diameter of approximately 6–7 mm (9–10 days after A20 tumor cell injection), mice with similar average tumor sizes were randomly divided into four groups. Then, we treated A20 tumor-bearing mice with IT injections of CpG-Stat3 siRNA (0.5 mg/kg body weight), intraperitoneal (i.p.) injections of CTLA4/PD-1 antibody (100 μg/mouse), or combination treatment with CpG-Stat3 siRNA and CTLA4 (clone 9D9, BioXcell)/PD-1(clone 29F.1A12 BioXcell) antibody every other day. For the IT injection, we used a fine (27G) needle to deliver a minimal volume of 50 μL per tumor. Tumor growth was monitored every other day with caliper measurement. For the B16 melanoma mouse model, C57BL/6 mice were s.c. injected with 2 × 10⁵ B16 tumor cells on the right and left flanks. The left flank tumors were treated by IT injections of CpG-STAT3 siRNA (0.5 mg/kg body weight), i.p injection of PD-1 antibody (100 μg/mouse), or combination treatment with CpG-STAT3 siRNA and αPD1 every other day, starting 7 days after challenge with 2 × 10⁵ B16 cells. Tumor growth was monitored as described above. For the human melanoma xenograft model, freshly isolated hPBMCs were mixed with cultured A2058 tumor cells at 1:4 (hPBMCs:A2058 cells) and implanted s.c. into the left flank of NSG mice. The tumors were treated by IT injections of CpG-Stat3 siRNA (2.5 mg/kg, body weight), i.p injection of anti-PD-1 antibody (100 μg/mouse), or CpG-Stat3 siRNA and anti-PD-1 every other day, starting day 15 post-tumor challenge (1.25 × 10⁶ hPBMCs and 5 × 10⁶ A2058 tumor cells). Tumor size was monitored every other day.

Oligonucleotide design and synthesis

The sequences of mouse cell-specific CpG1668(B)-siRNAs and the chemical modification have been described previously.²⁰ The sequences of single-stranded constructs are listed below.

Mouse Stat3 siRNA (SS): 5′-CAGGGUGUCAGAUCACAUGGGCUAA-3′.

CpG1668-mouse Stat3 siRNA (AS): 5′-TCCATGACGTTCCTGATGCT-linker-UUAGCCCAUGUGAUCUGACACCCUGAA-3′.

Human Stat3 siRNA (SS): 5′-GGAAGCUGCAGAAAGAUACGACUGA-3′.

CpG(A)-STAT3 siRNA (AS): 5′-G∗G∗TGCATCGATGCAGG∗G∗G∗G∗G-linker-UCAGUCGUAUCUUUCUGCAGCUUCCGU-3′.

Real-time qPCR

Total RNAs from various cell populations were purified with the RNeasy system according to the manufacturer’s instructions (QIAGEN). RNA (0.5–1 μg) was reverse transcribed to cDNA using the iScript cDNA Synthesis Kit (Bio-Rad), and real-time PCR reactions were performed as described previously.¹⁰ Specific primers for mouse Stat3 (Qiagen, PPM04643F), Actin (Qiagen, PPM02945B), and Gapdh (Qiagen, PPM02946E) were purchased from SA Bioscience and Qiagen. Each primer set was validated using a standard curve across the dynamic range of interest with a single melting peak. Samples were run in triplicate and expressed as means ± standard error of the mean (SEM).

hPBMC preparation

The use of anonymous discarded blood samples was approved by the City of Hope Institutional Review Board and exempt from the informed consent requirement. PBMCs were isolated from human blood by density gradient centrifugation using Ficoll-Paque Plus (GE Healthcare Biosciences).

Immunofluorescent staining and confocal microscopy

Mouse tumor tissue sections were obtained from melanoma tumor-bearing NSG mice and subsequently embedded in optimal cutting temperature (OCT) block, stained for immunofluorescence with specific primary and fluorophore-conjugated secondary antibodies, and imaged by confocal microscopy as previously described.¹⁰ Briefly, the tissue sections from the OCT-frozen tumor tissues were fixed in 4% paraformaldehyde, permeabilized with methanol, and blocked in PBS containing 5% goat serum. Samples were stained overnight at 4°C with the following primary antibodies: HMB45 (ab787, Abcam), Ki67 (VP-RM04, Vector), CD11b (clone LM2/1.6.11, Santa Cruz Biotechnology), CD8 (clone SP16, Thermo Scientific), GZMB (Cell Signaling Technology), and p-STAT3 (clone D3A7, Cell Signaling Technology). The next day, the slides were incubated with secondary antibodies for 2 h (Alexa Fluor 488 goat anti-rabbit, Alexa Fluor 555 goat anti-mouse, Thermo Scientific). Afterward, slides were mounted, and confocal imaging was performed with a Zeiss LSM 880 confocal microscope. Staining quantification was performed by ZEN 2.3 lite software and plotted in GraphPad Prism 9. For Ki67 analysis, the nuclear positive percentage within the tumor area was calculated using the software QuPath-0.4.2.

