Fasting-Mimicking Diet Reduces HO-1 to Promote T Cell-Mediated Tumor Cytotoxicity

Abstract

Immune-based interventions are promising strategies to achieve long-term cancer-free survival. Fasting was previously shown to differentially sensitize tumors to chemotherapy while protecting normal cells, including hematopoietic stem and immune cells, from its toxic side effects. Here, we show that the combination of chemotherapy and a fasting-mimicking diet (FMD) increases the levels of bone marrow common lymphoid progenitor cells and cytotoxic CD8(+) tumor-infiltrating lymphocytes (TILs), leading to a major delay in breast cancer and melanoma progression. In breast tumors, this effect is partially mediated by the downregulation of the stress-responsive enzyme heme oxygenase-1 (HO-1). These data indicate that FMD cycles combined with chemotherapy can enhance T cell-dependent targeted killing of cancer cells both by stimulating the hematopoietic system and by enhancing CD8(+)-dependent tumor cytotoxicity.

Copyright © 2016 Elsevier Inc. All rights reserved.

Figures

Figure 1. A Fasting-mimicking diet (FMD) alone or in combination with chemotherapy is as effective as short-term starvation (STS) in reducing tumor progression

(A-D) 12-week old female BALB/c mice were grafted with breast cancer (4T1) cells and subject to (A) multiple cycles of FMD or STS alone or to (C, D) a combination of FMD and the chemotherapy drugs (C) doxorubicin (DXR) or (D) cyclophosphamide (CP). (B) Circulating IGF-1 levels at the end of STS or FMD in (A) were measured. (E) 12-week old female C57BL/6 mice were grafted with melanoma (B16) cells and subject to multiple cycles of FMD alone or in combination with DXR. Tumor volume at multiple time-points (on the left) and immediately prior to euthanasia (on the right) are reported for each set of treatments. The animals receiving chemotherapy were injected at the end of each FMD cycle (shaded area). [(A, B) n=10, (C) n=13, (D) n=10, and (E) n=15]. Data represented as mean ± SEM. One-way ANOVA (Tukey post-analysis test) was performed. p-values <0.05, 0.01 and 0.001 are indicated as *, **, and ***, respectively. See also Figure S1.

Figure 2. FMD in combination with doxorubicin (DXR) promotes accumulation of tumor-infiltrating lymphocytes (TILs) in the tumor bed

(A) Bone marrow collected from BALB/c mice undergoing FMD/DXR treatments (see Figure S3A) was collected at the end of the experiment and analyzed with FACS (n=6) to assess the amount of Common Lymphoid Progenitors (CLP). (B) Breast cancer (4T1) tumor tissues collected at the end of the experiment (see Figure S3A) were analyzed using immunohistochemistry to assess CD3+/CD8+ TILs (n=8) (quantification on the right). (C, E) CD3+/CD8+ and (D, E) CD3+/CD4+/CD25+were also assessed by FACS analysis (n=7) in tumor tissue collected from a separate, equivalent experiment (see Figure 1C). (F) Melanoma (B16) tissue collected (see Figure 1E) was also processed to assess the levels of TILs (quantification on the right). Bar scale for DAPI, CD3 and CD8 is 75 μm. Bar scale for CD3+/CD8+ is 12.5 μm Data represented as mean ± SEM. The significance of the differences between experimental groups was determined by using one-way ANOVA (Tukey post-analysis test). p-values <0.05, 0.01 and 0.001 are indicated as *, **, and ***, respectively. See also Figure S2.

Figure 3. CD3+/CD8+ lymphocytes have a key role in FMD-mediated differential stress sensitization (DSS) to chemotherapy

(A) The effect of FMD on breast tumor growth (4T1) in immunodeficient BALB/c nude mice (n=15) (see Figure S4A-E); (B) The survival of wild type (WT) and nude mice injected with high dose of DXR and undergoing ad lib or FMD regimen. (C-H) 4T1 breast tumor-bearing mice were treated with DXR, FMD+DXR alone or in combination with [.alpha]CD8 monoclonal antibody and (C-E) circulating levels of (C, E) CD3+/CD8+ and (D, E) CD3+/CD4+/CD25+ were determined by FACS (n=7). (F) Tumor volumes of FMD+DXR and FMD+DXR+αCD8 were measured at multiple time points, and (G) TILs were also assessed in tumor samples collected from the same animals. (H) Circulating lymphocytes from DXR, FMD+DXR, DXR+αCD8, and FMD+DXR+αCD8 these animals were collected and cultured ex vivo with 4T1 cells for 24 hours and viability was assessed by MTT reduction. (I) Mice were immunized by subcutaneous inoculation (in the left flank) with 4T1 breast cancer cells preconditioned in vitro with either ad lib- (2 g/L glucose+10% FBS) or STS-medium (0.5 g/L glucose+1% FBS) with or without doxorubicin (5 μM). 7 days after the immunization, the same animals were inoculated with naïve 4T1 cells in the right flank (n=10). (J, K) Tumor progression for the immunization (left) and the naïve (right) sides are reported. Data represented as mean ± SEM. One-way ANOVA (Tukey post-analysis test) and Log-rank (Mantel-Cox) test (survival) was performed. Comparisons between groups were performed with Student's t test. p-values <0.05, 0.01 and 0.001 are indicated as *, **, and ***, respectively. See also Figure S3.

