Ex vivo recovery and activation of dysfunctional, anergic, monocyte-derived dendritic cells from patients with operable breast cancer: critical role of IFN-alpha

Abstract

Background: Dendritic cells (DCs) play a crucial role in initiating effective cell-mediated immune responses, but are dysfunctional and anergic in breast cancer. Reversal of this dysfunction and establishment of optimal DC function is a key prerequisite for the induction of effective anti-cancer immune responses.

Results: Peripheral blood DCs (PBDCs) and lymph node DCs (LNDCs) generated in vitro from adherent cultures of peripheral blood monocytes (PBMs) and lymph node monocytes (LNMs), respectively, using the 4 cytokine conditioned medium (CCM) (GM-CSF+IL-4+TNF-alpha+IFN-alpha) or 3 CCM (GM-CSF+IL-4+TNF-alpha) demonstrated a significantly higher degree of recovery and functional capacity in a mixed lymphocyte DC reaction (MLDCR, p < 0.001), expressed significantly higher levels of HLA-DR, CD86, compared with 2 CCM (GM-CSF+IL-4) or medium alone generated DCs from PBMs and LNMs (p < 0.001). The PBDCs generated with 3 CCM or 4 CCM showed a significantly (p < 0.001) enhanced macropinocytotic capability (dextran particles) and induced increased production and secretion of interleukin-12p40 (IL-12p40) in vitro (p < 0.001), compared with PBDCs generated from monocytes using 2 CCM or medium alone. Lipopolysaccharide (LPS) stimulation of PBDCs generated with 4 CCM demonstrated enhanced secretion of IL-6 but not IL-12p70, compared with control DCs unstimulated with LPS (p < 0.001).

Conclusion: Dysfunctional and anergic PBDCs and LNDCs from patients with operable breast cancer can be optimally reversed by ex vivo culturing of precursor adherent monocytes using a 4 CCM containing IFN-alpha. Maximal immunophenotypic recovery and functional reactivation of DCs is seen in the presence of IFN-alpha. However, 4 CCM containing IFN-alpha generated-PBDCs, do not produce and secrete IL-12p70 in vitro.

Figure 1

(A) Percentage of PBDCs (from patients with operable breast cancer) expressing HLA-DR, CD 86 and CD 40: Statistically significant increase in expression of HLA-DR was observed in the 3 and 4 CCM (cytokine conditioned medium) compared with 2 CCM generated DCs and medium only generated DCs (p < 0.001, ANOVA). Similarly, significant increase in expression of CD86 was observed in the 3 or 4 CCM compared with 2 CCM (cytokine conditioned medium) generated DCs or medium only generated DCs (p < 0.001, ANOVA). However, CD40 expression was increased (p < 0.01, ANOVA) in the 3 or 4 CCM generated DCs compared with medium only DCs. (B) Percentage of LNDCs (from patients with operable breast cancer) expressing HLA-DR, CD 86 and CD 40: Statistically significant increase in expression of HLA-DR, CD 86 and CD40 on LNDCs cultured with 3 and 4 CCM compared with medium only generated DCs from patients with breast cancer (p < 0.01, ANOVA). (C) Percentage of PBDCs (healthy donor) expressing HLA-DR, CD 86 and CD 40: Significant increase in expression of HLA-DR was observed in the 3 and 4 CCM (cytokine conditioned medium) compared with 2 CCM generated DCs and medium only generated DCs (p < 0.001, ANOVA). However, significant (p < 0.01, ANOVA) increase in expression of CD86 was observed in the 3 or 4 CCM (cytokine conditioned medium) compared with medium only generated DCs. Similarly, CD40 expression was significantly increased in 3 or 4 CCM generated DCs compared with medium only generated DCs. Results shown in 1 A, B and C are represented as mean ± standard deviation (error bars) are from ten different patients or healthy donors. Lineage-negative cells were selectively gated and analyzed in all conditions.

