Targeting Tumors with IL-10 Prevents Dendritic Cell-Mediated CD8+ T Cell Apoptosis

Article

Targeting Tumors with IL-10 Prevents Dendritic CellMediated CD8+ T Cell Apoptosis Graphical Abstract

Authors Jian Qiao, Zhida Liu, Chunbo Dong, ..., Chuanhui Han, Ting Xu, Yang-Xin Fu

Correspondence [email protected] (J.Q.), [email protected] (T.X.), [email protected] (Y.-X.F.)

In Brief Qiao et al. generate a Cetuximab-based IL-10 fusion protein for EGFR-targeted delivery of IL-10 to tumors, in which IL-10 regulates IFN-g production by dendritic cells and enhances CD8+ T celldependent antitumor responses. The fusion protein shows high efficacy alone and in combination with anti-PD-L1/ CTLA-4.

Highlights d

CmAb-(IL10)2 prolongs the half-life of IL-10 and minimizes its toxicity

d

Tumor-targeted IL-10 induces superior antitumor effects over nontargeted IL-10

d

d

CmAb-(IL10)2 prevents tumor-specific CD8+ TILs apoptosis through IL-10R on DCs CmAb-(IL10)2 overcomes resistance to immune checkpoint blockade

Qiao et al., 2019, Cancer Cell 35, 901–915 June 10, 2019 ª 2019 Elsevier Inc. https://doi.org/10.1016/j.ccell.2019.05.005

Cancer Cell

Article Targeting Tumors with IL-10 Prevents Dendritic Cell-Mediated CD8+ T Cell Apoptosis Jian Qiao,1,4,* Zhida Liu,1,4 Chunbo Dong,1,4 Yan Luan,3 Anli Zhang,1 Casey Moore,1,2 Kai Fu,3 Jianjian Peng,3 Yang Wang,1 Zhenhua Ren,1 Chuanhui Han,1 Ting Xu,3,* and Yang-Xin Fu1,2,5,* 1The

Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA Department of Immunology, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA 3Dingfu Biotarget Co. Ltd., Suzhou, Jiangsu 215125, China 4These authors contributed equally 5Lead Contact *Correspondence: [email protected] (J.Q.), [email protected] (T.X.), [email protected] (Y.-X.F.) https://doi.org/10.1016/j.ccell.2019.05.005 2The

SUMMARY

Increasing evidence demonstrates that interleukin-10 (IL-10), known as an immunosuppressive cytokine, induces antitumor effects depending on CD8+ T cells. However, it remains elusive how immunosuppressive effects of IL-10 contribute to CD8+ T cell-mediated antitumor immunity. We generated Cetuximab-based IL-10 fusion protein (CmAb-(IL10)2) to prolong its half-life and allow tumor-targeted delivery of IL-10. Our results demonstrated potent antitumor effects of CmAb-(IL10)2 with reduced toxicity. Moreover, we revealed a mechanism of CmAb-(IL10)2 preventing dendritic cell (DC)-mediated CD8+ tumor-infiltrating lymphocyte apoptosis through regulating IFN-g production. When combined with immune checkpoint blockade, CmAb-(IL10)2 significantly improves antitumor effects in mice with advanced tumors. Our findings reveal a DC-regulating role of IL-10 to potentiate CD8+ T cell-mediated antitumor immunity and provide a potential strategy to improve cancer immunotherapy.

INTRODUCTION Most immunotherapies focus on activating immune stimulatory pathways or blocking suppressive ones. However, this strategy risks inducing severe toxicity due to off-tumor-targeted therapeutic delivery and unregulatable overactive immune responses (Conlon et al., 2015; Floros and Tarhini, 2015; Gallagher et al., 2007; Horvat et al., 2015; Nishino et al., 2015; Sanmamed and Chen, 2018). Immunostimulatory cytokines, such as interleukin-2 (IL-2), are potent for activation of immune cells and have long been approved by the US Food and Drug Administration (FDA) but their modest antitumor effects, short half-life and severe toxicity limit their wide application for treatment of cancer patients (Floros and Tarhini, 2015). IL-10 has been classically known as an immune suppressive cytokine. The lack of IL-10 is associated with several types of

autoimmune and inflammatory diseases (Asadullah et al., 2003). In an experimental model, mice with defective IL-10 or IL-10 receptor (IL-10R) spontaneously develop inflammatory bowel disease (IBD), and administration of recombinant IL-10 (rIL-10) has demonstrated therapeutic efficacy in various experimental IBD models (Duchmann et al., 1996; Tomoyose et al., 1998). However, paradoxically, increasing evidence demonstrates that IL-10 can induce antitumor effects in an immunedependent manner (Berman et al., 1996; Emmerich et al., 2012; Fujii et al., 2001; Giovarelli et al., 1995; Mumm et al., 2011; Richter et al., 1993; Wang et al., 2016). Particularly, recent preclinical studies report that systemic delivery of a pegylated form of IL-10 (PEG-IL10) can inhibit tumor growth by enhancing intratumoral CD8+ T cell proliferation and function (Emmerich et al., 2012; Mumm et al., 2011). A phase I clinical trial with PEG-IL10 has also shown encouraging antitumor activity in the

Significance The severe adverse effects of nontumor-targeted delivery and short in vivo half-life have greatly limited the use of immunostimulatory cytokines as cancer therapeutics. The short half-life of IL-10 significantly limits its clinical application. Here, we generated an antibody-based CmAb-(IL10)2 that not only prolongs the half-life but also demonstrates tumor-targeted delivery of IL-10 to induce potent antitumor effects. In contrast to immunostimulatory cytokines, CmAb-(IL10)2 has significantly reduced toxicity. Moreover, we revealed a mechanism for CmAb-(IL10)2 of preventing DC-mediated CD8+ TIL apoptosis. In the context of immune checkpoint blockade, CmAb-(IL10)2 further prevents tumor-specific CD8+ T cell apoptosis and achieves significantly improved antitumor effects. Taken together, we have developed the next generation of IL-10-based therapeutic to improve immunotherapies with reduced toxicity. Cancer Cell 35, 901–915, June 10, 2019 ª 2019 Elsevier Inc. 901

treatment of patients with advanced solid tumors (Naing et al., 2016). However, systemic administration of PEG-IL10 to treat murine tumors still showed increased immune cells infiltration and pathological immune responses in several normal organs, possibly due to off-tumor-targeted delivery of PEG-10 (Mumm et al., 2011). Moreover, patients receiving the highest dose of PEG-IL10 showed treatment-related adverse events (Naing et al., 2016), in line with other clinical reports that systemic administration of rIL-10 induces undesired effects of stimulating B cells and promoting autoimmune diseases (Lauw et al., 2000; Tilg et al., 2002). Mechanistically, current studies have focused on promoting CD8+ T cell responses (Fujii et al., 2001; Mumm et al., 2011; Wang et al., 2016), and have shown that IL-10R on CD8+ T cells is required for activation and proliferation in a mouse tumor model (Emmerich et al., 2012). However, many immune cell populations express the IL-10R, and the immune suppressive effects of endogenous IL-10 have also been reported in studies of tumor immunity (Ruffell et al., 2014; Sun et al., 2015; Wilke et al., 2011). Therefore, these studies highlight that the therapeutic mechanisms of IL-10 in tumor immunity remain elusive. In this study, we sought to test whether an antioncogenic receptor antibody-based fusion protein can overcome the short half-life of rIL-10 while minimizing off-tumor toxicity, and to elucidate the mechanisms of how the fusion protein improves CD8+ T cell-mediated antitumor effects. RESULTS Generation, Characterization, and Toxicity Assessment of an Anti-EGFR-IL-10 Fusion Protein To overcome the short half-life and allow targeted delivery of IL-10 into tumor microenvironment (TME), we generated a bispecific fusion protein targeting both an oncogenic receptor and the IL-10R. For proof-of-concept studies, we chose the FDAapproved anti-epidermal growth factor receptor (anti-EGFR) antibody Cetuximab (Erbitux) for targeted delivery of IL-10 to EGFR+ tumors. Based on the heterodimeric Fc variant KiHss-AkKh platform (Wei et al., 2017a), we generated the bispecific fusion proteins in which one arm was derived from Cetuximab and the other arm was either an IL-10 dimer (Cetuximab-based IL-10 fusion protein [CmAb-(IL10)2]) (Figure 1A) or monomer (CmAb-IL10) (Figure S1A). We tested the bispecific proteins by gel electrophoresis under reducing and nonreducing conditions (Figures 1B and S1B), and further confirmed the purity by high-performance liquid chromatography analysis (Figure 1C). Under physiological conditions, IL-10 acts as a dimer to bind the IL-10R and mediate IL-10 signal transduction (Tan et al., 1993). We thus first compared the activity of CmAb-(IL10)2 versus CmAb-IL10 and observed that CmAb(IL10)2 had higher activity than CmAb-IL10 by measuring tumor necrosis factor alpha (TNF-a) inhibition (Figure S1C). We, therefore, chose CmAb-(IL10)2 for the rest experiments in the current study. We first evaluated the binding of CmAb-(IL10)2 to tumor cells expressing EGFR. Flow cytometry analysis showed that CmAb-(IL10)2 bound to cell surface EGFR with an affinity similar to that of Cetuximab (Figure 1D). To determine if IL-10 was biologically active in this form, the anti-inflammatory effect of 902 Cancer Cell 35, 901–915, June 10, 2019

