Employing Macrophage-Derived Microvesicle for Kidney-Targeted Delivery of Dexamethasone: An Efficient Therapeutic Strategy against Renal Inflammation and Fibrosis

Abstract

Although glucocorticoids are the mainstays in the treatment of renal diseases for decades, the dose dependent side effects have largely restricted their clinical use. Microvesicles (MVs) are small lipid-based membrane-bound particles generated by virtually all cells. Here we show that RAW 264.7 macrophage cell-derived MVs can be used as vectors to deliver dexamethasone (named as MV-DEX) targeting the inflamed kidney efficiently. Methods: RAW macrophages were incubated with dexamethasone and then MV-DEX was isolated from the supernatants by centrifugation method. Nanoparticle tracking analysis, transmission electron microscopy, western blot and high-performance liquid chromatography were used to analyze the properties of MV-DEX. The LC-MS/MS was applied to investigate the protein compositions of MV-DEX. Based on the murine models of LPS- or Adriamycin (ADR)-induced nephropathy or in-vitro culture of glomerular endothelial cells, the inflammation-targeting characteristics and the therapeutic efficacy of MV-DEX was examined. Finally, we assessed the side effects of chronic glucocorticoid therapy in MV-DEX-treated mice. Results: Proteomic analysis revealed distinct integrin expression patterns on the MV-DEX surface, in which the integrin α_Lβ₂ (LFA-1) and α₄β₁ (VAL-4) enabled them to adhere to the inflamed kidney. Compared to free DEX treatment, equimolar doses of MV-DEX significantly attenuated renal injury with an enhanced therapeutic efficacy against renal inflammation and fibrosis in murine models of LPS- or ADR-induced nephropathy. In vitro, MV-DEX with about one-fifth of the doses of free DEX achieved significant anti-inflammatory efficacy by inhibiting NF-κB activity. Mechanistically, MV-DEX could package and deliver glucocorticoid receptors to renal cells, thereby, increasing cellular levels of the receptor and improving cell sensitivity to glucocorticoids. Notably, delivering DEX in MVs significantly reduced the side effects of chronic glucocorticoid therapy (e.g., hyperglycemia, suppression of HPA axis). Conclusion: In summary, macrophage-derived MVs efficiently deliver DEX into the inflamed kidney and exhibit a superior capacity to suppress renal inflammation and fibrosis without apparent glucocorticoid adverse effects. Our findings demonstrate the effectiveness and security of a novel drug delivery strategy with promising clinical applications.

Keywords: dexamethasone; drug delivery; macrophage-derived microvesicles; renal fibrosis; renal inflammation.

Figure 1

Schematic illustration of the preparation of MV-DEX and the therapeutic effects of MV-DEX against renal diseases. RAW macrophages were treated with DEX and then supernatants were collected for MV isolation by centrifugation. MV-DEX treatment improved renal function, alleviated renal inflammation and fibrosis, and reduced side effects to a greater magnitude than free DEX. Mechanistically, MV-DEX efficiently targeted inflamed kidney through LFA-1 and VLA-4 on the surface. Further, MV-DEX increased the cellular levels of the GR and facilitated the entry of DEX into the cells.

Figure 2

Characterization of RAW 264.7 macrophages-derived DEX-packaging MVs. (A) Size distribution analysis of MV-DEX by nanoparticle tracking analysis. (B) Transmission electron microscopy images of MV-DEX showed double-membrane vesicles. Boxed area is enlarged in the right. (C) Markers of macrophage (CD68, F4/80 and CD206) and microvesicles (actinin 4) were confirmed in MV-DEX by Western blot. (D) HPLC analysis revealed the concentrations of DEX in MVs were dependent on the applied drug doses. (E) RAW cells were incubated with 30 μmol/L fluorescein-DEX. MVs were isolated and observed under confocal microscopy. DEX was shown as the green fluorescent (E1) (scale bar, 1 μm). GECs were incubated with fluorescein-DEX packaging MVs for 12 h and then observed under confocal microscopy (E2) (scale bar, 10 μm). n=3. Data are presented as mean ± SD.

Figure 3

DEX-packaging MVs efficiently target inflamed kidney. (A, B) GECs took up MV-DEX. PKH 26 or PKH 67-labeled MVs (5×10⁷) were incubated with GECs for 3 or 6 h, respectively, and then PKH 26-positive (red) cells were observed under confocal microscope (A) (scale bar, 10 μm), while PKH 67-positive cells were analyzed by flow cytometry (B). Compared with control cells, LPS-treated cells took up more MV-DEX. (C, D) For in vivo distribution, mice were injected intravenously with DID-labeled MVs (1×10¹⁰). Imaging of the kidneys for detection of DID-labeled MVs 24 h after injection (C). Representative micrographs and quantification of DID-labeled MVs in kidney sections (D) (scale bar, 25 μm). MV-DEX showed a great increase in accumulation into inflamed kidney tissue at 24h after injection. Data are presented as mean ± SD, * p<0.05, ** p<0.01, *** p<0.001, two-tailed t-test (A, B), one-way ANOVA (C, D). (E, F) Positive correlation between the renal average radiance of DID and renal IL-6 (E) or TNF (F) in LPS-treated mice. n=10 kidneys. The data were compared using the Spearmen correlation coefficient.

