Mechanism of action and therapeutic utility of stimulatory CpG oligonucleotides

细菌DNA中存在的未甲基化的CpG基序与Toll样受体9相互作用，触发促炎免疫反应。CpG DNA还可改善抗原提呈细胞功能，从而促进获得性免疫的发展。我的实验室证实，表达免疫刺激CpG基序的人工合成寡核苷酸(CpG ODN)可以与凋亡的肿瘤细胞偶联，生成可被专业APC迅速内化的肿瘤疫苗，促进DC成熟，并增强肿瘤特异性免疫的诱导。在多个小鼠模型中，我们发现接种CpG结合的凋亡细胞疫苗显着降低了对肿瘤攻击的易感性。这一效应在接种疫苗后的小鼠和在肿瘤攻击后免疫的动物中都观察到了。不幸的是，随着癌症负担的增加，这些疫苗根除肿瘤的能力减弱了。我们发现，这反映了大型已建立的肿瘤产生免疫抑制微环境的能力，该微环境能够抑制干扰CpG介导的免疫治疗的Ag特异性细胞反应。髓系来源的抑制细胞(MDSC)是这种免疫抑制环境的重要组成部分。大量的MDSC存在于已建立的肿瘤及其附近，并已被证明能抑制抗原特异性T细胞和NK细胞的活性。我们随后的研究表明，当CpG ODN注射到肿瘤本身时，单核细胞MDSC(MMDSC)的免疫抑制活性显著降低。我们假设，通过TLR9结扎提供的信号足以克服肿瘤细胞提供的抑制信号。在小鼠中，mMDSC表达TLR9，并对CpG刺激作出反应：1)失去抑制T细胞功能的能力；2)产生Th1细胞因子；3)分化为具有杀瘤能力的M-1样巨噬细胞。TLR7激动剂也观察到类似的活性，TLR7和TLR9激动剂的联合作用明显更有效。事实上，肿瘤内注射这种激动剂组合甚至导致了即使是较大的已建立的肿瘤的消退(最初治疗时为500mm3)。我们将这些研究扩展到评估TLR激动剂对从正常人供体和癌症患者中纯化的mMDSC的影响。结果表明，经TLR7/8激动剂刺激后，人骨髓间充质干细胞成熟，失去免疫抑制活性，在人骨髓间充质干细胞上恢复了TLR9结扎对小鼠骨髓间充质干细胞的作用。我们的结论是，可以利用TLR配体的组合来实现两个独立但相互支持的功能：提高抗肿瘤疫苗的效力和降低肿瘤部位细胞的活性，否则将降低这种抗肿瘤反应的有效性。我们随后的研究试图确定将TLR配体输送到肿瘤结节的最佳方法和治疗窗口。我们发现，吸附在微粒上的激动剂可以用在肺癌小鼠身上，当地的DC和巨噬细胞将这些微粒输送到肿瘤床上，在那里它们可以有效地防止肿瘤的生长。我们正在探索其他配方，旨在优化TLR激动剂向肺部的输送，用于癌症治疗。为了优化CpG ODN的治疗效果，需要详细了解它们(直接和间接)激活的细胞、它们的作用持续时间以及参与调节这些反应的调节通路。为了阐明这些问题，我们正在使用微阵列技术来识别CpG ODN诱导的免疫刺激的核心基因和网络。这些实验是在体外对高纯度的细胞亚群(包括人PDC和MDSC)进行的，并在小鼠体内进行研究，以监测生理条件下的基因表达。早期的结果表明，CpG ODN持续激活一组基因，这些基因在很大程度上依赖于自分泌I型干扰素(IFN)信号。目前的研究正在评估TLR激动剂与化疗药物(环磷酰胺)或PD1抑制剂在小鼠原位乳腺癌模型中对原发肿瘤和肺转移的进展的联合治疗。我们进一步证明，这一调控途径是通过IRF5通过核因子-kB发挥作用而介导的。事实上，邻近连接分析表明，IRF5与NF-kB共定位来完成这一任务。这种刺激活性被IRF8逆转，从而IRF8抑制CpG诱导的干扰素信号。人mMDSC成熟为M1VS M2样巨噬细胞的细胞因子也已被鉴定。高纯度的人mMDSC通过TNFa+IL-6或IL-10的组合被驱动分化为M1。相反，M-CSF驱动他们沿着M2途径成熟。功能研究支持这些独特细胞类型的表型鉴定。正在进行的研究旨在确定TLR激动剂可以诱导人单核细胞分化为M1或M2样巨噬细胞。结果表明，尽管大多数TLR激动剂支持促炎的M1样巨噬细胞的产生，但PAM3(TLR2/1激动剂)唯一地诱导M2样巨噬细胞的产生。对PAM3诱导的细胞因子/因子进行了鉴定，可以替代PAM3作为兴奋剂。此外，通过基因芯片、RNA测序和IPA分析相结合的方法，确定了负责驱动单核细胞分化为M1或M2样巨噬细胞的调控途径。

The unmethylated CpG motifs present in bacterial DNA interact with toll-like receptor 9 to trigger a pro-inflammatory immune response. CpG DNA also improves antigen presenting cell function, thereby facilitating the development of adaptive immunity. My laboratory established that synthetic oligonucleotides expressing immunostimulatory CpG motifs (CpG ODN) could be conjugated to apoptotic tumor cells to generate tumor vaccines that were rapidly internalized by professional APCs, promot DC maturation, and boost the induction of tumor-specific immunity. In multiple murine models we found that vaccination with CpG-conjugated apoptotic cell vaccines significantly reduced susceptibility to tumor challenge. This effect was observed both in mice vaccinated and then challenged and in animals immunized after tumor challenge. Unfortunately, the ability of these vaccines to eradicate tumors waned as cancer burden increased. We found this reflected the ability of large established tumors to generate an immunosuppressive microenvironment capable of inhibiting Ag-specific cellular responses that interferes with CpG-mediated immunotherapy. Myeloid-derived suppressor cells (MDSC) represent an important constituent of this immunosuppressive milieu. Large numbers of MDSC are present in and near established tumors and have been shown to inhibit the activity of antigen-specific T and NK cells. Our subsequent studies demonstrated that when CpG ODN were injected into the tumor itself, the immunosuppressive activity of monocytic MDSC (mMDSC) was significantly reduced. We hypothesize that the signal provided via TLR9 ligation was sufficient to overcome the inhibitory signals provided by the cells infiltrating the tumor. In mice, mMDSC express TLR9 and respond to CpG stimulation by i) losing their ability to suppress T cell function, ii) producing Th1 cytokines and iii) differentiating into M-1 like macrophages with