CmAb-(IL10)2 and rIL-10 was assessed by measuring TNF-a production from LPS-stimulated human peripheral blood mononuclear cells (Figure 1E). The results showed that CmAb-(IL10)2 effectively inhibited TNF-a production, although this activity (half maximal inhibitory concentration [IC50] = 4.77 pM) is slightly weaker than that of rIL-10 (IC50 = 1.968 pM). We next evaluated the serum levels of CmAb-(IL10)2 over time. C57BL/6J mice were injected intravenously (i.v.) with CmAb-(IL10)2, and blood samples were collected over time to quantify the concentration of IL-10. We observed that the halflife of CmAb-(IL10)2 is about 40 h (Table S1). Notably, at least 10% of the initial dose of CmAb-(IL10)2 was detected and remained in serum up to 4 days after administration (Figure 1F). By comparison, rIL-10 has a very short plasma half-life (2.7–4.5 h), which limits its clinical use (Huhn et al., 1997). In mice, rIL-10 was undetectable in serum 24 h after i.v. administration, and even PEG-IL10, which was specially designed to prolong its circulation time, was detected at less than 10% of the initial dose (Mattos et al., 2012). Together, these results suggest that CmAb-(IL10)2 significantly prolongs the half-life of IL-10, which may enhance the therapeutic potency of IL-10. Next, we tested whether systemic delivery of CmAb-(IL10)2 induces toxicity since a prolonged half-life could cause severe side effects. Given that IL-2 has been well documented in terms both its therapeutic effects and toxicity in clinical studies (Atkins et al., 1999; Siegel and Puri, 1991), and that, like human IL-10, human IL-2 can bind to murine IL-2 receptor and transduce IL-2 signal, we, therefore, chose IL-2 as a model to compare the toxicity with IL-10 in immune competent mice. We thus generated an anti-EGFR-IL-2 fusion protein (CmAb-IL2) in a similar way to generation of CmAb-IL10. We observed that mice treated with CmAb-IL2 at either high (2 mg/kg) or low (0.8 mg/kg) doses significantly induced inflammatory cytokine secretion accompanied by body weight loss (Figures 1G and 1H), and eventually all mice treated at the high dose died. By contrast, these effects were not observed in mice treated with CmAb-(IL10)2, even at a dosage (4 mg/kg) much higher than CmAb-IL2 (Figures 1G and 1H). These results suggest that CmAb-(IL10)2 could be a good candidate for the systemic treatment of cancer with a prolonged half-life and reduced toxicity. CmAb-(IL10)2 Has Superior Antitumor Effects than a Nontargeting IL-10 Fusion Protein To test our approach in immunocompetent syngeneic murine tumor models while minimizing the immunogenicity of human EGFR, we generated murine tumor cells expressing a chimeric EGFR (cEGFR). The cEGFR is full-length mouse EGFR with only six mutated amino acids that are critical for Cetuximab binding (Li et al., 2005). We showed that cEGFR could specifically and efficiently bind to Cetuximab (Figures 2A and S2A), as well as to CmAb-(IL10)2, in several tumor cell lines (Figures 2A and S2B). Having established the feasibility of mouse tumor studies with an anti-human EGFR antibody, we first tested whether systemic delivery of CmAb-(IL10)2 can induce antitumor effects by intraperitoneal injection of CmAb-(IL10)2 into C57BL/6J mice bearing B16-cEGFR tumors. The optimal dosage of 4 mg/kg was selected by assessing the dose-dependent antitumor activity in vivo (Figure S2C). A control IL-10 fusion protein (TmAb-(IL10)2) consisting

Figure 1. Generation and Characterization of the Anti-EGFR-IL-10 Bispecific Fusion Protein (A) Schematic structure of CmAb-(IL10)2. The Fab fragment of Cetuximab or the IL-10 dimer was fused to Fc region, respectively. These two fusion proteins form a bispecific CmAb-(IL10)2 protein. (B) SDS-PAGE analysis of CmAb-(IL10)2. (C) Size-exclusion chromatography-high-performance liquid chromatography analysis of the purity of CmAb-(IL10)2. (D) Binding affinity of CmAb-(IL10)2 to human EGFR overexpressed in B16 cells (B16-hEGFR). MFI, mean fluorescence intensity. (legend continued on next page)

Cancer Cell 35, 901–915, June 10, 2019 903

Figure 2. Targeting Tumors with IL-10 Achieves Effective Antitumor Effects (A) Assessment of cEGFR binding to Cetuximab, CmAb-(IL10)2, or TmAb-(IL10)2 (a control IL-10 fusion protein consisting of a nontargeting antibody instead of anti-EGFR), detected in B16 and B16-cEGFR cells by flow cytometry. (B) Tumor growth in C57BL/6J mice (n = 6–7) bearing B16-cEGFR tumors treated with Cetuximab, CmAb-(IL10)2, TmAb-(IL10)2, or control IgG (i.p., indicated by arrows). (C–E) Assessment of tumor targeting by CmAb-(IL10)2 in TC-1 (left flank) and TC-1-cEGFR (right flank) tumor-bearing NSG mice (C) (n = 7) injected intravenously by IRDye 800CW-labeled CmAb-(IL10)2 and fluorescence images (D) and quantification (E) of tumors collected at 96 h postinjection. (F and G) Tumor growth in C57BL/6J mice (n = 5–7) bearing TC-1-cEGFR (F) or MC38-cEGFR (G) tumors treated with CmAb-(IL10)2 or control IgG (i.p., indicated by arrows). (B and E–G) Data are shown as means ± SEM. ****p < 0.0001. See also Figure S2.

of a nontargeting antibody instead of anti-EGFR was generated in the same form as CmAb-(IL10)2. Following systemic treatment, CmAb-(IL10)2 led to complete tumor regression, whereas the nontargeting IL-10 fusion protein or Cetuximab had no

antitumor effect (Figure 2B). To confirm the targeting capability of CmAb-(IL10)2 in vivo, TC-1 and TC-1-cEGFR tumor cells were inoculated on the left and right flank of NSG mice, respectively. Tumor-bearing mice were given fluorescently labeled

(E) Human peripheral blood mononuclear cells (2 3 105/well) were incubated with LPS (2 mg/mL) in the presence or absence of CmAb-(IL10)2 or rIL-10 at indicated concentrations. Supernatants were collected and TNF-a was detected by ELISA. (F) Serum concentration of CmAb-(IL10)2 at the indicated time points after intravenous injection of CmAb-(IL10)2 at 1 mg/kg to C57BL/6J mice (n = 4). (G and H) Serum concentration of inflammatory cytokines (G) and body weight (H) of C57BL/6J mice (n = 5) that received intraperitoneal (i.p.) injections of CmAb-(IL10)2 or CmAb-IL-2 at the indicated doses on days 0, 2, and 4. For (G), blood was collected 20 h after the second treatment. (D–H) Data are shown as means ± SEM. ****p < 0.0001. See also Figure S1 and Table S1.

904 Cancer Cell 35, 901–915, June 10, 2019

CmAb-(IL10)2 via i.v. injection and imaged over time. Twenty-four hours later, CmAb-(IL10)2 could be observed in both TC-1-cEGFR and TC-1 tumors, whereas 48 h later only TC-1-cEGFR but not TC-1 tumors contained a significant amount of CmAb-(IL10)2 (Figures 2C–2E). These results suggest that systemic delivery of CmAb-(IL10)2 can target and retain IL-10 within tumor tissues, suggesting prolonged effects on tumor tissues. Similar to B16-cEGFR, C57BL/6J mice bearing TC-1-cEGFR or MC38-cEGFR tumors showed significant inhibition of tumor growth compared with control immunoglobulin G (IgG) (Figures 2F and 2G). Collectively, these results suggest that systemic administration of CmAb-(IL10)2 can induce potent antitumor effects through tumor-targeted delivery and prolonged retention of IL-10 within tumors. CmAb-(IL10)2 Requires Intratumoral CD8+ T Cells to Induce Antitumor Effects To test whether the antitumor effects of CmAb-(IL10)2 depend on intratumoral T cells, the mice were treated with FTY720 over the course of the experiment to block lymphocyte trafficking from lymphoid tissues (Tang et al., 2018; Thompson et al., 2010). The results showed that FTY720 treatment had no effect on efficacy (Figure S3A), which is consistent with a previous report in which response to PEG-IL10 was dependent on intratumoral CD8+ T cells (Emmerich et al., 2012). We thus focused on the effect of CmAb-(IL10)2 within the TME by intratumoral (i.t.) treatment with CmAb-(IL10)2 to rule out its effects outside TME. We evaluated the dose-dependent antitumor activity by i.t. injections of CmAb-(IL10)2 to select an optimal effective dosage of 0.4 mg/kg (Figure S3B) for further mechanism studies. To confirm that CmAb-(IL10)2 induces antitumor effects in an adaptive immune-dependent manner, we tested the antitumor efficacy in both immunocompetent syngeneic and immunocompromised mice. Consistent with systemic CmAb-(IL10)2 treatment, we observed significant inhibition of tumor growth after i.t. injections of CmAb-(IL10)2 into syngeneic C57BL/6J mice bearing B16-cEGFR tumors, and mice treated by anti-EGFR did not show antitumor efficacy compared with control IgG (Figure 3A), whereas in immunocompromised Rag1 / mice, neither CmAb-(IL10)2 nor anti-EGFR treatment inhibited tumor growth (Figure 3B). To better track tumor-specific T cell responses, we treated mice harboring established ovalbumin (OVA)-expressing B16-cEGFR tumors with CmAb-(IL10)2 or control antibody intratumorally. To this end, we used SIINFEKL-H2-Kb tetramer to detect OVA-specific CD8+ T cells in the tumor tissues. Mice treated with CmAb-(IL10)2 had significantly increased levels of OVA-specific CD8+ T cells compared with control antibodytreated mice (Figure 3C). Also supporting these results, we observed an increase in CD8+ T cells within tumors, but not in draining lymph nodes and spleens (Figure S3C). We further confirmed the tumor-specific CD8+ T cell responses with ELISPOT assay by re-stimulating splenocytes with the OVAderived SIINFEKL (OT1) peptide in vitro (Figure 3D). The nonspecific SIYRYYGL (SIY) used as a control peptide did not induce the responses. To determine whether CD8+ T cells are essential, we depleted CD8+ or CD4+ T cells to assess the antitumor effects of CmAb-(IL10)2. The results demonstrated that, in the absence of CD8+ T cells, the antitumor effect of CmAb-(IL10)2 was abrogated entirely, whereas this was not seen after depletion of CD4+

T cells (Figure 3E). In addition, we tested another tumor model and observed antitumor effects of i.t. injections of CmAb(IL10)2 into mice bearing MC38-cEGFR tumors (Figure S3D). Together, our results suggest that CmAb-(IL10)2 effectively acts within the TME and mediates CD8+ T cell-dependent antitumor effects. Furthermore, we investigated the antitumor efficacy of CmAb(IL10)2 in mice that had been reconstituted by human immune cell populations (humanized NSG-SGM3). We inoculated EGFR+ human epidermoid carcinoma A431cells into these mice, followed by i.t. injections of CmAb-(IL10)2. The results demonstrated that the antitumor effects were induced only in humanized mice, but not in their parental immune-deficient NSG-SGM3 mice (Figures 3F and 3G), indicating CmAb(IL10)2-mediated human adaptive immunity to induce antitumor effects. CmAb-(IL10)2 Suppresses DC-Mediated Tumor-Specific CD8+ T Cell Apoptosis Antigen-specific T cell activation often depends on antigen presentation by dendritic cells (DCs). Given that both DC and activated T cells express the IL-10R, we focused on investigating the interaction between DC and T cells. Bone marrow dendritic cells (BMDCs) and CD8+ OT1 T cells were obtained from bone marrow of C57BL/6J or C57BL/6-Tg(TcraTcrb)1100Mjb/J mice (OT1 transgenic mice) and were co-cultured with OVA antigen. The results showed that the proliferating CD8+ T cells were significantly increased with the addition of CmAb-(IL10)2 over time, indicating that the clonal expansion and survival of antigen-specific CD8+ OT1 cells were enhanced upon activation by and interaction with DCs (Figures 4A and 4B), and it was CmAb-(IL10)2 dose dependent (Figure S4). It is well known that activation of CD8+ T cells can be induced efficiently by DCs, but little is known about the role of DCs in the regulation of T cell death. Moreover, recent studies reported that the apoptosis of CD8+ tumor-infiltrating lymphocytes (TILs) might be a key limiting factor for T cell-mediated antitumor immune responses (Horton and Gajewski, 2018; Horton et al., 2018; Zhu et al., 2017); we, therefore, assessed the apoptosis of proliferating CD8+ T cells by detecting activated caspase-3+ cell population. Under the co-culture condition of DC and CD8+ OT1 in vitro, CmAb-(IL10)2 decreased activated caspase-3+ cells (Figure 4C). Most importantly, in vivo tumorspecific CD8+ T cells within tumors were protected from apoptosis when mice were treated by CmAb-(IL10)2 (Figure 4D). To further determine whether DCs were essential for the antitumor effect of CmAb-(IL10)2, we established tumors in mice with B6(Cg)-Zbtb46tm1(HBEGF)Mnz/J (zDC-DTR) bone marrow chimeras and selectively depleted conventional DCs (cDCs) with diphtheria toxin treatment. The results demonstrated that in the absence of DC cells, the antitumor effects of CmAb-(IL10)2 were completely abrogated (Figure 4E). It has been suggested that a subpopulation of cDCs, CD103+ DCs, plays a critical role in cross-presenting tumor antigens to CD8+ T cells intratumorally (Broz et al., 2014), we therefore further used Batf3 / mice lacking CD103+ DCs to test the efficacy and confirmed that DCs are essential for CmAb-(IL10)2-mediated antitumor effects (Figures 4F and 4G). Knowing that IL-10R-expressing macrophages also can present antigen to CD8+ T cells and may Cancer Cell 35, 901–915, June 10, 2019 905