Figure 4. Heme oxygenase-1 (HO-1) has a key role in mediating STS-dependent Differential Stress Sensitization (DSS)

(A) HO-1 expression levels of grafted 4T1 tumor and normal (liver and skeletal muscle) tissues collected from ad lib fed and mice undergoing STS or FMD regimen were analyzed with qRT-PCR. (B) HO-1 levels in 4T1 cells following 48 hours of in vitro STS were measured by Western blotting (n=3) (Blot was captured with Bio-Rad ChemiDoc, and unedited, representative bands are shown. (C, E, G) Viability of 4T1 cells was determined by MTT reduction following (C) DXR and hemin (10 M), (E) CP and hemin (10 M), and (G) CP and ZnPP (20 M). (D, F) Viability of 4T1 cell stably over-expressing HO-1 (pHO-1) or empty vector (pEV) was determined by MTT following (D) DXR and (F) CP under control and STS conditions. Data represented as mean ± SEM. The significance of the differences between experimental groups was determined by using one-way ANOVA (Tukey post-analysis test). Comparisons between groups were performed with Student's t test. p-values <0.05, 0.01 and 0.001 are indicated as *, **, and ***, respectively. See also Figure S5.

Figure 5. Down-regulation of HO-1 in the tumor during FMD/STS is required for the targeted attack of cancer cells

(A-C) 4T1 tumor-bearing mice were treated with (A) ZnPP (40 mg/kg/day; IP; n=7), (B) STS and hemin (30 mg/kg/day; IP; n=7), and (C) FMD and hemin and DXR. (D) Mice bearing 4T1 tumors stably over-expressing HO-1 (pHO-1), or empty vector (pEV) were treated with DXR under ad lib or FMD regimens (n=10-15) (see Figure 1C). Data represented as mean ± SEM. One-way ANOVA (Tukey post-analysis test) was performed. p-values <0.05, 0.01 and 0.001 are indicated as *, **, and ***, respectively. See also Figure S6.

Figure 6. Up-regulation or overexpression of HO-1 lead to decreased infiltration of TILs and accumulation of Tregs in the tumor bed

Mice bearing 4T1 or 4T1-pHO-1 breast tumors under FMD were treated with DXR and either hemin (30 mg/kg/day; IP) or CD8+ specific neutralizing monoclonal antibodies (αCD8; clone YTS 169.4) (n=13-15) (see Figures 3F and 5D). (A) Tumor volumes were measured at multiple time-points (left) and immediately prior to euthanasia (right). (B-D) CD3+/CD8+ (CTL) and CD3+/CD4+/CD25+ (Treg)lymphocytes from the tumor bed was analyzed by FACS. Data represented as mean ± SEM. One-way ANOVA (Tukey post-analysis test) was performed. p-values <0.05, 0.01 and 0.001 are indicated as *, **, and ***, respectively.

Figure 7. A model of FMD-dependent DSS in tumors

Two scenarios describing the rearrangement of tumor and immune system under different dietary interventions are shown. FMD+DXR increases the immunogenicity of tumors in an HO-1-dependent manner, and increases the recruitment of CD3+/CD8+ cytotoxic lymphocytes to the tumor bed (TILs), leading to the targeted attack of cancer cells. Such increase of CD3+/CD8+ TILs is accompanied by the reduction of CD3+/CD4+/CD25+ Tregs in the tumor bed. The latter scenario is also associated with an increase of CLP in the bone marrow and of circulating CD3+/CD8+ cytotoxic lymphocytes. Dendritic cells (DC) can recognize DXR-induced apoptotic tumor cells stimulate circulating CD3+/CD8+ CTLs, which, in our model, can be augmented by FMD cycles.