Local administration of lipid-silica nanohybrid-carried forskolin modulates thermogenesis in human adipocytes and impedes weight gain in mice

Abstract

Despite encouraging outcomes of the eight FDA approved drugs in lessening obesity burden, they have been associated with side effects caused by the lack of direct action on the adipose tissue. Therefore, we report a nanomedicine that promotes the transformation of local fat-storage white adipocytes, associated with obesity, into thermogenic adipocytes, indicative of enhanced metabolism, to both human cells and mice. Our nanomedicine consisting of a lipid-silica nanohybrid that allows accommodation of the natural supplement forskolin has resulted in successful browning of mature human adipocytes in vitro through upregulating thermogenesis biomarkers on mature human adipocytes and in vivo in mice tissue (achieving 400- and 80-fold increases in UCP1 and Cox7A1, respectively), while significantly enhancing glucose uptake and lipolysis in an acute fashion and significantly preventing weight gain in high fat diet (60% HFD) mice compared to other treatments including liposomal medicine.

Keywords: Adipose tissue; Forskolin; Human adipocytes; Lipids; Mesoporous silica nanoparticles; Obesity; Thermogenesis; Weight gain prevention.

Figure 1.

schematic summarizing the A) goal of the study to inject a reservoir of FSK that promotes the browning of white adipcoytes helping in obesity treatment. B) fabrication of the forskolin silica lipid nanohybrid via the step-by-step formation of LCMSN-FSK depicting the fabrication of liposomes-FSK in parallel to MSN-FSK and their fusion into the final construct LCMSN-FSK.

Figure 2.

Characterization of the nanoparticles. A) TEM images of implemented MSN. Scale bar = 200 nm (B) Hydrodynamic size, polydispersity index and zeta potential evolution of pure liposomes, MSNs and their corresponding FSK-loaded constructs during formation (n=3) C) stability of the construct with two FSK loading capacities in different biorelevant media (n=3) D) DFT approximation for pore size using two different assumptions.

Figure 3.. Loading and release of FSK from LCMSN-FSK

A) TGA curves of different samples providing the FSK%, B) Chromatograms on identical y-scale showing the decrease in signal for iso-forskolin (RT = 2.85 min.) and forskolin (RT = 3.30 min.) at the initial and final collection times for the release study, C) Representative full scan MS spectra for forskolin (RT = 3.30 min) and iso-forskolin (RT = 2.86 min.), D) MS/MS spectra confirming forskolin (RT = 3.30 min) and iso-forskolin (RT = 2.86 min) E) release % FSK using LC/MS in SBF and acetate D) release % of cy5 dimethyl in three different biorelevant media (SBF, acetate and serum), G) TEM micrographs of the remainder of nanoparticles upon 4 days incubation in the release media.

Figure 4.

Differentiation of hASC to mature adipocytes. (A-B) Representative phase microscopy image of (A) undifferentiated hASC and (B) differentiated mature adipocytes respectively. Scale bar = 100 μm. (C) Fluorescent image counterstained with BODIPY, of hASCs after 14 days of differentiation. Scale bar = 100 μm. (D) mRNAs levels of adiponectin and PPARγ were evaluated in adipocytes relative to hASC at d0 (undifferentiated cells). (E) Cell viability of mature adipocytes exposed to different concentrations of free FSK (1, 5, 25, 50 μg/mL of FSK) or LCMSN-FSK with 10% loading (1, 5, 25, 50 μg/mL of FSK loaded in 10, 50, 250, 500 μg/mL of LCMSN-FSK respectively) for 24 hours using CytoTox-Glo cytotoxicity assay. Values are means ± standard errors, from one-way ANOVA analysis followed with Tukey’s multiple comparisons test. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; n=3 biological replicates with 3 technical replicates per group.

Figure 5.