Figure 3. CmAb-(IL10)2-Mediated Antitumor Effects Depend on Host Immunity (A and B) Tumor growth in C57BL/6J (A) or Rag1 / (B) mice (n = 5) bearing B16-cEGFR tumors treated by intratumoral (i.t.) injection of Cetuximab, CmAb-(IL10)2, or control IgG (indicated by arrows). (C) Quantification of OVA tetramer-positive (OVA-specific) CD8+ T cells in tumor tissues collected from B16-cEGFR-OVA tumor-bearing C57BL/6J mice (n = 4–5) treated twice by i.t. injection with control IgG or CmAb-(IL10)2 on days 11 and 14 after tumor cell inoculation. Tumor tissues were collected 7 days after first treatment and analyzed by flow cytometry. (D) IFN-g ELISPOT assay of splenocytes collected from B16-cEGFR-OVA tumor-bearing C57BL/6J mice (n = 5–6) treated three times by i.t. injection of control IgG or CmAb-(IL10)2. The spleens were harvested 9 days after the first treatment. OT1 peptide, OVA-derived SIINFEKL peptide; SIY, a control peptide SIYRYYGL. (E) Tumor growth in C57BL/6J mice (n = 5) bearing B16-cEGFR tumors treated with control IgG or CmAb-(IL10)2 (i.t., indicated by arrows). a-CD8 or a-CD4 antibodies were administered for T cell depletion during the CmAb-(IL10)2 treatment. (F and G) Tumor growth in NSG-SGM3 (F) and NSG-SGM3 humanized (G) mice (n = 5) bearing A431 tumors treated with CmAb-(IL10)2 or Cetuximab on days 11, 14, 17, and 20 after tumor cell inoculation. (A–G) Data are shown as means ± SEM. **p < 0.01, ****p < 0.0001; ns, not significant. See also Figure S3.

affect CmAb-(IL10)2-mediated antitumor effects, we depleted macrophages and observed no significant impact on the therapeutic efficacy of CmAb-(IL10)2 (Figure S5A). Since Fc-mediated antibody-dependent cellular cytotoxicity may also contribute 906 Cancer Cell 35, 901–915, June 10, 2019

to the therapeutic efficacy, we generated an Fc mutant CmAb(IL10)2 to ablate its binding capacity to the receptor and observed no impact on antitumor efficacy (Figure S5B). Taken together, these results demonstrate that DCs are

Figure 4. DCs are Essential for the Antitumor Effects of CmAb-(IL10)2 by Preventing Apoptosis of Antigen-Specific CD8+ T Cells (A and B) Proliferation of CFSE-labeled CD8+ OT1 T cells co-cultured with BMDCs from C57BL/6J mice in the presence of OVA and treated with CmAb-(IL10)2, Cetuximab, or vehicle. The percentage (A) and the number (B) of proliferating CD8+T cells were assessed by flow cytometry at the indicated time points. (C) Apoptosis of proliferating CD8+ T cells at 72 h after co-culture as described in (A and B), assessed by flow cytometry. (legend continued on next page)

Cancer Cell 35, 901–915, June 10, 2019 907

required for CmAb-(IL10)2-mediated antitumor effects and that CmAb-(IL10)2 can prevent DC-mediated antigen-specific CD8+ T cell apoptosis, a possible critical barrier to effective immunotherapy (Horton and Gajewski, 2018). IL-10R Signaling on DCs Is Required to Protect Antitumor CD8+ T Cell Apoptosis We hypothesized that the direct role of CmAb-(IL10)2 on DCs might be its ability to suppress tumor-specific CD8+ T cell apoptosis. To this end, we used BMDCs derived from the IL10 receptor-2 (IL-10R2 or IL-10Rb)-deficient mice that are unresponsive to IL-10 due to lack of IL-10R signaling (Kotenko et al., 1997). Strikingly, we observed that a co-culture of IL-10R2-deficient DCs and antigen-specific CD8+ T cells in the presence of CmAb-(IL10)2 completely abrogates both the increase in T cell proliferation and the protection from apoptosis (Figures 5A–5C). Similar results were observed when testing Pmel-1 CD8+ T cells that recognized naturally expressed tumor-associated antigen glycoprotein 100 (Figure S6A). Our results suggest that DCs and CD8+ T cells play dominant roles for CmAb-(IL10)2-mediated antitumor effects. We thus next used Rag1 / or IL-10R2-deficient Rag1 / mice (called Il10r / Rag1 / ), and adoptively transferred tumor-specific CD8+ OT1 cell (mixed with wild-type [WT] CD8+ T cells to prevent homeostatic proliferation of tumor-reactive T cells) into these mice to determine whether IL-10R signaling on DCs prevents antigen-specific CD8+ T cell apoptosis in vivo (Figure 5D). Similar to what we observed in immunocompetent mice treated with CmAb-(IL10)2, following adoptive transfer of CD8+ T cells and CmAb-(IL10)2 treatment of Rag1 / mice, we saw decreased apoptosis and increased antigen-specific CD8+ T cells in tumors (Figures 5E and 5F), whereas neither of these effects was observed in Il10r / Rag1 / mice (Figures 5G and 5H). Interestingly, when WT DCs were cultured with Il10r / CD8+ OT1, CmAb-(IL10)2 treatment still prevented T cell apoptosis, similar to DCs and WT OT1 co-cultures treated with CmAb-(IL10)2 (Figures S6B and S6D). Taken together, these results suggest that direct activation of IL-10R signaling by CmAb-(IL10)2 on DCs is required to prevent tumor-specific CD8+ T cell apoptosis and to induce CD8+ T cell-mediated antitumor immunity. Next, we further investigated the molecular mechanisms linking DCs to CD8+ T cells, leading to protection of antigen-specific CD8+ T cell apoptosis. Activation-induced T cell apoptosis (AICD) is a key factor limiting T cell-mediated immune responses (Gattinoni et al., 2005; Scheffel et al., 2016; Snow et al., 2010). We thus tested whether CmAb-(IL10)2 prevents DC-mediated AICD. To determine this, we performed a re-stimulation experiment by co-culturing DCs with previously antigen-activated CD8+ OT1 cells. We observed a significant reduction of antigen-activated CD8+ T cell apoptosis when CmAb-(IL10)2 was added to the co-culture (Figure 6A). Given that interferon-g (IFN-g) plays a critical role in AICD and that IL-10 can increase

IFN-g production by activated CD8+ T cells (Mumm et al., 2011; Refaeli et al., 2002), we hypothesized that CmAb-(IL10)2 may act on DCs to regulate IFN-g production by DC-activated CD8+T cells which may consequently reduce IFN-g-mediated CD8+ T cell apoptosis. To address this hypothesis, we cocultured WT or Il10r / DCs and CD8+ OT1 cells and quantified the production of IFN-g in supernatants. The results showed that, indeed, CmAb-(IL10)2 did not reduce IFN-g production when co-culturing Il10r / DCs with WT CD8+ OT1, while a significant reduction of IFN-g was observed when co-culturing WT DCs with either WT or Il10r / CD8+ OT1 in the presence of CmAb-(IL10)2 (Figure 6B). Collectively, these results suggested that CmAb-(IL10)2 regulates IFN-g production through IL-10R signaling on DCs rather than on antigen-activated CD8+ T cells, which is in line with our finding that CmAb-(IL10)2 suppresses DC-mediated antigen-specific CD8+ T cell apoptosis through IL-10R signaling on DCs, but not on CD8+ T cells. To further provide evidence that CmAb-(IL10)2 regulates IFN-g production to prevent DC-mediated antigen-specific CD8+ T cell apoptosis, we added IFN-g to the co-culture of BMDCs and CD8+ OT1 cells in the presence of CmAb-(IL10)2. As expected, IFN-g dose-dependently induced apoptosis of antigen-activated CD8+ T cells, and an anti-IFN-g antibody rescued the viability of CD8+ T cells in co-cultures lacking CmAb-(IL10)2 (Figure 6C). On the other hand, since lacking DC IL-10R signaling led to unregulated IFN-g production (Figure 6B), we added anti-IFN-g antibody to the co-culture of Il10r / DCs with WT CD8+ OT1 cells. The results showed that neutralizing IFN-g efficiently reduced the apoptosis of CD8+ T cells co-cultured with Il10r / DCs in the absence or presence of CmAb-(IL10)2 (Figure 6D). Collectively, these results provide direct evidence that CmAb-(IL10)2 prevents IFN-g-mediated CD8+ T cell apoptosis in a DC IL-10R signaling-dependent manner. Given that DCs produce IL-12, an efficient inducer of IFN-g (Chan et al., 1992), we tested IL-12 production by DCs. We observed that CmAb-(IL10)2 reduced IL-12 production through IL-10R signaling on DCs (Figure S7A). We further confirmed that, in the presence of CmAb(IL10)2, the addition of IL-12 to the co-culture induced both IFN-g production and CD8+OT1 cell apoptosis, while an antiIL-12 antibody antagonized these effects (Figures S7B and S7C). Therefore, these results mirrored DC-mediated IL-10R signaling-dependent IFN-g production and apoptosis of antigen-activated CD8+ T cell. To address whether our in vitro observations hold in tumorbearing mice, we intratumorally injected antigen-activated OT1 cells into Rag1 / mice along with CmAb-(IL10)2 or anti-IFN-g antibody. We found a decrease of tumor-specific CD8+ T cell apoptosis in tumors treated by either CmAb-(IL10)2 or anti-IFN-g antibody, compared with the control group (Figure 6E). To corroborate the role of IFN-g, we experimented on IFN-g-deficient mice and observed no difference in tumor-specific CD8+

(D) Apoptosis of OVA tetramer-positive CD8+ T cells in B16-cEGFR-OVA tumor tissues from C57BL/6J mice (n = 4–6) treated twice by i.t. injection of control IgG or CmAb-(IL10)2. Tumor tissues were collected 7 days after first treatment and analyzed by flow cytometry. (E) Tumor growth in bone marrow-reconstituted zDC-DTR mice bearing B16-cEGFR tumors treated with control IgG or CmAb-(IL10)2 (i.t., indicated by arrows). Diphtheria toxin (DT) (4 mg/kg, i.p.) was injected every other day starting on day 10 to deplete dendritic cells. (F and G) Tumor growth in C57BL/6J (F) and Batf3 / (G) mice (n = 4–5) bearing B16-cEGFR tumors treated with control IgG or CmAb-(IL10)2 (i.t., indicated by arrows). (B–G) Data are shown as means ± SEM. *p < 0.05, ****p < 0.0001; ns, not significant. See also Figures S4 and S5.