Intracellular staining and flow cytometry

Antibodies

Fluorochrome-conjugated monoclonal antibodies against CD45 (clone I3/2.3), CD4 (clone GK1.5), and CD8 (clone 53-5.8) were purchased from BioLegend (San Diego, CA, USA). Antibodies against FoxP3 (clone FJK-16s), CD107α (clone 1D4B), GZMB (clone OA16A02), and IFNγ (clone XMG1.2) were obtained from eBioscience and BioLegend.

Single-cell suspension preparation

To prepare single-cell suspensions, tumor tissue was dissected into approximately 1–5 mm³ fragments and digested with collagenase type D (2 mg/mL; Roche) and DNase I (1 mg/mL; Roche) for 30–45 min at 37°C. Digests were filtered through 70 μm cell strainers and centrifuged at 1,500 rpm for 5 min. Single-cell suspensions from spleens were prepared as mentioned above. After red blood cell lysis (Sigma-Aldrich), single-cell suspensions were filtered, washed, and resuspended in fluorescence-activated cell sorting (FACS) buffer (2% FBS in Hank’s balanced salt solution without Ca⁺, Mg⁺, or phenol red).

Intracellular staining and flow cytometric analysis were performed as previously described.¹⁰ Single-cell suspensions (some of which were pooled from tumors harvested from 2–3 mice) were stimulated for 5 h with PMA (5 ng/mL, Sigma) and ionomycin (500 ng/mL, Sigma) in the presence of a protein transport inhibitor (monensin 1,000×, BioLegend). Cells were blocked with anti-CD16/CD32 antibody (TruStain FcX PLUS anti-mouse CD16/32, BioLegend) and incubated for 15 min on ice with PECy7-, Alexa Fluor 700-, Pacific Blue- (or v450), and APC-Cy7 (or Alexa Fluor-e780)-conjugated antibodies (1:100, CD4, CD8) purchased from BioLegend. After cell surface marker staining, cells were fixed and permeabilized using the BD Cytofix/Cytoperm Fixation/Permeabilization Solution Kit (BD) for IFNγ and GZMB staining or eBioscience Foxp3/Transcription Factor Staining Buffer Set (eBioscience) for FoxP3 staining. Cells were then incubated with FITC and PE-conjugated antibodies (1:100, IFNr or FoxP3, GZMB) purchased from BioLegend. Aqua LIVE/DEAD, used for cell viability, was purchased from Invitrogen. Cells were washed twice before analysis on the BD LSR Fortessa flow cytometer (Beckman Coulter Genomics).

Cytokine measurement

Splenic cells (1 × 10⁵) from B16 tumor-bearing mice were co-cultured with B16 tumor cells (2 × 10⁴) in a well of the 96-well plate for 3 or 4 days. Culture medium was collected to determine IFNγ and GZMB levels by ELISA (R&D).

Statistical analysis

Statistical analyses were performed using GraphPad Prism 7 software. Statistical comparisons between groups were performed using the unpaired Student’s t test to calculate two-tailed p values. Statistical significance values were set as ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Multiple group comparisons were conducted using one-way ANOVA. A p value less than 0.05 would be considered statistically significant, and ns means not significant. Data are presented as mean ± SEM. p value and n can be found in the main and supplemental figure legends.

Data and code availability

All reagents and data generated from this study are available from the corresponding author upon reasonable request.

Acknowledgments

We thank the staff members at the animal facilities at City of Hope for their dedication. We also acknowledge the contribution of staff members at the Analytical Cytometry Core and DNA/RNA Synthesis Core. This work is supported by the National Cancer Institute of the National Institutes of Health under grant numbers P50CA107399 (S.F.) and P30CA033572 (City of Hope). This work is also supported through sponsored research from Scopus Biopharma, Inc. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This study was also supported by the Billy and Audrey Wilder Endowment to H.Y. The diagrams in Figures 1E, 3C, and 5B were created with BioRender.com.

Author contributions

H.Y. and C.Z. developed the concept, designed the experiments, and prepared and wrote the manuscript. C.Z. provided guidance and carried out the experiments and statistical analyses. R.H. contributed to animal experiments and immunofluorescence staining. L.R. performed animal experiments, real-time qPCR, and flow cytometry experiments. A.M. performed immunofluorescence staining. J.S. performed animal experiments and the real-time qPCR assay. P.S. synthesized CpG-Stat3 siRNA. S.F. provided clinical insight on B cell lymphoma treatment, and M.K. contributed to the manuscript writing and discussion of the project.

Declaration of interests

M.K. and H.Y. are on the scientific advisory board of Scopus Biopharma, Inc., a licensee of the CpG-STAT3siRNA technology, with stock options. C.Z. also received stocks through CpG-STAT3siRNA technology licensing. M.K. is a scientific advisor to Duet Biotherapeutics, Inc.