Figure 2

Phenotypic profile of PBDCS (A) and LNDCs (B) generated with medium alone (no cytokine), 2, 3 and 4 CCM: Representative maximal fluorescent intensity from flow cytometry of the increasing expression of HLA-DR, CD 40 and CD 86 in different in vitro conditions. (Lineage negative cells were gated and analyzed only)

Figure 3

Phagocytosis (dextran uptake) for DCs generated in the presence of medium only (A), 2 CCM (B), 3 CCM (C) and 4 CCM (D): Representative maximal fluorescent intensity of mannose receptors in putative Ag uptake by incubation with Cascade Blue-conjugated dextran in different in vitro conditions (at 4°C and 37°C). Lineage negative cells were gated and analyzed only. 0

Figure 4

Box plot showing mannose receptors in putative antigen uptake by incubation with Cascade Blue – conjugated dextran in different in vitro conditions of PBDCs generated from 10 different patients: Percentage of uptake by cells of Cascade Blue conjugated-dextran were calculated from Lin-cells (Lin-cells were gated and analyzed). PBDCs grown in 4 CCM (GM-CSF+IL-4+TNF-α+IFN-α) or 3CCM (GM-CSF+IL-4+TNF-α) compared with medium only generated PBDCs or 2 CCM (GM-CSF+IL-4), both demonstrated significantly higher levels of dextran uptake (*** p < 0.001, ANOVA). Horizontal bars represent mean value for each category. Data shown are median (line), mean (dot), interquartile range (box) and maximum and minimum values (whiskers).

Figure 5

(A): Differential production of IL-10p40 and IL-12: Activation/maturation of PBDCs by the 3 and 4 CCM (cytokine conditioned media) leads to significantly increased production and secretion of the pro-inflammatory cytokine IL-12 p40, compared with medium only (*** p < 0.001, ANOVA) and 2 CCM generated DCs (***p < 0.001, ANOVA). In contrast, 3 and 4 CCM generated DCs produced and secreted significantly less inhibitory cytokine IL-10 compared with 2 CCM generated PBDCs (***p < 0.001, ANOVA). Cell free supernatants were obtained on day 7 of culture. Cytokine concentrations were determined by ELISA. The results shown are from ten different patients and are represented as mean ± standard deviation. Error bars represent standard deviations. (B): Production of IL-12p40 by PBDCs from healthy donors (controls): The amounts of cytokines in the samples were measured using ELISA. The levels of IL-10 and IL-12 p40 from PBDCs generated from monocytes cultured using all 4 cytokines (4 CCM) from 10 normal healthy donors were evaluated. The mean level of IL-12 p40 was 72.95 ± 4.47, which was significantly (p < 0.01, Student t-test) lower compared with 4 CCM generated PBDCs from breast cancer patients (Mean = 262 ± 15). However, IL-10 was not detectable in the supernatants and hence not shown in this Figure. (C): Cytokine production by PBDCs generated with 4 CCM from patients with operable breast cancer following stimulation with LPS (1 μg/300μl): IL-12p70 was undetectable before and after LPS stimulation and hence not shown in this Figure. IL-6 production and secretion was significantly increased after LPS stimulation (*** p < 0.001, Student t-test). NS: Not significant.

Figure 6

Increased blastogenesis in the MLDCR: The stimulatory capacity of DCs.(PB:peripheral blood, LN:lymph node) was estimated from the incorporation of ³H-thymidine (shown as cpm). During the last 18 h on the 5^th day of in vitro culture containing T cells (5 × 10⁴ cells/well) from a normal volunteer and irradiated DCs (2.5 × 10³ cells/well) from patients with operable breast cancer (PBDCs and LNDCs). DC:T cell ratio was 1:20. LNM and PBM-derived DCs with 3 and 4 CCM milieu showed a significantly increased allogeneic stimulatory capacity compared with 2 CCM (***p < 0.001, ANOVA) and medium only generated DCs (***p < 0.001, ANOVA). Results shown (mean ± standard deviation) are from ten different patients. Error bars represent standard deviations.