908 Cancer Cell 35, 901–915, June 10, 2019

Figure 5. IL-10R Signaling on DCs Is Required for Preventing Apoptosis of Antigen-Specific CD8+ T Cells (A) Proliferation of CFSE-labeled CD8+ OT1 T cells co-cultured with BMDCs from WT (left) or Il10r / (right) mice in the presence of OVA treated with CmAb-(IL10)2 or vehicle, assessed by flow cytometry at the indicated time points. (B) Cell number of proliferating CD8+ OT1 T cells co-cultured with BMDCs from Il10r / mice in the presence of OVA and treated with CmAb-(IL10)2 or vehicle, assessed by flow cytometry at the indicated time points. (C) Apoptosis of proliferating CD8+ T cells at 72 h after co-culture as described in (B), assessed by flow cytometry. (D) Scheme of adoptive transfer of CD8+ T cells (2 3 104 OT1 CD8+ mixed with 2 3 106 WT CD8+ T cells) and CmAb-(IL10)2 treatment of Rag1 / or Il10r / Rag1 / mice (n = 5) bearing B16-cEGFR-OVA tumors. (E–H) Quantification by flow cytometry of apoptotic (E and G) and viable (F and H) tetramer-positive CD8+ T cells in tumor tissues from Rag1 / (E and F) or Il10r / Rag1-/-(G and H) mice treated as in (D). (B–H) Data are shown as means ± SEM. *p < 0.05, **p < 0.01; ns, not significant. See also Figure S6.

Cancer Cell 35, 901–915, June 10, 2019 909

Figure 6. CmAb-(IL10)2 Prevents Antigen-Specific CD8+ T Cell Apoptosis through Regulating DC-Mediated IFN-g Production (A) Apoptosis assessment of re-stimulated CD8+ T cells by co-culturing antigen-activated CD8+ OT1 T cells with BMDCs from C57BL/6J mice in the presence of OVA and CmAb-(IL10)2 or vehicle, determined at 48 h after re-stimulation by flow cytometry. (B) IFN-g production from the indicated co-cultures of DCs and CD8+ OT1 T cells in the presence of OVA and treated with CmAb-(IL10)2 or vehicle. (C and D) Apoptosis of proliferating CD8+ T cells co-cultured with BMDCs from WT (C) or Il10r / (D) mice in the presence of OVA and treated with CmAb-(IL10)2, IFN-g, or a-IFN-g (10 mg/mL), assessed by flow cytometry. (E) Apoptosis of CD8+ T cells in B16-cEGFR-OVA tumor tissues from Rag1 / mice (n = 5–6) i.t. treated with 1 3 106 antigen-activated CD8+ OT1 T cells plus control IgG, CmAb-(IL10)2 or anti-IFN-g (150 mg, i.p.) on day 11 after tumor cell inoculation. Tumor tissues were collected 2 days after treatment and analyzed by flow cytometry. (F) Apoptosis of OVA tetramer-positive CD8+ T cells in B16-cEGFR-OVA tumor tissues from Ifng / mice (n = 7) i.t. treated with control IgG or CmAb-(IL10)2 on days 8 and 11 after tumor cell inoculation. Tumor tissues were collected 7 days after first treatment and analyzed by flow cytometry. (A–F) Data are shown as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, not significant. See also Figure S7.

T cell apoptosis between CmAb-(IL10)2 treatment and control group (Figure 6F). Thus, our data confirmed that CmAb-(IL10)2 suppresses DC-mediated tumor-specific CD8+ T cells apoptosis in the TME through regulating IFN-g production. CmAb-(IL10)2 Overcomes Resistance to Immune Checkpoint Blockade Therapy Our findings support a model in which IFN-g triggers CD8+ T cell apoptosis in the TME, while pharmacologically providing CmAb(IL10)2 acts on DCs through IL-10R signaling to prevent IFN-gmediated antigen-specific CD8+ T cell apoptosis. We, therefore, hypothesized that immunotherapies may result in unregulated increase in production of IFN-g, inducing T cell apoptosis. Indeed, a recent study showed intratumoral IFN-g can induce tumor-specific CD8+ T cell apoptosis in the MDSC-enriched TME (Zhu et al., 2017). Given that the combination of anti910 Cancer Cell 35, 901–915, June 10, 2019

CTLA-4 and/or anti-PD-1 therapy significantly increases expression of IFN-g by tumor-infiltrating T cells in murine models (Peng et al., 2012; Shi et al., 2016), we thus attempted to improve their therapeutic efficacy by combining CmAb-(IL10)2 treatment with immune checkpoint blockade (ICB). Notably, despite decreased apoptosis seen after ICB, we observed that CmAb-(IL10)2 can further prevent tumor-specific CD8+ T cell apoptosis (Figure 7A). Moreover, when CmAb-(IL10)2 was administered either intratumorally or systemically in combination with ICB, we achieved significantly improved therapeutic efficacy (Figures 7B–7E). We then re-challenged mice who had been treated by the combination of CmAb-(IL10)2 and ICB and became tumor-free for more than 4 weeks. All these tumor-free mice rejected their tumor growth after re-challenge with a 10-fold higher dose of tumor cells (Figure 7D). This suggests that the combined therapy of CmAb(IL10)2 with PD-1 and CTLA-4 blockade not only improves the

Figure 7. CmAb-(IL10)2 Can Prevent Antigen-Specific CD8+ TIL Apoptosis and Improve the Antitumor Effects of Immune Checkpoint Blockade in the Treatment of Advanced Tumors (A) Apoptosis of OVA tetramer-positive CD8+ T cells in B16-cEGFR-OVA tumor tissues from C57BL/6J mice (n = 7) treated twice by a-PD-L1 and a-CTLA-4 (immune checkpoint blockade [ICB]) in combination with control IgG or CmAb-(IL10)2. Tumor tissues were collected 7 days after first treatment and analyzed by flow cytometry. (B and C) Scheme (B) (top), tumor growth (B) (bottom), and survival curve (C) of the advanced B16cEGFR tumor-bearing (80–120 mm3) C57BL/6J mice (n = 6–7) treated with CmAb-(IL10)2, ICB, or the combination therapy as indicated. (D) Tumor growth after challenge with B16-cEGFR cells in treatment-naı¨ve mice or mice cured by the combination therapy for ICB and CmAb-(IL10)2. (E) Scheme of treatment (left) and tumor growth (right) of C57BL/6J mice (n = 5–7) bearing advanced B16-cEGFR tumors treated with CmAb-(IL10)2, ICB, or the combination therapy as indicated. (A–E) Data are shown as means ± SEM. *p < 0.05, ***p < 0.001, ****p < 0.0001.

efficacy but also boosts immune memory against tumors. These results demonstrated that CmAb-(IL10)2 can prevent apoptosis of CD8+ TILs and overcome ICB resistance. Together with recent observations (Horton et al., 2018; Zhu et al., 2017), our results suggest that the limited efficacy of ICB may be attributed to premature apoptosis of tumor-specific CD8+ TILs after ICB therapy.

DISCUSSION Immunostimulatory cytokines have been approved for systemic use in cancer treatment, but their short half-life, severe toxicity, and limited efficacy prevent their wide application as cancer therapeutics (Floros and Tarhini, 2015; Sanmamed Cancer Cell 35, 901–915, June 10, 2019 911

and Chen, 2018). IL-10 has been shown to induce antitumor effects in a CD8+ T cell-dependent manner (Emmerich et al., 2012; Fujii et al., 2001; Mumm et al., 2011; Wang et al., 2016). And PEG-IL10, designed to prolong its half-life, has demonstrated promising antitumor efficacy (Mumm et al., 2011; Naing et al., 2016), leading to initiation of clinical trials of PEG-IL10 treatment combined with ICB in solid tumors. The results that were very recently reported suggest that the combination of PEG-IL10 and anti-PD-1 had 42% of the overall response rate among 19 patients (Naing et al., 2018). However, PEG-IL10 treatment-related adverse events were observed either in mice at a therapeutic dose or in patients treated with a high dose of PEG-IL10 (Mumm et al., 2011; Naing et al., 2016). Here we have demonstrated that CmAb-(IL10)2 significantly prolongs the half-life of IL-10, reduces toxicity, and induces potent antitumor effects. Moreover, we revealed a mechanism that CmAb-(IL10)2 suppresses DC-mediated antigen-specific CD8+ T cell apoptosis through regulating IFN-g production. When combined with ICB, we achieved significantly improved antitumor effects. In most studies of antitumor immunity, endogenous IL-10 is considered a main factor contributing to immune suppressive TME (Ruffell et al., 2014; Sun et al., 2015; Wilke et al., 2011). However, increasing evidence shows pharmacological administration of IL-10 has antitumor effects by enhancing CD8+ T cell-mediated antitumor immunity (Fujii et al., 2001) (Emmerich et al., 2012; Mumm et al., 2011). These opposing effects may be due to several factors including high-dose and administration schedule of exogenous IL-10, and abundance, activation status, and the IL-10R expression on CD8+ TILs. Nevertheless, since DCs also express the IL-10R, understanding the role of IL-10 on DCs and linking this role to CD8+ T cell-mediated antitumor immunity is crucial. Here, we demonstrated that CmAb-(IL10)2 suppresses DC-mediated antigen-specific CD8+ T cell apoptosis leading to higher levels of tumor-specific CD8+ T cells within tumors in a DC IL-10R signal-dependent manner. The molecular mechanism linking this role of DC to T cell enhancement is that CmAb-(IL10)2 regulates the IL-12/ IFN-g production axis to suppress IFN-g-mediated apoptosis of antigen-specific CD8+ T cells. Taken together, we propose that IL-10 treatment regulating the interaction of CD8+ T cellDC is critical in the TME. First, IL-10 treatment increases IFN-g production by activated CD8+ T cells. This increase in IFN-g stimulates DCs to prime and reactivate CD8+ T cells, further increasing their secretion of IFN-g. This overproduction of IFN-g induces apoptosis of the tumor-specific CD8+ T cells. Furthermore, our data suggest that balance is maintained in this circuit through IL-10R-mediated downregulation of IL-12 production by DCs. This consequently prevents IFN-g levels from reaching overproduction, thus preventing CD8+ T cell apoptosis. Therefore, our results support an immunotherapy model in which effective responses can be achieved by a balanced stimulation to avoid dysfunctional apoptosis of CD8+ TILs. Systemic delivery of cytokines that have potent activation effects on immune cells often induces off-tumor target effects. As an example, IL-2 toxicity is well documented by increased inflammatory cell infiltration and cytokine production in normal organs, and capillary leak syndrome (Rafi-Janajreh et al., 912 Cancer Cell 35, 901–915, June 10, 2019

