The human lipodystrophy gene product Berardinelli-Seip congenital lipodystrophy 2/seipin plays a key role in adipocyte differentiation

Abstract

Mutations in the Berardinelli-Seip congenital lipodystrophy 2 gene (BSCL2) are the underlying defect in patients with congenital generalized lipodystrophy type 2. BSCL2 encodes a protein called seipin, whose function is largely unknown. In this study, we investigated the role of Bscl2 in the regulation of adipocyte differentiation. Bscl2 mRNA is highly up-regulated during standard hormone-induced adipogenesis in 3T3-L1 cells in vitro. However, this up-regulation does not occur during mesenchymal stem cell (C3H10T1/2 cells) commitment to the preadipocyte lineage. Knockdown of Bscl2 by short hairpin RNA in C3H10T1/2 cells has no effect on bone morphogenetic protein-4-induced preadipocyte commitment. However, knockdown in 3T3-L1 cells prevents adipogenesis induced by a standard hormone cocktail, but adipogenesis can be rescued by the addition of peroxisome proliferator-activated receptor-gamma agonist pioglitazone at an early stage of differentiation. Interestingly, pioglitazone-induced differentiation in the absence of standard hormone is not associated with up-regulated Bscl2 expression. On the other hand, short hairpin RNA-knockdown of Bscl2 largely blocks pioglitazone-induced adipose differentiation. These experiments suggest that Bscl2 may be essential for normal adipogenesis; it works upstream or at the level of peroxisome proliferator-activated receptor-gamma, enabling the latter to exert its full activity during adipogenesis. Loss of Bscl2 function thus interferes with the normal transcriptional cascade of adipogenesis during fat cell differentiation, resulting in near total loss of fat or lipodystrophy.

Figure 1

Bscl2 is expressed in adipose tissue and highly up-regulated during adipocyte differentiation. A, Bscl2 mRNA expression in mouse tissues as determined by real-time RT-PCR in epididymal (epW) and sc (scW) white adipose tissue, brown adipose tissue (BAT), brain (Brn), lung (Lng), small intestine (SI), liver (Lvr), adrenal gland (AD), heart (Hrt), skeletal muscle (Skm), testis, kidney (Kdy), and spleen (Spln) isolated from 4-h-fasted, 10-wk-old male C57BL/J6 mice. Data are normalized to cyclophilin A mRNA. Time course of Bscl2 and PPARγ mRNA expression during induced differentiation in 3T3-L1 cells was by DMI (B), mouse stromal vascular cells (SVF) by DMI (C), and human adipocyte stem cells (hAsc) by ADM (D) as described in Materials and Methods. Data are means ± sd (n = 3) relative to d 0. *, P < 0.05; **, P <0.005 vs. d 0. E, 3T3-L1 cells stably infected with retroviruses expressing GFP-tagged Bscl2 were induced to differentiate for 8 d and stained with anticalreticulin antibody. Merged images show overlay of the two proteins in which yellow indicates colocalization (×630). DAPI, 4′6-Diamidino-2-phenylindole.

Figure 2

Bscl2 is not induced and essential during the commitment of C3H10T1/2 mesenchymal stem cells to preadipocytes. A, Bscl2 and Pref-1 mRNA expression in d 0 preadipocytes and d 5 adipocytes with or without BMP4 treatment. Data are normalized to cyclophilin A and expressed as fold of expression compared with control at d 0 without BMP4 treatment. B, Oil-red O staining of d 8 adipocytes with or without BMP4 treatment. C, Bscl2 and Pref-1 mRNA expression were assayed by real-time RT-PCR in C3H10T1/2 cells stably infected with retroviruses expressing control shLuc or Bscl2 targeting shBs13a.in d 0 preadipocytes with or without BMP4 treatment. Data are normalized to cyclophilin A (ppia) and expressed as fold of expression compared with control at d 0 without BMP4 treatment. Data are means ± sd (n = 3). *, P < 0.05; **, P <0.005. D, Semiquantitative RT-PCR analyses of PPARγ expression in C3H10T1/2 cells stably infected with retroviruses expressing shLuc or shBs13a.in d 0 preadipocytes with or without BMP4 treatment. The amounts of cyclophilin A (ppia) are used as loading control. NS, Not significant.