Figure 4

LFA-1 and VAL-4 guide the homing of MV-DEX to inflamed kidney. (A) Uptake of MV-DEX in LPS-induced GECs. Cellular uptake was determined by flow cytometry after 6 h incubation with DID-labeled MVs (5×10⁷) pre-treated with or without proteinase K. Loss of surface proteins on MV-DEX significantly reduced their entry into inflamed GECs. (B) Western blot analysis showed ITGα_L, ITGβ₂, ITGα₄ and ITGβ₁ were robustly expressed on MV-DEX. (C) Western blot analysis showed VCAM-1 and ICAM-1 were significantly increased in LPS-treated GECs or kidneys from LPS-treated or ADR-treated mice. (D) Effect of LFA-1 and VLA-4 blocking on cellular uptake of MV-DEX in LPS-induced GECs. Cellular uptake was determined by flow cytometry after 6 h incubation with DID-labeled MVs (5×10⁷) pre-incubated with ITGβ₂ blocking antibody or VLA-4 blocking antibody. The uptake of MV-DEX was reduced 70% by inhibition of ITGβ₂ and VLA-4. (E, F) Effect of LFA-1 and VLA-4 blocking on biodistribution of MV-DEX in LPS-treated mice. Mice were injected intravenously with DID-labeled MVs (1×10¹⁰) pre-incubated with ITGβ₂ blocking antibody and VLA-4 blocking antibody. Imaging of the kidneys for detection of DID-labeled MVs 24 h after injection (E). Representative micrographs and quantification of DID-labeled MVs in kidney sections (F) (scale bar, 25 μm). MV-DEX showed a great decrease in accumulation into kidney tissue. Data are presented as mean ± SD, * p<0.05, ** p<0.01, *** p<0.001, # p<0.05, two-tailed t-test (A, C, E, F), one-way ANOVA (D).

Figure 5

Enhanced anti-inflammatory efficacy of MV-DEX in vitro. GECs were stimulated with LPS and treated with the indicated concentrations of MV-DEX or free DEX for 12 h. (A) Effects of MV-DEX or free DEX on NF-κB luciferase reporter activity in LPS-induced GECs. Compared with free DEX, MV-DEX showed a 5-fold increase in inhibiting NF-κB activity. (B, C) Real-time PCR analysis of inflammatory cytokines (TNF, IL-6, IL-β and CCL-2) in GECs. TNF (D) and IL-6 (E) levels in the supernatants were detected by ELISA. (F) The expression of p65 and p-p65 were analyzed by western blot. (G) Immunofluorescent staining against p-p65 was observed under confocal microscope (scale bar, 10 μm). Both MV-DEX and free DEX exhibited anti-inflammatory efficacy, but MV-DEX was much more superior. (H-J) MV-DEX and free DEX were stored at 37 ℃ for 10 days to destroy most MVs, whereas the drug concentration between two groups remained the same (with 5 μmol/L DEX). NF-κB luciferase reporter activity was shown as fold change to renilla luciferase activity (H). ELISA detection of TNF (I) and IL-6 (J) levels in supernatants. Following MV destruction, the MV-DEX group no longer showed a superior inhibition of inflammation compared with free DEX. Data are presented as mean ± SD, ** p<0.01, *** p<0.001 vs. LPS-treated group, # p<0.05, ## p<0.01, ### p<0.001, N.S., not significant, one-way ANOVA.

Figure 6

Enhanced renoprotective and anti-inflammatory efficacy of MV-DEX in LPS-treated mice. (A) Schematic diagram of the experimental design. In brief, mice were concurrently treated with MV-DEX or free DEX or vehicle after LPS treatment, and were sacrificed 48 h after disease induction. (B) Survival curve reveled enhanced survival in MV-DEX-treated mice with 45% survival compared to 30% in the free DEX-treated mice. n=8 mice (CTRL); n=20 mice (DEX, MV-DEX). (C) Albuminuria (urine albumin-to-creatinine ratio) and serum creatinine (D) were reduced to a much greater extent after MV-DEX treatment than with free DEX treatment. n=8 mice per group. ELISA detection of TNF (E) and IL-6 (F) levels in mice's serum. Real-time PCR analysis of cytokine (TNF, IL-6, IL-β and CCL-2) mRNA expression levels (G) and Western blot analysis of p65 and p-p65 (H) in renal cortex tissue lysates. (I) Immunostaining and quantification of macrophages and neutrophils in the tubulointerstitium (scale bar, 25 μm). MV-DEX had a more potent ability to alleviate renal inflammation. n=6 mice per group. Data are presented as mean ± SD, *** p<0.001 vs. LPS-treated mice, # p<0.05, ## p<0.01, ### p<0.001, one-way ANOVA.