1999; Siegel and Puri, 1991). In the case of IL-10 treatment, off-tumor target effects may also induce treatment-related adverse events, as observed when using PEG-IL10 in the treatment of preclinical murine tumor (Mumm et al., 2011), and as reported by clinical studies in which rIL-10 stimulates B cells and promotes autoimmune disease (Lauw et al., 2000; Tilg et al., 2002). To address whether CmAb-(IL10)2 induces toxicity, we compared CmAb-(IL10)2 versus CmAb-IL2. Our results showed that CmAb-(IL10)2 had significantly reduced toxicity in contrast to low-dose CmAb-IL2 (up to 5-fold lower than CmAb-(IL10)2), which induced severe toxicity indicated by significant increases in proinflammatory cytokines and body weight loss. CmAb-(IL10)2 demonstrates unique safety features because of its tumor-targeted delivery and retention of IL-10, and because of its regulatory role on proinflammatory cytokine production, which would provide a better safety profile than those potent immune-active cytokines. Recent studies have reported that apoptosis of CD8+ TILs may be a critical barrier to effective immunotherapy (Horton and Gajewski, 2018). In the treatment of advanced tumors, we observed that combination of CmAb-(IL10)2 with antiCTLA-4 and anti-PD-L1 blockade therapy greatly improved the therapeutic efficacy compared with either IL-10 or ICB therapy, and this improved efficacy was associated with further decrease of tumor-specific CD8+ TIL apoptosis. Despite an inhibition of apoptosis by blockade of the PD-L1/PD-1 pathway (Dong et al., 2002), our results suggest that, since checkpoint blockade therapy significantly increases expression of IFN-g by TILs (Peng et al., 2012; Shi et al., 2016), further prevention of T cell apoptosis by CmAb-(IL10)2 is likely due to regulating IFN-g production of CD8+ TILs. Meanwhile, the combination of anti-CTLA-4 and anti-PD-L1 antibodies may modulate the TME toward increasing CD8+ TILs and depleting immune suppressive regulatory T cells (Daud et al., 2016; Peng et al., 2012; Ribas et al., 2009; Simpson et al., 2013; Takahashi et al., 2000). Thus, this ICB plus CmAb-(IL10)2 therapy improves antitumor immunity and leads to the development of strong immune memory against tumors, as observed in our studies. Since anti-CTLA-4 and anti-PD-1 induce differential antitumor effects through specific mechanisms (Wei et al., 2017b), it will be necessary to dissect the role of IL-10 in combination with either anti-CTLA-4 or anti-PD-1 for better designing an optimal treatment regimen. A major issue with ICB therapy, particularly when combining anti-CTLA-4 and anti-PD-L1, is the frequency of immune-related adverse events (irAEs) (Brahmer et al., 2018; Wolchok et al., 2017). However, irAEs are rarely observed in mouse models (Liu et al., 2014; Olson et al., 2018). Although we did not observe toxicity in mice treated by either ICB or combined therapy with CmAb-(IL10)2, whether CmAb(IL10)2 would have limited toxicity in human remains to be determined. Taken together, our findings provide evidence that effective immunotherapy can be achieved by proper activation of T cells to avoid CD8+ TIL apoptosis. We demonstrated that CmAb(IL10)2 significantly improved ICB therapy, and it will be interesting to explore whether CmAb-(IL10)2 can improve therapeutic efficacy while minimizing toxicity when combined with other immunotherapies in the future.

STAR+METHODS

REFERENCES

Detailed methods are provided in the online version of this paper and include the following:

Asadullah, K., Sterry, W., and Volk, H.D. (2003). Interleukin-10 therapy–review of a new approach. Pharmacol. Rev. 55, 241–269.

d d d

d

d

KEY RESOURCES TABLE CONTACT FOR REAGENT AND RESOURCE SHARING EXPERIMENTAL MODEL AND SUBJECT DETAILS B Mice B Cell Lines B Humanized Mice and Human Peripheral Blood Mononuclear Cells METHOD DETAILS B Production of Bispecific Fusion Proteins B In Vivo Imaging B Tumor Growth and Treatment B IFN-g Enzyme-Linked Immunosorbent Spot Assay (ELISPOT) B Generation of Bone Marrow Chimeras B In Vitro Co-culture of Bone Marrow Dendritic Cells (BMDC) and T Cells + B Generation of Antigen Activated CD8 OT1 T Cells In Vitro B Flow Cytometry Analysis B Enzyme-Linked ImmunoSorbent Assay (ELISA) B RNA Extraction and Quantitative Real-Time PCR QUANTIFICATION AND STATISTICAL ANALYSIS

SUPPLEMENTAL INFORMATION Supplemental Information can be found online at https://doi.org/10.1016/j. ccell.2019.05.005. ACKNOWLEDGMENTS We thank the UT Southwestern Flow Cytometry Facility, Institutional Animal Care and Use Committee Animal Resources Center, and Animal Research Center. We thank Dr. Jinming Gao and Dr. Qiang Feng’s help of getting the in vivo imaging data. Y.-X.F. holds the Mary Nell and Ralph B. Rogers Professorship in Immunology. This work was in part supported by Cancer Prevention Research Institute of Texas (United States) grants RR150072 and RP180725 to Y.-X.F. AUTHOR CONTRIBUTIONS Z.L., C.D., and J.Q. performed the experiments and analyzed the data. J.Q., Z.L., C.D., and Y.-X.F. designed experiments and wrote the manuscript. Y.L., K.F., and J.P. characterized CmAb-(IL10)2 stability and purity under T.X.’s supervision. C.M. and C.H. participated in performing some experiments. C.M. contributed to manuscript preparation. A.Z., Y.W., and Z.R. provided some mice and reagents. J.Q. and Y.-X.F. supervised the study. DECLARATION OF INTERESTS Yan Luan, Kai Fu, Jianjian Peng, and Ting Xu are employees of Dingfu Biotarget and the antibody-IL-10 patent is in the process of application by Dingfu Biotarget. Received: November 11, 2018 Revised: March 7, 2019 Accepted: May 13, 2019 Published: June 10, 2019

Atkins, M.B., Lotze, M.T., Dutcher, J.P., Fisher, R.I., Weiss, G., Margolin, K., Abrams, J., Sznol, M., Parkinson, D., Hawkins, M., et al. (1999). High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J. Clin. Oncol. 17, 2105–2116. Berman, R.M., Suzuki, T., Tahara, H., Robbins, P.D., Narula, S.K., and Lotze, M.T. (1996). Systemic administration of cellular IL-10 induces an effective, specific, and long-lived immune response against established tumors in mice. J. Immunol. 157, 231–238. Brahmer, J.R., Lacchetti, C., Schneider, B.J., Atkins, M.B., Brassil, K.J., Caterino, J.M., Chau, I., Ernstoff, M.S., Gardner, J.M., Ginex, P., et al. (2018). Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: American Society of Clinical Oncology clinical practice guideline. J. Clin. Oncol. 36, 1714–1768. Broz, M.L., Binnewies, M., Boldajipour, B., Nelson, A.E., Pollack, J.L., Erle, D.J., Barczak, A., Rosenblum, M.D., Daud, A., Barber, D.L., et al. (2014). Dissecting the tumor myeloid compartment reveals rare activating antigenpresenting cells critical for T cell immunity. Cancer Cell 26, 638–652. Chan, S.H., Kobayashi, M., Santoli, D., Perussia, B., and Trinchieri, G. (1992). Mechanisms of IFN-gamma induction by natural killer cell stimulatory factor (NKSF/IL-12). Role of transcription and mRNA stability in the synergistic interaction between NKSF and IL-2. J. Immunol. 148, 92–98. Conlon, K.C., Lugli, E., Welles, H.C., Rosenberg, S.A., Fojo, A.T., Morris, J.C., Fleisher, T.A., Dubois, S.P., Perera, L.P., Stewart, D.M., et al. (2015). Redistribution, hyperproliferation, activation of natural killer cells and CD8 T cells, and cytokine production during first-in-human clinical trial of recombinant human interleukin-15 in patients with cancer. J. Clin. Oncol. 33, 74–82. Daud, A.I., Loo, K., Pauli, M.L., Sanchez-Rodriguez, R., Sandoval, P.M., Taravati, K., Tsai, K., Nosrati, A., Nardo, L., Alvarado, M.D., et al. (2016). Tumor immune profiling predicts response to anti-PD-1 therapy in human melanoma. J. Clin. Invest. 126, 3447–3452. Dong, H., Strome, S.E., Salomao, D.R., Tamura, H., Hirano, F., Flies, D.B., Roche, P.C., Lu, J., Zhu, G., Tamada, K., et al. (2002). Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat. Med. 8, 793–800. Duchmann, R., Schmitt, E., Knolle, P., Meyer Zum Buschenfelde, K.H., and Neurath, M. (1996). Tolerance towards resident intestinal flora in mice is abrogated in experimental colitis and restored by treatment with interleukin-10 or antibodies to interleukin-12. Eur. J. Immunol. 26, 934–938. Emmerich, J., Mumm, J.B., Chan, I.H., LaFace, D., Truong, H., McClanahan, T., Gorman, D.M., and Oft, M. (2012). IL-10 directly activates and expands tumor-resident CD8(+) T cells without de novo infiltration from secondary lymphoid organs. Cancer Res. 72, 3570–3581. Floros, T., and Tarhini, A.A. (2015). Anticancer cytokines: biology and clinical effects of interferon-alpha2, interleukin (IL)-2, IL-15, IL-21, and IL-12. Semin. Oncol. 42, 539–548. Fujii, S., Shimizu, K., Shimizu, T., and Lotze, M.T. (2001). Interleukin-10 promotes the maintenance of antitumor CD8(+) T-cell effector function in situ. Blood 98, 2143–2151. Gallagher, D.C., Bhatt, R.S., Parikh, S.M., Patel, P., Seery, V., McDermott, D.F., Atkins, M.B., and Sukhatme, V.P. (2007). Angiopoietin 2 is a potential mediator of high-dose interleukin 2-induced vascular leak. Clin. Cancer Res. 13, 2115–2120. Gattinoni, L., Klebanoff, C.A., Palmer, D.C., Wrzesinski, C., Kerstann, K., Yu, Z., Finkelstein, S.E., Theoret, M.R., Rosenberg, S.A., and Restifo, N.P. (2005). Acquisition of full effector function in vitro paradoxically impairs the in vivo antitumor efficacy of adoptively transferred CD8+ T cells. J. Clin. Invest. 115, 1616–1626. Giovarelli, M., Musiani, P., Modesti, A., Dellabona, P., Casorati, G., Allione, A., Consalvo, M., Cavallo, F., di Pierro, F., De Giovanni, C., et al. (1995). Local release of IL-10 by transfected mouse mammary adenocarcinoma cells

Cancer Cell 35, 901–915, June 10, 2019 913

does not suppress but enhances antitumor reaction and elicits a strong cytotoxic lymphocyte and antibody-dependent immune memory. J. Immunol. 155, 3112–3123.

