Cyclophosphamide resets dendritic cell homeostasis and enhances antitumor immunity through effects that extend beyond regulatory T cell elimination

Abstract

Using a model of established malignancy, we found that cyclophosphamide (Cy), administered at a dose not requiring hematopoietic stem cell support, is superior to low-dose total body irradiation in augmenting antitumor immunity. We observed that Cy administration resulted in expansion of tumor antigen-specific T cells and transient depletion of CD4(+)Foxp3(+) regulatory T cells (Tregs). The antitumor efficacy of Cy was not improved by administration of anti-CD25 monoclonal antibody given to induce more profound Treg depletion. We found that Cy, through its myelosuppressive action, induced rebound myelopoiesis and perturbed dendritic cell (DC) homeostasis. The resulting DC turnover led to the emergence of tumor-infiltrating DCs that secreted more IL-12 and less IL-10 compared to those from untreated tumor-bearing animals. These newly recruited DCs, originating from proliferating early DC progenitors, were fully capable of priming T cell responses and ineffective in inducing expansion of Tregs. Together, our results show that Cy-mediated antitumor effects extend beyond the well-documented cytotoxicity and lymphodepletion and include resetting the DC homeostasis, thus providing an excellent platform for integration with other immunotherapeutic strategies.

Figures

Fig. 1

a Nonmyeloablative conditioning with Cy augments endogenous antitumor immunity without the need for hematopoietic cell transplantation. Cy alone increases tumor-free survival. i Experimental schema. ii BALB/c mice were injected with 105 CT26 tumor cells using hemispleen injection technique and then received no additional treatment (open diamond) or were administered 200 cGy of TBI (open square) on day 10 after surgery, 200 mg/kg of Cy intraperitoneally on day 13 (dark filled diamond), combination of TBI and Cy (dark filled square) or Cy followed by vaccine on day 16 (open circle). P < 0.005 for Cy vs. no treatment; P < 0.01 for no treatment vs. TBI, TBI + Cy vs. no treatment, and Cy vs. Cy + vaccine. b Intact IFN-γ-mediated T cell responses are necessary for preservation of Cy-mediated antitumor effcacy. i Experimental schema. ii BALB/c or IFN-γKO mice were injected with CT26 and left untreated (dark filled diamond—BALB/c; dark filled square—IFN-γKO) or Cy was administered (open diamond—BALB/c; open square—IFN-γKO) on day 13. P < 0.0001 for BALB/c no treatment vs. BALB/c Cy, IFN-γKO Cy, and IFN-γKO no treatment. c Depletion of CD4+ T cells does not affect Cy efficacy, while CD8+ T cell elimination completely eliminates Cy's antitumor-effect. i Experimental schema. ii BALB/c mice were injected with CT26 and then treated intraperitoneally on days 6, 10, and 12 with 0.25 mg of IgG2b (dark filled diamond), anti-CD4 (open square), or anti-CD8 (dark filled square) and then injected with Cy on day 13 after tumor injection. P < 0.005 for Cy + isotype and Cy + anti-CD4 vs. Cy + anti-CD8. d Treg depletion after Cy administration does not improve survival of tumor-bearing animals. i Experimental schema. ii BALB/c mice were injected with CT26, and treated with Cy or left untreated. Animals then received 0.5 mg of anti-CD25 (open square—Cy treated; dark filled square—untreated) or isotype control (dark filled diamond—Cy treated; open diamond—untreated) on days 13 and 16 after tumor injection. P < 0.005 for Cy + isotype and Cy + anti-CD25 vs. isotype only and anti-CD25 only. In all experiments survival was monitored and is plotted as a function of time after tumor injection. Data are representative of at least three (a, b) or two (c, d) independent experiments

Fig. 2

Cy suppresses regulatory networks in the target organ and enables homing and persistence of tumor specific AH-1+CD8+ T cells despite heavy tumor burden. Twenty-three days after tumor injection, spleens and livers from three to five mice per treatment group were analyzed for the presence of T cell subsets. a Cy treatment increases CD8+ T cell content of tumor-infiltrated organ while exhibiting minimal effect on CD4+ T cell subset. b Greater proportion of CD8+ T cells from Cy-treated animals secrete IFN-γ. c Presence of tumor specific AH-1+CD8+ T cells is increased after treatment with Cy. d Cy-mediated depletion of Foxp3+CD4+ Tregs is of short duration and by day 10 after Cy injection, Cy treated and untreated animals cannot be distinguished based on composition of infiltrate at tumor site. e Reduced expression of CD25 by Foxp3+CD4+ T cells retrieved from livers of untreated and Cy-treated tumor-bearing mice. i Representative histogram. ii Bar graphs show mean fluorescent intensity (MFI) of CD25 on gated Foxp3+CD4+ T cells in liver on day 23 after tumor inoculation and day 10 after Cy treatment. Five mice per treatments group were analyzed. f Cy decreases the percentage of NK cells in the pool of liver-infiltrating leukocytes. P values were calculated using one-way ANOVA. Data are representative of three separate experiments