Figure 3

Bscl2 knockdown reduces adipogenesis in 3T3-L1 cells. 3T3-L1 cells were stably infected with retroviruses expressing control shLuc or Bscl2 targeting shBs13a and shBs14a. A, Bscl2 mRNA was assayed by real-time RT-PCR, and values were expressed as fold of expression by normalizing to control cells at d 0 relative to cyclophilin A expression. B, Cells expressing shLuc, shBs13a, or shBs14a were differentiated and lipid accumulation assessed by oil-red O staining (upper panel) and quantitated by triglyceride analysis kit after lipid extraction (lower panel) on d 8. C–E, Expression of mRNA encoding mature adipocyte marker proteins, adipocyte transcription factors, and major triglyceride synthesis enzymes analyzed by real-time PCR in d 6 differentiated shLuc-, shBs13a-, and shBs14a-transduced cells. Data are means ± sd (n = 3). *, P < 0.05; **, P <0.005 vs. control cells at d 6. GPAT, Glycerol phosphate acyltransferase; AGPAT2, l-acylglycerol-3-phosphate-O-acyltransferase 2; ADPN, adiponutrin; DGAT1, diacylglycerol acyltransferase 1; DGAT2, diacylglycerol acyltransferase 2; MGAT1, monoacyglycerol acyltransferase 1.

Figure 4

Rescue of differentiation blockade of Bscl2 knockdown 3T3-L1 cells in the presence of insulin, dexamethasone, and IBMX (DMI) by exogenous PPARγ ligand pio. At d 0 (2 d of confluence), 3T3-L1 cells stably infected with shLuc and shBs13a retroviruses were treated with DMI. One micromole pio was added at d 0, 2, 4, or 6 of differentiation and kept at that level until d 8. A, At d 8, cells were stained with oil-Red-O and photographed under digital camera and microscopy. RNA were extracted at d 8 for real-time RT-PCR analysis of Bscl2 (B), PPARγ (C), C/EBPα (D), and ap2 (E). Data are means ± sd (n = 3). **, P < 0.005 vs. control cells at d 8; ##, P < 0.005 vs. shBs13a cells at d 8.

Figure 5

Inhibition of PPARγ, C/EBPα, Srebp1c, and ap2 but not C/EBPβ induction by Bscl2 knockdown during early adipogenesis and pio rescue of PPARγ target gene expression in 3T3-L1 cells. Control cells or cells expressing Bscl2 targeting shBs13a without or with pio added at d 0 were induced to differentiate by DMI for different periods. The expression of mRNAs was determined by quantitative PCR for Bscl2, C/EBPβ, PPARγ, C/EBPα, Srebp1c, and ap2. Data were normalized to cyclophilin A mRNA expression and presented as fold of expression relative to d 0 control cells. Data are means ± sd (n = 3). *, P < 0.05; **, P < 0.005 vs. control cells at the same time point; #, P < 0.05 vs. control cells at d 0.

Figure 6

Basal Bscl2 expression not increased but required during pio-induced adipogenesis. A, Bscl2 mRNA levels during differentiation in 3T3-L1 cells in the constant presence of exogenous pio (1 μm) in the absence of DMI. B, 3T3-L1 cells stably infected with retroviruses expressing shLuc, shBs13a, and shBs14a in the presence of pio (1 μm) and no DMI. RNAs were extracted at d 0, 2, 4, and 8 for real-time RT-PCR analysis of Bscl2. C, At d 8, cells were stained with oil-Red-O and photographed under microscopy (×100). D, Relative mRNA levels of C/EBPβ, PPARγ, and C/EBPα and mature adipocyte marker ap2 quantified by real-time RT-PCR, normalized to cyclophilin A mRNA expression and relative to d 0 control cells. Data are means ± sd (n = 3). *, P < 0.05; **, P <0.005 vs. shLuc control cells at each time point; #, P < 0.05; ##, P <0.005 vs. control cells at d 0.