Refaeli, Y., Van Parijs, L., Alexander, S.I., and Abbas, A.K. (2002). Interferon gamma is required for activation-induced death of T lymphocytes. J. Exp. Med. 196, 999–1005.

Horton, B.L., and Gajewski, T.F. (2018). Back from the dead: TIL apoptosis in cancer immune evasion. Br. J. Cancer 118, 309–311.

Ribas, A., Comin-Anduix, B., Economou, J.S., Donahue, T.R., de la Rocha, P., Morris, L.F., Jalil, J., Dissette, V.B., Shintaku, I.P., Glaspy, J.A., et al. (2009). Intratumoral immune cell infiltrates, FoxP3, and indoleamine 2,3-dioxygenase in patients with melanoma undergoing CTLA4 blockade. Clin. Cancer Res. 15, 390–399.

Horton, B.L., Williams, J.B., Cabanov, A., Spranger, S., and Gajewski, T.F. (2018). Intratumoral CD8(+) T-cell apoptosis is a major component of T-cell dysfunction and impedes antitumor immunity. Cancer Immunol. Res. 6, 14–24. Horvat, T.Z., Adel, N.G., Dang, T.O., Momtaz, P., Postow, M.A., Callahan, M.K., Carvajal, R.D., Dickson, M.A., D’Angelo, S.P., Woo, K.M., et al. (2015). Immune-related adverse events, need for systemic immunosuppression, and effects on survival and time to treatment failure in patients with melanoma treated with ipilimumab at Memorial Sloan Kettering Cancer Center. J. Clin. Oncol. 33, 3193–3198. Huhn, R.D., Radwanski, E., Gallo, J., Affrime, M.B., Sabo, R., Gonyo, G., Monge, A., and Cutler, D.L. (1997). Pharmacodynamics of subcutaneous recombinant human interleukin-10 in healthy volunteers. Clin. Pharmacol. Ther. 62, 171–180. Kotenko, S.V., Krause, C.D., Izotova, L.S., Pollack, B.P., Wu, W., and Pestka, S. (1997). Identification and functional characterization of a second chain of the interleukin-10 receptor complex. EMBO J. 16, 5894–5903. Lauw, F.N., Pajkrt, D., Hack, C.E., Kurimoto, M., van Deventer, S.J., and van der Poll, T. (2000). Proinflammatory effects of IL-10 during human endotoxemia. J. Immunol. 165, 2783–2789. Li, S., Schmitz, K.R., Jeffrey, P.D., Wiltzius, J.J., Kussie, P., and Ferguson, K.M. (2005). Structural basis for inhibition of the epidermal growth factor receptor by cetuximab. Cancer Cell 7, 301–311. Liu, J., Blake, S.J., Smyth, M.J., and Teng, M.W. (2014). Improved mouse models to assess tumour immunity and irAEs after combination cancer immunotherapies. Clin. Transl. Immunol. 3, e22.

Richter, G., Kruger-Krasagakes, S., Hein, G., Huls, C., Schmitt, E., Diamantstein, T., and Blankenstein, T. (1993). Interleukin 10 transfected into Chinese hamster ovary cells prevents tumor growth and macrophage infiltration. Cancer Res. 53, 4134–4137. Ruffell, B., Chang-Strachan, D., Chan, V., Rosenbusch, A., Ho, C.M., Pryer, N., Daniel, D., Hwang, E.S., Rugo, H.S., and Coussens, L.M. (2014). Macrophage IL-10 blocks CD8+ T cell-dependent responses to chemotherapy by suppressing IL-12 expression in intratumoral dendritic cells. Cancer Cell 26, 623–637. Sanmamed, M.F., and Chen, L. (2018). A paradigm shift in cancer immunotherapy: from enhancement to normalization. Cell 175, 313–326. Scheffel, M.J., Scurti, G., Simms, P., Garrett-Mayer, E., Mehrotra, S., Nishimura, M.I., and Voelkel-Johnson, C. (2016). Efficacy of adoptive T-cell therapy is improved by treatment with the antioxidant N-acetyl cysteine, which limits activation-induced T-cell death. Cancer Res. 76, 6006–6016. Shi, L.Z., Fu, T., Guan, B., Chen, J., Blando, J.M., Allison, J.P., Xiong, L., Subudhi, S.K., Gao, J., and Sharma, P. (2016). Interdependent IL-7 and IFNgamma signalling in T-cell controls tumour eradication by combined alphaCTLA-4+alpha-PD-1 therapy. Nat. Commun. 7, 12335. Siegel, J.P., and Puri, R.K. (1991). Interleukin-2 toxicity. J. Clin. Oncol. 9, 694–704.

Mattos, A., de Jager-Krikken, A., de Haan, M., Beljaars, L., and Poelstra, K. (2012). PEGylation of interleukin-10 improves the pharmacokinetic profile and enhances the antifibrotic effectivity in CCl(4)-induced fibrogenesis in mice. J. Control Release 162, 84–91.

Simpson, T.R., Li, F., Montalvo-Ortiz, W., Sepulveda, M.A., Bergerhoff, K., Arce, F., Roddie, C., Henry, J.Y., Yagita, H., Wolchok, J.D., et al. (2013). Fc-dependent depletion of tumor-infiltrating regulatory T cells co-defines the efficacy of anti-CTLA-4 therapy against melanoma. J. Exp. Med. 210, 1695–1710.

Mumm, J.B., Emmerich, J., Zhang, X., Chan, I., Wu, L., Mauze, S., Blaisdell, S., Basham, B., Dai, J., Grein, J., et al. (2011). IL-10 elicits IFNgamma-dependent tumor immune surveillance. Cancer Cell 20, 781–796.

Snow, A.L., Pandiyan, P., Zheng, L., Krummey, S.M., and Lenardo, M.J. (2010). The power and the promise of restimulation-induced cell death in human immune diseases. Immunol. Rev. 236, 68–82.

Naing, A., Infante, J.R., Papadopoulos, K.P., Chan, I.H., Shen, C., Ratti, N.P., Rojo, B., Autio, K.A., Wong, D.J., Patel, M.R., et al. (2018). PEGylated IL-10 (Pegilodecakin) induces systemic immune activation, CD8(+) T Cell invigoration and polyclonal T cell expansion in cancer patients. Cancer Cell 34, 775–791.e3.

Sun, Z., Fourcade, J., Pagliano, O., Chauvin, J.M., Sander, C., Kirkwood, J.M., and Zarour, H.M. (2015). IL10 and PD-1 cooperate to limit the activity of tumorspecific CD8+ T Cells. Cancer Res. 75, 1635–1644.

Naing, A., Papadopoulos, K.P., Autio, K.A., Ott, P.A., Patel, M.R., Wong, D.J., Falchook, G.S., Pant, S., Whiteside, M., Rasco, D.R., et al. (2016). Safety, antitumor activity, and immune activation of pegylated recombinant human interleukin-10 (AM0010) in patients with advanced solid tumors. J. Clin. Oncol. 34, 3562–3569. Nishino, M., Sholl, L.M., Hodi, F.S., Hatabu, H., and Ramaiya, N.H. (2015). AntiPD-1-related pneumonitis during cancer immunotherapy. N. Engl. J. Med. 373, 288–290. Olson, B., Li, Y., Lin, Y., Liu, E.T., and Patnaik, A. (2018). Mouse models for cancer immunotherapy research. Cancer Discov. 8, 1358–1365. Park, C.Y., Majeti, R., and Weissman, I.L. (2008). In vivo evaluation of human hematopoiesis through xenotransplantation of purified hematopoietic stem cells from umbilical cord blood. Nat. Protoc. 3, 1932–1940. Peng, W., Liu, C., Xu, C., Lou, Y., Chen, J., Yang, Y., Yagita, H., Overwijk, W.W., Lizee, G., Radvanyi, L., and Hwu, P. (2012). PD-1 blockade enhances T-cell migration to tumors by elevating IFN-gamma inducible chemokines. Cancer Res. 72, 5209–5218. Rafi-Janajreh, A.Q., Chen, D., Schmits, R., Mak, T.W., Grayson, R.L., Sponenberg, D.P., Nagarkatti, M., and Nagarkatti, P.S. (1999). Evidence for the involvement of CD44 in endothelial cell injury and induction of vascular leak syndrome by IL-2. J. Immunol. 163, 1619–1627.

914 Cancer Cell 35, 901–915, June 10, 2019

Takahashi, T., Tagami, T., Yamazaki, S., Uede, T., Shimizu, J., Sakaguchi, N., Mak, T.W., and Sakaguchi, S. (2000). Immunologic self-tolerance maintained by CD25(+)CD4(+) regulatory T cells constitutively expressing cytotoxic T lymphocyte-associated antigen 4. J. Exp. Med. 192, 303–310. Tan, J.C., Indelicato, S.R., Narula, S.K., Zavodny, P.J., and Chou, C.C. (1993). Characterization of interleukin-10 receptors on human and mouse cells. J. Biol. Chem. 268, 21053–21059. Tang, H., Liang, Y., Anders, R.A., Taube, J.M., Qiu, X., Mulgaonkar, A., Liu, X., Harrington, S.M., Guo, J., Xin, Y., et al. (2018). PD-L1 on host cells is essential for PD-L1 blockade-mediated tumor regression. J. Clin. Invest. 128, 580–588. Thompson, E.D., Enriquez, H.L., Fu, Y.X., and Engelhard, V.H. (2010). Tumor masses support naive T cell infiltration, activation, and differentiation into effectors. J. Exp. Med. 207, 1791–1804. Tilg, H., van Montfrans, C., van den Ende, A., Kaser, A., van Deventer, S.J., Schreiber, S., Gregor, M., Ludwiczek, O., Rutgeerts, P., et al. (2002). Treatment of Crohn’s disease with recombinant human interleukin 10 induces the proinflammatory cytokine interferon gamma. Gut 50, 191–195. Tomoyose, M., Mitsuyama, K., Ishida, H., Toyonaga, A., and Tanikawa, K. (1998). Role of interleukin-10 in a murine model of dextran sulfate sodiuminduced colitis. Scand. J. Gastroenterol. 33, 435–440. Wang, Y., Sun, S.N., Liu, Q., Yu, Y.Y., Guo, J., Wang, K., Xing, B.C., Zheng, Q.F., Campa, M.J., Patz, E.F., Jr., et al. (2016). Autocrine complement inhibits

IL10-dependent T-cell-mediated antitumor immunity to promote tumor progression. Cancer Discov. 6, 1022–1035.