Figure 7

Enhanced renoprotective and anti-inflammatory efficacy of MV-DEX in ADR-treated mice. (A) Schematic diagram of the experimental design. In brief, 7 days after injection of ADR, mice were treated with free DEX or MV-DEX (0.5 mg/kg every 48 hours) until 3 weeks post-ADR treatment. (B) Albuminuria (urine albumin-to-creatinine ratio) and (C) serum creatinine were significantly reduced in MV-DEX-treated mice compared to free DEX-treated mice. (D) Representative images of PAS staining. Top, glomerular morphology; Bottom, tubulointerstitial morphology. (E) Representative images of Masson trichrome staining. Top, glomerulosclerosis; Bottom, tubulointerstitial fibrosis. scale bars, 25 μm. n=8 mice per group. (F) Electron microscopy was performed to assess ultrastructural changes in podocyte morphology (scale bar, 200 nm). n=4 mice per group. All the histological analysis showed a better renoprotection effect of MV-DEX. (G, H) ELISA detection of TNF and IL-6 in mice's serum. Real-time PCR analysis of inflammatory cytokine mRNA expression levels (I) and western blot analysis of p65 and p-p65 (J) in renal cortex tissue lysates. (K) Immunostaining and quantification of macrophages and T cells in the glomeruli (scale bar, 25 μm). Compared to free DEX treatment, MV-DEX showed more potent in reducing renal inflammation in ADR-induced nephropathy. n=6 mice per group. Data are presented as mean ± SD, * p<0.05, ** p<0.01, *** p<0.001 vs. ADR-treated mice, # p<0.05, ## p<0.01, ### p<0.001, one-way ANOVA.

Figure 8

Glucocorticoid receptor (GR) in DEX-packaging MVs facilitate the anti-inflammatory efficacy. (A) GR in MV-DEX was detected by Western blot. n=3. (B) RAW cells were transfected with RFP-tagged GR (B1) to trace the location of the receptor. MVs were isolated and RFP-tagged GR (red fluorescent) was observed in MVs (B2) (scale bar, 1 μm). GECs were incubated with GR-RFP packaging MVs for 12 h and then observed under confocal microscopy (B3) (scale bar, 10 μm). (C, D) GECs were transfected with siRNA or treated with RU486 to downregulate GR expression. NF-κB luciferase reporter activity revealed MV-DEX significantly reduced NF-κB activity, while the ability of free DEX was abrogated. n=4. (E-I) After intraperitoneal injection of RU486 for 7 days, LPS-induced nephropathy model and treatment protocol were built as previously described. (E) Survival curve showed mice received MV-DEX treatment had lower mortality compared to mice received free DEX treatment. n=8 mice (CTRL); n=20 mice (RU486+DEX, RU486+MV-DEX). Serum creatinine (F), TNF (G) and IL-6 (H) levels were also reduced to a much greater extent after MV-DEX treatment than with free DEX treatment. n=6 mice per group. (I) Western blot analysis of p65 and p-p65 in renal cortex tissue lysates showed MV-DEX was more superior in reducing NF-κB activity. n=3. (J) RAW cells were transfected with siRNA to reduce GR expression in the MV-DEX, also GECs were transfected to reduce endogenous GR expression. The experimental groups were as follows: GECs treated by MV-DEX with normal GR (MV-DEX group), GECs treated by MV-DEX with low GR (MV-DEX(GR-) group), GECs with low GR treated by MV-DEX with normal GR (MV-DEX+GEC(GR-) group), GECs with low GR treated by MV-DEX with low GR (MV-DEX(-)+GEC(GR-) group). Expression levels of p65 and p-p65 were detected by Western blot. Reducing GR expression in the MV-DEX or GECs both abated the anti-inflammatory efficacy of MV-DEX. n=3. Data are presented as mean ± SD, * p<0.05, ** p<0.01, *** p<0.001 vs. LPS-treated group, # p<0.05, ## p<0.01, ### p<0.001, N.S., not significant, one-way ANOVA.

Figure 9

MV-DEX therapy ameliorates the adverse effects of DEX in ADR-treated mice. (A) Fasting (6 h) blood glucose (change (Δ) between 14 days of treatment) measured in ADR-induced mice treated with vehicle, free DEX or MV-DEX. (B) Real-time PCR analysis of gluconeogenic genes in livers. (C) Real-time PCR analysis of steroidogenic genes in adrenals. (D)(E) ELISA detection of plasma markers of bone metabolism. n=5 mice per group. Data are presented as mean ± SD, ** p<0.01, *** p<0.001 vs. ADR-treated mice, ## p<0.01, ### p<0.001, one-way ANOVA.