SIRT6 Is Essential for Adipocyte Differentiation by Regulating Mitotic Clonal Expansion

Abstract

Preadipocytes initiate differentiation into adipocytes through a cascade of events. Mitotic clonal expansion, as one of the earliest events, is essential for adipogenesis. However, the underlying mechanisms that regulate mitotic clonal expansion remain elusive. SIRT6 is a member of the evolutionarily conserved sirtuin family of nicotinamide adenine dinucleotide (NAD)+-dependent protein deacetylases. Here, we show that SIRT6 deficiency in preadipocytes blocks their adipogenesis. Analysis of gene expression during adipogenesis reveals that KIF5C, which belongs to the kinesin family, is negatively regulated by SIRT6. Furthermore, we show that KIF5C is a negative factor for adipogenesis through interacting with CK2α', a catalytic subunit of CK2. This interaction blocks CK2α' nuclear translocation and CK2 kinase activity and inhibits mitotic clonal expansion during adipogenesis. These findings reveal a crucial role of SIRT6 in adipogenesis and provide potential therapeutic targets for obesity.

Keywords: CK2; KIF5C; SIRT6; adipogenesis; mitotic clonal expansion.

Figure 1.. SIRT6 Deficiency Impairs Adipogenesis In Vivo and In Vitro

(A) Representative immunofluorescence of subcutaneous adipocytes in WT and SIRT6 MT mice. n = 6 mice/genotype. Scale bars, 100 μm. (B) Quantification of (A). The number of perilipin A-positive cells in the subcutaneous area of six adjacent scapula sections was counted and averaged. One-way ANOVA was used for the statistical analysis. Data are represented as mean ± SD. **p < 0.01. (C and D) SIRT6 deficiency impairs adipogenesis of MEFs. Sirt6^flox/flox MEFs with or without Tam-Cre were treated with 4-HT for 2 days and induced with adipogenesis medium (day 0). (C) Morphological differentiation at 10 days. Top: western blot analysis of SIRT6 level. Center: stained plates. Bottom: representative fields under the microscope. Scale bars, 50 μm. (D) Western blot analysis of the expression of adipogenesis markers and SIRT6 during adipogenesis. (E and F) SIRT6 KD represses adipogenesis of 3T3-L1 cells. 3T3-L1 cells were infected with lentivirus-shSIRT6. Adipogenesis was induced after infection for 2 days (day 0). (E) Morphological differentiation at 8 days. Scale bars, 50 μm. (F) Western blot analysis of the expression of adipogenesis markers and SIRT6 during adipognesis.

Figure 2.. SIRT6 Deficiency Impairs Mitotic Clonal Expansion during Adipogenesis

(A) SIRT6 KD represses adipogenesis at the early stage. Shown are representative western blots and stained plates of WT and SIRT6 KD cells. (B–G) SIRT6 deficiency suppresses cell proliferation during adipogenesis. (B) 3T3-L1 cells or MEFs were induced for adipogenesis. The cells were trypsinized from the plates at different time points and then counted by a cell viability counter (Cellometer Auto 2000). (C) BrdU incorporation of 3T3-L1 cells with or without DMI. 3T3-L1 cells were cultured for different times (0, 16, or 20 hr) and labeled by BrdU for 1 hr. Cells were fixed and subsequently stained for FACS analysis. (D) Percentage of mitotic cells with or without DMI. 3T3-L1 cells were cultured for 20 hr with or without DMI and continued to be culture for 4 hr in the presence or absence of Noc. Next, cells were fixed and subsequently stained for FACS analysis. (E) Percentage of cell death 2 days after induction. 3T3-L1 cells were cultured for 2 days with or without DMI, and cell death was measured by AnnexinV/PI staining. (F) Representative western blot of cell-cycle proteins and SIRT6 during adipogenesis in 3T3-L1 cells. (G) Quantifications of the p27 level from (F). The results were normalized by β-actin level. One-way ANOVA was used for the statistical analysis; n = 3/group. Data are represented as mean ± SD. **p < 0.01, *p < 0.05.