(2017). Overall survival with combined Nivolumab and Ipilimumab in advanced melanoma. N. Engl. J. Med. 377, 1345–1356.

Wei, H., Cai, H., Jin, Y., Wang, P., Zhang, Q., Lin, Y., Wang, W., Cheng, J., Zeng, N., Xu, T., and Zhou, A. (2017a). Structural basis of a novel heterodimeric Fc for bispecific antibody production. Oncotarget 8, 51037–51049.

Wunderlich, M., Chou, F.S., Link, K.A., Mizukawa, B., Perry, R.L., Carroll, M., and Mulloy, J.C. (2010). AML xenograft efficiency is significantly improved in NOD/SCID-IL2RG mice constitutively expressing human SCF, GM-CSF and IL-3. Leukemia 24, 1785–1788.

Wei, S.C., Levine, J.H., Cogdill, A.P., Zhao, Y., Anang, N.A.S., Andrews, M.C., Sharma, P., Wang, J., Wargo, J.A., Pe’er, D., and Allison, J.P. (2017b). Distinct cellular mechanisms underlie anti-CTLA-4 and anti-PD-1 checkpoint blockade. Cell 170, 1120–1133.e17.

Yang, X.M., Zhang, X.M., Fu, M.L., Weichselbaum, R.R., Gajewski, T.F., Guo, Y.J., and Fu, Y.X. (2014). Targeting the tumor microenvironment with interferon-beta bridges innate and adaptive immune responses. Cancer Cell 25, 37–48.

Wilke, C.M., Wei, S., Wang, L., Kryczek, I., Kao, J., and Zou, W. (2011). Dual biological effects of the cytokines interleukin-10 and interferon-gamma. Cancer Immunol. Immunother. 60, 1529–1541.

Zhu, J., Powis de Tenbossche, C.G., Cane, S., Colau, D., van Baren, N., Lurquin, C., Schmitt-Verhulst, A.M., Liljestrom, P., Uyttenhove, C., and Van den Eynde, B.J. (2017). Resistance to cancer immunotherapy mediated by apoptosis of tumor-infiltrating lymphocytes. Nat. Commun. 8, 1404.

Wolchok, J.D., Chiarion-Sileni, V., Gonzalez, R., Rutkowski, P., Grob, J.J., Cowey, C.L., Lao, C.D., Wagstaff, J., Schadendorf, D., Ferrucci, P.F., et al.

Cancer Cell 35, 901–915, June 10, 2019 915

STAR+METHODS KEY RESOURCES TABLE

REAGENT or RESOURCE

SOURCE

IDENTIFIER

InVivoMAb anti-mouse CD4 (GK1.5)

BioXcell

Cat# BE0003-1

InVivoMAb anti-mouse CD8 (YTS 169.4)

BioXcell

Cat# BE0117

Antibodies

InVivoMAb anti-mouse PDL1 (10F.9G2)

BioXcell

Cat# BE0101

InVivoMAb anti-mouse CTLA-4 (9D9)

BioXcell

Cat# BE0164

InVivoMAb anti-mouse IFN g (R4-6A2)

BioXcell

Cat# BE0054

InVivoMAb anti-mouse IL-12 p40 (C17.8)

BioXcell

Cat# BE0051

Anti-CD45 (Flow cytometry, 30-F11)

BioLegend

Cat# 103126

Anti-CD8 (Flow cytometry, 53-6.7)

BioLegend

Cat# 100730

Anti-CD8a (Flow cytometry, KT15)

Invitrogen

Cat# MA5-16759

Anti- Active Caspase-3 (Flow cytometry, C92-605)

BD Biosciences

Cat# C92-605

Peroxidase AffiniPure Goat Anti-Human IgG (H+L)

Jackson ImmunoResearch

Cat# 109-035-088

AffiniPure Goat Anti-Human IgG, Fcg fragment specific

Jackson ImmunoResearch

Cat# 109-005-098

Annexin V (Flow cytometry)

BioLegend

Cat# 640912

Fixable Viability Dye eFluor 506

Thermo Fisher

Cat# 65-0866-18

7-AAD Viability Staining Solution (Flow cytometry)

BioLegend

Cat# 420404

iTAg Tetramer/PE - H-2 Kb OVA (SIINFEKL)

MBL

Cat# TB-5001-1

Donkey Anti-Human IgG (H+L)

Jackson ImmunoResearch

Cat# 709-116-149

Anti-FcgIII/II receptor (clone 2.4G2)

BD Biosciences

Cat# 553141

Erbitux (Cetuximab)

Pharmacy

N/A

FTY720 (hydrochloride)

Selleckchem

Cat# S5002

Clophosome-A - Clodronate Liposomes (Anionic)

FormuMax Scientific

Cat# F70101C-A

IRDye 800CW NHS Ester

Fisher: LI-COR

Cat# NC9690013

TMB Solution (1X)

eBioscience

Cat# 00-4201-56

Diphtheria toxin

Sigma- Aldrich

Cat# D0564

OVA257-264 (SIINFEKL)

Invivogen

Cat# vac-sin

SIYRYYGL (SIY) peptide

Sigma- Aldrich

N/A

Ovalbumin

Sigma- Aldrich

Cat# A2512

Sulfadiazine/ Trimethoprim (Aurora Pharmaceutical LLC)

UTSW-Veterinary Drug Services

Cat# 302

Chemicals, Peptides, and Recombinant Proteins

Dulbecco’s Modified Eagle’s Medium

Sigma- Aldrich

Cat# D6429

GE Healthcare Ficoll-Paque PLUS Media

Fisher

Cat# 45-001-750

Recombinant murine IFN-g

Fisher

Cat# 50-813-664

Recombinant murine IL-10

PeproTech

Cat# 210-10

Recombinant mouse IL-12

BioLegend

Cat# 577002

Recombinant mouse GM-CSF

BioLegend

Cat# 576306

Critical Commercial Assays BD Cytometric Bead Array (CBA) Mouse Inflammation Kit

BD Biosciences

Cat# 552364

BD Mouse IFN-g ELISPOT Sets

BD Biosciences

Cat# 551083

SsoAdvanced Uni SYBR Grn Supmix

Bio-Rad

Cat# 1725272

RNeasy Plus Mini Kit

Qiagen

Cat# 74134 Cat# 1725035

iScript gDNA Clear cDNA Synthesis Kit

Bio-Rad

True-Nuclear Transcription Factor Buffer Set

BioLegend

Cat# 424401

EasySep Mouse CD8+ T Cell Isolation Kit

STEMCELL

Cat# 19853 (Continued on next page)

e1 Cancer Cell 35, 901–915.e1–e4, June 10, 2019

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

EasySep Human CD34 Positive Selection Kit II

STEMCELL

Cat# 17856

CFSE Cell Division Tracker Kit

BioLegend

Cat# 423801

B16

ATCC

Cat# CRL-6322

TC-1

Gift from Dr. T.C. Wu

N/A

MC38

ATCC

N/A

Experimental Models: Cell Lines

FreeStyleTM 293-F

Thermo Fisher

Cat# R79007

A431

ATCC

Cat# CRL-1555

C57BL/6

UTSW breeding Core

Cat# 000664

NOD SCID IL-2 receptor g knockout (NSG)

UTSW breeding Core

Cat# 005557

B6.129S7-Rag1tm1Mom/J

Jackson Laboratory

Cat# 002216

NOD.Cg-Prkdcscid Il2rgtm1Wjl Tg(CMV-IL3, CSF2, KITLG) 1Eav/MloySzJ

Jackson Laboratory

Cat# 013062

Experimental Models: Organisms/Strains

B6(Cg)-Zbtb46 tm1(HBEGF)Mnz /J

Jackson Laboratory

Cat# 019506

B6.129S(C)-Batf3tm1Kmm/J

Jackson Laboratory

Cat# 013755

B6.129S2-Il10rbtm1Agt/J

Jackson Laboratory

Cat# 005027

B6.129S7-Ifngtm1Ts/J

Jackson Laboratory

Cat# 002287

B6.Cg-Thy1a/Cy Tg(TcraTcrb)8Rest/J

Jackson Laboratory

Cat# 005023

C57BL/6-Tg(TcraTcrb)1100Mjb/J

Jackson Laboratory

Cat# 003831

Mouse b-actin forward: ACACCCGCCACCAGTTCGC

This paper

N/A

Mouse b-actin reverse: ATGGGGTACTTCAGGGTCAGGATA

This paper

N/A

Mouse IL-12 p40 forward: GAAGCTGGTGCTGTAGTT

This paper

N/A

Mouse IL-12 p40 reverse: GAGTCATAGGCTCTGGAA

This paper

N/A

GraphPad Prism software 7.0

GraphPad Software, Inc.

https://graphpad.com/scientificsoftware/prism/

Image Lab Software

Bio-Rad

http://www.bio-rad.com/en-us/category/ image-analysis-software

CytExpert

Beckman Coulter, Inc

https://www.beckman.com/coulter-flowcytometers/cytoflex/cytexpert

FlowJo

Tree Star Inc.

https://www.flowjo.com/solutions/flowjo

Oligonucleotides

Software and Algorithms

CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Yang-Xin Fu ([email protected]). EXPERIMENTAL MODEL AND SUBJECT DETAILS Mice Six- to eight-weeks-old female C57BL/6J and NOD SCID IL-2 receptor g knockout (NSG) mice were purchased from UT Southwestern breeding core. B6.129S7-Rag1tm1Mom/J (Rag1-/-), NOD.Cg-Prkdcscid Il2rgtm1Wjl Tg(CMV-IL3, CSF2, KITLG) 1Eav/MloySzJ (NSG-SGM3), B6(Cg)-Zbtb46tm1(HBEGF)Mnz/J (zDC-DTR), B6.129S(C)-Batf3tm1Kmm/J (Batf3-/-), B6.129S2-Il10rbtm1Agt/J (Il10r-/-), B6.129S7-Ifngtm1Ts/J (Ifng-/-), C57BL/6J-Tg(TcraTcrb)1100Mjb/J (OT1 TCR transgenic mice), B6.Cg-Thy1a/Cy Tg(TcraTcrb)8Rest/ J (Pmel-1 TCR transgenic mice) mice were purchased from Jackson Laboratory. All mice were maintained in specific pathogenfree animal facility and all experiments were conducted according to regulations of the Institutional Animal Care and Use Committee of the University of Texas Southwestern Medical Center.

