Proof-of-concept for CRISPR/Cas9 gene editing in human preadipocytes: Deletion of FKBP5 and PPARG and effects on adipocyte differentiation and metabolism

Abstract

CRISPR/Cas9 has revolutionized the genome-editing field. So far, successful application in human adipose tissue has not been convincingly shown. We present a method for gene knockout using electroporation in preadipocytes from human adipose tissue that achieved at least 90% efficiency without any need for selection of edited cells or clonal isolation. We knocked out the FKBP5 and PPARG genes in preadipocytes and studied the resulting phenotypes. PPARG knockout prevented differentiation into adipocytes. Conversely, deletion of FKBP51, the protein coded by the FKBP5 gene, did not affect adipogenesis. Instead, it markedly modulated glucocorticoid effects on adipocyte glucose metabolism and, furthermore, we show some evidence of altered transcriptional activity of glucocorticoid receptors. This has potential implications for the development of insulin resistance and type 2 diabetes. The reported method is simple, easy to adapt, and enables the use of human primary preadipocytes instead of animal adipose cell models to assess the role of key genes and their products in adipose tissue development, metabolism and pathobiology.

Figure 1

Assessment of mutation efficiency at the DNA level. (a) Schematic representation of the experimental setup of the whole process from collecting human adipose tissue biopsy until an assessment of knockout efficacy. (b–e) Quantification of Sanger sequencing chromatograms by TIDE (Tracking of Indels by Decomposition) of representative transfection experiments of SVF cells transfected with the (b) FK-G57 guide, (c) FK-G54 guide and (d) FK-G66 guide and sequenced in both directions. Similar mutation results were obtained independently of the DNA strand that was sequenced. The size distribution of the insertions (plus) and deletions (minus) is shown on the x-axis and the percentage contribution of each indel to the total efficiency is shown on the y-axis. R² is the correlation coefficient calculated to assess the goodness of fit, and p is the estimated probability of each mutation event. (e) Average total mutation efficiency of 2 to 4 independent transfection experiments. Data are shown as mean ± SEM.

Figure 2

FKBP5 gene deletion and assessment of knockout efficacy. (a) FKBP5 mRNA levels in wild type, negative control, FK-G54 KO, FK-G57 KO, and FK-G66 KO cultures 48 hours after transfection. GUSB was used as a housekeeping gene. The qPCR data were quantified using 2^-dCt and shown as relative to wild type (n = 3 to 5) (b) FKBP5 gene expression in wild type, FK-G57 KO, and negative control cultures on days 0, 7 and 14 of differentiation. (c) FKBP51 protein levels in wild type and FK-G57 KO and negative control cultures on days 0, 7, and 14 of differentiation. A representative blot, along with the quantification graph, is shown. GAPDH was used as an endogenous control (n = 3). The blots are the cropped images from different parts of the same gel. The uncropped gel images are provided in the supplementary file. (d) Representative immunostaining images comparing the protein levels of FKBP51 in differentiated cells from wild type, negative control, and FK-G57 KO cultures. The nuclei are stained with Hoechst 33342 (blue) and FKBP51 stained with Alexa Fluor® 594 (red). (e) Quantification graph showing the percentage of cells positive for the FKBP51 protein in differentiated adipocytes (n = 3). (f) Effect of dexamethasone on the FKBP5 gene expression between wild type, negative control, and FK-G57 KO cultures. Data are shown as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001. WT wild type, KO knockout.

Figure 3

Deletion of PPARG, assessment of mutation and adipogenesis. (a) PPARG mRNA levels in wild type, negative control, PP-G1 KO, PP-G2 KO, and PP-G3 KO cultures 48 hours post-transfection. GUSB was used as a housekeeping gene. The data were quantified using 2^−dCt and shown as relative to wild type (n = 2) (b) A representative western blot image showing the protein levels of PPARG on day 14 of differentiation. GAPDH was used as an endogenous control. The blots are the cropped images from different parts of the same gel. The uncropped gel images are provided in the supplementary file. (c) Immunocytochemistry showing the loss of PPARG protein in differentiating adipocytes on day 7. The nuclei is stained with Hoechst 33342 (blue) and PPARG protein is stained with Alexa Fluor® 594 (red). (d) Quantification graph showing the percentage of cells positive for PPARG protein when counting >800 cells in representative cultures from one transfection experiment. (e) A representative bright-field image from wild type, negative control, and PP-G2 KO cultures depicting the progress of differentiation. For visualization purposes, the neutral lipids on day 14 of differentiation were stained with BODIPY and imaged under the GFP channel (green). (f) A graph showing the quantification of lipid accumulation on day 14 of differentiation (n = 1 to 3). WT wild type, KO knockout.

Figure 4

The loss of FKBP51 did not affect adipogenesis. Preadipocytes from wild type, FKBP51 (FK-G57) and PPARG (PP-G2) knockout cultures were differentiated for 14 days. (a–g) Differentiation was assessed by measuring the expression of adipogenic markers on days 0, 7, and 14 (n = 5). As a positive control for differentiation, the expression of adipogenic markers was also investigated from the PP-G2 knockout cultures (n = 2). (h) The degree of differentiation was evaluated by quantifying the amount of lipid accumulation on day 14 of differentiation. A representative bright-field image from wild type, negative control, and FK-G57 KO cultures depicting the progress of differentiation. For visualization purposes, the neutral lipids on day 14 of differentiation were stained with BODIPY and imaged under the GFP channel (green). (i) Quantification of lipid accumulation on day 14 of differentiation between wild type, negative control, and different FKBP51 knockout cultures. WT wild type, KO knockout.

Figure 5

Effect of dexamethasone on glucose uptake in differentiated adipocytes from wild type and FKBP51 knockout (FK-G57) cultures. Differentiated adipocytes were treated with or without dexamethasone (0.3 µM) for 24 hours, and basal and insulin-stimulated (1000 µU/ml) glucose uptake was measured (n = 6). The basal glucose uptake in wild type cultures was 9.0 ± 5.0 femtoliter/mg/s and in FK-G57 knockout cultures 5.0 ± 1.0 femtoliter/mg/s. Data are shown as mean ± SEM. *p < 0.05.

Figure 6

Dexamethasone increased CRN1 gene expression in FKBP51 knockout (FK-G57) adipocytes. Differentiated adipocytes were treated with or without dexamethasone (0.3 µM) for 24 hours. RNA was extracted, and the expression of the CNR1 gene was assessed using qPCR (n = 6). GUSB was used as an endogenous control. Data are shown as mean ± SEM. **p < 0.01.