P-glycoprotein acts as an immunomodulator during neuroinflammation

Abstract

Background: Multiple sclerosis is an inflammatory demyelinating disease of the central nervous system in which autoreactive myelin-specific T cells cause extensive tissue damage, resulting in neurological deficits. In the disease process, T cells are primed in the periphery by antigen presenting dendritic cells (DCs). DCs are considered to be crucial regulators of specific immune responses and molecules or proteins that regulate DC function are therefore under extensive investigation. We here investigated the potential immunomodulatory capacity of the ATP binding cassette transporter P-glycoprotein (P-gp). P-gp generally drives cellular efflux of a variety of compounds and is thought to be involved in excretion of inflammatory agents from immune cells, like DCs. So far, the immunomodulatory role of these ABC transporters is unknown.

Methods and findings: Here we demonstrate that P-gp acts as a key modulator of adaptive immunity during an in vivo model for neuroinflammation. The function of the DC is severely impaired in P-gp knockout mice (Mdr1a/1b-/-), since both DC maturation and T cell stimulatory capacity is significantly decreased. Consequently, Mdr1a/1b -/- mice develop decreased clinical signs of experimental autoimmune encephalomyelitis (EAE), an animal model for multiple sclerosis. Reduced clinical signs coincided with impaired T cell responses and T cell-specific brain inflammation. We here describe the underlying molecular mechanism and demonstrate that P-gp is crucial for the secretion of pro-inflammatory cytokines such as TNF-alpha and IFN-gamma. Importantly, the defect in DC function can be restored by exogenous addition of these cytokines.

Conclusions: Our data demonstrate that P-gp downmodulates DC function through the regulation of pro-inflammatory cytokine secretion, resulting in an impaired immune response. Taken together, our work highlights a new physiological role for P-gp as an immunomodulatory molecule and reveals a possible new target for immunotherapy.

Figure 1. Reduced clinical signs in Mdr1a/1b−/− mice during acute and progressive phase of EAE.

(A) Clinical signs of EAE induced by immunization of wild-type (WT) and Mdr1a/1b−/− mice with rMOG (1–125) showing mean clinical scores (+/− SEM) of two independent experiments (*p<0.002 Mann-Whitney, n = 23 mice per group). (B) Mean total EAE score per WT or Mdr1a/1b −/− mouse. *p<0.05.

Figure 2. Decreased brain inflammation in Mdr1a/1b−/− EAE lesions.

Brains were isolated from EAE mice on day 15 (A–D) and day 29 (E–H) after immunization and the cerebellum white matter was analyzed for the infiltration of CD45⁺ mac1⁺ macrophages (green; A,B,E,F) or CD45⁺ CD3⁺ T cells (green; C,D,G,H) and laminin positive (red) basement membranes around vessels in WT (left panel) or Mdr1a/1b−/− (right panel) mice. Images represent representative tissues from 4 mice per group. Magnification 200×.

Figure 3. Impaired Th1 and Th2 response in Mdr1a/1b−/− mice during EAE.

T cell proliferation in lymph node cells isolated from wild-type (WT) and Mdr1a/1b−/− mice was assessed upon re-stimulation with different concentrations of rMOG (1–125) on day 15 (A) or day 29 (B) of EAE. Supernatants from T cell proliferation assays upon re-stimulation with 10 µg/ml rMOG (1–125) were harvested and the production of the Th1 cytokines IFN-γ (C,D) or TNF-α (E,F) and the Th2 cytokines IL-10 (G,H) or IL-5 (I,J) was measured on different time points (48, 72 and 96 hr). Experiments were performed in triplicate using 4 mice per group and were presented as the mean +/− SEM. *p<0.05, **p<0.01.

Figure 4. P-gp does not affect polyclonal T cell activation.

Lymph node cells isolated from WT and Mdr1a/1b−/− mice were stimulated with anti-CD3/anti-CD28 for 5 hr and subsequently stained for T cell activation markers CD69 (A) or CD44 (B) on live cell gated CD4⁺ or CD8⁺ T cells. After anti-CD3/anti-CD28 stimulation, cytokine production of IFN-γ, TNF-α or IL-10 was measured (C) and IFN-γ and TNF-α transcripts were detected by RT-PCR and presented as relative expression compared to GAPDH (D). Intracellular IFN-γ production in CD4⁺ (E) or CD8⁺ T cells (F) was determined on permeabilized lymph node cells after anti-CD3/anti-CD28 stimulation. Experiments were performed in triplicate using 5 mice per group and were presented as the mean +/− SEM. *p<0.05, **p<0.01.

Figure 5. P-gp is necessary for CD4⁺ and CD8⁺ T cell proliferation.

Bone marrow derived DCs (BMDCs) were coated with MHC class I OVA_257–264 peptide or MHC class II OVA_323–339 peptide and were used as stimulators for T cell proliferation. For this, titrated BMDCs were cultured with purified OT-I (A) or OT-II (B) in the presence or absence of the P-gp inhibitor reversin 121 (10 µM) or its vehicle DMSO (0.1%). Cytokine production of IFN-γ (C,D) or TNF-α (E,F) was measured after 48 hr of co-culture. OVA peptide coated BMDCs were fixated with 0.1% PFA for 10 minutes after which OT-I (G) or OT-II (H) proliferation was measured together with IFN-γ (I,J) or TNF-α (K,L) production after 48 hr of co-culture. Representative results are depicted as the mean +/− SEM from three independent experiments performed in triplo. *p<0.05, **p<0.01, ***p<0.001.

Figure 6. P-gp mediates DC maturation via IFN-γ and TNF-α secretion.

(A) DC maturation of BMDCs isolated from WT (shaded histogram) or Mdr1a/1b−/− (open histogram) mice were determined by the expression of CD40, CD80, CD86 or MHCII on live cell gated and CD11c+ cells in the presence or absence of LPS. Quantification of DC maturation of CD11c+ cells positive for CD40 (B), CD80 (C), CD86 (D) and MHCII (E). Recombinant IFN-γ or TNF-α (5 ng/ml) was added to BMDCs derived from C57Bl/6 mice cultured in the presence or absence of LPS, the P-gp inhibitor reversin 121 (10 µM) or its vehicle DMSO (0.1%). Experiments were performed in triplicate using 5 mice per group (A–E) or three independent experiments (F–I) and were presented as the mean +/− SEM. *p<0.05.