Broccoli-Derived Nanoparticle Inhibits Mouse Colitis by Activating Dendritic Cell AMP-Activated Protein Kinase

Abstract

The intestinal immune system is continuously exposed to massive amounts of nanoparticles derived from food. Whether nanoparticles from plants we eat daily have a role in maintaining intestinal immune homeostasis is poorly defined. Here, we present evidence supporting our hypothesis that edible nanoparticles regulate intestinal immune homeostasis by targeting dendritic cells (DCs). Using three mouse colitis models, our data show that orally given nanoparticles isolated from broccoli extracts protect mice against colitis. Broccoli-derived nanoparticle (BDN)-mediated activation of adenosine monophosphate-activated protein kinase (AMPK) in DCs plays a role in not only prevention of DC activation but also induction of tolerant DCs. Adoptively transferring DCs pre-pulsed with total BDN lipids, but not sulforaphane (SFN)-depleted BDN lipids, prevented DSS-induced colitis in C57BL/6 (B6) mice, supporting the role of BDN SFN in the induction of DC tolerance. Adoptively transferring AMPK^+/+, but not AMPK^-/-, DCs pre-pulsed with SFN prevented DSS-induced colitis in B6 mice, further supporting the DC AMPK role in SFN-mediated prevention of DSS-induced colitis. This finding could open new preventive or therapeutic avenues to address intestinal-related inflammatory diseases via activating AMPK.

Keywords: AMPK; broccoli nanoparticles; colitis; edible plant and mammalia; gut immune homeostasis; sulforaphane; tolerogenic DCs; transkingdom interaction.

Figure 1

BDN Administration Protects from Intestinal Inflammation (A–D) BDN administration protects mice from dextran sulfate sodium (DSS)-induced colitis. C57BL/6 mice were given BDNs orally (250 μg/mouse in 200 μL PBS) before (every day for 10 days) and after (every 2 days for 12 days) administration of water containing DSS (2.5% DSS). (A) Weight loss in animals following the induction of colitis, measured as a reduction from initial weight until day of sacrifice. (B) Representative histology staining of colon and histological score. (C) Alcian blue staining of colon. (D) Colon length. (E–H) BDN administration protects mice from T cell-mediated colitis. C57BL/6 mice were given BDNs orally (250 μg/mouse in PBS), twice every week after adoptive transfer of naive CD4⁺ T cells isolated from C57BL/6 mice (0.5 × 10⁶, injected intraperitoneally [i.p.]). (E) Ratio of colon weight to length. (F) Colon thickness. (G) Histological scoring was performed on colon sections. (H) Representative FACS plots of intracellular staining of indicated cytokines in CD3⁺CD4⁺ T cells in colonic LPL, MLN, and spleen. Data are mean ± SEM (n = 7). *p < 0.05, **p < 0.01 (Student’s t test).

Figure 2

BDNs Block the Differentiation of the Inflammatory Gr1⁺ Monocytes into CD11b⁺ DCs during Experimental Colitis (A–D) The frequency (A and C) and cell number (B and D) of CD11b⁺ DCs isolated from colonic lamina propria (cLP) and MLN in DSS-induced colitis (A and B) or naive CD4 T cell-mediated colitis (C and D). (E) Phenotypic characterization of CD11b⁺ DCs from colon in CD4 T cell-mediated colitis. (F) Real-time PCR for the expression of genes in CD11b⁺ DCs sorted from pooled colonic LPL in CD4 T cell-mediated colitis. (G) 3 × 10⁶ sorted Gr1⁺ CD11b⁺ckit⁻CD11c⁻CD115⁺ monocytes from BM of naive C57BL/6 (B6) CD45.1 mice were injected i.v. into colitic B6 CD45.2 Rag1^−/− mice receiving PBS or BDNs as described in Materials and Methods. After 72 hr, the colon and MLN were harvested. Cells were gated on the CD45.1 population, and CD11b⁺CD11c⁺MHCII⁺ cells were examined. Staining of CD11b⁺ DCs gated on CD45.1⁺ cells from colitic mice that received Gr1⁺ CD115⁺ monocytes. Data are mean ± SEM (n = 5). *p < 0.05, **p < 0.01 (Student’s t test).

Figure 3

BDNs Inhibit the Anti-CD40 Antibody-Induced Colitis Rag1^−/− mice were given BDNs orally (250 μg/mouse in PBS) every day for 10 days before injection with anti-CD40 (200 μg) or with rat immunoglobulin G2a (IgG2a) control. (A) Weight loss as a percentage of the initial weight. (B) Representative immunohistochemistry colon sections and histological score from Rag^−/− mice (n = 5) at day 7 after injection with anti-CD40. (C) DAI score. (D) The frequency of CD11b⁺ DCs isolated from colon and MLN in anti-CD40-induced colitis. (E) Colon samples were stained with antibodies directed against F4/80 or CD11c and counterstained with hematoxylin. Graphic representation of the number of macrophages and DCs was per 1-mm² high power field in sections depicted in the left panel.