Analysis of LCMSNs uptake in mature adipocytes in vitro and ex vivo. (A) Fluorescence measurement of adipocytes in a microplate reader after treatment with cy3-labeled LCMSN. n = 3 (B) Brightfield microscopy image of mature adipocytes after 24 h of LCMSN treatment, demonstrated presence of red fluorescent LCMSN in adipocytes and not outside the cells. n = 2 biological replicates with 3 technical replicates per group, scale bar = 100 μm. (C) Flow cytometry of adipocytes treated with cy3-labeled LCMSNs at 3h, 6h, 24h and 48h. (D) Representative confocal image and analysis of mature adipocytes with cy3-labeled LCMSNs in between lipid droplets, confirming intracellular presence of nanoparticles, scale bar = 100 μm. E) The fluorescence intensity of LCMSN at different incubation time points was quantified inside adipocytes specified by a selected ROI. n= 25 adipocytes from images taken from 2 biological replicates with 3 technical replicates. Results are presented as mean ± standard deviation. For in vitro assays, data are shown as the mean from 6 different experiments. F) Schematic of what it is described and observed in confocal images. G) Ex vivo representative confocal microscopy image of adipose tissue after mice treatment with cy3-labeled LCMSN for 24 h. LCMSN (red, cy3) are observed in the cytoplasm but not colocalizing with lipid droplets (green, BODIPY) with and in the adipocyte nuclei (dark blue, DAPI), scale bar = 100 μm. Two different mice were used for ex vivo treatment of adipose tissue. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001

Figure 6.

In vitro efficacy studies in mature human adipocytes treated with either LCMSNs loaded with forskolin (LCMSN-FSK), or forskolin alone (FSK) compared to control cells, for different incubation time points: A-B) Gene expression levels of thermogenic marker UCP1 and Cox7A1, C) Glucose uptake and D) lipolysis efficacy. Values are means ± standard errors, from one-way ANOVA analysis followed with Tukey’s multiple comparisons test. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; n=3 biological samples with 3 technical replicates per group.

Figure 7.

A) Oxygen consumption rate (OCR) trace of human adipocytes after different treatments, was determined using a Seahorse XF96 Analyzer. B) Basal respiration, C) proton leak, D) ATP production, E) maximal respiration, and F) spare respiratory capacity were calculated. Results are reported as mean ± standard error of 3 technical replicates. * = p< 0.05, ** = p< 0.01, *** = p< 0.001, and **** = p< 0.0001.

Figure 8.

Biodistribution of DyLight 633-labeled LCMSNs-FSK (upper panels A-C) and FSK fluorescent surrogate cy5 dimethyl (lower panels D-F) after in vivo subcutaneous injection. A) KINO images of mice injected in both flanks with either PBS (left) or Dy633 LCMSN (right), B) ex vivo imaging of organs at t =48 h showing retention only on iWAT, C) quantitative measurement of iWAT retention vs other collected organs for up to 5 days, D-E) cy5 dimethyl distribution in whole mice in function of time and their organ biodistribution upom harvest at t= 3 hours vs t = 4 days. F) Fluorescence efficiency in different organs evolution over 4 days KINO imaging. Images are representatives of three independent experiments (n = 6).

Figure 9.

Thermogenic gene markers analysis, and adipose tissue lipolysis measurements after in vivo subcutaneous injection of LCMSNs-FSK. A-B) Gene expression of UCP1 and Cox7A1, and C) Lipolysis activity in iWAT at different time points post treatment injection. D) Protein expression of UCP1 in iWAT after 24 h treatment injection was assessed by western blot. E) UCP1 band intensity was quantified and normalized to total actin signal. n = 2 animals. Data are presented as mean ± standard error of the mean with *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001 as significant from one-way ANOVA followed with Tukey’s multiple comparison

Figure 10.

FSK dose dependent effect. A-B) Gene expression of UCP1 and COX7A1 C) Glucose uptake and D) lipolysis activity of adipocytes treated with different concentrations of free FSK and their equivalent of LCMSN-FSK. Adipocytes were also treated with different concentrations of unloaded LCMSN as controls. n=3 replicates for all in vitro studies. (G-H) Gene expression of UCP1 and COX7A1 of adipose tissue after 6h of in vivo subcutaneous injection with different doses of free FSK in the right subcutaneous fat pad and equivalent dose of FSK loaded in LCMSN-FSK in the left subcutaneous fat pad. Mice were injected with either saline (control), low dose of FSK (LD: 0.625 mg/kg), medium dose (MD: 2.5 mg/kg) or high dose (HD: 12.5 mg/kg). n= 1 animal per dose run in triplicates. Bars without a common superscript differ, p<0.05 by one-way ANOVA followed with Tukey’s multiple comparisons test.

Figure 11.

LCMSNs-FSK prevention of obesity in DIO C57BL/6J mice. A) Schematic showing the injection treatment frequency and the equivalent forskolin dose administered B) Mice food intake. C) Percentage body weight increase of mice over the six-weeks. n= 3 animals per group. Data are presented as mean ± standard error of the mean with *p<0.05, **p<0.01, ***p<0.001 as significant from one-way ANOVA followed with Tukey’s multiple comparisons test.