tumoricidal capability. Similar activity was observed with TLR7 agonists, and the combination of TLR7 plus TLR9 agonists was significantly more effective. Indeed, intra-tumoral injection of this agonist combination induced the regression of even large established tumors (500 mm3 at the time of initial treatment). We extended these studies to evaluate the effect of TLR agonists on mMDSC purified from normal human donors and cancer patients. Results show that stimulation with TLR7/8 agonists induce human mMDSC to mature and lose their immuno-suppressive activity, reproducing on human MDSC the activity that TLR9 ligation has on mouse MDSC. We conclude that a combination of TLR ligands may be harnessed to achieve two independent but mutually supportive functions: boosting the efficacy of anti-tumor vaccines and reducing the activity of cells at the tumor site that would otherwise reduce the efficacy of this anti-tumor response. Our subsequent research sought to identify the optimal means and therapeutic window for the delivery of TLR ligands to tumor nodules. We've found that agonists adsorbed onto microparticles can be administered to mice with lung cancer, and that local DC and macrophages transport these particles to the tumor bed where they can effectively prevent tumor growth. We are exploring other formulations designed to optimize the delivery of TLR agonists to the lungs for use in cancer therapy. Efforts to optimize the therapeutic utility of CpG ODN require a detailed understanding of the cells they activate (both directly and indirectly), their duration of action, and the regulatory pathways involved in mediating these responses. To clarify these issues, we are using microarray technology to identify the genes and networks central to the immune stimulation elicited by CpG ODN. Such experiments are conducted in vitro on highly purified cell subpopulations (including human pDC and MDSC) and in vivo studies of mice to monitor gene expression under physiologic conditions. Earlier results showed that CpG ODN consistently activated a set of genes that was largely dependent on autocrine type I interferon (IFN) signaling. Current studies are evaluating combination therapy with TLR agonists plus chemotherapeutic agents (Cytoxan) or PD1 inhibitors on the progression of primary tumors and lung metastases in a murine orthotopic breast cancer model. We further demonstrated that this regulatory pathway was mediated via IRF5 acting through Nf-kB. Indeed, proximity ligation assays showed that IRF5 col-localized with Nf-kB to accomplish this task. This stimulatory activity is reversed by IRF8, such that IRF8 inhibits CpG induced IFN signaling. The cytokines responsible for the maturation of human mMDSC into M1 vs M2 like macrophages have also been identified. Highly purified human mMDSC are driven to differentiate into M1 by a combination of TNFa plus either IL-6 or IL-10. In contrast, M-CSF drives them to mature down the M2 pathway. Functional studies support the phenotypic identification of these unique cell types. Ongoing studies are directed towards identifying TLR agonists that induce human monocytes to differentiate into either M1- or M2-like macrophages. Results indicated that although most TLR agonists support the generation of pro-inflammatory M1-like macs, PAM3 (a TLR2/1 agonist) uniquely induces the generation of M2-like macrophage. The cytokines/factors elicited by treatment with PAM3 were identified and could substitute as stimulants. In addition, the regulatory pathways responsible for driving monocytes to differentiate into either M1- or M2-like macrophage were identified by a combination of microarray, RNA sequencing, and IPA analysis.