Figure 3.. SIRT6 Negatively Regulates KIF5C Expression during Adipogenesis

(A) Venn diagram showing that two common genes significantly changed when compared between the SIRT6 KD and WT group after induction for 20 or 40 hr. (B) The mRNA levels of the Kif5c gene were measured by real-time PCR during adipogenesis. Top: 3T3-L1 cells. Bottom: Sirt6^flox/flox MEFs. (C) Representative western blot of KIF5C and SIRT6 during adipogenesis. Top: 3T3-L1 cells. Bottom: Sirt6^flox/flox MEFs. (D) SIRT6 negatively regulates KIF5C promoter activity. A luciferase reporter driven by the KIF5C promoter (pGL3-KIF5C Pro) or a control reporter (pGL3-Basic) was transfected into 3T3-L1 cells with an SIRT6-expressing vector or shRNA against the SIRT6 vector, and luciferase activity was measured 24 hr after transfection. Top: OX of SIRT6. Bottom: KD of SIRT6. (E) SIRT6 regulates the levels of H3K9Ac and H3K56Ac at the promoter of KIF5C during adipogenesis. 3T3-L1 cells were cultured for 12 hr with or without DMI and then harvested for ChIP. Top: western blot analysis of the SIRT6 level after infection with shRNA for 2 days. Bottom: the levels of H3K9Ac and H3K56Ac. One-way ANOVA was used for the statistical analysis; n = 3/group. Data are represented as mean ± SD. **p < 0.01, *p < 0.05.

Figure 4.. KIF5C Negatively Regulates Adipogenesis by Blocking Mitotic Clonal Expansion

(A) OX of KIF5C represses adipogenesis. 3T3-L1 cells were infected with a retrovirus expressing KIF5C for 1 day and induced for adipogenesis. Top: western blot analysis of the KIF5C level. Center: stained plates. Bottom: the amount of oil red O was quantified. (B and C) OX of KIF5C inhibited cell proliferation during adipogenesis. 3T3-L1 cells were induced for adipogenesis after infection with a retrovirus expressing KIF5C for 1 day. (B) The cell number was counted during adipogenesis. (C) BrdU incorporation in the control and KIF5C OX groups before and after induction for 16 hr. (D) KIF5C KD enhances adipogenesis in 3T3-L1 cells. Top: western blot analysis of the KIF5C level. Center: stained plates. Bottom: the amount of oil red O was quantified. (E) KIF5C KD increases BrdU incorporation during adipogenesis. (F–I) KIF5C KD partially rescues adipogenesis in SIRT6-deficient cells. For 3T3-L1 cells, the cells were infected with lentivirus-shRNAs against SIRT6 and KIF5C for 2 days. For MEFs, the cells were infected with an adenovirus expressing Cre and lentivirus-shRNA targeting KIF5C for 2 days. (F and H) The SIRT6 and KIF5C mRNA levels in 3T3-L1 cells (F) or MEFs (H) were measured by real-time PCR. (G and I) Morphological differentiation and quantification of the oil red O amount at 8 days (G, 3T3-L1 cells) or 12 days (I, MEFs). One-way ANOVA was used for the statistical analysis; n = 3/group. Data are represented as mean ± SD. **p < 0.01, *p < 0.05.