Cancer Cell 35, 901–915.e1–e4, June 10, 2019 e2

Cell Lines B16 and MC38 are murine melanoma and colon adenocarcinoma cell lines, respectively. TC-1 is a tumor cell line transformed from C57BL/6J primary mouse lung epithelial cells. A431 is a human epidermoid carcinoma cell line. B16-cEGFR, TC1-cEGFR, MC38-cEGFR were sorted and sub-cloned after being transduced by lentivirus expressing murine-human chimeric EGFR (cEGFR, is a full-length of the murine EGFR with six mutated amino acids that are critical for human EGFR binding to Cetuximab). B16, MC38 and A431 cell lines were purchased from American Type Culture Collection (ATCC). TC-1 cells were kindly provided by Dr. T. C. Wu at John Hopkins University. All the cells were cultured in 5% CO2 and maintained in vitro in Dulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated fetal bovine serum (Sigma-Aldrich), 100 U/mL penicillin, and 100 mg/mL streptomycin. FreeStyleTM 293-F cells (purchased from Invitrogen) are derived from the 293 cell line and used for large-scale transfection and protein production, the cells were cultured according to the manufacturer’s protocol. Humanized Mice and Human Peripheral Blood Mononuclear Cells Human cord blood samples were obtained from UT Southwestern (UTSW) Parkland Hospital according to the regulation and the use approval of human cord blood at UTSW medical center. Human peripheral blood mononuclear cells (PBMC) were purified from cord blood by Ficoll density gradient centrifugation. Human CD34+ cells were further purified by positive immunomagnetic selection with anti-human CD34+ microbeads according to the manufacturer’s protocol (STEMCELL Technologies). 1x105 CD34+ cells were intravenously injected into recipient mice. 4-weeks-old NSG-SGM3 female recipient mice were irradiated with 100 cGy (X-ray irradiation with X-RAD 320 irradiator) one day prior to CD34+ cells transfer. Irradiated mice were orally administrated with Sulfadiazine/Trimethoprim (Aurora Pharmaceutical LLC) water for 2 weeks (Park et al., 2008; Wunderlich et al., 2010). 12 weeks after engraftment, humanized mice with over 40% human CD45+ cells reconstitution were used for tumor study. METHOD DETAILS Production of Bispecific Fusion Proteins Based on the heterodimeric Fc variant KiHss-AkKh platform as previously described (Wei et al., 2017a), the Fab fragment of Cetuximab (anti-hEGFR) was fused with knob variant Fc region, and the IL-10 dimer was fused with hole variant Fc region. CmAb-(IL10)2 was generated by transient co-transfection of two arms of plasmids into FreeStyleTM 293-F cells. The supernatant containing the fusion protein was purified using Protein A affinity chromatography according to the manufacturer’s protocol. The heterogeneity was confirmed by SDS-PAGE and the purity was checked by SEC-HPLC. CmAb-IL10, CmAb-IL2, TmAb-(IL10)2 and Fc mutant CmAb-(IL10)2 were generated and produced in a similar way to CmAb-(IL10)2. In Vivo Imaging CmAb-(IL10)2 was labeled with IRDye 800CW Dye following the manufacturer’s instructions. TC-1 (left flank) and TC-1-cEGFR (right flank) tumor-bearing NSG mice were injected intravenously by IRDye 800CW-labeled CmAb-(IL10)2 (4 mg/kg) on day 8 after tumor cell inoculation. Fluorescence radiances was determined by LI-COR Pearl Trilogy in vivo imaging system at 0, 24, 48, 72 and 96 hr after injection. Tumors were collected for ex vivo fluorescence measurement at 96 hr post injection. Tumor Growth and Treatment B16-cEGFR (33105), MC38-cEGFR (33105), TC1-cEGFR (13106) or A431 (1x106) cells were injected subcutaneously on the right flank of mice. 7-11 days after tumor inoculation, mice were treated three times by intraperitoneal injection of CmAb-(IL10)2 at 4 mg/kg or intratumoral injection of CmAb-(IL10)2 at 0.4 mg/kg by every three days for a total of three times (same moles were used for control IgG, Cetuximab and TmAb-(IL10)2). Anti-CD8 antibodies (200 mg) and anti-CD4 antibodies (200 mg) were administrated intraperitoneally three times at an interval of three days during the treatment. Diphtheria toxin (DT) (4 mg/kg) was injected intraperitoneally every other day starting on 2 days prior to the first CmAb-(IL10)2 treatment. The macrophage depleting reagent (Clophosome, 150 mL) or control liposome was injected intraperitoneally one day prior to the first CmAb-(IL10)2 treatment, and again 5 days later. The combination of anti-PD-L1 (200 mg) and anti-CTLA-4 (200 mg) were administered intraperitoneally to mice at an interval of three days for a total of two or three times. FTY720 (1 mg/kg, i.p.) was administrated once a day for three times starting the first day of CmAb-(IL10)2 treatment, then decreased to every other day until the end of the experiment. Drinking water was maintained with 2 mg/mL FTY720 throughout the experiment. Tumor volumes were measured with a caliper by the length (a), width (b) and height (h) and calculated as tumor volume = abh/2. IFN-g Enzyme-Linked Immunosorbent Spot Assay (ELISPOT) B16-cEGFR-OVA (13106) were injected subcutaneously on the right flank of C57BL/6J mice. 9 days after the first CmAb-(IL10)2 or control antibody treatment, spleens were processed into single cell suspensions and re-suspended in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mmol/l L-glutamine, 100 U/mL penicillin, and 100 mg/mL streptomycin. A total of 43105 spleen cells were used for the assay. 5 mg/mL SIINFEKL peptide (OVA257-264) or control SIYRYYGL (SIY) peptide were used to restimulate the antigen specific T cells. After 48 hr of incubation, the IFN-g production was determined with an IFN-g ELISPOT assay kit according to the manufacturer’s protocol (BD Biosciences). The visualized spots were enumerated with the CTL-ImmunoSpot S6 Analyzer (Cellular Technology Limited). e3 Cancer Cell 35, 901–915.e1–e4, June 10, 2019

Generation of Bone Marrow Chimeras C57BL/6J mice were irradiated with a single dose of 10 Gray. Then the irradiated mice were adoptively transferred (i.v.) with 3x106 bone marrow cells from zDC-DTR transgenic donor mice at the same day. The mice were treated with sulfamethoxazole and trimethoprim (Bactrim) antibiotics diluted in drinking water for 4 weeks after reconstitution. After approximately 12 weeks, the mice were inoculated with tumor cells. In Vitro Co-culture of Bone Marrow Dendritic Cells (BMDC) and T Cells Single-cell suspensions of bone marrow (BM) cells were collected from tibias and femurs of C57BL/6J or Il10r-/- mice as previously described (Yang et al., 2014). The BM cells were placed in 24 wells plate and cultured with complete RPMI 1640 medium containing 20 ng/mL recombinant mouse GM-CSF (rmGM-CSF). Fresh media with rmGM-CSF was added into the culture on day 3 and day 6. The BMDCs were harvested and ready to use on day 7. CD8+ T cells were isolated from lymph nodes and spleens of OT1 transgenic, Il10r-/- OT1 transgenic or Pmel-1 transgenic mice with a negative CD8+ T cell isolation kit (Stemcell Technologies) following the manufacturer’s instructions. 23104 BMDCs were co-cultured with 23105 CFSE labeled CD8+ T cells in the presence of 100 mg/mL OVA protein with or without 100 ng/mL CmAb-(IL10)2. 24-72 hr later, the supernatants were collected for cytokines measurement and all the cells were collected for flow cytometry analysis. Generation of Antigen Activated CD8+ OT1 T Cells In Vitro Single cell suspensions from lymph nodes and spleens of OT1 transgenic mice were stimulated in the presence of SIINFEKL peptide (1 mg/mL) and IL-2 (50 IU/mL). 48 hr after activation, CD8+ T cells were isolated and used for co-culture in vitro or treatment in vivo. Flow Cytometry Analysis Single cell suspensions of cells were incubated with anti-FcgIII/II receptor (clone 2.4G2) for 20 minutes to block non-specific binding before stained with the conjugated antibodies. 7-AAD Viability Staining Solution or Fixable Viability Dye eFluor 506 was used to exclude dead cells. CFSE Cell Division Tracker Kit was used to label T cells. Activated caspase 3+ cells were stained intracellularly by using True-Nuclear transcription factor buffer set (BioLegend) following the manufacturer’s instructions. To assess the EGFR binding, EGFR expressing cells were firstly stained with Cetuximab, CmAb-(IL10)2, TmAb-(IL10)2 or Control IgG, then PE conjugated donkey anti-human IgG was used as a secondary antibody. BD Cytometric Bead Array (CBA) Mouse Inflammation Kit was used to measure the cytokines in the supernatants from in vitro cell culture or mice serum according to the manufacturer’s protocol (BD Biosciences). Data were collected on CytoFLEX flow cytometer (Beckman Coulter, Inc) and analyzed by using CytExpert (Beckman Coulter, Inc) or FlowJo (Tree Star Inc., Ashland, OR) softwares. Enzyme-Linked ImmunoSorbent Assay (ELISA) Microtiter plates (Corning Costar) were coated with 2 mg/mL (100 mL/well) capture antibody (AffiniPure Goat Anti-Human IgG, Fcg fragment specific) overnight at 4  C. After washing away the unbound antibody and block, serially diluted serum samples from CmAb-(IL10)2 (1 mg/kg) treated mice were added and incubated at 37 C for 1.5 hr. After washing, Peroxidase AffiniPure Goat Anti-Human IgG (H+L) was added and incubated at room temperature for 1.5 hr. Finally, the plates were visualized by adding 100 mL of TMB solution and read at 450 nm using the SPECTROstarNano (BMG LABTECH). RNA Extraction and Quantitative Real-Time PCR Total RNA from DCs was extracted with RNeasy Plus Mini Kit and reversed transcribed with iScript gDNA Clear cDNA Synthesis Kit. Real-time PCR was performed with SsoAdvanced Universal SYBR Green Supermix according to the manufacturer’s instructions and different primer sets (b-actin, forward primer: 5’- ACACCCGCCACCAGTTCGC, reverse primer: 5’-ATGGGGTACTTCAGGGTCAGGATA; IL-12p40, forward primer: 5’- GAAGCTGGTGCTGTAGTT, reverse primer: 5’-GAGTCATAGGCTCTGGAA) on CFX Connect Real-Time PCR Detection System (Bio-Rad Laboratories). The levels of gene expression were normalized to b-actin. QUANTIFICATION AND STATISTICAL ANALYSIS All data were analyzed using GraphPad Prism statistical software (version 7.04, GraphPad Software Inc.). Two way ANOVA was used to analyze the tumor growth, and unpaired two-tailed t tests was used to analyze the other data. A value of p < 0.05 was considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001).

Cancer Cell 35, 901–915.e1–e4, June 10, 2019 e4