Fig. 3

Cy treatment perturbs the DC homeostasis and increases presence of in vivo proliferating hepatic and splenic conventional DCs. Cy-treated or untreated tumor-bearing animals were sacrificed on day 23 after tumor inoculation, livers and spleens were harvested and DC subset composition in mononuclear infiltrate/splenocytes was analyzed. Cy-mediated increase in percentage, as well as in total liver CD11c+ DCs (a), is reflected in the composition of CD11c+ population (b). P values were calculated using one-way ANOVA. Data are representative of three independent experiments. c, d CD11c+ DCs in Cy-treated animals have increased proliferation rate in spleen and liver. On day 21 after tumor implantation tumor-bearing Cy-treated or untreated animals were injected with BrdU and kept on BrdU-containing water until analysis. Representative contour plots show Ki67 and BrdU profile of hepatic and splenic CD11c+MHCII+ DCs. Graphs show percentage of Ki67+BrdU+ among total hepatic and splenic CD11c+ DCs. P values were calculated using one-way ANOVA. Data are representative of at least two independent experiments

Fig. 4

Cy perturbs DC homeostasis by mobilizing DC precursors in naïve and tumor-bearing mice. a Cy treatment induces rebound monocytosis. Naïve BALB/c mice were intraperitoneally injected with 200 mg/kg of Cy and changes in circulating monocytes were sequentially monitored. b Monocytes are not the primary source of post-Cy occurring CD11c+ DCs in spleen and liver of tumor-bearing mice. Seven days after adoptive transfer of congenic CD45.1+CD115+ monocytes to Cy-treated or untreated tumor-bearing BALB/c (CD45.2+) mice, livers and spleens were analyzed for the presence of CD45.1+CD11c+ DCs. Bar graphs show % conversion of monocytes into CD11c+ DCs. c Gene expression profile of hepatic DCs is suggestive of their CDP, rather than monocytic, origin. On day 23 after tumor inoculation, total RNA was extracted from isolated hepatic CD11c+ DCs and analyzed for expression of MIP-1α and MCP-1 using quantitative RT-PCR. d Cy mobilizes primitive HSCs to peripheral circulation. After Cy administration, naïve BALB/c mice were serially bled and analyzed for presence of Lin−CD117+ HSCs. e, f Cy treatment mobilizes early DC progenitors in BM. e Representative contour plots and histograms show the profile of naïve and Cy-treated (day 10) Lin− BM cells as assessed through early DC progenitor-defining markers. f Changes in the total numbers of early DC progenitors in BM of Cy-treated animals. g Cy-mobilized early DC progenitors are recruited to livers and spleens of tumor-bearing animals. CT26 injected animals were treated with Cy or left untreated and analyzed for presence of Lin−CD117+CD135+ cells in livers and spleens 23 days after tumor injection. Bar graphs show threefold increase in percentage of early DC progenitors in livers and spleens of Cy-treated animals

Fig. 5

Cy influences the cytokine profile of DCs at the tumor site and reverses tumor-induced DC paralysis. a Liver CD11c+ cells were isolated on day 23 and total RNA was extracted. cDNA was generated by reverse transcription and analyzed for presence of TNF-α and IL-12p40 using quantitative RT-PCR. b On day 23 post-tumor implantation, isolated hepatic and splenic CD11c+ cells were incubated for 48 h in complete medium and production of IL-12p40 and IL-10 was assessed by ELISA. c Following Cy treatment, newly occurring liver CD11c+ cells are inefficient in stimulating naïve CD4+CD25+ T cells. Liver CD11c+ DCs were isolated on day 23 and cultured with naïve BALB/c CD4+CD25+ T cells at a 1:1 ratio for 72 h after which 3[H] Thymidine was added. Data are presented as stimulation index (determined by comparing proliferation of CD4+CD25+ T cells cocultured with isolated DCs with that of CD4+CD25+ T cells cultured in medium alone according to the formula: measured cpm/cpm medium). d Functional DC paralysis in the liver and spleen is reversed with Cy treatment. On day 23 after tumor injection CD11c+ cells were isolated from spleens and livers of naïve, Cy-treated and untreated tumor-bearing mice and cultured overnight in complete medium with addition of nothing or 5 μg/ml of CpG1826. The next day, DCs were irradiated and cocultured with 105 C57BL/6 LN cells (ratio of T cells to DCs in liver 1:1, in spleen 1:1 to 1,000:1). 3[H] Thymidine was added after 72 h of stimulation. Data are presented as stimulation index