Figure 5.. SIRT6 Deficiency Inhibits CK2 Activity during Adipogenesis by Blocking Nuclear Translocation of CK2α′ by KIF5C

(A and B) KIF5C has interaction with CK2α′. (A) CoIP analysis of the interaction of endogenous KIF5C and CK2α′ in 3T3-L1 cells. (B) CoIP analysis of the interaction of KIF5C and CK2α′ in 293T cells transfected with KIF5C- and Myc-CK2α′-expressing vector. (C and D) The effect of KIF5C OX on nuclear translocation of CK2α′ during adipogenesis. (C) The CK2α′ level in the nucleus and the KIF5C level in the cytoplasm were analyzed by western blot. T, total lysate. (D) Quantification of the nuclear CK2α′ level in (C). The CK2α′level was normalized by the PARP1 level. (E) SIRT6 deficiency impairs CK2 activity during adipogenesis. The nuclear lysate was extracted from 3T3-L1 cells at different time points after adipogenesis. CK2 activity in nuclear lysate was measured by the kit and normalized by nuclear protein level. (F and G) SIRT6 KD blocks nuclear translocation of CK2α′ in 3T3-L1 cells. (F) The levels of CK2α′ and SIRT6 in the nucleus were measured by western blot analysis. T, total lysate. (G) Quantification of the nuclear CK2α′ level in (F). The CK2α′ level was normalized by the PARP1 level. One-way ANOVA was used for the statistical analysis; n = 3/group. Data are represented as mean ± SD. **p < 0.01, *p < 0.05.

Figure 6.. CK2 Kinase Is Essential for Adipogenesis

(A) CK2α′ KD impairs adipogenesis in 3T3-L1 cells. Top: western blot analysis of CK2α′. Center: plates stained by oil red O. Bottom: the amount of oil red O was quantified. (B–D) CK2α′ KD inhibits cell proliferation during adipogenesis. (B) The cell number was counted at different time points after induction. (C) BrdU incorporation in the control and CK2α′ KD groups before and after induction for 16 hr. (D) Western blot analysis of p27 degradation during adipogenesis in 3T3-L1 cells (E) CK2β KD impairs adipogenesis in 3T3-L1 cells. Top: western blot analysis of CK2β. Center: plates stained by oil red O. Bottom: the amount of oil red O was quantified. (F and G) CK2β KD inhibits cell proliferation during adipogenesis. BrdU incorporation (F) and the percentage of mitotic cells (G) was measured by FACS analysis One-way ANOVA was used for the statistical analysis; n = 3/group. Data are represented as mean ± SD. **p < 0.01, *p < 0.05.

Figure 7.. Deletion of Kif5c Partially Rescues Defects of SIRT6 MT Mice

(A) KIF5C deficiency enhances adipogenesis of 3T3-L1. (B) The amount of oil red O in (A) was quantified. (C) Establishment of KIF5C MT mice. Shown is western blot analysis of the KIF5C level in brain tissues from two different deletion Kif5c mice. KIF5C MT mice have the same genetic background as SIRT6 MT mice (D) KIF5C deficiency enhances adipogenesis of primary white preadipocytes. Top: stained plates. Bottom: representative fields under the microscope. Scale bars, 50 μm. (E) Survival profile of Sirt6^+/+ Kif5c^−/− (n = 10), Sirt6^+/− Kif5c^−/− (n = 30), and Sirt6 ^−/− Kif5c^−/− (n = 19) female and male mice. (F) Relative body weight of mice after birth for 0.5 day. Sirt6^+/+ Kif5c^+/+, n= 12; Sirt6^+/+ Kif5c^−/−, n = 10; Sirt6^+/− Kif5c^−/−, n = 26; Sirt6^−/− Kif5c^−/−, n = 12; Sirt6^−/− Kif5c^+/+, n = 9. (G) Representative immunofluorescence of subcutaneous adipocytes in Sirt6^+/+ Kif5c^−/− and Sirt6^−/− Kif5c^−/− mice. Sirt6^+/+ Kif5c^−/−, n = 8; Sirt6^−/− Kif5c^−/−, n = 8). (H) Quantification of (G). The number of perilipin A-positive cells in the subcutaneous area of adjacent scapula sections was counted and averaged. One-way ANOVA was used for the statistical analysis. Data are represented as mean ± SD. **p < 0.01, *p < 0.05.