Figure 4

BND-Derived Lipid Endows CD11b⁺ DCs with Tolerogenic Potential (A and B) DMSO or BDN lipid-pretreated BMDCs were stimulated with LPS (100 ng/mL) for 24 hr. FACS analysis of co-stimulatory molecules in BMDCs (A) and ELISA analysis of cytokines (B) were examined. (C and D) CFSE-labeled OT-II T cells were co-cultured with LP-DCs and OVA peptide (5 μg/mL) in the presence of DMSO or BDN lipid. Proliferation of CFSE-labeled CD4⁺ T cells (C) and the level of IFN-γ and TNF-α (D) in the supernatant of co-cultured cells were determined. (E–G) DCs were pulsed with DMSO or BDN lipid for 24 hr, washed, and incubated with splenic CD4⁺ T cells in the presence of anti-CD3 (5 μg/mL) and anti-CD28 (5 μg/mL) for 4 days. ELISA of cytokine production in co-culture supernatants (E), RT-PCR analysis of the expression of genes in CD4⁺ T cells (F), and FACS analysis of T cell activation molecules (G) were determined after 4 days in culture. Data are mean ± SEM (n = 5). *p < 0.05, **p < 0.01 (Student’s t test).

Figure 5

BDN Lipid-Mediated Activation of DC AMPK Plays a Role in Preventing the Induction of Inflammatory DC Cytokines and Protection of Mice against Mouse Colitis (A–C) Mice receiving PBS or BDNs were exposed to PBS or 2.5% DSS. (A) Colonic lysates were prepared and analyzed for expression and phosphorylation of the indicated proteins by western blotting. (B) Immunohistochemistry of colon was analyzed by anti-pS6^ser235/236 or p70S6K antibody. (C) DMSO or BDN lipid-treated BMDCs from B6 mice were stimulated by LPS for the indicated time. Cell lysates were analyzed by immunoblotting with the indicated antibodies. (D) DMSO or BDN lipid-treated BMDCs from WT or AMPKα1^−/− mice were stimulated by LPS. The level of IL-12 and TNF-α in the supernatant of BMDCs was determined. (E) ELISA analysis of DC-T cell co-cultures similar to Figure 4E, but also including AMPKα1^−/− DCs. (F–I) Wild-type or AMPKα1^−/− BMDCs were treated with CBA (20 μg/mL) in the presence of DMSO or BDN lipid for 24–48 hr. 2 × 10⁶ cells were injected i.v. into each recipient (C57BL/6) mouse at day −1 and day +3 of 2.5% DSS administration, with PBS as a control for BMDCs transfer. (F) Body weight was measured at the indicated time points after transfer of BMDCs. (G) Histological scores from DC recipient animals with DSS-induced colitis. (H) Colon length from colitic mice. (I) Cytokine analysis by qRT-PCR of RNA recovered from whole colon. Data are mean ± SEM (n = 5). *p < 0.05, **p < 0.01 (Student’s t test).

Figure 6

SFN Induces Regulatory DCs via AMPK-mTOR Signaling (A and B) Mouse bone marrow-derived DCs differentiated with GM-CSF and IL-4 in the presence DMSO or SFN (10 μM) and were activated with PBS (A) or LPS (B) for 18 hr. FACS analysis of surface markers on BMDCs (A) and ELISA analysis of cytokines in the supernatant (B) were determined. Data are mean ± SEM (n = 5). (C and D) DMSO or SFN-derived BMDCs with or without LPS stimulation were incubated with splenic CD4⁺ T cells (DC/T cell ratio, 1:5) in the presence of anti-CD3 (5 μg/mL). ELISA of IFN-γ production of the supernatant (C) and FACS analysis of surface markers on CD4⁺ T cells (D) were determined. Data are mean ± SEM (n = 5). (E) Human DCs were derived with DMSO or SFN (10 μM) in the presence of GM-CSF and IL-4 and activated with LPS. Supernatants were collected after 18 hr and analyzed for IL-23 and IL-10 secretion by ELISA. (F) ELISA of IFN-γ production of the supernatants from allogeneic CD4⁺ T cells cultured for 5 days with human DCs derived from DMSO or SFN (DC/T cell ratio, 1:5). Data are mean ± SEM (n = 3). (G) SFN or DMSO pretreated human DCs were stimulated with 100 ng/mL LPS for the time points indicated. Cell lysates were analyzed for the phosphorylation of the indicated proteins by western blotting. **p < 0.01 (Student’s t test).

Figure 7

BND SFN Protects Colitis by Inducing Regulatory DCs (A) The lipids extracted from BDN-derived liposome-like nanoparticles (LNs) or LNs with SFN knockout (LN-SFN^−/−) or knockin (LN-SFN^+/+) and a standard SFN were separated on a thin-layer chromatography plate and developed. A representative image was scanned using an Odyssey scanner. (B–D) BMDCs were treated with CBA (20 μg/mL) for 24–48 hr in the presence of DMSO, LN-SFN^−/−, or LN-SFN^+/+ and were injected i.v. (2 × 10⁶) into each recipient (C57BL/6) mouse at day −1 and day +3 of 2.5% DSS administration. (B) Body weight was measured at the indicated time points after transfer of BMDCs. (C) Histological scores from DC recipient animals with DSS-induced colitis. (D) Colon length from colitic mice. (E) Cytokine analysis by qRT-PCR of RNA recovered from colonic LPL. (F) The cell number of CD11b⁺ DCs isolated from colon and MLN in DSS-induced colitis after adoptive transfer. Data are mean ± SEM (n = 5). *p < 0.05, **p < 0.01 (Student’s t test).